Molecular characterization of the SAUR gene family in sweet cherry and functional analysis of PavSAUR55 in the process of abscission

HOU Qian-dong ,HONG Yi ,WEN Zhuang ,SHANG Chun-qiong ,Ll Zheng-chun ,CAl Xiao-wei ,QlAO Guang,WEN Xiao-peng#

1 Key Laboratory of Plant Resource Conservation and Germplasm Innovation in Mountainous Region (Ministry of Education),College of Life Sciences/Institute of Agro-bioengineering, Guizhou University, Guiyang 550025, P.R.China

2 College of Forestry/Institute for Forest Resources &Environment of Guizhou, Guizhou University, Guiyang 550025, P.R.China

Abstract Small auxin up RNA (SAUR) is a large gene family that is widely distributed among land plants.In this study,a comprehensive analysis of the SAUR family was performed in sweet cherry,and the potential biological functions of PavSAUR55 were identified using the method of genetic transformation.The sweet cherry genome encodes 86 SAUR members,the majority of which are intron-less.These genes appear to be divided into seven subfamilies through evolution.Gene duplication events indicate that fragment duplication and tandem duplication events occurred in the sweet cherry.Most of the members mainly underwent purification selection pressure during evolution.During fruit development,the expression levels of PavSAUR16/45/56/63 were up-regulated,and conversely,those of PavSAUR12/61 were down-regulated.Due to the significantly differential expressions of PavSAUR13/16/55/61 during the fruitlet abscission process,they might be the candidate genes involved in the regulation of physiological fruit abscission in sweet cherry.Overexpression of PavSAUR55 in Arabidopsis produced earlier reproductive growth,root elongation,and delayed petal abscission.In addition,this gene did not cause any change in the germination time of seeds and was able to increase the number of lateral roots under abscisic acid (ABA) treatment.The identified SAURs of sweet cherry play a crucial role in fruitlet abscission and will facilitate future insights into the mechanism underlying the heavy fruitlet abscission that can occur in this fruit crop.

Keywords: sweet cherry,small auxin up RNA,gene family,expression profile,fruitlet abscission

1.lntroduction

Three special gene families in the auxin response signaling pathway are Aux/IAA (auxin/indole acetic acid),GH3 (Gretchen Hagen 3),and SAUR (small auxin up RNA),which are called the early auxin responsive genes (Luoet al.2018).These genes can respond to auxin induction in a very short time,such as the changes inSAURwithin 1 h after exogenous auxin treatment(Huet al.2018).SAURis a large gene family with many members.The firstSAURgene was discovered in the hypocotyl of soybean,and many additionalSAURgenes have been isolated from other species includingArabidopsisthaliana,rice,tomato,and apple(Stortenbeker and Bemer 2019).Many promoters ofSAURgenes contain at least one auxin response element(AuxRE) and downstream destabilizing element,30 of which were investigated inArabidopsis(Ren and Gray 2015).

TheSAURgene is involved in regulating plant growth and development,and some studies have found that it may promote cell elongation (van Mouriket al.2017).In soybean,the differential expression ofSAURmediated the gravitropism and phototropism movements of hypocotyls (Wang Xet al.2020).The expression of severalSAURmembers inArabidopsisinhibited the activity of protein phosphatase (PP2C.D,the D subfamily type 2C protein phosphatases) and induced AHAs(ArabidopsisH+-ATPases) to stimulate cell expansion(Ren and Gray 2015).InArabidopsis,SAUR19–24aggregated into the same subfamily and were extremely unstable;and the fusion protein overexpressingSAUR19showed the auxin-related phenotype indicative cell expansion,including hypocotyl elongation and leaf size increase (Spartzet al.2012).Ethylene and auxin might induce the expression ofSAUR76–78,and the overexpression of these genes promoted the growth of seeds;and compared with the wild type,SAUR76–78exhibited an elevated sensitivity to ethylene (Liet al.2015).The closure and opening of cotyledons before and after de-etiolation was specifically regulated bySAUR16andSAUR50,and influenced by light,which regulated the elongation of hypocotyls (Donget al.2019).SAUR is an unstable protein that is mainly distributed in the cell membrane (Spartzet al.2012;Liet al.2015).Recently,nine SAURs were shown to be related to the morphological formation and development of flower sepals in tomato (Liuet al.2020).Among the 70SAURgenes in citrus,23 are sensitive to IAA;and during citrus fruitlet dropping,14 members showed differential expression (Xieet al.2015).After the girdling and leaf removal that promote fruitlet drop,the expression ofLcSAUR1increased in the abscission zone of the litchi fruitlet,indicating that this gene might play an important role in the shedding process (Kuanget al.2012).

Sweet cherry (PrunusaviumL.) is an important economic fruit tree that is widely cultivated all over the world.During the past decade,it has been expanding rapidly in southern China.However,abnormal fruit dropping poses a dilemma for sweet cherry production,especially in southern China.In the process of fruit setting and small fruit development of sweet cherry,fruitlets are susceptible to the external environment,leading to non-physiological fruit abscission and ultimately to yield reductions and economic losses.SAURs participate in the regulation of fruit abscising in various fruits,such as citrus and litchi.In the transcriptional data of the abscising carpopodiums in sweet cherry,theSAURgene was found to be significantly differentially expressed (Qiuet al.2021).The genome sequences of the sweet cherry released in 2017 provide crucial biological information for genome-wide identification of gene families and other related genetic studies (Shirasawaet al.2017).To better understand the roles ofSAUR,the characteristics,evolution,and expression levels ofSAURgenes at the whole genome level of sweet cherry were carried out in this study.In addition,the functional verification ofPavSAUR55was also conducted to further elucidate the potential regulatory function of theSAURgene in the process of abscission,which may facilitate molecular-level insights into the mechanism underlying the heavy fruitlet abscission in this fruit species.

