The auxin response factor (ARF) gene family in Cyclocarya paliurus: genome-wide identification and their expression profiling under heat and drought stresses

Abstract

Auxin response factors (ARFs), as the main components of auxin signaling, play a crucial role in various processes of plant growth and development, as well as in stress response. So far, there have been no reports on the genome-wide identification of the ARF transcription factor family in Cyclocarya paliurus, a deciduous tree plant in the family Juglaceae. In this study, a total of 34 CpARF genes were identified based on whole genome sequence, and they were unevenly distributed on 16 chromosomes, with the highest distribution on chromosome 6. Domain analysis of CpARF proteins displayed that 31 out of 34 CpARF proteins contain a typical B3 domain (DBD domain), except CpARF12/ CpARF14/CpARF31, which all belong to Class VI. And 20 CpARFs (58.8%) contain an auxin_IAA binding domain, and are mainly distributed in classes I, and VI. Phylogenetic analysis showed that CpARF was divided into six classes (I–VI), each containing 4, 4, 1, 8, 4, and 13 members, respectively. Gene duplication analysis showed that there are 14 segmental duplications and zero tandem repeats were identified in the CpARF gene family of the C. paliurus genome. The Ka/Ks ratio of duplicate gene pairs indicates that CpARF genes are subjected to strong purification selection pressure. Synteny analysis showed that C. paliurus shared the highest homology in 74 ARF gene pairs with Juglans regia, followed by 73, 51, 25, and 11 homologous gene pairs with Populus trichocarpa, Juglans cathayensis, Arabidopsis, and rice, respectively. Promoter analysis revealed that 34 CpARF genes had cis-elements related to hormones, stress, light, and growth and development except for CpARF12. The expression profile analysis showed that almost all CpARF genes were differentially expressed in at least one tissue, and several CpARF genes displayed tissue-specific expression. Furthermore, 24 out of the 34 CpARF genes have significantly response to drought stress (P < 0.05), and most of them (16) being significantly down-regulated under moderate drought treatment. Meanwhile, the majority of CpARF genes (28) have significantly response to drought stress (P < 0.05), and most of them (26) are significantly down-regulated under severe drought treatment. Furthermore, 32 out of the 34 CpARF genes have significantly response to high, middle, and low salt stress under salt treatment (P < 0.05). Additionally, subcellular localization analysis confirmed that CpARF16 and CpARF32 were all localized to nucleus. Thus, our findings expand the understanding of the function of CpARF genes and provide a basis for further functional studies on CpARF genes in C. paliurus.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12298-024-01474-1.

Introduction

Cyclocarya paliurus, a deciduous tree plant belonging to the Juglandaceae family, is known as the cash cow tree due to its fruit shape resembling copper coins. It is also called sweet tea tree since its leaves have a sweet taste (Shu et al. 1995; Zhao et al. 2022). C. paliurus is mainly distributed in 420–2500 m highland regions of the south, central, and southwest of China (Li et al. 2017; Fan et al. 2013). The bark, roots, and leaves of C. paliurus have the effects of killing insects and itching, reducing inflammation and pain, and dispelling wind (Chen et al. 2022). The C. paliurus have attracted more attention in food science and functional nutrition for its amazing effect on heat-clearing, and reducing blood pressure, and blood glucose (Shen et al. 2023). The chemical studies showed that C. paliurus contains multiple bioactive ingredients, such as polysaccharides, flavonoids, triterpenoids, and amino acids, which contribute to various diseases, including anti-hyperglycemia, anti-hyperlipidemia, anti-hypertension, anti-oxidant, anti-tumor, and anti-microbial activities (Li et al. 2011; Zhao et al. 2022). Polysaccharides are the main effective ingredients in the leaves of C. paliurus. Studies have shown that C. paliurus polysaccharides alleviate the symptoms of type II diabetes by increasing certain bacteria that produce short-chain fatty acids in the intestine, promoting the production of short-chain fatty acids and upregulating glucagon-like peptides and peptide tyrosine-related sensory mediators (Yao et al. 2020). In addition, C. paliurus also plays an important role in the care of land pollution. It has been found that it can grow normally when planted on lead-contaminated land with a lead concentration of less than 2,000 mg/kg. Since Pb binds to carbonates in the body of C. paliurus in the form of compounds, the risk of lead leaching is significantly reduced. Therefore, it can be successfully applied as a plant management plant for the protection of lead-contaminated soils (Feng et al. 2021). Previous studies demonstrated that C. paliurus can adapt to adverse abiotic conditions, such as salt and drought stress, through its own morphological and physiological responses (Feng et al. 2020; Zhang et al. 2021; Li et al. 2023). It is demonstrated that the flavonoid content of C. paliurus leaves were increased under salt stress (Zhang et al. 2021). C. paliurus is regarded as an endangered plant under China national key protection. So, the study of salt and drought stress of C. paliurus is crucial for the growth, development, and quality formation of C. paliurus.

As one of the most important hormones in plants, auxin plays critical roles in a variety of physiological processes, such as growth and development, physiological processes, and stress response, etc. (Weijers and Wagner. 2016). The auxin signaling pathway is mainly controlled by two transcription factor gene families: ARF and auxin/indoleacetic acid (AUX/IAA) gene family (Weijers and Wagner. 2016). ARF binds to AUX/IAA proteins to form heterodimers, and its function is regulated by Aux/IAA repressors, which are targeted for degradation during auxin signal perception (Ulmasov et al. 1999). A typical ARF protein contains three conserved domains: an N-terminal DNA-binding domain (DBD), a middle region (MR), and a C-terminal interaction domain (CTD) which is similarly found in the C-terminus of Aux/IAAs (Chandler 2016). The DBD consists B3 DNA-binding domain and regulates auxin-responsive gene expression by binding to the auxin-responsive elements (AuxREs, sequence TGTCTC) located in the promoter regions (Guilfoyle and Hagen. 2007; Roosjen et al. 2018). The CTD domain contains two motifs III and motif IV that mediate the homodimerization or heterodimerization of ARF and inhibit the binding to the auxin-responsive elements at low auxin concentration levels (Roosjen et al. 2018).

Numerous investigations have revealed that ARF genes play critical roles in various growth and development processes, such as the control of floral organogenesis and pattern (Harper et al. 2000; Zhang et al. 2020a, b; Chen et al. 2021a), flower senescence and abscission (Ellis et al. 2005), lateral root formation (Okushima et al. 2007; Ren et al. 2017; Kirolinko et al. 2021); formation of epidermal cells and trichomes (Zhang et al. 2015); organogenesis (Chung et al. 2019); fruit development (Sagar et al. 2013; Yuan et al. 2018; Hu et al. 2023). Furthermore, the ARF genes are also involved in plant responses to numerous biotic and abiotic stress responses (Caruana et al. 2020; Kalve et al. 2020; Chen et al. 2021b; Shen et al. 2022; El Mamoun et al. 2023). For example, MdARF17 was negatively regulated during drought response, and involved in positive Mdm-miR160-MdARF17-MdHYL1 feedback loop of drought tolerance in apple (Shen et al. 2022). MdARF19-2 positively regulate the drought resistance in apple (Jiang et al. 2023). While, the SlARF2 gene was negatively regulated during salt and drought stress response in tomato (El Mamoun et al. 2023).

Since the important roles of ARF transcription factors in plants, the research on the ARF gene family has attracted increasing attention, and dozens of ARF family genes in plant species have been identified and analyzed, such as Arabidopsis (Okushima et al. 2007), rice (Wang et al. 2007), maize (Xing et al. 2011), physic nut (Tang et al. 2018), sorghum (Chen et al. 2019), sweet potato (Pratt and Zhang. 2021), pigeonpea (Arpita et al. 2023), Linum usitatissimum (Qi et al. 2023), Populus trichocarpa (Liu et al 2023) and Coix (Zhai et al. 2023). However, there are currently no reports on the genome-wide identification analysis of the ARF family genes in C. paliurus. The released genome sequence data of C. paliurus provides an opportunity for us to study the ARF gene family in the genome-wide (Qu et al. 2023). In this study, we comprehensively and systematically analyzed the complete ARF gene family in C. paliurus. Family members of the ARF gene were identified and classified according to phylogenetic analysis. Their exon–intron structure, motif composition, cis-acting regulatory elements, duplication events, and synteny analysis with other species were also investigated. In addition, the gene expression pattern in the tissue of C. paliurus and its response to drought and salt were explored.

