Dehydroascorbate Reductase Affects Leaf Growth, Development, and Function

Abstract

Ascorbic acid (Asc) is a major antioxidant in plants that detoxifies reactive oxygen species (ROS) and maintains photosynthetic function. Expression of dehydroascorbate reductase (DHAR), responsible for regenerating Asc from an oxidized state, regulates the cellular Asc redox state, which in turn affects cell responsiveness and tolerance to environmental ROS. Because of its role in Asc recycling, we examined whether DHAR is important for plant growth. Suppression of DHAR expression resulted in a preferential loss of chlorophyll a, a lower steady state of Rubisco as measured by the amount of the large subunit of Rubisco (RbcL), and a lower rate of CO₂ assimilation. As a consequence, a slower rate of leaf expansion and reduced foliar dry weight were observed. In addition, an accelerated rate of loss of chlorophyll, RbcL, light-harvesting complex II, and photosynthetic functioning was observed in mature leaves, resulting in premature leaf aging. Reduced growth rate as measured by plant height and leaf number was consistent with the DHAR-mediated reduction of photosynthetic function. Increasing DHAR expression maintained higher levels of chlorophyll, RbcL, light-harvesting complex II, and photosynthetic functioning, resulting in delayed leaf aging. The effect of DHAR expression on leaf aging inversely correlated with the level of lipid peroxidation, indicating that DHAR functions to protect against ROS-mediated damage. These observations support the conclusion that through its Asc recycling function, DHAR affects the level of foliar ROS and photosynthetic activity during leaf development and as a consequence, influences the rate of plant growth and leaf aging.

Ascorbic acid (Asc) is a major antioxidant that serves many functions in plants. Asc is involved in the detoxification of reactive oxygen species (ROS), e.g. superoxide, singlet oxygen, ozone, and hydrogen peroxide (H₂O₂), which are produced during aerobic metabolic processes such as photosynthesis or respiration (Asada and Takahashi, 1987). Asc also participates in the regeneration of α-tocopherol (vitamin E) from the tocopheroxyl radical (Asada, 1994). In addition, Asc functions as a cofactor for enzymes such as prolyl and lysyl hydrolases, violaxanthin deepoxidase, and ethylene-forming enzyme (Davies et al., 1991; McGarvey and Christoffersen, 1992; Smith et al., 1992; Eskling et al., 1997) as well as for 2-oxoacid-dependent dioxygenases required for the synthesis of abscisic acid (ABA) and gibberellic acid (Arrigoni and De Tullio, 2000, 2002; Smirnoff, 2000). Asc is involved in the regulation of cell elongation and progression through the cell cycle (for review, see Smirnoff, 1996; Horemans et al., 2000).

Once used, Asc is oxidized to the monodehydroascorbate (MDHA) radical that can be reduced to Asc in the chloroplast or cytosol by MDHA reductase (MDHAR) in an NAD(P)H-dependent reaction (Asada, 1999). In the chloroplast, the MDHA radical can also be reduced to Asc by thylakoid-associated, reduced ferredoxin that is more effective than reduction by MDHAR (Miyake and Asada, 1994; Asada, 1999). MDHA produced in the thylakoid lumen by violaxanthin deepoxidase or following donation of electrons from Asc to PSII or PSI (Mano et al., 1997; Mano, 1999), however, is not available as a substrate for reduction by ferredoxin or MDHAR but rapidly disproportionates to Asc and dehydroascorbate (DHA) when the pH of the lumen is low (Asada, 1999). DHA is then reduced to Asc by DHA reductase (DHAR) in a reaction requiring glutathione. Because the apoplast contains little glutathione or DHAR, DHA, which predominates in the apoplast, must reenter the cell for reduction to Asc. In the absence of sufficient DHAR, however, DHA undergoes irreversible hydrolysis to 2,3-diketogulonic acid.

Given that Asc is present in most cellular compartments and that several pathways exist to ensure that Asc is recycled, it might be expected that perturbations in one recycling pathway would be offset by the activity of the remaining pathways to maintain the cellular Asc redox state. However, changes in DHAR expression result in substantial alterations in the cytosolic and apoplastic Asc redox state; overexpression of DHAR in tobacco (Nicotiana tabacum) leaves increased the Asc redox state (i.e. was more reduced), whereas suppression of DHAR had the opposite effect (Chen et al., 2003; Chen and Gallie, 2004, 2005). These observations suggest that DHAR is expressed in rate-limiting amounts and that DHAR contributes significantly to establishing the cellular Asc redox state, at least in leaves.

H₂O₂ functions as a signaling intermediate downstream of ABA to promote stomatal closure (Price et al., 1994, 2000; Murata et al., 2001; Schroeder et al., 2001a, 2001b; Zhang et al., 2001). Guard cells of plants expressing DHAR exhibit a higher (i.e. more reduced) Asc redox state, reduced levels of H₂O₂, decreased responsiveness to H₂O₂ or ABA signaling, and greater stomatal opening (Chen and Gallie, 2004). Suppression of DHAR expression results in higher levels of H₂O₂ in guard cells and enhanced stomatal closure under normal growth conditions or following water stress. Thus, the level of DHAR expression is important for appropriate signaling in guard cells, because it establishes the efficiency of H₂O₂ scavenging. Increasing DHAR expression also provided enhanced tolerance to environmental ROS, e.g. ozone, despite the increase in stomatal conductance (Chen and Gallie, 2004).

