Nitric oxide accumulation in Arabidopsis is independent of NOA1 in the presence of sucrose

Authors


(fax +713 348 5154; e-mail braam@rice.edu).

Summary

Nitric oxide signals diverse responses in animals and plants. Whereas nitric oxide synthesis mechanisms in animals are well understood, how nitric oxide is synthesized and regulated in plants remains controversial. NOA1 is a circularly permuted GTPase that is important for chloroplast function and is implicated in nitric oxide synthesis. However, the reported consequences of a null mutation in NOA1 are inconsistent. Whereas some studies indicate that the noa1 mutant has severe reductions in nitric oxide accumulation, others report that nitric oxide levels are indistinguishable between noa1 and the wild type. Here, we identify a correlation between the reported ability of noa1 to accumulate nitric oxide with growth on sucrose-supplemented media. We report that noa1 accumulates both basal and salicylic acid-induced nitric oxide only when grown on media containing sucrose. In contrast, nitric oxide accumulation in wild type is largely insensitive to sucrose supplementation. When grown in the absence of sucrose, noa1 has low fumarate, pale green leaves, slow growth and reduced chlorophyll content. These phenotypes are consistent with a defect in chloroplast-derived photosynthate production and are largely rescued by sucrose supplementation. We conclude that NOA1 has a primary role in chloroplast function and that its effects on the accumulation of nitric oxide are likely to be indirect.

Introduction

Nitric oxide (NO) is a critical signal in animal cells, important for diverse physiological functions, including blood pressure regulation and neurotransmission (Schmidt and Walter, 1994; Davis et al., 2001). In plants, NO has been implicated in stomatal aperture regulation (Garcia Mata and Lamattina, 2001; Desikan et al., 2002), pathogen defense (Dangl, 1998; Delledonne et al., 1998; Ma et al., 2008), developmental transitions (He et al., 2004) and abiotic stress responses (Durner et al., 1999; Garces et al., 2001). Despite the importance of NO in plant development and environmental responses, how NO is generated in plant cells remains unclear and controversial (Crawford et al., 2006; Moreau et al., 2008; Gas et al., 2009). Unlike in animals, plant NO synthase (NOS) enzymes or genes have not yet been identified (Gas et al., 2009). Arabidopsis Nitric Oxide Associated 1 (NOA1), also identified as RIF1 (Flores-Pérez et al., 2008), was originally reported to encode a protein with NOS activity (Guo et al., 2003). However, evidence is accumulating that NOA1 has a function distinct from NO synthesis (Crawford et al., 2006; Moreau et al., 2008; Gas et al., 2009). The rif1-1 allele of noa1 was isolated because it confers fosmidomycin (FSM) resistance, possibly through altered post-transcriptional regulation of the methylerythritol phosphate (MEP) pathway, and results in chloroplast biogenesis defects (Flores-Pérez et al., 2008). Supplementation with NO-producing sodium nitroprusside (SNP) improves the rif1-1 (Flores-Pérez et al., 2008) and noa1 (Guo et al., 2003) growth phenotypes, but fails to fully rescue the rif1-1 MEP pathway defects (Flores-Pérez et al., 2008). Therefore, the defects of the rif1-1 mutant may not be fully ascribed to a reduction in NO accumulation. Furthermore, NOA1 and a related homolog from Bacillus subtilis, YqeH, have no detectable NOS activity in vitro, but instead display GTPase activity (Moreau et al., 2008; Sudhamsu et al., 2008). YqeH, which functions to assemble ribosomes in B. subtilis (Uicker et al., 2007), can rescue noa1 phenotypes when produced in plants (Flores-Pérez et al., 2008; Sudhamsu et al., 2008). Despite these data that question a direct role for NOA1/RIF1 in NO production, multiple reports indicate that the noa1 mutant has significantly reduced NO levels relative to the wild type (Guo et al., 2003; Zeidler et al., 2004; Zhao et al., 2007; Guo and Crawford, 2005; Bright et al., 2006; Li et al., 2009; Chen et al., 2010), supporting the idea that NOA1/RIF1 is a critical player in plant NO biosynthesis mechanisms.

