| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
First published online February 9, 2007; 10.1104/pp.106.094771 Plant Physiology 143:1720-1738 (2007) © 2007 American Society of Plant Biologists OPEN ACCESS ARTICLE
Specific Roles of
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
ABSTRACT |
|---|
 TOP ABSTRACT RESULTSDISCUSSIONMATERIALS AND METHODSLITERATURE CITED |
|---|
- and 
-tocopherol are predominant with their ratio being under spatial and temporal control. While -tocopherol accumulates predominantly in photosynthetic tissue, seeds are rich in -tocopherol. To date, little is known about the specific roles of - and -tocopherol in different plant tissues. To study the impact of tocopherol composition and content on stress tolerance, transgenic tobacco (Nicotiana tabacum) plants constitutively silenced for homogentisate phytyltransferase (HPT) and -tocopherol methyltransferase (-TMT) activity were created. Silencing of HPT lead to an up to 98% reduction of total tocopherol accumulation compared to wild type. Knockdown of -TMT resulted in an up to 95% reduction of -tocopherol in leaves of the transgenics, which was almost quantitatively compensated for by an increase in -tocopherol. The response of HPT and -TMT transgenics to salt and sorbitol stress and methyl viologen treatments in comparison to wild type was studied. Each stress condition imposes oxidative stress along with additional challenges like perturbing ion homeostasis, desiccation, or disturbing photochemistry, respectively. Decreased total tocopherol content increased the sensitivity of HPT:RNAi transgenics toward all tested stress conditions, whereas -TMT-silenced plants showed an improved performance when challenged with sorbitol or methyl viologen. However, salt tolerance of -TMT transgenics was strongly decreased. Membrane damage in -TMT transgenic plants was reduced after sorbitol and methyl viologen-mediated stress, as evident by less lipid peroxidation and/or electrolyte leakage. Therefore, our results suggest specific roles for - and -tocopherol in vivo.
; Alscher and Heath, 2002).
Eight structurally related tocochromanols are commonly addressed as vitamin E, namely tocopherols and tocotrienols, which can be distinguished by the saturation of their prenyl moiety and by the methylation degree of their chromonal ring. Subcellular distribution and enzymatic steps of vitamin E biosynthesis have been characterized in detail (for recent review, see DellaPenna and Pogson, 2006) and a number of molecular approaches to manipulate vitamin E synthesis have been published (summarized in Herbers, 2003; DellaPenna and Last, 2006; Dörmann, 2007).
Vitamin E synthesis is restricted to photosynthetic organisms including plants, algae, and some cyanobacteria. The first committed steps in tocopherol and tocotrienol biosynthesis are the prenylation of homogentisic acid by either phytyl diphosphate or geranylgeranyl diphosphate yielding 2-methyl-6-phytyl-benzoquinone or methyl geranylgeranyl benzoquinone, respectively. These reactions are catalyzed by homogentisate phytyltransferase (HPT) and homogentisate geranylgeranyl transferase. The two subsequent steps catalyzed by tocopherol cyclase (TC) and -tocopherol methyltransferase (-TMT) are in common for tocopherol and tocotrienol biosynthesis (see Fig. 1
). With the recent characterization of phytol and phytyl phosphate kinase from Arabidopsis (Arabidopsis thaliana; Ischebeck et al., 2006; Valentin et al., 2006), it could be demonstrated that the prenyl moiety for -tocopherol biosynthesis is derived from free phytol in seeds, substantiating the assumption that phytol gets recycled during chlorophyll breakdown (Peisker et al., 1989; Rise et al., 1989; Dörmann, 2007).
|
). Usually, leaves commonly accumulate -tocopherol, whereas seeds are rich in -tocopherol. 
- and 
-tocopherol are not very abundant in most plant species. Therefore, -tocopherol has been proposed to participate in the detoxification of ROS together with the hydrophilic antioxidants glutathione and ascorbate (Foyer and Noctor, 2003). Based on inhibitor studies it has been shown that tocopherol acts as singlet oxygen scavenger in PSII of Chlamydomonas reinhardtii (Trebst et al., 2002; Kruk et al., 2005). The inhibitory effect could be overcome by addition of membrane-permeable short chain - and -tocopherol derivatives, suggesting that -tocopherol can be substituted by -tocopherol in leaves. A similar conclusion was drawn from the biochemical analysis of the Arabidopsis -TMT-deficient mutant vte4-1, which accumulates comparable amounts of -tocopherol instead of -tocopherol (Bergmüller et al., 2003). Photosynthetic parameters in vte4-1 were found to be indistinguishable from wild type under a variety of stress conditions (Bergmüller et al., 2003). In contrast to leaves, -tocopherol is predominant in seeds, where total tocopherol contents are 10 to 100 times higher than in leaves (Table I
). In oil-storing seeds like Arabidopsis, -tocopherol has been shown to protect polyunsaturated fatty acids (PUFAs) from oxidation, thereby increasing seed longevity (Sattler et al., 2004). This suggests that -tocopherol is involved in desiccation tolerance of seeds.
|
; Boudsocq and Laurière, 2005). The interplay of the foliar antioxidant network and cellular signaling is pivotal for stress sensing. Leaf ascorbate content was shown to modulate the response to ABA (Pastori et al., 2003), while ABA was shown to stimulate antioxidant capacity and transient ROS production in parallel (Zhang et al., 2006, and refs. therein). As proposed by Foyer and Noctor (2005), the balance of the antioxidant and ROS responses finally determines whether the cells respond to water stress by acclimation and hardening or with programmed cell death.
