Transcriptome analysis reveals comprehensive responses to cadmium stress in maize inoculated with arbuscular mycorrhizal fungi

Ecotoxicology and Environmental Safety 186 (2019) 109744

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety journal homepage: www.elsevier.com/locate/ecoenv

Transcriptome analysis reveals comprehensive responses to cadmium stress in maize inoculated with arbuscular mycorrhizal fungi

T

Longjiang Gua,b,1, Manli Zhaoc,1, Min Gec, Suwen Zhua,c, Beijiu Chenga,c,∗∗, Xiaoyu Lia,b,∗ a

National Engineering Laboratory of Crop Stress Resistance Breeding, Anhui Agricultural University, Hefei, 230036, China Anhui Province Key Laboratory of Crop Biology, Anhui Agricultural University, Hefei 230036, China c Schools of Life Sciences, Anhui Agricultural University, Hefei, 230036, China b

A R T I C LE I N FO

A B S T R A C T

Keywords: Cadmium stress Arbuscular mycorrhizal fungi Maize RNA-Seq Differentially expressed genes

Biological strategy of utilization of plants-microbe's interactions to remediate cadmium (Cd) contaminated soils is effective and practical. However, limited evidence at transcriptome level is available about how microbes work with host plants to alleviate Cd stress. In the present study, comparative transcriptomic analysis was performed between maize seedlings inoculated with arbuscular mycorrhizal (AM) fungi and non-AM fungi inoculation under distinct concentrations of CdCl2 (0, 25, and 50 mg per kg soil). Significantly higher levels of Cd were found in root tissues of maize colonized by AM fungi, whereas, Cd content was reduced as much as 50% in leaf tissues when compared to non-AM seedlings, indicating that symbiosis between AM fungi and maize seedlings can significantly block translocation of Cd from roots to leaf tissues. Moreover, a total of 5827 differentially expressed genes (DEG) were determined and approximately 68.54% DEGs were downregulated when roots were exposed to high Cd stress. In contrast, 67.16% (595) DEGs were significantly up-regulated when seedlings were colonized by AM fungi under 0 mg CdCl2. Based on hierarchical clustering analysis, global expression profiles were split into eight distinct clusters. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed that hundreds of genes functioning in plant hormone signal transduction, mitogen-activated protein kinase (MAPK) signaling pathway and glutathione metabolism were enriched. Furthermore, MapMan pathway analysis indicated a more comprehensive overview response, including hormone metabolism, especially in JA, glutathione metabolism, transcription factors and secondary metabolites, to Cd stress in mycorrhizal maize seedlings. These results provide an overview, at the transcriptome level, of how inoculation of maize seedlings by AM fungi could facilitate the relief of Cd stress.

1. Introduction It is a global environmental puzzle that increasingly larger cultivated land has being contaminated with toxic metals such as lead, mercury and cadmium (Cd). Cadmium is a widespread naturally occurring heavy metal that negatively affects the plant development and growth (Rasmussen et al., 2000). Excess Cd in plants can induce massively toxic reactive oxygen species (ROS), alter physiological processes, and disrupt nutrient homeostasis (DalCorso et al., 2014; Lin and Aarts, 2012). All of these effects are accompanied by damage to nucleic acids, proteins and plasma membrane lipids (Clemens, 2006; Garnier et al., 2006). A recent study illustrated that excessive Cd can inhibit later root formation by disrupting oscillating signals in the root (Xie et al., 2019). More seriously, Cd accumulation in crops reduces yield

production, contaminates edible plant parts and threatens global food safety (Feng et al., 2016). Previous reports illustrated that plants have developed diverse ways to deal with Cd including chelation, efflux (Hu et al., 2013), detoxification or sequestration (Cobbett and Goldsbrough, 2002). More importantly, numerous genes have been functionally characterized for their roles in counteracting the stress stimuli of Cd. The mRNA expression level of one transcription factor, AtWRKY12, was sharply reduced by Cd stress, while loss-of-function atwrky12 mutant exhibited enhanced Cd tolerance and decreased Cd accumulation (Han et al., 2019, p. 12). OsCAL1, encoding a defensin-like protein, can reduce cytosolic Cd content and inhibit long-distance transport from roots to aerial parts via chelating Cd in the cytosol and facilitating Cd secretion to extracellular space (Luo et al., 2018). Variation in OsCd1 was

∗

Corresponding author. National Engineering Laboratory of Crop Stress Resistance Breeding, Anhui Agricultural University, Hefei, 230036, China. Corresponding author. National Engineering Laboratory of Crop Stress Resistance Breeding, Anhui Agricultural University, Hefei, 230036, China. E-mail addresses: [email protected] (B. Cheng), [email protected] (X. Li). 1 Longjiang Gu and Manli Zhao contributed equally to this work. ∗∗

https://doi.org/10.1016/j.ecoenv.2019.109744 Received 26 June 2019; Received in revised form 27 September 2019; Accepted 29 September 2019 0147-6513/ © 2019 Elsevier Inc. All rights reserved.