Here, we test the hypothesis that the defective chloroplasts of noa1/rif1 mutants lead to reduced fixed carbon availability and indirectly affect the accumulation of NO. We find that exogenous sucrose restores 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (cPTIO)-sensitive 4-amino-5-methyl-amino-2′,7′-di-fluorofluorescein diacetate (DAF-FM DA) fluorescence in noa1, consistent with the restoration of NO accumulation. The small, pale and slow-growing noa1 phenotypes are also largely rescued when grown in the presence of sucrose. Non-symbiotic hemoglobin (AtHB1) gene expression, energy status measurements and nitrate reductase activities are not significantly different between noa1 and the wild type and are therefore unlikely to contribute to the altered NO levels of noa1 grown without sucrose. These data verify that NOA1/RIF1 is not essential for NO production but instead probably plays only an indirect role in the accumulation of NO.

Results

Reports of NO accumulation in noa1 largely correlate with seedling growth on media supplemented with sucrose

The literature includes significant variation as to whether noa1 is defective in NO accumulation. We hypothesized that growth conditions could account for the differing reports. Investigation of plant growth conditions in 16 published reports suggested a strong, although not perfect, correlation between the reported ability of noa1 to promote NO accumulation and growth on media containing sucrose (Table 1). Seven papers report that noa1 is defective in NO accumulation; all but one of these reports analyzed plants grown on media containing <1% sucrose (Table 1). In contrast, noa1 was reported to be unaffected in NO accumulation in five reports in which plants were grown with 1% sucrose or higher, regardless of the stimulus applied (Table 1). However, the correlation is not perfect: there are four cases when noa1 was reported to produce NO, even when grown in the absence of sucrose. These data suggest that the ability of noa1 to accumulate NO may be affected by growth conditions, and supplementation of growth media with sucrose may promote the accumulation of NO in noa1.

Table 1.   Publications reporting the ability of noa1 to accumulate NO in response to various stimuli
ReferencesSucroseNO accumulationStimulus
Guo et al. (2003) 0NoAbscisic acid (ABA)
Zeidler et al. (2004) 0NoLipopolysaccharides
Guo and Crawford (2005) 0.5%NoArginine
Bright et al. (2006) SoilNoABA
Zhao et al. (2007) 0NoSalt stress
Li et al. (2009) 0NoExtracellular calmodulin
Chen et al. (2010) 1%NoIron deficiency, auxin (NAA)
Arnaud et al. (2006) 1%YesIron
Bright et al. (2006) SoilYesH2O2
Zottini et al. (2007) 0YesSalicylic acid (SA)
Kolbert et al. (2008) 0YesIndole butyric acid (IBA)
Shi and Li (2008) 3%Yes Verticillium dahliae
Tun et al. (2008) 2%YesZeatin
Lozano-Juste and Leon (2009) 1%YesABA
Zhao et al. (2009) 0YesCold acclimation/freezing
Xuan et al. (2010) 3%YesHeat stress

Growth on media with sucrose rescues the ability of noa1 to accumulate NO

To test directly whether the presence of sucrose in the growth media can rescue the ability of noa1 to accumulate NO, we compared DAF-FM DA fluorescence in seedlings of the wild-type background, Col-0, with noa1 seedlings, grown with and without 1% sucrose. DAF-FM DA is membrane permeable and fluoresces in the presence of oxidized NO (NO+ and N2O3) (reviewed in Planchet and Kaiser, 2006) and is used widely to monitor NO levels in plants (e.g. Zottini et al., 2007; Zhao et al., 2009; Chen et al., 2010; Lozano-Juste and Leon, 2010; Wang et al., 2010). To verify the specificity of DAF-FM DA fluorescence for NO-derived compounds, we compared fluorescence detected in the presence of cPTIO, an NO scavenger.

Figure 1 demonstrates that basal DAF-FM DA fluorescence is detected in wild-type Col-0 roots, and that this basal fluorescence level is not significantly affected by the presence of sucrose in the growth media. In contrast, noa1 seedling roots have nearly undetectable DAF-FM DA fluorescence when grown on media lacking sucrose. However, in the presence of sucrose, noa1 seedlings show restored DAF-FM DA fluorescence (Figure 1). Figure 1(b) quantifies the fluorescence levels detected in both Col-0 and noa1 seedling roots and reveals that cPTIO reduces, at least partially, the DAF-FM DA fluorescence, consistent with the interpretation that some, if not most, of the DAF-FM DA fluorescence is caused by NO. These results demonstrate that noa1, when provided with an exogenous source of sucrose, is rescued in its ability to accumulate basal levels of NO.

Figure 1.