Materials and methods

Identification of ARF genes and chromosomal localization of CpARF

The whole genome sequence of C. paliurus was downloaded from the National Genomics Data Center (https://ngdc.cncb.ac.cn/gsa/, GWHGWHBKKY00000000) (Qu et al. 2023). Two methods were employed to identify the CpARF gene family members. The first method is to download the AtARF protein sequence of Arabidopsis from TAIR (https://www.Arabidopsis.org/). And use them to perform a blast search query on the C. palius genome (E value < 10–5). The second method was as follows: the hidden Markov model (HMM) files corresponding to the Auxin Resp (PF06507) domains were downloaded from Pfam protein family database (http://pfam.xfam.org/), and the hmm build command was used to identify the candidate ARF genes from C. paliurus genome (E-value < 10^–5) using Tbtools (v1.09874) software. All candidate CpARF sequences obtained were culled from repeat members and further confirmed in NCBI-CDD (https://www.ncbi.nlm.nih.gov/cdd), Pfam (https://pfam-legacy.xfam.org/), and SMART (http://smartembl.de/) databases. After identification of the CpARF gene family members, the candidate CpARF gene is designated based on their loci in the chromosome. The chromosomal location of each CpARF gene was obtained from MG2C (MapGene2Chrom, http://mg2ciask.in/mg2c_v2.0/) and plotted from the whole genome sequencing information of C. paliurus. The physicochemical properties (isoelectric point (pI) and molecular weight (Mw)) of CpARF proteins were predicted by ExPASY(https://web.expasy.org/protparam/). Subcellular location of the CpARF gene was predicted by online tools WoLF_PSORT (https://wolfpsort.hgc.jp/), Plant-mPLoc (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi/), and CELLO v.2.5 (http://cello.life.nctu.edu.tw/).

Phylogenetic analysis of CpARF genes

The ARF gene family protein sequences of rice (Wang et al. 2007) and Arabidopsis (Okushima et al. 2007) were obtained and performed for phylogenetic analysis, and then a neighbor-joining (NJ) phylogenetic tree with 1000 bootstrap replicates was constructed using MEGA7.0 software (Kumar et al. 2018). The specific parameters are as follows: the selected alignment sequence = MUSCLE method; Gap opening and gap extension = − 2.9 and 0, respectively; Poisson model = alternative model; Uniform rate = gap; missing data = two by two deletes, other values remain at the default value. The phylogenetic tree was visualized by Evolview (http://www.evolgenius.info/evolview/).

Gene duplication and collinearity analysis

The CpARF gene duplication in C. paliurus was analyzed using MCScanX software with default parameters. Collinearity analysis between C. paliurus and other five plant species (Arabidopsis, rice, Juglans regia, J. cathayensis, and P. trichocarpa) was performed using TBtools software (v1.098774) with the one-step MCScanx command. The synonymous substitution rates (Ks), nonsynonymous substitution rates (Ka), and the Ka/Ka ratio of CpARF gene pairs were calculated by KaKs_Calculator 2.0 software. The divergence time (T) was calculated using the Formula T = Ks/2r × 10^–6 Mya (r = 6.5 × 10⁻⁹).

Conservative motif observation and gene structure analysis

The conserved motifs of 34 CpARFs were identified using online MEME tools (https://meme-suite.org/meme/tools/meme), and it was concluded that their primitive widths were between 6 and 50 and the maximum primitive width was 2, and then TBtools were used to analyze the domains, exons-introns of CpARFs.

Analysis of cis-acting elements in CpARF promoters

The cis-acting elements of the upstream 2,000 bp sequence of CpARFs were predicted separately through the PlantCARE online database (https://bioinformatics.psb.ugent), and then visually analyzed by TBtools.

Functional analysis of CpARF

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (www.kegg.jp/kegg/kegg1.html) annotation evaluations were conducted by submitting CpARF protein sequences to eggNOG-mapper (Cantalapiedra et al. 2021; Kanehisa et al. 2023). Then, GO and KEGG enrichment evaluation and visualization were conducted using TBtools. The protein–protein interaction (PPI) network of CpARF proteins was generated using STRING database V11.5 (https://cn.string-db.org/).

CpARF gene expression profiling analysis

The raw data of the transcriptomic data were downloaded from NCBI publicly Expression Atlas Data. In detail, tissue expression analysis of CpARF genes during various tissues viz. stem, bark, tender leaf, old leaf, and flower were carried out in silico using publicly RNA-seq Data (BioProject number: PRJNA894718). The available data for drought treatment (BioProject number: PRJNA953807) was obtained from three PEG6000 concentrations (m/v): 0 (CK), 15% (moderate drought, MD), and 25% (severe drought, SD) to simulate drought stress (Li et al. 2023). The available data for salt treatment (BioProject number: PRJNA700136) was obtained from four salt concentration treatments (CK: 0 mM NaCl; LS: 0.15% mM NaCl; MS: 0.30% NaCl, and HS: 0.45% NaCl) under two times (T1: 15 d and 30 d) (Zhang et al. 2022). The TPM value of RNA-seq data was analyzed using the CLC workbench 20.0 software. Then, the log2-transformed TPM values were visualized using Lianchuan Biotech Cloud Platform (https://www.omicstudio.cn).

Plant treatment and tissue collection

The C. paliurus plant used for drought stress was collected from the nursery in Shaoyang City, Hunan Province, which was planted for 2 years with a tree height of 30 ± 3 cm. The samplings are placed in an 18 cm glass bottle container with a completely random design, and 1/2 strength Hoagland's nutrient solution (pH = 6.0 ± 0.2) is placed in the bottle. Based on previous studies, the treatment was set to three PEG6000 concentrations (0, 10%, 20%) treatment (drought stress) (Li et al. 2022), with three biological replicates per treatment. Several leaves were collected from the treated samplings at 2 h, 12 h, and 48 h, respectively, and then instantly frozen in liquid nitrogen and stored at − 80 ℃. Each selected gene has three biological replicates, and each biological replicate has three technical replicates.

Validation of CpARF gene expression by qRT‒PCR

Total RNA was extracted by Plant Total RNA Isolation Kit Plus Kit (Tiangen, 

Beijing, China), and cDNA was synthesized using the EasyScript One-Step gDNA Removal and cDNA Synthesis SuperMix Kit (TransGen Biotech, Beijing, China) according to the manufacturer’s instructions. qRT‒PCR was performed using THUNDERBIRD qPCR Mix QPS-201 (Toyobo, Shanghai, China) and an ABI 7900HT Real-Time PCR System (Applied Biosystems, Waltham, CA, USA). 18SRNA was selected as the internal reference, PCR reaction conditions were as follows: 10 min at 95 ℃, with 40 cycles of 15 s at 95 ℃, 60 s at 55 ℃, and 60 s at 72 ℃. The relative expression levels of the CpARF genes were calculated using the 2^−ΔΔCt method, and significance analysis was conducted by one-way ANOVA using Microsoft Excel 2016 software.

Subcellular localization of CpARF16 and CpARF32 proteins

In order to verify the subcellular localization of ARFs in C. paliurus, the CpARF16 and CpARF32, which active in drought and salt stresses, were selected as templates. The full-length open reading frames (ORF) of CpARF16 and CpARF32 were amplified and then cloned into the CaMV35S-GFP vector linearized by DNA restriction endonuclease Bam HI using TOROIVD® One Step Fusion Cloning Mix (FCM-050, TOROIVD Co., Ltd, China). After sequencing and confirming that there were no mutations, the CpARF16-GFP and CpARF32-GFP plasmids were introduced into Agrobacterium tumefaciens strain GV3101 and infiltrated into 4-week-old Nicotiana benthamiana mesophyll cells by acetosyringone for transient expression. H2B-mcherry served as a nuclear marker. Following a 48 h incubation, the fluorescence signals was visualized using a laser scanning microscopy (LSM 700; Carl Zeiss, Germany). The required primer sequences are shown in Table S8.

