Physiological implications of SWEETs in plants and their potential applications in improving source–sink relationships for enhanced yield

Summary

The sugars will eventually be exported transporters (SWEET) family of transporters in plants is identified as a novel class of sugar carriers capable of transporting sugars, sugar alcohols and hormones. Functioning in intercellular sugar transport, SWEETs influence a wide range of physiologically important processes. SWEETs regulate the development of sink organs by providing nutritional support from source leaves, responses to abiotic stresses by maintaining intracellular sugar concentrations, and host–pathogen interactions through the modulation of apoplastic sugar levels. Many bacterial and fungal pathogens activate the expression of SWEET genes in species such as rice and Arabidopsis to gain access to the nutrients that support virulence. The genetic manipulation of SWEETs has led to the generation of bacterial blight (BB)‐resistant rice varieties. Similarly, while the overexpression of the SWEETs involved in sucrose export from leaves and pathogenesis led to growth retardation and yield penalties, plants overexpressing SWEETs show improved disease resistance. Such findings demonstrate the complex functions of SWEETs in growth and stress tolerance. Here, we review the importance of SWEETs in plant–pathogen and source–sink interactions and abiotic stress resistance. We highlight the possible applications of SWEETs in crop improvement programmes aimed at improving sink and source strengths important for enhancing the sustainability of yield. We discuss how the adverse effects of the overexpression of SWEETs on plant growth may be overcome.

Introduction

SWEETs are the most recently identified class of sugar transporters. SWEETs are present in all kingdoms of life, being conserved across prokaryotes and eukaryotes (Chen et al., 2010; Jia et al., 2017). The eukaryotic SWEET proteins are predicted to have seven transmembrane domains (Anjali et al., 2020). Phylogenetically, plant SWEETs are divided into four clades (clades I, II, III and IV) (Chen et al., 2010; Yuan and Wang, 2013). They function in the control of plant growth and defence, facilitating the long‐distance sugar transport from source (leaves) to sink (such as seeds and flowers) organs, with crucial roles in, among others, senescence, nectar secretion, hormone signalling, pathogenesis and abiotic stress responses (Figure 1) (Breia et al., 2021; Chardon et al., 2013; Chen et al., 2010, 2012; Jeena et al., 2019; Mathan et al., 2021a; Sosso et al., 2015; Wang et al., 2019a). Of the many functions of SWEETs in plant biology, their critical roles in partitioning photosynthates to sink organs such as seeds and in pathogenicity are perhaps the most important because they can directly influence crop yield.

Figure 1.

Different aspects of plant growth, development and physiology regulated by SWEETs.

The transport of photosynthates from source leaves to sink tissues is an essential feature of the whole‐plant source–sink balance that requires both symplastic and apoplastic phloem loading mechanisms. It is widely assumed that the pre‐phloem transport of photoassimilates from mesophyll cells to phloem parenchyma cells occurs predominantly through plasmodesmata, which create a symplastic route (Liesche and Schulz, 2012; Miras et al., 2022; Schulz, 2015). In species such as maize, potato and Arabidopsis that use apoplastic phloem loading mechanism, sucrose from phloem parenchyma cells is exported out of the cytoplasm into the apoplast/cell wall space by SWEETs (Abelenda et al., 2019; Bezrutczyk et al., 2018a; Chen et al., 2012). Sucrose is then actively taken up from the apoplast by sucrose/H⁺ symporters known as sucrose transporters (SUTs, also called SUCs), which load sucrose into the sieve element–companion cells complex for translocation to distant sink tissues (Gottwald et al., 2000; Riesmeier et al., 1994; Slewinski et al., 2009). SWEETs and SUTs also function in sugar unloading in sink organs such as seeds (Scofield et al., 2002; Wang et al., 2021; Yang et al., 2018).

Plant pathogens require a supply of sugars from the host cells to support their growth and proliferation. The SWEET proteins play a vital role in the virulence strategies of many bacterial and fungal pathogens in species such as Arabidopsis, rice and cotton (Chen et al., 2010; Chu et al., 2006; Cox et al., 2017). Bacterial and fungal pathogens activate the expression of specific SWEETs to facilitate sugar secretion from cells at the sites of infection and ensure propagation (Chen et al., 2010; Gupta et al., 2021). Many bacterial pathogens, e.g. Xanthomonas oryzae pv. oryzae (Xoo) and X. citri subsp. Malvacearum, attack hosts, such as rice and cotton, by secreting effector molecules known as transcription activator‐like effector (TALE) proteins via the type III secretion system (TIIISS) (Antony et al., 2010; Cox et al., 2017; Zhou et al., 2015). TALEs are transcription factors that activate the expression of their cognate SWEET genes by binding to the effector‐binding element (EBE) sites present in the promoters of the SWEET genes (Figure 2)(Antony et al., 2010; Boch et al., 2009; Chen et al., 2010; Moscou and Bogdanove, 2009).

Figure 2.

TAL effectors dependent activation of SWEETs by pathogenic bacteria. Pathogenic bacteria such as Xanthomonas oryzae pv. oryzae (Xoo), invading plant cells secrete transcription activator‐like (TAL) effectors that bind to the effector‐binding element (EBE) sites in the promoter regions of some SWEET genes to activate their expression. Increased expression of SWEETs results in enhanced efflux of sugars from cells to the apoplastic space that facilitates bacterial growth.

Current concepts suggest that genetic manipulation of SWEETs can result in improved partitioning of photosynthates, which in turn can lead to increased yield and enhanced resistance to pathogens. However, recent reports concerning the overexpression of clade III SWEETs to increase photosynthate translocation to sink tissues have shown that plant growth is adversely affected (Abelenda et al., 2019; Eom et al., 2015; Fatima and Senthil‐Kumar, 2021; Gao et al., 2018; Kim et al., 2020; Singh et al., 2021; Yuan et al., 2009). Surprisingly, the transgenic plants overexpressing the clade III SWEETs showed increased tolerance to pathogens, which are known to induce these SWEET genes (Fatima and Senthil‐Kumar, 2021; Kim et al., 2020; Singh et al., 2021; Yuan et al., 2009). Such findings suggest that sugar transport and pathogenesis are tightly coupled via SWEETs, at least in cases where the same SWEET genes regulate both processes. Manipulation of SWEETs, therefore, has implications for plant growth as well as pathogen resistance.

Here, we review the role of SWEETs in plant pathogenesis and carbohydrate translocation in different tissues and their potential for improving source–sink relationships to enhance yield. We consider the emerging roles of SWEETs in abiotic stress tolerance, evolution of C4 photosynthesis and highlight the new knowledge gained from overexpression studies concerning how clade III SWEETs can simultaneously modulate plant growth and susceptibility to disease.