 DAF-FM DA fluorescence is increased when noa1 is grown on media containing sucrose.
(a) Five-day-old Col-0 and noa1 seedlings grown without (− sucrose) or with (+ sucrose) sucrose were pre-incubated for 1 h without (− cPTIO) or with (+ cPTIO) an NO scavenger, and then stained with the NO-sensitive fluorescent dye, DAF-FM DA, for 15 min. Fluorescence was detected with a fluorescein isothiocyanate (FITC) filter; excitation, 490 nm; emission, 515 nm. Scale bars: 500 μm.
(b) Average fluorescence intensity was quantified in the distal 900 μm of roots using ImageJ; the standard error is shown (n ≥ 3). This experiment was conducted at least five times with similar results.

We next tested whether the presence of exogenous sucrose affected the ability of noa1 to accumulate NO in response to an inducing stimulus. Figure 2 shows that 1 mm salicylic acid (SA) induces DAF-FM DA fluorescence in Col-0 wild-type roots after 1 and 2 h of incubation. To avoid saturation of the high fluorescence signals from the SA-treated plants, the grayscale dynamic range was broadened to obtain the images shown in Figure 2(a) and the quantitative data shown in Figure 2(c). To visualize basal fluorescence from seedlings incubated for 2 h in the absence of SA, the grayscale dynamic range was decreased twofold for analysis of the same data (Figure 2b). Sucrose in the media caused an additional moderate increase in wild type fluorescence at the later time point (Figure 2a,c). noa1 fluorescence increased in the presence of salicylic acid only when the seedlings were grown on sucrose (Figure 2). The rescue by sucrose was not complete, in that the fluorescence levels of the mutants were not as high as in the wild type at the later time point (Figure 2c). These results are consistent with the conclusion that SA-induced NO production in noa1, but not in the wild type, is dependent upon an exogenous supply of sucrose.

Figure 2.

 Salicylic acid (SA) induces NO accumulation in noa1 only in the presence of sucrose.
(a) Five-day-old Col-0 and noa1 seedlings grown with (+ sucrose) or without (− sucrose) sucrose were incubated in the presence of DAF-FM DA for 15 min. Following DAF-FM DA incubation, seedlings were incubated without SA for 2 h (2 h 0 mm SA) as controls or with SA for 1 or 2 h. Fluorescence was detected with a fluorescein isothiocyanate (FITC) filter; excitation, 490 nm; emission, 515 nm. Scale bars: 500 μm. The grayscale range was adjusted relative to that used in Figure 1 to accommodate the much higher fluorescence emitted by the SA-treated roots.
(b) The same fluorescence data as in (a) for the control seedlings (2 h 0 mm SA) with a twofold reduction of the grayscale range to enable visualization of the lower basal fluorescence.
(c) Average fluorescence intensity was quantified in the distal 900 μm of roots using ImageJ. The standard error is shown. This experiment was conducted at least twice with similar results.

Together, the data in Figures 1 and 2 demonstrate that growth conditions, in particular sucrose availability, affect the ability of noa1 to accumulate NO. We conclude that noa1 is only conditionally defective in both basal and SA-induced NO accumulation and hence that NOA1 is not required for NO production.

noa1 is defective in fumarate accumulation

The ability of exogenous sucrose to restore NO accumulation in noa1 suggests that noa1 may be deficient in endogenous fixed carbon levels. To test this idea, we compared fumarate accumulation levels in noa1 versus wild type. Fumarate is a major storage form of fixed carbon in Arabidopsis (Chia et al., 2000). We quantified fumarate in noa1 seedlings, grown with and without sucrose, and compared these levels with the wild type. When grown on media lacking sucrose, noa1 had only 47% fumarate accumulation relative to the wild type (Figure 3). However, the availability of sucrose in the growth media promoted fumarate accumulation in noa1 to levels comparable with those of the wild type (Figure 3). Therefore, noa1 has the ability to generate fumarate if supplied with fixed carbon. These results indicate that an impairment resulting from the loss of NOA1 is reduced carbon fixation, consistent with the report that noa1 has defective chloroplasts (Flores-Pérez et al., 2008). Reduced fumarate in noa1 may also underlie the finding that noa1 displayed low activity in a citrulline-based NOS activity assay (Guo et al., 2003; Zhao et al., 2007), because the presence of fumarate in tissue extracts analyzed with the citrulline assay can result in the production of compounds that can be misinterpreted as NOS reaction products (Tischner et al., 2007). As explained more fully in the discussion, reduced fumarate levels in noa1 might have led to an erroneous interpretation that noa1 has reduced NOS activity.