2.Materials and methods

2.1.ldentification of the SAUR family members in the sweet cherry genome

The genomic sequence,protein sequence and coding sequence (CDS) of sweet cherry were downloaded from the Genome Database Rosaceae (GDR,https://www.rosaceae.org/).In total,79 protein sequences of theAtSAURgene family present inArabidopsiswere obtained from theArabidopsisInformation Resource (TAIR,https://www.arabidopsis.org/index.jsp),and the hidden Markov model profile (HMM,Auxin_inducible,PF02519) of theSAURfamily was downloaded from the Pfam database(http://pfam.xfam.org/).The 79AtSAURsequences were used as a query to perform BLASTP with an e-value of 1e–10searched against the sweet cherry genome sequences to obtain the candidate sweet cherrySAURmembers.To make the results more accurate,HMM was used as the query with an e-value of 1e–20to search the whole cherry protein sequence data.After the redundant sequences were removed from the two sets of search results,Conserved Domain Database (CDD,https://www.ncbi.nlm.nih.gov/cdd/) and Pfam were used to determine the conservative auxin-inducible domain.All of the genes were assigned asP.aviumSAURs (PavSAURs).These genes were named based on their positions on the chromosomes (PAV_r1.0chr).The numbers of amino acids (aa),molecular weights (MW),theoretical isoelectric points (pI) and grand averages of hydropathicity(GRAVY) of the PavSAURs were calculated using the Expasy ProtParam tool (ExPASy,https://web.expasy.org/protparam/).

2.2.Gene structure and conserved motif analysis

The gene exon/intron structures were analyzed and visualized using the TBtools (ToolBox for Biologists)by comparing the generic feature format file (Chenet al.2020).The online program Multiple Em for Motif Elicitation (MEME,version 5.3.3,http://meme-suite.org/index.html) was used to identify PavSAUR motifs with the site number of motif set to 5 (Liet al.2017),the expected motif sites to be distributed in sequences were set as Zero or One Occurrence Per Sequence (Zoops),and the motif width was set to between 6 and 50 wide (Baileyet al.2009).

2.3.Multiple sequence alignment and phylogenetic analysis of SAURs

The protein sequences ofA.thaliana,OryzasativaandP.aviumwere used for multiple sequence alignment,and the alignment was performed using Clustal W under default parameters.Phyre 2 (http://www.sbg.bio.ic.ac.uk/phyre2/html/) and SWISS-MODEL (https://swissmodel.expasy.org/) were employed for secondary and tertiary structure predictions.Based on the results of multiple sequence alignment,MEGA 7 Software was used to construct a phylogenetic tree under the neighborjoining method,Poisson Model,1 000 bootstraps and other default parameters.The phylogenetic tree was esthetically improved using the EvolView tool (https://www.evolgenius.info/evolview/).

2.4.Synteny analysis and chromosome localization

PavSAURmembers were mapped on the chromosomes using Mapgene2chrom (v2.1,http://mg2c.iask.in/mg2c_v2.1/).Gene synteny was assessed through the Multiple Collinearity Scan toolkit (MCScanX) and the results were achieved using TBtools (Wanget al.2012;Chenet al.2020).The syntenic blocks were used to construct a synteny analysis map within theSAURmembers ofP.avium,as well as between the SAURs ofP.aviumand those from theA.thaliana,O.sativa,Glycine max,Solanumlycopersicum,Malusdomestica,Citrus sinensis,andPrunuspersicagenomes.Gene duplication events of the PavSAURs were confirmed based on two conditions: the alignment region covered more than 70% of the longer gene and the identity of the aligned regions was >70% (Yanget al.2008).To explain the patterns of macroevolution for each paralogous gene,the nonsynonymous substitutions rate (Ka) and synonymous substitution rate (Ks) were calculated using the KaKs_Calculator 2.0.Based on the Ka/Ks ratio,a Ka/Ks ratio >1 indicates positive selection,a Ka/Ks ratio of 1 indicates neutral selection,and a ratio that is <1 indicates purifying or negative selection.The evolutionary duplication time(T) was calculated as T=Ks/2λ×10–6(Mya),where λ is taken to be 1.5×10?8synonymous substitutions per site per year for dicotyledonous plants (Kochet al.2000).

2.5.Cis-element analysis

The 2 000 bp upstream sequences of thePavSAURgenes were selected and submitted to PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) for analysis of thecis-acting elements.

2.6.Analyzing the expression profiles of Pav-SAURs using RNA-seq and quantitative real-time PCR

To identify the expression patterns ofPavSAURgenes in the abscising carpopodium of sweet cherry,the RNAseq expression data from ‘Santina’ were downloaded from NCBI-SRA (https://www.ncbi.nlm.nih.gov/).The raw expression reads data from abscising carpopodiums and retention carpopodium of ‘Santina’ were retrieved from NCBI access number PRJNA636209 (Qiuet al.2021).Furthermore,RNA-seq expression data from various time points throughout the fruit development of sweet cherry cultivars ‘13–33’ (yellow) and ‘Tieton’ (red) were retrieved from NCBI Bio-project SRP044388 (Weiet al.2015).In addition,the transcriptome data of sweet cherry (P.avium cv.‘Bing’) flowering induction and floral bud differentiation were obtained from NCBI Bio-project PRJNA529895(Binget al.2020).The FPKM (fragments per kilobase of transcript per million fragments mapped reads) value of each gene was calculated using Hisat2 and Stringtie software.The heat map was constructed to visualize the expression through TBtools with a log2(FPKM+1) and normalized row scale (Chenet al.2020).The VENNY 2.1 tool (https://bioinfogp.cnb.csic.es/tools/venny/index.html) was used to identify the FPKM>1 gene numbers in different stages.Based on the homologous genes ofA.thaliana,STRING (https://string-db.org/) was used to predict the interactions of sweet cherry SAURs.

Seven-year-old cherry trees were grown in rain shelters in Weining County (104°12′E,27°25′N),Guizhou Province,China.The total RNA was extracted from the retention (size 0.5–0.7 cm) as well as abscising fruitlets(size 0.5–0.7 cm) from the sweet cherry cultivar ‘Brooks’.A total of 18 PavSAURs that were differentially expressed in the abscising carpopodium RNA-seq were selected for comparing their expression levels in the retention fruitlets and abscising fruitlets by qPCR analysis using the primers listed in Appendix A.The cDNA first strand was synthesized using the PrimeScriptTMRT reagent Kit(TaKaRa,Dalian,China) and the product was diluted 10-fold by adding ddH2O.The qPCR was performed on the CFX ConnectTMReal-Time System (BIO-RAD,USA)with PowerUpTMSYBRTMGreen Master Mix (Applied Biosystems,China) with a thermal cycling condition of:95°C for 30 s,followed by 40 cycles of 95°C for 3 s,and 60°C for 30 s.Three technical replications were applied and the expression levels were calculated using the 2–ΔCtmethod (Schmittgen and Livak 2008).