Ecotoxicology and Environmental Safety 186 (2019) 109744

L. Gu, et al.

2.2. Plant materials and growth condition

responsible for divergence of both root Cd uptake and grain Cd accumulation between two subspecies (indica and japonica) of cultivated rice (Yan et al., 2019). Despite these highlighted progresses in the characterization of key genes that were responsive to Cd stress, however, it remains difficult to utilize these genes to breed highly tolerant plants and eventually to bioremediate Cd contaminated areas. AM fungi, which is a typical symbiont fugus, can establish symbiotic relations with the roots of 80% of all terrestrial plant species (Cabral et al., 2015). The phenomenon that host plants inoculated with AM fungi could remarkably alleviate the Cd toxicity were repeatedly observed in various species, such as in trigonella plants (Abdelhameed and Metwally, 2019), Medicago truncatula (Aloui et al., 2012), Triticum aestivum (Sharma et al., 2016), Solanum photeinocarpum (Tan et al., 2015) and Zea mays (Zhang et al., 2019b). Therefore, utilization of AM fungi appears to be a promising, effective and economical strategy to reduce Cd toxicity and to bioremediate polluted soil. Under different concentrations of CdCl2, AM inoculation could increase various growth parameters, as well as stimulate the production of antioxidant enzymes (Abdelhameed and Metwally, 2019). Moreover, secondary metabolites, such as isoflavonoids and their derivates, significantly accumulated in roots of Medicago truncatula under 2 mg/kg Cd stress, while these metabolites greatly reduced when plant were colonized by AM fungi [16]. Symbiosis between AM fungi and Solanum photeinocarpum significantly enhanced phosphate acquisition and growth under severe Cd stress. Furthermore, the activities of antioxidant parameters, such as catalase and guaiacol peroxidase, were substantially improved when seedlings were inoculated with AM fungal inoculums (Tan et al., 2015). Moreover, AM colonization could change the subcellular distribution of Cd and promote its translocation to vacuoles (Zhang et al., 2019b). These outstanding researches have clearly demonstrated that symbiosis between AM fungi and host plants can moderate Cd phytotoxicity and reduce growth inhibition. However, the molecular mechanism involved in this process still remains elusive. Questions that need to be addressed include: how many genes are differentially expressed? How is the global transcriptome changed? Are there hub genes that play important roles in this process? In order to explore molecular mechanism on how AM fungi could help to alleviate Cd phytotoxicity, AM-inoculated and non-inoculated maize seedlings under distinct concentrations of CdCl2, were subjected to high throughput RNA sequencing, respectively. In line with our expectation, maize seedlings that establish firm symbiotic relation with AM fungi exhibited obviously better growth vigor. In particular, genes involved in JA metabolism and peroxidases were significantly upregulated in colonized seedlings under severe Cd stress. We also identified several key transcription factors that may have roles in regulation of Cd adsorption and translocation in maize seedlings that were inoculated with AM fungi. Our research provided a modest comprehensive transcriptome profiles for deciphering the molecular mechanism involved in the enhanced Cd tolerance of plants that form symbiotic relationships with AM fungi.

Firstly, 200 uniformly sized seeds of maize inbred line B73 were surface-sterilized using 75% alcohol for 5 min, and soaked in water for 1 d. Germinated seeds were then transferred to seed germination fabrics bags filled with vermiculite. After 7 d, a total of 54 seedlings with considerable growth vigor were transplanted into 18 pots (3 seedlings per pot). Afterwards, three pots each, prefilled with 2.5 kg soil matrix, were watered with different solutions containing 0, 62.5, and 125 mg CdCl2, respectively, making the final concentration of CdCl2 to 0, 25 and 50 mg per kg soil matrix, and herein were designated as C0, C25 and C50. While, three pots each were treated with 0, 25, and 50 mg CdCl2 per kg soil matrix and simultaneously inoculated with 5 mL of an AM fungus inoculum (Glomus intraradices, DAOM-197198). Approximately 1000 spores, inspected via hemocytometer, were applied. These treatments were designated as C0I, C25I and C50I, respectively. All seedlings were randomly arranged in a greenhouse for 30 d, with daytime temperature ranging from 25 to 30 °C. Extra artificial illumination was supplemented, with light intensity of 300 μmol m2 s−1 under a 16:8 (L:D) photoperiod. Moreover, a 500 mL Hoagland's (pH = 6.41) nutrient solution was applied to each pot every 3 d throughout the plant growth period. 2.3. Determination of root colonization rate To detect root colonization event, seedlings at 30 d after transplantation were carefully collected and washed with deionized water to remove adherent soil particles. Subsequently, randomly selected fresh root samples were cut into pieces with 1 cm in length, which were then dipped in 10% potassium hydroxide (KOH) solution for 24 h (Phillips and Hayman, 1970). Afterwards, sub-samples were further stained with 0.05% trypan blue in lactoglycerol for 3 h (Giovannetti and Mosse, 1980). Finally, the stained sub-samples were microscopically examined to determine the percentage of root pieces containing arbuscules, vesicles and hyphae. Identification of AM fungi is based mainly on the morphology of spores that formed around or within root systems as referred by Chang et al. (2018). 2.4. Measurement of Cd content Leaves and roots were harvested separately at 30 d after transplantation, and then carefully washed with deionized water. Plant tissues were oven-dried at 80 °C for 3 d. Dried plant tissue was then ground into fine powder. Subsequently, 1 g of tissue powder was digested into 3 parts of diacid mixture containing 1 M HNO3 and 1 M HCl (3:1 ratio, v/v) at 160 °C, which was used to eliminate interfering organic substances. Then the concentration of Cd was measured by a graphite furnace atomic absorption spectrophotometer (ICETM 3300AAS, ThermoFisher, USA). 2.5. RNA-seq and data analysis pipeline

2. Materials and methods

At 30 d post inoculation, three individual seedlings from each group, namely, C0, C25, C50, C0I, C25I and C50I, were separately collected as three biological replicates. Afterwards, root tissue of each seedling was repeatedly washed with deionized water. RNA was isolated and subjected to high throughput sequencing on the Illumina Hiseq 4000 platform at Beijing Genomics Institute (BGI; Shenzhen, China). After trimming low-quality bases and adaptor sequences from the raw sequencing dataset, clean reads were aligned to the reference genome sequence of maize inbred line B73 (Jiao et al., 2017) by Bowtie2 (Langmead and Salzberg, 2012) software. The resulting BAM alignment files were piped to RSEM software (https://github.com/ deweylab/RSEM) to accurately quantify the relative gene expression level. Princomp and cor function in R (https://www.r-project.org/) were used to perform principal component analysis, and Pearson

