Genome-Wide Identification of SnRK1 Catalytic α Subunit and FLZ Proteins in Glycyrrhiza inflata Bat. Highlights Their Potential Roles in Licorice Growth and Abiotic Stress Responses

Sucrose non-fermenting-1-related protein kinase-1 (SnRK1) and its scaffolding proteins, FCS-like zinc finger proteins (FLZs), are well conserved in land plants and involved in various processes of plant growth and stress responses. Glycyrrhiza inflata Bat. is a widely used licorice species with strong abiotic stress resistance, in which terpenoids and flavonoids are the major bioactive components. Here, we identified 2 SnRK1 catalytic α subunit encoding genes (GiSnRK1α1 and GiSnRK1α2) and 21 FLZ genes in G. inflata. Polygenetic analysis showed that the 21 GiFLZs could be divided into three groups. A total of 10 representative GiFLZ proteins interact with GiSnRK1α1, and they display overlapped subcellular localization (mainly in the nucleus and the cytoplasm) when transiently expressed in Nicotiana benthamiana leaf cells. Coinciding with the existence of various phytohormone-responsive and stress-responsive cis-regulatory elements in the GiSnRK1α and GiFLZ gene promoters, GiFLZs are actively responsive to methyl jasmonic acid (MeJA) and abscisic acid (ABA) treatments, and several GiFLZs and GiSnRK1α1 are regulated by drought and saline-alkaline stresses. Interestingly, GiSnRK1α and 20 of 21 GiFLZs (except for GiFLZ2) show higher expression in the roots than in the leaves. These data provide comprehensive information on the SnRK1 catalytic α subunit and the FLZ proteins in licorice for future functional characterization.


Introduction
Licorice (Glycyrrhiza) belongs to the genus Leguminosa Glycyrrhiza linn, and there are around 20 species/subspecies worldwide [1,2]. Licorice is a group of a commercially valuable plants, with a wide range of uses in the pharmaceutical, tobacco, cosmetics, and food industries [3]. Owing to its beneficial effects for reducing toxicity and increasing the efficacy of other herbal medicines when used together, licorice is one of the most prescribed herbs [4,5]. Around 400 total bioactive compounds have been isolated from licorice [6,7]. For an example, glycyrrhizic acid, a kind of triterpenoid mainly accumulated in the roots and rhizomes [8], shows good inhibitory effects on HIV [9], hepatitis [10], leukemia [11], gastric cancer [12], and SARS-CoV-2 [13,14]; it is also commonly used as a spice and

Identification of SnRK1 Catalytic α Subunit and FLZ Proteins in G. inflata
In this study, we used the protein sequence of AtSnRK1.1 (AT3G01090.1) as a query to perform BLASTP against our sequenced G. inflata genome and successfully identified two SnRK1 catalytic α subunit encoding genes, named GiSnRK1α1 and GiSnRK1α2 (Table 1). Using the same strategy (the protein sequence of the FLZ domain in AtFLZ1 (AT5G47060.1) used as the query), 21 FLZ protein encoding genes were identified (Table 1), and named as GiFLZ1 through GiFLZ21, based on their chromosomal localization ( Figure S1). Detailed information about each GiSnRK1α and GiFLZ gene, including gene name, locus ID, chromosome, open reading frame (ORF) and protein length, predicted protein molecular weight (MW), isoelectric point (pI), and subcellular localization, is listed in Table 1. Their gene structures were shown in Figure S2. All GiFLZ genes contain 2 exons and 1 intron ( Figure S2), the ORFs range from 348 to 1338 bp, and the encoded proteins contain 115 to 445 amino acids, with MW from 13.00 to 48.91 kDa, and the pI from 5.92 to 10.48 (Table 1).

Phylogenetic Analysis of FLZ Proteins from G. inflata and Arabidopsis thaliana
To investigate the phylogenetic relationships of GiFLZ gene family members, a total of 39 FLZ proteins from G. inflata (21) and Arabidopsis thaliana (18) were used to construct a neighbor-jointing phylogenetic tree. These FLZ proteins were roughly clustered into three major groups (groups I to III) ( Figure 1A). Among them, group III contains the largest number of GiFLZs-up to 10-followed by Group I-containing 8-and Group II-containing 3-GiFLZs ( Figure 1A). Multiple sequence alignment analysis confirmed that all of the 39 FLZ proteins have an FLZ domain sharing highly conserved sequence identity (CX 2 CX 3 LX 3-4 DX 3 YX 4-5 FCSX 2 CR) ( Figure 1B).

