Genome-wide sequence identification and expression analysis of N⁶-methyladenosine demethylase in sugar beet (Beta vulgaris L.) under salt stress

Materials

The salt-tolerant strain “O”68 of sugar beet was used as the experimental material in this experiment (Shi et al., 2008). The seeds were soaked in running water for 12 h, disinfected and sown onto wet sponge and cultured in the dark for 2 days. After germination, the seedlings were transferred to culture pots (43.5 × 20 × 14 cm, 10 plants per pot) containing quarter-strength Hoagland solution. The seedlings were cultivated under 16/8 light photoperiod at 24 °C/18 °C day/night temperature in a phytotron (Friocell 707, Germany). After the formation of three pairs of true leaves, 300 mM NaCl was used to replace the nutrient solution for 24 h, and other conditions were kept unchanged. The control group was treated with nutrient solution without salt treatment. After the salt stress treatment, leaves and roots were sampled. Plant samples from the same treatment were premixed and divided into 0.2 g small packages, immediately frozen in liquid nitrogen and stored at −80 °C until used for analysis.

Screening and identification of sugar beet m⁶A demethylases

The whole genome database of sugar beet has been published (http://bvseq.molgen.mpg.de/index.shtml). The encoding motif sequence of the demethylase conserved domain 2OG-Fe II-oxy (PF13532) was downloaded from the Pfam database. The e-value <1e ⁻⁵ was set on HMMER (http://www.hmmer.org/) and the beet genome-wide database was searched. The Pfam online tool was used to analyze the domains of candidate proteins, and the proteins with the conserved domain were considered to be BvALKB proteins. Multiple sequence alignment of BvALKB proteins was performed using DNAMAN7.0 and its conserved domain was identified using Weblogo (http://weblogo.berkeley.edu/).

Bioinformatic analysis of BvALKB family

ExPASY (https://web.expasy.org/protparam/) was used to analyze the physical and chemical properties of proteins, including the average molecular weight, isoelectric point and average number of amino acids (Gasteiger et al., 2003). Protein subcellular localization was predicted by CELLO (http://cello.life.nctu.edu.tw/). MapGene2Chrom (http://mg2c.iask.in/mg2c_v2.0/) was used to map the position of each gene on a chromosome. MEME (http://meme-suite.org/tools/meme) was used to predict protein motifs (Bailey et al., 2006), and the number of motifs was set to 20, with other parameters for tacit recognition. Gene intron and exon structures were analyzed in Splign (https://www.ncbi.nlm.nih.gov/sutils/splign/splign.cgi?textpage=online&level=form). A phylogenetic tree (1000 replicates) was constructed by neighbor-joining method using MEGA7 for protein sequence progression and multi-sequence alignment between Arabidopsis and sugar beet (Kumar, Stecher & Tamura, 2016).

Expression analysis of BvALKB genes and gene cloning

All samples were ground in liquid nitrogen. Total RNA was extracted using TRIzol reagent and the concentration of RNA was determined using a MicroDrop spectrophotometer. Total RNA was reverse-transcribed into cDNA by using PrimeScript ™ II first strand cDNA Synthesis Kit (TaKaRa, Japan). In order to detect the gene expression level, qRT-PCR was performed using the CFX96 real-time system and the iTaq™ Universal SYBR Green Supermix Kit (BIO-RAD, USA). The primers were designed using Primer 5 and their sequences are listed in Table 1. Referring to our previous studies (Li et al., 2020), UBQ5 and PP2A were used as internal controls for leaves, and 25S rRNA and PP2A were used as internal controls for roots. All experiments were repeated at least three times. Data analysis was performed by 2^−△△Ct method. The relative expression of each gene was presented as mean ± standard deviation. The PCR primers were designed for BvALKBH10B cloning as follows:

F: 5′-GGAATTCATGTCGCCGGCGGCGGGACCATTGT-3′,

R: 5′-GGGATCCTCACATTATCCTTCCTTCCACACCTGGGTCAGACATGGT-3′.