One means by which Asc reduces photoinhibition is by promoting conversion of violaxanthin to zeaxanthin in the xanthophyll cycle to dissipate excess excitation energy as part of nonphotochemical quenching (NPQ). The importance of the Asc pool size in supporting growth was shown with the Arabidopsis (Arabidopsis thaliana) vitamin C1 (vtc1) mutant that is defective in GDP-Man pyrophosphorylase and accumulates only 25% to 30% of the wild-type level without altering the Asc redox state (Conklin et al., 1996, 1999; Veljovic-Jovanovic et al., 2001). In addition to being hypersensitive to ozone, sulfur dioxide, or UVB light, vtc1 plants exhibit slower shoot growth, smaller leaves, and reduced shoot fresh weight and dry weight (Veljovic-Jovanovic et al., 2001). vtc1 plants exhibited no significant difference in the light saturation curves for CO₂ assimilation or chlorophyll fluorescence under these growth conditions, suggesting that the effect on growth was not due to decreased photochemical efficiency or reduced photosynthetic capacity (Veljovic-Jovanovic et al., 2001). No change in the amount of H₂O₂ and only a slight reduction in NPQ were observed in vtc1 leaves (Veljovic-Jovanovic et al., 2001). Expression of cytosolic ascorbate peroxidase (APX) increased and that of chloroplast APX isoforms was either unchanged or slightly decreased, suggesting that the level of Asc is involved in the regulation of the compartmentalization of the antioxidant system in Arabidopsis (Veljovic-Jovanovic et al., 2001). The vtc2 mutant has only 10% to 30% of wild-type levels of Asc and is slightly deficient in the feedback deexcitation component of NPQ that causes the dissipation of excess light as heat (Muller-Moule et al., 2003). The vtc2 mutant experiences chronic photooxidative stress in high light and photobleaches when transferred from low to high light accompanied by increased lipid peroxidation and photoinhibition (Muller-Moule et al., 2003, 2004).

Because changes in DHAR expression result in substantial changes in the Asc redox state not seen in the vtc mutants, we examined whether the efficiency of Asc recycling as determined by DHAR influenced plant growth and leaf function. Plants suppressed in DHAR expression exhibited a slower rate of leaf expansion, slower shoot growth, delayed flowering time, and reduced foliar dry weight. These phenotypes correlated with reduced leaf function as measured by the preferential loss of chlorophyll a, a reduced level of Rubisco large subunit (RbcL), and a lower rate of CO₂ assimilation in mature leaves. The effect of DHAR expression on leaf aging inversely correlated with the level of lipid peroxidation, indicating that the efficiency of Asc recycling was important in regulating ROS-mediated damage. These results suggest that DHAR contributes to plant growth by maintaining photosynthetic functioning through efficient Asc recycling that limits ROS-mediated damage that slows leaf aging.

RESULTS

Chlorophyll Pool Size, CO₂ Assimilation, and Plant Growth Rate Correlate with the Level of Foliar DHAR Activity

To investigate how the level of DHAR activity correlates with leaf function at the whole plant level, DHAR activity and protein levels were measured in every second leaf of mature tobacco plants just prior to flowering. DHAR activity was highest in the youngest leaves and declined with leaf age (Fig. 1A). DHAR protein levels also were highest in the youngest leaves and declined with leaf age, although the decline in DHAR protein was not as great as the decline in DHAR activity, suggesting that DHAR activity may be posttranslationally regulated. The changes in DHAR activity largely correlated with the change in the chlorophyll a and b pool sizes (Fig. 1B) as well as the rate of CO₂ assimilation (Fig. 1C). The only exception to this was that DHAR activity was near maximum in the youngest leaves when the chlorophyll a and b pool size and the rate of CO₂ assimilation were not yet at maximum.

Figure 1.

DHAR activity, chlorophyll pool size, and rate of CO₂ assimilation during leaf development. A, DHAR activity was measured from every second leaf of adult tobacco just prior to flowering. DHAR protein levels were measured by western analysis. B, Chlorophyll a (♦) and b (○) levels and rate of CO₂ assimilation (C) were determined from every second leaf of adult tobacco. Leaf 2 was the youngest leaf and leaf 20 the oldest leaf tested.

In its antioxidant function, Asc acts to maintain photosynthetic function through the detoxification of ROS as well as to maintain electron flow through PSI and PSII (Asada, 1999). Because DHAR contributes substantially to the regulation of the Asc redox state (Chen et al., 2003; Chen and Gallie, 2004), we examined whether alterations in DHAR expression affected components of the photosynthetic machinery or photosynthetic activity. To investigate this possibility, we used DHAR-overexpressing (D_OX) or DHAR-suppressed (D_KD) tobacco that have been described previously (Chen et al., 2003; Chen and Gallie, 2004). D_OX lines were generated following the introduction of a wheat (Triticum aestivum) DHAR (DHAR_Ta) cDNA under the control of the cauliflower mosaic virus (CaMV) 35S promoter. A wheat DHAR cDNA was used because introduction of a tobacco DHAR cDNA under the control of the CaMV 35S promoter resulted only in suppression of endogenous tobacco DHAR expression. Although the nucleotide sequence of the wheat DHAR cDNA has diverged enough that it did not silence endogenous tobacco DHAR expression, at the protein level, wheat DHAR shares 73% identity and 85% similarity with the tobacco ortholog. Therefore, expression of wheat DHAR was used to generate tobacco with increased DHAR activity (i.e. the aggregate of wheat DHAR activity and endogenous tobacco DHAR), whereas expression from the tobacco DHAR transgene was used to generate tobacco with reduced DHAR activity. In each case, it is the level of total DHAR activity present within a line that is correlated with changes in leaf growth and function. The wheat and tobacco DHAR genes used encode a cytosolic DHAR. Of the nine independent D_OX lines initially characterized (three of which have been previously described; Chen et al., 2003) and six D_KD lines (one of which has been previously described; Chen and Gallie, 2004), a reduction in DHAR activity correlated with reduced chlorophyll content and rate of growth, whereas an increase in DHAR activity correlated with higher chlorophyll content (data not shown). Detailed growth measurements were then performed on one representative D_OX and D_KD line.

To examine leaf function and growth at different leaf ages, expanding, mature, and presenescent leaves of D_OX, D_KD, and control plants were used for the analysis. An expanding leaf is defined as one that has achieved approximately 50% of its final size but has not reached a maximum rate of CO₂ assimilation (e.g. leaf 4, Fig. 1C). A mature leaf is defined as a fully expanded leaf that exhibits a maximum rate of CO₂ assimilation (e.g. leaf 6, Fig. 1C). Thus, each leaf type collected from the three lines examined was defined by its developmental stage, in the same position with regard to the apical leaf whorl, and of similar chronological age. Presenescent leaves exhibited reduced rates of CO₂ assimilation relative to the maximum exhibited by a mature leaf (e.g. leaf 18, Fig. 1C) but still contained chlorophyll, albeit at significantly reduced levels. Presenescent leaves were defined according to their position (numbering from the plant base) so that leaves of equivalent chronological age could be collected from all lines to permit a comparison of various parameters of leaf aging, e.g. chlorophyll content, level of Rubisco or light-harvesting chlorophyll-protein complex, lipid oxidation, and leaf dry weight. All measurements were made at the same time each day during the course of an experiment to avoid any possible diurnal effects.

