Characterization and Functional Analysis of GhNAC82, A NAM Domain Gene, Coordinates the Leaf Senescence in Upland Cotton (Gossypium hirsutum L.)

In the process of growth and development, cotton exhibits premature senescence under various abiotic stresses, impairing yield and fiber quality. NAC (NAM, ATAF1,2, and CUC2) protein widely distributed in land plants, play the critical role in responding to abiotic stress and regulating leaf senescence. We have identified and functional analyzed a NAM domain gene GhNAC82 in upland cotton, it was located on the A11 chromosome 4,921,702 to 4,922,748 bp, only containing one exon. The spatio-temporal expression pattern analysis revealed that it was highly expressed in root, torus, ovule and fiber development stage. The results of qRT-PCR validated that GhNAC82 negatively regulated by salt stress, drought stress, H2O2 stress, IAA treatment, and ethylene treatment, positively regulated by the ABA and MeJA treatment. Moreover, heterologous overexpression of GhNAC82 results in leaf premature senescence and delays root system development in Arabidopsis thaliana. The phenotype of delayed-senescence was performed after silencing GhNAC82 by VIGS in premature cotton. Taken together, GhNAC82 was involved in different abiotic stress pathways and play important roles in negatively regulating leaf premature senescence.


Introduction
Leaf senescence, an internal factor, is the adaptive mechanism across long evolutionary distances and it is a prevalent phenomenon during plant development [1]. In addition, it is usually taking place in the final stage of plant growth, accompanied by complex and highly coordinated gene regulatory networks [2]. The phenomenon of plants emerges advanced chlorophyll degradation, the photosynthesis rate decreases, the accumulation of reactive oxygen species (ROS), and oxidative enzyme activity reduced in the appropriate growing season, termed premature senescence [3]. The occurrence of premature aging is the result of a combination of multiple factors. Extreme environmental conditions adversely affect the growth of plants and accelerate leaf senescence, including drought, salt, pathogenic bacteria, and osmotic stress, which seriously restricts crop yield and quality [4][5][6][7]. Otherwise, genetic factors also play a crucial role, such as gene mutations leading to the premature-aging. The regulation of leaf senescence involves the dynamic changes of related gene expression and protein synthesis, and the transcription factors (TFs) and the downstream gene promoters cis-elements are combined to regulate the senescence process [8][9][10]. Transcriptional regulation plays a critical role in plant maturation and senescence, and TFs mediate senescence regulatory networks by activating or inhibiting target genes [11].
NAC is a family of plant-specific transcription factors that regulate the normal growth and development of plants under various abiotic stress response and aging process [12,13]. All of the N-terminals of their encoding proteins contain a conserved sequence of about 150 amino acid residues, which plays an important role in plant growth and development [14]. It could be degraded by protease or ubiquitin-proteasome under specific environmental signals, and the homologous or heterodimer of the activated NAC protein combines with the promoter of its target gene to control its expression during the nuclear signal input [15,16]. Overexpression of OsNAC9 and OsNAC10 regulates root architecture to enhance drought resistance and increase the grain yield of rice [17,18]. Similar results were found in Arabidopsis, AtNAC1 and AtNAC2 are induced by auxin to promote lateral root elongation, and both AtNTL8 and AtLOV1 overexpressed showed slow growth and delayed flowering [19][20][21][22]. Involvement of ANAC003, ANAC010, ANAC042, and ANAC075 under dark stress tolerance was corroborated by the phenotype of Arabidopsis tic55-II knock-out (KO) mutants and found the promoter region of ANAC003 bound to MYB108 in yeast one-hybrid (Y1H), taking into account the above indicate that the regulation network of MYB-NAC is involved in the process of aging under dark stress [23]. ANAC072 positively regulates natural and dark-induced leaf senescence through chlorophyll degradation and activating the transcription of NYE1, which is induced by multiple stresses [24].
NAC as a regulatory factor of leaf senescence has been confirmed in the major crops and the regulatory network can be mediated by changes in endogenous hormone levels [25][26][27]. Phytohormones play a key role in response to biotic and abiotic stresses and have been found to involved in the regulation of leaf senescence; specifically, indole-3-acetic acid (IAA), cytokinins (CTK), and gibberellin (GA) are thought to delay leaf senescence, whereas ethylene (ETH), abscisic acid (ABA), methyl jasmonate (MeJA), salicylic acid (SA), brassinosteroids (BR) are thought to induce it. A previous study reported that the content of IAA gradually increases in Arabidopsis leaves during later development stages, indicating auxin is involved in the regulation mechanism of senescence [28]. The content of endogenous cytokinins decreased promotes cell division synergistically with auxin to delay leaf senescence, and cytokinin-mediated AtAHK3 controls aging by phosphorylating AtARR2 [29,30]. The root system also affects leaf senescence in G. hirsutum directly or indirectly via increased the biosynthesis and transport of abscisic acid [31]. Exogenous treatment of methyl jasmonate reduces plant growth in rice by decreasing chlorophyll contents and photosynthetic capacity and induces the expression of OsDOS genes [32]. Plant hormones respond to signaling responses to biotic and abiotic stresses by interacting with or antagonizing NAC transcription factors and their upstream or downstream genes to maintain the stability of the intracellular environment [33]. Likewise, exogenous treatment of plant hormones inducible expression of NAC transcription factor, and the ATAF subfamily positive function in hormones-dependent stress-response pathways, ANAC019 and ANAC055 are negative regulators of stress [18,34].
The completion of genome sequencing has laid the foundation for the research on the regulation of leaf senescence by cotton NAC transcription factors at the genome level [35,36]. Several upland cotton NAC genes (such as GhNAC2, GhNAC4, and GhNAC6) have been isolated in response to abiotic stresses: drought, salt, high and low temperatures by the ABA signal transduction pathway [37,38]. Overexpression of GhNAC79 enhanced the expression level in the cotyledon and fiber development stage significantly improved the resistance to drought stress of upland cotton and A. thaliana [39]. It has been found that seven genes in upland cotton (GhNAC7-GhNAC13) were highly expressed in roots indicating these genes might be involved in responses to abiotic stress during root development [40]. In addition, GhNAC63 and GhNAC78 have already been identified to involved in the regulation of leaf senescence [41,42]. These results all indicated that NAC gene family members play an essential role in the regulation of leaf senescence, stress resistance, and phytohormoneresponsive in cotton. Therefore, it is necessary to unravel the conduction mechanisms of NAC genes in aspects of stress-responsive and leaf senescence, which further improves crop yield and quality.
In this study, we revealed the function and molecular mechanism of GhNAC82 in upland cotton leaf senescence. The expression patterns of GhNAC82 in response to different abiotic stresses and hormonal treatments were analyzed, and the results showed that it is particularly sensitive to changes imposed by exogenous hormones. The overexpression of GhNAC82 in A. thaliana and silence in differentially aging cotton shows that overexpress led to premature aging, silence delayed leaf senescence. These results show the function of GhNAC82 in vegetative growth, and it will provide the theoretical basis for the breeding of new cotton varieties with early maturity but no premature senescence.