2.7.Cloning of PavSAUR55 and function identification

Amplification primers (Appendix A) containing the full length of the entire coding region were designed according to the sequence ofPavSAUR55in the genome.Prime Star Max (TaKaRa,Dalian,China) was used for PCR amplification and ligated to the BLUNT vector for sequencing.The pCambia-35S was double digested withKpnI andBamHI (TaKaRa,Dalian,China),andPavSAUR55and the vector were ligated with T4 ligase (TaKaRa) to form a pCambia-35S-PavSAUR55recombinant plasmid.After the recombinant plasmid was introduced intoAgrobacteriumtumefaciensstrain GV3101 competent cells using the heat-shock method,the wildtype Columbia typeA.thaliana(Col-0) was genetically transformed using the tidbit infection method.The positive plants were screened successively for the T3generation by the hygromycin (Hyg) and PCR methods.Young leaves,old leaves,flower buds,flowers,turning color period fruit,mature fruit and annual stems were selected for the expression analysis ofPavSAUR55.For determining the germination time and germination rate,about 40 seeds were spotted on the same 1/2 MS plate and observed for germination activity.The Col-0 and over-expression (OE)lines were spotted on 1/2 MS plates (six plants per line)for observation.Further,75 mmol L–1NaCl (seven plants per line),250 mmol L–1mannitol (six plants per line) and 0.2 mmol L–1abscisic acid (ABA) (five plants per line) were added for the treatments.After 7 days of growth,the root length of each plant was measured.In addition,the lateral roots treated with ABA were also counted.AllA.thalianawere grown in an incubator,and RNA extraction and cDNA first-strand synthesis were performed on the Col-0 and OE lines that grew in the same way for 14 days.Transgene expression was assessed using qRT-PCR primers.Based on the previous study (Renet al.2018),the expression changes of thePP2C.Dgene inA.thalianawere further analyzed.The state of normal growth in nutrient soil for 20 days after seed germination was observed and counted(for eight plants per line).The position of the flower was counted from the first white flower at the top,and counting was conducted from top to bottom to the ninth position.

3.Results

3.1.ldentification of the SAUR genes in sweet cherry

A total of 86 SAURs were obtained by HMM analysis,and 79 sequences were found by local BLASTP.After removing redundant sequences,the remaining sequences were submitted to CDS and Pfam to verify the conserved domains.As a result,86SAURgenes were identified in theP.aviumgenome and namedPavSAUR1toPavSAUR86according to their distribution on the chromosomes.Gene names,gene IDs,CDS lengths,chromosomal locations,and the values for amino acid length,MW,pI,and GRAVY are listed in Appendix B.The lengths of thePavSAURgenes ranged from 273(PavSAUR34/73/74/75/82) to 5 054 bp (PavSAUR35);CDS ranged from 225 (PavSAUR81) to 855 bp(PavSAUR10),and amino acid numbers ranged from 74 to 284.The MWs of these PavSAUR proteins ranged from 1.38 kDa (PavSAUR76) to 31.49 kDa (PavSAUR10),and their pIs ranged from 1.78 (PavSAUR49) to 10.64(PavSAUR64).The 61PavSAURmembers were unevenly distributed on eight chromosomes (Appendices B and C).

3.2.Gene structure and protein motif analysis of PavSAURs

To gain some insight into the diversification of theSAURgenes in sweet cherry,the conserved motifs and exonintron organization were further analyzed (Fig.1).The sequence and width information of the conserved motifs is shown in Appendix D.As shown in Fig.1-B,a total of five kinds of Motifs were obtained,of which Motifs 1 and 2 were found in most members,and only the four genes (PavSAUR26/27/28/81) did not possess Motif 2.PavSAUR27may be a special gene,which is evolutionarily far apart from the other members,and it contains none of these five kinds of Motif.In addition,except forPavSAUR81,which has only one conserved motif (Motif 1),all the other members have two or more conserved motifs.From an evolutionary point of view,the structures of these genes are similar.Members of the same class have roughly the same Motif features,and only a few members have changed.Motif 1 is conserved in the sequences of SAUR fromArabidopsis,rice,and sweet cherry.Motif 5 exists only inPavSAUR6–11,which may be unique to sweet cherry (Figs.1 and 2-A).

Fig.1 Phylogenetic tree (A),conserved motif distribution (B),and exon–intron structures (C) of the SAUR family members in Prunus avium.Motifs,coding sequence (CDS) and untranslated regions (UTRs) are represented by differentially colored boxes.

Fig.2 SAUR family conservative Motif logos (A),multiple sequence alignment (B) and protein tertiary structure predictions (C).In Fig.2-B,the long black lines represent the SAUR domains,which are composed of Motif2/1/3.α-helices are displayed as squiggles.β-strands are rendered as arrows,with strict β-turns as TT letters and strict α-turns as TTT.

Gene structure can somewhat reflect gene duplication or insertion events in the process of gene evolution,and it may also be more intuitive for reflecting the differences between genes.The gene structures herein showed that most members ofPavSAURdo not contain introns,and only a few members have one or two introns (Fig.1-C),probably leading to the shorter gene lengths for members of this gene family.PavSAUR35is special,with longer introns resulting in the longest gene length.Combined with the conservative motif analysis,this gene may have undergone fragmented replication events during evolution.

3.3.Phylogenetic analysis of the SAUR gene family

To systematically investigate the molecular evolution of theSAURfamily,a phylogenetic tree was constructed from three species based on the full-length protein sequences of 79 AtSAURs,56 OsSAURs and 86 PavSAURs.A total of 221 SAURs were clustered into seven groups (A–G) (Fig.3).PavSAURmembers are distributed in all groups,and Group A only hasPavSAURandAtSAURmembers.Groups A,D,and G have more members,and Group A ranks first.In general,regardless of the phylogenetic tree or the number of gene members,the PavSAURs are genetically closer to AtSAURs in comparison with OsSAURs,which is also consistent with the fact that both are dicotyledons.

