The cultivar K326 was used to analyze the expression of GASA genes in tobacco. Seeds were germinated in a nursery tray, seedlings were grown in a greenhouse with a cycle of 14 h light at 28°C/10 h dark at 25°C and relative humidity at 50–60%. Different tissues (root, stem, leaf, axillary bud, and flower) were collected at the flowering stage to analyze NtGASA expression. For phytohormone treatments, 3-week-old seedlings were transferred to plates containing 10 μM ABA, 10 μM GA, 10 μM IAA, 10 μM SA, 50 μM methyl-jasmonate (MeJA), or 1% (v/v) dimethyl sulfoxide (control) and incubated for 5 h under the same photoperiod, temperature, and humidity conditions (Kretzschmar et al., 2012; Zhang et al., 2018).

For NtGASA identification, 15 GASA sequences were obtained from the Arabidopsis database (TAIR; http://www.arabidopsis.org) and used as queries for BLAST search against the Solanaceae Genomics Network (https://solgenomics.net/). Subsequently, the Hidden Markov Model-based profile of the GASA domain PFAM 02704 was used to verify the presence of the complete GASA domain in NtGASA sequences. The non-redundant putative NtGASA sequences with a conserved GASA domain were used for further bioinformatics (phylogenetic relationships, chromosomal locations, Cis-regulatory elements, etc) and expression analysis.

The isoelectric point, number of amino acids, and molecular weight of NtGASA were predicted using the ExPASy tool (http://web.expasy.org/protparam/). The sequences of GASA from Arabidopsis (AtGASA), rice (OsGASA), grapevine (Vitis vinifera; VvGASA), and tobacco (NtGASA) were used to construct a phylogenetic tree using MEGA 7.0 with the neighbor-joining (NJ) method and a bootstrap test of 1,000 replicates (Supplementary Table S1) (Tamura et al., 2007). The exon/intron structure of each NtGASA genes was illustrated using the Gene Structure Display Server (http://gsds.cbi.pku.edu.cn). The conserved motifs of NtGASA proteins were analyzed using MEME 5.1.1 (http://meme-suite. org/tools/meme) (Bailey et al., 2006).

To obtain the chromosomal locations of NtGASA genes, the DNA sequence of each gene was mapped using MG2C 2.0 (http://mg2c.iask.in/mg2c_v2.0/). Segmental and tandem duplicated gene pairs within the tobacco genome, as well as collinear gene pairs among the Arabidopsis, rice, grapevine, and tobacco genomes, were identified using MCScanX (Wang et al., 2012). The collinearity map was constructed using Circos (Krzywinski et al., 2009). The synonymous and non-synonymous substitution rates (Ks and Ka, respectively) were calculated using KaKs_Calculator 2.0 (Wang et al., 2010).

Plant samples were collected from root, flower, leaf, stem and axillary bud of tobacco at flowering stage, total RNA was isolated from frozen samples using Trizol reagent (TaKaRa, Kusatsu, Shiga, Japan), and cDNA synthesis was performed using the M-MLV reverse transcriptase Kit (Thermo Fisher Scientific, Waltham, MA, United States), according to the manufacturer’s instructions. Quantitative reverse-transcription (qRT)-PCR was carried out using the Bio-Rad CFX96 real-time system (Bio-Rad, Hercules, CA, United States) with SYBR Green Master Mix (Bio-Rad). The NtGADPH gene was used as the internal control for data normalization, and the relative expression levels of selected genes were calculated using the 2^−ΔΔCt method (Schmittgen and Livak, 2008). The primers used for qRT-PCR are listed in Supplementary Table S3.

The 3 kb DNA sequence upstream of the start codon of NtGASA genes was examined for the presence of cis-regulatory elements. Cis-regulatory elements in the promoters of each NtGASA gene were analyzed using the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) and classified according to their regulatory functions.