Plant Materials, Growth Conditions and Stress Treatments
One the premature senescence upland cotton (cv. CCRI 10) and normal senescence upland cotton (cv. CCRI 50) were used in this study. Plump seeds were selected and cultivated in seedlings pot and grew up in an artificial climate box (16 h light 28 • C/8 h dark 25 • C cycle), normal water and fertilizer management was administrated until the end of the experiment.
Select seedlings with consistent growth for the first sampling when cotyledons spread out. Then the samples were taken every one week to observe the senescence change of the cotyledons until the cotyledons turn completely aged after seven weeks. Selecting the root, stem, leaf, torus, petals, stamen, and pistil form healthy plants, quickly stored in liquid nitrogen. Samples of ovules and fibers were taken at 0, 5, 10, 15, 20, and 25 DPA (days post anthesis), respectively. After the unfolding of the third leaf, seedlings were irrigated with 20% PEG 6000 or 200 mM NaCl at 5-day intervals and sampling at 0, 5, 10, 15, 20, and 25 days after drought/salt treatment. For hormone treatment at the same period, seedlings were sprayed with 20 µM auxin, 50 µM ABA, 200 µM MeJA, and 200 µM ethephon, and controlled with water. The leaf samples were collected at 0, 4, 12, 24, and 48 h after treatment. All of the above samples contain at least 3 biological repetitions.

Gene Cloning and Sequence Analysis
cDNA sequence of GhNAC82 (1047 bp, GenBank accession number: KF669793) was cloned from the leaf of CCRI 10. The basic information and physicochemical properties of the GhNAC82 were predicted by the smart program via the ExPASy Proteomics Server [43]. The gene structure was scaled by Gene Structure Display Server (GSDS, http://gsds.cbi. pku.edu.cn/, accessed on 10 May 2020). Nucleotide sequences of 2000 bp in upstream regions of GhNAC82 were obtained from CottonFGD [44], while performing cis-acting elements analysis by PlantCare [45]. Multi-sequence alignments and phylogenetic analyses of Arabidopsis, G. arboreum, G. raimondii, G. hirsutum, G. barbadense, cocoa, soybean, grape, and sorghum NAC proteins were carried out with the ClustalW [46], phylogenetic tree based on the alignment using the method of Neighbor-Joining by MEGA 7.0 (bootstrap replications set to 1000) [47]. Chromoal localization is illustrated using the MG2C v2.1 (http://mg2c.iask.in/mg2c_v2.0/, accessed on 11 May 2020) based on the specific location of GhNAC82 on the genome. RNA-Seq data was downloaded from Cotton Omics Database (http://cotton.zju.edu.cn/10.rnasearch, accessed on 10 June 2021) to analyze GhNAC82 under different tissues and abiotic stresses (salt, PEG, low and high temperature), the TBtools to visualize it [36,48]. The DNA of Arabidopsis was extracted from the TIANGEN plant genome DNA extraction kit (centrifugal column type, DP305), and the DNA of upland cotton was extracted using the plant DNA pickup kit (YK00A5). RNA extraction was following the manufacturer's protocol for plants rich in polyphenols and amylase total RNA extraction kit Quantitative RT-PCR used ChamQ universal SYBR qPCR Master Mix (Vazyme, Q711), with GhUBQ7 as the reference gene. The running procedure was established as described in the manual, the expansion curve and dissolution curve of real-time PCR were confirmed after the reaction, and the results were analyzed using 2 −∆∆Ct data [49]. The entire trials were performed under low-light conditions, each data contains three technical repetitions and three biological repetitions at least. All primer sequences are listed in Table S1.