Nevertheless, excess salt and sorbitol are thought to have different impact on cellular metabolism. After its uptake into the symplast, sodium will be sequestered in the vacuole by Na+/H+ antiporters (Fukuda et al., 1999; Gaxiola et al., 1999; Shi et al., 2000) to prevent poisoning of enzyme activities in the other cellular compartments (Serrano et al., 1999). As this process utilizes the pH gradient across the tonoplast, high salinity also impairs cellular pH homeostasis (Serrano et al., 1999). In contrast, sorbitol accumulates as compatible solute in Rosaceae (e.g. apple [Malus domestica]; Li and Li, 2005; Sircelj et al., 2005) and prokaryotes and possesses a lower toxicity than NaCl. Sorbitol accumulation of up to 3 µmol g fresh weight–1 was tolerated in transgenic tobacco (Nicotiana tabacum; Sheveleva et al., 1998). Excess sorbitol more specifically effectuates desiccation of the cellular compartments, inevitably leading to increased ROS production at the photosystem reaction centers when the hydration status in chloroplasts drops. Consequently, increasing ROS quenching capacity by overexpression of superoxide dismutase (Gupta et al., 1993) or glutathione S-transferase and glutathione peroxidase (Roxas et al., 1997) had the potential to alleviate oxidative stress.
In this study, we have constitutively silenced HPT and -TMT posttranscriptionally in tobacco plants with a dsRNAi approach. Assessing the response of HPT and -TMT knockdown lines to salt stress, sorbitol stress, and methyl viologen enabled us to compare the effects of a general tocopherol deficiency (in HPT:RNAi) versus a replacement of -tocopherol with -tocopherol (in -TMT:RNAi) in oxidative stress scenarios. Silencing of both HPT and -TMT resulted in an elevated susceptibility to salt stress. In contrast, the substitution of - for -tocopherol in -TMT transgenics resulted in an elevated osmotolerance when plantlets were subjected to sorbitol. This was characterized by substantially increased membrane integrity compared to the wild type, as determined by ion leakage and lipid peroxidation. The -TMT:RNAi plants exhibited a similar benefit when ROS production in the thylakoids was increased by methyl viologen treatment. Our results provide evidence that -tocopherol is more potent than -tocopherol in mediating osmoprotection in vivo.
|
RESULTS |
|---|
|
TOPABSTRACTRESULTSDISCUSSIONMATERIALS AND METHODSLITERATURE CITED |
|---|
- to -Tocopherol Is Tissue Dependent in Tobacco
To investigate the plasticity of tocopherol biosynthesis in tobacco, we examined the composition of the tocopherol pool in a variety of vegetative and reproductive organs of wild-type Samsun NN (Table I). The tocopherol pool in seedlings, leaves, roots, and stigma consisted more than 90% of -tocopherol, female flower organs contained between 70% and 80% -tocopherol, sepals and petals contained roughly equal amounts of - and -tocopherol, while stamina (80% -tocopherol) were dominated by -tocopherol. Both total tocopherol pool size and the fraction of -tocopherol increased progressively with seed maturation (data not shown). The tocochromanol pool in mature tobacco seeds is dominated by -tocopherol and -tocotrienol with negligible amounts of -tocopherol (e.g. Falk et al., 2003).
In summary, -tocopherol accumulates in vegetative organs of tobacco, while reproductive tissues are enriched in -tocopherol, especially when they are desiccation tolerant like stamina or seeds.
dsRNAi-Mediated Silencing of HPT Results in Tocopherol Deficiency in Tobacco Plants
To manipulate tocopherol metabolism in transgenic tobacco, potato (Solanum tuberosum) cDNA fragments of StHPT and St-TMT (see next paragraph) were introduced into intron-spliced hairpin RNA (RNAi) silencing constructs driven by the constitutive cauliflower mosaic virus 35S promoter (Figs. 2A
and 3A
). Posttranscriptional silencing of tobacco HPT mediated by pBin-StHPT-RNAi (Fig. 2A) resulted in 68 primary transformants in the T0 generation, which were selected for kanamycin resistance. All positive lines were transferred to the greenhouse and screened for tocopherol deficiency by HPLC analysis. Tocopherol content in the transgenic lines ranged from more than 130% to less than 2% of wild type (data not shown), demonstrating efficient silencing of tobacco HPT by the heterologous potato dsRNAi construct. HPT catalyzes the committed step in tocopherol biosynthesis, the condensation of the precursors homogentisic acid and phytyl diphosphate (see Fig. 1). Therefore, silencing of HPT affected all tocopherols to the same extent, leaving the ratio between individual tocopherols constant (data not shown).
