Identification and annotation of LBD gene family in switchgrass
To evaluate the LBD gene family in switchgrass, 43 annotated AtLBD proteins were chosen to screen the switchgrass genome database. SMART and NCBI conserved domain search was used to check the domains of the LBD sequences, and sequences without a typical LOB domain were excluded. Finally, the confirmed LBD genes in switchgrass amounted to 69 (Table 1). Gene characteristics, including the size of the protein sequence (amino acid), protein molecular weight, isoelectric point (pI), and subcellular localization were listed in Table 1. These LBD genes were predicted to encode proteins 179-599 amino acids in length, with putative molecular weights ranging from 19 to 64 kDa, and isoelectric points (pI) ranging from 4.15 to 9.73. The predicted subcellular location of the LBD proteins was the nucleus in switchgrass.
Phylogenetic analysis of PvLBD genes
MEGA5.0 software was used to construct the phylogenetic tree, examining the evolutionary patterns of 69 LBDs in switchgrass and 43 LBDs in Arabidopsis using the full-length protein sequences. The results showed that all LBD proteins were divided into two classes, named class I and class II. Class I was further divided into four subgroups (Ia, Ib, Ic, and Id), comprising 57 PvLBDs and 37 AtLBDs, while class II comprised 12 PvLBDs and 6 AtLBDs (Fig. 1).
Gene structure, motif composition, and chromosome location of PvLBD genes
In general, alterations in the exon-intron structure can provide important evidence for gene function divergence, especially for duplicate genes. In this study, the exon-intron structure of 69 PvLBD genes was investigated (Fig. 2). The results revealed that PvLBD genes have a different exon-intron organization, with the exon number of PvLBD genes ranging from one to four. Among the 69 PvLBD genes, only one gene (Pavir.Fa00418) had four exons, four PvLBD genes contained three exons, and 18 PvLBD genes had one exon, without any intron disrupting the coding sequence. Most of the PvLBD genes contained two exons. We further analyzed the conserved motifs of the LBD family using the MEME program, and six conserved motifs were identified, with lengths ranging from 41 to 50 amino acids (Table S1).
The distribution of identified PvLBD genes on the chromosomes was carried out using MapDraw and 34 PvLBD genes were unevenly distributed across the seven chromosomes, based on their physical positions (Fig. 3). The distribution of the PvLBD genes was different on each chromosome. Chromosomes 1 and 4 contained one gene, chromosomes 3 and 5 contained two genes, chromosomes 6 and 2 contained four and five PvLBD genes, respectively, and chromosome 9 contained the highest number of genes. Additionally, it was observed that two PvLBD genes were present on chromosome 9b in the form of clusters, implying that these two genes may be part of a single QTL.
Cis‑element analysis of PvLBD genes promoter region
To further understand the transcriptional regulation of the identified switchgrass LBD family genes, we analyzed the cis-regulatory elements of the PvLBD gene promoter region in the vicinity of the upstream 2000 bp region using the online prediction software Plant CARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). In this study, 62 cis-acting elements were identified from switchgrass LBD genes (Fig. 4). TATA box and CAAT box motifs, which play an important role in the precise localization of transcription initiation, presented in the most LBD genes promoter regions. In most cis-acting elements, we found that they can respond to light, control plant growth and development and regulate circadian rhythm. In addition, there are many important cis-acting elements in the promoter region of the PvLBD genes responding to stress. TGACG-motif and CGTCA-motif as cis-acting regulatory elements are involved in MeJA-responsiveness. GARE-motif, P-box, and TATC-box are involved in the gibberellin response. AuxRR-core, TGA-box, and TGA-elements are associated with the auxin response. ERE, ABA responsive element (ABRE), and TCA elements are involved with the ethylene, abscisic acid, and salicylic acid responses, respectively. TC-rich repeats motif is related to defense and stress response. Long terminal repeat is involved in low-temperature response. Myeloblastosis (MYB) binding sites (MBS) is involved in drought response. It was also shown that several switchgrass LBD genes contain ABRE cis-acting elements, and MBS responds to ABA signaling, involved in drought induction. Adenine and uridine-rich elements and GC-motif are related to anaerobic induction. WUN-motif is involved in wound response. Finally, a few motifs are related to protein biosynthetic regulation. MBS is the MYB-binding site involved in flavonoid biosynthetic gene regulation. The O2-site is involved in zein metabolism regulation. These results show that the above structures have important roles in transcription initiation of the LBD gene family (Fig. 4).
Expression patterns of PvLBD in different tissues and developmental stages
In order to deeply analyze the tissue expression profiles of the PvLBD family, the switchgrass Gene Expression Atlas was used to analyze the PvLBD expression patterns in 26 different tissues and developmental stages. We found that only 12 PvLBD genes had differential expression patterns in different organs and tissues; most genes had similar expression patterns (Fig. 5). Interestingly, the Gene Expression Altas showed that four PvLBD genes displayed relatively higher expression levels in lignified tissues (vascular bundle, root, and stem) (Fig. 5), indicating that they may be involved in secondary cell wall strengthening or lignification processes. Taking full account of the importance of lignin content in lignocellulosic grass, understanding the function of PvLBD39 and PvLBD40 in switchgrass could provide new insights into the lignification process.