Vector Construction and Genetic Transformation of A. thaliana
PCR amplification products of GhNAC82 were inserted into the EcoRI and XbaI sites of the binary vector pBI121 to generate 35S::GhNAC82 plasmids, and the 35S::GhNAC82 recombinant plasmid was introduced into A. thaliana wild type (Col-0) by the Agrobacterium (GV3101). Positive plants were selected on 1/2 MS media containing kanamycin (100 mg/L), and the green plants were transplanted in the seedling pot and further confirmed by PCR and qRT-PCR after 14 days.

Determination of the Related Physiological Parameters
The malondialdehyde (MDA) contents were quantified from 100 mg cotton leaves using an assay kit as the instructions described (Solarbio, BC0025, Beijing, China). The chlorophyll content was determined according to the method as the previous study described [51]. The soluble protein was determined according to the method described by Enzyme Free Company Plant Soluble Protein (sprotein) ELISA kit and Coomassie brilliant blue G250 method.

Identification and Evolution Analysis of the GhNAC82
NAC transcription factor family exerted a vital regulatory effect on leaf senescence, stress resistance, and phytohormone-responsive in cotton. GhNAC82 was selected for further functional studies. GhNAC82 contained only one exon, the coding sequences (CDS) with the same nucleotide composition as genomic DNA (gDNA) sequences, and it was isolated from the cotton variety CCRI 10. The full-length cDNA includes a 1047 bp open reading frame ORF encoding a 348 amino acid protein with a molecular weight (MW) is 38.727 kDa. It was found to be localized on chromosome A11: 4,921,702-4,922,748 bp.
According to the domain scrutiny and relationships analysis, 27 proteins were identified from upland cotton (Figure 1a). Of these, the closest relative to GhNAC82 (Gh_A11G0524) was another copy on the D subgroup (Gh_D11G0608), followed by GhNAC17 (Gh_D02G0682, Gh_A02G0634, Gh_A05G2095, Gh_D05G2348). Physicochemical properties and gene information of these NAC proteins in G. hirsutum are presented in Table S2.
According to the domain scrutiny and relationships analysis, 27 proteins were identified from upland cotton (Figure 1a). Of these, the closest relative to GhNAC82 (Gh_A11G0524) was another copy on the D subgroup (Gh_D11G0608), followed by GhNAC17 (Gh_D02G0682, Gh_A02G0634, Gh_A05G2095, Gh_D05G2348). Physicochemical properties and gene information of these NAC proteins in G. hirsutum are presented in Table S2.  Figure 1b). The phylogenetic analysis characterized the closest evolutionary relationship to GhNAC82 (Gh_A11G0524) were GOBAR_AA17024 from G. barbadense A subgenome and Ga11G347 from the diploid ancestor G. arboreum. Meanwhile, the closest evolutionary relationship to Gh_D11G0608 (another homologous gene in D genome of upland cotton) was GOBAR_DD27691 from G. barbadense D subgenome and Gorai.007G065300 from the diploid ancestor G. raimondii. In the evolutionary between cotton and other representative species, the closest phylogenetic relationship was T. cacao (TcNAC082), secondly is A. thaliana (AT5G09330 and AT5G64060), followed by G. max (Gm_05G191300), S. bicolor (SORBI_3007G047700), O. sativa (Os08g0157900), and V. vinifera (VIT_18s0089g01120). These genes may have a similar biological function to GhNAC82, and might be the connection to leaf senescence and stress tolerance.

Predicted of Cis-Elements in Promoter Region of GhNAC82
The cis-regulatory elements in the promoter usually regulate gene expression and respond to environmental and growth factors [52]. Cis-acting regulatory elements were analyzed in the GhNAC82 upstream 2000 bp promoter region by bioinformatics using the With MEGA 7.0 software construct a phylogenetic tree from the aligned sequences, showing evolutionary relationships between GhNAC82 and other homologous genes from upland cotton and representative plant species (including G. arboreum, G. raimondii, G. hirsutum, G. barbadense, A. thaliana, Theobroma cacao, Oryza sativa, Glycine max, Vitis vinifera, and Sorghum bicolor, Figure 1b). The phylogenetic analysis characterized the closest evolutionary relationship to GhNAC82 (Gh_A11G0524) were GOBAR_AA17024 from G. barbadense A subgenome and Ga11G347 from the diploid ancestor G. arboreum. Meanwhile, the closest evolutionary relationship to Gh_D11G0608 (another homologous gene in D genome of upland cotton) was GOBAR_DD27691 from G. barbadense D subgenome and Gorai.007G065300 from the diploid ancestor G. raimondii. In the evolutionary between cotton and other representative species, the closest phylogenetic relationship was T. cacao (TcNAC082), secondly is A. thaliana (AT5G09330 and AT5G64060), followed by G. max (Gm_05G191300), S. bicolor (SORBI_3007G047700), O. sativa (Os08g0157900), and V. vinifera (VIT_18s0089g01120). These genes may have a similar biological function to GhNAC82, and might be the connection to leaf senescence and stress tolerance.