Protein Interaction between GiSnRK1α1 and GiFLZs
Previous studies demonstrated that FLZ proteins could interact with SnRK1 catalytic α subunits in Arabidopsis, rice, and maize [28][29][30][31]36,37]. To determine whether GiFLZs also interact with GiSnRK1α, we selected ten GiFLZs from all three different groups: GiFLZ4 from Group I; (GiFLZ2, GiFLZ15/16-both CDS and protein sequences are identical with different chromosomal localization- Table 1 and Figure S1), GiFLZ18 and GiFLZ21 from Group II; and GiFLZ3, GiFLZ9, GiFLZ10, GiFLZ14, and GiFLZ17 from Group III ( Figure 1A, marked with asterisks) to perform a yeast two-hybrid experiment with GiSnRK1α1 as the example (Figure 2). To this end, the full-length GiFLZs were fused to the GAL4 activation domain, while GiSnRK1α1 was fused to the DNA binding domain, to produce AD-GiFLZ and BD-GiSnRK1α1 vectors, respectively. As expected, all the ten AD-GiFLZs interacted with BD-GiSnRK1α1, but not with the empty BD vector, as judged by the growth of the co-transfected yeast cells on selective medium plates ( Figure 2). As shown in Figure 2A The phylogenetic tree was constructed based on the m tiple sequence alignments of FLZ proteins and generated with MEGA 5.0 software using the ne bor-joining method. The red asterisks marked the selected genes used in the subsequent study Protein sequence alignment of the FLZ domain. ClustalW was used for protein sequence alignm The identical and similar amino acids in the FLZ domains were highlighted with dark and light g backgrounds, respectively. The asterisks marked the conserved amino acids.

Subcellular Localization of GiSnRK1α1-GFP and GiFLZs-GFP Fusions
Protein subcellular localization is important for understanding its function and aids the identification of potential interacting partners [38]. The WoLF PSORT prediction revealed that GiSnRK1α proteins might reside in the cytoplasm and the nucleus (Table 1). To validate this prediction, a fusion of GiSnRK1α1 to green florescence protein (GFP) under the control of the AtUBQ10 promoter was co-expressed with Histone3-mCherry (nuclear marker) in 6-week-old N. benthamiana leaf epidermal cells. The fluorescence signals indicated that GiSnRK1α1-GFP is localized to both the nucleus and the cytoplasm ( Figure 3A), similar to that of AtSnRK1.1 [26,31].

Subcellular Localization of GiSnRK1α1-GFP and GiFLZs-GFP Fusions
Protein subcellular localization is important for understanding its function and aids the identification of potential interacting partners [38]. The WoLF PSORT prediction revealed that GiSnRK1α proteins might reside in the cytoplasm and the nucleus (Table 1). To validate this prediction, a fusion of GiSnRK1α1 to green florescence protein (GFP) under the control of the AtUBQ10 promoter was co-expressed with Histone3-mCherry (nuclear marker) in 6-week-old N. benthamiana leaf epidermal cells. The fluorescence signals indicated that GiSnRK1α1-GFP is localized to both the nucleus and the cytoplasm ( Figure 3A), similar to that of AtSnRK1.1 [26,31]. Subcellular localization of GiFLZ proteins predicted by WoLF PSORT suggested that they were localized in diverse subcellular compartments, including the nucleus, cytoplasm, chloroplast, and mitochondria (Table 1). To confirm this result, the above selected 10 GiFLZs were fused with a GFP tag, and the resultant chimeric proteins were transiently co-expressed with Histone3-mCherry in N. benthamiana leaves. Confocal scanning microscopy revealed that the GFP signals of GiFLZ4, 9, 10, 14, 15/16, 18, and 21 fusions were found both in the nucleus and the cytoplasm, as was the case for GFP protein ( Figure 3B). Interestingly, GiFLZ2-GFP and GiFLZ3-GFP were shown to be localized only in the nucleus ( Figure 3C). The results are, in general, consistent with the bioinformatics prediction, in which they mainly localize in the nucleus (Table 1 and Figure 3). Furthermore, the overlapped distribution of GiSnRK1α1 and GiFLZs in the subcellular compartments allows for their potential interaction in plant cells.