Results

Identification and sequence analysis of the ARF genes in C. paliurus

A total of 34 CpARF genes were identified based on BLAST and HMMER strategy in the genome of C. paliurus. The names of CpARFs were designated according to their chromosomal location (Table 1). The length of the genome DNA of CpARF genes are ranged from 681 bp (CpARF26) to 25,864 bp (CpARF27), and the coding sequence (CDS) length of CpARFs varied from ranges from 681 bp (CpARF26) to 5115 bp (CpARF27) with a mean length of 2,220 bp, and their encoded protein size ranges from 226 amino acids (CpARF26) to 1704 amino acids (CpARF27). The predicted molecular weight (MW) of CpARFs ranges from 25.2 kD to 187.4 kD, with the theoretical isoelectric points (pI) ranging from 5.35 (CpARF23) to 8.62 (CpARF26), with a mean value of 6.50. Almost all the CpARFs were characterized as hydrophilic (Grand Average of Hydropathicity < 0) and unstable (instability index > 40) proteins. The subcellular localization prediction based on Wolf, Plant mPLoc, and CELLO shows that most CpARFs are predicted to be located to the nucleus, except that CpARF12 to nucleus and cytoplasmic, CpARF13 and CpARF20 to peroxisome, CpARF14 and CpARF31 to cytoplasmic according to Wolf; CpARF8, CpARF12, and 31 to nucleus and cytoplasmic, CpARF14 to cytoplasmic and nucleus and mitochondrial, CpARF27 to chloroplast according to CELLO (Table 1). These predicted non-nuclear CPARF genes indicate that their functions may not have any other characteristic functions other than serving as transcription factors.

Table 1.

ClARF gene family in C. paliurus

Gene name	Sequence ID	Start	End	±	CDS Length	Amino Acid	Molecular Weight (kD)	pI	Aliphatic Index	Grand average of hydropathicity	Instability Index	Num of Exon	Domains	Wolf	Plant-mPLoc	CELLO
CpARF1	CpaM1st00554.t1	6,133,254	6,140,311	−	1791	596	65.45	5.99	71.95	− 0.341	51.05	4	Auxin_resp, B3	Nucleus	Nucleus	Nucleus and Chloroplast
CpARF2	CpaM1st02735.t1	36,666,575	36,672,131	−	2178	725	81.08	7.26	72.76	−0.452	58.71	15	Auxin_resp, B3, AUX_IAA	Nucleus	Nucleus	Nucleus
CpARF3	CpaM1st03908.t1	45,829,934	45,854,160	−	3432	1143	126.53	6.91	72.21	− 0.399	57.03	20	Auxin_resp, B3, AUX_IAA	Nucleus	Nucleus	Nucleus
CpARF4	CpaM1st22045.t1	12,530,943	12,540,885	−	3351	1116	124.50	6.28	71.85	− .639	70.06	13	Auxin_resp, B3, AUX_IAA	Nucleus	Nucleus	Nucleus
CpARF5	CpaM1st24690.t1	42,867,921	42,878,486	−	2556	851	95.41	5.84	71.81	− 0.483	60.41	14	Auxin_resp, B3, AUX_IAA	Nucleus	Nucleus	Nucleus
CpARF6	CpaM1st27253.t1	32,826,538	32,834,897	+	3330	1109	122.63	6.28	73.35	− 0.546	64.29	15	Auxin_resp, B3	Nucleus	Nucleus	Nucleus
CpARF7	CpaM1st28900.t1	9,431,199	9,436,132	+	2085	694	77.49	6.05	71.18	− 0.497	55.72	14	Auxin_resp, B3, AUX_IAA	Nucleus	Nucleus	Nucleus
CpARF8	CpaM1st30968.t1	37,745,968	37,751,126	−	2763	920	102.18	5.53	70.87	− 0.489	49.46	18	Auxin_resp, B3, AUX_IAA	Nucleus	Nucleus	Nucleus and Cytoplasmic
CpARF9	CpaM1st32014.t1	8,918,016	8,924,449	−	2526	841	92.94	5.35	72.54	− 0.443	56.2	12	Auxin_resp, B3	Nucleus	Nucleus	Nucleus
CpARF10	CpaM1st32694.t1	15,214,287	15,220,751	+	2112	703	78.18	6.87	73.3	− 0.342	52.73	14	Auxin_resp, B3, AUX_IAA	Nucleus	Nucleus	Nucleus
CpARF11	CpaM1st34516.t1	662,771	671,239	+	2415	804	90.10	5.82	74.07	− 0.458	58.2	14	Auxin_resp, B3, AUX_IAA	Nucleus	Nucleus	Nucleus
CpARF12	CpaM1st35746.t1	11,713,654	11,714,334	+	681	226	25.22	8.62	87.43	− 0.059	39.98	1	Auxin_resp	Nucleus and Cytoplasmic	Nucleus	Cytoplasmic and Nucleus
CpARF13	CpaM1st36501.t1	23,661,846	23,687,709	+	5115	1704	187.40	5.8	79.25	− 0.467	60.98	14	Auxin_resp, B3, AUX_IAA	peroxisome	Nucleus	Nucleus
CpARF14	CpaM1st37131.t1	32,552,399	32,555,188	−	2091	696	78.85	6.63	97.26	− 0.175	45.83	2	Auxin_resp, AUX_IAA	Cytoplasmic	Nucleus	Cytoplasmic and Nucleus and Mitochondrial
CpARF15	CpaM1st38220.t1	7,917,116	7,921,654	+	2106	701	77.07	6.24	68.16	− 0.344	52.15	4	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF16	CpaM1st39230.t1	22,790,927	22,798,022	−	2769	922	102.06	6.28	75.29	− 0.438	68.08	15	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF17	CpaM1st39966.t1	29,999,829	30,005,975	+	2256	751	82.15	6.28	68.83	− 0.402	50.42	12	Auxin_resp, B3	nucl	Nucleus	Nucleus
CpARF18	CpaM1st41984.t1	18,558,106	18,563,808	−	2079	692	77.10	6	79	− 0.403	45.68	14	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF19	CpaM1st44464.t1	12,207,364	12,215,034	−	2136	711	80.38	6.21	75.34	− 0.381	50.44	14	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF20	CpaM1st04275.t1	2,720,117	2,723,534	−	1860	619	67.95	7.59	69.14	− 0.439	49.85	3	Auxin_resp, B3	peroxisome	Nucleus	Nucleus
CpARF21	CpaM1st07399.t1	5,584,152	5,590,173	−	2184	727	79.50	6.33	69.23	− 0.426	53.85	11	Auxin_resp, B3	nucl	Nucleus	Nucleus
CpARF22	CpaM1st07915.t1	10,801,599	10,809,732	+	2721	906	100.67	6.57	77.47	− 0.412	66.83	14	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF23	CpaM1st08991.t1	26,764,358	26,768,875	+	2112	703	77.37	6.3	69.93	− 0.373	50.11	4	Auxin_resp, B3	nucl	Nucleus	Nucleus
CpARF24	CpaM1st10585.t1	15,715,549	15,720,150	+	2139	712	78.60	8.31	73.4	− 0.414	44.47	5	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF25	CpaM1st10673.t1	16,724,661	16,733,140	+	2886	961	105.82	5.74	75.35	− 0.373	47.14	15	Auxin_resp, B3	cyto	Nucleus	Nucleus
CpARF26	CpaM1st12847.t1	6,637,160	6,642,180	+	2370	789	88.73	6.31	67.43	− 0.539	60.84	16	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF27	CpaM1st14226.t1	24,857,423	24,871,569	+	1734	577	63.02	6.62	69.1	− 0.321	54.59	4	Auxin_resp, B3	nucl	Nucleus	Chloroplast
CpARF28	CpaM1st15605.t1	13,414,140	13,418,716	+	2115	704	77.97	6.82	71.9	− 0.405	50.75	4	Auxin_resp, B3	nucl	Nucleus	Nucleus
CpARF29	CpaM1st15655.t1	14,062,825	14,072,296	+	2367	788	87.27	6.18	73.06	− 0.423	48.85	12	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF30	CpaM1st17147.t1	3,062,940	3,074,437	−	3324	1107	122.47	6.04	78.93	− 0.467	61.45	14	Auxin_resp, B3	nucl	Nucleus	Nucleus
CpARF31	CpaM1st17834.t1	12,509,362	12,510,829	−	1404	467	53.43	6.88	96.77	− 0.235	49.26	2	Auxin_resp	Cytoplasmic	Nucleus	Cytoplasmic and Nucleus
CpARF32	CpaM1st18921.t1	593,421	599,876	+	2820	939	104.86	7.14	72.03	− 0.499	49.96	18	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF33	CpaM1st20192.t1	17,976,952	17,981,027	−	2049	682	76.09	6.17	67.01	− 0.551	59.04	15	Auxin_resp, B3, AUX_IAA	nucl	Nucleus	Nucleus
CpARF34	CpaM1st47162.t1	2,614,932	2,618,544	+	1989	662	72.61	7.63	73.01	− 0.388	52.52	4	Auxin_resp, B3	nucl	Nucleus	Nucleus