Overexpression of wheat DHAR resulted in a substantial increase in DHAR activity in expanding, mature, and presenescent leaves (Fig. 2A). The level of endogenous DHAR protein (DHAR_Nt) remained unaffected in expanding and mature leaves of D_OX plants as quantitated from the western analysis but was 77% higher in presenescent leaves than in control leaves (Fig. 2B). This is consistent with the prolonged maintenance of leaf function in D_OX plants (see below). In contrast, D_KD tobacco exhibited substantially reduced levels of DHAR activity in leaves at the same developmental stages (Fig. 2A). The level of endogenous DHAR_Nt in expanding, mature, and presenescent leaves of D_KD plants was reduced to 47%, 35%, and 44%, respectively, of the level in the corresponding leaves of control plants (Fig. 2B), data that are in good agreement with the reduction in DHAR activity (Fig. 2A). Although the level of DHAR activity declined with leaf age in D_OX, D_KD, and control plants, the level of activity was always higher in D_OX leaves and lower in D_KD leaves relative to the control (Fig. 2A).

Figure 2.

DHAR activity correlates with ROS-mediated damage and the level of RbcL and LHCII. A, The level of DHAR activity, lipid peroxidation (as measured by the TBARS assay), and total APX was measured in expanding, mature, and presenescent leaves of control (C), D_OX, and D_KD tobacco. The average and sd from three replicates are reported. B, The level of DHAR (DHAR_Nt, DHAR_Ta), RbcL, and LHCII were measured by western analysis. C, RT-PCR analysis of CP1 expression (top) in mature leaves, presenescent leaves, senescent stage 1 leaves (approximately 2 weeks older than presenescent leaves), and senescent stage 2 leaves (approximately 4 weeks older than presenescent leaves) of control, D_OX, and D_KD plants. RT-PCR analysis of actin expression (bottom) was also performed as a control using the same leaf samples.

The level of lipid peroxidation, as measured by thiobarbituric acid reactive substances (TBARS) assay, increased with leaf age; however, it was consistently higher in mature and presenescent leaves of D_KD plants than in the corresponding leaves of D_OX plants (P < 0.05 and P < 0.005; n = 3, respectively) when grown in moderate light (approximately 500 μmol m⁻² s⁻¹; Fig. 2A). Lipid peroxidation in presenescent D_KD leaves was also significantly higher than in the corresponding leaves of control plants (P < 0.05; n = 3), indicating that the extent of lipid peroxidation may be affected by the level of foliar DHAR activity. Although the level of lipid peroxidation was consistently lower in D_OX leaves relative to the control, the difference was not significant. No significant difference in lipid peroxidation was observed in expanding leaves among the three lines, suggesting that the differences in lipid peroxidation accumulated over time. Although total APX activity declined with leaf age in D_OX, D_KD, and control plants, no consistent change in its activity with respect to DHAR activity was observed, in good agreement with previous results (Chen et al., 2003). The level of Rubisco, as measured by the amount of RbcL and the level of light-harvesting complex II (LHCII) declined with leaf age, but their relative abundance in expanding or mature leaves of D_OX, D_KD, and control plants was not substantially different. In presenescent leaves, however, the level of RbcL and LHCII in D_KD plants was reduced to 64% and 38%, respectively, of the level present in the corresponding leaves of control plants, and the level of LHCII in D_OX plants was 29% higher than in the control (Fig. 2B). The reduction in the amount of LHCII in D_KD plants was not observed in vtc2 mutant plants (Muller-Moule et al., 2004). The correlation between DHAR activity and the abundance of RbcL and LHCII suggests that the level of DHAR activity may be important in maintaining these proteins during leaf aging.

To examine whether changes in the level of DHAR activity influenced the induction of a senescence-related gene, expression of tobacco CP1, the ortholog of the senescence-associated gene SAG12 of Arabidopsis, was examined in D_OX, D_KD, and control plants using reverse transcription (RT)-PCR. Little to no CP1 expression was detected in mature or presenescent leaves from D_OX, D_KD, and control plants (Fig. 2C), supporting the conclusion that the presenescent leaves used in this study had not initiated senescence. CP1 expression was detected in senescent stage 1 D_KD leaves, which were approximately 2 weeks older than the presenescent leaves used in this study (Fig. 2C). No CP1 expression was detected, however, in senescent stage 1 leaves of D_OX or control plants. CP1 expression was also detected in senescent stage 2 D_KD leaves (approximately 4 weeks older than the presenescent leaves used in this study), whereas expression was just detectable in senescent stage 2 D_OX or control leaves (Fig. 2C). These data suggest that a reduction in the level of DHAR activity accelerates the onset of senescence.

To determine whether the level of DHAR activity may affect the pool size of chlorophyll a and b, chlorophyll levels were measured in expanding, mature, or presenescent leaves of D_OX, D_KD, and control plants. The level of chlorophyll a was significantly higher in expanding leaves of D_OX plants than in control plants (P < 0.005; n = 4) as it was in mature or presenescent leaves (P < 0.05 and P < 0.005; n = 4, respectively; Fig. 3), whereas it was significantly lower in D_KD mature and presenescent leaves (P < 0.005 and P < 0.005; n = 4, respectively) but not in expanding D_KD leaves (P = 0.085; n = 4) relative to the control. However, the level of chlorophyll a was significantly lower in expanding D_KD leaves relative to expanding D_OX leaves (P < 0.005; n = 4). Although a small increase in the level of chlorophyll b was observed in expanding D_OX leaves relative to the control, this difference was only significant in presenescent leaves (P < 0.005; n = 4; Fig. 3). No significant difference in the level of chlorophyll b was observed in expanding D_KD leaves relative to the control (P = 0.797; n = 4), but its level was significantly reduced in mature or presenescent leaves of D_KD plants relative to the control (P < 0.05 and P < 0.05; n = 4, respectively). The chlorophyll a to b ratio was significantly lower in expanding, mature, and presenescent leaves of D_KD plants than in D_OX leaves (P < 0.05, P < 0.05, and P < 0.005; n = 4, respectively). The chlorophyll a to b ratio was also significantly lower in presenescent D_KD leaves than in control leaves (P < 0.01; n = 4; Fig. 3).