|
|
|
-TMT Gene by dsRNAi Results in an Accumulation of - Instead of -Tocopherol in Leaves
Transforming the pBin-St-TMT-RNAi construct (Fig. 3A) for constitutive silencing of -TMT into tobacco yielded 82 kanamycin-resistant T0 transformants, in which the tocopherol composition was surveyed by HPLC. Silencing of -TMT lead to diminished -tocopherol contents in leaves of the transgenics. The T0 transformant St-TMT:RNAi-55 showed the strongest -tocopherol-deficient phenotype with less than 5% of -tocopherol wild-type levels (Fig. 3B). St-TMT:RNAi-82 and -22 were also strongly affected, accumulating 7% to 10% of wild-type -tocopherol, respectively.
The reduction in -tocopherol was mostly paralleled by an increase in -tocopherol in the -TMT transformants (Fig. 3C), so that strong silencing of -TMT correlated with a decrease of the - to -tocopherol ratio in leaves of the transgenics (Fig. 3E). The intermediate lines St-TMT:RNAi-1, -21, and -13 displayed equal contents of - and -tocopherol, while the three lines with the most dramatic change in the ratio of - to -tocopherol (to far below 1), namely St-TMT:RNAi-55, -82, and -22 (see Fig. 3E), were selected for closer study. For comparison, the wild type displayed an - to -tocopherol ratio of about 15 (Fig. 3E).
In the lines St-TMT:RNAi-22, -55, and -82, total tocopherol contents remained rather unchanged in relation to the wild type (Fig. 3D), resulting in a net substitution of - for -tocopherol in leaves. The strongest line St-TMT:RNAi-55 was characterized in more detail along with StHPT:RNAi-28, as described below.
Silencing of -TMT Results in Increased Biomass Production under Osmotic Stress, while Silencing of Both HPT and -TMT Decreases Tolerance to Salt Stress
In contrast to recent reports on the functional characterization of plants with altered tocopherol composition (Porfirova et al., 2002; Bergmüller et al., 2003; Hofius et al., 2004; Havaux et al., 2005; Kanwischer et al., 2005; Maeda et al., 2006), we aimed at examining the oxidative and osmotic stress tolerance instead of focusing on high light and cold stress. Wild-type tobacco and the transgenics were grown on Murashige and Skoog medium supplemented with increasing amounts of 0 to 400 mM sorbitol or sodium chloride in axenic culture; plant growth (this section), metabolite contents, and physiological parameters (other paragraphs) were scored after 4 weeks of either stress.
Biomass production of wild-type plants negatively correlated with increasing sorbitol or NaCl concentrations and was diminished by 90% and 65% on 400 mM sorbitol and 400 mM sodium chloride, respectively (Tables III
and IV
, column 1). Vegetative growth of HPT and -TMT transgenics was not significantly different from wild type on control plates and on plates supplemented with up to 200 mM sorbitol or sodium chloride, except for a 25% growth retardation of HPT knockdown plants on 200 mM NaCl (Tables III and IV; see Figs. 4, A–C
and 5, A–C
for the corresponding phenotypes). Sorbitol and sodium chloride of equal or more than 300 mM provoked distinct phenotypes for HPT:RNAi and -TMT:RNAi tobacco (Figs. 4, A–C and 5, A–C for lines StHPT:RNAi-28 and St-TMT:RNAi-55; see Supplemental Figs. S1 and S2 for the performance of lines StHPT:RNAi-49 and St-TMT:RNAi-82 challenged with sorbitol and NaCl, respectively).
|
|
|
|
As in HPT:RNAi tobacco, silencing of -TMT also aggravated the sensitivity of the transgenics toward salt stress: this effect was less dramatic than for the HPT knockdown plants at 300 mM NaCl, but similar at 400 mM sodium chloride (Table IV; Fig. 5, A and B; Supplemental Fig. S2). Surprisingly, silencing of -TMT led to a substantial growth benefit of more than 2-fold over the wild type in osmotic stress of 300 and 400 mM sorbitol (Table III). Despite a decrease in growth of 75% on 400 mM sorbitol compared to the control plates without osmolyte, -TMT transgenics remained virescent at 400 mM sorbitol, even while wild-type and HPT:RNAi plants exhibited advanced chlorosis (Fig. 4, A–C; Supplemental Fig. S1).
Foliar Tocopherol Pools Exhibit Distinct Responses to Sorbitol and Salt Stress in the Tobacco Transgenics
It is well documented that the amount of tocopherol increases during stress (e.g. Collakova and DellaPenna, 2003; Havaux et al., 2005; Kanwischer et al., 2005; Maeda et al., 2006). Therefore, we examined how the disturbances in tocopherol biosynthesis in HPT:RNAi and -TMT:RNAi affected tocopherol accumulation in the studied stress conditions. To this end, tocopherol contents in stress-acclimated leaves were determined after 4 weeks of osmotic and salt stress.
