Genome-wide identification and characterization of FAD family genes in barley

Identification and characterization of FAD family members in the Hordeum vulgare genome

To effectively identify the FAD family members of Hordeum vulgare cv. Morex genome, candidate protein sequences corresponding to the structural domains of FA_desaturatase (PF00487), FA_desatase 2 (PF03405), and TMEM189 (PF10520) were downloaded from the Pfam protein family database (http://Pfam.xfam.org/). Using the SMART database (http://SMART.embl-heidelberg.de/), Markov model (HMM) files were utilized to confirm each putative HvFADs. TBtools software was employed to analyze the physicochemical properties, transmembrane structural domains, and signaling peptides.

Chromosomal distribution of HvFADs

The locations of HvFADs on chromosomes were extracted from the genome annotation files. The genes densities of the whole chromosome were determined and visualized using TBtools software (Chen et al., 2020). Only HvFADs anchored on chromosomes were displayed, for analysis of conserved motifs and domains. Conserved motif analysis of the FAD protein sequences was performed using the classical motif discovery model of MEME-suite 5.3.3 (Bailey et al., 2006). The parameter of the optimum motif was set from 6 to 200, and the maximum number of motifs was set to 10. Conserved domain analysis was performed using TBtools.

Collinearity analysis

To investigate the mechanisms underlying the evolution of HvFADs, MCScanX (Wang et al., 2012) was used to analyze tandem, proximal, and dispersed duplications. Collinearity analysis of HvFAD was conducted locally using the TBtools that map the genes to chromosomal locations (Xie et al., 2018). We identified collinearity resulting from segmental duplication events, as well as tandem duplicates arising from gene duplication events. The MCScanX algorithm inferred that one sequence emerged as a result of the duplication of its counterpart, indicating a gene duplication event.

Phylogenetic analysis

To construct a phylogenetic tree, the protein sequences of T. aestivum L. (68), G. max (30), A. thaliana (27), and O. sativa L (18) were analyzed using the neighbor-joining method. The coding sequences of the full-length genes were aligned using fast Fourier transform (MAFFT) (Kazutaka, Rozewicki & Yamada, 2019). The phylogenetic trees were constructed using MEGA7 software (Arizona State University, Tempe, AZ, USA) with 1,000 bootstrap replicates. The FAD protein sequences of Arabidopsis, Glycine max and rice were downloaded from the Arabidopsis (Philippe et al., 2012), Glycine max (https://www.soybase.org/) and rice (Kawahara et al., 2013) databases for constructing the phylogenetic tree, and the tree was visualized using the iTOL online website (https://itol.embl.de/) (Letunic & Bork, 2019).

Analysis of cis-acting elements in HvFADs promoters

The 1,500 bp sequences upstream the initiation codon (ATG) of each HvFADs were extracted from the Hordeum vulgare genome database using TBtools and subjected to cis-acting elements and transcription factor binding sites prediction analysis using PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) and PlantTFDB (http://plantregmap.gao-lab.org/binding_site_prediction.php), respectively (Chen et al., 2020).

Protein-protein interaction network

To predict the interactions between HvFAD proteins, FAD protein sequences were submitted to STRING V10 (Szklarczyk et al., 2015). The protein-protein interaction (PPI) networks were then visualized using Cytoscape V3.7.2 (Shannon et al., 2003).

Plant material culture, saline alkali treatment, and quantitative real-time polymerase chain reaction analysis