Figure 3.

DHAR activity correlates with the chlorophyll pool size and chlorophyll a/b ratio. Chlorophyll a, chlorophyll b, and the chlorophyll a/b ratio were measured in expanding, mature, and presenescent leaves of control (C), D_OX, and D_KD tobacco. The average and sd from four replicates are reported.

The chlorophyll pool size is determined by rates of its synthesis and degradation. To investigate whether the level of DHAR activity influences the rate of chlorophyll loss, young leaves from D_OX, D_KD, and control plants were dark treated, and the rate of chlorophyll loss was followed over time. Although the chlorophyll pool size in D_OX leaves was larger than in control leaves, the rate of chlorophyll loss from D_OX leaves (i.e. loss of 38.1 μg/g fresh weight/d) was only slightly lower than that from control leaves (i.e. loss of 42.4 μg/g fresh weight/d; Fig. 4). The rate of chlorophyll loss from D_KD leaves was more than 2-fold greater (i.e. loss of 87.1 μg/g fresh weight/d) than that from control leaves despite the use of D_KD leaves with an initial chlorophyll pool size similar to control leaves.

Figure 4.

Suppression of DHAR expression accelerates loss of chlorophyll. Chlorophyll (a + b) was measured every 4 d from dark-treated leaves of control (C, ♦), D_OX (▪), and D_KD tobacco (▴). The average and sd from three replicates are reported.

To examine whether the observed differences among D_OX, D_KD, and control leaves correlated with differences in CO₂ assimilation, gas exchange was measured from every second leaf. The rate of CO₂ assimilation in expanding and mature leaves of D_OX plants was similar to those of control plants despite a higher rate of transpiration (Fig. 5A). The rate of stomatal conductance was too high to measure differences in these same leaves between D_OX and control plants. In older, fully expanded leaves as well as in presenescent leaves, a higher rate of CO₂ assimilation was observed in D_OX plants. Although this correlated with a higher rate of transpiration, a lower substomatal CO₂ concentration was also observed, suggesting a higher rate of CO₂ assimilation (Fig. 5A). The rate of CO₂ assimilation was consistently lower in all leaves of D_KD plants relative to the control (Fig. 5A). In mature and presenescent leaves, this correlated with reduced rates of transpiration and stomatal conductance previously reported for D_KD plants (Chen and Gallie, 2004), but the higher level of substomatal CO₂ concentration in these D_KD leaves indicated that the internal concentration of CO₂ was not limiting and thus suggested that D_KD leaves were less efficient in assimilating CO₂. In young D_KD leaves, the rate of transpiration was similar to that of control leaves, and therefore the observed reduced rate of CO₂ assimilation in the young D_KD leaves supported the conclusion of less efficient CO₂ assimilation. These data suggest that the reduced rate of CO₂ assimilation in young D_KD leaves is likely not a consequence of changes in the level of RbcL or LHCII but do correlate with changes in the chlorophyll pool size and chlorophyll a/b ratio. The reduced rate of CO₂ assimilation in older D_KD leaves and higher rate of CO₂ assimilation in older D_OX leaves correlate with similar changes in the levels of RbcL, LHCII, and chlorophyll pool size as well as the chlorophyll a/b ratio (Figs. 2 and 3).

Figure 5.

DHAR activity affects the efficiency of CO₂ assimilation and leaf dry weight. A, The rates of CO₂ assimilation, transpiration, and stomatal conductance, and the substomatal CO₂ concentration were measured from every second leaf of control (C, ♦), D_OX (▪), and D_KD tobacco (▴). Leaf 2 was the youngest leaf and leaf 18 the oldest leaf tested. Expanding leaves include leaves 2 to 4, mature leaves include leaves 6 to 8, and presenescent leaves include leaves 16 to 18. B, The dry weight as measured as percent of fresh weight was determined for expanding, mature, and presenescent leaves. The average and sd from three replicates are reported.

The reduced rate of CO₂ assimilation in D_KD leaves correlated with a reduced leaf dry weight of expanding and mature leaves relative to the control (P < 0.005 and P < 0.01; n = 3, respectively), although the difference was not significant in presenescent leaves (P = 0.11; n = 3; Fig. 5B). Leaf dry weight in expanding, mature, and even presenescent D_KD leaves was significantly lower than D_OX leaves (P < 0.01, P < 0.05, and P < 0.05; n = 3, respectively; Fig. 5B). No significant difference in leaf dry weight was observed between expanding, mature, or presenescent leaves of D_OX and control plants (P = 0.19, P = 0.83, and P = 0.08; n = 3, respectively; Fig. 5B), correlating with their similar rates of CO₂ assimilation during leaf growth (Fig. 5A). To examine whether the differences in CO₂ assimilation observed in D_OX and D_KD plants are accompanied by other changes in growth, parameters of plant growth were measured. Plant growth as measured by height over time was little altered in D_OX plants relative to the control, and no change in leaf number was observed (Fig. 6). D_OX plants flowered at the same time as control plants (Table I). D_KD plants exhibited a significantly reduced rate of growth (P < 0.005 at weeks 5–7; n = 8) but caught up during late growth (P = 0.130 and P = 0.161 at weeks 8 and 9, respectively; n = 8). D_KD plants had significantly fewer leaves during growth (P = 0.05, P < 0.005, and P < 0.05 at weeks 5, 6, and 7, respectively; n = 8) but eventually produced the same number of leaves as the control (Fig. 6). Consistent with this, D_KD plants flowered significantly later (Table I) than control plants (P = 0.001; n = 8). The average leaf internode distance of fully grown D_KD plants was slightly less than that of control plants, although the difference was not significant (P = 0.9; n = 18). Although no significant difference in leaf area was observed between D_OX and control plants, leaves from D_KD plants were consistently smaller in size relative to the control (P = 0.0014; Fig. 6). These data indicate that the growth rate but not final plant height is reduced in D_KD tobacco, suggesting that efficiency of Asc recycling is important in maintaining maximum growth. Because increasing DHAR expression did not increase plant growth, the endogenous level of DHAR activity appears to be sufficient to support a maximum growth rate.