The wild-type Samsun NN displayed a more than 4-fold increase in total tocopherol from around 70 ng cm2 in controls to about 400 ng cm2 at 400 mM sorbitol or NaCl (Figs. 4D and 5D). The total tocopherol content correlated directly with the concentration of sorbitol and sodium chloride in the medium, with -tocopherol contributing to more than 90% of the tocopherol pool in any condition in the wild type.
The total tocopherol content in unchallenged -TMT:RNAi plants was comparable to wild type, between 60 to 80 to ng cm2 (see Figs. 4D and 5D), except that - substituted -tocopherol in the -TMT transgenics. Likewise, the -tocopherol-deficient -TMT:RNAi transgenic tobacco plants accumulated similar amounts of -tocopherol instead of -tocopherol during osmotic and salt stress (Figs. 4D and 5D), which was up to 350 ng cm2 -tocopherol. Compared to unstressed wild type, this would determine a 70-fold increase in -tocopherol. In conclusion, however, tocopherol biosynthesis got induced to a similar extent in -TMT:RNAi and in wild-type tobacco, resulting in the accumulation of - and -tocopherol, respectively.
In contrast, silencing of HPT abrogated the capacity for tocopherol biosynthesis in the transgenics, so that the applied stress conditions did not result in significant tocopherol abundance. Nevertheless, total tocopherol content was induced from 1.6 (i.e. 2% of unstressed wild type) to 9.8 ng cm2 (i.e. 12% of unstressed and 1% of stressed wild type, respectively) after 4 weeks of either osmotic or salt stress in the HPT transgenics (Figs. 4D and 5D).
Pigment Analysis of Tobacco Transgenics and Wild-Type Plants Subjected to Sorbitol and Salt Stress
Tocopherols are involved in protecting lipids and the photosynthetic apparatus against oxidative damage together with the antioxidants ascorbate, glutathione, and the photosystem-associated xanthophyll cycle (Porfirova et al., 2002; Sattler et al., 2004; Foyer and Noctor, 2005; Havaux et al., 2005; Kanwischer et al., 2005). Therefore, we studied how chlorophyll and carotenoid contents were affected in the transgenic tobacco plants under osmotic and salt stress.
In control and low stress conditions (100 and 200 mM sorbitol or salt) there were no significant differences in chlorophyll and carotenoid content of wild type and both transgenic plants (Fig. 6, A, C, E, and F
). In contrast, chromophore contents in wild type and transgenics were diminished whenever a chlorotic phenotype appeared, paralleling the observations described in Figures 4, A to C and 5, A to C and Supplemental Figures S1 and S2. Compared to the controls, chlorophyll and carotenoid contents were substantially decreased by about 25% each in HPT:RNAi and -TMT:RNAi leaves at sodium chloride concentrations of 300 mM and higher, while the wild type showed a 30% reduction in pigments only on 400 mM NaCl (Fig. 6).
|
-TMT:RNAi plants remained constant irrespective of the sorbitol concentration in the medium (Fig. 6, D and E). Interestingly, the chlorophyll a/b ratio decreased from 2.4 on control plates to approximately 1 for both wild type and HPT:RNAi transgenics on 400 mM sorbitol, due to a strong decline in chlorophyll a, but not b (Fig. 6F). This suggests that the photosystem core and accessory complexes become more damaged under osmotic stress than antenna complexes, which are rich in chlorophyll b.
Altering Tocopherol Composition and Quality by Silencing HPT and -TMT Does Not Lead to Strong Effects on the Ascorbate Pool in Response to Salt and Osmotic Stress
The ascorbate pool size is reportedly increased in response to oxidative stress (Noctor and Foyer, 1998) and there is strong experimental evidence that ascorbate constitutes the interface of the lipophilic tocopherols to the hydrophilic antioxidant network represented by the ascorbate-glutathione cycle (as summarized in Munné-Bosch and Alegre, 2002; Foyer and Noctor, 2005). However, it is unclear whether the ascorbate pool gets induced in the tocopherol-deficient Arabidopsis vte1 mutant, as contrasting results have been obtained (Havaux et al., 2005; Kanwischer et al., 2005).
After 4 weeks of treatment, total ascorbate contents increased in the wild-type Samsun NN with increasing amounts of sorbitol and salt (Fig. 6, B and E). Salt treatment effectuated a more pronounced increase of the ascorbate pool than sorbitol at identical concentrations (compare Figs. 7, B and E ), e.g. 11.7 nmol cm–2 ascorbate on 400 mM NaCl compared to 7.4 nmol cm–2 ascorbate on 400 mM sorbitol.
|
Silencing of -TMT resulted in 30% less accumulation of ascorbate than in the wild type during salt and osmotic stress of equal or more than 200 mM sodium chloride or sorbitol, respectively (Fig. 6). Consistently, -TMT transgenics produced more biomass and exhibited less chlorosis on sorbitol than the wild type (Table IV; Fig. 4; Supplemental Fig. S1). In contrast, knockdown of -TMT resulted in elevated susceptibility toward salt (Table III; Fig. 5; Supplemental Fig. S2), while no substantial effect on the ascorbate pool was evident in salt-stressed -TMT transgenics compared to wild type.
