Functional analysis of MdSUT2.1,a plasma membrane sucrose transporter from apple

ZHANG Bo ,FAN Wen-min ,ZHU Zhen-zhen ,WANG Ying ,ZHAO Zheng-yang#

1 State Key Laboratory of Crop Stress Biology for Arid Areas,College of Horticulture,Northwest A&F University,Yangling 712100,P.R.China

2 Shaanxi Research Center of Apple Engineering and Technology,Yangling 712100,P.R.China

Abstract Sugar content is a determinant of apple (Malus×domestica Borkh.) sweetness. However,the molecular mechanism underlying sucrose accumulation in apple fruit remains elusive. Herein,this study reported the role of the sucrose transporter MdSUT2.1 in the regulation of sucrose accumulation in apples. The MdSUT2.1 gene encoded a protein with 612 amino acid residues that could be localized at the plasma membrane when expressed in tobacco leaf protoplasts.MdSUT2.1 was highly expressed in fruit and was positively correlated with sucrose accumulation during apple fruit development. Moreover,complementary growth assays in a yeast mutant validated the sucrose transport activity of MdSUT2.1. MdSUT2.1 overexpression in apples and tomatoes resulted in significant increases in sucrose,fructose,and glucose contents compared to the wild type (WT). Further analysis revealed that the expression levels of sugar metabolism-and transport-related genes SUSYs,NINVs,FRKs,HXKs,and TSTs increased in apples and tomatoes with MdSUT2.1 overexpression compared to WT. Finally,unlike the tonoplast sugar transporters MdTST1 and MdTST2,the promoter of MdSUT2.1 was not induced by exogenous sugars. These findings provide valuable insights into the molecular mechanism underlying sugar accumulation in apples.

Keywords: apple,MdSUT2.1,sugar,transport,plasma membrane

1.Introduction

Sucrose,the main product of plant photosynthesis,is transported from source leaves to sink organsviathe phloem to sustain their growth and development (Alukoet al.2021;Yoonet al.2021). Sucrose distribution among competing sinks,which influences whole-plant productivity and crop yield (Alukoet al.2021),is largely governed by sink-located transport and transfer processes (Durandet al.2018;Mathanet al.2021).

Sucrose transporters (typically named SUC or SUT proteins) are implicated in long-distance sucrose transport;they load sucrose into the leaf phloem and take up sucrose into sink tissues,such as fruits and flowers (Sauer 2007;Milneet al.2018). In plants,the first clonedSUTgene,SoSUT1,was isolated from spinach (SpinaciaoleraceaL.) (Riesmeieret al.1992).Subsequently,multipleSUTgenes were identified from other plant species (Reinderset al.2012).Arabidopsis thalianahas a family of nineSUTgenes: one putative pseudogeneAtSUC7and eight functionally active sucrose transporter protein genes (Sauer 2007;Rottmannet al.2018). Evolutionary analysis has established that the SUT family can be divided into three subfamilies: SUTI,SUTII,and SUTIII (Kühn and Grof 2010;Lalonde and Frommer 2012). Interestingly,type I SUTs are found exclusively in true dicots. The SUTII subfamily can further be divided into two evolutionary branches,namely SUTIIA and SUTIIB,with the latter being monocot-specific. It is worth pointing out that all land plant species contain type IIA and type III SUT(Reinderset al.2012;Penget al.2014).