Figure 6.

The level of DHAR activity affects the rate of plant growth. Plant height and leaf number were measured in control (C, ♦), D_OX (▪), and D_KD tobacco (▴) beginning at 4 weeks after germination until the appearance of the inflorescence. Leaf area was measured following the vegetative to floral transition when all leaves had reached their final size. Leaf 1 was the youngest leaf and leaf 9 was the oldest leaf measured. The average and sd from eight replicates are reported.

Table I.

DHAR is required for correct internode length, leaf growth, and flowering time

	Internode Length	Leaf Expansiona	Flowering Time
	cm	cm2/d	d
Control	9.32 ± 0.68	36.5	60.25 ± 1.09
DOX	9.42 ± 0.70	36.6	60.13 ± 1.27
DKD	8.89 ± 0.74	29.7	64.38 ± 2.18

^aExpansion from the sixth true leaf was measured.

The Level of DHAR Activity Correlates with the Maintenance of Chlorophyll a and Leaf Growth Rate

The previous analyses provided insight into how DHAR activity affects growth at the level of the whole plant. However, it is not valid to assume that expanding, mature, and presenescent leaves from a single plant are representative of the developmental stages that a specific leaf undergoes during its development, as not all leaves on a given plant are biochemically or functionally equivalent (e.g. leaf size, chlorophyll content). Therefore, to investigate how DHAR activity affects growth of a specific leaf, we followed the growth of the sixth true leaf (as numbered from the plant base) in D_OX, D_KD, and control plants. We first measured the area of the sixth true leaf when the leaf was fully expanded, a value that remained constant for a given line when the plants were grown under identical conditions. From the area of the sixth true leaf in its fully expanded state, sixth true leaves could be analyzed during their growth when they were 8%, 50%, or 100% of the fully expanded size. DHAR activity and protein level on a fresh weight basis declined during leaf expansion in control plants as it did in D_OX and D_KD plants (Fig. 7, A and B). Correlating with the decline in DHAR was a decrease in the Asc pool size and redox state (i.e. more oxidized) during the expansion of control leaves (Fig. 7A). The decrease in protein and Asc pool size may be explained in part by the dilution during the cell expansion that occurs during leaf growth. In contrast, dilution does not account for the oxidation of the Asc redox state. As reported previously (Chen and Gallie, 2004), increasing DHAR expression in D_OX plants results in an increase in the Asc pool size and redox state, both of which decline during leaf expansion although always remaining higher than control leaves (Fig. 7A). Suppressing DHAR expression in D_KD plants results in higher levels of DHA (Chen and Gallie, 2004), consistent with reduced Asc recycling, although the total ascorbate pool size (i.e. Asc + DHA) shows slight to no reduction depending on leaf age (Fig. 7A). As with control and D_OX leaves, the Asc pool size and redox state in D_KD plants declined with leaf expansion and were consistently lower than in control leaves. The relative level of RbcL was not substantially different between D_OX and control plants during leaf expansion. The relative level of RbcL was also not substantially different between D_KD and control plants when leaves were 8% of their fully expanded size, but when D_KD leaves reached 50% or 100% of their fully expanded size, the level of RbcL was reduced to 87% and 64% of that present in control leaves (Fig. 7B). No consistent change in the relative level of LHCII was observed in D_OX, D_KD, and control leaves during their expansion (Fig. 7B).

Figure 7.

The level of DHAR activity correlates with the foliar Asc redox state and level of RbcL and LHCII during leaf expansion. A, The level of DHAR activity, Asc (black bars), DHA (white bars), and the Asc redox state (Asc/DHA) were measured in the sixth true leaf during its growth from 8%, 50%, and 100% of fully expanded size of control (C), D_OX, and D_KD tobacco. The average and sd from three replicates are reported. B, The level of DHAR, RbcL, and LHCII were measured in the same leaves as in A by western analysis.

The pool size of chlorophyll a and b in control leaves declined concomitantly during leaf expansion such that the chlorophyll a/b ratio was largely unaltered (Fig. 8A). Although the pool size of chlorophyll a in D_KD leaves was not significantly different from control leaves when the leaves were 8% of their fully expanded state (P = 0.437; n = 4), it was significantly lower than the control when the leaves reached 50% or 100% of the fully expanded state (P < 0.05 and P < 0.05; n = 4; Fig. 8A). In contrast, the pool size of chlorophyll a in D_OX leaves was significantly higher than in control leaves at 8%, 50%, and 100% of the fully expanded state (P < 0.05, P < 0.05, and P < 0.01, respectively; n = 4), although the pool size of chlorophyll b was not significantly different from the control (Fig. 8A). Thus, although the chlorophyll a/b ratio in D_OX and D_KD leaves was not significantly different when the leaves were 8% of their fully expanded state (P = 0.189; n = 4), it was significantly lower in D_KD leaves when the leaves reached 50% or 100% of the fully expanded state (P < 0.05 and P < 0.01; n = 4; Fig. 8A). These data suggest that increasing DHAR activity increases the pool size of chlorophyll a from early leaf development up to senescence, whereas decreasing DHAR activity does not affect the initial pool size of chlorophyll a but does result in a preferential loss of chlorophyll a during subsequent leaf expansion and aging.

Figure 8.

The level of DHAR activity correlates with the chlorophyll a pool size, the chlorophyll a/b ratio, and leaf dry weight during leaf expansion. A, Chlorophyll a, chlorophyll b, and the chlorophyll a/b ratio were measured in the sixth true leaf during its growth from 8%, 50%, and 100% of fully expanded size of control (C), D_OX, and D_KD tobacco. B, The dry weight as measured as the percent of fresh weight was determined at the same stages of leaf growth as in A. The average and sd from four replicates are reported.