SWEETs in pathogenic and symbiotic interactions of plants

The rice genome contains 22 SWEET genes, which show distinct tissue‐specific expression patterns (Wu et al., 2022; Yuan et al., 2014; Yuan and Wang, 2013). Of these, activation of six clade III SWEETs, which have sucrose transport activity, namely OsSWEET11/Os8N3/Xa13, OsSWEET11b, OsSWEET12, OsSWEET13/Xa25, OsSWEET14/Os11N3 and OsSWEET15, results in the development of BB disease symptoms triggered by Xoo. Hence, these SWEETs are called ‘susceptibility genes’ for Xoo (Streubel et al., 2013; Wu et al., 2022). Only OsSWEET11, OsSWEET13 and OsSWEET14 are targeted by at least one of the TALEs secreted by Xoo, leading to infection (Antony et al., 2010; Yang et al., 2006; Zhou et al., 2015). Activation of the expression of the OsSWEET11, OsSWEET13 and OsSWEET14 genes by Xoo enhances sucrose efflux from phloem parenchyma cells to the apoplast, a process that enhances Xoo growth and multiplication (Eom et al., 2019; Oliva et al., 2019). In addition, the rice OsSWEET11 interacts with the copper transporters COPT1 and COPT5 to remove copper from the xylem vessels, which are the sites of Xoo multiplication and disease progression (Yuan et al., 2010). Copper is an essential plant micronutrient used in many pesticides to inhibit Xoo growth. Thus, removing copper from xylem vessels via OsSWEET11 and copper transporters is another Xoo strategy to evade plant defences and enhance virulence (Yuan et al., 2010).

In addition to Xoo, Rhizoctonia solani infection induces the expression of OsSWEET11 and OsSWEET14 in rice leaves (Gao et al., 2018; Kim et al., 2020). The expression of a mutated inactive form of the protein (ossweet11) under the control of the RUBISCO promoter in rice leaves led to the formation of dysfunctional OsSWEET11 oligomers that lacked sugar transport activity. These transgenic rice plants showed improved resistance to R. solani with no yield penalty. The results suggest that R. solani induces OsSWEET11 in rice leaves to acquire sugars for pathogenesis (Gao et al., 2018). Oligomerization of the SWEET proteins is necessary to form a functional pore to transport sugars. It has previously been shown that a mutated version of the SWEET protein can form oligomers with functional SWEET units and abolish sugar transport activity (Xuan et al., 2013).

Like rice, pathogens modulate the expression of SWEETs in other plant species, either in a TALE‐dependent or independent manner, to gain access to host carbon resources (Chen et al., 2010; Gupta et al., 2021; Veillet et al., 2017). Activation of the expression of the GhSWEET10 gene by X. citri subsp. malvacearum and the MeSWEET10a gene by X. axonopodis pv. manihotis is associated with the progression of bacterial blight in cotton and with water‐soaking symptoms in cassava, respectively, suggesting that bacteria exploit them for growth and disease progression (Cohn et al., 2014; Cox et al., 2017). The induced expression of CsSWEET1 by X. citri subspecies citri (Xcc306), the causal agent for citrus bacterial canker, did not lead to the characteristic symptoms of pustule formation (Hu et al., 2014). Interestingly, CsSWEET1 is a homologue of OsSWEET11 and OsSWEET14 in rice, which are disease‐susceptibility genes (Hu et al., 2014). In addition to CsSWEET1, TALEs secreted from Xcc306 induced the expression of the CsLOB1 (lateral organ boundaries 1) transcription factor that causes the symptoms of citrus bacterial canker (Hu et al., 2014). CsLOB1 induces the expression of many downstream genes involved in cell expansion, which causes hypertrophy resulting in pustule formation (Hu et al., 2014). The bacterial spot disease of pepper is caused by X. campestris pv. Vesicatoria (Xcv). This pathogen induces the expression of many UPA (up‐regulated by AvrBs3 effector) genes, including UPA20 and UPA16, a SWEET family protein. Like CsSWEET1, UPA16 activation is not associated with hypertrophy in pepper (Kay et al., 2009). Activation of UPA20, which is a bHLH (basic helix–loop–helix) transcription factor, causes hypertrophy via the transactivation of the expansin‐encoding UPA7 gene (Kay et al., 2007). These results suggest that Xanthomonas‐induced hypertrophy in pepper and citrus depends on the activation of transcription factors and downstream genes that are either directly or indirectly involved in cell expansion. The CsSWEET1 and UPA16 genes in citrus and pepper, respectively, may be ‘collateral targets’ of Xanthomonas infection, with as‐yet‐unknown significance. However, the TALE‐driven activation of CsSWEET1 and UPA16 indicates that bacterial virulence strategies are highly complex. It has been suggested that some UPA genes are involved in hypertrophy‐unrelated functions. For example, they can have a role in disseminating Xcv between plants under field conditions (Kay et al., 2009).

The Arabidopsis genome contains 17 SWEET genes, most of which are induced upon exposure to either bacterial or fungal pathogens (Chen et al., 2010; Gupta et al., 2021). The expression of the clade III sucrose transporters AtSWEET10, 11, 12 and 15 is significantly increased when the plants are challenged with bacterial or fungal pathogens (Chen et al., 2010). Infection of Arabidopsis plants by the protist Plasmodiophora brassicae, the causative agent of clubroot disease in Brassica crops, led to the phloem‐specific accumulation of AtSWEET11 and 12 proteins at the site of infection, which facilitated the delivery of sugars to the pathogen (Walerowski et al., 2018). The atsweet11 single‐ and atsweet11;12 double‐knockout mutants had impaired sugar distribution to pathogens and were found to be less susceptible to P. brassicae, as demonstrated by their slower gall formation compared to their respective wild types (Li et al., 2018; Walerowski et al., 2018). These reports suggest that pathogens exploit the AtSWEET11 and 12 sugar transporters to acquire sugars for growth and propagation.

The atsweet11;12 double‐knockout mutants were resistant to the hemibiotrophic fungus Colletotrichum higginsianum. When the atsweet11;12 mutants were exposed to C. higginsianum, the leaf starch, soluble sugars and salicylic acid (SA) contents increased (Gebauer et al., 2017). Reduced leaf sugar contents were negatively correlated with C. higginsianum infection in Arabidopsis. Moreover, the low availability of carbohydrates in starchless Arabidopsis knockout mutants of phosphoglucomutase‐1 and ADP‐glucose pyrophosphorylase1‐1 resulted in reduced SA levels, which led to weakened SA‐regulated defence responses and increased susceptibility to C. higginsianum (Engelsdorf et al., 2013). On the other hand, elevated carbon status of the atsweet11;12 mutants correlated with high SA contents and the activation of SA‐mediated expression of pathogenesis‐related (PR) defence genes, factors that led to increased C. higginsianum resistance (Gebauer et al., 2017). The expression of the grapevine glucose transporter VvSWEET4 is strongly up‐regulated upon infection with the necrotrophic fungus Botrytis cinerea (Chong et al., 2014). The Arabidopsis atsweet4 knockout mutants (the closest homologue of VvSWEET4) were resistant to B. cinerea (Chong et al., 2014). These findings suggest that the infection‐induced transcriptional activation of VvSWEET4 increases glucose efflux from the cytoplasm to the apoplast to augment fungal growth (Chong et al., 2014). Overexpression of VvSWEET4 in grapevine hairy roots conferred resistance to the necrotrophic pathogen Pythium irregulare (Meteier et al., 2019). The resistant phenotype was ascribed to the higher glucose levels in roots following infection, which provided energy for synthesizing flavonoids with antifungal properties. VvSWEET4‐overexpressing hairy root lines had significantly higher flavanol contents than controls (Meteier et al., 2019).

