|Ali, Hazem Abd El-Rahman Obiadalla: Understanding of Carbon Partitioning in Tomato Fruit |
The Arabidopsis model system has been contributed much to the remarkable advances in plant molecular biology during the last decade. The major reasons for the success of Arabidopsis are its small size, short life cycle, small genome (Leutwiler et al., 1984) and easy of transformation (Bechtold et al., 1993) These features facilitate the genetic dissection of any trait through the screening of large populations saturated in mutants for the various genes involved in the trait. Nevertheless, despite the considerable advantages of Arabidopsis, the knowledge acquired in this species cannot always be applied to other plant species. Having a silique type of fruit makes Arabidopsis a good model for species of the Brassicaceae but not for those with a fleshy fruit.
Tomato (Lycopersicon esculentum) offers a good model for other crop species whose fruit is also a fleshy berry. It is one of the most important crops in the fresh vegetable market as well as in the food processing industry (Rick and Yoder, 1988; Hille et al., 1989). It is well characterised genetically; it has a relatively small diploid genome (n=12) and is readily transformable (McCormick et al., 1986). One disadvantage of tomato is that the plants have a large size and relatively long live cycle.
A new cultivar (Micro-Tom) has, however, recently been developed that overcomes these problems (Meissner et al., 1997). The plants of this variety grow to a similar size as Arabidopsis and have a considerably shorter life cycle than other tomato varieties, routinely producing seed within twelve weeks of being planted.
We wish to study tomato fruit carbohydrate metabolism in the Micro-Tom cultivar using genetic engineering techniques. As an initial study we decided to examine the activity of enzymes in untransformed fruit during development to see whether this new variety is equivalent to other varieties that have been studied previously, and as such to ascertain if it represent a useful model for fruit carbohydrate metabolism.
The development of fruit of the Micro-Tom cultivar are shown in Fig. 4. The fruits were small and green at 20 DAF and remained green until 45-50 DAF when they reached the breaker stage. By 60 DAF they were fully ripe.
Figure 4: Developmental series of tomato fruits from Micro-Tom cultivar.
DAF = Days after Flowering.
Starch and soluble sugar contents were determined in both the pericarp and placental tissues between 20-60 DAF Starch accumulated transiently in the fruits of the Micro-Tom variety in both the pericarp and placental tissues (Fig. 5A). In both these tissues the starch content was approximately 45µmol hexose (g FW)-1 at 20 DAF. In the pericarp the decrease was quickly to under 10µmol hexose (g FW)-1 at 30 DAF, before decreasing more slowly to barely detectable amounts at 45 DAF. In the placental tissue the decrease was slower and more linear, reaching barely detectable amounts at 60 DAF. There were no significant differences in starch concentrations between the pericarp and placental tissues between 25 and 35 DAF. After that the concentrations were significantly decreased in the pericarp in comparison with the placenta.
46Sucrose levels remained relatively constant in the different tissues throughout fruit development, although they were generally greater in the placental tissue in comparison with the pericarp (Fig. 5B). There were no significant differences in sucrose concentrations between the pericarp and placenta, except for three time points. At 20, 25 and 30 DAF in the placental tissue there were significantly increased sucrose in comparison with the pericarp. In both tissues the concentration stayed below 7µmol hexose (g FW)-1 throughout development.
Figure 5: Starch and soluble sugar contents in pericarp and placental tissues of tomato cultivar Micro-Tom during development. (A) Starch. (B) Sucrose. (C) Fructose. (D) Glucose. Data represent the mean of five independent measurements + SE.
47Changes in both fructose (Fig. 5C) and glucose (Fig. 5D) concentrations showed a similar pattern. In the pericarp they remained relatively constant between 20-50 DAF, before increasing at the end of development. Fructose levels also remained relatively constant in the placenta between 20-45 DAF whilst over this time period glucose levels decreased slightly. After 45 DAF the concentrations of both fructose and glucose increased. Both glucose and fructose levels were generally lower in the placenta than in the pericarp, but in both tissues their concentrations were an order of magnitude higher than the concentration of sucrose. Fructose concentrations were significantly increased in the pericarp in comparison with the placenta, but only at four time points (20, 35, 45 and 60 DAF). Glucose concentrations were also significantly increased in the pericarp in comparison with the placenta, but at six time points (30, 35, 40, 45, 55 and 60 DAF).
Sucrose synthase (SuSy) activity was initially about twice as high in the pericarp than in the placenta (Fig 6A). The Susy activity was significantly reduced in the placenta in comparison with the pericarp until 40 DAF, but at 60 DAF the activity was significantly increased in the placenta in comparison with the pericarp. The activity in both tissues decreased over time.