To examine how alterations in the efficiency of Asc recycling resulting from changes in DHAR expression affected CO₂ assimilation as a leaf ages, gas exchange was measured in expanding leaves every 3 d. The rate of CO₂ assimilation in D_OX leaves was similar to control leaves during leaf expansion but declined at a slower rate as the leaf aged, correlating with a higher stomatal conductance (Fig. 9). The rate of CO₂ assimilation was consistently lower in D_KD leaves throughout its expansion and subsequent aging. The reduced level of CO₂ assimilation in D_KD leaves could not be explained by a corresponding reduction in stomatal conductance but did correlate with a higher level of substomatal CO₂ concentration (Fig. 9), suggesting less efficient assimilation of CO₂.

Figure 9.

The level of DHAR activity affects the efficiency of CO₂ assimilation during leaf expansion. The rates of CO₂ assimilation, transpiration, and stomatal conductance, and the substomatal CO₂ concentration were measured in the sixth true leaf during its growth from control (C, ♦), D_OX (▪), and D_KD tobacco (▴). Measurements were taken every third day, starting when the leaves were 8% of fully expanded size until the leaves were 100% of fully expanded size (approximately day 9) and continued for an additional 9 d. The average and sd from three replicates are reported.

The rate of leaf expansion was followed for the same cohort of leaves and revealed that D_OX and control leaves expanded at the same rate (P = 0.91; n = 8), whereas the expansion of D_KD leaves was significantly delayed relative to the control (P = 0.028; n = 8; Fig. 10; Table I). The reduced rate of CO₂ assimilation and slower rate of expansion of D_KD leaves correlated with a reduced leaf dry weight throughout leaf expansion (Fig. 8B) that was similar to the reduced leaf dry weight observed for all leaves of adult plants (Fig. 5B). No significant difference in leaf dry weight during leaf expansion was observed between D_OX and control leaves (Fig. 8B), correlating with their similar rates of CO₂ assimilation during early leaf growth (Fig. 9).

Figure 10.

The level of DHAR activity affects the rate of leaf growth. Leaf area was measured for the sixth true leaf during its growth in control (C, ♦), D_OX (▪), and D_KD tobacco (▴) every 3 d beginning at leaf emergence. The average and sd from eight replicates are reported.

DISCUSSION

In this study, we present evidence suggesting that the level of foliar DHAR activity influences the rate of leaf aging, and, as a consequence, the rate of plant growth. We were able to take advantage of the fact that DHAR is expressed in limiting amounts with regard to the Asc redox state (Chen et al., 2003; Chen and Gallie, 2004) to test whether alterations in DHAR expression would perturb plant growth. Suppressing DHAR expression resulted in less efficient Asc recycling and as a consequence, a lower (i.e. more oxidized) Asc redox state. The effect that suppressing DHAR expression had on Asc recycling increased with leaf age and correlated with an increase in ROS (Chen and Gallie, 2004), lipid peroxidation, and slower leaf and plant growth. This reduced growth rate correlated with a smaller chlorophyll pool size and a lower chlorophyll a/b ratio resulting from the preferential loss of chlorophyll a during leaf expansion. The lower chlorophyll a/b ratio correlated with a reduced rate of CO₂ assimilation in all leaves. Although reducing DHAR expression has been shown to reduce stomatal conductance, which might be expected to limit CO₂ diffusion into the leaf interior, a reduced rate of CO₂ assimilation was also observed in expanding and newly expanded leaves in which stomatal conductance was not substantially affected. Moreover, despite the lower stomatal conductance in older D_KD leaves, the substomatal CO₂ concentration was higher than it was in control leaves, indicating that CO₂ diffusion into the leaf interior was not limited but rather CO₂ was not being used efficiently. The preferential loss of chlorophyll a in D_KD leaves may have been a consequence of the higher level of ROS present and may account for the reduced photosynthetic activity observed in D_KD leaves. In addition to the lower chlorophyll a/b ratio in older D_KD leaves, the reduced rate of CO₂ assimilation correlated with reduced levels of RbcL and LHCII. Although it is not known which of these may be rate limiting in older leaves, these data suggest that the efficiency of Asc recycling as determined by the level of DHAR activity plays an important role in leaf aging. The reduced rate of leaf expansion observed in D_KD plants is likely a consequence of the accelerated loss of photosynthetic activity in mature leaves that may have prematurely reduced their photosynthate that they could provide to sink tissues. This would also explain the slower plant growth rate as measured by plant height and rate of leaf production over time. Changes in the level of DHAR expression affect the Asc pool size and redox state within the cytosol and apoplast (Chen and Gallie, 2005), and as Asc is readily transported throughout the cell (Horemans et al., 2000), changes in chloroplastic Asc would also be expected. Whether changes in chloroplastic or cytosolic Asc are responsible for the observed effects on leaf aging remain to be determined.

The observation that increasing DHAR expression correlated with higher levels of RbcL and chlorophyll and a higher rate of CO₂ assimilation in presenescent leaves, whereas reducing DHAR expression in the same leaves correlated with a lower level of RbcL and LHCII, a smaller pool size of chlorophyll, and reduced rate of CO₂ assimilation indicates that the level of foliar DHAR activity can affect the rate of loss of leaf function. The progressive loss of DHAR as a function of leaf age, correlating with a loss of chlorophyll and CO₂ assimilation rate, is consistent with a potential role for this enzyme in influencing the rate of leaf aging. The fact that reducing DHAR expression correlated with slower growth is also consistent with a role of DHAR in influencing leaf aging. The observation that increasing the level of DHAR activity correlated with reduced lipid peroxidation whereas reducing DHAR activity correlated with increased lipid peroxidation particularly during leaf aging is consistent with previous observations demonstrating an inverse relationship between the level of DHAR expression and the foliar level of ROS (Chen and Gallie, 2004). It also suggests that the level of DHAR activity may affect leaf function and aging by regulating the level of ROS and, as a consequence, affect the amount of ROS-generated damage that occurs to the photosynthetic machinery. Reducing DHAR expression in D_KD leaves also resulted in an accelerated onset of senescence as revealed by the premature induction of expression of the SAG12 ortholog, CP1. These observations support the conclusion that through its Asc recycling function, the level of DHAR activity affects the basal level of ROS during leaf development and, as a consequence, influences the rate of leaf aging. In addition, the observation that increasing DHAR expression did not substantially increase leaf expansion or plant growth indicates that the endogenous level of DHAR expression provides a level of Asc recycling sufficient to support maximum growth. This is in contrast to the role of DHAR in regulating guard cell function, where increasing DHAR expression results in substantial changes in stomatal behavior as a result of changes in ROS levels in the guard cells (Chen and Gallie, 2004).