Fig.3 Phylogenetic analysis of SAUR family members in Prunus avium (Pav),Arabidopsis thaliana (At) and Oryza sativa (Os).Different colors represent the seven sub-families (A–G).The tree was constructed in MEGA 7.0 using the neighbor joining method with 1 000 bootstraps.

The multiple sequence alignments of the proteins indicate that the SAUR proteins are considerably conservative,especially in the position of Motif 1,which may be the core region for SAUR functions (Fig.2-B).There are α helices and β turns in the secondary structures of the proteins,and the tertiary structure of a protein lays the foundation for realizing the biological functions of genes.The tertiary structure of the sweet cherry SAUR members is uniform,and also relatively simple,which may be the basis for the rapid realization of the function of SAURs (Fig.2-C).In addition,this may also be the reason for the short half-lives of SAURs in cells.

3.4.Gene duplication and genome synteny

The 61PavSAURmembers were unevenly distributed on eight chromosomes,and 25 members could not be assigned to any chromosome of the reference genome due to technical difficulties (Appendix C).The collinearity analysis,including whole genome duplication,segmental duplication and tandem duplication,was performed for comparisons inP.aviumby multiple alignments of the SAURs,in order to expound the mechanism underlying the expansion and duplication of theSAURfamily in sweet cherry.In total,96 orthologous gene pairs were identified,14 of which were predicted to be in tandem duplication.There were three tandem-duplicated gene pairs on chromosome 2 (PavSAUR10-PavSAUR11,PavSAUR7-PavSAUR6andPavSAUR15-PavSAUR14);one tandemduplicated gene pair on chromosome 6 (PavSAUR34-PavSAUR35);and five tandem-duplicated gene pairs on chromosome 8 (PavSAUR43-PavSAUR42,PavSAUR53-PavSAUR54,PavSAUR51-PavSAUR52,PavSAUR47-PavSAUR46andPavSAUR43-PavSAUR42).Five tandem repeat genes were located on chromosome 0.The expansion patterns of thePavSAURgenes were analyzed using circos analysis,which revealed the duplication of various genes.In addition to tandem repeat genes,76 pairs of segmental-duplicated genes or whole genome duplications were also investigated in theP.aviumgenome (Fig.4-A).On the whole,some genes,such asPavSAUR10,PavSAUR11,PavSAUR7andPavSAUR6,have tandem repeat gene pairs in addition to segmentalduplicated gene pairs.These paralogous gene pairs might have played an important role in the evolutionary process,and probably came from the same ancestral gene.

Fig.4 Synteny relationships (A) and pairwise comparisons of the synonymous substitution rate (Ks) values for SAUR genes (B)in Prunus avium.Different colors are used to mark the different chromosomes.A,the red lines represent the duplication events in the same chromosome,and the green lines represent the duplication events in different chromosomes.For genes that are not assembled on the chromosomes (chromosome 0),the blue lines are used to represent duplication events.B,the lines representing the SAUR genes are red,and box plots show the Ks values for the SAURs in P.avium.

Selective pressure and the duplication types ofPavSAURparalogous gene pairs were studied by their synonymous and nonsynonymous rates.The Ka/Ks ratio predicts the evolutionary event of a gene region or gene.In the current study,the Ka/Ks ratios of homologous gene pairs ranged from 0.1 to 50,and most gene pairs had values of <1 (Appendix E),which suggested that they had experienced purifying selection pressure and only limited functional divergence occurred after segmental duplication.Only four pairs of genes(PavSAUR65-PavSAUR77,PavSAUR7-PavSAUR6,PavSAUR11-PavSAUR7andPavSAUR15-PavSAUR14)gave ratios of >1,indicating that these gene pairs might have experienced relatively rapid evolution following duplication.No gene pairs gave ratios equal to 1.To estimate the divergence time,the Ks values of theP.aviumSAURgenes ranged from 0.00007 to 1.5 and were focused around approximately 0.22 (～7.33 Mya)(Fig.4-B;Appendix E).Based on the Ks value,the latest time for the separation of theSAURgene in sweet cherries might be about 24 000 years ago,and the earliest event occurred about 49.8 million years ago.

To further understand the phylogenetic mechanisms of theSAURfamily,we constructed six comparative syntenic maps ofP.aviumassociated with six representative species: monocotyledonousO.sativaandZeamays,and dicotyledonousA.thaliana,G.max,C.sinensisandP.persica.ThePavSAURgenes were homologous to genes in the other plant species,and syntenic conservation was observed amongO.sativa(eight orthologous gene pairs in chromosomes 1,2,3,5 and 7),Z.mays(seven orthologous gene pairs in chromosomes 5 and 7),A.thaliana(40 orthologous gene pairs),G.max(68 orthologous gene pairs),C.sinensis(19 orthologous gene pairs) andP.persica(55 orthologous gene pairs that were dispersed).In the dicotyledonous genomes,one or moreSAURgenes in the sweet cherry chromosomes have a collinearity relationship (Fig.5).Some gene members,such asPavSAUR40,have four pairs of collinear relationships with the soybean genome,further indicating that these members might have undergone a large number of replication events during the evolutionary process,and these gene pairs may have come from the same ancestral gene (Fig.5).

Fig.5 Synteny relationships of SAUR gene pairs among Prunus avium,Oryza sativa,Zea mays,Arabidopsis thaliana,Glycine max,Citrus sinensis and Prunus persica.Green represents sweet cherries,and the other species are represented by different colors.The red line indicates the collinearity that exists between species.

3.5.Promoter analysis of the PavSAURs members

The 2 000 bp sequence upstream of the 86PavSAURgenes was used to analyze thecis-acting elements.The results revealedcis-acting elements related to phytohormone metabolism,abiotic stress,secondary metabolite synthesis and organogenesis in thePavSAURpromoter region,indicating that this gene family may play an important role in the growth and development of sweet cherry (Appendix F).

Among the hormone-related elements,many responding to ABA and methyl jasmonate,and auxin responsive elements exist in the promoter regions of most members.Among them,PavSAUR16/72have many elements that respond to ABA and methyl jasmonate.In addition,response elements related to gibberellin and salicylic acid were also investigated.Among the elements involved in stress resistance,there are main elements responding to low temperature and drought,and many elements related to anaerobic induction.