Acid-invertase activity was significantly greater in the placenta than in the pericarp at six time points (20, 25, 40, 45, 50 and 55 DAF), but at 60 DAF the activity was significantly increased in the pericarp in comparison with the placenta (Fig 6B). The activity in both tissues increased slightly between 20-55 DAF, before increasing quickly at 60 DAF.
UDP-glucose pyrophosphorylase (UGPase) activity was very high in both tissues (Fig 6C). It increased initially in both tissues until 45 DAF, after which it decreased slightly. The activity was significantly reduced in the placenta in comparison with the pericarp but only at three time points (30, 35, and 50 DAF).
Phosphoglucomutase (PGM) activity was significantly greater in the pericarp at all time points than the placenta. The activity stayed approximately the same until 40-45 DAF, after which it decreased (Fig 6D).
ADP-glucose pyrophosphorylase (AGPase) activity was also significantly greater in the pericarp than in the placenta but only at three time points (20, 35 and 45 DAF). In the placental tissue it decreased from a high activity to a low one between 20 and 30 DAF, after which it stayed at the low activity throughout the rest of development. The activity in the pericarp, on the other hand, decreased over the entire developmental stage (Fig. 6E).
Hexokinase (HK) activity increased in a similar manner in both pericarp and placental tissues until 45 DAF, after which it decreased (Fig. 7A). The activity was significantly lower in the placenta than in the pericarp but only at two time points (30 and 35 DAF).
Fructokinase (FK) activity was initially significantly higher in the pericarp than in the placenta until 40 DAF, but after 50 DAF no differences could be detected. The activity decreased after 45 DAF in both tissues (Fig. 7B).
Phosphoglucose isomerase (PGI) activity increased in both pericarp and placental tissues until 45 DAF, wherafter it decreased (Fig. 7C). The activity was significantly lower in the placenta than in the pericarp at four time points (40, 45, 50 and 60 DAF), but at 25 DAF the activity was significantly increased in the placenta in comparison with the pericarp.
Fructose 1,6 bisphosphate (FBPase) activity was significantly higher in the placenta than in the pericarp at all time points except at 20 DAF. The activity of this enzyme decreased over time (Fig. 7D).
Phosphofructokinase dependent pyruphosphate (PPi-PFK) activity decreased rapidly in both tissues from an initial relatively high activity to a basal activity at 35 DAF. Thereafter the activity remained relatively constant (Fig 7E). The activity was significantly increased in the placenta in comparison with the pericarp at three time points (20, 25 and 30 DAF).
Phosphofructokinase (PFK) activity was initially relatively high in both tissues and increased slightly until 45 DAF, after which it decreased (Fig. 7F). Its activity in the placenta was significantly lower than in the pericarp but only at two time points (30 and 35 DAF)
Triose phosphate isomerase (TPI) activity increased in both tissues until 40-45 DAF, after which it decreased (Fig 7G). The activity in the placenta was significantly lower than in the pericarp but only at three time points (30, 35 and 50 DAF).
Glyceraldehyde 3-phosphate dehydrogenase (G3P DH) activity was significantly greater in the placenta than in the pericarp at all time points except at 20 DAF. In the placenta it increased until 35 DAF, after which it decreased. In the pericarp it decreased from the first time point until 50 DAF after which it increased slightly (Fig. 7H).
Phosphoglycerate kinase (PGK) activity was significantly greater in the pericarp than in the placenta until 50 DAF. In both tissues it rose between 20-45 DAF, after which it decreased (Fig 7I).
Enolase activity was significantly lower in the placenta than in the pericarp at all time points except two time points (50 and 55 DAF). In the pericarp, however, the activity was greater and stayed relatively stable until 40 DAF after which it decreased (Fig. 7J).
Figure 6: Activities of enzymes involved in the conversion of sucrose to starch in the pericarp and placental tissues of fruit of the tomato cultivar Micro-Tom. (A) SuSy. (B) Acid invertase. (C) UDPase. (D) PGM. (E) AGPase. Data represent the mean of five independent measurements + SE.
Figure 7: Activities of some glycolytic and clavin cycle enzymes in pericarp and placental tissues of fruit from the tomato cultivar Micro-Tom during its development.(A) HK. (B) FK. (C) FGI. (D) FBPase. (E) PPi-PFK. (F) PFK. (G) TPI. (H) G3P DH. (I) PGK. (J) Enolase. (K) PK. (L) PEP phosphatase. Data represent the mean of five independent measurements+ SE.
51Pyruvate kinase (PK) activity was significantly greater in the placenta than in the pericarp at all time points except one time point (at 60 DAF). It remained constant until 35 DAF. Between 35-40 DAF the activity increased, before decreasing afterwards (Fig 7K).
Phosphoenolpyruvate phosphatase (PEP phosphatase) activity was also significantly greater in the placenta than in the pericarp at all time points except one time point (at 55 DAF). Generally, the alterations in activities of PEP phosphatase was similar to those of PK (Fig. 7L).