Different members of each SUT evolutionary branch perform distinct functions (Milneet al.2018). The function of plasma membrane-localized type I SUTs is critical in apoplasmic phloem-loading species. AtSUT1 (AtSUC2)is responsible for loading sucrose into the phloem,and mutations inAtSUT1stimulate source leaves of plants to excessively store starch and sucrose,eventually leading to delayed development,stunted growth,and sterility(Gottwaldet al.2000). Similar to Arabidopsis AtSUT1,functional loss of tobacco (Nicotianatabacum) NtSUT1(Burkleet al.1998),maize (Zeamays) ZmSUT1 (Slewinskiet al.2010),and potato (Solanumtuberosum) StSUT1(Riesmeieret al.1994) can lead to severe stunting of plant growth and excessive accumulation of soluble sugars in leaves. Type II AtSUT2 (AtSUC3) can transport sucrose in yeast and is highly expressed in a number of non-photosynthetic cells and tissues,such as trichomes,guard cells,germinating pollen,stipules,seed coat,and root tips,suggesting that AtSUT2 may play a role in sucrose input sink tissues (Meyeret al.2000;Meyeret al.2004). Likewise,PlantagomajorPmSUC3 mediates the transport of sucrose and was identified in pollen tubes and root tips (Barthet al.2003). Tomato (Lycopersicon esculentum) LeSUT2 directly participates in sugar unloading in fruit,thereby impacting fruit yield in tomato plants (Hackelet al.2006). In contrast to type I and type II SUT proteins,type III SUT is primarily localized to the vacuolar membrane. In addition,vacuoles release sucrose via SUT4 transporters localized to the vacuolar membrane (Eomet al.2011;Payyavulaet al.2011;Schneideret al.2012).

A total of sixMdSUTgenes (MdSUT1.1/1.2/2.1/2.2/4.1/4.2) have been identified in apples (Penget al.2020a).Furthermore,the functions of type II MdSUT2.2 and type III MdSUT4.1 in apples have been elucidated. Unlike type II SUT2,which is localized to the plasma membrane in all species identified so far,apple MdSUT2.2 is localized to the vacuolar membrane (Maet al.2019).Expression ofMdSUT2.2in a sucrose uptake-deficient mutant yeast strain restored yeast uptake of sucrose.However,overexpression ofMdSUT2.2inArabidopsisand apple calli did not lead to elevated sucrose content under normal growth conditions. Only under NaCl,ABA,and mannitol stress conditions did the transgenic plants show increased sucrose content to resistance stress compared to wild-type apple calli (Maet al.2016). Overexpression ofMdSUT4.1in apple calli and strawberries had a net negative impact on sugar accumulation,leading to the re-release of sucrose from the vacuole (Penget al.2020a).

Apple is a popular fruit crop worldwide. Given that sucrose is known to affect the flavor and sweetness of apples (Maet al.2015),it is critical to understand the mechanism underlying sucrose accumulation in apples.The function of type II SUT proteins has been explored inArabidopsis(Meyeret al.2004),Plantago(Barthet al.2003),tomatoes (Hackelet al.2006),and other species(Dasguptaet al.2014;Yanet al.2021). Nonetheless,the function of type II MdSUT2.1 in apples remains elusive. In this study,a plasma membrane-localized type II sucrose transporterMdSUT2.1gene was cloned from apples,and its function was analyzed using yeast,tomatoes,and apples. The results revealed that MdSUT2.1 could transport sucrose and increase the soluble sugar content in apples and tomatoes. This study laid the foundation for a deeper understanding of the sugar accumulation mechanisms in fruit and provided potential strategies to increase sucrose levels in fruit,thereby improving fruit quality and yield.

2.Materials and methods

2.1.Plant materials and sugar content analysis

This study used six-year-old apple trees (Malus×domestica‘Starkrimson Delicious’) grown and maintained at Baishui Apple Experimental Farm of Northwest A&F University,Baishui,Shaanxi Province,China. The fruit were harvested 30,60,90,120,and 140 days after full bloom (DAFB). For each sample,18 fruit were harvested;6 fruit were considered as one biological repeat. Apple tissues were sampled from the following: young leaves just unfolding on new shoots;old leaves from the middle part of growing branches in the lower part of the tree;flowers at anthesis;unopened flower buds;stems (near the apices of newly growing shoots,3-4 mm in diameter);and mature fruit sampled 140 DAFB. Soluble sugars were quantified using high-performance liquid chromatography(HPLC) (Waters,USA) according to a previously defined protocol (Liuet al.2013).

2.2.Gene structure and phylogenetic analysis of apple MdSUT2.1 proteins

Transmembrane domains were predicted using the TMHMM tool (https://services.healthtech.dtu.dk/service.php?TMHMM-2.0). The phylogenetic tree was generated using the MEGA 7.0 Software and the neighbor-joining method.

