The N. tabacum L. K326 plants were cultured in a greenhouse with conditions described in previous report (Li Z. et al., 2021). For expression profiles of the NtLHT genes in different tissues, total RNA was isolated from the roots, axillary buds, stems, flowers, and leaves of tobacco at the flowering stage. For drought treatment, 3-week-old seedlings were not watered, the obviously wilted plants were termed as drought stress. For cold treatment, 3-week-old seedlings were exposed to 4°C for 5 h (Xie et al., 2021).

For NtLHT identification, the protein sequences of AtLHTs were retrieved from TAIR¹ (Rhee et al., 2003), the obtained sequences were then used as a query to search putative NtLHT genes from the China tobacco genome database (data not shown). Top hits for putative NtLHT proteins were retained on the basis of high scores (score ≥ 0) and low E-values (E value ≤ 0.1), the NCBI conserved domain database (CDD)² was used to analyze the conserved domain (PF01490) of candidate NtLHT proteins (Rabby et al., 2022). After manually removing the redundant sequences, 23 NtLHT genes were identified and used for further bioinformatics and expression analysis.

The physiochemical properties of NtLHT proteins were predicted using the ExPASy tool (Artimo et al., 2012). The subcellular location of NtLHT proteins were predicted using The WOLF PSORT II program.³ The PROTTER 1.0 was used to predict the presence of transmembrane helices,⁴ and the tertiary protein structures of NtLHT proteins were analyzed by PHYRE server v2.0.⁵

The protein sequences of LHT from tobacco (NtLHT), tea (CsLHT), Arabidopsis (AtLHT), and rice (OsLHT) were used to construct a phylogenetic tree in MEGA X program via the neighbor-joining (NJ) method with 1,000 bootstrap replicates (Supplementary Table 1). The coding sequence of each NtLHT (Supplementary Table 2) was aligned with its corresponding genomic sequence to identify the exon–intron structure by GSDS tool.⁶ The conserved motifs of NtLHT proteins were analyzed using MEME tool with default parameter (Bailey et al., 2006).

The chromosomal positions of NtLHT genes were mapped using MG2C 2.0.⁷ Segmental and tandem duplicated gene pairs among the tobacco, tea, Arabidopsis, and rice genomes were conducted using MCScanX. The collinearity map was drawn with Circos. The non-synonymous substitution (Ka) and synonymous substitution rate (Ks) of duplicated genes were determined using KaKs Calculator 2.0.

To assess the cis-regulatory elements of the NtLHT promoters, the 2 kb DNA sequence upstream from the transcription start site (start codon ATG) of NtLHT genes were extracted, the obtained sequences were analyzed using the PlantCARE program⁸ and classified according to their regulatory functions.

Total RNA was isolated from plant tissue samples using the Plant RNA Kit (Transgen, Beijing, China). The first strand cDNA synthesis was performed according to the manufacturer’s directions (Thermo Fisher Scientific, Waltham, MA, United States). The quantitative real time PCR (qRT-PCR) was carried out using the CFX96 (Bio-Rad, Hercules, CA, United States) with 20 μL volume (containing 1 μL of 1:10 diluted cDNA, 200 nM of each gene-specific primer, and SYBR Green Mix from Bio-Rad). PCR cycling parameters were set as following: 95°C for 5 min, 40 cycles of 10 s at 95°C and 30 s at 60°C, then a final melting curve at 65°C for 5 s. The stably expressed gene NtL25 (ribosomal protein gene) was used as the internal control for data normalization. Expression data was carried out with three biological and technical replicates and calculated using the 2^–ΔΔCt method (Schmittgen and Livak, 2008). The primers used in this study are listed in Supplementary Table 3.

The coding sequence (CDS) of NtLHT22 without the stop codon was amplified and cloned into pEarlyGate101 vector (Invitrogen, Carlsbad, CA, United States) to produce 35S: NtLHT22-YFP construct, the empty pEarlyGate101 vector was used as a control. Tobacco protoplasts preparation and transformation for subcellular localization experiments were carried out as previously described (Schweiger and Schwenkert, 2014). After transformation, the fluorescence signals was observed via Leica SP8-X confocal laser scanning microscope (Leica, Wetzlar, Germany).

NtLHT22 was amplified and cloned into the modified overexpression vector pEarlyGate101 (Invitrogen, Carlsbad, CA, United States) to produce 35S: NtLHT22-Myc construct. The CRISPR/Cas9 vector was constructed as previously described (Gao et al., 2018). The CRISPR-P 2.0 software⁹ was used to design the target sites of NtLHT22. Two 20 bp DNA target sites were synthesized and inserted to the pORE-Cas9 binary vector, the resulting constructs was verified via PCR and sequencing. The overexpression and pORE-Cas9 vector was transformed into Agrobacterium tumefaciens (LBA4404), respectively. The Agrobacterium-mediated transformation and regeneration of transgenic tobacco plants were carried out as previously described (Wang et al., 2018; Supplementary Figure 1). Two independent knockout and overexpression lines of NtLHT22 were identified and used for subsequent experimental studies.

