Transcriptome Analysis of Genes Involved in Aluminum Stress Responses in Peanut (Arachis Hypogaea L.)

Aluminum (Al) contamination inhibits plant growth and development, however, mechanisms involved in Al stress tolerance in peanut (Arachis hypogaea L.) were rarely studied. The present study was comprised of four Al levels i.e., 0, 1.25, 2.5 and 5 mmol l −1 AlCl 3 .18H 2 O regarded as Al0, Al1, Al2, and Al3. The respective concentrations were added in Hoagland nutrient solution and replaced every three days. Results revealed that seeding length low Al concentration (Al1) treatment had no noticeable effect on seeding lenght, while higher Al concentration (Al2 and Al3) treatment signicantly inhibited seeding lenght. The differentially expressed genes (DEGs) of plant hormone metabolism pathway were signicantly enriched whereas the contents of salicylic acid (SA) and abscisic acid (ABA) were up-regulated, and jasmonic acid (JA) were down-regulated to different levels. Moreover, transcription factors (TFs) and ALMT9 and FRDL1 genes were up-regulated at higher Al concentration and down-regulated at the lowest Al concentration (Al1).

Plants can deal with the higher Al ions by external exclusion mechanism and internal compartmentalization [5]. The roots are the foremost organ that come into contact with the Al ions in growing medium [6], and high-concentration Al toxicity could cause noticeable morphological and structural changes in peanut root tip cells, manifested by inhibited root elongation and programmed cell death [7]. The aging-related AhSAG gene cloned in peanut was related to Al-induced cell death [8]. Moreover, Al could produce organic acids in plants, such as malic acid, citric acid, and oxalic acid which could combine with Al 3+ to form a non-toxic complex and thus prevents Al 3+ from entering the root tip cells [9]. There were are two signi cant types of transporter families related to the secretion of organic acids, which were multidrug and toxic compound extrusion (MATE) and aluminum-activated malate transporter (ALMT). The AhFRDL1 encoding MATE family proteins had been cloned in peanut. The expression of the peanut AhFRDL1 gene was up-regulated and increased the secretion of citric acid in the root tip, which affect the tolerance ability of the root system to Al stress [10]. In addition to the external exclusion mechanism, isolating or detoxifying the Al entering the plant is the primary mechanism of internal tolerance mechanism. Moreover, the cell wall could adsorb Al 3+ to improve aluminum resistance in plants [11] whereas the higher pectin content in cell wall induced pectin methylesterase (PME) family genes and enhanced PME activity, thereby preventing Al 3+ from entering the plant cells under Al stress [12]. In addition, transcription factors (TFs) STOP1, ART1 and WRKY were involved in the mechanism of plant resistance to Al toxicity [13,14,15], however, TFs involved in Al stress resistance in peanut were rarely investigated.
Plant hormones also play an important role in Al tolerance, for example, exogenous salicylic acid (SA) treatment could reverse the Al stress-induced pectin accumulation and PMD reduction in Panax notoginseng roots, thereby reducing the Al content of root pectin. Moreover, Al stress activated the endogenous SA content and NO signaling pathway of Panax notoginseng [16]. Exogenous indole acetic acid (IAA) could reduce the Al concentration of soybean roots and stimulate Al-induced citrate secretion and plasma membrane Hþ-ATPase activity under aluminum stress [17]. The jasmonic acid (JA) receptor COI1 mediated JA signal was involved in the regulation of ALMT1-mediated malic acid secretion, thereby expelling Al from the cell [18]. Exogenous abscisic acid (ABA) could enhance al-induced citrate secretion in soybean roots [19]. Overall, negative impacts of Al toxicity on the growth and yield of crop plants [20], however, as per our knowledge, molecular basis of Al stress tolerance in peanut were rarely studied. The present study used transcriptome analysis to explore the changes in the transcriptional regulation mechanism of peanut under Al stress and the effects of plant hormones on this process. The preliminary analysis of the molecular mechanism of peanut tolerance to Al stress provides a theoretical rationale for cultivating Al tolerant peanut cultivars in Al contaminated lands.

Results
Effect of different concentrations of Al on peanut seedling length Compared with Al0, the seedling lenght of Guihua58 was not signi cantly affected under Al1 treatment, whilst decreased by 37.66 % and 43.77 % under Al2 and Al3, respectively.

Sequencing statistics
Sequencing data of all samples has been quality evaluated. The original bases of the sequencing data of each sample were between 5777435056 and 7275220472, the GC content of the sequence was between 45.43 % and 46.01 %, the Q20 value was above 97.39 % and the Q30 value was above 92.77 % ( Table 1).

Analysis of sample relationship
The clean reads were compared with the reference genome, and the number of effective sequences that could be aligned on the reference genome for each sample sequence was between 36691674 and 45441812, and the contrast ratio was above 94.04 % and 8.46 % of multiple mapped reads on the reference genome, the unique mapped reads on the reference genome reached more than 84.8 % ( Table   2). The data showed that the sequencing quality of the transcriptome data was good, the experimental process was pollution-free, and the selected reference genome information could meet the needs of subsequent analysis.