Predicted of Cis-Elements in Promoter Region of GhNAC82
The cis-regulatory elements in the promoter usually regulate gene expression and respond to environmental and growth factors [52]. Cis-acting regulatory elements were analyzed in the GhNAC82 upstream 2000 bp promoter region by bioinformatics using the PlantCARE database. As shown in Figure 2, the promoter sequence of GhNAC82 contains the basal cis-elements TATA-box and CAAT-box. The cis-elements involved in plant growth and development, stress response, and hormone response were found. Among these cis-elements, including the light responded elements: such as ACE, AE-box, ATCT-motif, Box 4, MRE, and TCCC-motif. Furthermore, the binding site of AT-rich DNA binding protein (ATBP-1, AT-rich element), endosperm expression elements (GCN4_motif), and the anaerobic induction element (ARE) was found to be related to plant growth and development. It was also found that the stress-related elements contained low-temperature  Table  S3). These results suggest that the GhNAC82 may be involved in the stress responses of plant growth and stress.
the basal cis-elements TATA-box and CAAT-box. The cis-elements involved in growth and development, stress response, and hormone response were found. A these cis-elements, including the light responded elements: such as ACE, AE-box, A motif, Box 4, MRE, and TCCC-motif. Furthermore, the binding site of AT-rich DNA ing protein (ATBP-1, AT-rich element), endosperm expression elements (GCN4_m and the anaerobic induction element (ARE) was found to be related to plant growt development. It was also found that the stress-related elements contained low-tem ture responsiveness (LTR). Phytohormone correlated elements include MeJA (CG motif, TGACG-motif), gibberellin (GA, TATC-box), and salicylic acid (SA, TCA-elem In addition, MYC and MYB-binding sites involved in dehydration-responsive, the ements involved in wound-responsiveness, such as W-box, WRE3, and WUN-moti ure 2, Table S3). These results suggest that the GhNAC82 may be involved in the responses of plant growth and stress.

Expression Patterns Analysis of GhNAC82 in Various Tissues
Spatiotemporal expression files of GhNAC82 were appraised in seven differe sues (including root, stem, leaf, torus, petals, stamen, and pistil) and development p (including 0-, 5-, 10-, 20-, and 25-DPA ovule; 10-, 20-, and 25-DPA fiber) using RN data from G. hirsutum. It was found that GhNAC82 was highly expressed in root, p 10 DPA ovule and fiber, 20 DPA fiber, and 25 DPA fiber. Secondly, the lower expr in torus, 5 DPA ovule, and 20 DPA ovule. The results of the gene expressed in ovu fiber suggested that it may play a key role in regulating expression during embryon fiber development.
To detect the expression patterns of GhNAC82 in different tissues, the roots, and leaves in the seedling stage; torus, petals, stamen, and pistil in the flowering p were collected. At the same time, ovules in 0 DPA, 5 DPA, 10 DPA, 20 DPA, 25 DPA fibers in 10 DPA, 20 DPA, and 25 DPA were collected. The pattern of gene expr assessed by qPCR was basically consistent with transcriptome profiling data, as sho Figure 3. However, it demonstrated different patterns, it was found that GhNAC8 expressed in stem, leaf, stamen, pistil, 0 DPA ovule, and 25 DPA ovule using rea PCR. Overall, these results indicated that GhNAC82 may play a critical regulatory r root growth and development.

Expression Patterns Analysis of GhNAC82 in Various Tissues
Spatiotemporal expression files of GhNAC82 were appraised in seven different tissues (including root, stem, leaf, torus, petals, stamen, and pistil) and development period (including 0-, 5-, 10-, 20-, and 25-DPA ovule; 10-, 20-, and 25-DPA fiber) using RNA-seq data from G. hirsutum. It was found that GhNAC82 was highly expressed in root, petals, 10 DPA ovule and fiber, 20 DPA fiber, and 25 DPA fiber. Secondly, the lower expression in torus, 5 DPA ovule, and 20 DPA ovule. The results of the gene expressed in ovule and fiber suggested that it may play a key role in regulating expression during embryonic and fiber development.
To detect the expression patterns of GhNAC82 in different tissues, the roots, stems, and leaves in the seedling stage; torus, petals, stamen, and pistil in the flowering period were collected. At the same time, ovules in 0 DPA, 5 DPA, 10 DPA, 20 DPA, 25 DPA, and fibers in 10 DPA, 20 DPA, and 25 DPA were collected. The pattern of gene expression assessed by qPCR was basically consistent with transcriptome profiling data, as shown in Figure 3. However, it demonstrated different patterns, it was found that GhNAC82 was expressed in stem, leaf, stamen, pistil, 0 DPA ovule, and 25 DPA ovule using real-time PCR. Overall, these results indicated that GhNAC82 may play a critical regulatory role in root growth and development.