2.3.RNA extraction and RT-qPCR analysis

Total RNA was isolated from the samples with TRIzol RNA Plant Plus Reagent (Tiangen,Beijing,China),and cDNA was synthesized using HiScript?II Q RT SuperMix for qPCR (Vazyme,Nanjing,China). Finally,RT-qPCR was performed on an ABI StepOnePlus?Real-Time PCR Detection System (Applied Biosystems,Waltham,MA,USA) usingMdActinas the internal control. The primers used in the experiment are listed in Appendix A.

2.4.Subcellular localization

The coding sequence (CDS) ofMdSUT2.1was cloned from cDNA extracted from the ‘Starkrimson Delicious’apple. The35S:MdSUT2.1-eGFPconstruct was acquired by inserting theMdSUT2.1CDS without a stop codon into the pCAMBIA2301-eGFPvector under the control of theCauliflowermosaicvirus35S(CaMV35S)promoter. Protoplasts were isolated from four-weekold tobacco (N.benthamiana) leaves and infiltrated withA.tumefaciensaccording to a previously reported method(Yooet al.2007). Excitation/emission wavelengths were 488 nm/498-540 nm for the enhanced green fluorescent protein (eGFP).

2.5.Complementation assays for MdSUT2.1 in yeast

For complementation assays in yeast (Saccharomyces cerevisiae) cells,the CDSs ofMdSUT2.1andAtSUT1(AT1G22710) were cloned into the vector pDR196.Next,the pDR196,pDR196-MdSUT2.1,and pDR196-AtSUT1constructs were transformed into the sucrose uptake-deficient yeast strain SUSY7/ura3. Thereafter,transformants of the SUSY7/ura3 yeast strain were grown in a liquid SD (synthetic deficient)/-uracil medium supplemented with 2% glucose. Afterward,serial dilutions of the SUSY7/ura3 yeast cell suspension were added dropwise onto a solid SD/uracil medium containing 2%sucrose. Finally,the yeast cells were cultured in the dark at 30°C for 3-4 days.

2.6.Artificial microRNA vector construction

In order to silence theMdSUT2.1gene in apple calli and fruit,an artificial microRNA (amiRNA) vector was constructed according to a previously described method(Ossowskiet al.2008). The mature sequences of two amiRNAs (amiRNA1: 5′-TATTTAGTTGGGGATCATCGA-3′and amiRNA2: 5′-TGCTGATACATAGTGAGTGGA-3′)targetingMdSUT2.1were designed on the WMD3 website (http://wmd3.weigelworld.org/cgi-bin/webapp.cgi?page=Home;project=stdwmd). Next,amiRNA1,amiRNA1＊,amiRNA2,and amiRNA2＊ were engineered into the Arabidopsis precursor MIR319a to replace ath-miR319a and ath-miR319a＊ and generate MIR319a-amiRNA1 and MIR319a-amiRNA2,respectively. Then,the MIR319aamiRNA1 and MIR319a-amiRNA2 sequences were fused with35Spromoter,NOS terminator,and E9 terminator to form35S:MIR319a-amiRNA1:NOS and35S:MIR319aamiRNA2:E9,respectively. Finally,35S:MIR319aamiRNA1:NOS and35S:MIR319a-amiRNA2:E9 were inserted together into the vector pCAMBIA2300 to generate35S:amiR-MdSUT2.1.

2.7.Generation of transgenic plant materials

‘Orin’ apple calli were transformed according to a previously reported method (Zhanget al.2021). First,the calli were infected with theA.tumefaciensstrain carrying35S:MdSUT2.1and35S:amiR-MdSUT2.1and then transferred to a co-cultivation medium. After a 3-day co-cultivation in darkness,the calli were transferred to a screening medium to screen the transformants.To overexpress and silence theMdSUT2.1gene in apple fruit,the35S:MdSUT2.1andMdSUT2.1-amiRNA recombinants were each transformed into theAgrobacteriumstrain GV3101. Transient fruit infiltration was performed according to a previously described method (Liet al.2016,2020). Briefly,250 μL of infiltration buffer was injected into the fruit on the tree using a syringe at 120 DAFB. TheAgrobacteriumharboring the empty vector pCAMBIA2301 was used as a control.The fruit were harvested 7 days later to determine sugar content. Ten fruit were considered as one biological replicate,and three biological replicates were prepared.Solanumlycopersicum‘Micro-Tom’ was transformedvia Agrobacterium-mediated techniques by following the procedures described by Sunet al.(2006).