In general, the oxidation state of the ascorbate pool was comparable for the three genotypes in all tested conditions (Fig. 6, B and E).
Lipid Peroxidation and Membrane Damage in Response to Oxidative Stress Are Increased by Silencing HPT and Decreased when -TMT Is Silenced
The observed increase in the pools of the antioxidants ascorbate and especially tocopherol under severe sorbitol and salt stress (Fig. 6) suggests that wild type and transgenics suffer from oxidative stress in both conditions. Malondialdehyde (MDA) occurs as a soluble side product during lipid peroxidation and has proven to be a reliable marker for oxidative stress (Müller-Moulé et al., 2003; Baroli et al., 2004). Excess lipid oxidation will eventually result in membrane damage. The extent of membrane damage can be quantified by ion leakage from the symplast (Rizhsky et al., 2002). Unfortunately, determining ion leakage from salt-stressed plants is prone to artifacts due to the elevated ion concentrations, so that only data for sorbitol stress were recorded.
Four weeks after the onset of salt or sorbitol stress, lipid peroxidation in wild type and both transgenics was correlated in a near-linear manner with increasing salt and sorbitol stress (Fig. 7). Lipid peroxidation in the -TMT transgenics was significantly lowered compared to wild type in either salt or sorbitol of equal or more than 200 mM (Fig. 7). This is in good accordance with the decreased accumulation of ascorbate observed in -TMT:RNAi leaves in both salt and sorbitol stress (Fig. 6, B and E). Together, this points toward a diminished oxidative stress in challenged -TMT:RNAi leaves, even though chlorosis is increased under salt stress compared to wild type (Fig. 5).
However, ion leakage from -TMT:RNAi and wild-type leaves was similar over a broad range of stress conditions (Table V
). On 400 mM sorbitol, ion leakage remained comparable to lower sorbitol concentrations in -TMT knockdowns, while the wild type displayed a 2-fold increase compared to the transgenics (Table V).
|
-TMT transgenics, MDA content was significantly elevated in HPT-silenced plants above 100 mM salt or sorbitol compared to the wild type (Fig. 7). At 400 mM of either stressor, HPT:RNAi tobacco accumulates 1.5-fold more MDA than the wild type (Fig. 7). In addition, ion leakage was already increased by 25% in HPT:RNAi leaves in the absence of sorbitol or sodium chloride supplements on control plates (Table V). Ion leakage in the HPT transgenics reached similar values than in the wild type when 400 mM sorbitol was supplied.
To evaluate whether a substitution of - for -tocopherol in -TMT:RNAi plants improved the resilience of the transgenics toward oxidative stress in a less artificial system, we sprayed leaves of soil-grown plants with different amounts of methyl viologen (Fig. 8
). With the methyl viologen treatments, we aimed to specifically impose oxidative stress on chloroplasts, mimicking the scenario in chloroplasts of sorbitol-stressed leaves. Ion leakage from wild-type leaves correlated with the amount of applied methyl viologen, rising from 20% in the controls to 85% when 50 µM methyl viologen had been sprayed (Fig. 8). With the exception of the untreated control leaves, ion leakage in HPT knockdown lines was similar to the wild type. As in the in vitro experiment, untreated HPT:RNAi leaves displayed elevated ion leakage in the controls relative to the wild type (compare Fig. 8 and Table V).
|
-TMT transgenics exhibited a 35% lower ion leakage than the wild type in the concentration range of 5 to 20 µM methyl viologen (Fig. 8). When 50 µM methyl viologen was applied, the ion leakage from -TMT:RNAi leaves was not significantly different from the other two genotypes. However, membrane damage got already saturated in the wild-type and HPT:RNAi plants after spraying 20 µM methyl viologen (Fig. 8), indicating that 50 µM methyl viologen might be too harsh to be overcome by the antioxidant system.
Taken together, our results indicate that a lack of tocopherols in HPT:RNAi decreases membrane integrity in the absence of stress and exacerbates lipid oxidation in mild stress conditions already, while lipid peroxidation is diminished and membrane integrity is elevated in the -TMT transgenics during oxidative stress caused by either sorbitol or methyl viologen.