2.1. Pot experiment The pot experiment was carried out in a completely randomized manner, taking three variables into consideration: three distinct Cd concentrations (0, 25, and 50 mg CdCl2 per kg soil), two treatments (inoculation by AM fungi and non-inoculation by AM fungi), and three replicates for each treatment. Therefore, a total of 18 pots were used in the present study. The pot volume was 4.5 L, and pot dimensions were top diam. 25 cm, bottom diam. 13 cm, and depth 16 cm. Each pot was filled with 2.5 kg of a soil matrix (sandy soil:vermiculite = 1:1, v/v, pH = 6.72), which was previously sterilized at 121 °C for 1 h to eliminate microbes and fungal spores. 2

Ecotoxicology and Environmental Safety 186 (2019) 109744

L. Gu, et al.

3. Results

correlation analysis, respectively. Furthermore, R package DEGseq was explored to determine differentially expressed genes (DEGs) as previously described (Anders and Huber, 2010). Both GO and KEGG analysis were conducted using ClusterProfiler package (Yu et al., 2012) in Bioconductor (Version 3.6). GO items or KEGG pathways with FDR≤0.01 were considered to be significantly enriched. Furthermore, MAPMAN software was used to display the expression profiles of differentially expressed genes onto various metabolic pathways (Thimm et al., 2004). The Venn diagram was drawn using the BioVenn online tool (Hulsen et al., 2008).

3.1. Arbuscular mycorrhizal fungi promote growth of maize seedlings exposed to cadmium Dry weight of root tissues of C50 declined as much as 39.17% on average when compared to seedlings from C0 (Figure 1A). Whereas, difference in dry weight of shoot tissues between seedlings from C0 and C50 was statistically insignificant (two-tailed Student's t-test, Pvalue = 0.8367, Figure 1B). Moreover, significantly greater dry weight of both roots and shoots were repeatedly observed for seedlings in C50I compared to seedlings in C50 (two-tailed Student's t-test, Pvalue = 0.00052, Figure 1A and 1B). These results indicated that high concentration of CdCl2 can cause growth damage to root tissues. Symbiosis between AM fungi and maize seedlings can greatly alleviate the damage caused by Cd stress.

2.6. Validation of RNA-Seq via qRT-PCR assay Root tissues of the 18 samples were carefully collected and total RNA was extracted and purified with DNase by Beijng Genomics Institute (BGI; Shenzhen, China). The first-strand cDNA was generated using the reverse transcription system from Promega company (Promega, Madison, WI, USA). The qRT-PCR was carried out using SYBR Green Master (Roche, Basel, Switzerland) on an ABI 7300 Real Time PCR system (Applied Biosystem, Förster City, CA, USA), and the reaction condition was based on a previous report (Zha et al., 2019): 50 °C for 2 min, then 95 °C for 10 min, followed by 38 cycles of 95 °C for 15 s and 60 °C for 1 min. To ensure the robustness of our data, three technical replicates were quantified for each cDNA sample. The raw dataset was further dealt with 2_DDCt method. Additionally, the expression level of maize Actin1 gene was utilized as internal control to normalize the detection threshold. The primer sequences used in the present study were summarized in Table S1.

3.2. Arbuscular mycorrhizal fungi facilitate absorption and affects translocation of cadmium from roots to aerial parts of maize seedlings In order to further determine if the superior phenotypic performances were aided by AM fungi, root sections were collected and dyed with trypan blue accordingly. In accordance with our expectation, a larger number of hyphae and vesicle were found in the inoculated group than in the non-inoculated group (Figure 2), indicating that a robust symbiosis relation was formed between maize seedlings and AM fungi. Moreover, seedlings in C50I showed fewer hyphae as well as vesicle compared to C0I, suggesting that high concentration of CdCl2 can damage the symbiotic process between the host plants and AM fungi. Afterwards, Cd content was carefully measured in root, stem and leaf tissues of maize seedlings. Surprisingly, significantly higher levels of Cd were found in root tissues of maize inoculated with AM Fungi (two-tailed Student's t-test, P-value = 0.033), whereas, no significant difference was observed in stem tissues in AM inoculated and non-AM

Fig. 1. Growth parameters and Cd content of seedlings from AM and non-AM groups. (A) Dry weight of roots; (B) Dry weight of shoots; (C) Cadmium content in root, stem and leaf of maize seedlings with either inoculated AM fungi or not inoculated; and (D) Cadmium content of rhizosphere soil and bulk soil. 3

Ecotoxicology and Environmental Safety 186 (2019) 109744

L. Gu, et al.

Fig. 2. Influence of Cd stress on the symbiosis between AM fungi and root tissues via trypan blue staining. (A) Root sections of C0; (B) Root sections of C0I; (C) Root sections of C50; and (D) Root sections of C50I. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

separately from samples from AM-group based on PC1, suggesting that symbiosis between AM fungi and maize seedlings remarkably changed their transcriptome profiles.

seedlings (Figure 1C). Intriguingly, compared to seedlings that were not inoculated with AM fungi, Cd content declined by as much as 50% in leaves tissues of seedlings that were inoculated by AM fungi (Figure 1C). Furthermore, determination of Cd content in soil provided extra evidence. Cadmium content in rhizosphere soil around mycorrhizal roots declined by 17.5 mg per kg soil, which was much more than that of non-AM group (Figure 1D). These results illustrated that AM fungi could significantly prohibit translocation of Cd from roots to leaf tissues. Thereby, lower level of Cd in leaves might be less harmful to the normal development and photosynthesis processes.