SWEETs are bidirectional sugar transporters and can thus facilitate both the efflux and influx of sugars, depending on the sugar concentration on either side of the membrane. Expression of the tonoplast‐localized glucose transporter AtSWEET2 was induced by more than 10‐fold in roots by infection with the soil‐borne oomycete P. irregulare (Chen et al., 2015a). Higher levels of the AtSWEET2 protein inhibit the growth of P. irregulare because they can sequester more glucose in the vacuoles and limit glucose availability in the rhizosphere. The knockout mutants of atsweet2 were more susceptible to the oomycete than the wild type and exhibited growth defects after infection (Chen et al., 2015a). Thus, the infection‐induced activation of SWEETs can act either in favour or against the pathogen. In some species, such as tomatoes and bananas, down‐regulation of SWEETs expression was observed during the pathogen attack (Asai and Kobayashi, 2016; Miao et al., 2017). Down‐regulation of SWEETs might positively regulate plant defence responses by retaining and redistributing sugars for biochemical processes such as antioxidant defence reactions and producing secondary metabolites that are important under stress conditions. Decreased expression of SWEETs also reduces the flux of sugars to infection sites that inhibit pathogen growth. On the other hand, suppression of SWEET genes by pathogens might aid in their virulence as it can interfere with plant immune responses dependent on the sugar signalling defence pathway. SWEETs‐mediated release of sugars can trigger signalling cascades that can induce defence genes (Bezrutczyk et al., 2018b). Some fungal pathogens secrete small RNA effectors that hijack the RNAi machinery of the host plant and so silence immunity‐related genes (Weiberg et al., 2013). Further characterization of different SWEET mutants is required to understand how SWEETs modulate plant–pathogen interactions.

The activation of SWEETs is also observed in legumes that have nitrogen‐fixing symbioses with soil bacteria (rhizobia) and with arbuscular mycorrhizal (AM) fungi, suggesting that these transporters play a role in facilitating the infection efficiency by providing sugars to the symbiotic microorganisms (Desrut et al., 2020; Kryukov et al., 2021; Manck‐Götzenberger and Requena, 2016). The expression of LjSWEET3 is up‐regulated in Lotus japonicus following mycorrhizal or rhizobial infection (Sugiyama et al., 2017). Similarly, the expression of MtSWEET1b (a glucose transporter) and MtSWEET11 (a sucrose transporter) is increased in M. truncatula root cells after contact with the AM fungus Rhizophagus irregularis and the soil bacterium Sinorhizobium meliloti respectively (An et al., 2019; Kryvoruchko et al., 2016). Overexpression of MtSWEET1b in M. truncatula roots stimulated the growth of the intraradical mycelium during AM symbiosis (An et al., 2019). This finding indicates that more glucose was transported to the roots of the MtSWEET1b overexpressors, which enhanced AM growth (An et al., 2019). However, mycorrhizal colonization levels were unchanged when MtSWEET1b sugar transport activity was disrupted (An et al., 2019). Similarly, impairment of the activities of the LjSWEET3 and MtSWEET11 transporters did not affect symbiotic N₂ fixation (Kryvoruchko et al., 2016; Sugiyama et al., 2017). These data suggest that there is functional redundancy in the sugar transporters or that alternatives of sugar supplies are available to the symbiotic root tissues.

SWEET proteins for improved source–sink relationships

SWEETs in sink tissues

SWEETs in seeds and fruits

Understanding source–sink relationships is key to improving crop productivity, resilience and nutritional quality (Foyer and Galtier, 2017). More efficient photosynthate partitioning between photosynthesizing leaves (source) and heterotrophic sink tissues, such as seeds, stems, roots and flowers, is an agreed strategy to increase photosynthesis and yield (Ainsworth and Bush, 2011). In globally important food and oilseed crops such as rice, maize and soybean, SWEETs mediate sugar partitioning to seeds, thus determining seed carbohydrate, protein and oil contents, as well as seed size (Fei et al., 2021; Ma et al., 2017; Sosso et al., 2015; Wang et al., 2020; Yang et al., 2018). Early farmers and breeders selected larger seeds for their high caloric value, and thus SWEETs became selection targets inadvertently throughout crop domestication (Sosso et al., 2015). For example, the maize and rice transporter SWEET4, which mediates the transport of hexose sugars across endosperm membrane during grain filling, was selected during domestication for larger seed sizes. Knockout mutations in zmsweet4c and its rice orthologue ossweet4 resulted in defective seed filling in both maize and rice, respectively, indicating impaired import of hexoses into the endosperm (Sosso et al., 2015). The superior alleles of GmSWEET10a, GmSWEET10b and GmSWEET39, which are expressed in seeds, were selected during domestication for increased seed size and oil contents in soybean (Miao et al., 2020; Wang et al., 2020). The GmSWEET10a and 10b are expressed in the seed coat, and they transport both sucrose and glucose to the embryo during development. The CRISPR/Cas9‐edited gmsweet10a and gmsweet10b knockout mutants produce seeds but are smaller in size with reduced oil contents than the wild‐type controls. Double‐knockout mutants of gmsweet10a and gmsweet10b had a significantly reduced seed size and oil content compared to the single‐knockout mutants or wild‐type plants, suggesting their functional redundancy (Wang et al., 2020). Conversely, overexpression of GmSWEET10a and 10b by genetic transformation of their full‐length genomic sequences increased seed size and oil contents in transgenic soybean plants, possibly by enhancing sugar transport to the embryo (Wang et al., 2020). Similarly, the inbred soybean lines possessing superior GmSWEET39 alleles had higher oil contents than their parental accessions (Miao et al., 2020). Two soybean SWEET paralogues, GmSWEET15a and 15b, regulate embryo and seed development by mediating sucrose transport from endosperm to embryo (Wang et al., 2019b). Overexpression of GmSWEET15b in soybean resulted in increased seed sucrose contents (Wang et al., 2019b).

In Arabidopsis, AtSWEET11, 12 and 15 are expressed in the seed coat and endosperm to import sucrose into the embryo to support normal seed development (Chen et al., 2015b). In comparison to the single‐ and double‐knockout mutants of atsweet11, 12 and 15, the atsweet11;12;15 triple‐knockout mutants had more severe phenotypes such as retarded embryo development and lower seed weight with reduced starch and lipid contents, suggesting their functional redundancy during seed development (Chen et al., 2015b). Similarly, the rice OsSWEET11, 14 and 15 sucrose transporters, which show distinct expression profiles in seeds, are involved in carbohydrate import into the seeds. The ossweet11, 14 and 15 single‐knockout mutants had reduced seed weight and lowered seed starch contents because of aberrant grain filling (Antony et al., 2010; Fei et al., 2021; Li et al., 2021; Ma et al., 2017; Yang et al., 2018). Endosperm development and grain filling were much more severely affected in the ossweet11;15 double‐knockout mutants due to impaired sucrose transport in seeds, indicating their redundant roles in seed development (Yang et al., 2018).