Silencing of HPT and -TMT Alters Sugar and Amino Acid Metabolism in Stress Conditions
Severely tocopherol-deficient maize (Zea mays) and potato plants are reportedly compromised in sugar export and accumulate soluble sugars and starch, paralleled by the formation of callose plugs at the plasmodesmata (Provencher et al., 2001; Hofius et al., 2004). The Arabidopsis HPT mutant vte2 displayed this phenomenology only upon chilling (Maeda et al., 2006). Carbon and nitrogen metabolism interdepend on each other (see Fritz et al., 2006 for a very recent report and the refs. cited therein). In an oversimplified view, carbon availability triggers nitrogen metabolism (e.g. Henkes et al., 2001; Matt et al., 2001; Gibon et al., 2004), and vice versa, nitrogen availability modulates carbon flow in the leaf (e.g. Scheible et al., 1997; Geiger et al., 1999). Therefore, we assessed how sugar and amino acid metabolism are affected in the HPT and -TMT-silenced tobacco plants during salt and sorbitol stress.
Salt and sorbitol stress provoked clearly distinct responses in sugar metabolism in the wild type, although 300 mM of both salt and sorbitol provoked an accumulation of starch and soluble sugars (Fig. 9, A–J ). Compared to unsupplemented plates, starch accumulated 3- and 20-fold in wild type on 300 mM sorbitol and salt, respectively, with salt being about 5 times more potent than sorbitol (compare Fig. 9, D–J); less than 300 mM of either stressor did not result in a considerable accumulation of starch. The accumulation of soluble sugars in the wild type was found to correlate directly with salt and sorbitol concentrations. The monosaccharides Glc and Fru accounted for the increase in total soluble sugar content with increasing sorbitol stress (Fig. 9, G–I), while Suc accumulation correlated with increasing salt stress (Fig. 9, A–C). Glc and Fru contents were finally increased about 20-fold on 300 mM sorbitol compared to the controls, while Suc was found to accumulate more than 10-fold on 300 mM sodium chloride.
|
-TMT transgenics displayed comparable steady-state pools of soluble sugars as the wild type, with the remarkable exception that Glc and Fru accumulation were 2-fold lower in sorbitol stress of 200 mM or higher compared to wild type (Fig. 9, G and H).
Abrogating HPT activity in HPT:RNAi abolished the correlation of sugar or starch accumulation with the degree of salt or sorbitol stress. Sugar and starch pools in the HPT transgenics did not respond to salt stress at all (Fig. 9D); soluble sugar contents in HPT:RNAi leaves were already substantially elevated compared to wild type or -TMT:RNAi lines in low NaCl (Fig. 9, A–C), with starch contents in the HPT transgenics being as high as in wild type or -TMT transgenics grown on 300 mM sorbitol (Fig. 9, D and J). In sorbitol stress, up to 200 mM sorbitol, starch, and soluble sugar contents in HPT transgenics remained comparable to untreated wild type on control plates (Fig. 9, G–J). In contrast, HPT:RNAi leaves accumulated 30% more Suc, 75% more Glc, and 100% more Fru than wild type or -TMT on 300 mM sorbitol, while starch contents rose to similar levels as in the two other genotypes. In contrast to maize sxd1, Arabidopsis vte2, and TC-silenced potato plants, we failed to observe a sugar export block in the tocopherol depleted HPT transgenics in any of the tested conditions, which would involve much stronger accumulation of starch and Suc. Attempts to probe plasmodesmatal obstruction in HPT transgenics remained inconclusive (data not shown).
Characteristic changes in the composition of the free amino acid pool coincided with the accumulation of starch or soluble sugars (Supplemental Fig. S3). The compatible solute Pro accumulated 15- and 20-fold on 300 mM sorbitol and NaCl compared to the unsupplemented controls in all genotypes (Fig. 9, K and E, respectively), as did its precursor Glu in salt stress (see Supplemental Fig. S3). Proline accounted for 40% to 50% of the free amino acid pool when 300 mM of either salt or sorbitol were supplied (compare values in Fig. 9, E and F, or Fig. 9, K and L). It is noteworthy that a 9-fold accumulation of Pro could be observed when as little as 100 mM sodium chloride were supplied, while 200 mM sorbitol led to a 4-fold Pro accumulation only (compare Fig. 9, E and K).
Apart from this indicative amino acid, we also observed correlations between the accumulation of Asn, Gln, His, most branched-chain, aromatic amino acids and hexose contents with increasing sorbitol stress (Supplemental Fig. S3). Almost the same subset of amino acids accumulated 4- to 9-fold in the wild type on 400 mM NaCl, while the transgenics did not accumulate these amino acids (Supplemental Fig. S3). In general, the transgenics did not show pronounced differences in free amino acid contents to the wild type in sorbitol stress, with -TMT:RNAi plants being indistinguishable from the wild type (Supplemental Fig. S3).