Figure 3.

noa1 has reduced fumarate relative to Col-0. Shoots of 2-week-old Col-0 and noa1 seedlings grown under constant light without (open bars) or with (filled bars) sucrose were assayed for fumarate content by GC-MS. Standard deviation is shown (n = 3).

noa1 phenotypes of delayed maturation and pale green leaves are rescued by growth on sucrose-containing media

noa1 displays a number of visible phenotypes. Mutant plants accumulate less biomass and have pale green leaves when compared with the wild type grown under comparable conditions (Figure 4) (Guo et al., 2003; Flores-Pérez et al., 2008). To address whether the availability of fixed carbon affects the overt phenotypes of noa1, we compared wild-type and noa1 phenotypes grown with and without sucrose supplementation in the growth media.

Figure 4.

noa1 phenotypes of delayed maturation and pale green leaves are largely rescued by growth on sucrose-containing media.
(a) Photographs of 16-day-old Col-0 and noa1 seedlings grown without (top) and with (bottom) sucrose. Scale bar: 1 cm.
(b) Seedling fresh weight of 16-day-old wild-type (Col-0) and noa1 plants (n ≥ 25).
(c) Primary root lengths of 15-day-old Col-0 and noa1 seedlings (n ≥ 7). (b) and (c) Growth on media without (open bars) or with (filled bars) sucrose. Standard errors are shown. Statistical differences are indicated by lower case letters.

The noa1 mutant is significantly delayed in the development of both shoots and roots (Figure 4). After 2 weeks of growth on MS medium, noa1 plants were very small with short roots and pale shoots when compared with Col-0 (Figure 4a). Remarkably, this growth phenotype greatly improved when sucrose was supplied to the growth medium (Figure 4a). The shoot biomass of noa1 grown without sucrose supplementation was reduced by approximately 75% compared with Col-0 (Figure 4b). When sucrose was supplied, noa1 leaves were darker green and developed more rapidly than in the absence of sucrose (Figure 4a), although noa1 shoots grown on sucrose media were still significantly smaller (approximately 30%) than wild-type shoots. Exogenous sucrose also improved noa1 root growth. Without sucrose supplementation, noa1 roots were almost 90% shorter than those of Col-0 (Figure 4c); this growth defect was less severe, only 30% less than the wild type, when sucrose was present in the medium.

To determine whether the pale green phenotype of noa1 is caused by reduced chlorophyll accumulation, we measured chlorophyll levels in noa1 and wild-type Col-0 grown in the absence and presence of sucrose. When grown without sucrose, noa1 chlorophyll levels were only 50% that of the wild type (Figure 5). When grown in the presence of sucrose, wild-type levels of chlorophyll per fresh weight decreased, and noa1 accumulated chlorophyll levels comparable with those of the wild type. Together, these results indicate that the presence of exogenous sucrose largely rescues the growth, leaf coloration and chlorophyll level defects of noa1.

Figure 5.

noa1 has reduced chlorophyll levels. Shoots of 2-week-old Col-0 and noa1 seedlings grown without (open bars) or with (filled bars) sucrose were assayed for total chlorophyll content per fresh weight. Standard errors are shown (n ≥ 5).

noa1 does not have reduced ATP, ADP, NAD(P)H or NAD(P) levels

We have demonstrated that noa1, grown in the absence of sucrose, has altered phenotypic consequences, including the reduced accumulation of NO. Understanding why noa1 has lower NO accumulation when grown on media lacking sucrose may shed light on how NO is generated in Arabidopsis. Reduced fumarate levels and the pale green, slow-growing phenotypes of noa1 suggest that the mutant has a defect in accumulating fixed carbon, perhaps because of defective photosynthetic chloroplasts. Therefore, we sought to test whether reduced fumarate correlated with altered energy status and/or redox potential. We compared ATP, ADP, NAD+/NADH and NADP+/NADPH levels in noa1 versus Col-0, and found that the accumulation levels of these cellular metabolites were not significantly different between the wild type and noa1 plants (Figure S1). Therefore, the chloroplast and carbon fixation defects in noa1 do not affect these aspects of energy status or redox potential, suggesting that these aspects are unlikely to be responsible for the reduced accumulation of NO.