3.6.Expression profiles of the PavSAURs

Gene expression levels may reflect the potential biological functions of PavSAURs.The expression profiles of PavSAURs in the abscising carpopodium (CA)and retention carpopodium (CN) of sweet cherry were investigated using RNA-seq data.The results showed that 17PavSAURmembers were expressed in the petiole of the sweet cherry fruitlet (Fig.6-A),among which,PavSAUR13/16/25/26were highly expressed.Compared with CN,PavSAUR13/16/25/39were up-regulated in CA,suggesting they might function as the positive regulators in fruitlet abscission;whilePavSAUR26/55were downregulated so they might play a role as negative regulators.The changes in expression levels indicate that these genes may play an essential regulatory role in the abscission of sweet cherry fruitlets.

Fig.6 Expression of SAUR genes in sweet cherry carpopodium RNA-seq.A,hierarchical clustering analysis of PavSAUR expression levels in abscising carpopodium (CA) and non-abscission carpopodium (CN).B,predicted protein–protein interaction networks of PavSAUR13/16 using STRING.C,predicted protein–protein interaction networks of PavSAUR55 using STRING.

To further analyze the functions of SAURs,based on an in-depth study of the interaction network of homologous genes inArabidopsis,the online tool STRING was used to analyze the interactions of a single PavSAUR protein sequence.The results demonstrated that the two up-regulated genes (PavSAUR13/16) in abscising carpopodium were homologous toArabidopsisAT2G46690 (SAUR-like),and their identities were 68.2 and 78.4%,respectively.These two genes interact with some auxin-related genes (such as AUX/IAA and LAX2)and other SAURs,suggesting that they may be involved in the auxin response process (Fig.6-B).PavSAUR55is a down-regulated gene,homologous toArabidopsisAT2G21220 (80.6% identity),that interacts with otherSAUR-likegenes (AT5G20820,AT1G72430,AT1G17345 and AT3G12955),indicating that this gene may be widely involved in the early auxin response signal pathway(Fig.6-C).The change in auxin is one of the significant factors for plant organ shedding.Therefore,as an early auxin response gene,PavSAURs may have an important function in the auxin response pathway and participate in the formation of the abscission zone.

Based on RNA-seq data (Weiet al.2015),the expression profiles of PavSAURs were investigated in order to unravel their possible biological functions in the different fruit development stages.Transcript abundance was obtained using FPKM values and these transcripts were assembled hierarchically in a heat map (Fig.7-A).The results demonstrated that 55PavSAURmembers were differentially expressed in the four developmental stages of the fruit.In R-S1 of red (R) fruits (Tieton),PavSAUR1/2/12/24/58/60gave higher transcription levels than in other developmental stages.In R-S3,PavSAUR3/36/84were also up-regulated.In the yellow(Y) fruit Y-S1,PavSAUR13/14/61demonstrated higher transcription levels,and starting from S2,more gene expression levels were maintained in a higher expression state.In the four stages of the two cultivars,the genes with FPKM values of >1 included 24 genes in R-S1,28 genes in R-S2,29 genes in R-S3,30 genes in R-S,24 genes in Y-S1,35 genes in Y-S2,35 genes in Y-S3,and 32 genes in Y-S4 (Fig.7-B).Among them,15 and 19 genes had FPKM >1 in the red fruit and yellow fruit development stages,respectively (Fig.7-C),and the FPKM values of 14 members were greater than 1 in these eight organizations (Fig.7-D).The FPKM value ofPavSAUR16was <1 only in R-S1,and all the values in the other seven tissues were >1,so the higher FPKM value indicates that this gene may play an important role in the fruit development of sweet cherry.In addition,PavSAUR55had higher FPKM values in all eight samples.Therefore,these genes may play a significant role in the development of sweet cherry fruit.

Fig.7 Expression of SAUR genes in fruits of sweet cherry at different developmental stages by RNA-seq.A,hierarchical clustering analysis of PavSAUR expression levels in sweet cherry cultivars ‘Tieton’ (red,R) and ‘13–33’ (yellow,Y),where S1 to S4 represent the four different periods of fruit development.B,number of genes with FPKM greater than 1.C,an overview of SAUR genes with values greater than 1 in R or Y at the four fruit development stages.D,an overview of the number of SAURs with FPKM greater than 1 in the two varieties.

The gene expression levels were also evaluated during flowering induction and floral bud differentiation(Fig.8).The results showed that the transcription levels of theSAURgenes fluctuated greatly at the different stages (S1–S4 and D) of flower bud differentiation,implying that theSAURgene has an important regulatory effect on flower bud differentiation (Fig.8-A).And 65 members are expressed in these organizations.Five(PavSAUR10/11/31/29/44) showed higher transcription levels in S1,andPavSAUR48was highly expressed in S2.PavSAUR54andPavSAUR69were up-regulated in S3.In the D stage (June floral bud sample,dormant bud),only 12 members gave FPKM values greater than one,suggesting that most of the genes were negatively regulated during this period (Fig.8-B).In the four stages of flower bud differentiation,15 genes have FPKM>1,indicating that these genes may play a positive regulatory role in the process of flower bud differentiation (Fig.8-C).

Fig.8 Expression of SAUR genes by RNA-seq during flower bud differentiation in sweet cherry.A,hierarchical clustering analysis of PavSAUR transcription levels in Prunus avium (‘Bing’) flowering induction (L) and floral bud stages (S1–S4,D),where S1–S4 denote sampling dates (S1,December 16;S2,January 15;S3,February 15;and S4,March 15) and D represents dormant flower buds.B,FPKM>1 number in six diverse samples.C,an overview of the numbers of SAURs with FPKM>1 in the four stages of floral buds.