Chromosomal distribution of the CpARF gene family

The CpARFs were examined for their distribution on the chromosomes C. paliurus based on gene coordinate annotation data. The identified 34 CpARF genes were unevenly and randomly distributed on all 16 chromosomes of C. paliurus genome, except that CpARF34 was not assigned to any chromosomes. Chr13 harbored the largest amount of CpARFs (four members), followed by Chr 1, Chr 3, and Chr 14 (each with three members). There were two CpARF members presented on Chr 4, Chr 5, Chr 6, Chr 7, Chr 8, Chr 9, Chr 11, and Chr 12. While Chr 2, Chr 10 and Chr 15 each carried only a single CpARF gene (Fig. 1).

Fig. 1.

Chromosomal distributions of ClARF genes in the C. paliurus genome. The length of the chromosomes is represented by the scale on the left. The chromosome number is located at the top of each chromosome, and the ClARF gene names are labeled with red color

Phylogenetic analysis and gene duplication of CpARF genes

To identify the evolutionary relationship between the CpARF protein in C. paliurus and the ARF protein in the existing model plants rice and Arabidopsis, the evolutionary relationship of ARF gene family was constructed by using 82 ARF amino acid sequences of 23 AtARF proteins in Arabidopsis (Okushima et al. 2007), 25 OsARF proteins in rice (Wang et al. 2007), and 34 CpARF proteins in C. paliurus. As shown in Fig. 2a, the ARF protein phylogenetic tree is divided into six categories (ClassI–ClassVI), which contain 10, 10, 3, 32, 10, and 17 members, respectively (see Fig. 2 for details). The 34 CpARFs were compared and classified into all six categories, of which Class I included CpARF6/30/19/25, Class II included CpARF18/27/14/20, Class III included CpARF23, Class IV included CpARF1/11/7/29/8/12/4/34, Class V included CpARF5/31/9/13, and Class VI included CpARF24/ 33/2/10/16/22/3/15/28/26/32/17/21. Interestingly, there are 6 CpARFs clustered in Class VI, including CpARF7/29/8/12/4/34, displayed sisters clusters with AtARF16, which suggested that these 6 CpARFs were expanded in the C. paliurus genome. In total, the CpARFs are more closely related to Arabidopsis than rice according to the evolutionary trees (Fig. 2a).

Fig. 2.

Phylogenetic relationships and duplication analysis of ARF proteins in C. paliurus. a. Phylogenetic relationships ARF proteins C. paliurus (Cp), Arabidopsis (At), and rice (Os). Full-length ARF protein sequences were aligned using the Clustal X 1.83 software, and the neighbor-joining (NJ) phylogenetic tree was constructed using MEGA 7.0 with 1,000 bootstrap replicates. The classes I-VI were distinguished by distinct colors. b. The duplication analysis CpARF used by multiple collinear scanning toolkits (MCScanX) in Tbtools. The gray lines represent the collinear blocks within the C. paliurus genome, and the red lines highlight the syntenic pairs of ARF genes. The number of chromosomes is displayed in the middle of each chromosome

Gene duplication events, including tandem duplication and segmental duplication, are the main driving force for the expansion of the gene family and generate new biological functions (Qiao et al. 2019). Duplication analysis of CpARFs showed that there are 30 segmental duplication events, while there are no tandem duplication events occurred in the CpARF gene family of the C. paliurus genome (Fig. 2b). As shown in Fig. 2b, 30 pairs of segmental duplications among CpARF genes were unevenly distributed on all 16 chromosomes. The segmental duplications are as following: CpARF2/CpARF26, CpARF1/CpARF27, CpARF2/CpARF32, CpARF2/CpARF8, CpARF20/CpARF28, CpARF20/CpARF15, CpARF20/CpARF34, CpARF23/CpARF24, CpARF21/CpARF17, CpARF22/CpARF16, CpARF23/CpARF15, CpARF24/CpARF28, CpARF25/CpARF29, CpARF24/CpARF15, CpARF24/CpARF34, CpARF26/CpARF32, CpARF26/CpARF8, CpARF28/CpARF15, CpARF28/CpARF34, CpARF30/CpARF6, CpARF33/CpARF7, CpARF32/CpARF8, CpARF33/CpARF18, CpARF5/CpARF11, CpARF4/CpARF13, CpARF7/CpARF18, CpARF10/CpARF19, and CpARF15/CpARF34. However, no tandem duplication was found among CpARF genes. Therefore, segmental duplication, rather than tandem duplication, played a vital role in the expansion of ARF genes in the C. paliurus genome (Fig. 2b).

Synteny analysis and Ka/Ks of CpARFs

To further investigate the phylogenetic relationship of ARF genes, the potential evolution link and collinearity of members of the ARF gene family between C. paliurus and 5 other species, including 4 dicot plants (Arabidopsis, J. regia, J. cathayensis, and P. trichocarpa) and one monocot plant (O. sativa), were analyzed. As shown in Fig. 3, ARF genes in C. paliurus showed different degrees of collinear relationship with other species. The most orthologous CpARF gene pairs (74) were detected between C. paliurus and J. regia, followed by P. trichocarpa, J. cathayensis, Arabidopsis, and rice with 73, 51, 25, and 11 homologous pairs with C. paliurus, respectively (Fig. 3). The results indicated that the collinear genes between C. paliurus and dicots were far more than that between C. paliurus and monocots.

Fig. 3.

Collinearity analysis of ARF genes between C. paliurus and 5 other species. Gray lines indicate collinear blocks within the C. paliurus genome and other genomes, and the red curve indicates collinear ARF genes

To understand the evolutionary information on the CpARF gene family, the Ka/Ks ratio of CpARF gene pairs was calculated. The Ka/Ks ratio of all duplicate gene pairs is less than 0.6, indicating that the CpARF genes had undergone strong purification selection during evolution (Table 2). The divergence times of duplicated gene pairs ranged from 20.09 to 123.33 Mya and averaged 66.71 Mya. Fourteen segmental duplicated events occurred in the same era (averaged 24.75 Mya) and another sixteen segmental duplicated events occurred in another same era (averaged 103.42 Mya) (Table 2). The results demonstrated that the CpARF gene family may have undergone two large-scale segmental duplicated expansions.

Table 2.