Non-symbiotic hemoglobin (AtHB1) transcript levels in noa1 are indistinguishable from those in the wild type

Non-symbiotic hemoglobin (AtHB1) has been implicated in scavenging NO in plants (Dordas et al., 2004; Perazzolli et al., 2004); therefore, one possibility is that noa1 has reduced accumulation of NO as a result of elevated accumulation of AtHB1. To test this possibility, we evaluated whether noa1 when grown without sucrose has elevated AtHB1 expression relative to wild-type Col-0. AtHB1 transcript levels were indistinguishable between noa1 and the wild type, in seedlings grown both with and without sucrose (Figure S2). Therefore, AtHB1, at least at the level of transcript abundance, is unaltered in noa1, and thus is unlikely to be responsible for the reduction in detectable NO.

noa1 is not deficient in nitrate reductase activities or accumulation

Nitrate reductase is one proposed enzymatic source of NO in plants. To test whether the ability of noa1 to accumulate NO correlated with altered nitrate reductase activity, we measured active and total nitrate reductase in Col-0 and noa1 extracts. Extracts from nia2, a mutant in the more highly expressed nitrate reductase-encoding gene, were used as controls to verify that the activities measured were the result of nitrate reductase. As reported previously (Wilkinson and Crawford, 1991), nia2 had low nitrate reductase levels (Figure S3). Although the presence of sucrose may lower the levels of both active and total nitrate reductase in both wild type and noa1 plants, there were no significant differences in activated nitrate reductase activities in noa1 compared with the wild type Col-0 (Figure S3). The only difference observed is that the total nitrate reductase level may be somewhat higher in noa1 than in the wild type (Figure S3). The active levels, however, which would be directly relevant to NO synthesis, were comparable (Figure S3). Furthermore, we would have predicted that if nitrate reductase were responsible for the reduced NO accumulation in noa1, nitrate reductase should have been reduced in noa1 relative to the wild type.

NIA1-derived activity has been proposed to be responsible for the conversion of nitrite to NO (Bright et al., 2006). Therefore, we tested whether reduced NIA1 expression may underlie the conditional phenotypes of noa1. Semi-quantitative RT-PCR of RNA purified from noa1 and Col-0 revealed that NIA1 expression was not altered in noa1 relative to Col-0, nor does the addition of sucrose have a detectable effect on NIA1 expression levels (Figure S4).

Discussion

NOA1, despite original reports (Guo et al., 2003; Guo and Crawford, 2005), is not an NO synthase (Crawford et al., 2006; Moreau et al., 2008; Sudhamsu et al., 2008; Gas et al., 2009). However, whether NOA1 might have an essential role in NO production remains a matter of speculation because loss of NOA1 was reported to dramatically reduce NO accumulation (Guo et al., 2003; Zeidler et al., 2004; Zhao et al., 2007; Guo and Crawford, 2005; Bright et al., 2006; Li et al., 2009; Chen et al., 2010). However, in conflict with this conclusion were numerous reports indicating that NO can accumulate in noa1 (Arnaud et al., 2006; Bright et al., 2006; Zottini et al., 2007; Kolbert et al., 2008; Shi and Li, 2008; Tun et al., 2008; Zhao et al., 2009; Lozano-Juste and Leon, 2010; Xuan et al., 2010). When comparing growth conditions used in these previous publications, we were intrigued by an apparent correlation between noa1 reportedly accumulating NO when plants were grown with sucrose, whereas reported failures to detect NO in noa1 generally did not provide sucrose supplementation to the plants (Table 1).

We found that NOA1 is not required for basal or SA-induced NO: when supplemented with sucrose, the ability of noa1 plants to accumulate NO is comparable with that of Col-0 wild-type plants (Figures 1 and 2). Therefore, NO accumulation in noa1 depends on the growth conditions.

NOA1, also known as RIF1, is required for chloroplast biogenesis (Flores-Pérez et al., 2008). Defective chloroplasts and photosynthesis may underlie the reduced fumarate accumulation we observed in noa1 (Figure 3). Indeed, noa1 can generate fumarate if given sucrose, an alternative form of fixed carbon (Figure 3). It is possible, therefore, that the primary requirement for NOA1 is efficient chloroplast function in generating photosynthate. The other prominent phenotypes of noa1, including slow growth, pale coloration and reduced chlorophyll levels, are also largely rescued by supplementation with sucrose (Figures 4 and 5), consistent with the interpretation that reduced levels of photosynthate resulting from defective chloroplasts is the primary physiological defect resulting from the loss of NOA1.

The finding of reduced fumarate levels in noa1 may underlie the reported conclusion that noa1 had low NO synthase activity in comparison with the wild type (Guo et al., 2003; Zhao et al., 2007). Fumarate, abundant in Arabidopsis extracts (Chia et al., 2000), interferes with the standard assay designed to monitor l-citrulline formation from l-arginine as an indicator of NO synthase activity (Tischner et al., 2007). Plant extracts generate argininosuccinate from fumarate and arginine (Tischner et al., 2007). Under standard assay conditions, argininosuccinate can be mistaken for l-citrulline accumulation, and thereby result in inaccurate interpretations of NO synthase activity levels (Tischner et al., 2007). Low fumarate in noa1 is likely to be responsible for reduced argninosuccinate accumulation, and thus for the misinterpretation that noa1 has low NO synthase activity.