3.4. Global transcriptome profile changes in roots of maize seedlings under cadmium stress and inoculation by AM fungi Given the fact that Cd stress could cause severe damage to the development of maize seedlings, how could Cd stress affect the phenotypic performance at the transcriptome level? Collectively, four comparisons, namely, C0–C25, C0–C50, C0I–C25I, and C0I–C50I, were conducted to identify genes that were differentially expressed under Cd stress. Three comparisons, including C0–C0I, C25–C25I, and C50–C50I, were performed to determine genes enhancing tolerance to Cd when maize was inoculated with AM fungal propagules. After filtering out genes with extremely low expression, 5827 DEGs were detected at least in one comparison (Table S2). The largest number of DEGs (2335) was detected in the C0I–C50I comparison, and an equal number of DEGs were detected with respect to the comparison of C0I–C25I (2323) (Figure 3B, Table S3). While, the least number (886) of DEGs was observed in C0–C0I comparison (Figure 3B). Interestingly, approximately 68.54% DEGs were downregulated when roots were exposed to high Cd stress. In contrast, among the DEGs detected in the C0–C0I comparison, 67.16% (595) were upregulated when maize was inoculated with AM fungal propagules (Figure 3B). Furthermore, 122 DEGs were shared in all four comparisons (Figure 3C). Similarly, among the 886 DEGs detected between the C0–C0I comparison, 441 (49.77%) were shared by either C25–C25I or C50–C50I, and 77 common DEGs were detected for all three comparisons (Figure 3D, Table S4). In order to further examine the expression patterns among the 18 RNA libraries, Genesis software was utilized. The expression profiles of

3.3. Transcriptome profiles of 18 RNA libraries from roots of maize seedlings exposed to combinations of cadmium and arbuscular mycorrhizal fungi To further investigate the global transcriptome changes of maize seedlings in response to Cd and AM fungi, root tissues of seedlings from C0, C25, C50, C0I, C25I and C50I were collected and subjected to highthroughput RNA sequencing. To ensure the robustness of our data, three individual seedlings were randomly selected from each treatment and used as three biological replicates. Therefore, a total of 18 independent RNA libraries were constructed. Overall, 730 million raw paired-end (PE) reads were obtained and about 35–37 million clean reads were retained per library after trimming low-quality bases and adaptor sequences (Table 1). To provide an overview of the transcriptome profile, a principal component analysis was performed using normalized read counts from DESeq package. Most of biological replicates from the same treatment clustered together, indicating that the datasets were of satisfactory quality. The PC1 and PC2 could explain 39% and 19% of transcriptome variations among 18 samples, respectively (Figure 3A). Moreover, samples from non-AM group clustered 4

Ecotoxicology and Environmental Safety 186 (2019) 109744

L. Gu, et al.

Table 1 Summary of RNA Sequencing and mapping using the Zea mays genome as the reference. Sample

C0_1 C0_2 C0_3 C25_1 C25_2 C25_3 C50_1 C50_2 C50_3 C0I_1 C0I_2 C0I_3 C25I_1 C25I_2 C25I_3 C50I_1 C50I_2 C50I_3

Treatments Cd (mg/kg)

AMF

0 0 0 25 25 25 50 50 50 0 0 0 25 25 25 50 50 50

– – – – – – – – – + + + + + + + + +

Total reads

Clean reads

Clean bases (Gb)

Filter (%)

GC (%)

Total mapped (%)

Uniquely mapped (%)

40,818,620 40,818,498 40,819,732 40,819,434 40,819,050 40,818,526 40,820,118 40,819,960 40,819,870 40,818,930 40,818,600 39,186,226 40,818,264 40,819,056 39,185,892 39,186,034 40,818,702 40,819,996

36,509,422 36,795,100 36,492,396 36,543,974 36,235,922 36,524,182 37,028,172 36,426,654 36,523,848 36,103,496 36,171,714 35,835,686 36,766,566 36,513,284 35,673,962 36,073,314 36,942,744 36,927,298

5.48 5.52 5.47 5.48 5.44 5.48 5.55 5.46 5.48 5.42 5.43 5.38 5.51 5.48 5.35 5.41 5.54 5.54

10.56 9.86 10.60 10.47 11.23 10.52 9.29 10.76 10.52 11.55 11.38 8.55 9.93 10.55 8.96 7.94 9.50 9.54

56.95 56.29 58.78 57.14 57.5 58.4 56.03 56.57 55.86 56.85 57.6 57.38 59.47 57.28 61.14 56.49 55.48 56.35

84.82 85.26 84.98 83.83 84.07 84.63 84.3 85.07 86.45 85.23 83.48 87.64 84.45 84.67 87.01 86.59 78.92 85.12

74.12 74.62 73.97 73.39 73.88 73.69 73.44 74.74 75.8 74.95 73.07 77.17 73.54 74.34 76.31 76.27 69.65 74.83

gradually down-regulated as the Cd concentration increased. The highest expression of the cluster1 genes was in C0 and C0I, while the lowest expression was in C50 and C50I. Intriguingly, genes were slightly up-regulated in C0I and C25I when compared to C0 and C25,

5827 DEGs were then divided into 8 distinct clusters (Figure 4 and Figure S1). KEGG analysis was subsequently performed for each cluster using the Clusterprofiler package in Bioconductor resources. As is depicted in Figure 4, the mRNA expression level of genes in cluster1 was

Fig. 3. Identification and classification of differentially expressed genes. (A) Principal component analysis for normalized reads for all 18 samples; PCA1 and PCA2 denote the first two principle components. (B) Bar plot of DEGs for 7 comparisons; FC denotes the fold change for gene expression. (C) Venn diagram for samples watered with diverse concentrations of Cd solution. (D) Venn diagram for comparisons between AM-group and non-AM group, regardless of Cd concentration. 5

Ecotoxicology and Environmental Safety 186 (2019) 109744

L. Gu, et al.

Fig. 4. Heatmap and KEGG pathway enrichment analysis performed for DEGs from each cluster.

and microtubule cytoskeleton (Figure 4 and Figure S7).