The Ka/Ks ratio of ARF gene pairs in C. paliurus

Gene pair	Duplicated type	Ka	Ks	Ka/Ks	Purify selection	Divergence time (Mya)
CpARF2/CpARF26	Segmental	0.110865977	0.40163326	0.276038	Yes	30.89
CpARF1/CpARF27	Segmental	0.081897632	0.363652014	0.225209	Yes	27.97
CpARF2/CpARF32	Segmental	0.283337049	1.485494998	0.190736	Yes	114.27
CpARF2/CpARF8	Segmental	0.266068235	1.353447801	0.196586	Yes	104.11
CpARF20/CpARF24	Segmental	0.245571147	1.197509238	0.205068	Yes	92.12
CpARF20/CpARF28	Segmental	0.246328325	1.141320794	0.215827	Yes	87.79
CpARF20/CpARF15	Segmental	0.226804516	1.327846104	0.170806	Yes	102.14
CpARF20/CpARF34	Segmental	0.084191754	0.350566939	0.240159	Yes	26.97
CpARF23/CpARF24	Segmental	0.20280738	1.529859164	0.132566	Yes	117.68
CpARF23/CpARF28	Segmental	0.214295977	1.290856427	0.166011	Yes	99.30
CpARF21/CpARF17	Segmental	0.106487524	0.269557461	0.395046	Yes	20.74
CpARF22/CpARF16	Segmental	0.055785449	0.261232753	0.213547	Yes	20.09
CpARF23/CpARF15	Segmental	0.051260939	0.354578617	0.144569	Yes	27.28
CpARF24/CpARF28	Segmental	0.085150614	0.307452842	0.276955	Yes	23.65
CpARF25/CpARF29	Segmental	0.074221844	0.304042771	0.244116	Yes	23.39
CpARF24/CpARF15	Segmental	0.202712706	1.323209993	0.153198	Yes	101.79
CpARF24/CpARF34	Segmental	0.260387554	1.364030406	0.190896	Yes	104.93
CpARF26/CpARF32	Segmental	0.345491841	1.40931306	0.245149	Yes	108.41
CpARF26/CpARF8	Segmental	0.320248015	1.603252789	0.199749	Yes	123.33
CpARF28/CpARF15	Segmental	0.20638685	1.228187693	0.168042	Yes	94.48
CpARF28/CpARF34	Segmental	0.254864511	1.025951312	0.248418	Yes	78.92
CpARF30/CpARF6	Segmental	0.083849167	0.315160484	0.266052	Yes	24.24
CpARF33/CpARF7	Segmental	0.073679867	0.304985235	0.241585	Yes	23.46
CpARF32/CpARF8	Segmental	0.093882256	0.345522554	0.271711	Yes	26.58
CpARF33/CpARF18	Segmental	0.245658768	1.423677595	0.172552	Yes	109.51
CpARF5/CpARF11	Segmental	0.053364525	0.286640567	0.186172	Yes	22.05
CpARF4/CpARF13	Segmental	0.079501381	0.341154452	0.233036	Yes	26.24
CpARF7/CpARF18	Segmental	0.26523003	1.531236285	0.173213	Yes	117.79
CpARF10/CpARF19	Segmental	0.165226483	0.297976348	0.554495	Yes	22.92
CpARF15/CpARF34	Segmental	0.220233441	1.277123566	0.172445	Yes	98.24

CpARF gene structure and conserved domain analysis of CpARFs

Analysis of the conserved motifs of the CpARF protein using HMMER online analysis software detected a total of 10 motifs, of which 2 common motifs are motif2 and motif9. The DBD domain is related to motif6 and motif4, and the CTD domain is related to motif10 and motif5 (Fig. 4a, b). Therefore, phylogenetic analysis of the CpARF gene can also be obtained by the pattern and structure of the primitives.

Fig. 4.

Phylogenetic relationships, gene structures, conserved protein motifs, and CCD domain of the CpARFs. a The NJ phylogenetic tree was constructed based on the full-length sequences of CpARF proteins using MEGA 7.0. b Motif distribution of the CpARF proteins. The conserved motifs of CpARF proteins were identified by MEME tool (http://meme-suite.org/ tools/meme). The motifs, numbered 1–10, are displayed in different colored boxes. c The conserved domain of the CpARF proteins were analyzed by the online Conserved Domain Database (CDD) tool (https://www.ncbi.nlm.nih.gov/cdd/). d Exon–intron structures of the CpARF genes. Orange boxes indicate exons, and black lines indicate introns; green boxes are represented UTRs; yellow boxes indicated coding sequences

To further understand the domain architecture of CpARF proteins, the conserved domain distributions in CpARFs were analyzed based on Batch Web CD-Search Tool (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi). Most of CpARFs had the highly conserved B3-type DNA-bind domain (DBD) in the N-terminal portion, except CpARF26/28 (Table 1). Moreover, 13 out of 27 ClARFs belonged to Class IV and Class V subfamilies and lacked CTD domain (Aux/IAA domain) (Fig. 4c).

Exon–intron structure analysis will provide valuable information for evolutionary relationships among plants. To investigate the gene structure of CpARF genes, their exon–intron structure was analyzed by using TBtools based on the annotated data. All the identified 34 CpARF genes contained exons varied from 1 (CpARF26) to 20 (CpARF3) with the average number of exons being 11. The majority (23) of the CpARFs contained more than 10 exons, except CpARF9 contained 5 exons, CpARF1/7/10/11/28/34 contained 4 exons, CpARF4 contained 3 exons, and CpARF15/28 contained 2 exons (Fig. 4d, Table 1). Most CpARF members within a given class have similar exon–intron structures in terms of intron number and exon length. For example, CpARF genes in class IV have 3 to 5 introns, which is lower than that of most other classes. The conserved exon–intron structure shared within each class provides important support for the classification and evolutionary relationships of members of the CpARF family.

GO term, KEGG pathway enrichment and analysis of the PPI network of CpARFs

To further elucidate the functions of the CpARF genes, GO and KEGG enrichment analysis was performed. As shown in Fig. 5, CpARF proteins were enriched in three main categories of GO, including 8 terms of molecular function, 6 terms of cellular components, and 34 terms of molecular function (Supplementary Table S1). In terms of molecular function, CpARF proteins were classified as transcription regulator activity (GO:0140110), DNA-binding transcription factor activity (GO:0003700), DNA binding (GO:0003677), etc. In terms of cellular components, CpARF proteins were components of the nucleus (GO:0005634). In biological process terms, the high-enriched GO term includes RNA biosynthetic process (GO:0032774), DNA-templated transcription (GO:0006351), and regulation of DNA-templated transcription (GO:0006355), etc. (Fig. 5a). In addition, the KEGG pathway enrichment analysis identified three pathways involved in the different functions of the CpARF proteins. The highly enriched pathways include Plant hormone signal transduction (04075), Environmental information processing (A09130), and Signal transduction (B09132) (Fig. 5b; Supplementary Table S2).

Fig. 5.

Diagram of Cis-regulatory element in the promoter of CpARF genes. a The length of the CpARF gene promoter was represented by black line. The different colored boxes on the right represent cis-acting elements with different functions. b The numbers of different type of cis-acting elements in the promoter of CpARF genes. c The heatmap of different type of cis-acting elements in the promoter of CpARF genes

To determine interactions among CpARFs and related proteins, a PPI interaction network was constructed using the STRING database based on Arabidopsis protein orthologs. A total of 34 CpARF proteins had Arabidopsis orthologs with identities ranging from 40.6 to 74.3%. As shown in Fig. 5c, the PPI network consisted of 13 nodes and 20 edges, suggesting that these CpARFs interacted with each other and with other proteins and participated in some biological processes. For example, CpARF18 shared a high degree of interaction with other CpARF members, including CpARF3, CpARF16, and CpARF27, which hinted that CpARF18 may form heterologous complexes with these ARF proteins to exert biological functions. CpARF14 was associated with ARATH, SKIP, K16L22.5, T10O24.21, and PRP19B proteins. SKIP is a splicing factor involved in the post-transcriptional regulation of the circadian clock, flowering time, osmotic tolerance, and ABA signaling in Arabidopsis (Wang et al. 2012; Feng et al. 2015; Zhang et al. 2022). The results of GO, KEGG, and PPI analysis demonstrated that CpARF genes were related to the growth, development, and abiotic stresses.