To better understand the role ofSAURexpression in sweet cherry fruit droppings,16 genes that are differentially expressed in the carpopodium of sweet cherry were selected for qRT-PCR analysis using the retention and dropped fruits.The results showed that these members were differentially expressed in the dropped fruits of sweet cherry (Fig.9).The expression levels ofPavSAUR2/3/5/26/55/61/63were significantly down-regulated in the two different periods of fruit abscission.Among them,PavSAUR55/61demonstrated the same expression trend as in the fruit carpopodium,both of which were down-regulated during shedding.Compared with normal fruits in the same period,PavSAUR13/16/25/39/52/53were up-regulated in both fruit dropping stages,indicating that these genes may have a positive regulatory effect on sweet cherry fruit abscission.Interestingly,the two genesPavSAUR13andPavSAUR16also showed up-regulated expression in the RNA-seq of the fruit carpopodium,suggesting that these genes may play an important role in fruit drop.Based on the evidence from both RNA-seq in the fruit carpopodium and qRT-PCR quantification in the dropped fruit,PavSAUR13/16/55/61may have an enormously crucial regulatory role in the fruit abscission of sweet cherry.

Fig.9 Expression of 16 PavSAURs in two different periods of abscission of fruitlet.A,the relative expression levels of SAUR genes expressed in fruit carpopodium in two different stages of fruitlet abscission.Data represent the mean±SD of three samples.＊ and ＊＊,P<0.05 and P<0.01,respectively.B,the morphology of two different stages of fruitlet abscission.

3.7.Analyzing transgenic Arabidopsis plants overexpressing PavSAUR55

To further explore the functions ofSAURin sweet cherry,PavSAUR55was overexpressed inA.thaliana.Nine transgenic overexpression (OE)PavSAUR55lines were obtained (Fig.10-A),among which the three with the highest expression levels were OE3,OE5,and OE8(Fig.10-C).Seven different organizations in sweet cherries were selected for qPCR testing.As a result,PavSAUR55showed the highest expression in mature fruits,which was higher than in the turning color period fruits (Fig.10-B),suggesting that this gene may be associated with fruit ripening.In addition,its expression was barely detectable in aging leaves (Fig.10-B),suggesting its possible involvement in the regulation of leaf senescence.Generally,SAUR interacts with PP2C.D and promotes cell expansion by inhibiting the activity of PP2C.D (Spartzet al.2014).In this case,the expression levels ofPP2C.D1,PP2C.D3,PP2C.D3,PP2C.D5,andPP2C.D6were all changed in comparison with the wild type (Col-0) in the three most overexpressed lines(Fig.10-D),indicating that these genes might interact withPavSAUR55.Notably,althoughPP2Cgene expression was up-regulated in the OE lines,the expression levels of these genes were still low relative to that ofPavSAUR55(Fig.10-C and D).After growth on 1/2 MS plates,the three OE lines showed no significant differences in seed germination time compared to Col-0 (Fig.10-E and F).Furthermore,the phenotypic observations of the T3generation of lines OE3,OE5 and OE8 showed that the overexpression ofPavSAUR55could substantially promote the early flowering ofA.thaliana(Fig.10-G and H).Interestingly,the numbers of rosette leaves of the Col-0 and OE lines were not significantly different (Fig.10-I).

Fig.10 Genetic transformation of Arabidopsis thaliana with PavSAUR55.A,PCR positive detection of transgenic plants.M represents the D2000 marker;ddH2O is a negative control,recombinant plasmid is a positive control (vector),and Col-0 is a wild-type control.B,expression levels of PavSAUR55 in seven different tissues.C,expression levels of PavSAUR55 in transgenic plants.D,expression of the PP2C.D genes in transgenic (OE) and wild-type Arabidopsis thaliana (Col-0).Data represent the mean±SD of three repetitions,and error bars indicate standard deviation.＊ indicates significance at P<0.05.E,Arabidopsis seeds were grown on 1/2 MS medium for 10 days to observe the status.F,germination time statistics.Data represent the mean±SD of 35–42 seeds,and error bars indicate standard errors.G,phenotypic observations of OE plants(normal growth for 25 days after seed germination).H,statistics of A.thaliana flowering time for growth 25–32 days after germination.Data represent the mean±SD of eight plants.I,statistics on the number of rosettes in A.thaliana during growth for 25 days after germination.Data represent the mean±SD of eight plants.The error bars represent the standard deviation.＊ indicates significance at P<0.05.

To explore the root phenotype of transgenicA.thalianaand the growth status under different stress treatments,trials were conducted with NaCl and mannitol added to the 1/2MS.In the 1/2MS medium,the root lengths of the three OE lines were significantly higher than that of Col-0(Fig.11-A and B).In addition,the root lengths of the OE lines were significantly longer than that of Col-0 after exposure to either 75 mmol L–1NaCl or 250 mmol L–1mannitol (Fig.11-C and F),suggesting thatPavSAUR55may be involved in root development and enhance the tolerances to salt and drought stress by improving root elongation.Interestingly,under the 0.2 mmol L–1ABA treatment,there was no significant difference in root length between OE and Col-0,which may indicate that the ABA compensated for thePavSAUR55overexpression(Fig.11-G).However,the lateral root lengths of the OE lines were obviously increased under ABA treatment(Fig.11-G and H).

Fig.11 Growth of over-expression (OE) lines and wild-type Columbia type Arabidopsis thaliana (Col-0) for 7 days under different treatments.A and B,growth and root length comparisons of OE line and Col-0 in 1/2MS medium.Data represent the mean±SD of six plants.C and D,growth phenotype observations and root length comparison of OE line and Col-0 in 1/2 MS medium supplemented with 75 mmol L–1 NaCl.Data represent the mean±SD of seven plants.E and F,growth observations and root length comparison of OE lines and Col-0 in 1/2 MS medium supplemented with 250 mmol L–1 mannitol.Data represent the mean±SD of six plants.G and H,growth and number of lateral roots of A.thaliana treated with 0.2 mmol L–1 abscisic acid (ABA).Data represent the mean±SD of five plants.＊ indicates significance at P<0.05.

SincePavSAUR55is down-regulated in abscission fruit and fruit carpopodium,the petal phenotype of the T3generationA.thalianawas further examined to investigate the potential regulatory role of this gene in abscission.The results showed that the petals in the OE line all shed later compared to those in Col-0(Fig.12).Specifically,neither the OE lines nor the Col-0 phenotype were significantly altered in positions 1–4(Fig.12-A).From positions 4 to 5,however,the petal shedding of Col-0 was remarkably accelerated.In the three OE lines,the petal changes from positions 4 and 5 were still especially noticeable (Fig.12-B).At positions 6 and 7,the petals on the Col-0 pods had almost fallen off,while the petals in the OE lines began to fall off at position 5.At position 8,the Col-0 petals had completely detached,but there were some petals left in the three OE series.