Because no change in the photosynthetic capacity or in H₂O₂ levels was observed in the Arabidopsis vtc1 mutant, it was suggested that the low Asc pool size did not impair photosynthetic functioning or result in oxidative stress (Veljovic-Jovanovic et al., 2001). The observation that vtc1 plants continued to exhibit a slower shoot growth rate when grown under conditions of high CO₂ that suppress oxidative stress resulting from photorespiration supported the conclusion that the slow growth phenotype was unrelated to a higher oxidative load (Veljovic-Jovanovic et al., 2001). vtc2 mutants that contain only 10% to 30% of the wild-type level of Asc exhibit sensitivity to high light and photobleach when transferred from low to high light (Muller-Moule et al., 2002, 2003). When grown in high light, vtc2 plants contain about 40% of the wild-type level of Asc and are able to grow despite signs of oxidative stress, such as lower electron transport, a lower rate of oxygen evolution, and lower PSII quantum efficiency (Muller-Moule et al., 2004). These mutants demonstrate that although some degree of photooxidative stress may result from the reduced levels of Asc, a significant reduction in Asc does not prevent growth under conditions of high light.

In contrast to vtc mutants, D_KD plants showed reduced photosynthetic function and reduced growth in moderate light. Plants suppressed for DHAR are not substantially Asc deficient but rather have a decreased Asc redox state (i.e. more oxidized) that may explain the differences observed between the present plants and the vtc mutants. However, the fact that perturbations to the Asc pool through changes in Asc biosynthesis (vtc mutants) or recycling (DHAR-silenced lines) can affect photosynthetic function under certain growth conditions suggests that the Asc pool is important to leaf function. The correlations between changes in DHAR activity and changes in the foliar level of H₂O₂ (Chen and Gallie, 2004), rate of CO₂ assimilation, and the level of RbcL, LHCII, and chlorophyll are consistent with the conclusion that the slow growth phenotype exhibited by plants suppressed for DHAR is related to reduced photosynthetic functioning and increased oxidative load. The reduced photosynthetic function in D_KD leaves may lead to an earlier onset of carbon starvation that can induce autophagy (Moriyasu and Ohsumi, 1996; Brouquisse et al., 1998). The higher rate of chlorophyll loss in carbon-starved (i.e. dark-treated) D_KD leaves relative to control leaves (Fig. 4) is consistent with this possibility. The higher level of ROS (such as H₂O₂) present in D_KD leaves (Chen and Gallie, 2004) may also promote leaf aging and senescence through increased signaling that induces expression of senescence-related genes known to be enhanced in leaves experiencing a higher oxidative load (Navabpour et al., 2003). It is also possible that a lower (i.e. more oxidized) Asc redox state in plants suppressed for DHAR contributes to slower rate of cell division and/or elongation through mechanisms that are separate from a lower fixed carbon supply resulting from a lower rate of photosynthesis. Asc promotes G1 to S progression of cells within the onion (Allium cepa) root meristem and pericycle (for review, see Smirnoff, 1996) and reversed the inhibition of cell division caused by lycorine treatment, which reduces Asc content (Arrigoni, 1994). However, any such effect would be limited to growing tissues and would not account for the observed reduction in chlorophyll, RbcL, or CO₂ assimilation in expanding D_KD leaves or the accelerated loss of chlorophyll, RbcL, LHCII, or CO₂ assimilation from mature D_KD leaves. Moreover, the prolonged maintenance of leaf function in plants overexpressing DHAR supports a role for this enzyme in leaf aging. Thus, these observations support the conclusion that an important function of DHAR in leaves is to maintain photosynthetic function by limiting ROS-mediated damage.

MATERIALS AND METHODS

Plant Transformation and Growth Conditions

Full-length wheat (Triticum aestivum) and tobacco (Nicotiana tabacum) DHAR cDNAs (accession nos. AY074784 and AY074787, respectively) were isolated as described previously (Chen et al., 2003). Transgenic tobacco (cv Xanthi) expressing the His-tagged wheat DHAR from the CaMV 35S promoter (in the binary vector, pBI101) was generated using Agrobacterium tumefaciens as described (Chen et al., 2003). Transgenic tobacco plants suppressed for DHAR were identified following the introduction of a tobacco DHAR construct in pBI101.

All plants were grown in commercial soil in a glasshouse supplied with charcoal-filtered air. Experiments were carried out and repeated from fall to spring to avoid excessive heat or light. Plants were watered to saturation twice/day (7 am and 1 pm) to ensure even soil moisture. Plants were grown under natural light conditions in a 10-h light and 14-h dark cycle. The average temperature during the day was 25.9°C ± 0.6°C and 20.2°C ± 0.5°C during the night. The average light intensity in the morning (9 am) was 514 ± 206 μmol m⁻² s⁻¹ and in the afternoon (1 pm) was 1,191 ± 244 μmol m⁻² s⁻¹.

To determine leaf area during leaf expansion, digital images were taken at 3-d intervals. Leaf area was calculated using Adobe Photoshop (version 6.0) by normalizing total leaf size to an internal standard included in each image. Eight leaves at each developmental stage were measured, and the data were processed using Microsoft Excel. Leaf area of whole plants was determined using the same approach by measuring the area of every other leaf. For leaf expansion studies, we used the sixth true leaf as numbered from the plant base. The area of the sixth true leaf of D_OX, D_KD, and control plants measured when the leaf was fully expanded remained constant for a given line when the plants were grown under identical conditions. Knowing the area of the sixth true leaf in its fully expanded state, corresponding leaves were collected from subsequently grown plants when they were 8%, 50%, or 100% of the size of the fully expanded state.