Data treatment and statistical analysis

For the data of qPCR analysis, the mean and SD were calculated from three repeats of each treatment, and the differences were analyzed by SPSS 20.0 using independent-samples t-test (p < 0.05).

Identification of sugar beet m⁶A demethylases

The seed sequence of the conserved domain (PF13532) was downloaded from Pfam and used as search bait in the beet genome database by HMMER. A total of 10 homologous proteins were identified, and they were named BvALKBH1B through 10B (Table 2). The e-value of all other proteins was less than 1e⁻⁵ except BvALKBH10B, where it was 0.016. The ten proteins were blastp-aligned with the NCBI database, as shown in Table 2. Information from NCBI suggests that BvALKBH2B, BvALKBH3B, BvALKBH8B and BvALKBH10B have not been described before and belonged to new ALKB family members while the other proteins have been confirmed to belong to the ALKB family.

The domains of the ten candidate proteins were analyzed by Pfam (Fig. 1). All proteins have 2OG-Fe II-Oxy domain except for BvALKBH10B, indicating that these domains are highly conserved. In terms of domain distribution, the domains of BvALKBH7B were at the N-terminus, and the domains of BvALKBH1B, BvALKBH2B, BvALKBH3B, BvALKBH4B, BvALKBH5B, and BvALKBH9B were all at the C-termini. The RRM domain of BvALKBH5B was related to mRNA and rRNA processing, RNA output and RNA stability by query. However, due to low sequence similarity, the e-value of BvALKBH10B in Pfam database comparison was 0.023, and it has a high possibility of possessing the 2OG-Fe II-Oxy domain, so it is regarded as a member of this family for subsequent analysis.

The alignment results of DNAMAN7.0 showed certain homology but low conservation in the domain sequences (Fig. 1). The homology was very high at the sites 162, 212, 215, 222, 255, 259, etc., which might be related to the function of the domain and amino acids at these specific locations.

Analysis of physicochemical properties of BvALKB proteins

Physical and chemical properties analysis showed that the average length of the coding region of the ten genes was 1,260 bp (and for all numbers of 1,000), the average number of amino acids per protein was 416 (260–584), the average molecular weight was 46.41 kDa (28.91–64.97 kDa), and the average isoelectric point was 7.12 (5.11–9.02) (Table 3).

Chromosomal localization of genes

Sugar beet is diploid with 2n =18 chromosomes. Chromosome localization analysis showed that BvALKBH6B was located on chromosome 3, BvALKBH7B was located on chromosome 4, BvALKBH5B was located on chromosome 5, BvALKBH1B and BvALKBH8B were located on chromosome 6, BvALKBH2B, BvALKBH3B and BvALKBH9B were located on chromosome 7, and BvALKBH4B was located on chromosome 8 (Fig. 1). Analysis of BvALKBH10B showed that it was located on chromosome 7, but lacked localization information, which may be due to gaps in genome assembly. These results indicate that the chromosome distribution of BvALKB genes is relatively scattered, without cluster distribution and lack of chromosome preference.

Phylogenetic relationships and gene structure analysis of BvALKB

Multiple sequence alignment was performed on 14 ALKB family proteins of Arabidopsis and ten proteins of sugar beet using MEGA7, and the alignment diagram of protein local domain was analyzed (Fig. 2). The results showed that the differences in amino acid sequences of ALKB proteins led to low domain homology and therefore belonged to different subclasses, but remained highly conserved at individual sites.