The watermelon hexose transporter, ClSWEET3, the most highly expressed SWEET sugar transporter in parenchymal cells (or storage cells) during fruit development, is involved in hexose uptake from the apoplast (Ren et al., 2021). The expression level of ClSWEET3 is positively correlated with the sugar content of watermelon fruit. Overexpression of ClSWEET3 increased the sugar contents of watermelon fruit in transgenics compared to the wild type, whereas lack of ClSWEET3 activity reduced fruit sugar contents (Ren et al., 2021). It has been suggested that during watermelon domestication and evolution, ClSWEET3 was selected with other sugar‐metabolizing and transporting proteins such as ClAGA2 (C. lanatus alkaline alpha‐galactosidase 2), ClVST1 (vacuolar sugar transporter 1) and ClTST2 (tonoplast sugar transporter 2) for higher sugar content and fruit size (Ren et al., 2020a, 2021). The tomato sucrose transporter SlSWEET15, which functions in sucrose phloem unloading and its subsequent uptake into the storage parenchyma cells in fruits, is essential for fruit development (Ko et al., 2021). The grape hexose transporters VvSWEET10 and VvSWEET15, which are strongly expressed during fruit ripening, were shown to regulate the sugar contents of grape berries (Ren et al., 2020b; Zhang et al., 2019). The high expression of many SWEETs is associated with fruit development and the sugar contents of the fruits of other horticultural crops also, such as tomato, banana, loquat, apple, pineapple and litchi (Feng et al., 2015; Guo et al., 2018; Li et al., 2020; Miao et al., 2017; Xie et al., 2019; Zhen et al., 2018). Taken together, these reports show the potential of effectively targeting the import capacity of SWEETs allowing efficient unloading of sugars into harvestable sink organs, which may be used as an important strategy for crop yield improvements.

SWEETs in male and female reproductive organs

The glucose uniporters AtSWEET1 and AtSWEET8 (also known as RPG1, ruptured pollen grain 1) play a role in the nutrition of growing pollen tubes in Arabidopsis (Chen et al., 2010). In addition, AtSWEET8 is strongly expressed in male gametophytic tissues such as microspores and the tapetum (Chen et al., 2010; Guan et al., 2008; Yuan and Wang, 2013). The atsweet8 knockout mutants are male sterile, suggesting a role for AtSWEET8 in supplying glucose for pollen nutrition (Chen et al., 2010; Guan et al., 2008). The AtSWEET8 gene is also expressed in the embryo sac of female gametophytes, suggesting a role in the development of female reproductive organs (Yuan and Wang, 2013). The AtSWEET5 (also called VEX1), which transports glucose and galactose, shows high expression in vegetative pollen grains, where it may serve to supply sugar to the generative cells (Chen et al., 2010; Engel et al., 2005; Wang et al., 2022b). The tomato SlSWEET5 is expressed at high levels in stamens, where it functions in apoplastic hexose import for the development and maturation of pollen (Ko et al., 2022). The Arabidopsis AtSWEET13 and 14 sucrose transporters are involved in gibberellic acid (GA) transport, which is an important factor in male fertility (Kanno et al., 2016). The atsweet13;14 double‐knockout mutants with defective GA transport showed delayed anther dehiscence (Kanno et al., 2016). Exogenous GA application restored the normal anther dehiscence phenotype in atsweet13;14 double‐knockout mutants. These mutants also exhibited reduced pollen viability, fertility and germination (Kanno et al., 2016). Genetic complementation of atsweet13;14 knockout mutant with AtSWEET9, which transports sucrose but not GA, fully restored the normal pollen viability and germination phenotypes, suggesting that sucrose limitation was responsible for defective pollens (Wang et al., 2022a). These findings are validated by another study, in which the impaired pollen viability and germination phenotypes of atsweet13;14 double‐knockout mutants were rescued by the sucrose‐selective AtSWEET13 structural mutant (that transported only sucrose) but not by the GA‐selective AtSWEET13 mutant (high‐capacity GA transporter) (Isoda et al., 2022).

The rice OsSWEET11 has a role in pollen development. The RNAi mutants of ossweet11 were shown to be defective in pollen development (Chu et al., 2006; Yang et al., 2006). However, CRISPR/Cas9‐generated knockout mutants of ossweet11 had normal pollen development and viability (Kim et al., 2019; Yang et al., 2018). Recently it has been shown that, unlike the single‐knockout mutants, the double‐knockout mutants of ossweet11a;11b (OsSWEET11 is referred to as OsSWEET11a after the discovery of OsSWEET11b) are male sterile (Wu et al., 2022). The rice OsSWEET11a and 11b genes are expressed in the anther peduncle and the anther veins respectively. Simultaneous mutations in OsSWEET11a and 11b resulted in an insufficient supply of sugars to the pollen and lower pollen starch contents, together with defective pollen development (Wu et al., 2022). Reduced male fertility was also observed in ossweet11 RNAi mutants (Yang et al., 2006). This observation is suggested to be caused by co‐suppression of OsSWEET11b by the hairpin RNAi targeted against OsSWEET11 (Wu et al., 2022).

SWEETs in senescing and developing leaves

The expression of SWEETs in senescing leaves is associated with the remobilization of sugars from them to other sink tissues, which is vital for the reutilization of carbon resources in other plant organs. The Arabidopsis AtSWEET15 (SAG29 – senescence‐associated gene 29) gene and its homologue PbSWEET4 in pear are abundantly expressed in senescing and older leaves, respectively, suggesting a role in the remobilization of carbon resources to sink organs through these sugar exporters for their re‐assimilation in different plant tissues (Chen et al., 2010; Ni et al., 2020; Seo et al., 2011). Overexpression of PbSWEET4 is associated with accelerated leaf senescence and reduced sugar concentrations in leaves (Ni et al., 2020). The rice galactose transporter OsSWEET5, expressed in senescing leaves and other organs such as flowers and roots, has been suggested to control plant growth and senescence through crosstalk with auxin signalling (Zhou et al., 2014). Imbalances in sugar metabolism and transport led to stunted growth and premature senescence in OsSWEET5 overexpressors (Zhou et al., 2014).

The Arabidopsis glucose and fructose transporter, AtSWEET4, is expressed in the veins of leaves, where it is thought to be involved in sugar unloading and import into leaves (Liu et al., 2016). The RNAi atsweet4 lines were shorter in size and had lower levels of hexoses in the leaves than in the wild type. In contrast, the overexpression lines had greater biomass and accumulated higher glucose and fructose levels, suggesting that AtSWEET4 is involved in the import of hexose sugars to sink tissues (Liu et al., 2016). The tomato glucose transporter SlSWEET1a is involved in phloem unloading. It imports glucose into the phloem parenchyma cells from the apoplast of developing leaves (Ho et al., 2019). Silencing SlSWEET1a reduced the hexose contents of young leaves by 50%, with a concomitant increase in the hexose levels of mature source leaves (Ho et al., 2019). The fructose content of Arabidopsis leaves is dependent on the activity of a tonoplast‐localized fructose transporter called AtSWEET17 (Chardon et al., 2013). The leaves of the atsweet17 knockout mutants accumulate fructose to high levels compared to the wild type, confirming that AtSWEET17 is required to export fructose out of the vacuoles (Chardon et al., 2013).

SWEETs in flowers, stems and roots

Nectaries and tubers are important sink and storage organs for sugars derived from photosynthesis. The presence of chloroplasts in nectary parenchyma cells suggests that this tissue is capable of photosynthesis. Nectaries release a variety of sugars in the apoplastic space in order to attract pollinators. The sucrose synthesized from starch stored in A. thaliana, Brassica rapa and Nicotiana attenuate nectaries is secreted into the apoplast via the SWEET9 sucrose transporter (Lin et al., 2014). The petunia AtSWEET9 homologue, PhSWEET9 (PhNEC1), is found in nectaries. The expression of PhSWEET9 is inversely correlated with the starch content of the nectaries, suggesting functional similarities to AtSWEET9 (Ge et al., 2000; Lin et al., 2014). In Arabidopsis, FT is found to induce the expression of AtSWEET10 during the floral transition (Andrés et al., 2020). Transformed plants overexpressing AtSWEET10 flowered earlier than controls, suggesting that this transporter plays a role in providing sugars to sustain increased growth and cell division during the floral transition (Andrés et al., 2020). The potato StSWEET11, expressed in phloem, stolon and tuber parenchyma cells, mediates sucrose transport into the apoplastic space from sink tissues (Abelenda et al., 2019). The mode of sucrose unloading switches at the onset of tuberization from apoplastic to symplastic routes, which requires inhibition of StSWEET11 activity. The tuberization‐specific flowering locus T (FT) homologue StSP6A (self‐pruning 6A) physically interacts with StSWEET11 to block its activity and thus sucrose transport, favouring the symplastic sucrose transport routes (Abelenda et al., 2019). These reports highlight the crosstalk between SWEETs and FT that regulates the photoperiod‐dependent movement of sugars for the development of different sink organs.