2.8.Exogenous sugar incubation test

The 2 000-bpMdSUT2.1promoter was inserted in the pCAMBIA0390-GUSvector to generateMdSUT2.1pro:GUSand then transiently transformed into four-week-old tobacco leaves (Sparkeset al.2006). Next,tobacco leaves were cut off after 3 days ofAgrobacteriuminfiltration and placed in 2% sugar solutions (glucose,fructose,and sucrose) or water for 24 hours before measuring GUS protein activity (Zhuet al.2021).

Exogenous sugar incubation test of fruit was conducted using the method outlined by Berüter and Studer Feusi(1995) and Archbold (1999) with marginal modifications.The standard MES buffer consisted of the following: 50 mmol L-1MES (pH 5.5),5 mmol L-1CaCl2,5 mmol L-1MgCl2,1 mmol L-1EDTA,and 5 mmol L-1ascorbic acid. Fruits harvested at 100 DAFB were divided into discs of 1.0 cm diameter and 0.1 cm thickness. The discs were equilibrated in an equilibration buffer (standard MES buffer+200 mmol L-1mannitol) for 30 min and then used for the exogenous sugar incubation test. We prepared 30 mL of buffer solutions containing different sugars (standard MES buffer+200 mmol L-1sorbitol/fructose/glucose),and the osmotic potential of each buffer solution was adjusted to equal that of mannitol. The equilibrium buffer was used as a control. We mixed 10 g of fruit discs with 30 mL of buffer solution in a 100-mL triangular flask and shook it on a shaker at 25°C for 16 h. The discs were finally washed with ddH2O and used for RNA extraction to determine the expression of relevant genes.

3.Results

3.1.Gene structure and phylogenetic analysis of apple MdSUT2.1

Previous studies have reported two types of IISUTgenes in apples:MdSUT2.1(MD13G1074600) andMdSUT2.2(MD16G1075900) (Penget al.2020a). Furthermore,the function of MdSUT2.2 has previously been described (Maet al.2017). Therefore,theMdSUT2.1gene,with an ORF of 1 839 bp long (a protein with a mass of approximately 65.7 kD),was cloned. The predicted membrane structural domains signaled that the protein has 11 transmembrane structural domains (Fig.1-A). Next,the amino acid sequences of MdSUT2.1 were compared with those of type II SUT proteins from different plants,including Arabidopsis,grapes,oranges,rice,and tomatoes. Then,a phylogenetic tree was constructed using the neighborjoining method. The results demonstrated that apple MdSUT2.1 had the highest homology with Arabidopsis AtSUT2 (Fig.1-B). Similar to AtSUT2,MdSUT2.1 had an extended structural domain at the N-terminal region (～33 amino acid residues long) and a central cytoplasmic loop(～59 amino acid residues long) compared to type I and type III SUT proteins (Fig.1-C).

Fig.1 The structural analysis of the MdSUT2.1 gene. A,transmembrane domains prediction of MdSUT2.1. B,phylogenetic relationships of SUT2 in different species,including Citrus sinensis (CsSUT2,NP_001275773),Vitis vinifera (VvSUC12,AAF08330),Arabidopsis thaliana (AtSUT2,AT2G02860),Plantago major (PmSUC3,CAD58887),Lycopersicon esculentum(LeSUT2,AAG12987),Solanum tuberosum (StSUT2,AAP43631),Oryza sativa (OsSUT4,BAC67164),Sorghum bicolor (SbSUT2,Sb04g038030) and Zea mays (ZmSUT2,AAS91375). C,comparison of the apple MdSUT2.1 sequence with sequences of the previously characterized Arabidopsis transporters AtSUC1-4. The black background indicates amino acids with 100% homology.

3.2.Subcellular localization of MdSUT2.1

TheeGFPsequence was fused to the C-terminal region ofMdSUT2.1to examine the location of the MdSUT2.1 protein. TheMdSUT2.1-eGFPfusion construct was transiently expressed in tobacco protoplasts;eGFP was located in the plasma membrane. Contrastingly,in control tobacco cells transiently transformed with35S:eGFP,eGFP was present throughout the cells,including the nucleus and cytoplasm (Fig.2). This finding indicated that MdSUT2.1 is a plasma membrane-localized protein.