Analysis of cis-acting regulatory elements in the CpARF genes promoters

To investigate the regulation elements on the CpARF promoter, we analyzed the cis-acting elements on 2, 000 bp upstream of the gene starting site of CpARF genes. A total of 593 cis-acting elements were found present in the promoter of CpARF genes, most of which were associated with development, hormone, light, and stress-responsive elements. Moreover, the cis-acting elements of most CpARF promoters within a given class have similar types (Fig. 6a). We further divided the cis-acting elements into four categories, including development, hormone, light, and stress-response elements (Fig. 6b). A higher number of light-responsive elements were found among the 26 out of 34 CpARF gene promoters. The majority of the CpARFs contain a higher number of MeJA-responsive elements (88 in total, 44 CGTCA-motif and 44 TGACG-motif), followed by ABA-responsive elements (77 ABRE), salicylic acid-responsive elements (25), auxin-responsive elements (23 in total, 21 TGA-element and 2 AuxRR-core), and gibberellin-responsive element (18 in total, 9 P-box, 5 TATC-box and 4 GARE-motif). Furthermore, there are many types of elements in the stress response, such as anaerobic induction (ARE motif: AAACCA) (63 in total), low-temperature response (LTR motif: CCGAAA) (30 in total), drought inducibility (MBS motif: CAACTG) (22 in total), and anoxic (GC-motif: CCCCCG) (10 in total) (Fig. 6b, 6c; Supplementary Table S3). All CpARFs contain at least one cis-regulatory element associated stress response, suggesting that the functions of CpARFs were related to abiotic responses. Interestingly, some CpARF genes contain multiple copy cis-acting elements (Fig. 6c; Supplementary Table S3). For example, CpARF7/9/24/33 contains multiple ABRE elements, each containing 11, 6, 6, and 6 copies, respectively, suggesting that CpARF7 might be involved in ABA and drought response.

Fig. 6.

Gene ontology (GO), KEGG enrichment analysis, and predicted protein–protein interaction (PPI) networks of CpARFs. a GO enrichment analysis of CpARFs. The X and Y axes represent the -log10 (P value) and the information on GO terms, respectively. b KEGG enrichment analysis of CpARFs. The X and Y axes represent the -log10 (P value) and the information on the KEGG pathway, respectively. C PPI network of significant CpARF proteins in C. paliurus. Nodes represent proteins and gray lines indicate interactions between nodes. Different thicknesses of grey edges indicated the degree of protein–protein associations

Expression profiling of CpARF genes in different tissues

To investigate the expression profiling of CpARF genes in different tissues, the TPM (transcripts per kilogram base per million mapped reads) values of 34 CpARF genes were acquired from the transcriptome data of five distinct tissue samples (stem, bark, tender leaf, old leaf, and flower) based on the RNA-seq data (PRJNA894718). A heatmap was generated with the corresponding log2 TPM values of the five tissues, and almost all CpARF genes, except CpARF12, could detect to have different degrees of expression levels in five different tissues (Fig. 7; Supplementary Table S4). According to the overall clustering results, the expression patterns between stems and bark, as well as between flowers and tender leaves, are similar, while the expression patterns of old leaves differ from them. It was mentioned that several CpARF genes displayed tissue-specific expression. For example, CpARF10/19 was preferentially expressed in flower and tender leaf, but was very low expressed in stem, bark, and old leaf. CpARF10 was highest expressed in flower, tender leaf, and stem, but was rarely expressed in other tissues. CpARF22/25/34 was highly expressed in flower, stem, bark, and tender leaf, but was hardly detected in old leaf. CpARF16 was highest expressed in the stem, followed by bark tender leaf, flower, and was lowest expressed in old leaf. Moreover, CpARF22 is preferentially expressed in bark and stem compared to other tissues. These differential expressions of CpARF genes in different tissues suggested that these CpARF genes probably play specific functions in the growth, development, and reproductive process of C. paliurus.

Fig. 7.

Expression pattern of 34 CpARF genes in different tissues of C. paliurus. The heatmap was generated based on the relative expression values of 34 CpARF genes obtained by RNA-seq data in five different tissues (BioProject number: PRJNA894718)

Analysis of CpARF gene expression in response to drought and salt stresses

Analysis of cis-acting elements showed that most of the promoters of CpARF genes harbored stress response elements. To identify the potential role of CpARF genes under drought and salt stresses, the TPM values of 34 CpARF genes were acquired from the transcriptome data based on the RNA-seq data (PRJNA700136 and PRJNA894718). The results showed that most of CpARF genes (24) have s significant (P < 0.05) response to drought stress at some or all times of 2 h, 12 h, and 48 h under moderate drought treatment. Among these drought-responsive CpARF genes, 16 were down-regulated, and three (CpARF3/13/20) were up-regulated at some or all times of 2 h, 12 h, and 48 h treatment. Whereas the expression of 4 CpARF genes (CpARF6/14/16/32) showed a trend of first decreasing and then increasing, and CpARF7 showed a trend of first increasing and then decreasing under moderate drought stress. In addition, the majority of CpARF genes (28) have significantly responses to drought stress at some or all times of 2 h, 12 h, and 48 h under severe drought treatment (P < 0.05), among them, only two CpARF genes (CpARF14/20) were significantly up-regulated (Fig. 8a, Supplementary Table S5).

Fig. 8.

Expression pattern of CpARF genes under drought and salt stress treatments. a Expression pattern of CpARF genes under drought stress. The available data for drought treatment (BioProject number: PRJNA953807) was obtained from three PEG6000 concentrations (m/v): 0% (CK), 15% (moderate drought, MD), and 25% (severe drought, SD) to simulate drought stress (Li et al. 2023). b Expression pattern of CpARF genes under salt stress. The available data for salt treatment (BioProject number: PRJNA700136) was obtained from four salt concentration treatments (CK:0 mM NaCl; LS: 0.15% mM NaCl; MS: 0.30% NaCl, and HS: 0.45% NaCl) under two times (T1: 15 d and 30 d) (Zhang et al. 2022)

The ARF family not only has a significant response to drought stress, but participates in the response to salt stress. The majority of CpARF genes (27) could respond to salt stress at high, middle, and low under T1 salt treatment, of which 16 (CpARF1/3/5/6/7/8/9/13/14/21/25/30/31/32/33) were significantly up-regulated, and 7 (CpARF10/15/16/1723/28/34) were significantly down-regulated. In addition, two of them (CpARF10/16) showed a trend of up-regulated at LS, down-regulated at MS, and returning to normal levels at HS. CpARF20 was at normal level at LS, up-regulated at MS, and down-regulated at HS. Regarding T2 salt treatment, a total of 29 CpARF genes could respond to salt stress at high, middle, and low under, of which 15 of them were significantly down-regulated at some or all treatments of LS, MS, and HS, and only one of them (CpARF1) were significantly up-regulated at all treatment. Interestingly, one gene (CpARF7) was significantly down-regulated at LS and significantly up-regulated at MS and HS under T2 salt treatment. Six genes (CpARF16/26/28/29/30/32) were significantly up-regulated at LS, and down-regulated at MS and/or HS condition. Furthermore, two genes (CpARF20/23) showed down-regulated at LS, and up-regulated at MS, and then down-regulated at HS (Fig. 8b, Supplementary Table S6).

To validate the results obtained from RNA-seq data, 12 CpARF genes significantly regulated by drought stress were selected to conduct qRT-PCR (Supplementary Table S7). The expression trends of most selected CpARF genes from the results of qRT-PCR were overall consistent with the RNA-seq data, except for some CpARF genes having some differences in expression trends at different time points. For example, CpARF7 was down-regulated at all times of 2 h, 12 h, and 48 h under moderate and severe drought treatment. In addition, CpARF20 was significantly up-regulated at MD-12 h and SD-12 h, while was not significantly regulated at MD-2 h, MD-48 h, SD-2 h, and SD-48 h from RNA-seq data, but was significantly up-regulated at all times of 2 h, 12 h and 48 h under moderate and severe drought treatment (Fig. 9). The difference in expression of ARF gene between RNA-seq and qRT-PCR may be caused by the different sensitivity of treated materials or seedling ages to drought. These results revealed that these CpARF genes may play critical roles in drought response.

Fig. 9.

qRT–PCR verify of 12 CpARF genes under drought stress. The column indicates mean ± standard deviation. All experiments were performed independently at least three times. * P < 0.05, **P < 0.01

Subcellular localization of CpARF16 and CpARF32 proteins

The subcellular localization prediction showed that most CpARFs are predicted to be located to the nucleus, except that CpARF12 to nucleus and cytoplasmic, CpARF13 and CpARF20 to peroxisome, CpARF14 and CpARF31 to cytoplasmic (Wolf), CpARF8, CpARF12, and 31 to nucleus and cytoplasmic, CpARF14 to cytoplasmic and nucleus and mitochondrial, CpARF27 to chloroplast (CELLO). To test these results, we select CpARF16 and CpARF32 genes, which active in drought and salt stresses, to perform the transient expression in the leaves of N. benthamiana via Agrobacterium-mediated transformation. The full-length ORF of CpARF16 and CpARF32 was fusion to the n-terminus of green fluorescent protein (GFP) in the same reading frame derived by CaMV35S promoter and generated CpARF16-GFP and CpARF32-GFP fusion construct, and then transformed into N. benthamiana mesophyll cells. As expected, the fluorescence signals of CpARF16-GFP and CpARF32-GFP were co-located with H2B-mCherry in the nucleus. The results confirmed that the CpARF16 and CpARF32 proteins were located to the nucleus (Fig. 2g) (Fig. 10).