The clade I rice glucose transporter OsSWEET3a, which is expressed in vascular bundles in the basal regions of seedlings such as the scutella and mesocotyl, is known to transport GA (Morii et al., 2020). Functional analysis of ossweet3a knockout mutants suggests that OsSWEET3a mediates the loading of both glucose and GA from the endosperm to the phloem for transport to the shoot in order to promote seedling growth and development (Morii et al., 2020). Differential expression patterns of the sucrose transporter genes AtSWEET11‐15 are observed in stems and roots of Arabidopsis under different physiological conditions (Durand et al., 2018), suggesting their implication in carbon partitioning during different stages of plant development and under variable growth conditions. For example, AtSWEET11 and 12 are strongly expressed in roots under osmotic stress conditions to enhance sucrose phloem unloading and increased carbon allocation to roots for improved osmotic stress tolerance (Durand et al., 2016, 2018).

SWEETs in source tissues

SWEETs in source tissues (photosynthesizing leaves) mediate sucrose efflux from phloem parenchyma cells to the apoplastic space, an essential step in the apoplastic sucrose phloem loading pathway (Chen et al., 2012). In Arabidopsis, clade III sucrose transporters AtSWEET11 and 12, which are localized to the plasma membrane of phloem parenchyma cells, mediate this step. The atsweet11;12 double‐knockout mutants had defective phloem loading and consequently restricted sucrose transport to sink tissues resulting in stunted growth and accumulation of starch and soluble sugars in leaves (Chen et al., 2012). In maize, the essential role of sucrose transport into the apoplast for phloem loading is carried out by three close paralogues ZmSWEET13a, b and c, which are highly expressed in leaves (Bezrutczyk et al., 2018a). The triple‐knockout mutants of zmsweet13a;b;c exhibited severe growth retardation and carbohydrate accumulation in leaves, indicating impaired phloem loading (Bezrutczyk et al., 2018a). Similarly, RNAi mutants of StSWEET11 also exhibited features of blocked phloem loading, such as reduced yield and elevated carbohydrate levels in leaves, suggesting that StSWEET11 is essential for apoplastic sucrose phloem loading in potatoes (Abelenda et al., 2019). Transcript and GUS expression studies have shown a high expression of IbSWEET10 in the leaf vasculature of mature sweet potatoes, suggesting that IbSWEET10 plays a role in sucrose export from source leaves and in phloem loading (Li et al., 2017).

The OsSWEET13 is the most highly expressed clade III SWEET gene in the rice leaf phloem, whereas OsSWEET11 and 14 are expressed at relatively low levels in this tissue (Eom et al., 2019; Ma et al., 2017; Mathan et al., 2021a; Yuan et al., 2014). All these genes encode sucrose transporters and are thought to fulfil roles in phloem loading (Eom et al., 2019). Analysis of single‐knockout mutants of ossweet11, 13 and 14 revealed no growth penalty. These mutants grew similar to the wild types (Eom et al., 2019; Fei et al., 2021; Ma et al., 2017). Similarly, double‐knockout mutants of ossweet11;14 and ossweet13;14 showed no noticeable phenotypic differences compared to the wild types, suggesting that SWEET‐mediated apoplastic sucrose phloem loading is not the predominant pathway used in rice for sucrose phloem transport (Eom et al., 2019; Fei et al., 2021). However, a previous report had shown that rice sweet14 T‐DNA insertional knockout mutants had a shorter plant phenotype and produced smaller seeds than the wild type (Antony et al., 2010), suggesting that phloem loading is impaired in these mutants.

Studies based on anatomical, physiological and genetic approaches support the idea that sucrose phloem loading in rice predominantly takes the apoplastic route mediated by SUTs and SWEETs (Braun, 2022; Li et al., 2022; Scofield et al., 2007; Wang et al., 2021). Chemical inhibition of SUTs in rice significantly reduced sucrose uptake, exudation and distribution to sink tissues such as stems and panicles (Li et al., 2022; Wang et al., 2021). Reducing the expression of OsSUT1, OsSWEET11 and OsSWEET14 in rice by knocking down/out OsDOF11 (DNA binding with one finger 11), decreased rates of sucrose uptake and transport in leaves resulting in stunted plants with fewer tillers and smaller panicles (Wu et al., 2018). The OsDOF11 binds to the promoter regions of OsSUT1, OsSWEET11 and OsSWEET14 genes to activate their expression (Wu et al., 2018). Conversely, increased expression of OsSUTs, OsSWEET11 and OsSWEET14 contributed to enhanced sucrose phloem loading and transport of photoassimilates, resulting in increased grain filling percentage and hence grain yield in rice under low nitrogen conditions (Li et al., 2022). The ossut1 knockout mutants showed reduced plant height and photosynthesis, together with lower thousand‐grain weights than the wild types, indicating that OsSUT1 is an important player in phloem loading and seed filling (Wang et al., 2021). However, the mild reduction in plant height and photosynthesis in ossut1 knockout mutants differs significantly from the severe growth phenotypes observed in principal sut/suc gene knockout mutants responsible for sucrose phloem loading in species using apoplastic phloem loading mechanism (Braun, 2022; Gottwald et al., 2000; Slewinski et al., 2009). Functional redundancy among rice SUTs could be one of the reasons for mild phenotypes in ossut1 mutants. The generation of higher‐order mutants of different SUTs would help reveal the precise path of phloem loading in rice.

All the SWEET genes that show source preferential expression patterns and are implicated in sucrose loading and transport to sink tissues belong to the clade III of sucrose transporters (Abelenda et al., 2019; Bezrutczyk et al., 2018a; Chen et al., 2012; Mathan et al., 2021a). Interestingly, the expression of the Arabidopsis AtSWEET11 and 12 genes and their rice orthologues OsSWEET11 and 14 is induced by the bacterial and fungal pathogens to acquire sugars for virulence and growth (Antony et al., 2010; Chen et al., 2010, 2012; Fatima and Senthil‐Kumar, 2021; Gao et al., 2018). Therefore, the increased expression of these SWEETs could alter the plant defence responses to pathogens as well as modulate the differential distribution of carbohydrates.