respectively. Several pathways, including plant-pathogen interaction, plant hormone signal transduction and MAPK signaling, were significantly enriched in cluster1 (Fisher's exact test, P-value = 0.0016, Figure 4). Reactive oxygen species metabolic process, peroxidase activity, and phosphatase activity were significantly enriched for genes in cluster1 based on GO analysis (Figure S2). In contrast, genes in cluster3 were down-regulated in seedlings treated with 25 mg CdCl2 per kg soil (including C25 and C25I), and remarkably up-regulated in seedlings of C50 and C50I. Particularly, genes from cluster3 showed highest expression in all 18 libraries. Surprisingly, great majority of DEGs in cluster3 were involved in DNA processes, such as base excision repair and homologous recombination (Figure 4 and Figure S3). Response to the symbiotic fungus was also significantly enriched in cluster3, indicating that DEGs in cluster3 might play important roles in regulation of the cell cycle and symbiosis. In cluster4, cluster5 and cluster6, the DEGs were up-regulated to levels proportional to the increased concentration of Cd (C25 and C50), and their expressions were further upregulated after inoculation by AM fungi. However, the mRNA expression levels were down-regulated in the C50I treatment (Figure 4). These DEGs were mostly enriched in phenylpropanoid biosynthesis, which is a critical step in cell wall synthesis. GO analysis repeatedly demonstrated that cell wall related processes were significantly enriched. Moreover, fatty acid metabolic process, lipid transport, and lipid binding process were also enriched (Figures S4 and S5). Genes that were responsive to karrikin and various phytohormones, such as salicylic acid and gibberellin, were also enriched (Figure S6). In cluster7, all DEGs were significantly up-regulated in seedlings inoculated with AM fungi in comparison to non-inoculated seedlings (C0–C0I, C25–C25I, and C50–C50I). These DEGs were mainly involved in flavonoid biosynthesis and photosynthetic membrane (Figure 4). With respect to cluster8, all genes were continuously expressed in all samples but C25I, and overwhelmingly larger number of DEGs were responsible for basic metabolic processes, such as DNA polymerase complex, RNA modification,

3.5. Inoculation by AM fungi upregulated various aspects of biological processes to promote growth In order to decipher the molecular mechanism for obviously better phenotypic performance when seedlings were inoculated with AM fungi, transcriptome profiles were analyzed with MapMan tools, which would give us an overview concerning about the global metabolic changes. As shown in Figure 5, a total of 4485 DEGs that were mapped to annotations in MapMan in the C25–C25I comparison. Of these, 36.74% (1648) were uniquely enriched in biotic stress. Numerous genes were mapped to hormone signaling process, including auxins, ethylene, abscisic acid, and JA (Table S5). This indicated that hormone signaling, especially JA, is important in plant-microbe symbiosis and probably contributes to reduction of the Cd stress. In accordance with previous results (Figure 5), we repeatedly found that the majority of genes, involving in redox state, peroxidases and glutathione-S-transferase, were up-regulated in C25I compared to that in C25 (Table S6), indicating that inoculation by AM fungi can help maize balance the inner redox state. Moreover, large proportions of genes encoded MAPKs and participated in signaling transduction processes in C25I, leading to upregulation of dozens of transcription factors, including WRKY, ERF, MYB and etc. (Table S7). More importantly, overwhelming proportion of genes were mapped to various secondary metabolites (Table S8), indicating that secondary metabolites probably played critical roles in helping host plant to adapt to the severe Cd stress. Collectively, when seedlings suffer from severe Cd stress, symbiosis with AM fungi can invoke systematic changes at the transcriptome level to help the host plant survive.

6

Ecotoxicology and Environmental Safety 186 (2019) 109744

L. Gu, et al.

Fig. 5. Representative image of biotic stress pathways accomplished via MapMan software for DEGs detected in comparison between C25I and C25.

4. Discussion

Zm00001d017292 (homolog to OsCAL1), and Zm00001d051936 (homolog to AtPP2-B15), were significantly upregulated in root tissues of seedlings inoculated with AM fungi (Figure S8). As described previously, OsCAL1 (Luo et al., 2018), OsHAM2 (Takahashi et al., 2012), and AtPP2-B15 (Hou et al., 2017) were well functionally characterized genes, playing key roles in absorption of Cd into root tissues. Moreover, the Cd content in aerial part, especially in the leaves of maize seedlings inoculated with AM sharply reduced in comparison to non-AM seedlings. This suggested that symbiosis between AM and host plants blocks the translocation of Cd from roots to leaves. The leaves are therefore protected from Cd damage to photosynthesis and normal growth. Collectively, the symbiosis of AM fungi with host plants such as maize can help accumulation of toxic metals in root and straw part, then it may contribute to reduced toxicity of Cd to maize leaves. It is also noteworthy to point out that heavy metal ecotoxicity can also be greatly affected by several factors, including active concentration of metal in soil solution, presence of other cations (Ca2+ and Mg2+), and particularly the pH of the soil solution (Sydow et al., 2017; Thakali et al., 2006). Toxicity of Cu2+ and Ni2+ can be greatly reduced with pH change from 6.5 to 4.5, and reduction of metal toxicity can be enhanced with presence of Ca2+ or Mg2+ (Thakali et al., 2006). In the present study, the pH of the soil solution is almost neutral (pH = 6.7), which suggested that the toxicity of Cd should not be influenced by the pH of the studied soil. The symbiosis of AM fungi and maize modified the plant hormone metabolism, especially JA, which probably played an important role in helping the host plants cope with Cd stress. Currently, a highlighted previous research illustrated that Cd treatment can sharply increase the mRNA expression level of several key genes mainly involved in endogenous synthesis of JA in Arabidopsis thaliana (Lei et al., 2019). A loss-of-function mutant of a critical JA synthesis gene (AtAOS) exhibited increasing sensitivity to Cd stress and accumulated much more Cd in both roots and shoots (Lei et al., 2019). In line with these results, we observed that large proportion of genes involved in JA metabolism were differentially expressed in C25I compared to C25. Apart from JA pe se, it has been reported that robust root architecture was of great importance when plant is exposed to severe heavy metal pollution. Thus, root architecture remodeling can provide additional strategy to avoid heavy metal stress. Interestingly, root architecture patterning and