To identify the GASA genes in tobacco, we used 15 AtGASA sequences as queries for BLAST search, and identified 18 putative NtGASA based on amino acid similarities. As shown in Table 1, the total and coding sequence lengths of NtGASA genes were 186 to3,715 bp and 186 to 444 bp, respectively. The deduced NtGASA proteins varied from 61 to 147 amino acids with a molecular weight of 6.6–16.17 kDa, and the isoelectric point ranged from 6.66 to 9.75. Apart from these, the instability index for most of the proteins (77.8%) were more than 35. According to the Grand average of Hydropathicity (GRAVY), the NtGASA proteins were hydrophilic except for NtGASA3, NtGASA4, and NtGASA9. The amino acid content of NtGASA was conserved, cysteine, lysine, and leucine were predominant amino residues. Most NtGASA proteins were localized in the extracellular membrane, chloroplasts, and mitochondria. Detailed information about NtGASA physicochemical characteristics is presented in Table 2.

To characterize the phylogenetic relationships among GASA genes from Arabidopsis, rice, grapevine and tobacco, an unrooted NJ tree was constructed aligning 15 AtGASA, 10 OsGASA, 14 VvGASA, and 18 NtGASA. According to the phylogenetic tree, GASA genes could be classified into three subfamilies: subfamily I included six AtGASA genes (AtGASA1/2/3/9/11/13), three OsGASA genes (OsGASA3/5/7), six VvGASA genes (VvGASA1/5/8/10/11/12), and nine NtGASA genes (NtGASA1–9). Subfamily II included five AtGASA genes (AtGASA4/5/6/12/15), four OsGASA genes (OsGASA4/6/9/10), five VvGASA genes (VvGASA4/6/7/13/14), and four NtGASA genes (NtGASA10–13). Subfamily III included four AtGASA genes (AtGASA7/8/10/14), three OsGASA genes (OsGASA1/2/8), three VvGASA genes (VvGASA2/3/9), and five NtGASA genes (NtGASA14–18) (Figure 1). Therefore, subfamily I had more GASA members from Arabidopsis, grapevine, and tobacco, whereas subfamily II had more GASA members from rice.

The localizations of the NtGASA genes in the chromosomes of tobacco were further determined. Using a simplified physical map, we found that the 18 NtGASA genes were unevenly distributed in 11 chromosomes in the tobacco genome. Chromosome (Chr.) 1, 4, 6, 8, and 21 contained two copies each, whereas Chr. 2, 10, 12, 14, 15, 16, 17, and 18 contained one copy each (Figure 2).

Tandem and segmental duplicates play an important role in the expansion of gene families. Two genes (NtGASA10 and NtGASA11) were tandemly duplicated on Chr.1. In addition, five pairs (NtGASA3/NtGASA4, NtGASA4/NtGASA5, NtGASA6/NtGASA7, NtGASA15/NtGASA16, and NtGASA17/NtGASA18) were segmental duplicated (Figure 2). All tandem and segmental duplicates had Ka/Ks values less than 1 (Table 3), indicating that the six gene pairs evolved under the influence of purifying selection.

We constructed a collinearity plot of the tobacco, rice, grapevine and Arabidopsis GASA gene families to further explore the evolutionary relationships among GASA genes from different species. A total of 2, 4, and 19 collinear gene pairs were identified between tobacco and rice, tobacco and Arabidopsis, and tobacco and grapevine, respectively. Most collinear relationships were many-to-one matches, such as (NtGASA3, NtGASA4, NtGASA5)/AtGASA1, (NtGASA16, NtGASA18)/OsGASA2, (NtGASA2, NtGASA3, NtGASA4, NtGASA5)/VvGASA10 and (NtGASA6, NtGASA7)/VvGASA11. There were also one-to-many matches, such as NtGASA9/(VvGASA1, VvGASA5, VvGASA8), NtGASA15(VvGASA2, VvGASA3, VvGASA9),NtGASA16/(VvGASA2, VvGASA3, VvGASA9) and NtGASA18/(VvGASA2, VvGASA3). The one-to-one matches were NtGASA16/AtGASA7,NtGASA13/VvGASA4 and NtGASA17/VvGASA3 (Figure 3, Supplementary Table S2). These results indicate that GASA genes were relatively conserved between different species and might originate from the same ancestor.