The overexpression of OsSWEET11/OsSWEET14 in rice resulted in stunted growth and lower grain weights (Figure 3) rather than stimulating growth and yield (Gao et al., 2018; Kim et al., 2020; Yuan et al., 2009). Similarly, Arabidopsis plants overexpressing AtSWEET11 and 12 were smaller than their respective wild types (Eom et al., 2015; Fatima and Senthil‐Kumar, 2021). Constitutive overexpression of OsDOF11, which regulates the expression of OsSUT1, OsSWEET11 and OsSWEET14 in rice, resulted in stunted plants with lower thousand‐grain weights (Kim et al., 2020; Wu et al., 2018). This is caused by the activation of OsSWEET14 expression in these lines (Kim et al., 2020). Overexpression of StSWEET11 in potatoes resulted in impaired growth, with the formation of aerial tubers and the outgrowth of above‐ground stolon‐like axillary buds, which commonly occurs in plants with blocked phloem transport (Abelenda et al., 2019). The phenotypes of the transgenic plants overexpressing clade III SWEET genes resemble those of their respective SWEET knockout mutants of atsweet11;12 in Arabidopsis (Chen et al., 2012) and ossweet14 in rice (Antony et al., 2010), and RNAi mutants of stsweet11 in potatoes (Abelenda et al., 2019), which show impaired phloem loading and reduced sugar translocation to sink tissues. SWEETs‐mediated sugar translocation is a highly regulated and directional process limited to specific cell types, such as vascular parenchyma cells in leaves (Eom et al., 2015). Ubiquitous and constitutive overexpression could thus result in random and directionless movement of sugar molecules from cells that normally do not function in sugar export. This can reduce sugar flux for phloem loading, negatively impacting plant growth (Eom et al., 2015; Feng and Frommer, 2015).

Figure 3.

Trade‐off between plant growth and immunity induced by SWEETs. Transgenic rice plants overexpressing clade III sucrose exporting SWEETs, OsSWEET11 and 14, have been shown to exhibit reduced growth parameters. However, these transgenics showed increased resistance to pathogens (Kim et al., 2020; Singh et al., 2021; Yuan et al., 2009). Similarly, overexpression of clade III ASWEET11 and 12 in Arabidopsis resulted in growth retardation, and increased tolerance to pathogens in plants overexpressing AtSWEET12 (Eom et al., 2015; Fatima and Senthil‐Kumar, 2021). These reports indicate that overexpression of SWEETs induced a trade‐off between plant growth and defence.

We recently simultaneously co‐expressed OsSUT1, OsSWEET11 and OsSWEET14 in rice under the control of their native promoters. The aim was to maintain tissue specificity and enhance sucrose translocation from leaves to sink tissues (Singh et al., 2021). However, the transgenic lines showed multiple phenotypes, such as stunting, reduced tiller numbers and decreased grain yields. These phenotypes were similar to those observed in transgenic rice plants that individually overexpress either OsSWEET11 or OsSWEET14 (Gao et al., 2018; Kim et al., 2020; Yuan et al., 2009). This finding suggests that the phenotypes result from increased OsSWEET11 and OsSWEET14 expression, not OsSUT1 overexpression. Rather, OsSUT1 overexpression has been shown to stimulate growth and yield in rice (Feng et al., 2018). Further analysis of the lines that co‐express OsSUT1, OsSWEET11 and OsSWEET14 revealed that these plants accumulated starch in their leaves at the end of the dark period and had reduced sucrose transport to sink tissues, indicative of impaired phloem loading (Singh et al., 2021). Moreover, the photosynthesis and sucrose synthesis rates were significantly reduced in transgenic plants compared to the wild type, leading to lower leaf‐soluble sugar contents (Singh et al., 2021).

It is predicted that the overexpression of clade III SWEETs would enhance sucrose availability in the apoplast leading to increased pathogen susceptibility. Surprisingly, the transgenic rice plants overexpressing OsSWEET11 and OsSWEET14 showed improved resistance to pathogens such as Xoo and R. solani that are known to induce the expression of these genes for virulence (Kim et al., 2020; Singh et al., 2021; Yuan et al., 2009). Similarly, constitutive overexpression of AtSWEET12 in Arabidopsis resulted in enhanced resistance to bacterial pathogens (Fatima and Senthil‐Kumar, 2021). Also, enhanced expression of BjSWEET11 in Brassica juncea has been shown to attenuate the growth and multiplication of aphids feeding on transgenic plants compared to controls (Zauva et al., 2020). Up‐regulation of SWEETs could trigger defence responses via the translocation of sugars to the sites of infection, which may include enhanced expression of sugar transporter proteins (STPs) (Bezrutczyk et al., 2018b). STPs are proton–hexose symporters that retrieve hexoses from apoplastic space to the cytoplasm in response to the perception of pathogens to reduce the sugar levels in the apoplastic space in order to restrict pathogen growth, supporting the ‘pathogen starvation hypothesis’ (Figure 4) (Bezrutczyk et al., 2018b; Lemonnier et al., 2014). It is also plausible that increases in SWEETs activities transiently elevate apoplastic sugar levels to activate defence reactions via ‘sugar signalling’ pathways that confer pathogen resistance (Bezrutczyk et al., 2018b). Overexpression of BjSWEET11 in Brassica juncea is accompanied by transcriptional activation of many defence‐related genes such as ABC transporter G family member 36 (ABCG36), pathogenesis‐related protein 1 (PR1) and WRKY transcription factor 70 (WRKY70), which are known to regulate plant defence responses (Zauva et al., 2020).

Figure 4.

A possible model to explain the compromised growth and increased resistance to pathogens in plants overexpressing clade III SWEET genes. Overexpression of SWEETs that are induced by pathogens for virulence such as AtSWEET11 and 12 in Arabidopsis and OsSWEET11 and 14 in rice is perceived as pathogen attack in plants. This activates plant defence machinery for ‘pathogen starvation’ resulting in down‐regulation of biosynthetic processes such as photosynthesis and sucrose synthesis. This reduces sugar concentrations in the leaves and apoplastic space which limits pathogen growth and also impacts plant growth negatively (Singh et al., 2021). Overexpression of SWEETs could also increase secretion of sucrose into the apoplast. Sucrose in the apoplast is partially cleaved by cell wall invertases into glucose and fructose. Increased sugar levels in the apoplast serve as pathogen elicitors to activate STPs (proton–hexose symporters) for retrieval of hexoses from the apoplastic space. This reduces the concentration of sugars in the apoplast, thereby inhibiting pathogen growth and reducing sucrose flux to sink tissues affecting plant growth (Bezrutczyk et al., 2018b; Lemonnier et al., 2014).

We found that co‐expression of OsSWEET11 and OsSWEET14 in rice reduced sucrose synthesis and transport (Singh et al., 2021). This indicates that the up‐regulation of OsSWEET11 and OsSWEET14 in rice mimics the attack of Xoo. The ‘pathogen starvation hypothesis’ response strategy serves to prevent Xoo growth and proliferation by inhibiting sucrose synthesis via decreased expression of the genes encoding the enzymes of the sucrose synthesis pathway, as well as restricting sucrose transport (Figure 4) (Bezrutczyk et al., 2018b; Singh et al., 2021). Supporting evidence for this concept has come from another very recent study in which the overexpression of AtSWEET12 in transgenic Arabidopsis plants conferred resistance to bacterial pathogens. The transgenic lines had 50% less sucrose in the apoplast than the wild types (Fatima and Senthil‐Kumar, 2021). Similarly, overexpression of the IbSWEET10 gene in sweet potatoes reduced leaf sugar levels, which led to enhanced resistance to Fusarium oxysporum. Increased resistance to pathogens, however, came at a cost in the transgenic rice and Arabidopsis plants overexpressing OsSWEET11/14 and AtSWEET12 respectively. Reduced sucrose levels in the leaves of transgenic plants resulted in poor plant growth and yield (Fatima and Senthil‐Kumar, 2021; Singh et al., 2021). These results indicate that there is a trade‐off between plant growth and defence that is modulated by the SWEETs (Figure 3).