Cis-Regulatory Elements in GiSnRK1α and GiFLZ Gene Promoters
Cis-elements in promoter regions are bound by different types of transcription factors to determine gene expression. We used the 2.0-kb genomic sequence upstream of the start code ATG of each gene to identify cis-elements against the PlantCARE database [39]. In total, 10 kinds well-studied stress-responsive and 12 kinds of phytohormone-responsive cis-elements were found (Figure 4 and Figure S3). As presented in Figure 4, each gene promoter contains 3-19 stress-responsive and 2-15 phytohormone-responsive cis-elements, respectively ( Figure 4). In the stress-responsive group, anaerobic induction (ARE, GCmotif), drought-inducibility (MBS, DRE-core, DRE1), low temperature (LTR), stress (STRE, TC-rich repeats), and wounding responsiveness (WRE3, WUN-motif) elements were found ( Figure 4 and Figure S3). In the phytohormone-responsive group, Me-JA (MYC, TGACGmotif), ABA (ABRE), salicylic acid (TCA-element), ethylene (ERE), auxin (TGA-element, AuxRR-core, TATC-box), and gibberellin (GARE-motif, P-box) elements were discovered ( Figure 4 and Figure S3). These results suggest that the GiSnRK1α and GiFLZ genes might actively respond to various stresses and phytohormone signals, and that they can be regulated by different types of transcription factors.

Expression Pattern of GiSnRK1α and GiFLZ Genes in Response to MeJA and ABA Treatment
MeJA has been reported to regulate the biosynthesis of multiple specialized metabolites [40], for example, flavonoid and soyasaponin biosynthesis in G. uralensis [41,42], and 2 GiSnRK1α and 18 GiFLZ genes contain at least one JA-responsive cis-element ( Figure 4).

Expression Pattern of GiSnRK1α and GiFLZ Genes in Response to MeJA and ABA Treatment
MeJA has been reported to regulate the biosynthesis of multiple specialized metabolites [40], for example, flavonoid and soyasaponin biosynthesis in G. uralensis [41,42], and 2 GiSnRK1α and 18 GiFLZ genes contain at least one JA-responsive cis-element ( Figure 4). Therefore, we checked the responsiveness of GiSnRK1α and GiFLZ genes to MeJA treatment. As shown in Figure 5A, GiSnRK1α2 showed obvious repressed expression, and 14 GiFLZ genes were MeJA-responsive, with 8 members being downregulated and 6 upregulated in hair roots with MeJA treatment for 2 h ( Figure 5A). To further confirm this result, we performed a qRT-PCR assay. It is worth noting that, because of the high percentage of sequence identity of GiFLZ13/14 (97.8%), GiFLZ15/16 (100%), and GiFLZ18/19 (98.2%) gene pairs, we could not distinguish them in the qRT-PCR assay, and the expression of GiFLZ15/16 was too low to detect. As shown in Figure 5B, GiFLZ2 was induced about 6-fold by MeJA application, whereas GiFLZ3, 4, 9, 10, 17, 18/19, and 21 were repressed, and GiSnRK1α1 did not change significantly. Among them, the expression pattern of other genes in the qRT-PCR assay were consistent with the results obtained from the RNA-seq analysis, except for GiFLZ3, 4 and 13/14. These results showed that most GiFLZ genes are MeJA responsive and might be involved in JA signaling to exert function in JA-regulated specialized metabolic processes. MeJA responsive and might be involved in JA signaling to exert function in JA-regulated specialized metabolic processes. ABA is the most important stress phytohormone [43], and 2 GiSnRK1α and 15 GiFLZ genes contain at least one ABRE element (Figure 4). To understand how the GiSnRK1α and GiFLZ genes respond to ABA, we performed qRT-PCR analysis in the roots of 5-dayold seedlings under 100 μM ABA treatment for 2 h. The results showed that ABA treat- ABA is the most important stress phytohormone [43], and 2 GiSnRK1α and 15 GiFLZ genes contain at least one ABRE element (Figure 4). To understand how the GiSnRK1α and GiFLZ genes respond to ABA, we performed qRT-PCR analysis in the roots of 5-day-old seedlings under 100 µM ABA treatment for 2 h. The results showed that ABA treatment could significantly induce the expression of GiSnRK1α1, GiFLZ10, 13/14, and 21, but repress those of GiFLZ2 and 4. The expression of GiFLZ3, 9, 17, and 18/19 showed no obvious change ( Figure 5C).