|
DISCUSSION |
|---|
|
TOPABSTRACTRESULTSDISCUSSIONMATERIALS AND METHODSLITERATURE CITED |
|---|
The rationale behind employing salt and sorbitol stress in our study was to elicit specific stress patterns. Both severe salt and sorbitol stress finally result in oxidative stress caused by hyperosmolarity (as reviewed by Wang et al., 2003). However, salt and sorbitol stress will primarily target different subcellular compartments. Excess symplastic salt concentrations endanger protein integrity and therefore, have to be sequestered in the vacuole or extruded from the cells, concomitantly disturbing pH and ion homeostasis across the tonoplast and the plasma membrane (Serrano and Rodriguez-Navarro, 2001). Increased salt tolerance was achieved by generating transgenic plants overexpressing the plasma membrane Na+/H+ antiporter SOS1 (Shi et al., 2003) or the vacuolar Na+/H+ antiporter AtNHX1 (Apse et al., 1999; Zhang et al., 2001; Zhang and Blumwald, 2001), demonstrating that the sequestration capacity for sodium is limiting in high salinity. Our data on carbohydrate and amino acid metabolism provide evidence that the assumed mode of sodium chloride toxicity also holds true for our experimental setup: in transgenics and the wild type, we observed a 9-fold accumulation of Pro (and its precursor Glu) in 100 mM salt stress (Fig. 9; Supplemental Fig. S3), while Pro biosynthesis, evidenced by steady-state contents of itself and its precursor Glu, was not induced in 100 mM sorbitol stress, indicating that the constraint to protect macromolecules with the accumulation of compatible solutes is more specific for salinity. Furthermore, a high salinity of 400 mM NaCl led to a dramatic accumulation of starch and Suc as well as Asp and pyruvate-derived amino acids in the wild type (Fig. 9; Supplemental Fig. S3), indicating a severely disturbed central cellular metabolism. Likewise, this was not observed in sorbitol stress. The correlation of Suc and starch accumulation with increasing salt stress also argues for a progressing sugar export block from source leaves in response to poisoning by salinity, as commonly observed under various stress conditions (e.g. Voll et al., 2003; Sam et al., 2004; Maeda et al., 2006).
In contrast to sodium chloride, sorbitol is a compatible solute, which does not interfere with enzymatic activities in plasmatic compartments. Sorbitol would more specifically cause desiccation of the cells in addition to hyperosmotic stress. In the second line, reduced water availability would finally increase oxidative stress in the chloroplasts by stimulating the generation of ROS and, concomitantly, lipid peroxy radicals in illuminated thylakoids. Paralleling the increased tolerance toward sorbitol stress, the -TMT transgenics were less susceptible than wild type toward the electron donor methyl viologen that specifically induces oxidative stress in the thylakoids by stimulating the generation of superoxide anions at PSI reaction centers (Fig. 8). Taken together, this indicates that -tocopherol in -TMT:RNAi leaves is capable of mitigating ROS-induced stress in chloroplasts better than -tocopherol in wild-type leaves. Increasing ROS quenching capacity by, for example, overexpression of superoxide dismutase (Gupta et al., 1993) or glutathione S-transferase and glutathione peroxidase (Roxas et al., 1997), has already been shown to possess the potential to alleviate oxidative stress.
In support of our model of sorbitol action, we noted an enhanced hexose accumulation in wild type and -TMT:RNAi under sorbitol stress compared to salt stress (Fig. 9), indicating a rearrangement of carbohydrates in favor of osmotically competent monosaccharides like Glc and Fru. The compatible solute Pro accumulated in elevated sorbitol stress of more than 300 mM only (Fig. 9). As concluded from the metabolic analysis, the specific accumulation of soluble sugars indicates that desiccation predominates over sorbitol toxicity and oxidative stress in sorbitol treatments. An accumulation of hexoses during drought (i.e. desiccation) stress has already been described for tobacco (e.g. Karakas et al., 1997) and was also observed for barley (Hordeum vulgare) leaves (Villadsen et al., 2005) and sweet potato (Ipomoea batatas) suspension cells (Wang et al., 1999) subjected to drought and osmotic stress. Mostly, drought stress was accompanied by a decrease in starch biosynthesis in favor of soluble sugars (Karakas et al., 1997; Villadsen et al., 2005).
Tocopherol Depletion by Silencing HPT Decreases Stress Tolerance in Transgenic Tobacco
HPT catalyzes the committed prenyl transferase step in tocopherol biosynthesis at the branch point of tocopherol and tocotrienol biosynthesis (Fig. 1). By posttranscriptional silencing of HPT activity, we could obtain transgenic tobacco with less than 2% of wild-type foliar tocopherol content. Although tobacco seeds are capable to synthesize tocotrienols (Falk et al., 2003), wild-type tobacco leaves are devoid of tocotrienols (Cahoon et al., 2003; Rippert et al., 2004). Even in severe stress, we could only detect traces of tocochromanols in HPT:RNAi leaves (data not shown), so that it can be ruled out that the depletion of tocopherols in HPT knockdown tobacco is functionally compensated by tocotrienols.
