In fleshy fruits, soluble sugars, including sucrose, fructose, and glucose, are not only essential for fruit growth and development but also central to fruit quality. Fruit taste and flavor is closely related to the composition and concentration of sugars and their balance with acids. As the composition and concentration of sugars at fruit maturity is determined by metabolic and transport processes during fruit development, understanding these processes and their regulation is important for fruit quality improvement. At the center of sugar metabolism in sink cells is the Sucrose cycle, previously named the Sucrose–Sucrose cycle or the futile Sucrose recycle, which consists of the breakdown of sucrose by invertase and sucrose synthase, the phosphorylation of the resulting hexoses and the interconversion between hexose phosphates and UDP-glucose, and the re-synthesis of sucrose via sucrose-6-phosphate synthase and sucrose-6- phosphate phosphatase. This metabolic system connects sugar metabolism with many other metabolic pathways such as glycolysis and tricarboxylic acid cycle, starch synthesis, and cellulose synthesis, and its coordination with the sugar transport system on the tonoplast is expected to determine the partitioning of sugars between metabolism in the cytosol and accumulation in the vacuole. In fleshy fruits, the concentration and distribution of sugars in parenchyma cells are affected via this cycle by developmental processes and environmental factors. However,plastic pots for planting the biochemical regulation of the cycle and the associated transport system is not fully understood.

In apple and many other tree fruit species of the Rosaceae family, sorbitol is a primary end product of photosynthesis and a major phloem-translocated carbohydrate, accounting for 60–80% of the photosynthates produced in apple leaves and transported in the phloem. In source leaves, sorbitol is synthesized from glucose-6-phosphate in a two-step process: G6P is first converted to sorbitol-6-phosphate via aldose-6-phosphate reductase , then followed by dephosphorylation of S6P to sorbitol via S6P phosphatase. The loading of both sorbitol and sucrose into the companion cell-sieve element complex in the phloem is passive and symplastic in apple, but their phloem unloading in fruit involves an apoplastic step.Once released from the SE-CC complex of the phloem in apple fruit, sorbitol is taken up into the cytosol of parenchyma cells by plasma membrane-bound sorbitol transporters and then converted to fructose by sorbitol dehydrogenase ; sucrose is either directly taken up into parenchyma cells by sucrose transporters , or first converted to glucose and fructose by cell wall invertase and then transported into the parenchyma cells via hexose transporters. Compared with plants that transport and utilize only sucrose, such as Arabidopsis, tomato , and poplar , apple is unique in that both sorbitol and sucrose are transported in the phloem and are metabolized in sink organs. It is estimated that >80% of the total carbon flux goes through fructose in apple. Once taken up into parenchyma cells of fruit, both sorbitol and sucrose feed into the Sucrose cycle to meet the carbon requirement for fruit growth and development while excess carbon is converted to starch for storage in plastids or transported into vacuole by sugar transporters for accumulation. Although we have characterized the genes and proteins involved in sugar metabolism and accumulation in apple, it remains unclear how apple trees adjust the Sucrose cycle and the transport system in response to altered supply of sorbitol and sucrose from source leaves.

In transgenic apple trees with antisense suppression of A6PR, leaf sorbitol concentration is dramatically decreased, whereas sucrose concentration is significantly elevated in the source leaves, but neither leaf CO2 assimilation nor plant vegetative growth is altered. The decreased sorbitol synthesis leads to significant changes in the expression profile of key genes in leaf starch metabolism and many stress response genes. In addition to being a key metabolite in carbohydrate metabolism, sorbitol also acts as a signal regulating stamen development and pollen tube growth and resistance to Alternaria alternata in apple. In the shoot tips of the A6PR transgenic plants, both the activity and transcript level of SDH are downregulated, whereas those of sucrose synthase are upregulated in response to a lower sorbitol but higher sucrose supply. Teo et al.reported that fruit of the transgenic apple trees accumulated a higher level of glucose and lower levels of fructose and starch at maturity, but no significant difference was detected in the activity of key enzymes in sugar metabolism, CWINV, neutral invertase , fructokinase , hexokinase , or SPS between the transgenic lines and the untransformed control . Considering that antisense suppression of A6PR has drastically decreased leaf sorbitol level and increased sucrose level, leading to less sorbitol but more sucrose being transported in the phloem; and both transcript levels and activities of SDH and SUSY responded to the altered sorbitol and sucrose supply in the shoot tips of the transgenic plants, we predicted that the decreased supply of sorbitol and increased supply of sucrose would lead to down regulation of sorbitol metabolism and upregulation of sucrose metabolism in the transgenic fruit as well. The discrepancy between the data obtained by Teo at al. and our predicted responses on the activities of sucrosemetabolizing enzymes in the transgenic fruit has prompted us to re-evaluate sugar metabolism and accumulation in the fruit of these transgenic plants to better understand how the Sucrose cycle and the sugar transport system respond to an altered supply of sorbitol and sucrose.