Expression Patterns of GiSnRK1α and GiFLZ Genes in Response to Drought and Saline-Alkaline Stresses
Recent studies have suggested that FLZ and SnRK1 genes are involved in signaling and in the response to abiotic stimuli [36,37,44]. In the field, the most frequent abiotic stresses to licorice are drought and soil saline-alkaline stresses [16]. To uncover the potential roles of GiSnRK1α and GiFLZ genes in plant response to these stresses, we performed a qRT-PCR analysis in the roots of 5-day-old seedlings under drought (20% PEG6000) and mixed saline-alkaline (A5, B5 and C5, referred to Peng et al., 2008 [45]) treatments. After 2 h of 20% PEG6000 treatment, GiFLZ2, 3, 9, and 10 showed a significant decrease in transcription, whereas GiFLZ21 displayed a slight increase, and the others did not show any obvious change, when compared to the mock treatment ( Figure 6A). With high salt treatment (A5), GiFLZ2 was strongly induced by about six fold, while GiFLZ3, 9, 13/14, 18/19, and 21 showed significantly decreased expression ( Figure 6B), and the expression levels of GiFLZ4, 10, 17, and GiSnRK1α1 showed no change. Upon slight (B5) and moderate strength (C5) mixed saline-alkaline stresses, GiFLZ2 was strongly induced; GiFLZ10, 17, 19, 21, and GiSnRK1α1 also showed increased expression (about two fold), while GiFLZ9 and 13/14 obviously showed repressed expression ( Figure 6C,D). These data indicated that the GiSnRK1α1 and GiFLZ genes are actively responsive to the environmental stresses occurring in the natural habitat of G. inflata, and GiFLZ2 showed the highest sensitivity to these stresses.

Expression Pattern of GiSnRK1α and GiFLZ Genes in Root and Leaf Tissues
Because the major beneficial compounds are accumulated more in roots than in the leaves, and the roots are the major medicinal elements [7,8], we checked the expression of the GiSnRK1α and GiFLZ genes in the roots and leaves. Interestingly, in our RNA-seq analysis, we found that these two GiSnRK1α, and most GiFLZ genes, exhibited greater expression in the roots than in the leaves, and this trend is more obvious in 2− and 3−year-old plants ( Figure 7A). For instance, in 3-year-old plants, there are 19 GiFLZ genes showing obvious higher expression in the roots than in the leaves ( Figure 7A). The only outlier is GiFLZ2, which displayed lower expression in the roots of 1-and 2-year-old plants, but comparable expression in roots and leaves in 3-year-old plants ( Figure 7A). A similar expression pattern of FLZ genes in 2-year-old G. uralensis and G. glabra plants was also observed ( Figure S4), indicating that this specific expression pattern of FLZ genes is conserved in different licorice species. To validate the results obtained from RNA-seq, we checked the gene expression of the selected GiSnRK1α1 and GiFLZs with qRT-PCR assay, using roots and leaves harvested from 1-year-old G. inflata ( Figure 7B). As shown in Figure 7B, only GiFLZ2 showed lower expression in roots than leaves, in good agreement with the results of the RNA-seq assay, and all other GiFLZ genes and GiSnRK1α1 showed higher expression in the roots than in the leaves. These results indicated that these GiFLZ and GiSnRK1α genes mainly function in the roots, and they might be involved in the biological processes occurring mainly in roots, i.e., glycyrrhizin biosynthesis and accumulation [8]. strength (C5) mixed saline-alkaline stresses, GiFLZ2 was strongly induced; GiFLZ10, 19, 21, and GiSnRK1α1 also showed increased expression (about two fold), while GiF and 13/14 obviously showed repressed expression ( Figure 6C,D). These data indica that the GiSnRK1α1 and GiFLZ genes are actively responsive to the environmental stre occurring in the natural habitat of G. inflata, and GiFLZ2 showed the highest sensitivit these stresses. Figure 6. Expression pattern of selected GiSnRK1α and GiFLZ genes in abiotic stress treatments. qRT-PCR data were assayed for the expression of selected GiSnRK1α and GiFLZ genes in the r of 5-day-old seedlings subjected to 20% PEG6000 (A), A5 (B), B5 (C), and C5 (D) treatment for The fold change of treatment (T) in relation to mock (M) are presented as mean ± SD (n = 9), and data on the graph indicate the ratios. * p < 0.05; ** p < 0.01-determined by Student's t-test.