Western Analysis

Anti-DHAR antiserum raised against DHAR purified from wheat seedlings was used for western analysis. Protein extracts were resolved using standard SDS-PAGE and the protein transferred to 0.22 μm polyvinylidene difluoride membrane by electroblotting. Following transfer, the nitrocellulose membranes were blocked in 5% milk in TPBS (0.1% Tween 20, 137 mm NaCl, 2.7 mm KCl, 10 mm Na₂HPO₄, 1.4 mm KH₂PO₄, pH 7.4) followed by incubation with primary antibodies diluted typically 1:1,000 to 1:2,000 in TPBS with 1% milk for 1.5 h. The blots were then washed twice with TPBS and incubated with goat anti-rabbit horseradish peroxidase-conjugated antibodies (Southern Biotechnology Associates) diluted to 1:5,000 to 1:10,000 for 1 h. The blots were washed twice with TPBS and the signal detected typically between 1 to 15 min using chemiluminescence (Amersham). Each western was repeated three to four times and representative results presented. The results from the western analysis were quantitated using MicroComputer Imaging Device Elite image processing software (version 7.0, Imaging Research).

Enzyme Assays

DHAR activity was assayed essentially as described (Hossain and Asada, 1984). Tobacco leaves were ground in extraction buffer (50 mm Tris-HCl, pH 7.4, 100 mm NaCl, 2 mm EDTA, 1 mm MgCl₂) and soluble protein obtained following a 5-min centrifugation at 13,000 rpm. DHAR was assayed from an equal amount of protein as described (Bradford, 1976) in 50 mm K₂HPO₄/KH₂PO₄ pH 6.5, 0.5 mm DHA, and 1 mm reduced glutathione and its activity followed by an increase in A₂₆₅. APX activity was determined as described (de Pinto et al., 2000). Each experiment was repeated two to three times and representative results presented.

Asc and DHA Measurements

Asc was measured as described (Foyer et al., 1983). Leaves were ground in 2.5 m HClO₄ and centrifuged at 13,000 rpm for 10 min. Two volumes of 1.25 m Na₂CO₃ were added to the supernatant and following centrifugation, 100 μL was added to 895 μL 100 mm K₂HPO₄/KH₂PO₄, pH 5.6. Asc was determined by the change in A₂₆₅ following the addition of 0.25 units ascorbate oxidase. The total amount of reduced and oxidized Asc (i.e. Asc and DHA) was determined by reducing DHA to Asc (in a reaction containing 100 mm K₂HPO₄/KH₂PO₄, pH 6.5, 2 mm reduced glutathione, and 0.1 μg recombinant wheat DHAR protein incubated at 25°C for 20 min prior to measuring Asc). The amount of DHA was determined as the difference between these two assays.

TBARS Assay

TBARS were measured following the approach essentially described by Larkindale and Knight (2002). A total of 0.5 g of leaf was ground in liquid nitrogen in a 1.5-mL microfuge tube using a micropestle. Then, 0.5 mL of 0.5% (w/v) thiobarbituric acid in 20% (w/v) trichloroacteic acid and 0.5 mL of buffer (50 mm Tris-HCl, pH 8.0, 175 mm NaCl) was added. Following heating at 95°C for 25 min and pelleting the cell debris, the absorbance of the supernatant was measured at 532 nm, with the A₆₀₀ subtracted to account for nonspecific turbidity. The amount of malonaldehyde was calculated using an exciting coefficient of 155 mm⁻¹ cm⁻¹, in agreement with a standard curve relating malonaldehyde concentrations to absorbance. The measurements were repeated two times and representative results presented.

Chlorophyll Measurements

Chlorophyll a and b were measured spectrophotometrically as described (Jeffrey and Humphrey, 1975). In brief, leaf samples were ground in liquid nitrogen and extracted with 90% (v/v) acetone. The A₆₆₄ and 647 nm was determined and used to calculated chlorophyll a and b content by the equations: Chl a = 11.93A₆₆₄ − 1.93A₆₄₇ and Chl b = 20.36A₆₄₇ − 5.50A₆₆₄, respectively. Each experiment was repeated two to three times and representative results presented.

Leaf Function Assays

For all experiments, plants were grown under daylight conditions of approximately 500 μmol m⁻² s⁻¹ in the morning and 1,200 μmol m⁻² s⁻¹ in the afternoon. For all experiments, measurements were collected at the same time each day (between 10:30 am and 11:30 am) during the course of the experiment. In situ rates of CO₂ assimilation, transpiration, and stomatal conductance were measured with TPS-1 portable photosynthesis system. For whole plant assays, every second leaf on a plant was measured. Each experiment was repeated two times and representative results presented.

RT-PCR Analysis

Total nucleic acid was isolated from whole leaves using the RNeasy Plant Mini kit (Qiagen). Residual genomic DNA was removed by on-column DNAse I digestion, using RQ1 RNase free DNase I (Promega). We used 0.5 μg total RNA for cDNA synthesis using Omniscript RT (Qiagen) with oligo(dT)₂₀ as the primer. PCR reactions contained 1 μL of the RT reaction, and 1.5 μg of forward and primers in a total reaction volume of 25 μL and were performed with the following conditions: initial denaturing step, 94°C/15 min; 35 cycles of 94°C/15 s, 52°C/30 s, 72°C/60 s, and a final extension step of 72°C/5 min. PCR products were visualized on an ethidium bromide-stained 1.2% agarose gel. Primers used were: actin (X63603) forward, 5′-CGCGAAAAG-ATGACTCAAATC-39 and reverse, 5′-AGATCCTTTCTGATATCCACG-3′; tobacco CP1 (AY881011) forward, 5′-CTTTATCAGAGCAAGAGCTTG-3′ and reverse, 5′-TTTTGATGCGCATATATCCAC-3′.

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers AY074784 and AY074787.