H. vulgare cv. Morex were cultivated in a light incubator until two leaves and one heart stage. The plants were then exposed to 0, 3, 12 and 24 h of saline alkali treatment. The leaves were collected for RNA extraction. Similarly, roots, stems, leaves and tiller tissues of healthy barley plants were collected at the two leaves and one heart stage for RNA extraction. The quality and concentration of RNA were measured using a NanoDrop 2000 UV spectrophotometer. First-strand cDNA was synthesized using the Novizen Reverse Transcription Kit according to the manufacturer’s protocol. The cDNA was diluted to 100 ng/μl and used as a template for quantitative real-time polymerase chain reaction (qRT-PCR) reaction. Gene-specific primers for RT-PCR and qRT-PCR were designed using NCBI (Table S1). Five genes belonging to four FAD subfamilies were selected to analyze the expression in different tissues, with amplified size ranging from 190 to 230 bp. The PCR program was performed as follows: 94 °C for 2 min; 39 cycles of 95 °C for 5 s, 60 °C for 30 s; 95 °C for 5 s, 65 C for 5 s, and 95 °C for 5 s.

Gene Expression Analysis^™ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) and Taq Pro Universal SYBR qPCR Master Mix (Q712-03) Master Kit (Novozymes, Bagsværd, Denmark) were used for gene expression analysis by qRT-PCR experiments using FX96 Touch. Relative fold differences were calculated based on the comparative cycle threshold (2-Δct) using Tublin (F: AGTGTCCTGTCCACCCACTC; R: AGCATGAAGTGGATCCTTGG) as the reference gene. The PCR mixture (20 μl) included 10 μl of 2× Taq Pro Universal SYBR qPCR Master Mix (Novozymes, Bagsværd, Denmark), 10 μm of each primer, 100 ng cDNA template, and nuclease-free water. The PCR program was performed as follows: 94 °C for 2 min; 39 cycles of 95 °C for 5 s, 60 °C for 30 s; 95 °C for 5 s, 65 °C for 5 s, and 95 °C for 5 s.

Transient expression of HvFAD14

Subcellular localization of HvFAD14 in barley protoplasts was analyzed using a transient transformation system as follows. Brifely, a full-length open reading frame (ORF) fragment (1,156 bp) of HvFAD14 was cloned in an empty GFP vector. The resulting construct (HvFAD14-GFP) was then transformed into Fast-T1 chemoreceptor cells. After that, the cells were harvested by shaking the bacteria and centrifugation for bulk plasmid extraction. Protoplasts were extracted from barley yellowing seedling leaves using cellulase and dissociative enzyme solutions, and the reconstructed GFP plasmid was transformed into protoplasts using a 40% PEG solution. After 16 h of dark expression, the expression of HvFAD14 were observed using a Leica fluorescence focusing microscope.

Identification and analysis of HvFADs in barley

We identified a total of 24 HvFADs from barley genome. These genes were named based on their chromosomal locations and phylogenetic relationship (Fig. 1, Table 1), ranging from HvFAD1 to HvFAD24. Systematic analysis of their physical and chemical properties showed that the CDS length of these 24 HvFAD genes ranged from 987 to 1,329 bp, while the lengths of the deduced amino acid sequences ranged from 329 to 469 amino acids. Additionally, the predicted molecular weights (MW) ranged from 37,724.61 Da (HvFAD9) to 52,548.61 Da (HvFAD18), and the theoretical isoelectric points (pI) ranged from 6.04 to 9.37 (Table 1).

Phylogenetic relationship analysis of HvFADs

In order to unravel the phylogenetic relationship among HvFADs, we constructed a phylogenetic tree using FADs from Triticum aestivum with 68 members, Glycine max with 30 members, Arabidopsis thaliana with 27 members and Oryza sativa with 18 members. As shown in Fig. 2, we observed that these FADs were categorized into five distinct subfamilies: FAB2, FAD4, FAD3/FAD7/FAD8, FAD2/FAD6, and DES/SLD. The gene number among these FADs of the five species is different from each other, the homology of genes in the same subfamily is high. Among these subfamilies, the HvFAB2 subfamily consists of 14 genes such as FAD1, FAD2, FAD3, FAD4, FAD5, FAD8, FAD10, FAD11, FAD12, FAD13, FAD16, FAD19, FAD20 and FAD24. The HvFAD3/FAD7/FAD8 subfamily comprises FAD6, FAD14, FAD15 and FAD17 four genes. The HvFAD2/FAD6 subfamily includes FAD7, FAD2, FAD22 and FAD23. The DES/SLD subfamily has FAD9 and FAD18 in HvFADs. Furthermore, the FAD4 subfamily does not contain any HvFADs (Fig. 2). So HvFADs have four subfamilies, which is different from other species FAD gene family.