Phenotypic Changes and Expression Patterns of GhNAC82 during Cotyledon Senescence
During the leaf grew and developed, chlorophyll contents and photosynthetic efficiency were gradually decreased as mature leaves yellowing. As shown in Figure 4a, with the growth and development, cotyledon color was progressively changed and CCRI 10 was significantly higher than CCRI 50 based on teh degree of yellowing. The cotyledon of CCRI 10 was obviously yellowing on 49 days, completely yellow even withered on 56 days, while the CCRI 50 only a small section of the cotyledon was with brown spots or shrinking. Figure 4b shows that GhNAC82 was highest expressed in the 21 days after cotyledons unfolding of CCRI 10. Subsequently, the expression gradually decreased. From the patterns of GhNAC82 during cotyledon senescence, we presumed that GhNAC82 likely has an important role in the early cotyledon development stage. Differing from the patterns in CCRI 10, the expression of GhNAC82 was relatively gentle in CCRI 50, suggesting that elevated GhNAC82 expression may be responsible for premature senescence.

Phenotypic Changes and Expression Patterns of GhNAC82 during Cotyledon Senescence
During the leaf grew and developed, chlorophyll contents and photosynthetic efficiency were gradually decreased as mature leaves yellowing. As shown in Figure 4a, with the growth and development, cotyledon color was progressively changed and CCRI 10 was significantly higher than CCRI 50 based on teh degree of yellowing. The cotyledon of CCRI 10 was obviously yellowing on 49 days, completely yellow even withered on 56 days, while the CCRI 50 only a small section of the cotyledon was with brown spots or shrinking. Figure 4b shows that GhNAC82 was highest expressed in the 21 days after cotyledons unfolding of CCRI 10. Subsequently, the expression gradually decreased. From the patterns of GhNAC82 during cotyledon senescence, we presumed that GhNAC82 likely has an important role in the early cotyledon development stage. Differing from the patterns in CCRI 10, the expression of GhNAC82 was relatively gentle in CCRI 50, suggesting that elevated GhNAC82 expression may be responsible for premature senescence. The content of chlorophyll, malonaldehyde (MDA), and soluble protein are the integral part and indicators of the degree of leaf senescence [53]. Chlorophyll is the main pigment for photosynthesis and is closely related to the aging process in plant growth and development. Malondialdehyde (MDA) is the oxidation final product of free radicals acting on the lipid with peroxidation reaction in the organism, which is cytotoxic and is closely related to the stress-related response [54]. Soluble protein content increased in the stressed seedlings stage, and it also was a key abiotic stress indicator in plants [55]. As Figure 4c shows, the chlorophyll content has peaked on 14 and 35 days of cotyledon development, with an overall, showed downward trend. In addition, the chlorophyll in CCRI 50 exhibited higher content than CCRI 10 after 21 days. The MDA content in CCRI 10 and CCRI 50 showed a uniform upward trend, then the MDA content of CCRI 10 is peaked at 28 days, and CCRI 50 is peaked at 35 days but the increasing trend was relatively flat in CCRI 50. After reaching the peak, followed by a slow downward trend (Figure 4d). The content of soluble protein in CCRI 10 always remained higher than CCRI 50 after 14 days, and it showed two peaks at 21 days and 35 days (Figure 4e). It was coincided with the relative expression of GhNAC82 and chlorophyll content among different cultivars and indicated the content of soluble protein is strongly correlated with the degree of leaf senescence. Both results suggest that probably exist a potential regulatory mechanism is via the expression of GhNAC82 mediates the physiological indicators to regulate the leaf senescence.  The content of chlorophyll, malonaldehyde (MDA), and soluble protein are the integral part and indicators of the degree of leaf senescence [53]. Chlorophyll is the main pigment for photosynthesis and is closely related to the aging process in plant growth and development. Malondialdehyde (MDA) is the oxidation final product of free radicals acting on the lipid with peroxidation reaction in the organism, which is cytotoxic and is closely related to the stress-related response [54]. Soluble protein content increased in the stressed seedlings stage, and it also was a key abiotic stress indicator in plants [55]. As Figure 4c shows, the chlorophyll content has peaked on 14 and 35 days of cotyledon development, with an overall, showed downward trend. In addition, the chlorophyll in CCRI 50 exhibited higher content than CCRI 10 after 21 days. The MDA content in CCRI 10 and CCRI 50 showed a uniform upward trend, then the MDA content of CCRI 10 is peaked at 28 days, and CCRI 50 is peaked at 35 days but the increasing trend was relatively flat in CCRI 50. After reaching the peak, followed by a slow downward trend (Figure 4d). The content of soluble protein in CCRI 10 always remained higher than CCRI 50 after 14 days, and it showed two peaks at 21 days and 35 days (Figure 4e). It was coincided with the relative expression of GhNAC82 and chlorophyll content among different cultivars and indicated the content of soluble protein is strongly correlated with the degree of leaf senescence. Both results suggest that probably exist a potential regulatory mechanism is via the expression of GhNAC82 mediates the physiological indicators to regulate the leaf senescence.