Fig.2 Analysis of the subcellular localization of MdSUT2.1. Excitation/emission wavelengths of 488 nm/498-540 nm were utilized for the enhanced green fluorescent protein (eGFP).

3.3.Expression profiles of MdSUT2.1 gene in apple tissues

RT-qPCR was performed to assess the spatio-temporal expression levels ofMdSUT2.1using RNA extracted from different tissues of ‘Starkrimson Delicious’ apple plants.The results revealed thatMdSUT2.1was expressed in flower buds,flowers,young leaves,mature leaves,stems,and mature fruit;nevertheless,the expression was higher in flowers and fruit (Fig.3-A). The expression level ofMdSUT2.1and sucrose content in fruit at different developmental stages were subsequently analyzed(Fig.3-B and C) to determine the relationship betweenMdSUT2.1expression and sucrose content. Correlation analysis demonstrated a highly significant positive correlation betweenMdSUT2.1and sucrose content(r=0.97,P=0.006). These results imply thatMdSUT2.1may play a fundamental role in the modulation of sucrose accumulation in apple fruit.

Fig.3 Expression profiles of MdSUT2.1 gene in apple tissues. A,relative expression levels of MdSUT2.1 in different tissues from flower buds,flowers,young leaves,mature leaves,stems,and mature fruit. The expression level in young leaves was set to 1. B,expression pattern of MdSUT2.1 during fruit development in the ‘Starkrimson Delicious’ apple. C,the sucrose content in the fruit of the ‘Starkrimson Delicious’ apple at five time points. DAFB,days after full bloom. The bars represent the mean value±SD (n=3).

3.4.Transport activity of MdSUT2.1

The SUSY7/ura3 yeast strain inserted with the potatosucrosesynthase(SUSY) gene cannot grow in a medium with sucrose as the sole source of sugar owing to a lack of extracellular invertase and sucrose transporter(Riesmeieret al.1992;Grimes and Overvoorde 1996).In order to investigate whether MdSUT2.1 possesses sucrose transport activity,pDR196-MdSUT2.1was transformed into SUSY7/ura3 cells to generate SUSY7/ura3-MdSUT2.1cells. Negative control SUSY7/ura3-pDR196 cells (the SUSY7/ura3 cells transformed with empty vector pDR196),positive control SUSY7/ura3-AtSUT1cells (the SUSY7/ura3 cells transformed with pDR196-AtSUT1),and SUSY7/ura3-MdSUT2.1cells could grow on an SD/-Ura/Glucose medium. However,only the SUSY7/ura3-AtSUT1cells and SUSY7/ura3-MdSUT2.1cells were able to grow in an SD/-Ura/Sucrose medium,indicating MdSUT2.1 could transport sucrose in yeast. Moreover,SUSY7/ura3-MdSUT2.1cells grew slowly compared to SUSY7/ura3-AtSUT1cells in the SD/-Ura/Sucrose medium (Fig.4).

Fig.4 Heterologous expression of MdSUT2.1 in SUSY7/ura3 yeast. SUSY7/ura3 carrying different genes grown on SD/-Ura/Glucose (left) and SD/-Ura/Sucrose (right) medium. Empty vector (pDR196) and pDR196-AtSUT1 (AT1G22710) were used as negative and positive controls,respectively.

3.5.Overexpression of MdSUT2.1 increased sucrose,fructose,and glucose contents,whereas silencing the gene significantly inhibited sucrose,fructose,and glucose levels in apples