We wanted to examine how the accumulation of mRNA coding for of various plastidial transporters changed during development of the fruit. mRNA coding for the triose phosphate transporter (TPT) accumulated most in young, green, fruit. As the fruit developed the amount of mRNA decreased, but was still present throughout most of the developmental period (Fig. 8A and B).
Figure 8: RNA blot analysis of some plastidial transporters throughout fruit development in the tomato cultivar Micro-Tom. TPT in (A) pericarp and (B) placental tissues. Glc-6-P transporter in (C) pericarp and (D) placental tissues. ATP/ADP transporter in (E) pericarp and (F) placental tissues.
52The accumulation of mRNA coding for the Glc-6-P transporter was greatest between 25-30 DAF. There appeared to be no expression after 40 DAF (Fig. 8C and D). The expression in the placental tissue was greater than in the pericarp.
The ATP/ADP transporter was only expressed at low levels, but was present throughout development (Fig. 8E and F).
This study was initiated to examine carbohydrate metabolism in a new variety of tomato that is beginning to be used as a model system. Much of the data is, therefore, descriptive, but it also leads to some novel conclusions. The first of these is that the metabolism in the pericarp is different to that in the placenta. The placenta accumulates starch over a longer period than the pericarp, and has different concentrations of soluble sugars within it. This is presumably because the different tissues serve different roles. The placental tissue acts as a conduit for nutrients going to the developing seeds, while the pericarp protects the seeds within the fruit. The starch that accumulates over a longer period in the placenta may act as a nutrient reserve in case the flow of sucrose coming to it from the leaves becomes disrupted. The chloroplasts in the pericarp, on the other hand, differentiate to chromoplasts during ripening, and the starch is presumably degraded as a source of soluble sugars to make the fruit more palatable for the dispersal of seeds.
There were also differences in the activities of enzymes measured in the pericarp in comparison with the placental tissue. Although all of the enzyme activities shown were calculated on the basis of the amount of protein in the extract, it is also possible to do so based on the fresh weight. Although they showed minor differences with the data presented, these were not large enough to alter the conclusions. The pericarp consistently had increased activities of several enzymes in comparison with the placenta. These were SuSy, PGM, enolase, PGK and UGPase. With the exception of enolase all of these enzymes are closely associated with the degradation of sucrose. It is interesting to note that the pericarp had a lower concentration of sucrose in comparison with the placenta during development. It may be that the higher activities of these enzymes led to it being metabolised faster to its lower concentration.
The conversion of sucrose to starch has been relatively well studied in tomato fruits from varieties other than Micro-Tom (Robinson et al., 1988; Yelle et al., 1988; Schaffer and Petreikov, 1997a) and it is, thus, possible to compare the data from this study with that from
53those. As in this study, it has often been found that sucrose concentrations are lower than both glucose and fructose (Damon et al., 1988; Klann et al., 1996; Schaffer and Petreikov, 1997a), although the wild tomato relative Lycopersicon chmielewskii accumulates higher levels of sucrose than the other soluble sugars (Yelle et al., 1988). The reason for this accumulation of sucrose in the wild relative is due to a reduction in the activity of acid invertase (Klann et al., 1996). The invertase activity in this study increased dramatically in the final stages of fruit development, especially in the pericarp. This again is similar to what has been found previously in wild-type tomato fruits (Klann et al., 1996) indicating that the Micro-Tom cultivar is not significantly altered in this respect.
SuSy activity is often considered to be a major determinant of sink strength, although the evidence in tomato for this is contradictory. One study found repression of SuSy using genetic engineering techniques let to a decrease in a fruit set (D‘Aoust et al., 1999), whilst a second found no effect (Chengappa et al., 1999). This might be because the first study used a constitutive promoter to reduce SuSy activity, whilst the second used a fruit specific promoter. In this study SuSy activity decreased during development in both the pericarp and placental tissues. Although there is variation between different studies as to what occurs to SuSy activity during fruit development, this type of pattern is not unusual (for example Robinson et al., 1988; Klann et al., 1996) and furthermore parallels the switch from symplastic to apoplastic unloading that occurs during development (Ruan and Patrick, 1995).
AGPase activity has often been correlated with starch accumulation in tomato fruits. In this study that was also the case, with AGPase activity being below detectable levels after 30 DAF in the placenta and decreasing in activity in the pericarp. At this time point there is net degradation of starch, and it might be expected, therefore, that enzymes involved directly in its synthesis would be down -regulated.