Expression of the GhNAC82 in Leaves Was Significantly Induced by Abiotic Stresses
To investigate whether are responsive regulatory proteins under abiotic stresses in different degrees of prematurity species, NaCl, PEG, and H 2 O 2 treatment of the CCRI 10 and CCRI 50 cotton seedlings were performed. As shown in Figure 5a,e,i, the expression levels of GhNAC82 significantly changed after salt, drought, and H 2 O 2 treatment, and the expression in CCRI 50 was higher than in CCRI 10 all the time (except for the early H 2 O 2 -stress, 0 h and 2 h).
The two cotton species (phenotype of premature senescence and non-premature senescence) showed a peak at the time 10 days after treatment and gradually decreased subsequently. A similar trend was observed in the changes in chlorophyll contents with expression pattern (Figure 5b). Malondialdehyde content was increased gradually after the peak time of 10 days (Figure 5c). After drought stress, the expression of GhNAC82, chlorophyll and MDA contents exhibited a continuously rising trend (Figure 5e-g). H 2 O 2treated plants show expression peaked at 12 h, and decline followed (Figure 5i). Chlorophyll content in leaves under H 2 O 2 stress reached a maximum in 12 h too (Figure 5j). MDA contents peaked relatively early under the H 2 O 2 treatment condition, and it appeared to maximum advanced one stage (at 8 h, Figure 5k). All the soluble protein content under abiotic stress conditions exhibit decreased slowly with treatment time (Figure 5d,h,l). From the above results, we can conclude that the GhNAC82 gene is involved in multiple abiotic stress responses, and its' expression may be negatively regulated by salt, drought, and H 2 O 2 treatment. In addition, chlorophyll and MDA content are tightly linked to the expression of GhNAC82, soluble protein content showed a dynamic decrease as the abiotic stress progressed.

Expression of the GhNAC82 in Leaves Was Significantly Induced by Abiotic Stresses
To investigate whether are responsive regulatory proteins under abiotic stresses in different degrees of prematurity species, NaCl, PEG, and H2O2 treatment of the CCRI 10 and CCRI 50 cotton seedlings were performed. As shown in Figure 5a,e,i, the expression levels of GhNAC82 significantly changed after salt, drought, and H2O2 treatment, and the expression in CCRI 50 was higher than in CCRI 10 all the time (except for the early H2O2stress, 0 h and 2 h). The two cotton species (phenotype of premature senescence and non-premature senescence) showed a peak at the time 10 days after treatment and gradually decreased subsequently. A similar trend was observed in the changes in chlorophyll contents with expression pattern (Figure 5b). Malondialdehyde content was increased gradually after the peak time of 10 days (Figure 5c). After drought stress, the expression of GhNAC82, chlorophyll and MDA contents exhibited a continuously rising trend (Figure 5e-g). H2O2treated plants show expression peaked at 12 h, and decline followed (Figure 5i). Chlorophyll content in leaves under H2O2 stress reached a maximum in 12 h too (Figure 5j). MDA contents peaked relatively early under the H2O2 treatment condition, and it appeared to maximum advanced one stage (at 8 h, Figure 5k). All the soluble protein content under abiotic stress conditions exhibit decreased slowly with treatment time (Figure 5d,h,l). From the above results, we can conclude that the GhNAC82 gene is involved in multiple abiotic stress responses, and its' expression may be negatively regulated by salt, drought, and H2O2 treatment. In addition, chlorophyll and MDA content are tightly linked to the expression of GhNAC82, soluble protein content showed a dynamic decrease as the abiotic stress progressed.

The Expression Patterns of GhNAC82 under Different Hormones
Different hormone treatments were performed for cotton seedlings at CCRI 10 and CCRI 50, respectively, and the expression of GhNAC82 was detected during different treatment periods. The expression level of GhNAC82 changed significantly in response to phytohormone treatments.
In the MeJA treatment initial periods (2 h after treatment), there was induction reaching a significance difference, and the expression of GhNAC82 in CCRI 10 increased to approximately 3 times. After that, the expression falls rapidly, 24 h a second peak was reached (Figure 6a). However, only the expression peaked at 24 h, and the expression was lower than CCRI 50 all the time. A trend of first rising and then falling was seen in soluble protein content, but the content of chlorophyll and malondialdehyde did not show a clear trend (Figure 6b-d).
The expression pattern of GhNAC82 in CCRI 10 demonstrated similar trends under ABA treatment compared with MeJA treatment, and it was peaked at 8 h after treatment (Figure 6e). The content of chlorophyll and malondialdehyde also reached a peak at 8h after treatment (Figure 6f,g). Although the soluble content was on the rise and peaked at 24 h, there is no obvious trend (Figure 6h). ing a significance difference, and the expression of GhNAC82 in CCRI 10 increased to approximately 3 times. After that, the expression falls rapidly, 24 h a second peak was reached (Figure 6a). However, only the expression peaked at 24 h, and the expression was lower than CCRI 50 all the time. A trend of first rising and then falling was seen in soluble protein content, but the content of chlorophyll and malondialdehyde did not show a clear trend (Figure 6b-d). The expression pattern of GhNAC82 in CCRI 10 demonstrated similar trends under ABA treatment compared with MeJA treatment, and it was peaked at 8 h after treatment (Figure 6e). The content of chlorophyll and malondialdehyde also reached a peak at 8h after treatment (Figure 6f,g). Although the soluble content was on the rise and peaked at 24 h, there is no obvious trend (Figure 6h). In contrast to the expression trend in MeJA and ABA response, the expression levels under IAA and ethylene treatment of GhNAC82 in CCRI 50 exhibited higher than in CCRI 10. This pattern of expression is similar to that described in abiotic stresses ( Figure 5). Meanwhile, a peak expression under IAA and ethylene at 12 h and 8 h, respectively (Figure 6i,m). The content of chlorophyll and MDA also showed a trend of the first decline and then rise, timing to reach peak emergence broadly matched with the expression peak appeared (Figure 6j,k,n,o). The trend of changes in soluble protein is basically to decline (Figure 6l,p). These results all indicated that the induction expression of GhNAC82 in phytohormonetreated, and was positively induced by MeJA and ABA treatment, negatively induced by IAA and ethylene treatment. Furthermore, the changes in chlorophyll and MDA content coincided with the expression patterns of GhNAC82. Additionally, the changes in soluble protein content basically displayed a decreasing trend under various hormone treatments.