Provision of exogenous sucrose enables noa1 to accumulate NO (Figures 1 and 2). This finding explains discrepancies in the literature and helps to identify the underlying cause of noa1 phenotypes, but leads to the question of why fixed carbon may be necessary for NO accumulation in Arabidopsis. We found no evidence for roles of energy status (Figure S1), hemoglobin (AtHB1) expression (Figure S2), nitrate reductase activity (Figure S3) or NIA1 expression regulation (Figure S4) in the accumulation of NO. Further experimentation is necessary to elucidate how the presence of fixed carbon enables the accumulation of NO in Arabidopsis.

Experimental Procedures

Plant materials

Arabidopsis seeds were surface sterilized, rinsed, stratified and sown on either half-strength MS agar media (Murashige and Skoog, 1962), with or without 1% sucrose, or in soil. Plates were placed in incubators held at 22°C under 16 h of light or under constant light, as indicated; soil-containing posts were placed in growth rooms under 16 h of light or under constant light.

Fresh weights and root length determinations

Plants were grown vertically on plates for 15 days in 16-h photoperiods. Fresh weights were measured with a microscale balance. For root length measurements, plants were photographed and then analysed with ImageJ (W.S. Rasband, US National Institutes of Health, Bethesda, MD, USA). Statistical analysis was performed using a one-way anova and Student’s Newman–Keuls test in spss (IBM SPSS Statistics, http://www-01.ibm.com/software/analytics/spss/products/statistics).

Chlorophyll measurements

Seedlings (14-days old) grown on plates under 16-h photoperiods were harvested, and chlorophyll was extracted in ice-cold 80% acetone for 6 h in the dark. A 100-μl volume of extract was pipetted into clear 96-well plates (Greiner Bio-One, http://www.greinerbioone.com). Absorbances were read at 645 and 663 nm in a Tecan Infinite M1000 plate reader (Tecan, http://www.tecan.com). Chlorophyll content was calculated using the formula chlorophyll (μg) = [20.2(A645) + 8.02(A663)], and normalized to fresh weight.

DAF-FM fluorescence

Five-day-old seedlings grown on plates under constant light were placed in a loading buffer [5 mm 2-(N-morpholino)ethanesulfonic acid (MES)-KOH, pH 5.7, 0.25 mm KCl, 1 mm CaCl2] with or without 500 μm cPTIO for 1 h prior to the addition of 5 μm DAF-FM DA for 15 min, and then rinsed in loading buffer for 15 min. For SA treatment, following the rinse, seedlings were placed in loading buffer with or without 1 mm SA for 1 or 2 h. The fluorescence of distal root regions was monitored with an Axioplan fluorescence microscope (Zeiss, http://www.zeiss.com). Similarly treated seedlings were used as controls to report autofluorescence levels: autofluorescence was subtracted from all experimental samples. Z-stacks were reconstructed into three-dimensional images, and the fluorescence within the area encircling the distal 900 μm of root tips was quantified using ImageJ. Two different grayscales were used to report fluorescence in Figure 2 to enable both visualization and quantification of the high fluorescence levels (Figure 2a,c) and visualization of the low basal fluorescence levels (Figure 2b).

ATP and ADP determinations

ATP and ADP levels were measured as described by Holm-Hansen and Karl (1978), with modifications. Whole seedlings were frozen and ground in liquid nitrogen using the Qiagen Tissue Lyser II (Qiagen, http://www.qiagen.com). Total adenylates were extracted in ice-cold 10% perchloric acid (1 ml per 100 mg of sample fresh weight) by vortexing, followed by neutralization with a quarter of the volume of 5 m KOH/1 m triethanolamine. Precipitated potassium perchlorate was removed by centrifugation at 5000 g for 5 min at 4°C. The volume of the supernatants was measured, and enzyme assays were set up for the determination of ATP and ADP as described by Holm-Hansen and Karl (1978), with the following modifications: 100 μl of sample was added to 70 μl of enzyme mixture (for ATP, 75 mm potassium phosphate buffer, pH 7.8, 15 mm MgCl2; for ADP, 75 mm potassium phosphate buffer, pH 7.8, 15 mm MgCl2, 0.5 mm phosphoenol pyruvate; Sigma-Aldrich, http://www.sigmaaldrich.com), 20 μg desalted pyruvate kinase (Sigma-Aldrich) and incubated at 30°C for 30 min. The enzyme reaction was stopped by 5 min of boiling and centrifugation at 5000 g. ATP was measured by bioluminescence using firefly luciferase (Sigma-Aldrich). Firefly luciferase extract was prepared as instructed by the manufacturer. A 100-μl volume was pipetted into a white 96-well polystyrene plate (Greiner Bio-One). A 50-μl volume was added to each row using a multichannel pipette. Bioluminescence emission was measured in luminescence mode 20 sec after the addition of the sample to a Tecan Infinite M1000 plate reader, with an integration time of 500 ms (Tecan). ATP was quantified after background subtraction using standards ranging from 1 to 50 μm in neutralized perchloric acid. ADP values were obtained after the subtraction of ATP values. Statistical analysis was performed using a one-way anova and Student’s Newman–Keuls test in spss.