Although not so well studied in tomato as enzymes involved in sucrose to starch conversion, we also measured some enzymes involved in glycolysis or the Calvin cycle. Most glycolytic enzymes showed a peak of activity at about 40 DAF. Some of these, specifically HK, FK, PGI, PFK, TPI, G3P DH, and PGK, showed a gradual increase and decrease. Two others though, PK and PEP phosphatase showed a dramatic increase in activity between 35 and 40 DAF, which then declined very quickly to a lower level at 45 DAF. It is interesting that PK showed this sudden increase in activity as it is thought to exert significant control over flux through the glycolytic pathway (Plaxton, 1990). The peaking of activities of glycolytic
54enzymes at 45 DAF is presumably due to climacteric respiration. It may be that the sudden up-regulation of PK just before this point indicates that this enzyme is important in increasing flux through glycolysis to generate ATP for climacteric respiration.
It has been speculated that climacteric respiration in fleshy fruits, such as tomato, only occurs when they are detached from the plant. This idea was based on studying internal CO2 concentrations in both tomato and muskmelon, and not finding expected increases when the fruits were attached to the plant (Saltveit, 1993; Shellie and Saltveit, 1993). It has been argued, however, that the interpretation of these studies was incorrect as; they did not take into account the effect of photosynthesis on internal CO2 concentration (Knee, 1995). Our data are consistent with tomatoes acting as climacteric fruits when attached to the vine as they show an up-regulation of glycolysis just prior to the onset of ripening.
We also decided to study the expression of some transporters that are present in the plastidial membrane. Expression of the TPT has previously been studied in tomato fruits (Schünemann et al., 1996; Büker et al., 1998). In these studies it was shown that both mRNA coding for the TPT protein, and the protein itself, accumulated in both green and red fruits. Our data do not disagree with these findings, showing maximal accumulation of mRNA in green fruits, with less accumulation during development. The expression of the Glc-6-P transporter has not previously been studied in tomato fruits. In maize, mRNA coding for it has been demonstrated to be present only in tissues containing non-green chloroplasts (Kammerer et al., 1998). Our data indicate that in tomato fruits, the mRNA coding for the Glc-6-P transporter is expressed maximally in tissues containing green chloroplasts, with reduced amounts in red fruits. This is opposite to that found in maize, but indicates the difference of fruit of chloroplasts in comparison with those in leaves. The sole source of sugars in leaf chloroplasts comes directly from photosynthesis, while fruit chloroplasts can import Glc-6-P from the cytoplasm also (Büker et al., 1998). This Glc-6-P is the result of the catabolism of imported sucrose in the cytosol. The expression of the Glc-6-P transporter appears to correlate with accumulation of starch in the pericarp and placental tissues. It is tempting to speculate that the Glc-6-P transporter is expressed at times of maximal starch accumulation to supplement carbon being fixed through photosynthesis in the chloroplast. This is especially so as it is much more strongly expressed in the placental tissues, which are in the centre of the fruit and will, therefore, receive less light for photosynthesis. This tissue would have to rely more on sucrose to supply starch synthesis, than on any photosynthate it may be able to produce. This is of course true for all pathways that would utilise Glc-6-P in the plastide, but
55starch constitutes the major sink for carbon in plastids and so would have the greatest influence. The ATP/ADP transporter was weakly expressed throughout development of the fruits. This transporter has been demonstrated to have a great influence on the rate of starch accumulation in potato tubers (Tjaden et al., 1998), but ATP is used in practically, all biosynthetic pathways, so maintaining a constant into the plastids would be expected.
One aim of this study was to examine whether the Micro-Tom tomato cultivar was a suitable candidate to act as a model system for the study of carbohydrate metabolism in tomato fruit generally. It might be that the mutations leading to the dwarf phenotype lead to pleotropic effects on the metabolism of the fruit, which would make it an unsuitable candidate. All the data in this study indicate that the metabolism of the Micro-Tom cultivar is not greatly altered in comparison with reports of other cultivars, showing that suitable for such studies.
From the previous data presented in this investigation, it can be concluded that: (A) The metabolism in the pericarp is different to that in the placenta. (B) Starch was degraded more slowly in the placenta in comparison to the pericarp, while soluble sugars accumulated to a greater extent in the pericarp. (C) There were also differences in the activities in enzymes involved in conversion of sucrose to starch measured in the pericarp in comparison with the placental tissue. (D) The pericarp consistently had increased activities of several enzymes SuSy, PGM, enolase, PGK and UGPase. SuSy, PGM and UGPase in comparison with the placenta. (E) The activities of glycolytic enzymes tended to peak at 40 DAF. (F) Two of these, PEP phosphatase and PK, showed a dramatic increase in activity just before this peak possibly indicating a role in up-regulating glycolysis to generate ATP for climacteric respiration. (G) Both the TPT and Glc-6-P transporter were expressed greatest in green fruits, before declining. (H) The expression of the TPT was greater than that of the Glc-6-P transporter. (I) The ATP/ADP transporter was expressed to a low level throughout fruit development.
© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 2.0||
der Humboldt-Universität zu Berlin
|HTML - Version erstellt am:|
Thu Sep 18 11:25:27 2003