Gene structures and conserved motifs of HvFADs

To investigate the distribution of these HvFADs motifs, we identified 10 motifs in barley. Conserved motif analysis using MEME software revealed (Fig. 3A) that HvFAD1, HvFAD3, HvFAD4, HvFAD5, HvFAD24, HvFAD12, HvFAD19, HvFAD11, HvFAD13, HvFAD8, HvFAD16, and HvFAD20 in the FAB2 subfamily have seven identical motifs. Additionally, HvFAD2 and HvFAD1 have six identical motifs. HvFAD10 and HvFAD1 have five identical motifs. HvFAD9, HvFAD18, and HvFAD7 in the DES/SLD subfamily have two identical motifs. Lastly, HvFAD21, HvFAD22, HvFAD23, HvFAD6, HvFAD15, HvFAD14, and HvFAD17 in the FAD2/FAD6 subfamily and FAD3/FAD7/FAD8 subfamily have three identical motifs. In addition, it was found that the conserved motif composition of each subfamily was generally similar. Moreover, the conserved motif composition of FAD3/FAD7/FAD8, FAD2/FAD6 and DES/SLD subfamily is obviously different from that of FAB2 subfamily.

We further analyzed the exon-intron structure of each HvFAD gene (Fig. 3B). The number of introns ranges from 1 to 9, while the number of exons ranges from 1 to 10. Among them, the HvFAD7 has the highest number of introns and exons, with nine introns and 10 exons. Except for HvFAD5, HvFAD10 and HvFAD18, FAB2 and DES/SLD subfamily possess less introns. HvFAD22 and HvFAD23 have no intron. Other members in FAD3/FAD7/FAD8 and FAD2/FAD6 subfamilies have relatively more introns.

Chromosomal location and colinearity analysis of HvFADs

The HvFAD genes randomly and unevenly distributed on all barley chromosomes except chromosome 1 (Chr1H). Chr2H contains the most HvFADs (nine members), followed by 5 HvFADs on Chr5H, 4 HvFADs on Chr3H, 3 HvFADs on Chr6H, 2 HvFADs on Chr4H, and 1 HvFAD gene on Chr7H (Fig. 1). Collinearity analysis was performed to further investigate gene duplication events in the HvFADs family. Our data revealed a single pair of segmentally duplicated genes (Fig. 4). Segmental duplication events may be a main cause for the HvFAD gene family expansion.

Analysis of cis-acting elements in HvFADs promoters

A comprehensive investigation was conducted to clarify the cis-regulatory elements found in the promoter domains of HvFADs. The identified cis-acting components within these promoters were classified into four main groups: light-responsive elements, hormone-responsive elements, stress-responsive elements, and elements essential for plant growth and maturation (Fig. 5). Light-responsive elements were present in all HvFAD promoters, with the Sp1 element being the most prominent with 148 occurrences. However, it is important to note that the promoters of HvFAD11, HvFAD23, and HvFAD24 did not contain this particular element.

For hormone-responsive elements, the methyl jasmonate (MeJA) response element has emerged as the predominant one with 118 instances, closely followed by the abscisic acid (ABA) response element with 113 instances. All HvFAD promoters, except HvFAD8, contain ABA response motifs (specifically ABRE and DRE1). The cis-regulatory motifs associated with the MeJA response, including the TGACG and CGTCA patterns, are present in the promoters of 20 HvFADs. HvFAD4, HvFAD6, HvFAD13 and HvFAD12 possess promoters that feature growth hormone, ethylene, SA and gibberellin response motifs, respectively. Moreover, all HvFAD promoters have at least two hormone-responsive motifs. Notably, HvFAD4 and HvFAD5 promoters were characterized by the inclusion of all the previously mentioned hormone-responsive motifs.