Heavy metals, including Cd, lead, and arsenic, are deleterious abiotic factors reducing plant growth, decreasing food production, and directly affecting human health (Luo et al., 2018). Therefore, comprehensive understanding about how plants absorb and translocate these heavy metals are becoming increasingly important (Jalmi et al., 2018). Up to date, many highlighted researches have characterized dozens of genes involved in Cd absorption, translocation, chelation, and detoxification in model species, such as Oryza sativa (Das et al., 2017) and Arabidopsis thaliana (Amaral dos Reis et al., 2018). Currently, the use of naturally occurring symbiotic relationships between AM fungi and host plants to remediate contaminated soils is considered to be another applicable, effective and economical strategy (Abdelhameed and Metwally, 2019). Numerous previous studies have demonstrated that symbiotic relationship between AM fungi and host plant can alleviate Cd stress via affecting the accumulation and partitioning of Cd and altering synthesis process of glutathione (Zhang et al., 2019a). However, a comprehensive analysis of dynamic transcriptome profiles will enable us to gain a global overview of how symbiosis between AM fungi and host plants contributes to alleviation of severe Cd stress. In accordance to previous reports, Cd stress can substantially impair growth of maize seedlings, but inoculation by AM fungi reduced Cd toxicity. Maize seedlings colonized by AM fungi had increased biomass increased biomass as well as robust root structure (Abdelhameed and Metwally, 2019). It has been well recognized that nutrients exchange between host plants and AMF is the main benefit for the two symbiotic partners (Wipf et al., 2019). For example, Pi can be transported from rhizosphere to plant organs via phosphate transporter (PHT) protein family (Victor Roch et al., 2019). The mRNA expression of such phosphate transporter can be greatly induced by AM fungi (Liu et al., 2018). Enhanced phosphate acquisition was also detected in Solanum photeinocarpum inoculated with AM fungi under severe Cd stress (Tan et al., 2015). Therefore, reduced Cd toxicity may be conferred via enhanced nutrients exchange between AM fungi and host plants. Interestingly, we found that Cd content was much higher in roots of inoculated plants, indicating greater absorption activities compared to non-AM seedlings. Three genes, Zm00001d014669 (homolog to OsHAM2), 7

Ecotoxicology and Environmental Safety 186 (2019) 109744

L. Gu, et al.

Appendix A. Supplementary data

lateral root formation can be partially modulated by phytohormones (De Smet et al., 2015), indicating that phytohormones can indirectly help plants survive when faced with severe heavy metal pollution. In Arabidopsis, Cu2+ toxicity leads to variation in cytokinin and auxin accumulations within root tips (Lequeux et al., 2010). The JrVHAG1 gene functions as a CdCl2 stress response regulator by participating in ABA-signal pathway (Xu et al., 2018). We observed that maize seedlings inoculated with AM fungi exhibited a robust root structure and higher biomass under Cd stress, which might be regulated by various plant hormones (Figure 5). It has been well recognized that Cd stress can increase the production of ROS, which damage membranes, destroy cellular organelles, and then disrupt normal physiology and development of plants (Gupta et al., 2017). To scavenge these harmful ROS, plants up-regulate the mRNA expression level of various antioxidative enzymes, such as superoxide dismutase, catalase and glutathione reductase and eventually accumulate glutathione and phytochelatin (Zhang et al., 2019a). We also observed that hundreds of genes, involving peroxidases and glutathione-S-transferase, were up-regulated in C25I compared to C25, indicating that inoculation of maize seedlings by AM fungi increase the mRNA expression further to maintain ROS balance. Supporting this hypothesis, two genes encoding glutathione synthetase, namely, Zm00001d007670 (homolog to AtGSH2 (Han et al., 2019)) and Zm00001d040146, were upregulated in C25 and C50 compared to C0, indicating that these genes were induced by Cd stress. Interestingly, the mRNA expression was much higher in both C25I and C50I in comparison to C25 and C50, respectively (Figure S8). These results indicated that inoculation of maize plants by AM fungi can stimulate the expression of genes involved in antioxidant synthesis.