Exploiting SWEETs for improved plant growth and defence

The above discussion has highlighted the tight linkage between plant growth and pathogen defence and demonstrated that these traits are inversely related in plants overexpressing clade III SWEETs that serve a dual function in sucrose export from source tissues and in pathogenesis. Any changes in the expression levels of clade III SWEETs disturb the balance between sugar transport and pathogen susceptibility. Novel and innovative approaches that uncouple pathogenesis and sugar translocation are required to improve source–sink relationships and yield sustainability.

Sugar signalling is a central component of plant defence responses activated during plant–microbe interactions. The sucrose non‐fermenting‐1‐related protein kinase 1 (SnRK1) and the target of rapamycin (TOR) kinase are key regulatory nodes in the sugar‐sensing cascade. These kinases regulate the growth–immunity trade‐off via multiple transcriptional and post‐translational mechanisms, hormonal crosstalk and carbohydrate partitioning (Margalha et al., 2019). The SnRK1 positively regulates immunity while inhibiting growth. In contrast, TOR promotes growth and limits immune responses to pathogens (Choudhary et al., 2022). Transgenic rice plants overexpressing OsSnRK1 showed reduced growth and yield but exhibited broad‐spectrum resistance to bacterial and fungal pathogens such as Xoo and R. solani (Filipe et al., 2018). Similar phenotypes of impaired growth and enhanced resistance to Xoo and R. solani were also observed in transgenic rice plants overexpressing OsSWEET11 and 14 (Kim et al., 2020; Singh et al., 2021; Yuan et al., 2009). These findings hint towards the involvement of SnRK1 in regulating the growth–immunity trade‐off in SWEET transgenics. SnRK1 is, therefore, an attractive candidate for targeted modulation of plant growth and defence responses in transgenics overexpressing SWEETs.

The CRISPR‐Cas‐mediated tissue‐specific expression of OsDOF11 increased grain yields, as well as resistance to sheath blight, via activation of OsSWEET14 (Kim et al., 2020). The regulated expression of SWEETs in a tissue‐specific manner could, therefore, have positive effects on both yield and disease resistance. The indirect regulation of SWEET gene expression could also be beneficial. For example, the FT genes in Arabidopsis and potatoes regulate the expression of clade III SWEET genes that encode sucrose transporters (Abelenda et al., 2019; Andrés et al., 2020). The FT gene in Arabidopsis regulates flowering by enhancing the expression of AtSWEET10. Similarly, the FT homologue StSP6A inhibits StSWEET11 activity to promote tuberization (Abelenda et al., 2019; Andrés et al., 2020). Thus, the expression of FT genes could be tailored to modulate the expression of SWEETs and hence influence yield‐related traits such as early flowering and tuber formation.

Increased oligomeric status of SWEETs enhances their sugar transport activity by forming a functional pore (Anjali et al., 2020; Xuan et al., 2013). The phosphorylation of AtSWEET11 and 12 enhanced oligomerization and, consequently, the sucrose transport activity of these SWEETs. This resulted in increased sucrose content in the roots leading to improved growth under drought stress (Chen et al., 2022a). Therefore, increasing the activity of SWEETs in source tissues through post‐translational modifications such as protein phosphorylation could be an attractive strategy for enhanced photoassimilate transport to sink tissues. The in silico analysis of SWEET proteins from various species revealed many putative phosphorylation sites in the C‐terminal cytosolic region. The functions of each of these phosphorylation sites can be tested experimentally to determine whether they play a role in the regulation of SWEET activity (Anjali et al., 2020).

Enhancement in photosynthesis could translate to increased biomass and yield (Foyer et al., 2017). Some members of the SWEET family have been shown to regulate photosynthesis rates. Overexpression of the putative glucose transporter ZjSWEET2.2 in Ziziphus jujuba increased photosynthetic carbon assimilation in the leaves, presumably through improved phloem loading (Geng et al., 2020). Enhanced phloem loading is expected to decrease the carbohydrate levels in leaf mesophyll cells, which will enhance photosynthetic gene expression (Geng et al., 2020). The maize CST1 (CLOSED STOMATA 1) gene, which is a homologue of AtSWEET1, regulates stomatal movements and photosynthesis (Wang et al., 2019a). CST1 is a glucose transporter localized on the plasma membrane of subsidiary cells. CST1 regulates stomatal movements by regulating the sugar contents of the subsidiary cells (Wang et al., 2019a). Loss of CST1 functions resulted in decreased stomatal opening and lower photosynthesis rates, leading to lower grain yields (Wang et al., 2019a). Therefore, the regulated expression of CST1 might be exploited to optimize leaf CO₂ concentrations under different environmental conditions to optimize photosynthesis (Wang et al., 2019a).

Genome‐wide association mapping and transcriptional analysis in species such as rice, maize, soybean and watermelon revealed candidate SWEET genes that were selected during the process of crop domestication for yield‐related traits such as bigger grains and fruits, and seeds with higher oil content (Mathan et al., 2021a; Miao et al., 2020; Ren et al., 2021; Sosso et al., 2015; Wang et al., 2020). Genome‐wide association studies (GWAS) in other crop species could help identify novel sugar transporters, which can be targeted to develop new crop varieties with improved yields through genetic engineering and breeding approaches. Multi‐locus genome‐wide association mapping in bread wheat identified a novel candidate SWEET transporter that could influence spike‐related traits and hence the grain yield component (Malik et al., 2021). Similarly, GWAS and QTL mapping approaches can help identify SWEET alleles for disease resistance. Meta‐QTL analysis on QTLs obtained from 58 different studies involving 62 different mapping populations for multiple disease resistance in wheat identified 194 differentially expressed candidate genes, including a SWEET sugar transporter (Saini et al., 2022).

Role of SWEETs in C4 photosynthesis

C4 plants such as maize, Setaria viridis and Sorghum bicolor possess high rates of photosynthesis and have high expression of sugar transporters to facilitate phloem loading of photosynthates. The loading of photoassimilates into the phloem of C4 leaves via sugar transporters is required because there are fewer plasmodesmata between the bundle sheath cells and vascular parenchyma than between the bundle sheath and mesophyll cells (Emms et al., 2016). The multiple paralogues of SWEET13, which are highly expressed in the bundle sheath cells of mature leaves in maize (three paralogues), Setaria viridis (two paralogues) and Sorghum bicolor (three paralogues), arose during the evolution of C4 plants that encompasses independent and parallel duplication events. This has led to multiple copies of this gene in C4 plants compared to the single copy in rice, i.e. orthologue OsSWEET13 (Emms et al., 2016). The maize SWEET13a, b and c paralogues are highly expressed in leaves and have been shown to transport sucrose from the bundle sheath cells to the apoplast (Bezrutczyk et al., 2018a). Similarly, based on transcript abundance and immunolocalization studies, SWEET13 paralogues in Setaria viridis and Sorghum bicolor have been suggested to facilitate apoplastic phloem loading of photoassimilates (Chen et al., 2022b). The SvSWEET13a and 13b transporters have high glucose and sucrose transport capacity. They are localized in the bundle sheath and phloem parenchyma cells around both minor and major veins, where they are suggested to load photosynthates into the phloem (Chen et al., 2022b). The duplication of SWEET13 paralogues, therefore, played an important role in the evolution of C4 photosynthesis.