In regard to stress-responsive motifs, the drought-responsive elements, specifically MYC, as-1 and MBS were found to be present with a total count of 167. However, the MYC element was not found in the promoters of HvFAD11 and HvFAD16, but was present in the remaining 22 HvFAD promoters. The as-1 motif had a wide distribution, only being absent from the promoters of HvFAD9, HvFAD13, HvFAD17 and HvFAD24. The MBS elements were found in a more limited number, featured in only 16 HvFAD promoters. Additionally, defense- and stress-responsive motifs, including STRE, CARE and TCrich repeat sequences, were highly prevalent, totaling 77. STRE patterns were found in 18 promoters, CARE in a single promoter, and TCrich repeat sequences in six promoters. Elements related to drought, trauma, pathogens (WRE3, box, and WUN motifs), defense and stress, temperature fluctuations, dehydration (DRE core), anaerobic induction (ARE) and hypoxia-specific induction (GC motif) were distributed among 24, 21, 19, 11, 7, 16 and 9 HvFADs, respectively, as shown in Fig. 5.

Analysis of protein-protein interaction network of HvFADs

As shown in Fig. 6, members of HvFADs interact with each other. The highest efficiency of interaction was found between HvFAD7 and HvFAD17. The possible interactions among HvFADs may provide the information for the research on their biological functions (Fig. 6).

HvFAD expression under different stresses by qRT-PCR

HvFADs have been reported to function under various abiotic and biotic stresses (Saini & Kumar, 2019). In this study, qRT-PCR was utilized to detect changes in the expression of HvFADs under saline and alkaline stress conditions. According to the transcriptomic data of barley treated with salt and alkali, six genes with higher expression levels and significant changes in expression levels under alkali and salt treatment were selected for qRT-PCR analysis: three genes from the FAB2 subfamily (FAD8, FAD11, FAD13), two genes from the FAD3/FAD7/FAD8 subfamily (FAD14, FAD15), and one gene from the FAD2/FAD6 subfamily (FAD21). As shown in Fig. 7, the expression levels of HvFAD8, HvFAD11, HvFAD13, and HvFAD21 were significantly up-regulated under salt treatment, reaching between 1.3 and seven times, respectively. Among them, the expression of HvFAD11 changed the most, up to seven times. In addition, the expression levels of HvFAD14 and HvFAD15 were significantly up-regulated at 3 h of salt treatment, and the expression level of HvFAD15 was up-regulated to about 65 times at 3 h of salt treatment, but gradually decreased with the increase of salt treatment time. The results showed that these genes responded to salt stress (Fig. 7).

As shown in Fig. 8, the expression levels of HvFAD13 and HvFAD14 were significantly up-regulated with the increase of alkali treatment, reaching about nine to 14 times, respectively. The expression levels of HvFAD8 and HvFAD21 had no significant change after alkali treatment for 3 h, but were significantly down-regulated after alkali treatment for 12 h. The expression levels of HvFAD21 were significantly up-regulated after alkali treatment for 24 h. The expression of HvFAD11 was significantly down-regulated to about 0.5 times under alkaline treatment. The expression of HvFAD15 was significantly up-regulated at 3 h with alkali treatment, reaching about 50 times, but significantly down-regulated at 12 and 24 h with alkali treatment. Finally, after 24 h alkali treatment, the expression levels of HvFAD13, HvFAD14 and HvFAD21 were significantly up-regulated, while the expression levels of HvFAD11 were significantly down-regulated. The results showed that these genes also responded to alkali stress (Fig. 8).

Localization analysis of HvFAD14 protein

In order to investigate the subcellular location of HvFADs, HvFAD14 with high expression in barley FAD gene family was cloned. The localization of HvFAD14 protein was distributed in the endoplasmic reticulum compared to the control (Fig. 9).