Fig. 10.

Subcellular localization of CpARF16 and CpARF32 proteins in the mesophyll cells of Nicotiana benthamiana leaves. H2B-mCherry is a nucleus marker. Bar = 20 μm

Discussion

ARF gene family play vital roles in regulating growth and development, metabolism progress, and stress response (Ulmasov et al. 1999; Guilfoyle and Hagen. 2007; Roosjen et al. 2018). With the rapid increase in whole-genome sequencing, the ARF gene family of nearly 50 plant species has been identified at the whole-genome wide level. However, there have been no reports on the identification of the ARF gene family at the genome level in C. paliurus until now. In the present study, we identified and characterized the CpARF gene family on the genome-wide scale.

Characterization of ARFs in the C. paliurus

Although the number of motif members varies across different classes, the group motif pattern is strongly conservative. The phylogenetic tree analysis of 10 conserved motifs in CpARF revealed that each class of CpARFs contains common or specific motifs. For example, 2 common motifs (motif2 and motif9) which correspond to the auxin response domain, were highly conserved in the CpARF gene family. In addition, all 10 motifs were found presented in Class I and II, while the remaining classes lack one or more motifs (Fig. 2a; Table 1), which is consistent with the results in other species (Okushima et al. 2007; Wang et al. 2007; Zhai et al. 2023; Chen et al. 2023). The presence or absence of other motifs may be related to new plant functions and further research is needed. By analyzing the conserved domains of CpARF proteins, all CpARFs contain an auxin_respsonse domain, and most of CpARFs contain a typical B3 domain (DBD domain), except CpARF12/CpARF14/CpARF31, which belonging to Class VI. Lack of B3 domain in ARFs has also been found in other species, indicating that these genes cannot be recognized and bind to AuxREs on the target gene promoter, and lose their function as transcription factors. Moreover, only 20 CpARFs (58.8%) contain an auxin_IAA binding domain, and they are mainly distributed in classes I, and VI. The absence of the B3 domain in ARF indicates that they may not be able to form heterodimers with AUX/IAA or homodimers with other ARF proteins (Ulmasov et al. 1999; Guilfoyle and Hagen. 2007). According to the gene structure analysis, the variation range of exon number in CpARF genes varies from 1 to 20, which is larger than that of Arabidopsis (from 2 to 15) (Okushima et al. 2007) and rice (from 3 to 16) (Wang et al. 2007). Interestingly, Class IV members contained 3–5 exons, which were relatively few that most of other class (Fig. 4d), which suggested that the ARF members of Class IV have certain specificity in structure and functions. In general, our results are similar to those in previously reported species, hinting that the structure and function of the ARF gene family are highly conserved in plant species.

ARF proteins are transcription factors and are typically located to the nucleus. Similar to most of the ARFs in other species, most of CpARF proteins were predicted to be located to the nucleus, except that CpARF12 to nucleus and cytoplasmic, CpARF13 and CpARF20 to peroxisome, CpARF14 and CpARF31 to cytoplasmic according to Wolf; CpARF8, CpARF12, and 31 to nucleus and cytoplasmic, CpARF14 to cytoplasmic and nucleus and mitochondrial, CpARF27 to chloroplast according to CELLO (Table 1). Interestingly, CpARF14 and CpARF31 also lacked the B3 domain (Table 1), suggesting they may not function as transcription factors. Previous reports have found that several ARFs are not localized to the nucleus. For example, EgARF6 in Oil palm (Elaeis guineensis Jac) was predicted to be localized to cytoplasm, EgARF11 to cytoplasm, and EgARF7 to mitochondria, respectively (Jin et al. 2022). There are 22 MsARFs in Medicago sativa predicted to the chloroplast, and 9 MsARFs to the cytoplasm (Chen et al. 2023). These ARFs that are not located to nucleus may have undergone structural changes during long-term evolution, leading to functional differentiation.

Expansion of CpARF gene family and collinearity analysis

A total of 34 CpARF genes were identified and analyzed based on the whole genome of C. paliurus, the number of CpARF genes was similar to that in Z. mays (36) (Xing et al. 2011), apple (31) (Luo et al 2014), M. truncatula (39) (Shen et al. 2015), P. trichocarpa (35) (Kalluri et al. 2007; Liu et al. 2023), but more than that in Arabidopsis(23) (Okushima et al. 2005), rice (25) (Wang et al. 2007), grapevine (19) (Wan et al. 2014), pigeonpea (12) (Arpita et al. 2023), Eucalyptus grandis (17) (Yu et al. 2014), physic nut (17) (Tang et al. 2018), and Coix (27) (Zhai et al. 2023), which suggested that the CpARF gene family was expanded compared to these species. The genome size of the two diploid types (PA-dip, PG-dip) of C. paliurus is 587 Mb and 583 Mb respectively (Qu et al. 2023), which is about 4.7 times the size of the Arabidopsis genome (125 Mb), 1.5 times the size of the rice genome (385.7 Mb), 1.8 times the size of physic nut genome (320 Mb), and 27.2% of the size of the maize genome (2.16 Gb), but the fact is that the number of ARF family members is independent of the genome size.

Gene duplications, such as tandem and segmental duplications, are the important driving forces for genome evolution and the gene expansion of multiple gene families, in which whole genome segmental duplications were supposed to contribute to gene functional redundancy, while tandem duplication can help plants adapt to constantly changing environments and acquire novel functions (Ulmasov et al. 1999; Qiao et al. 2019). In our report, a total of 30 segmental duplication events, and no tandem duplication events were found in the CpARF gene family of the C. paliurus genome (Fig. 2b), suggested that segmental duplications, rather than tandem duplications contributed to CpARF gene family expansion. It is to some extent consistent with previous results that segmental duplication was the main reason for the expansion of the ARF gene family (Wen et al. 2019; Zhai et al. 2023; Liu et al. 2023).

To explore the evolution of ARF genes, the collinearity relationships between C. paliurus and five species were analyzed. The collinear genes between C. paliurus and dicots were far more than those between C. paliurus and monocots (Fig. 3), which implied that ARF genes have undergone extensive evolution and duplicated events after differentiation in monocot and dicot plants, leading to the unique features of monocot and dicot plants. The same results were found in P. trichocarpa (Kalluri et al. 2007), Prunus avium (Hou et al. 2021), M. sativa (Chen et al. 2023), L. usitatissimum (Qi et al. 2023).