SWEETs and abiotic stress response in plants

The intracellular and intercellular levels of sugars and sugar alcohols are regulated by the SWEETs expressed in the vacuolar and plasma membranes (Breia et al., 2020, 2021). These transporters can contribute to acclimation to stress conditions. High levels of soluble sugars such as sucrose, glucose, trehalose and sugar alcohols mitigate abiotic stress by acting as osmoprotectants and antioxidants that stabilize membrane structures by interacting with the lipid bilayer (Ahmad et al., 2020; Singh and Thakur, 2018). The functional characterization of the plasma membrane‐localized AtSWEET4, AtSWEET11 and AtSWEET12 genes and the Arabidopsis tonoplast‐localized AtSWEET16 and AtSWEET17 genes revealed a role in freezing tolerance (Chardon et al., 2013; Guo et al., 2014; Klemens et al., 2013; Le Hir et al., 2015; Liu et al., 2016). Transgenic Arabidopsis plants overexpressing AtSWEET4 and AtSWEET16 showed a higher freezing tolerance than the controls linked to the higher accumulation of glucose and sucrose, both of which have anti‐freezing properties. The transgenics also showed a reduced leakage of electrolytes following exposure to cold stress, a finding which indicates that membrane integrity is maintained (Klemens et al., 2013; Liu et al., 2016). Similarly, the heterologous expression of the Camellia sinensis CsSWEET1a and CsSWEET17 genes, which are expressed upon exposure to cold stress, improved freezing tolerance in transgenic Arabidopsis plants (Yao et al., 2020). The study also reported a role of the alternative splicing in these genes in cold stress responses in C. sinensis. An alternative splice form of CsSWEET17, which is found to be localized in the cytosol, is strongly expressed during winter (Yao et al., 2020).

The AtSWEET17 fructose transporter is highly expressed in the tonoplast of roots to regulate the levels of fructose in roots in response to cold and drought stress (Guo et al., 2014; Valifard et al., 2021). The AtSWEET17 is essential for the fructose‐induced modulation of root growth and architecture, which is critical for drought stress tolerance (Valifard et al., 2021). The expression of AtSWEET10 was significantly enhanced during waterlogging, which induces ethylene production (Phukan et al., 2018). The waterlogging‐responsive ethylene response factor MaRAP2‐4 from Mentha arvensis was shown to regulate the expression of AtSWEET10 and hence the sugar contents of transgenic Arabidopsis plants by binding to the stress‐responsive elements on the promoter region of AtSWEET10 (Phukan et al., 2018). The OsSWEET13 and OsSWEET15 genes were strongly expressed in response to drought, salt and abscisic acid (ABA) treatment (Mathan et al., 2021b). These sucrose transporters were linked to increased levels of soluble sugars in the leaves, roots and phloem sap of plants exposed to drought or salinity stress. The ABA‐responsive OsbZIP72 transcription factor binds to the promoters of the OsSWEET13 and OsSWEET15 genes, activating their expression (Mathan et al., 2021b). Understanding the gene regulatory networks that control the stress‐induced expression of SWEETs could help improve abiotic stress tolerance in crop plants.

Enhanced expression of AtSWEET11 and 12 under water‐deficit conditions increased carbohydrate export from leaves to roots to counter drought stress (Durand et al., 2016). A recent study revealed the mechanistic basis of increased AtSWEET11 and 12 activities on exposure to drought stress (Chen et al., 2022a). Drought‐induced ABA signalling in Arabidopsis leads to the activation of SnRK2s, which phosphorylate two highly conserved serine residues, Ser237 and Ser248, in the cytosolic carboxy‐terminal regions of the key phloem sucrose transporters AtSWEET11 and 12 (Chen et al., 2022a). Phosphorylation at these sites enhances AtSWEET11 and 12 protein oligomerization, a process that results in increased sucrose transport activity to roots and a concomitant enhancement in root:shoot ratio of biomass and drought tolerance (Chen et al., 2022a). Crucially, the transgenic plants expressing phospho‐mimic mutants (Ser‐to‐Asp/Glu mutations) of AtSWEET11 or 12 showed the development of a deeper, broader and improved root system that provided growth advantages under both normal and drought stress conditions (Chen et al., 2022a). Thus, enhanced source strength and transport can be an effective strategy for combating the adverse effects of stress without negative effects on plant growth.

Conclusions

Since the discovery of SWEETs nearly a decade ago, an overwhelming body of data has demonstrated that SWEETs regulate diverse physiological processes in plants, particularly plant–pathogen and source–sink interactions (Breia et al., 2021; Chen et al., 2010, 2012; Cox et al., 2017; Hu et al., 2014; Ren et al., 2021; Wang et al., 2020). Thus, the genetic manipulation of SWEETs is considered an effective strategy for developing disease‐resistant crops and increasing source and sink strength, which can boost crop productivity. With a few exceptions, pathogens activate the expression of SWEET genes to acquire nutrition to support growth. Remarkable success has been achieved in generating broad‐spectrum BB‐resistant rice varieties that have a higher yield potential under biotic stress conditions, specifically by the gene editing of OsSWEET11, 13 and 14, which Xoo induces to cause BB (Eom et al., 2019; Li et al., 2012; Oliva et al., 2019; Xu et al., 2019). Similarly, the SWEET genes in other crops are attractive potential targets for improved disease resistance.

In contrast, the significance of SWEETs in improving carbohydrate export from leaves to seeds and thus enhancing photosynthesis and crop yield remains unclear. The overexpression of many SWEET genes such as OsSWEET11 and 14 in rice, AtSWEET11 and 12 in Arabidopsis and StSWEET11 in potatoes resulted in stunted growth and reduced yields, parameters indicative of impaired phloem loading (Abelenda et al., 2019; Eom et al., 2015; Gao et al., 2018; Kim et al., 2020; Singh et al., 2021; Yuan et al., 2009). Alternate strategies such as tissue‐specific controlled overexpression via CRISPR/Cas and enhancement in the activity by increasing the oligomeric state through phosphorylation of SWEET transporters could be employed to augment the rate of sugar transport to sink tissues. Such approaches may potentially reduce the risks of unintended consequences of uncontrolled overexpression.

The SWEET family is the most recently discovered class of sugar transporters and is, therefore, understudied. Thus, the physiological roles of the many SWEET genes present in different plant species must be studied in detail to identify suitable candidates for yield enhancements in crops through improved carbon partitioning. GWAS and QTL mapping are powerful tools that can be used to reveal candidate SWEET genes controlling traits for carbohydrate partitioning in source and sink tissues.

Based on the reports implicating SWEETs in plant responses to environmental stress (Chen et al., 2022a; Klemens et al., 2013; Le Hir et al., 2015), it is possible to suggest that the expression of SWEETs can be tweaked to enhance plant resistance to different abiotic stress conditions. Hence, the modulation of SWEETs may have a significant role in enhancing plant resilience and the sustainability of yield in the face of climate change and accompanying temperature variations.

Conflict of interest

The authors declare no conflict of interest.

Authors' contributions

Conceptualization, J.S. and J.K.T. Manuscript preparation, J.S., J.K.T., C.H.F., A.R., K.J.G. and S.D. Funding acquisition, J.K.T. All authors have read and agreed to the published version of the manuscript.