Antisense suppression of A6PR significantly decreased sorbitol concentration but increased sucrose concentration while largely maintaining fructose and glucose concentrations in source leaves throughout fruit development in the two transgenic lines relative to the untransformed CK . Sorbitol concentration in the source leaves of antisense line A27 was decreased to ~70% initially and 13% at harvest of that detected in CK. For antisense line A04, sorbitol concentration was decreased to 32% initially and 10% at harvest of the CK level. By contrast, sucrose concentration in the source leaves of A27 and A04 was much higher than in CK throughout fruit development, with larger differences detected at later developmental stages . Concentrations of sorbitol and sucrose were also measured for source leaves, leaf petioles, and fruit pedicels at 75 days after bloom . Compared with CK, antisense lines A27 and A04 had lower concentration of sorbitol, higher concentration of sucrose, and lower molar ratio of sorbitol to sucrose in the source leaves,strawberries in a pot leaf petioles, and fruit pedicels. The abundance of sorbitol followed the order of source leaves > leaf petioles > fruit pedicels .Average fruit fresh weight did not differ significantly between the two antisense lines and CK during fruit development except for about a 10% lower value detected for A27 and A04 at 108 DAB and at harvest . Average fruit dry weight did not show any significant difference throughout fruit development . Dark respiration was ~1.5–1.9-fold higher in A27 and A04 fruits than in CK fruits between 40 and 108 DAB during fruit development, but no significant difference was detected at harvest . Fruit yield per tree was significantly lower in the two antisense lines than in CK, largely due to lower average fruit weight at harvest as fruit number per tree was not significantly different between the two antisense lines and CK .Suppression of sorbitol synthesis in source leaves led to a significant decrease in sorbitol concentration in the fruit of two antisense lines A27 and A04 throughout fruit development, particularly in A04 . However, sucrose concentration was similar in the fruits of the two antisense lines and CK during fruit development with a higher level detected in the transgenic fruit only at 74 DAB. Fructose concentration showed no difference between the transgenic fruit and CK except being slightly lower at 108 DAB in the transgenic fruit. Compared with CK, concentrations of glucose and galactose were much higher throughout fruit development, with larger differences detected at later developmental stages. Concentrations of G6P and fructose-6- phosphate decreased during fruit development and were significantly lower in A27 and A04 than in CK from40 to 108 DAB . At fruit maturity , total soluble solids concentration was significantly higher in A27 and A04 than in CK . Fruit starch concentration did not show obvious difference between the transgenic lines and CK before 74 DAB but was slightly lower in A27 and A04 than in CK after 108 DAB .SDH activity decreased during fruit development and was significantly lower in both A27 and A04 than in CK at each developmental stage . CWINV activity dropped dramatically from 40 to 74 DAB and then remained fairly constant to maturity, but no significant difference was detected between the two antisense lines and CK.

NINV activity decreased throughout fruit development and was ~1.5–2.0-fold higher in A27 and A04 than in CK from 74 DAB to fruit maturity . Vacuolar acid invertase activity showed no significant difference between the two antisense lines and CK except a slightly higher activity detected in A27 and A04 than in CK at 108 DAB. SUSY activity declined during fruit development and was significantly higher in A27 and A04 than in CK from 40 to 108 DAB. FK activity decreased during fruit development and was significantly lower in A27 and A04 than in CK at 40 and 74 DAB. HK activity decreased during fruit development and was significantly higher in both A27 and A04 than in CK from 74 to 134 DAB. SPS activity increased slightly from 40 to 108 DAB and then dramatically to fruit maturity, with a significantly lower activity detected in both A27 and A04 from 40 to 108 DAB.Our data clearly showed that sorbitol concentration was significantly lower, whereas sucrose concentration was significantly higher in the source leaves of 5-year-old transgenic “Greensleeves” apple trees with antisense suppression of A6PR compared with the untransformed CK throughout fruit development. These results are consistent with those reported for the 1-year-old transgenic trees. The higher sucrose concentration in the source leaves is an indication that a larger proportion of the photosynthetically fixed carbon ends up in sucrose over a 24-h period because most of the starch accumulated during the day breaks down for sucrose synthesis at night in the transgenic plants although no difference in the carbon flux to sucrose during the day was detected. As both sorbitol and sucrose diffuse into SE-CC complex from mesophyll cells via plasmodesmata,accumulation of a higher level of sucrose in leaves is expected to facilitate the transport of sucrose in the phloem when less sorbitol is translocated in the transgenic plants. The lower concentration of sorbitol and higher concentration of sucrose in both leaf petiole and fruit pedicel and a smaller ratio of sorbitol to sucrose indicate that significantly less sorbitol but much more sucrose is translocated from leaves to fruit in the transgenic trees, which is consistent with a lower sorbitol but a higher sucrose concentration in the phloem exudates collected from fruit pedicels of these plants. The total amount of carbon translocated to fruit is expected to be very similar between the transgenic lines and the CK because all the trees had a very similar cropload and no significant difference was detected in average fruit dry weight between the transgenic lines and the CK at fruit maturity . These data clearly demonstrate that, when sorbitol synthesis is decreased in the source leaves, more sucrose is synthesized in the leaves and translocated to the fruit, thereby largely maintaining fruit growth and development. This is also consistent with the homeostasis of vegetative growth observed in the transgenic lines. The transgenic trees with decreased sorbitol synthesis grown under our experimental conditions were only slightly smaller after 5 years of growth than those of the untransformed CK . This is consistent with comparable photosynthetic rates measured in the transgenic lines and the untransformed CK throughout the growing season , with the lower rates detected only at fruit harvest being largely related to the leaf brown spots caused by Alternaria alternata in the transgenic lines. However, Teo et al.found that the transgenic trees were much smaller than the CK trees. This discrepancy is likely due to differences in growing conditions between the two locations. As sorbitol is implicated in drought-stress tolerance in apple, these trees might have experienced more drought stress under warm and dry conditions in California than under cool and humid conditions in upstate New York.