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ecoenv.2019.109744. Research data The raw sequencing dataset has been submitted to National Genomics Data Center at the Beijing Institute of Genomics, under accession number CRA001924. References Abdelhameed, R.E., Metwally, R.A., 2019. Alleviation of cadmium stress by arbuscular mycorrhizal symbiosis. Int. J. Phytoremediation 21, 663–671. Aloui, A., Dumas-Gaudot, E., Daher, Z., van Tuinen, D., Aschi-Smit, S., Morandi, D., 2012. Influence of arbuscular mycorrhizal colonisation on cadmium induced Medicago truncatula root isoflavonoid accumulation. Plant Physiol. Biochem. 60, 233–239. Amaral dos Reis, R., Keunen, E., Mourato, M.P., Martins, L.L., Vangronsveld, J., Cuypers, A., 2018. Accession-specific life strategies affect responses in leaves of Arabidopsis thaliana plants exposed to excess Cu and Cd. J. Plant Physiol. 223, 37–46. Anders, S., Huber, W., 2010. Differential expression analysis for sequence count data. Genome Biol. 11, R106. Cabral, L., Soares, C.R.F.S., Giachini, A.J., Siqueira, J.O., 2015. Arbuscular mycorrhizal fungi in phytoremediation of contaminated areas by trace elements: mechanisms and major benefits of their applications. World J. Microbiol. Biotechnol. 31, 1655–1664. Chang, Q., Diao, F., Wang, Q., Pan, L., Dang, Z., Guo, W., 2018. Effects of arbuscular mycorrhizal symbiosis on growth, nutrient and metal uptake by maize seedlings (Zea mays L.) grown in soils spiked with Lanthanum and Cadmium. Environ. Pollut. 241, 607–615. Clemens, S., 2006. Toxic metal accumulation, responses to exposure and mechanisms of tolerance in plants. Biochimie 88, 1707–1719. Cobbett, C., Goldsbrough, P., 2002. PM: roles in heavy metal detoxification and homeostasis. Annu. Rev. Plant Biol. 53, 159–182. DalCorso, G., Farinati, S., Furini, A., 2014. Regulatory networks of cadmium stress in plants. Plant Signal. Behav. 5, 663–667. Das, N., Bhattacharya, S., Bhattacharyya, S., Maiti, M.K., 2017. Identification of alternatively spliced transcripts of rice phytochelatin synthase 2 gene OsPCS2 involved in mitigation of cadmium and arsenic stresses. Plant Mol. Biol. 94, 167–183. De Smet, S., Cuypers, A., Vangronsveld, J., Remans, T., 2015. Gene networks involved in hormonal control of root development in Arabidopsis thaliana: a framework for studying its disturbance by metal stress. Int. J. Mol. Sci. 16, 19195–19224. Feng, S.J., Liu, X.S., Tao, H., Tan, S.K., Chu, S.S., Oono, Y., Zhang, X.D., Chen, J., Yang, Z.M., 2016. Variation of DNA methylation patterns associated with gene expression in rice (Oryza sativa) exposed to cadmium: DNA methylation in Cd-exposed rice. Plant Cell Environ. 39, 2629–2649. Garnier, L., Simon-Plas, F., Thuleau, P., Agnel, J.-P., Blein, J.-P., Ranjeva, R., Montillet, J.L., 2006. Cadmium affects tobacco cells by a series of three waves of reactive oxygen species that contribute to cytotoxicity. Plant Cell Environ. 29, 1956–1969. Giovannetti, M., Mosse, B., 1980. An evaluation of techniques for measuring vesicular arbuscular mycorrhizal infection in roots. New Phytol. 84, 489–500. Gupta, D.K., Pena, L.B., Romero‐Puertas, M.C., Hernández, A., Inouhe, M., Sandalio, L.M., 2017. NADPH oxidases differentially regulate ROS metabolism and nutrient uptake under cadmium toxicity. Plant Cell Environ. 40, 509–526. Han, Y., Fan, T., Zhu, X., Wu, X., Ouyang, J., Jiang, L., Cao, S., 2019. WRKY12 represses GSH1 expression to negatively regulate cadmium tolerance in Arabidopsis. Plant Mol. Biol. 99, 149–159. Hou, J., Liu, X., Cui, B., Bai, J., Wang, X., 2017. Concentration- dependent alterations in gene expression induced by cadmium in Solanum lycopersicum. Environ. Sci. Pollut. Res. 24, 10528–10536. Hu, J., Wu, S., Wu, F., Leung, H.M., Lin, X., Wong, M.H., 2013. Arbuscular mycorrhizal fungi enhance both absorption and stabilization of Cd by Alfred stonecrop (Sedum alfredii Hance) and perennial ryegrass (Lolium perenne L.) in a Cd-contaminated acidic soil. Chemosphere 93, 1359–1365. Hulsen, T., de Vlieg, J., Alkema, W., 2008. BioVenn–a web application for the comparison and visualization of biological lists using area-proportional Venn diagrams. BMC Genomics 9, 488. Jalmi, S.K., Bhagat, P.K., Verma, D., Noryang, S., Tayyeba, S., Singh, K., Sharma, D., Sinha, A.K., 2018. Traversing the links between heavy metal stress and plant signaling. Front. Plant Sci. 9. Jiao, Y., Peluso, P., Shi, J., Liang, T., Stitzer, M.C., Wang, B., Campbell, M.S., Stein, J.C., Wei, X., Chin, C.-S., Guill, K., Regulski, M., Kumari, S., Olson, A., Gent, J., Schneider, K.L., Wolfgruber, T.K., May, M.R., Springer, N.M., Antoniou, E., McCombie, W.R., Presting, G.G., McMullen, M., Ross-Ibarra, J., Dawe, R.K., Hastie, A., Rank, D.R., Ware, D., 2017. Improved Maize Reference Genome with Single-Molecule Technologies. Nature advance online publication. Langmead, B., Salzberg, S.L., 2012. Fast gapped-read alignment with Bowtie2. Nat. Methods 9, 357–359. Lei, G.J., Sun, L., Sun, Y., Zhu, X.F., Li, G.X., Zheng, S.J., 2019. Jasmonic acid alleviates cadmium toxicity in Arabidopsis via suppression of cadmium uptake and translocation. J. Integr. Plant Biol. https://doi.org/10.1111/jipb.12801. Lequeux, H., Hermans, C., Lutts, S., Verbruggen, N., 2010. Response to copper excess in

5. Conclusions Plants inoculated with AM fungi exhibit better physiological and morphological potential when exposed to Cd stress, which was considered to result of increased uptake of immobile nutrients, such as P, S and Cu, and decreased metal toxicity. Herein, we found that large number of genes involved in antioxidant synthesis and responsive to various plant hormones were significantly enriched. Moreover, transcription factors and secondary metabolites may also play important roles in facilitating the relief of Cd stress and contribute to better phenotypic performance. This study provided an overview of the global transcriptomic responses in maize seedlings inoculated by AM fungi when exposed to severe Cd stress. Special attention should also be payed to elucidate the potential effect at the reproductive stages, for example, yield, and Cd content in maize kernels. If the Cd content can be significantly reduced in maize kernels when symbiosis is formed between AM fungi and maize plants, it will be of great importance to ensure safe production of maize in Cd-contaminated areas. Additionally, future works should focus on identification of hub genes and construction of regulatory networks that contributed to the alleviation of Cd toxicity when maize is inoculated by AM fungi.

Competing interests The authors declare that they have no competing interests.