NADH and NADPH measurements

Nicotinamide adenine dinucleotides were determined using a cycling assay, as described by Gibon and Larher (1997), as follows. Nicotinamide adenine dinucleotides were extracted by the addition of 200 μl of either 0.5 m NaOH (for the extraction of NADH and NADPH) or 0.5 m HCl (for the extraction of NAD+ and NADP+) per 25 mg of ground fresh weight, vortexed and boiled for 5 min. This was followed by centrifugation at 10 000 g for 10 min at 4°C. A 50-μl volume was neutralized with the same volume of either 0.5 m HCl or 0.5 m NaOH in 100 mm Tricine-KOH, pH 9.0. A 100-μl volume of determination mix [200 mm Tricine-KOH, pH 9.0, 2 mm 3-(4,5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromidem (MTT), 0.4 mm phenazine ethosulfate (PES), 16 mm EDTA; for determination of NAD(H): 4 units of alcohol dehydrogenase in 100 mm Tricine-KOH pH 9.0 and 10% ethanol; for determination of NADP(H): 2 units of glucose-6-phosphate dehydrogenase and 20 mm glucose-6-phosphate were added and samples were incubated in the dark at 37°C for 1 h. Reactions were stopped by adding sodium chloride to a concentration of 2.65 m, and then centrifuging at 12 000 g for 15 min at 4°C. The resulting formazan pellets were resuspended in 500 μl ethanol, and the absorbance at 570 nm was measured in the Tecan Infinite M1000 plate reader. Concentrations of nicotinamide adenine dinucleotides were determined after background subtraction using standards ranging from 1 to 10 μm.

Nitrate reductase assay

Adapting a reported protocol (Kaiser et al., 2000), nitrate reductase activity was measured as follows. Leaves of 2-week-old seedlings grown on plates with or without sucrose were collected and ground to a powder using a mortar and pestle. Extraction buffer was added to the frozen powder (50 mm Hepes-KOH, pH 7.6, 1 mm DTT, 10 μm FAD, 10 mm MgCl2 and 50 μm cantharidin). Dephosphorylation of nitrate reductase was prevented by the addition of cantharidin (Alexis Biochemicals, now Enzo Life Sciences, http://www.enzolifesciences.com/alexis). The samples were centrifuged at 16 000 g for 10 min at 4°C. To determine activated nitrate reductase levels (NRact), 100-μl aliquots of the extraction buffer supernatant were added to the reaction buffer (900 μl, 50 mm Hepes-KOH, pH 7.6, 10 μm FAD, 1 mm DTT, 10 mm MgCl2, 5 mm KNO3, 0.2 mm NADH), and incubated for 5 min at 23°C. In another 100-μl aliquot of the extraction buffer supernatant used to determine total nitrate reductase levels (NRmax), 5′-AMP and EDTA were added to final concentrations of 5 mm and 15 mm, respectively, and incubated at room temperature for 12 min. An 895-μl volume of reaction buffer (with 15 mm EDTA in place of MgCl2) was added, and the reaction was incubated for 5 min. To stop both NRact and NRmax reactions, 125 μl of 0.5 m zinc acetate was added, followed by centrifugation at 16 000 g for 10 min. The supernatant was removed and any unreacted NADH was oxidized by treatment with 10 μm phenazine methosulfate for 15 min in the dark. A 500-μl portion of each sample was then reacted with 250 μl of 1% (w/v) sulfanilamide (3 N HCl) and 250 μl 0.02% (w/v) N-(1-naphthyl)ethylenediamine (dH2O) for 20 min at room temperature. The samples were centrifuged at 16 000 g for 1 min, and the nitrite formed was assessed by colorimetric determination at OD546 nm using the Tecan Infinite M1000 plate reader. Absorbances were normalized to the protein concentration of the sample, as determined by the bicinchoninic acid (BCA) protein assay (ThermoFisher Scientific, http://www.thermofisher.com), using bovine serum albumin as a standard.