MdSUT2.1-overexpressing apple calli lines (OL1 and OL2) andMdSUT2.1-silenced lines (SL1 and SL2)were generated (Fig.5-A) to investigate the role of MdSUT2.1 in sugar accumulation. The RT-qPCR analysis determined thatMdSUT2.1was increased in OL1/2 and decreased in SL1/2 (Fig.5-B). The growth rate of apple calli was significantly increased in OL1/2 but suppressed in SL1/2 (Fig.5-A). Meanwhile,the soluble sugar content was examined in apple calli;the results uncovered that overexpression ofMdSUT2.1promoted the contents of sucrose,fructose,and glucose compared to WT (Fig.5-C),while suppression ofMdSUT2.1expression led to their decrease. We also analyzed the expression of specific genes related to sugar metabolism and transport,including threesucrosesynthases(MdSUSY1/3/4),threevacuolar acidinvertases(MdvAINV1/2/3),twoneutralinvertases(MdNINV7.1/7.2),twofructokinases(MdFRK1/2),twohexokinases(MdHXK1/2),and twotonoplastsugar transporters(MdTST1/2). The findings revealed that the expression levels ofMdSUSY4,MdNINV7.1/7.2,MdHXK2,MdFRK1/2,andMdTST1/2were significantly increased or decreased inMdSUT2.1-overexpressing lines orMdSUT2.1-silenced lines compared to WT,but the expression levels ofMdvAINV1/2/3were not significantly altered in the overexpressing and silenced lines (Fig.5-D).

Fig.5 Overexpression and silencing of MdSUT2.1 in apple calli. A,the growth of the transformed calli on medium. B,the expression levels of MdSUT2.1 were measured in the overexpressing lines (OL1 and OL2) and silencing lines (SL1 and SL2). C,sucrose,fructose,and glucose contents in transformed apple calli. D,the expression levels of genes related to sugar metabolism and transport in transformed apple calli. WT,wild type. The bars represent the mean value±SD (n=3). ＊＊,P＜0.01 (Student’s t-test).

AnAgrobacteriumtumefaciensinfiltration-based transient expression assay was used to overexpress and inhibitMdSUT2.1in the ‘Starkrimson Delicious’ apple. The apple fruit were harvested 8 days following the transient transformation. The expression level ofMdSUT2.1was significantly higher inMdSUT2.1-overexpressing fruit but markedly lower inMdSUT2.1-silenced fruit. The contents of sucrose,fructose,and glucose increased significantly by 20,14,and 13% in fruit overexpressingMdSUT2.1,respectively,compared to fruit infiltrated with the empty vector,whereas the contents decreased by 10,11,and 9%in silenced fruit compared to fruit infiltrated with the empty vector (Appendix B). These results signified that alteringMdSUT2.1expression in apples can directly affect sucrose,fructose,and glucose levels.

3.6.Overexpression of MdSUT2.1 increased sugar content in tomato fruit

TwoMdSUT2.1-overexpressing tomato lines (MdSUT2.1-L1/L3) were generated (Fig.6-A) to further investigate the impact of increasedMdSUT2.1expression on sugar accumulation in plants. RT-PCR analysis revealed thatMdSUT2.1was highly expressed in ripening transgenic fruit (Fig.6-B). Furthermore,the plant phenotype and the soluble sugar content in the fruit ofMdSUT2.1-L1/L3 were analyzed. There was no significant fluctuation in the growth ofMdSUT2.1-L1/L3,but the size of the fruit increased (Fig.6-A). The carbohydrate (sucrose,fructose,and glucose) levels and soluble solids content(SSC) in theMdSUT2.1-overexpressing tomato fruit were significantly increased compared to WT (Fig.6-C).Furthermore,the expression of genes related to sugar metabolism and transport was examined by RT-qPCR.The results revealed that the expression levels ofSlSUSY1/3/4,SlNINV4/5,SlFRK1/3,SlHXK1/2,andSlTST1/2were increased compared to WT (Fig.6-D).These findings suggested that overexpression ofMdSUT2.1in tomatoes increased their sugar content and affected the expression of the genes related to sugar metabolism and transport.

Fig.6 Overexpression of MdSUT2.1 in tomato. A,the phenotype of transgenic tomato fruit. B,detection of the expression levels of MdSUT2.1 in two MdSUT2.1-overexpressing lines (MdSUT2.1-L1/L3) by RT-PCR. C,the carbohydrate (sucrose,fructose,and glucose) levels and soluble solids content (SSC) in MdSUT2.1-L1/L3. D,the expression levels of genes related to sugar metabolism and transport in MdSUT2.1-L1/L3. The bars represent the mean value±SD (n=3). ＊＊,P＜0.01 (Student’s t-test).