Free amino acid contents were measured according to the ninhydrin method with some modifications (Fang et al., 2013). 2.0 g sample was placed in 80% ethanol (10.0 mL) at 95°C for 20 min. The collected extracts were placed at 80°C in a drying oven to remove the ethanol, the residues were re-dissolved in 1 ml water. After centrifugation at 12000 g for 15 min, the supernatant was filtered through a 0.45 μm membrane. Amino acid contents were determined using an LA8080 automatic amino acid analyzer (Hitachi, Tokyo, Japan).

To investigate the LHT genes in tobacco, we employed 10 AtLHT protein sequences as queries for BLAST search. As a result, 23 LHT genes were obtained from the tobacco genome. These genes were named in the order of their locations on the chromosomes and scaffolds. The gene and CDS lengths of NtLHTs ranged from 912 to 17381 bp and 912 to 1716 bp, respectively (Table 1). The length of NtLHT proteins ranged from 303 to 571 amino acids with a molecular weight of 33.4–61.9 kDa, and their pIs ranged from 5.74 to 9.51. The predominant amino acid residues were valine, alanine, and leucine. The instability index for most of the proteins (77.8%) were less than 40. Apart from these, the GRAVY values of all NtLHT proteins were positive, indicating that all NtLHT proteins were hydrophobic. The predicted subcellular locations suggested that most of the NtLHT proteins were localized in the plasma membrane, cytoplasm, and vacuoles (Table 2).

To comprehensive understand the information of NtLHT proteins, the transmembrane regions were predicted by PROTTER 1.0. The number of transmembrane regions in NtLHT proteins ranged from 5 to 11 (Supplementary Figure 2). The predicted protein structures for all NtLHT proteins revealed the presence of α helices, random coils, extended strands, and β turns. All NtLHT proteins have α helices, while β turns were the least common (Supplementary Figure 3).

To explore the evolutionary relationships among LHT genes from Tobacco, tea, Arabidopsis, and rice, an unrooted NJ tree was constructed based on alignment of 23 NtLHT genes in tobacco, 7 CsLHT genes in tea, 10 AtLHT genes in Arabidopsis, 6 OsLHT genes in rice. According to the created phylogenetic tree, these LHT genes could be classified to two subgroups. Subgroup I contained AtLHT1/2/3/5/6/8/9/10, OsLHT1/2/3/4, CsLHT1/6, and NtLHT1/3/4/5/6/7/10/11/12/13/14/17/20/21/22/23. Subgroup II contained AtLHT4/7, OsLHT5/6, CsLHT2/3/4/5/7, and NtLHT2/8/9/15/16/18/19 (Figure 1). Therefore, subgroup I had the largest number of LHT members in Arabidopsis (8 genes), rice (4 genes), and tobacco (16 genes), whereas subgroup II had more LHT members from tea (5 genes).

To clarify the evolutionary relationships of NtLHT genes, a phylogenetic tree was generated between NtLHT genes. In concordance with the previous results, all the NtLHT genes were divided into two subgroups. Subgroup I contained 16 NtLHT genes, and subgroup II contained 7 NtLHT genes (Figure 2A). The gene structure analysis of NtLHT was performed by the GSDS program. The number of exons in subgroup I ranged from 0 to 8, while each member of subgroup II contained 5 exons (Figure 2B). The conserved motif analysis of NtLHT proteins was captured by MEME software. Several motifs were widespread among NtLHT proteins, such as motif 1, 2, 3, 4, 8, 9. The number of conserved motifs in subgroup I ranged from 6 to 11, while most of the members in subgroup II, except NtLHT2, contained 10 motifs (Figure 2C). These results indicating that subgroup II members is more conserved than subgroup I.

The chromosomal distribution of NtLHT genes in the tobacco genome were further determined. In the tobacco genome, 11 NtLHT genes were unevenly distributed in 8 chromosomes (Chr). There were three NtLHT genes on Chr 17, two NtLHT genes on Chr 6, Chr 3, 8, 18, 19, 20, and 21 contained one NtLHT gene each. The remaining 12 NtLHT genes could not be mapped on chromosomes but mapped on some scaffolds (Figure 3).

Tandem and segmental duplicates play significant roles for the evolution of species (Xu et al., 2012). Two genes (NtLHT6 and NtLHT7) were tandemly duplicated, while six pairs (NtLHT1/NtLH12, NtLHT5/NtLHT23, NtLHT6/NtLHT10, NtLHT6/NtLHT22, NtLHT8/NtLHT15, and NtLHT14/NtLHT21) were segmental duplicated (Figure 4). The Ka/Ks values of all tandem and segmental duplicated gene pairs were less than 1 (Supplementary Table 4), indicating that these NtLHT genes were evolved under the influence of purifying selection.