Differentially expressed genes (DEGs)
Through the comparison between samples, the results showed that Al0-vs. The expression levels of ALMT9 and FRDL1 genes were measured in peanut leaves under different Al concentrations treatment. Compared with Al0, the gene expression of ALMT9 and FRDL1 were signi cantly up-regulated under Al2 and Al3 treatments which were up-regulated by 1.3-1.5 times and 1.7-8.7 times, respectively. Compared with Al0, the ALMT9 gene expression was signi cantly reduced by 77 % under Al1 treatment (Figure 5a). Compared with Al0, the expression of the FRDL1 gene was not signi cantly different under the Al1 treatment (Figure 5b).

Effect of different Al concentrations on the regulation of TFs expression
The key TFs associated with Al treatments in peanut were exhibited in Figure 6. Among them, MYB (4), bHLH (9), NAC (6) and AP2 (2) were differently expressed in different treatments (Figure 6a). Ten genes from the differentially expressed transcription factor genes were randomly selected by qRT-PCR analysis ( Figure 6b). The randomly selected gene expression were consistent with the results calculated by the FPKM value obtained by sequencing, indicating that the transcriptome data was reliable.

Discussion
Nowadays, with the development of transcriptome sequencing technology, transcriptome analysis is becoming an e cient and reliable research method to deeply understand plants' gene expression in a speci c growth environment. Transcriptome analysis can analyze the biological processes and molecular mechanisms of plants in response to environmental stress. This study found that the plant hormone metabolic pathway was more obvious in KEGG pathway shared by the four sample groups, indicating that the plant hormonal pathway was an essential metabolic pathway in response to Al tolerance in peanut ( Figure 3). Previously, Tian and Li [21] reported that Al stress inhibited the root elongation in watermelon (Citrullus lanatus) whereas exogenous SA application improved the root growth in growing watermelon seedlings under Al toxic conditions. In addition, the IAA was also involved in regulating the growth and development of plant roots under Al stress [22]. Wang [23]found that the black soybean roots were inhibited by adding the IAA transport inhibitor TIBA under Al stress, indicating that IAA participated in the resistance to Al stress. Expression of JA receptor COI1 in Arabidopsis root tips was up-regulated under Al stress whereas ethylene had also been involved in regulating JA signal induced by Al in root tips, indicating that JA regulated and resisted Al stress through interaction with ethylene [24]. Al stress increased the accumulation of endogenous ABA in soybean roots and leaves and accelerated the transport of ABA, it showed that ABA might act as an Al stress response signal to regulate the Al resistance of soybeans [25]. Studies in buckwheat (Fagopyrum Mill) had found that high Al concentrations could promote Al detoxi cation by directly activating the ABA-like gene ALS3 or stimulating the increase of ABA levels [26]. This study found that with the increase of Al concentrations, the SA and ABA contents in peanut leaves were signi cantly up-regulated, and the content of JA was signi cantly down-regulated, indicating that peanut could resist Al stress by changing the content of SA, ABA and JA. There was no signi cant difference in IAA content, which may be due to the fact that IAA in leaves did not respond to Al toxicity ( Figure 4).
Sasaki [27] isolated TaALMT1 from wheat with different aluminum tolerance for the rst time. The homologous gene AtALMT1 isolated from Arabidopsis thaliana was one of the critical genes that regulated the aluminum tolerance mechanism in Arabidopsis [28]. At present, the AhFRDL1 gene had been cloned from peanut, which had been con rmed to be a citrate transporter gene. The expression of peanut AhFRDL1 gene was up-regulated and the secretion of citric acid from the root tip increased under Al stress [10]. The WRKY is a large family of TFs, involved in the transduction of plant defense signaling mechanism [29]. Schluttenhofe [30]found that 13 AhWRKY genes may be involved in SA and JA signaling pathways in cultivated peanut. In this study, the expression of ALMT9, FRDL1 and WRKY genes were upregulated under high concentration of Al i.e., Al2 and Al3, indicating that peanut leaves also increased the expression of organic acid secretion transporter family (ALMT9 and FRDL1) and WRKY genes ( Figure 5 and Figure S1). In Arabidopsis, the C2H2-type transcription factor STOP1 induced the AtALMT1 gene to secrete malate to chelate and detoxify aluminum [31,32,33]. This was consistent with the results of previous studies. However, the expression of ALMT9, FRDL1 and WRKY genes were all down-regulated under low concentration of Al (Al1) ( Figure 5 and Figure S1), which may be due to low concentration of Al (Al1) promoted the absorption of other elements by the root system, thereby maintaining the growth of seeding lenght, making it offset part of the impact of Al poisoning has been eliminated [34], however further studies in this regard are still needed.