Fig.12 Observations of petal status (A) and the statistics of abscission position (B) in the over-expression (OE) lines and wild-type Columbia type Arabidopsis thaliana (Col-0).A total of nine positions (1–9) of petals were observed and counted in A.Data in B represent the mean±SD of four plants.

4.Discussion

4.1.Evolution of SAUR genes in the plant kingdom

The phytohormone auxin regulates plant growth and development by adjusting the basic processes of cell expansion,division,differentiation and architecture formation (Zhao 2018).As one of the early auxinresponsive gene families,SAURcan quickly respond to auxin induction and promote cell elongation (Stortenbeker and Bemer 2019).Since the firstSAURgene was isolated in soybean hypocotyls in 1987,SAURs have been discovered in a variety of species such asA.thaliana,rice (Jainet al.2006),cotton (Liet al.2017),maize (Chenet al.2014),apples (Wang Pet al.2020)and others.Previous research has indicated that these genes may regulate cell life activities,and the growth and development of tissues and organs,thereby playing a vital regulatory role in the phytohormone response process and changes in the external environment (Ren and Gray 2015).In this study,86 SAURs were identified fromP.aviumusing bioinformatics methods and characterized.Compared with most of theSAURfamily numbers reported in other plants,more members were specific in sweet cherry,suggesting that theSAURfamily in sweet cherry probably underwent an abundant expansion during its evolutionary process.In terms of gene structure,most of PavSAURs are intron-less,which is also seen in other species such as watermelon,poplar and mulberry(Huanget al.2016;Zhanget al.2017;Huet al.2018).Previous studies have found that introns have potential biological functions,such as mediating gene expression and alternative splicing recombination (Roy and Gilbert 2006;Roseet al.2016).Alternative splicing is generally considered to be one of the foundations for allowing genes to carry more genetic information.In addition,it also makes genes more plastic (Mastrangeloet al.2012).In the sweet cherry genome,only 14PavSAURmembers have introns,which are similar to the genes that contain introns in maize,and these genes may have stronger expression capabilities (Chenet al.2014).In addition,the existence of introns may play an irreplaceable role in genetic evolution,although this requires further research(Rogozinet al.2005).

The auxin signal appeared relatively late in evolution compared with cytokinin and ethylene,and it initially occurred in streptophyte algae (Chenget al.2019).However,as a plant-specific gene family,noSAURhomologs have been found in streptophyte algae,although 18SAURmembers were identified in the bryophytePhyscomitrellapatenswhich is more advanced than the algae (Rensinget al.2008;Chenget al.2019).In the early genomes,e.g.,MesostigmavirideandChlorokybusatmophyticus,some gene families such as ARF and PIN,which are involved in the auxin response,have evolved to allow for adaptation to the rapidly changing environment (Flores-Sandovalet al.2015;Katoet al.2018).These gene families have been found in many plants,from streptophyte algae to higher land plants(Mutteet al.2018).With the advancement of evolution,AUX/IAA-ARF-mediated auxin signals have been found in bryophytes (Baez and Nemhauser 2021).However,SAURappeared late in evolution,which may be ascribed to the evolution of new genes for adaptation and survival(Lauet al.2009;Wang Set al.2020).In higher plants,the number ofSAURmembers varies from 60 to 140,and they are scattered as clusters on the chromosomes,indicating that they have undergone significant fragment duplication and tandem duplication events during species evolution (Jainet al.2006;Chenet al.2014;Liet al.2017;Zhanget al.2017,2021;Wang Pet al.2020).In addition,phylogenetic tree Group A contains only proteins fromArabidopsisand the cherrySAURmembers,but it does not include any rice genes (Fig.2),partially suggesting that this subfamily plays an important role in the evolution of dicotyledons,which is also similar to the evolutionary relationship in maize (Chenet al.2014).

In the evolution from lower plants to higher plants,the genome has become more complex.During the long evolutionary process,a large number of replication events and genome rearrangement events have led to the current gorgeous plant kingdom (Soltis and Soltis 2016;Landiset al.2018).Gene duplication is a primary pattern for expanding the gene members of a species.To amass further information over the evolution ofPavSAURs,we investigated their genome duplication events.Overall,PavSAURgene expansion in sweet cherry was mainly due to the segmental duplication,while tandem duplication played only a minor role.TheSAURduplicated gene pairs have experienced purifying selection and consequently played a key in dominating the genomic variability in natural populations (Fig.4;Appendix E).Compared with the gene duplication events in otherPrunusspecies,the gene duplication events in sweet cherry rank second,just below almond (Wang Jet al.2020),indicating that theSAURfamily in sweet cherry was subjected to a massive expansion during its evolutionary history.Similar processes of purifying selection and gene expansion were also documented in cotton (Liet al.2017),maize (Chenet al.2014),and poplar (Huet al.2018).Gene expansion is one of the sources for the function diversification of plants,and also triggers the adaptation to rapid environmental changes (Lespinetet al.2002).Based on the Ka/Ks ratio,the extensive duplication events involving PavSAURs were predicted to have occurred within the last 0.24–49.8 million years.In apples,MdSAURs occurred within the last 3.98–20.52 million years (Wang Pet al.2020).Thus,the apple genome duplication occurred after its divergence from sweet cherry,and may end earlier than in cherry.By comparing the collinearity relationships between sweet cherry and other species,SAURdemonstrates more homologous gene pairs among the dicotyledonous plants.Some genes,such as Pav_sc0000129.1_g1670.1.mk(PavSAUR40),have collinearity gene pairs in rice,maize and other dicotyledons (Fig 5;Appendix E).These genes may have existed before the separation of monocotyledons and dicotyledons,and played an important role in plant evolution,partially reflecting the notion thatSAURis considerably conserved in land plants (Mutteet al.2018;Stortenbeker and Bemer 2019;Zhanget al.2021).