We constructed a collinearity plot to investigate the orthologous LHT genes in tobacco, tea, Arabidopsis, and rice (Figure 4). Two collinear gene pairs were identified between tobacco and rice, the collinear relationships between them were many-to-one matches. Three collinear gene pairs were identified between tobacco and Arabidopsis, the collinear relationships between them include many-to-one matches and one-to-one matches. Four collinear gene pairs were identified between tobacco and tea, the collinear relationships between them were one-to-many matches (Supplementary Table 5).

The gene expression profiling in different tissues could provide preliminary clues for their function. To explore the tissue specific expression profiles of NtLHT genes, different tissues from tobacco were analyzed, including root, stem, leaf, flower, and axillary bud. The NtLHT genes showed diverse expression patterns among different tissues. NtLHT6/13/16/18/19 were expressed in all these tissues. Ten NtLHT genes, NtLHT2/3/4/7/10/11/14/20/21/23 showed high expression in flower, among which NtLHT2 also displayed high expression in root and NtLHT23 had high expression in leaf. NtLHT8/9/12/15 had the highest expression levels in root. NtLHT1 and NtLHT22 showed high expression in leaf. Notably, NtLHT5 was highly expressed in axillary bud (Figure 5). The results showed that the expression of NtLHT have a certain preference and most of these genes were highly expressed in flower, root, and leaf.

To explore the regulatory pathways of NtLHT genes, we isolated the 2 kb upstream sequences of the NtLHT genes for cis-element analysis. The largest number of cis-elements identified in the NtLHT genes promoter was involved in light-responsiveness. Moreover, cis-elements associated with anaerobic induction were detected in the promoters of most of the NtLHT genes. Apart from these, cis-elements associated with hormone response (e.g., abscisic acid, auxin, MeJA, salicylic acid, and gibberellins), stress response (e.g., low temperature, defense and stress, and drought), and meristem expression were also detected in the promoter sequences of NtLHT genes (Figure 6). The diversity of cis-elements suggest that NtLHT genes performed multiple functions in physiological and biological processes.

To analyze the roles of NtLHT genes to cold and drought stress, we analyze the expression profiling of NtLHT genes after cold and drought treatment. The expression levels of most NtLHT genes were differently changed under cold or drought stress (Figure 7). Under cold stress, the expression of NtLHT2/6/10/15/17 were significantly upregulated, the expression of NtLHT9/13/19/22 were significantly downregulated. After drought treatment, the expression of NtLHT1/2/3/5/6/8/12/15/23 were significantly upregulated, while the expression of NtLHT13/16/18/19/22 were significantly downregulated. In addition, the expression of NtLHT7 and NtLHT21 were not significantly changed after cold and drought treatment. Furthermore, four genes (NtLHT4/11/14/21) highly expressed in flower were not detected under drought and cold stresses. The results implied that NtLHT genes may play distinctive roles in plant response to environmental abiotic stresses.

To explore the potential functions of the NtLHT genes, the subgroup I member NtLHT22 was selected for subcellular localization analysis. The 35S: NtLHT22-YFP vector and the 35S:YFP control vector were transiently expressed in tobacco protoplasts, subsequently, the subcellular localization of the YFP signal was detected by confocal laser scanning microscope. In control, the YFP signal was spread throughout the whole protoplasts. In contrast, the signal of the YFP protein fused to NtLHT22 was only observed in the plasma membrane, suggesting that NtLHT22 was targeted to the plasma membrane (Figure 8).

To further explore the functions of NtLHT22 in amino acids homeostasis, transgenic plants with overexpressing and knock out of NtLHT22 were generated, respectively. Two overexpression (OE3 and OE6) and two mutant lines (ntlht22-1 and ntlht22-2)were chosen for further experiments (Supplementary Figures 4, 5). In leaf tissues, threonine, glycine, serine, aspartic acid, and alanine were the main amino acids. Compared with WT, the levels of six individual amino acids were significantly increased in ntlht22 mutants, including aspartic acid, serine, phenylalanine, lysine, arginine, and methionine, while the contents of threonine, aspartic acid and glycine were decreased in NtLHT22-OE plants (Figure 9A). In root tissues, the levels of threonine, alanine and serine were decreased in NtLHT22-OE plants than WT. In ntlht22 mutants roots, the levels of alanine was decreased but the levels of aspartic acid and lysine were increased than WT. In addition, the content of arginine in NtLHT22-OE plants and ntlht22 mutants roots were significantly higher than WT (Figure 9B). The amino acids content were significantly reduced in NtLHT22-OE plants than WT, however, total of amino acids were not changed between ntlht22 mutants and WT (Figures 9C,D). Taken together, the amino acids profile was significantly changed in NtLHT22-OE plants and ntlht22 mutants, suggesting that NtLHT22 is participate in amino acids homeostasis in tobacco.