3.7.MdSUT2.1 expression was not induced by exogenous sugars

Previous studies have evinced that tonoplast sugar transporters (TSTs) could be substantially up-regulated after exogenous sugar treatment in apples and Arabidopsis (Wormitet al.2006;Liet al.2022). In order to determine whether exogenous sugars induceMdSUT2.1,isolated tissue discs of apple fruit were incubated in different sugar solutions,followed by RTqPCR measurement;the result showed that the mRNA level ofMdSUT2.1was not induced by sucrose,fructose,and glucose (Fig.7-A). Additionally,theMdSUT2.1promoter was cloned in front of theGUSsequence. TheMdSUT2.1pro:GUSconstruct was transiently transformed into tobacco leaves,and GUS activity was measured after exogenous sugar feeding. The result showed thatMdSUT2.1promoter activity could not be induced by glucose,fructose,and sucrose (Fig.7-B).

Fig.7 MdSUT2.1 expression was not induced by exogenous sugars. A,isolated tissue discs of apple fruit were incubated in different sugar solutions. B,the influence of sugar feeding on MdSUT2.1 promoter activities. Tobacco leaves transiently transformed with MdSUT2.1pro:GUS were fed with 2% different exogenous sugar for 24 h before GUS activity was measured. The bars represent the mean value±SD (n=3).

4.Discussion

The Arabidopsis type II sucrose transporter protein AtSUT2 is structurally distinct from the other type I and type III SUT proteins identified. It possesses an extended structural domain at the N-terminal region and a central cytoplasmic loop (Lalondeet al.1999;Barkeret al.2000). This study affirmed that MdSUT2.1 shares a high degree of structural similarity with AtSUT2;they share the same extended structures (Fig.1),indicating that their functions may be identical. Previous studies have suggested that type II AtSUT2,PmSUC3,and SbSUT2 were able to transport sucrose in yeast (Meyeret al.2000;Barthet al.2003;Milneet al.2013). In the current study,MdSUT2.1was transferred into sucrose uptake-deficient yeast mutant SUSY7/ura3 and partially restored sucrose uptake in yeast (Fig.4),implying that MdSUT2.1 mediates sucrose transport. However,the growth of SUSY7/ura3-MdSUT2.1yeast was slower compared to SUSY7/ura3-AtSUT1yeast,which could be attributed to the lower affinity of MdSUT2.1 for sucrose. AtSUT2 had a lower sucrose affinity than AtSUT1 in Arabidopsis (Meyeret al.2000).

Subcellular localization is pivotal for the function of transport proteins (Schulzet al.2011). The localization of MdSUT2.1 in tobacco protoplasts signified that it functioned as a transporter in the plasma membrane.Alanine-scanning mutagenesis revealed that an N-terminal LXXXLL motif in ESL1 (EDR6-like 1) was essential for its localization at the vacuole membrane (Yamadaet al.2010). Except for AtSUT4,which contained the sequence KRVLL,the N-terminal region of all angiosperm type III SUTs localized to the vacuolar membrane contained a perfect LXXLL motif,suggesting that it could be a vacuolar membrane-targeting motif. However,this motif was not present in type I and II SUTs located at the plasma membrane (Reinderset al.2012). Herein,the N-terminal region of the sequence of MdSUT2.1 was also devoid of the LXXXLL motif,the KRVLL sequence,and the LXXLL motif. This result provided additional evidence that MdSUT2.1 is localized to the plasma membrane.