Acknowledgments This research was supported by grants from the National Natural Science Foundation of China (Nos. 31870415, 31640057, and 31701436); The special project of local science and technology development guided by the central government of Anhui province (No. 2018080503B0015). We thank LetPub (www.letpub.com) for its linguistic assistance during the preparation of this manuscript. 8

Ecotoxicology and Environmental Safety 186 (2019) 109744

L. Gu, et al.

Thimm, O., Bläsing, O., Gibon, Y., Nagel, A., Meyer, S., Krüger, P., Selbig, J., Müller, L.A., Rhee, S.Y., Stitt, M., 2004. MapMan: a user-driven tool to display genomics data sets onto diagrams of metabolic pathways and other biological processes. Plant J. 37, 914–939. Victor Roch, G., Maharajan, T., Ceasar, S.A., Ignacimuthu, S., 2019. The role of PHT1 family transporters in the acquisition and redistribution of phosphorus in plants. Crit. Rev. Plant Sci. 38, 171–198. Wipf, D., Krajinski, F., van Tuinen, D., Recorbet, G., Courty, P., 2019. Trading on the arbuscular mycorrhiza market: from arbuscules to common mycorrhizal networks. New Phytol. 15775 nph. Xie, Y., Wang, J., Zheng, L., Wang, Y., Luo, L., Ma, M., Zhang, C., Han, Y., Beeckman, T., Xu, G., Cai, Q., Xuan, W., 2019. Cadmium stress suppresses lateral root formation by interfering with the root clock. Plant Cell Environ. 13635 pce. Xu, Z., Ge, Y., Zhang, W., Zhao, Y., Yang, G., 2018. The walnut JrVHAG1 gene is involved in cadmium stress response through ABA-signal pathway and MYB transcription regulation. BMC Plant Biol. 18. Yan, H., Xu, W., Xie, J., Gao, Y., Wu, L., Sun, L., Feng, L., Chen, X., Zhang, T., Dai, C., Li, T., Lin, X., Zhang, Z., Wang, X., Li, F., Zhu, X., Li, J., Li, Z., Chen, C., Ma, M., Zhang, H., He, Z., 2019. Variation of a major facilitator superfamily gene contributes to differential cadmium accumulation between rice subspecies. Nat. Commun. 10, 1–12. Yu, G., Wang, L.-G., Han, Y., He, Q.-Y., 2012. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS A J. Integr. Biol. 16, 284–287. Zha, K., Xie, H., Ge, M., Wang, Z., Wang, Y., Si, W., Gu, L., 2019. Expression of Maize MADS Transcription factor ZmES22 negatively modulates starch accumulation in rice endosperm. Int. J. Mol. Sci. 20, 483. Zhang, B.-L., Guo, C.-C., Ding, F., Lu, Y.-T., Fu, Z.-W., 2019a. 14-3-3s function in plant cadmium response by changes of glutathione and glutathione synthesis in Arabidopsis. Environ. Exp. Bot. 163, 69–77. Zhang, X.-F., Hu, Z.-H., Yan, T.-X., Lu, R.-R., Peng, C.-L., Li, S.-S., Jing, Y.-X., 2019b. Arbuscular mycorrhizal fungi alleviate Cd phytotoxicity by altering Cd subcellular distribution and chemical forms in Zea mays. Ecotoxicol. Environ. Saf. 171, 352–360.

Arabidopsis thaliana: impact on the root system architecture, hormone distribution, lignin accumulation and mineral profile. Plant Physiol. Biochem. 48, 673–682. Lin, Y.-F., Aarts, M.G.M., 2012. The molecular mechanism of zinc and cadmium stress response in plants. Cell. Mol. Life Sci. 69, 3187–3206. Liu, F., Xu, Y., Han, G., Wang, W., Li, X., Cheng, B., 2018. Identification and functional characterization of a maize phosphate transporter induced by mycorrhiza formation. Plant Cell Physiol. 59, 1683–1694. Luo, J.-S., Huang, J., Zeng, D.-L., Peng, J.-S., Zhang, G.-B., Ma, H.-L., Guan, Y., Yi, H.-Y., Fu, Y.-L., Han, B., Lin, H.-X., Qian, Q., Gong, J.-M., 2018. A defensin-like protein drives cadmium efflux and allocation in rice. Nat. Commun. 9. Phillips, J.M., Hayman, D.S., 1970. Improved procedures for clearing roots and staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessment of infection. Trans. Br. Mycol. Soc. 55 158-IN18. Rasmussen, N., Lloyd, D.C., Ratcliffe, R.G., Hansen, P.E., Jakobsen, I., 2000. 31P NMR for the study of P metabolism and translocation in arbuscular mycorrhizal fungi. Plant Soil 226, 245–253. Sharma, V., Parmar, P., Kumari, N., 2016. Differential cadmium stress tolerance in wheat genotypes under mycorrhizal association. J. Plant Nutr. 39, 2025–2036. Sydow, M., Chrzanowski, Ł., Cedergreen, N., Owsianiak, M., 2017. Limitations of experiments performed in artificially made OECD standard soils for predicting cadmium, lead and zinc toxicity towards organisms living in natural soils. J. Environ. Manag. 198, 32–40. Takahashi, R., Ishimaru, Y., Shimo, H., Ogo, Y., Senoura, T., Nishizawa, N.K., Nakanishi, H., 2012. The OsHMA2 transporter is involved in root-to-shoot translocation of Zn and Cd in rice: characterization of OsHMA2. Plant Cell Environ. 35, 1948–1957. Tan, S.-Y., Jiang, Q.-Y., Zhuo, F., Liu, H., Wang, Y.-T., Li, S.-S., Ye, Z.-H., Jing, Y.-X., 2015. Effect of inoculation with Glomus versiforme on cadmium accumulation, antioxidant activities and phytochelatins of Solanum photeinocarpum. PLoS One 10, e0132347. Thakali, S., Allen, H.E., Di Toro, D.M., Ponizovsky, A.A., Rooney, C.P., Zhao, F.-J., McGrath, S.P., Criel, P., Van Eeckhout, H., Janssen, C.R., Oorts, K., Smolders, E., 2006. Terrestrial biotic ligand model. 2. Application to Ni and Cu toxicities to plants, invertebrates, and microbes in soil. Environ. Sci. Technol. 40, 7094–7100.

9