Overexpression of GhNAC82 in Transgenic Arabidopsis Promotes Leaf Senescence
According to the agarose gel electrophoresis (Figure 7a), showed a high band brightness, and the length of the cloned GhNAC82 gene CDS was 1047 bp, consistent with the sequencing gene CDS fragment size. Under normal growth conditions, the expression levels of senescence-related marker genes involved in wild type and overexpressed GhNAC82 transgenic A. thaliana were detected, respectively. The internal reference primers (AtActin2) and other primers were presented in Table S1. As shown in Figure 7b, the overexpression-GhNAC82 transgenic plants were exhibited a shorter root length. According to the leaf senescence phenotype characterized by the wild type and transgenic type Arabidopsis, GhNAC82-overexpressing plants showed an earlier senescence phenotype than the wildtype ones (Figure 7c). The expression level of the GhNAC82 gene in the overexpressed transgenic plants was significantly higher than that in the wild type. In addition, the expression of AtUBQ7 and AtCAT2 (senescence-related gene) was higher than that in the wild type too, but the expression of AtSAG12 was lower than wild type (Figure 7d). The data above indicate that the transgenic type plants' root system grew slowly and promoted early leaf senescence compared with the wild type.

Virus-Induced Gene Silencing of GhNAC82 Exhibited Stronger Vegetative Growth
The GhNAC82 gene was silenced in CCRI 10 and CCRI 50 using virus-induced gene silencing (VIGS). In addition, the expression level of GhNAC82 was detected by qRT-PCR after the albino phenotype appeared on the leaves of positive control material PYL156::GhPDS. As shown in Figure 8, after the virus infected cotton seedlings, the expression level of the GhNAC82 gene in both CCRI 10 and CCRI 50 decreased to 1/20-1/10 times of negative control. It was indicating that VIGS effectively inhibited the expression of GhNAC82. The expression levels of other senescence related marker genes in CCRI 50 were higher than that in CCRI 10, and the expression levels of GhNAC82 in negative control plants were higher than in GhNAC82-silenced plants. The expression levels of other senescence-related genes in negative control plants were lower than those in silenced plants, which may be the result of other senescence genes playing more roles after silencing GhNAC82. Compared with negative control plants, GhNAC82-silenced expression increased plant height in CCRI 10, but there were no noticeable changes in CCRI 50.

Virus-Induced Gene Silencing of GhNAC82 Exhibited Stronger Vegetative Growth
The GhNAC82 gene was silenced in CCRI 10 and CCRI 50 using virus-induced gene silencing (VIGS). In addition, the expression level of GhNAC82 was detected by qRT-PCR after the albino phenotype appeared on the leaves of positive control material PYL156::GhPDS. As shown in Figure 8, after the virus infected cotton seedlings, the expression level of the GhNAC82 gene in both CCRI 10 and CCRI 50 decreased to 1/20-1/10 times of negative control. It was indicating that VIGS effectively inhibited the expression of GhNAC82. The expression levels of other senescence related marker genes in CCRI 50 were higher than that in CCRI 10, and the expression levels of GhNAC82 in negative control plants were higher than in GhNAC82-silenced plants. The expression levels of other senescence-related genes in negative control plants were lower than those in silenced plants, which may be the result of other senescence genes playing more roles after silencing GhNAC82. Compared with negative control plants, GhNAC82-silenced expression increased plant height in CCRI 10, but there were no noticeable changes in CCRI 50.

Discussion
Abiotic stress affects the productivity of major agricultural crops worldwide. NAC transcription factor is the largest family of plant-specific transcription factors that play crucial roles in plant growth and developmental processes, withstand abiotic stress, and delayed senescence [56]. It is distribution widespread in bryophytes to dicots genome, received much attention as an important regulator in gene function and regulation of expression [57,58]. In our study, we identified the orthologous genes in GhNAC82 upland cotton and other species, which found the closest phylogenetic relationship genes was from T. cacao, second, is A. thaliana, followed by G. max, S. bicolor, O. sativa, and V. vinifera (Figure 1). The relatedness among the paralogous gene pairs was similar to the phylogenetic relatedness between species. This illustrates that these genes may play similar roles to leaf senescence and abiotic stress response. Meanwhile, the GhNAC82 promoter regions have multiple growth and development elements, abiotic stress-responsive elements, light-and phytohormone-responsive elements (Figure 2). The cis-elements of the upstream promoter at the gene transcription start site regulate the expression of GhNAC82 in response to various abiotic stresses and hormone treatment.