Sucrose transporter proteins play a key role in plants,but the members of the SUT family have contrasting functional roles,including sucrose loading into the phloem and sucrose uptake into sink cells (Braun and Slewinski 2009;Reinderset al.2012). Several sucrose transporter proteins are abundantly expressed in various sink organs,such as barley (Hordeumvulgare)HvSUT1,fava bean (Viciafaba)VfSUT1,and PlantagoPmSUC1in developing seeds (Gahrtzet al.1996;Weberet al.1997;Weschkeet al.2000);OsSUT1-5in various sink organs and tissues of rice,including developing grains (Aokiet al.2003;Sunet al.2011);LotusjaponicusLjSUT4in green seedpods and roots (Flemetakiset al.2003);AtSUC9in floral tissues (Sivitzet al.2007);tomatoLeSUT2in fruit(Hackelet al.2006);andStSUT1andStSUT4in sink tubers (Kühnet al.1997;Chincinskaet al.2008). In the present study,MdSUT2.1was highly expressed in the fruit (Fig.3),supporting its role in sugar accumulation in fruit. Mutant and transgenic studies have demonstrated that these SUTs are essential for unloading sucrose from the phloem to the sink. For example,knockout mutants ofAtSUC9exhibited an early flowering phenotype (Sivitzet al.2007). In another instance,inhibition ofLeSUT2gene expression significantly reduced seed and fruit development and pollen germination (Hackelet al.2006).TheStSUT4-RNAi plant phenotype includes high tuber yield,early flowering,and reduced sensitivity to light at far-red-rich wavelengths (Chincinskaet al.2008). In the present study,overexpression or silencing ofMdSUT2.1significantly increased or decreased sucrose content in apple fruit. These results implied that MdSUT2.1 promoted sucrose uptake in sink cells. The increased fruit size and sucrose content in tomato fruit overexpressingMdSUT2.1corroborated this conclusion (Fig.6).

Plants have developed an intricate system for sugar metabolism and accumulation in sink cells. Once translocated into the cells,sucrose is converted to glucose and fructose by NINV or to fructose and UDPglucose (UDPG) by sucrose synthase. Fructose and glucose can be phosphorylated to fructose-6-phosphate(F6P) and glucose-6-phosphate (G6P) by fructokinase and hexokinase,respectively. F6P can participate in the glycolytic and TCA cycles,generating energy and intermediates for other processes. Most of the unmetabolized soluble sugars,such as sucrose,glucose,and fructose,are transported to the vacuole,which is the main organelle for sugar accumulation in fruit cells,viaspecial transporter proteins located in the vacuolar membrane (Nguyen-Quoc and Foyer 2001;Liet al.2012;Hedrichet al.2015;Hussainet al.2020;Chenet al.2021). Tonoplast sugar transporters (TSTs) can transport sucrose,glucose,and fructose in several species,including sugar beets,tomatoes,apples,melons,and peaches (Junget al.2015;Chenget al.2018;Penget al.2020b;Zhuet al.2021;Liet al.2022). In this study,the expression levels of the genes involved in sugar transport and metabolism (TSTs,SUSYs,NINVs,FRKs,andHXKs)were elevated inMdSUT2.1-overexpressed apples and tomatoes (Figs.5 and 6). Therefore,we posited that sucrose was transported from the extracellular space to the cytoplasm by MdSUT2.1 at the plasma membrane and was subsequently broken down into glucose,fructose,and UDPG by SUSY and NINV,resulting in an increase in glucose and fructose levels in the cytoplasm.Earlier studies have highlighted that elevated levels of sucrose,glucose,and fructose in the cytoplasm promoted the expression of TSTs,which transport sugar (sucrose,glucose,and fructose) into the vacuole (Zhuet al.2021).Thus,it is understandable that sucrose,glucose,and fructose levels were elevated in both apples and tomatoes withMdSUT2.1overexpression.

5.Conclusion

This study systematically explored the function of apple MdSUT2.1,a type II SUT family member. Subcellular localization revealed that MdSUT2.1 was localized to the plasma membrane. Moreover,heterologous expression ofMdSUT2.1allowed SUSY7/ura3 yeast cells to grow on a sucrose medium. Overexpression ofMdSUT2.1increased sucrose,fructose,and glucose contents in apples,while silencing of the gene suppressed their levels. Finally,overexpression ofMdSUT2.1in tomatoes increased sugar content and impacted the expression of genes related to sugar metabolism and transport in the fruit. The findings add to a growing body of evidence supporting the idea that MdSUT2.1 plays a role in regulating sugar accumulation in apples.

Acknowledgements

This work was supported by the earmarked fund for China Agriculture Research System (CARS-27).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendicesassociated with this paper are available on http://www.ChinaAgriSci.com/V2/En/appendix.htm