Expression Pattern of GiSnRK1α and GiFLZ Genes in Root and Leaf Tissues
Because the major beneficial compounds are accumulated more in roots than in leaves, and the roots are the major medicinal elements [7,8], we checked the expressio the GiSnRK1α and GiFLZ genes in the roots and leaves. Interestingly, in our RNAanalysis, we found that these two GiSnRK1α, and most GiFLZ genes, exhibited gre expression in the roots than in the leaves, and this trend is more obvious in 2− and 3−ye 7B, only GiFLZ2 showed lower expression in roots than leaves, in good agreement with the results of the RNA-seq assay, and all other GiFLZ genes and GiSnRK1α1 showed higher expression in the roots than in the leaves. These results indicated that these GiFLZ and GiSnRK1α genes mainly function in the roots, and they might be involved in the biological processes occurring mainly in roots, i.e., glycyrrhizin biosynthesis and accumulation [8]. qRT-PCR assay. The relative expression from root to leaf is presented as mean ± SD (n = 3), and the data on the graph indicate the ratios. * p < 0.05; ** p < 0.01-determined by Student's t-test.

Discussion
SnRK1 is evolutionarily conserved in all eukaryotes, and FLZ genes first emerged in liverwort and were conserved in land plants [32]. Genome-wide identification and characterization of FLZ genes have recently been reported in Arabidopsis [31], rice [36], maize [37], and wheat [35]. However, they have not been studied in medicinal plants.
Here, we report the identification of two SnRK1 catalytic α subunit encoding genes and 21 FLZ genes in G. inflata (Table 1). The Y2H assay showed that GiFLZs extensively interact with GiSnRK1α1 (Figure 2), similar to the cases observed in Arabidopsis, rice, and maize [28][29][30][31]36,37], indicating that the SnRK1α-FLZ module is also conserved in licorice. Subcellular localization analysis showed that GiSnRK1α1 and most examined GiFLZs are localized both in the nucleus and the cytoplasm, meeting the requirement for their interaction at the subcellular level ( Figure 3). Furthermore, the gene expression patterns of GiSnRK1α1 and GiFLZs are similar and actively responsive to phytohormone or abiotic stress treatments (Figures 5-7), meeting the requirement for their interaction at the tissue level. More work should be carried out to explore the biological functions of the GiSnRK1α-GiFLZ module in the future.
MeJA has been reported to regulate flavonoid and soyasaponin biosynthesis in G. uralensis [41,42]. In this work, we found that the gene expression level of GiSnRK1α1 is not altered by exogenous MeJA treatment ( Figure 5), but 14 of 21 GiFLZ genes are actively responsive to MeJA, and among them, 8 GiFLZ genes are downregulated, and 6 are upregulated (Figure 5), implying the critical roles of GiFLZs during licorice JA signaling or JA-mediated biological processes. In Arabidopsis, AtFLZs are found to interact with several transcription factors, including DELLA proteins [30], which interact with the core component in JA signaling MYC2 [46] to modulate anthocyanidin biosynthesis [47]. Therefore, in future, it will be interesting to explore the function of GiSnRK1α-GiFLZ modules in regulating plant specialized metabolism.
Wild G. inflata plants usually grow in arid, semi-arid, and saline-alkaline regions, and therefore, they have the ability to adapt to these harsh conditions [16][17][18][19]. Thus, G. inflata makes a good research material to explore the stress-resistant genes for crop breeding. In Arabidopsis, the ectopic expression of wheat TaSRHP and TaFLZ1D results in enhanced tolerance to salt stress [34,35], and the overexpression of AtSnRK1.1 could confer resistance to drought stress [24]. Recently, a report proved that LbSnRK1 positively regulates the tolerance of plants to high salt and drought stresses using the transgenic method in the sweet potato [48]. In this study, we found that the expression of GiSnRK1α1 is induced by PEG and mixed saline-alkaline stresses (Figure 6), and most of the examined GiFLZ genes are also actively responsive to PEG and saline-alkaline treatments ( Figure 6). ABA is called the "stress hormone," and it is involved in the abiotic stress responses tested above. We checked the gene expression pattern under ABA treatment, and similar results were found ( Figure 5C), indicating that GiSnRK1α and GiFLZs might also be involved in the stress responses in licorice. Among them, GiFLZ2 is very special because it is strongly repressed by drought, but highly induced by salt and salt-alkali treatments (Figure 6), implying its important roles regarding the stress responses in licorice.
Interestingly, the RNA-seq data and qRT-PCR results showed that most GiFLZ genes and GiSnRK1α are highly expressed in the roots, and this trend is gradually enhanced following plant growth (Figure 7). Furthermore, similar expression profiles of FLZ genes were also observed in the other two Glycyrrhiza species, G. uralensis and G. glabra (Figure S4), indicating that the function of the FLZ genes may be similar in different licorice species. Since many active compounds accumulate specifically in licorice roots [7,8], the rootspecific expression pattern of these genes points to their potentials in regulating specialized metabolism. In the future, it will be interesting to investigate the roles of GiSnRK1α and GiFLZs in regulating licorice growth and/or specialized metabolic processes.