Expression profiles of PvLBD genes under abiotic stress and hormone treatment
Based on the results from cis‑element analysis of the PvLBD genes promoter region, some genes were shown to respond to abiotic stress (Fig. 4). To further analyze the expression patterns of PvLBD under 350 mM NaCl, 20% PEG, 100 uM ABA, and low-temperature (cold and chilling) stress conditions, 11 PvLBD genes were analyzed by qRT-PCR (Fig. 6). Ten PvLBD genes responded to low-temperature stress (cold and chilling stress), and, except for PvLBD (Pavir.J03238), the expression levels of other genes increased > 10-fold. Seven PvLBD genes were upregulated under saline condition, and PvLBD (Pavir.J13923) had the highest expression level after 24 h when treated with 350 mM NaCl. Five PvLBD genes were upregulated under 20% PEG treatment, and the expression level of PvLBD (Pavir.J19063) increased over seven-fold when treated with 20% PEG for 24 h. Three PvLBD genes were upregulated under ABA treatment. Of these, three genes were upregulated at all three treatment points. However, the up-regulation of PvLBD (Pavir.J19063) in response to these abiotic stresses appeared to be independent of ABA. These results showed that PvLBD genes had different expression patterns under abiotic stress conditions, indicating that they may play important roles in abiotic stress response.
GO and KEGG enrichment analyses of PvLBD target genes under salt and drought stress
To further analyze the regulatory pathways of PvLBDs in abiotic stress, we identified the potential downstream target genes regulated by PvLBDs. LBD motifs were used to search the 2.0-kb promoter sequences upstream of the switchgrass protein coding genes, and target genes were identified for further annotation in Table S3. We also analyzed the binding sites of PvLBD target genes, and the results are shown in Table S4. Members of the LBD family play important roles in plant stress, so we analyzed the GO and KEGG enrichment of PvLBD target genes under salt and drought stress.
PvLBD target genes were classified into three major classes based on GO annotation and enrichment analysis, which involve cellular components, molecular function, and biological processes (Fig. S1). The strongest enrichment was observed in biological process, followed by the molecular function. The top 20 GO terms of enriched target genes were determined. Under salt stress, differentially expressed target genes in the spermidine biosynthetic (GO:0008295) and glucose catabolic processes (GO:0006007) were upregulated (Fig. 7A). Similarly, the results showed that differentially expressed target genes in the oxylipin (GO:0031408), cinnamic acid (GO:0009800), and spermidine (GO:0008295) biosynthetic processes were also upregulated under drought stress, and other GO terms in the enriched target gene set were involved in response to salt stress (GO:0009651) and water deprivation (GO:0009414) (Fig. 8A). Interestingly, the target gene set enriched in the spermidine biosynthetic process was upregulated under salt and drought stress. We also analyzed the differentially expressed PvLBD target genes between salt stress and drought stress, and target genes involved in photosynthesis (GO:0015979), the proline catabolic process (GO:0006562), regulation of the ABA biosynthetic process (GO:0010115), and regulation of the jasmonic acid-mediated signaling pathway (GO:2000022) were all upregulated (Fig. S2A). These results indicated that PvLBDs play vital roles in the whole growth process and response to environmental stresses.
Likewise, among the top 20 KEGG pathways represented by the differentially expressed target genes were carbon metabolism (ko01200), oxidative phosphorylation (ko00190), and biosynthesis of amino acids (ko01230) under salt stress. Additionally, PvLBD target genes were also involved in arginine and proline metabolism (ko00330), pentose phosphate pathway (ko00030), and ABC transporters (ko02010) (Fig. 7B). These results showed that PvLBD target genes were involved in different biological processes and achieved multiple catalytic functions. Similarly, ABC transporters (ko02010), carbon metabolism (ko01200), and biosynthesis of amino acids (ko01230) were also enriched under drought stress (Fig. 8B). Remarkably, the highest number of differentially expressed target genes was involved in plant hormone signal transduction (ko04075) between salt vs. drought stress (Fig. S2B). These results suggest that PvLBDs can influence multiple pathways by regulating their target genes under salt and drought stress.
PvLBD12 gene positively regulate salt tolerance in switchgrass
Based on the above results, we selected PvLBD12 (Pavir.J13923) as a candidate gene for functional analysis, and gained the transgenic switchgrass overexpressing PvLBD12. We found that all transgenic plant displayed vigorous growth and appeared much bigger than WT plants under standard greenhouse conditions (Fig. 9A).
For salinity stress in sand culture, we grew plants in pots in the presence or absence of 0 mM, 150 mM and 300 mM NaCl. While the leaves of transgenic plants were less wilted and appear greener than WT after 7 days of 300 mM salt stress (Fig. 9B). To further evaluate the consequences of physiological changes in transgenic plants, we measured MDA and proline contents under 0 mM, 150 mM and 300 mM NaCl salt stress condition (Fig. 9C-D). After 7 days, the MDA and proline content in the leaves of transgenic plants appeared similar to WT under 0 mM NaCl salt stress condition, but high salt stress significantly increased the MDA and proline content both in the leaves of transgenic and WT plants (Fig. 9C-D). Among these results, MDA content in WT was 1.39-, 1.64- and 1.48-fold higher than that in OE-2, OE-5 and OE-6 under 300 mM NaCl, respectively (Fig. 9C); and the maximum proline level was observed under 300 mM NaCl treatment, at which point the proline level was 1040 μg/g in OE-2, 1129 μg/g in OE-5 and 1071 μg/g in OE-6 compared with about 531 ug/g in WT (Fig. 9D).