4.2.Diversity of SAUR functions and involvement in organ abscission

As one of the early auxin-responsive gene families,SAURplays a crucial role in the auxin signal transduction pathway.Thecis-element analysis in the 2 000 bp upstream promoter regions of PavSAURs revealed that most members include at least one type of auxinresponsive element (Appendix F).TheSAUR15Apromoter in soybean contains two connected sequences,TGTCTC and GGTCCCAT,which are considered to be auxin response elements (Liet al.1994).In plants,theSAURgenes are a diverse family with responsiveness to auxin,and Clade I comprises six genes inA.thaliana,all of which were up-regulated by auxin under at least one condition (Paponovet al.2008).After IAA treatment in cotton,11 of the 16 GhSAURs were up-regulated and three were down-regulated (Liet al.2017).In maize,a comparison of the distribution of regulatory elements in the promoter sequence revealed differences in the promoter regions of the replicated gene pairs,indicating that the replicated genes may not have common regulatory characteristics (Chenet al.2014).In addition,ARF7 could directly bind to the promoter region ofSAURand asymmetrically activated the expression ofSAUR19in the tropic response of plants (Wang Xet al.2020).TheSAUR63promoter contains a motif that binds to TCP,and theSAUR63subfamily gene was directly activated through the conservative target site in its promoter to regulate GA-dependent stamen filament elongation(Gastaldiet al.2020).The evidence presented in these cases indicate that theSAURgene has extensive interactions with other transcription factors to regulate the growth and development of plants.

The abscission of plant organs is a necessary event for the occurrence of some life activities.For example,the shedding of mature fruits or seeds away from the parent plant is the beginning of reproduction.In this study,the four PavSAURs (PavSAUR13/16/55/61) were differentially expressed in the carpopodium and small fruits of the abscission,which were also present in citrus and litchi.After the treatment of girdling plus defoliation in litchi,the transcript levels ofLcSAUR1mRNAs were increased in both the abscission zone and fruitlet,and the expression level in AZ (the abscission zone) was much higher than that in the fruitlet (Kuanget al.2012).In citrus,23 CitSAURs responded to the treatment of IAA,and the expression levels of 14 members changed significantly in the AZ of the fruit (Xieet al.2015).Auxin plays an important role in the process of regulating shedding,and it is one of the necessary conditions for preventing plant organs from abscising.Many lines of evidence have shown that the reduction of auxin in the abscission zones of plant organs make the separation zone more sensitive to ethylene and accelerate the shedding process (Robertset al.2002;Jinet al.2015).The level of auxin is higher before the fruit matures,and it gradually decreases as the fruit matures,and may cause the fruit to fall off at a low critical point (Maet al.2021).The change in the abscission zone also causes the transport of additional auxin to be inhibited,which reduces the polar transport of auxin and leads to abscission (Patharkar and Walker 2018;Maet al.2021).Studies have shown thatSAURhas a positive effect on the transport of auxin.Upregulating the expression ofSl-SAUR69in tomato leads to premature fruit ripening,while the overexpression ofSl-SAUR69reduces the polar transport of auxin,indicating the positive role ofSAURin the regulation of auxin transport (Shinet al.2019).Blocking or weakening the polar transport of auxin by alteringSAURexpression is a potential node for regulating fruit drop,which may provide a new opportunity for fruit tree genetic breeding.

Auxin induction will lead to an increase in H+,and the resulting decrease in pH is known to promote the growth of plants (Rayle and Cleland 1980).In growing organs,SAUR inhibits PP2C.D phosphatase activity and prevents the dephosphorylation of H+-ATPase at the plasma membrane,thereby promoting cell expansion (Wonget al.2021).There is a conserved motif responsible for catalytic function in the close C-terminal region of the PP2C.D protein,and SAUR can bind to this motif (Wonget al.2019).Under the action of auxin,SAUR interacts with protein phosphatase to prevent the dephosphorylation of H+-ATPases,thereby increasing its activity,and ultimately inducing the acidification of cell membranes (Spartzet al.2014).Abscission is accompanied by a decrease in auxin,which reduces the acidity of the cells in the isolated zone;and the relationship between alkalization and shedding has been confirmed inA.thalianaand tomato(Patharkar and Walker 2018).In the fruit abscission of sweet cherry,the level of IAA was lower than that of the retention fruit,suggesting that the reduction of auxin is a vital factor in fruit falling (Guoet al.2020).In cherry fruit,early abscission begins with the abscission of the pedicel–peduncle boundary that functions during immature fruit abscission (Tranbarger and Tadeo 2020).The level of auxin in the abscising fruit stalk of sweet cherry was lower than that in the normal fruit carpopodium (Qiuet al.2021).This also indicates that the transport of auxin in abscising fruit may be inhibited.Therefore,SAURplays important roles in cell elongation and auxin transport in sweet cherry,and is also involved in the regulation of fruit abscission in this species.

5.Conclusion

In the current study,a total of 86SAURmembers were identified from the sweet cherry genome v1.0,and these genes were divided into seven subfamilies (A–G).Most of the genes lack introns.The genes displayed diverse and analogous promotercis-acting elements and gene duplication events,and most members have at least one auxin response element.The RNA-seq data and qRTPCR results indicate that some PavSAURs play a key role in regulating fruit development and fruitlet abscission,and four members (PavSAUR13/16/55/61) demonstrate significant regulatory roles in fruit dropping.Heterologous overexpression ofPavSAUR55inA.thalianapromoted reproductive growth and affectedPP2C.Dgene expression.PavSAUR55improved root elongation in overexpression lines,which showed somewhat higher tolerances to NaCl and mannitol treatments.In particular,the overexpression lines were insensitive to NAA and increased the number of lateral roots under ABA treatment.In addition,PavSAUR55overexpression can delay the abscission of petals.Therefore,these results broaden our knowledge of the role of theSAURgene family in regulating fruit development and abscission in sweet cherry,provide new insights into the role of the biological function ofSAURgenes in plants,and facilitate future efforts to prevent fruit from abscising.

Acknowledgements

This project was supported by grants from the National Natural Science Foundation of China (32272649) and the Core Program of Guizhou Education Department,China(KY 2021-038).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendicesassociated with this paper are available on https://doi.org/10.1016/j.jia.2023.04.031