To further explore the phylogenetic relationships among NtGASA genes, an unrooted tree was constructed between NtGASA genes. In concordance with the phylogenetic tree including the tobacco, Arabidopsis, grapevine, and rice GASA genes, this analysis also supported the classification of NtGASA genes into three subfamilies (Figure 4A). The number of conserved motifs in NtGASA proteins varied from three to 6 (Figure 4B). The highly conserved motifs 1, 2, and three were detected in all 18 NtGASA proteins, whereas motif five was only found in NtGASA6 and NtGASA7, motif eight were only found in NtGASA9 and NtGASA10. The diversity of motifs in different subfamilies suggests that NtGASA functions have tended to diversify during evolution.

Structural analysis revealed that the length, arrangement, and position of introns in NtGASA genes were relatively less conserved. For instance, subfamilies I and II contained one to three introns and subfamily III contained one intron, except for NtGASA8 that had only one exon and no intron (Figure 4C). Intron gain and loss is a frequent phenomenon during evolution and can increase the complexity of gene structures.

In previous findings, putative GASA protein possesses highly conserved C-terminal domain that containing 12 conserved cysteines (Aubert et al., 1998). Amino-acid sequence comparison of AtGASA, OsGASA, VvGASA, and NtGASA revealed that all putative NtGASA proteins shared a conserved GASA domain, except for NtGASA17, in which GASA domains were mutated by the insertion of several amino acids (Figure 5).

The spatio-temporal expression analysis of genes can provide information about gene function. We performed qRT-PCR for expression profiling of the NtGASA genes in the root, flower, leaf, stem, and axillary bud. The expression profiling showed that most NtGASA genes had diverse expression patterns in different tissues. NtGASA3, NtGASA11, NtGASA17, and NtGASA18 were expressed relatively ubiquitously. Whereas many NtGASA genes showed high expression in specific tissues, such as NtGASA9 had the highest expression levels in the stem, NtGASA7 in the leaf, NtGASA16 in the axillary bud, and NtGASA2, NtGASA5, NtGASA6, NtGASA10, NtGASA13, NtGASA14, and NtGASA15 in the flower. Notably, NtGASA12 had the lowest expression levels in the stem. In general, most NtGASA genes were highly expressed in reproductive organs (i.e., flower) compared with vegetative parts (i.e., leaf and stem) (Figure 6).

The study of cis-elements could provide clues about regulatory pathways of gene expression, then we analyzed the 3,000-bp upstream promoter sequences of NtGASA genes. The largest number of cis-elements observed across the NtGASA genes was associated with light-responsiveness. In addition, cis-elements involved in phytohormone (i.e., ABA, GA, IAA, SA, and MeJA) and stress (i.e., low temperature) responses were also identified in the promoter sequences of NtGASA genes (Figure 7). The diversity in response elements indicated the regulatory roles of NtGASA genes in various physiological and biological processes.

The results of the cis-element analysis indicated that NtGASA genes might be related to many plant hormone responses. To elucidate the expression pattern of NtGASA genes and their possible roles in phytohormone signaling pathway, the transcript levels of all NtGASA genes under ABA, GA, IAA, MeJA, SA treatment were investigated. The expression profiling of NtGASA under different phytohormone treatments showed diverse patterns compared with the control. For instance, ABA significantly upregulated the expression of NtGASA1/2/3/4/8/9/13/14, but inhibited the expression of NtGASA5/10/17/18. Most of the NtGASA genes were highly expressed by GA treatment, except for NtGASA6/7/10/16/17/18. After IAA treatment, the expression of NtGASA1/3/4/8/9/14/15 were significantly upregulated, NtGASA11 was downregulated. The expression of NtGASA1/2/3/4/8/9/11/12 were significantly upregulated by SA treantment, NtGASA6/7/10/15/16 were downregulated. Moreover, after MeJA treantment, the expressions of NtGASA3/4/8/9/11/12/13/16 were significantly upregulated, and the expression of NtGASA16 was only upregulated by MeJA treatment, NtGASA1/2/5/6/7/15 were downregulated. Interestingly, the expression of NtGASA17 and NtGASA18 were downregulated by all phytohormones (Figure 8). These findings indicated that different NtGASA genes might play distinctive roles in response to various phytohormone signals.