Fumarate assay

Extraction and measurement of fumarate was carried out as described by Chia et al. (2000), with minor modification. In brief, leaf tissue was collected from intact plants and immediately frozen in liquid nitrogen. Samples were finely ground using a mortar and pestle while frozen, then quickly weighed and approximately 300 mg were transferred to a 4-ml screw-top Supelco vial (Sigma-Aldrich) containing 1 ml of 1 N methanolic-HCl, and incubated at 80°C for 1 h. Equal volumes of 0.9% (w/v) NaCl and hexane were added to each sample, followed by vigorous shaking for 2 min. Tubes were subsequently centrifuged at 2000 g for 5 min at room temperature. The hexane layer (containing the dimethyl ester of fumaric acid) was then analyzed by GC-MS using a Hewlett Packard 6890 series gas chromatograph (HP, http://www.hp.com) coupled to an Agilent Technologies 5973 network mass selective detector operated in electronic ionization (EI) mode (Agilent Technologies, http://www.agilent.com). A 1-μl portion of sample was injected in 1:40 split mode at 250°C and separated using a Restek Rtx-35ms GC column (30 m × 0.25 mm × 0.1 mm) held at 100°C for 1 min after injection, and then at increasing temperatures programmed to ramp at 25°C min−1 to 160°C (Restek, http://www.restek.com). This was followed by temperature increases of 15°C min−1 until reaching 220°C, and then by an incubation at 220°C for 4.5 min. Helium was used as the carrier gas (constant flow rate 0.7 ml min−1). Measurements were carried out in full scan mode with the retention time for the dimethyl ester of fumaric acid being 2.28 min.

Quantitative reverse-transcription polymerase chain reaction (Q-PCR)

Seedlings (14-days old) were ground in liquid nitrogen, and RNA was extracted using Tri reagent (Molecular Research Center, http://www.mrcgene.com), according to the manufacturer’s instructions. Extracted RNA was quantified using a Thermo-Fisher Scientific NanoDrop 1000 spectrophotometer, and 1 μg was reverse transcribed using Invitrogen Superscript III reverse transcriptase, as instructed by the manufacturer (Invitrogen, http://www.invitrogen.com) after DNase treatment (Roche Diagnostics, http://www.roche.com). Quantitative real-time PCR was performed with Thermocycler ABI SYBR Green PCR master mix (Applied Biosystems, http://www.appliedbiosystems.com) in an ABI PRISM 7000 (Applied Biosystems) system with primers designed for the AtHB1 gene (forward primer, 5′-AATCCAAAGCTCAAGCCTCACG-3′; reverse primer, 5′-GGCTGGCTCCAAGTCTCTTCAA-3′). β-Tubulin primers (forward primer, 5′-CTGTTTCCGTACCCTCAAGC-3′; reverse primer, 5′-AGGGAAACGAAGACAGCAAG-3′) were used as a control to normalize gene expression in each sample. Transcript abundances were quantified within the logarithmic amplification phase, as previously described (Tsai et al., 2007).

Semi-quantitative polymerase chain reaction

Seedlings (14-day old) were ground in liquid nitrogen and RNA was extracted as described above. Semi-quantitative PCR ran for 28, 34 and 36 cycles in an Eppendorf Mastercycler with primers designed for the NIA1 gene (right primer, 5′-TGCACACGTTGGTCCTAATC-3′; left primer, 5′-TTCTGGTGCTGGTGTTTCTG-3′). β-Tubulin primers (same as used in Q-PCR) were used as a control.

Accession numbers

Sequence data from this article can be found in the EMBL/GenBank data libraries under accession number(s): NIA1, At1g77760; NIA2, At1g37130; NOA1/RIF1, At3g47450; AtHB1, At2g16060.

Acknowledgements

We thank Michael Grusak for scientific discussions and advice, Bonnie Bartel for feedback on our manuscript, Seiichi Matsuda for equipment use and Adriano Nunes-Nesi for the detailed NAD(P)H quantification protocol. Thanks to the Braam lab members for critical feedback. This material is based upon work supported by the National Science Foundation under grant no. MCB 0817976 to JB.

Ancillary