Data Search and Analyses
The genomic data of G. inflata used in this study were sequenced and annotated by our lab (https://ngdc.cncb.ac.cn/;NGDC; CRA009044; accessed on 25 November 2022); Arabidopsis SnRK1α and FLZ data were downloaded from the TAIR database (https:// www.arabidopsis.org/; accessed on 5 March 2022); The transcriptome data of roots and leaves or MeJA-treated hair roots have been deposited in the NCBI GEO repository (http: //www.ncbi.nlm.nih.gov/geo; PRJNA574093; accessed on 25 September 2019). We used the BlastP program to search for all SnRK1α and FLZ proteins in the G. inflata genome with the BioEdit 7.2 software (https://bioedit.software.informer.com/; accessed on 10 March 2022). The protein sequences of AtSnRK1.1 (AT3G01090.1) and the FLZ domain of AtFLZ1 (AT5G47060.1) were used as queries to blast against the G. inflata genome (e ≤ 1 × 10 −5 , identities ≥ 0.5). The phylogenetic trees of the FLZ proteins from A. thaliana and G. inflata were constructed using the neighbor-joining (NJ) method by MEGA5.0 with a bootstrap 1000 [48]. The alignment of the protein sequences of the FLZ domain were conducted using MEGA5.0 software with the ClusterW program and edited with GeneDoc 2.7 (https://genedoc.software.informer.com/; accessed on 25 March 2022). The chromosomal locations of the GiFLZ genes were provided by the G. inflata genome data, and ShinyCircos software (https://shiny.hiplot.com.cn/shiny-circos/; accessed on 10 April 2022) was used to draw the location images. The WoLF PSORT web server (http: //www.genscript.com/psort/wolf_psort.html; accessed on 10 May 2022) was used to predict the subcellular localizations of the GiSnRK1α and GiFLZ proteins, and the first two items are listed in Table 1. Gene structure information of each GiSnRK1α and GiFLZ was obtained from the generic feature format file of G. inflata genome data and TBtools (https: //github.com/CJ-Chen/TBtools; accessed on 20 November 2022) was used to display their intron-exon structures. For cis-elements analysis, the promoter sequence (−2000 bp) of each gene was inferred on PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/ html/; accessed on 20 November 2022), and the stress-responsive and phytohormoneresponsive cis-elements were recorded. The heat maps of gene expression profiles were generated with the OmicShare tools platform (http://www.omicshare.com/tools; accessed on 15 May 2022).

Plant Materials and Stress Treatments
The roots and leaves of 1−, 2−, and 3−year-old licorice plants were collected in the test field at the Northwest Biological Agricultural Center (Ningxia, China) and flush frozen in liquid nitrogen. For the stress treatments, G. inflata seeds were surface sterilized and germinated on wet filter papers in darkness for 2 days at 25 • C, then transferred to 1/2MS plates and grown at 25 • C under a 16/8 h light/dark photo-cycle. After 5 days of growth, the seedlings were subjected to different treatments (water-mock, 100 µM MeJA, 100 µM ABA, 20% PEG6000, and saline-alkaline stresses, marked as A5, B5, and C5, all with salinity 160 mM and pH 7.26, 8.30, and 9.02, respectively, according to [45]). After 2 h, the roots of five seedlings were pooled as one sample, flush frozen with liquid nitrogen, and stored at −80 • C. All experiments were carried out with three independent biological replicates.