As expected, HPT:RNAi leaves displayed an increased sensitivity toward salt and sorbitol stress, although membrane damage and lipid peroxidation were already elevated in unstressed HPT transgenics (Table V). Sorbitol stress, despite some minor changes in carbon and nitrogen metabolism, rather specifically affected lipid peroxidation in HPT transgenics compared to the two other genotypes (Fig. 7). In contrast, the extent of lipid peroxidation, pigment breakdown, and the changes in carbon and nitrogen metabolism were aggravated in the HPT transgenics compared to the wild type in salt stress (Tables III–V
; Figs. 6, 7, and 9; Supplemental Fig. S3). Aside from a direct impact of elevated lipid peroxidation, the increased susceptibility of HPT:RNAi tobacco toward oxidative stress caused by salt and (to a lesser extent) sorbitol might be facilitated by enhanced ROS signaling in the transgenics early in the acclimation process. Transcriptome analysis of unchallenged Arabidopsis vte2 seedlings revealed that the increased lipid peroxidation in this HPT mutant triggered ROS-regulated genes, as during pathogen challenge or oxidative stress-like ozone (Sattler et al., 2006). Interestingly, the induced genes in vte2 seedlings were not positively regulated by methyl jasmonate (Sattler et al., 2006), the synthesis of which is elevated when lipid peroxidation occurs. In contrast, Arabidopsis TC mutants (vte1) challenged with low temperature and high light exhibited a transient increase in methyl jasmonate (Munné-Bosch et al., 2007). However, the diverging results obtained for vte1 and vte2 could be explained by the different age of the examined plants or by the accumulation of the tocopherol precursor 2,3-dimethyl-5-phytyl-1,4-benzoquinone in vte1 (Sattler et al., 2003), which is absent in vte2.
Despite the evidence that Arabidopsis vte2 mutants exhibit an enhanced ROS response (Sattler et al., 2006), the soluble antioxidants ascorbate and glutathione were not substantially induced in the Arabidopsis tocopherol-deficient vte1 and vte2 mutants (Havaux et al., 2005; Kanwischer et al., 2005). We also found that the ascorbate pool was not significantly increased in tocopherol-deficient HPT knockdown tobacco compared to wild type (Fig. 6). But as tocopherols connect the lipophilic xanthophyll cycle to the soluble antioxidant network governed by the ascorbate-glutathione cycle (Foyer and Noctor, 2003), high light sensitivity of vte1 was greatly enhanced when other antioxidant systems like the zeaxanthin or the glutathione-ascorbate cycle were compromised concomitantly (Havaux et al., 2005; Kanwischer et al., 2005). This demonstrates compensatory capacity in the foliar antioxidant network, which might also account for the comparable response of the ascorbate pool in HPT transgenics and wild type during acclimation to the imposed oxidative and osmotic stress.
The Substitution of - for -Tocopherol in -TMT-Silenced Tobacco Increases Osmotolerance
We achieved an almost quantitative substitution of - for -tocopherol by silencing -TMT in tobacco with only 5% of foliar -tocopherol left compared to wild type. To date, mutants with decreased -TMT activity have been isolated from Arabidopsis (vte4), sunflower (Helianthus annuus), and Synechocystis (slr0089) and all of them accumulate -tocopherol instead of -tocopherol (Shintani and DellaPenna, 1998; Bergmüller et al., 2003; Hass et al., 2006). Interestingly, the total tocopherol content in sunflower and Arabidopsis mutant leaves remained similar to the corresponding wild type (Bergmüller et al., 2003; Hass et al., 2006), while the Synechocystis mutant lacked -tocopherol without any compensation in -tocopherol (Shintani and DellaPenna, 1998; Sakuragi et al., 2006). Among these -TMT mutants, only the Arabidopsis vte4 mutant was tested for abiotic stress tolerance and did not exhibit an altered stress response toward heat, cold, and high light compared to the wild type (Bergmüller et al., 2003). In contrast, we imposed sorbitol, methyl viologen, and sodium chloride to probe whether replacing -tocopherol with -tocopherol affects oxidative stress tolerance in tobacco.
Silencing -TMT in tobacco, thereby exchanging - for -tocopherol, resulted in an elevated susceptibility toward salt, but a diminished susceptibility toward osmotic stress and methyl viologen compared to the wild type (Tables III and IV; Figs. 4, 5, and 8). Ascorbate pool size and lipid peroxidation were reduced in the -TMT transgenics compared to the wild type in salt and sorbitol treatments (Fig. 7), indicating less oxidative stress in the transgenics in both stress conditions. As judged from carotenoid and chlorophyll contents (Fig. 6), the photosynthetic apparatus in -TMT transgenics remained intact in sorbitol, but not in NaCl stress. While the chlorophyll a to b ratio collapsed in wild-type leaves on 400 mM sorbitol, leaves of -TMT:RNAi plants retained the same ratio on 400 mM sorbitol as on control plates.
Our results shed new light on the in vivo function of -tocopherol, allowing two major conclusions. First, -tocopherol is more potent than -tocopherol in conferring desiccation tolerance in vivo. This is presumably mediated by the higher in vivo lipid an
- and 
-Tocopherol in Abiotic Stress Responses of Transgenic Tobacco