Then a phylogenetic tree (1000 replicates) was constructed using neighbor-joining method to observe the evolutionary relationships between Arabidopsis and sugar beet proteins (Fig. 3). Most of the bootstrap values were greater than 70, indicating high reliability. The proteins were divided into five categories: Class I (AtALKBH9-like) included BvALKBH6B and BvALKBH8B, which are similar to AtALKBH9; Class II (AtALKBH10-like) only contained BvALKBH10B, which is similar to AtALKBH10; Class III (AtALKBH2-like) contained only one BvALKB protein; Class IV (AtALKBH6/8-like) consisted of BvALKBH5B, BvALKBH7B and BvALKBH9B; Three members were assigned to Class V (AtALKBH1-like), including BvALKBH1B, BvALKBH2B and BvALKBH3B (Fig. 3). AtALKBH9B and AtALKBH10B in the first two classes were confirmed to be m⁶A demethylases, so BvALKBH6B, BvALKBH8B and BvALKBH10B are also likely to have demethylase function, however, future experimental validation is required to confirm the function of these three ALKBH proteins.

The gene structure analysis revealed that genes within the same group showed similar intron and exon organization. The BvALKBH6B and BvALKBH8B of Class I had six exons, whereas BvALKBH2B and BvALKBH3B of Class V had four exons, based on their sequence similarity. The BvALKBH5B and BvALKBH7B in Class IV were similar in structure, although the number of exons was different. The structures of BvALKBH1B and BvALKBH9B were different to varying degrees from those of the above genes (Fig. 3).

Motif analysis and subcellular localization prediction of BvALKB proteins

The motif analysis of BvALKB proteins is shown in Fig. 4. In general, the ten proteins (except BvALKBH10B) had the motifs 1, 2, 4, and 8, which are probably important components of the 2OG-Fe II-Oxy domain. Proteins belonging to the same group had similar motif composition. BvALKBH6B, BvALKBH8B and BvALKBH10B homologous to AtALKBH9B/10B differed from other proteins in motif composition because they had closely connected motif 3 and motif 6, which may be related to the demethylase function.

The scores of different locations of CELLO predicted proteins showed that most of the proteins were located in the nucleus and mainly performed function of demethylation in the nucleus (Table 3). The proteins BvALKBH10B and BvALKBH7B were both located to the cytoplasm. However, BvALKBH7B was additionally predicted to be in the extracellular region presumably performing other extra-nuclear functions.

Quantitative analysis of BvALKB genes in sugar beet under salt stress

N⁶-methyladenosine plays an important role in response to abiotic stresses. In order to understand the changes in the potential m⁶A demethylase genes in sugar beet under salt stress, we compared the expression levels of the genes under normal and salt stress conditions. The phenotypic changes of sugar beet grown to the stage of three pairs of true leaves were observed under 300 mM salt stress, and the expression of each gene was analyzed by qRT-PCR.

In leaves, all genes were up-regulated or down-regulated to varying degrees except BvALKBH1B. BvALKBH2B, BvALKBH4B and BvALKBH10B were up-regulated and in particular BvALKBH10B was highly up-regulated (Fig. 5). BvALKBH3B, BvALKBH5B, BvALKBH6B, BvALKBH7B, BvALKBH8B and BvALKBH9B were down-regulated, and BvALKBH9B showed a significant and striking degree of expression down-regulation. In roots, BvALKBH1B, BvALKBH3B, BvALKBH6B, BvALKBH8B and BvALKBH9B were up-regulated, while the other five genes were down-regulated. BvALKBH2B, BvALKBH4B and BvALKBH5B were significantly down-regulated (Fig. 5). Under salt stress, the expression levels of BvALKBH7B, BvALKBH9B and BvALKBH10B in leaves were significantly affected, whereas that of BvALKBH2B, BvALKBH4B and BvALKBH5B in roots was only mildly modified by stress treatment. These results indicate that these genes had tissue-specific activity in regulating the response to salt stress in sugar beet.

Since BvALKBH10B and AtALKBH10B were highly homologous and the expression level of BvALKBH10B changed significantly under salt stress, we cloned and sequenced the BvALKBH10B gene from Beet “O” 68. The sequencing results were submitted to the Genbank database (MZ358117), which was consistent with the whole genome database of sugar beet.