RNA Extraction and Quantitative Reverse-Transcription PCR (qRT-PCR) Analysis
Total RNA was extracted using the HiPure Total RNA Mini Kit (Code No. R4151-03, Magen, China). Then, 1 µg of total RNA was used for cDNA synthesis with the PrimeScript ® RT reagent Kit with gDNA Eraser (Cat No. RR047A, Takara, Dalian, China), according to the manufacturer's instructions. The qRT-PCR reactions were performed in 384-well blocks using TB Green Premix ExTaq II (Tli RNaseH Plus) (Code No. RR820D, Takara, Dalian, China) on a LightCycler 480 (Roche, Switzerland), according to manufacturers' instructions. The 2 −∆∆CT method was used to normalize and calculate the expression levels of each tested gene relative to the internal reference gene (GiCS) [49,50]. All the primers used in this study are listed in Table S1. Three biological replicates and three technical replicates were performed. Data were presented as the mean ± SD. Statistical analysis was performed in Microsoft Excel using TTEST at the 0.05 (*) and 0.01 (**) level of significance.

Subcellular Localization Assays in Nicotiana benthamiana Leaves
The subcellular localizations of GiSnRK1α and GiFLZ were tested by transient expression assays in 6-week-old N. benthamiana leaf epidermal cells. The full-length CDS of GiSnRK1α1 and GiFLZs were cloned into the binary vector pCambia1300-UBQ-GFP in a frame with a GFP at the C-terminals, with the primers used listed in Table S1. The fusion constructs were co-transfected into tobacco leaves with a nuclear localization marker (Histone3-mCherry) using an agroinoculation assay [36]. A confocal fluorescence microscope (Laica, SP8) was used to determine the localization of the target proteins 3 days after inoculation.

Yeast Two-Hybrid Assay
The yeast two-hybrid (Y2H) assay was performed following the manufacturer's instructions (Clontech, http://www.clontech.com/; accessed on 12 March 2022) [36,37]. To generate the activation domain (AD)-fused GiFLZs, the full-length CDS of the respective genes were fused in-frame to the pGADT7 vector. For DNA-binding domain (BD)-fused GiSnRK1α1, the full-length CDS of GiSnRK1α1 was PCR-amplified and separately cloned into the pGBKT7 vector. The primers used are listed in Table S1. Yeast AH109 cells were co-transformed with each set of different vector combinations, as indicated. All yeast transformants were grown on the SD/−Trp−Leu (SD-2) medium and screened on the SD/−Trp−Leu−His (SD-3) or SD/−Trp−Leu−His−Ade (SD-4) medium for interaction tests.

Conclusions
In this study, we identified 2 SnRK1 catalytic α subunit encoding genes and 21 GiFLZ genes in G. inflata. The 21 GiFLZs could be roughly divided into 3 groups. A total of 10 representative GiFLZ proteins interacted with GiSnRK1α1, and they displayed overlapped subcellular localization. RNA-seq analysis showed that 20 GiFLZ genes were highly expressed in the root, and 14 members are responsive to MeJA treatment. Furthermore, qRT-PCR analysis showed that the expression levels of GiSnRK1α1 and several GiFLZ genes (especially GiFLZ2) were significantly changed upon drought, saline-alkaline, or ABA stresses, highlighting their potential roles during the environmental adaptation of G. inflata. The comprehensive understanding of the GiSnRK1α and GiFLZ genes provides useful information for further functional studies to elucidate their regulation mechanisms and build a foundation for the future cultivation of high-quality cultivars of licorice through molecular breeding methods.

Data Availability Statement:
The genomic data that support the findings of this study are available from the National Genomics Data Center (https://ngdc.cncb.ac.cn/; NGDC; CRA009044; accessed on 25 November 2022). The transcriptome data are available from the NCBI GEO repository (http: //www.ncbi.nlm.nih.gov/geo; PRJNA574093; accessed on 25 September 2019).

Conflicts of Interest:
The authors declare no conflict of interest.