Discussion
Abiotic stress affects the productivity of major agricultural crops worldwide. NAC transcription factor is the largest family of plant-specific transcription factors that play crucial roles in plant growth and developmental processes, withstand abiotic stress, and delayed senescence [56]. It is distribution widespread in bryophytes to dicots genome, received much attention as an important regulator in gene function and regulation of expression [57,58]. In our study, we identified the orthologous genes in GhNAC82 upland cotton and other species, which found the closest phylogenetic relationship genes was from T. cacao, second, is A. thaliana, followed by G. max, S. bicolor, O. sativa, and V. vinifera (Figure 1). The relatedness among the paralogous gene pairs was similar to the phylogenetic relatedness between species. This illustrates that these genes may play similar roles to leaf senescence and abiotic stress response. Meanwhile, the GhNAC82 promoter regions have multiple growth and development elements, abiotic stress-responsive elements, light-and phytohormone-responsive elements (Figure 2). The cis-elements of the upstream promoter at the gene transcription start site regulate the expression of GhNAC82 in response to various abiotic stresses and hormone treatment.
We found that the expression of GhNAC82 was significantly induced for abiotic stresses and hormone treatment, which were negatively regulated by salt stress, drought stress, H 2 O 2 stress, MeJA, and ABA treatment, positively regulated by IAA and ethylene treatment (Figures 5 and 6). This finding was in agreement with the previous study on other plants. Overexpression of SiNAC1 in Arabidopsis regulated the abscisic acid (ABA) accumulation to promote natural leaf senescence [59]. Ectopic expression of OsNAC2 accelerated leaf senescence by up-regulated the expression levels of the abscisic acid (ABA) biosynthesis genes (OsNCED3 and OsZEP1) and abscisic acid catabolism gene OsABA8ox1 [60]. Similarly, VviNAC33 has been noted to promote leaf senescence by inhibiting the relative expression of auxin transport-related gene PIN1 [61]. Apart from the above, the contradictory results among stress treatments and some phytohormone treatments (i.e., salt and ethylene treatment). Plants respond and adapt to stress by upregulating ethylene signal as the negative feedback regulators, this is consistent with our findings under ethylene treatment.
Previously reported that the NAC transcription factors regulate the formation of lateral roots, AtNAC2 promotes lateral root development and inhibited the primary root growth to mitigate damage in A. thaliana [62]. In our study, transcriptome data and qRT-PCR results showed that GhNAC82 was highest expressed in root, suggesting that GhNAC82 might be related to root system development to regulated abiotic stress response ( Figure 3). In parallel, we have generated the transgenic Arabidopsis overexpressing GhNAC82, and transgenic plants demonstrated a shorter primary root compared to wild-type (Figure 7). This result is basically consistent with the results of the previous study in Arabidopsis, Solanum tuberosum, and Zea mays [59,63]. In addition, AtORS1-overexpression (AtORS1: NAC transcription factor) has been shown to promote premature senescence, and it was up-regulated the expression of multiple genes related to salt and H 2 O 2 stress [64]. Combined the results of GhNAC82-silenced expression increased plant height and delayed senescence indicating GhNAC82 is a negative regulator of leaf senescence. The combination of these results and our study illustrating NAC transcription factor family member GhNAC82 plays an important role in the regulation of leaf senescence, hormone, and abiotic stresses responses.
Similar to in other plants, premature senescence in cotton is a complex polygenic trait, and it is linked to multiple signal transduction pathways. Based on available studies of premature senescence, regulating expression levels of NAC genes to achieve elite germplasm resources that are tolerant to multiple abiotic stresses using gene engineering method and genome wide association studies (GWAS) [65]. In parallel, investigating the molecular mechanisms of senescence to develop and grow normally. In summary, elucidation of the functions of GhNAC82 has important implications to develop crop precision breeding.

Conclusions
In this study, we performed the protein sequences and evolutionary relationships analysis of GhNAC82 and orthologous genes from upland cotton and other species by taking advantage of the published genome data to date. The expression patterns of GhNAC82 under various tissues and development stages, different abiotic stresses, and phytohormonetreated were carried out using qRT-PCR. It was found that GhNAC82 was higher expressed in root, torus, 10 DPA ovules, and fibers in 10 DPA, 20 DPA, and 25 DPA. In addition, these results indicate that GhNAC82 is negatively regulated by salt stress, drought stress, H 2 O 2 stress, IAA treatment, and ethylene treatment. Meanwhile, it is positively regulated by the ABA and MeJA treatment. Expression patterns change of GhNAC82 in response to phytohormone treatments and abiotic stress, the relevant physiological indices are highly correlated mutually. The phenotype of shorter root length and premature senescence was exhibited in the 35S::GhNAC82 overexpressing Arabidopsis compared with wild-type. Lines of GhNAC82-VIGS (virus-induced gene silencing) were showed stronger vegetative growth compared to the negative control in premature senescence cotton. Based on the