Conclusion
Al toxicity suppressed the seedling growth and root system of peanut whilst the effects were more severe at higher concentrations. Higher Al concentrations up-regulated the expression of transcription factors (TFs), and ALMT9 and FRDL1 genes to resist the stress of high Al concentrations whereas transcriptome analysis revealed that Al stress tolerance is closely related to endogenous hormone contents i.e., salicylic acid (SA), abscisic acid (ABA), and jasmonic acid (JA), transcription factors (TFs) and the expression of ALMT9 and FRDL1 genes.

Plant materials and treatment conditions
The hydroponic experiment was carried out in the Guangzhou Key Laboratory for Research and Development of Crop Germplasm Resources, Zhongkai University of Agriculture and Engineering, Guangzhou, China (23104 N, 113281 E). The peanut cultivar 'Guihua58' was provided by Guangxi Academy of Agricultural Sciences, Nanning, China. Homogenous seeds were surface sterilized and germinated in a petri dish with wet lter paper with 20 seeds per petri dish. After three days, the seeds were transferred to a plastic culture bowl and continued to cultivate with Hoagland nutrient solution. When the seedlings grow to one leaf and one heart (one week of growth) at 27℃, the photoperiod was 12h day/12h night. The seeding lenght was measured every 24 hours the following four levels of Al were employed i.e., i.e., 0, 1.25, 2.5 and 5 mmol l -1 AlCl 3 .18H 2 O regarded as Al0, Al1, Al2, and Al3, respectively.
The nutrient solution was replaced once every three days.

Raw sequencing data
After RNA extraction, the magnetic beads with Oligo (dT) were used to enrich eukaryotic mRNA. A fragmentation buffer was added to interrupt mRNA randomly. mRNA was used as a template to synthesize the rst cDNA strand with six-base random hexamers, then added buffer, dNTPs, RNase H and DNA polymerase I to synthesize the second cDNA strand and puri ed the cDNA with AMPure XP beads.
The puri ed double-stranded cDNA was repaired, A-tailed and connected to the sequencing adapter, and then AMPure XP beads were used for fragment size selection. Finally, the cDNA library was obtained by PCR enrichment. Qubit2.0 and Agilent 2100 were used to detect the library's concentration and insert size, and the library's effective concentration was accurately quanti ed using the Q-PCR method to ensure the quality of the library. High-throughput sequencing was performed with NovaSeq 6000, and the sequencing read length was PE150. Raw sequencing data have been uploaded in the NCBI Gene Expression Omnibus under the accession number PRJNA754251 (http://www.ncbi.nlm.nih.gov/geo).

Enrichment Analysis
The raw data analysis was performed using BMKCloud (www.biocloud.net). The differentially expressed genes (DEGs) between the comparison groups were obtained based on certain standardized processing and screening conditions. The default parameters were FDR=0.1 and FC=2. The KEGG database was used for functional annotation, classi cation statistics and metabolic pathway analysis of DEGs in the comparison group.
Determination of JA, ABA, SA and IAA content The cartridge was activated with 4 mL of methanol and 2 mL of 0.1 M aqueous ammonia solution. Fresh sample (100 mg) were homogenized with 1 mL of extraction solution (acetonitrile: water=1:1, containing sodium diethyldithiocarbamate)-ice bath for 4 h, 4 ℃, 12000 rpm, 10 min. The supernatant was concentrated in vacuo, 0.1 M aqueous ammonia solution was added to the volume to 2 mL, and then passed through the MAX cartridge. The MAX cartridge was washed with 2 mL of 0.1 M ammonia solution and 2 mL of 0.1 M ammonia solution with 60% methanol and nally added 0.2 ml methanol to dissolve it. The chromatographic system used an ultra-high performance liquid system (Vanquish, Thermo, USA), and the mass spectrometry system uses a Q executive high-resolution mass spectrometry detection system (Vanquish, Thermo, USA). A liquid chromatography column was used Waters HSS T3 (50*2.1 mm, 1.8 μm). The sample volume was 2 μL. The column temperature was 40°C.
Real-time quantitative RT-PCR RNA extraction was performed using RNAprep Pure Plant Kit (TianGen Biotech, Beijing, China). The RNA quality detection was performed using Micro-Spectrophotometer (Allsheng, Nano-300, Hangzhou, China). cDNA synthesis was used PrimeScriptRT reagent Kit with gDNA Eraser kit (TaKaRa, Beijing, China). 2×SYBR Green qPCR Mixture kit (Hlingene Corporation, Shanghai, China), Option Real-Time PCR System (Bio-Rad, CFX96, California, USA) instrument was used for real-time uorescent quantitative PCR. cDNA was used as a template, three biological replicates were set for each sample, actin gene was used as an internal reference gene, and gene-speci c primers were synthesized by Sangon Biotech, Shanghai, China.

Experimental design and statistical analyses
The experimental treatments were arranged in completely randomized design (CRD) in triplicate. Data were compiled using Microsoft Excel 2010 (Microsoft, Chicago, USA). SPSS Statistics 20.0 (IBM, Chicago, USA) was used for one-way analysis of variance, and the Tukey's test at the 5% signi cance level was used to separate treatment means. The R Programming Language was used for mapping.

Declarations
Ethics approval and consent to participate Not applicable.