Phylogenetic analysis of CpARF and its putative functions

From the phylogenetic tree maps of C. paliurus and Arabidopsis and rice, 34 CpARF proteins were unevenly distributed in 6 subfamilies, and most of the CpARF proteins were closer to the members of the AtARF family than that of the OsARF family (Fig. 2a). Class VI contained the most CpARF(13) (CpARF2/3/7/8/10/12/14/18/19/26/31/32/33) and was closely related to AtARF1/2/11. The AtARF1 and AtARF2 genes are involved in floral organ development, senescence, and abscission of flower organs (Ellis et al. 2005), suggesting these 13 genes probably function in floral organ development or senescence. In Class VI, OsARF1 gene may interact with OsIAA1 to regulate rice morphogenesis (Waller et al. 2002) and interact with OsIAA6 to regulate leaf inclination by suppressing auxin signaling (Xing et al 2022). OsARF4, also belonging to class VI, is involved in leaf inclination regulation via auxin and brassinosteroid pathways in rice (Qiao et al. 2022). class IV contained 8 CpARF proteins and was closely related to AtARF10/16/17.Interestingly, the clade corresponding to AtARF16 has 6 CpARFs (CpARF23/15/24/28/20/34) in C. paliurus, suggesting these six CpARF genes have expanded in C. paliurus. AtARF10 and AtARF16 gene are involved in root crown cell formation (Wang et al. 2005). Moreover, AtARF10 gene also plays roles in the seed germination and post germination process, as well as in the formation of the primary outer wall and pollen development (Liu et al. 2007; Yang et al. 2013). Class II contained 4 CpARF members (CpARF3/6/13/30) that were closely related to AtARF7/19. AtARF7 gene has been shown to function in hypocotyl response to blue light and reduce the auxin response (Harper et al. 2000). The hypocotyl and roots of the AtARF7/AtARF19 genes double mutant exhibit aberrant lateral root formation and abnormal gravitropism (Narise et al. 2010). OsARF19, belonging to Class II, positively regulates leaf inclination by binding the promoter region of OsBRI1 in rice (Zhang et al. 2015). Class I contained 4 CpARF members (CpARF5/11/16/22) that were closely related to AtARF6/8. AtARF6 and AtARF8 genes are thought to promote the production of jasmonic acid, and regulate petal, stamen, and gynoecium growth at anthesis (Tabata et al. 2010). In class I, OsARF12 gene controls root elongation and regulates iron accumulation in rice (Qi et al. 2012), and OsARF17 gene plays important roles in plant defense against different types of plant viruses (Zhang et al. 2020a, b). Class III contained 1 CpARF member (CpARF9) that was closely related to AtARF5, AtARF5 gene is essential for embryonic and vascular tissue development, as well as in female and male gametophyte development (Liu et al. 2018a, b). Class V contained 4 CpARF members (CpARF17/21/25/29) that were closely related to AtARF3/4, AtARF3/4 genes are involved in the development of leaf development, polarity specification, and floral meristem determinacy (Sessions et al. 1997; Liu et al. 2014). Various researches have revealed that ARF genes play crucial roles in regulating abiotic stress responses, such as drought and salinity in plants. For example, overexpression of AtARF3 can induce the expression of drought-stress-responsive genes in Arabidopsis, while, these ARF-activated genes maintain very low expression levels in the ARF3 mutant under drought stress (Zheng et al. 2018). The expression of SlARF8A is strongly induced in leaves, primary roots, and lateral roots under salt stress, while SlARF10A is comparatively stronger inducted in roots than in leaves under salt stress (Bouzroud et al. 2018). In addition, OsARF12, OsARF16, and OsARF11 can interact with IAA10, in which OsARF12 and OsARF16 positively regulate rice resistance to viruses, while OsARF11 negatively regulates rice resistance to viruses (Qin et al. 2020). Therefore, it is speculated that the CpARF genes may be involved in various aspects of the growth and development process, and biotic and abiotic stress response in C. paliurus, and its function is worth further research.

Expression pattern of CpARF genes and their potential roles

The expression pattern of genes in different tissues provides important information for predicting their potential biological functions. Almost all the 34 ClARF genes, except CpARF12, expressed varied greatly in five different tissues (Fig. 7; Supplementary Table S4). CpARF12 lacks B3 and CTD domains (Fig. 4c; Table 1), suggesting that CpARF12 may have lost its function during evolution or only express and perform specific biological functions under specific conditions. Unsurprisingly, several segmental duplication CpARF gene pairs displayed similar expression patterns. For example, CpARF20/CpARF15, CpARF20/CpARF34, CpARF22/CpARF16, CpARF23/CpARF15, CpARF24/CpARF15, CpARF26/CpARF32, CpARF26/CpARF8, CpARF32/CpARF8, CpARF33/CpARF18, CpARF5/CpARF11, and CpARF10/CpARF19, totally showed very similar expression patterns in different tissues (Figs. 2b; 7; Supplementary Table S4). These segmental duplication CpARF gene pairs with similar expression patterns may have redundant functions. Otherwise, those segmental duplication CpARF gene pairs with different expression patterns may have developed novel biology functions during the evolution process in C. paliurus. Several CpARF genes, such as CpARF10/19/16/22/25/34 displayed specific expression patterns, indicating these CpARF genes probably play distinct roles in the growth, development, and reproductive process. The same results are also found in other species, such as Oil palm (Jin et al. 2022), kiwifruit (Su et al. 2021), and Coix (Zhai et al. 2023).

It has been reported that the expression of ARFs is regulated by various phytohormones, such as auxin, ABA, and SA, and various abiotic stresses. For example, numerals ARF genes, such as SbARF10/16/21 in sorghum (Chen et al. 2019), 14 GmARFs in Glycine max (Ha et al. 2013), BdARF15/17/18 in Brachypodium distachyon (Liu et al. 2018a, b), majority of EgARF (19 out of 23) in Oil palm (Jin et al. 2022), and 23 MsARF genes in M. sativa (Chen et al. 2023), were changed their expression under various abiotic treatments (such as salt and drought stresses). In this study, numerous cis-elements of each CpARF gene promoter involved in phytohormone and stress response, including 88 MeJA, 77 ABA, 25 SA, 23 auxin, and 18 GA responsive elements, were identified, suggesting that CpARFs may participate in phytohormone stimuli and stress defense. Especially, a total of 77 ABRE elements were found in 24 CpARFs, indicating that these genes may be related to drought stress response (Fig. 4). Moreover, GO, KEGG, and PPI analysis showed that CpARF genes were related to growth, development, and abiotic stresses (Fig. 6a–c; Supplementary Table S3). The expression pattern of 34 CpARF genes from RNA-seq data showed that most of the CpARF genes (24) have s significant response to drought stress (P < 0.05) at some or all times of 2 h, 12 h and 48 h under moderate drought treatment (Fig. 8a, Supplementary Table S5). Among these drought-responsive CpARF genes, the most of them (16) were down-regulated, and three (CpARF3/13/20) were up-regulated. The results were consistent with those reported in other species, such as EgARFs in Oil palm (Jin et al. 2022), NsARFs in Nitraria sibirica (Liu et al. 2022), MsARFs in M. sativa (Chen et al. 2023), and ClARFs in Coix (Zhai et al. 2023), and the expression level of most ARF genes were down-regulated under drought stress. It has been reported that most of BdARFs were inhibited by NaCl and PEG6000 stress (Liu et al. 2018a, b). In the present results, the majority of CpARF genes (27) could respond to salt stress at high, middle, and low under T1 salt treatment, of which 16 were significantly up-regulated, and 7 were significantly down-regulated (Fig. 8a, Supplementary Table S6). The results are generally consistent with the previous results, such as BdARFs in B. distachyon (Liu et al. 2018a, b), JcARFs in Jatropha curcus (Tang et al. 2018), AcARFs in Actinidia chinensis (Su et al. 2021), indicating that ARF genes must also involve in salt stress response. It should be noted that the number of ARF genes affected by drought and salt stress in this study is higher than that of other species, possibly due to the sensitivity of C. paliurus to drought and salt stress, or possibly due to the longer salt treatment time of C. paliurus. Totally, our results provide a basic information for exploring the molecular and functional roles of ARF genes in C. paliurus.

Additionally, subcellular localization showed that both the selected CpARF16 and CpARF32 were localized to the nucleus, which was consistent with the previous predictions, suggesting that CpARFs likely act as transcription factors and co- regulate the expression of stress resistance genes with other factors.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ARF

Auxin response factor

DBD

DNA-binding domain

AuxREs

Auxin-responsive elements

PI

Isoelectric point

Aux/IAA

Auxin/indole-3-acetic acid

MW

Molecular weight

ABA

Abscisic acid

SA

Salicylic acid

qRT-PCR

Reverse transcription-quantitative PCR

Author contribution

ZG, YW, ML, LD and DW conceived and designed the experiments. ZG, JL, YL, and YH performed the experiments; JL, YC, and YH validated the data; QJ and DW revised the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This research work is financially supported by Key Research and Development Program of Zhejiang Province (2023C02017); Chinese herbal medicine industry technical team project of Zhejiang Agricultural and Rural Department (N2), and the Zhejiang Sci-Tech University scientific research fund (19042142-Y).

Data availability

All data and materials used in this research are publicly available. Other supporting data are provided as supplementary files with the manuscript.

Declarations

Conflict of interest

The authors declare no competing interests.

Institutional review board statement

Not applicable.

Informed consent statement

Not applicable.