A Deficiency in the Flavoprotein of Arabidopsis Mitochondrial Complex II Results in Elevated Photosynthesis and Better Growth in Nitrogen-Limiting Conditions

Mitochondrial complex II (succinate dehydrogenase [SDH]) plays roles both in the tricarboxylic acid cycle and the respiratory electron transport chain. In Arabidopsis thaliana its flavoprotein subunit is encoded by two nuclear genes, SDH1-1 and SDH1-2 . Here we characterize heterozygous SDH1-1/sdh1-1 mutant plants displaying a 30% reduction in SDH activity as well as partially silenced plants obtained by RNA interference. We found that these plants displayed significantly higher CO 2 assimilation rates and enhanced growth than wild type plants. There was a strong correlation between CO 2 assimilation and stomatal conductance and both mutant and silenced plants displayed increased stomatal aperture and density. By contrast no significant differences were found for dark respiration, chloroplastic electron transport rate, CO 2 uptake at saturating concentrations of CO 2 or biochemical parameters such as the maximum rates of carboxylation by Rubisco and of photosynthetic electron transport. Thus, photosynthesis is enhanced in SDH deficient plants by a mechanism involving a specific effect on stomatal function which results in improved CO 2 uptake. Metabolic and transcript profiling revealed that mild deficiency in SDH results in limited effects on metabolism and gene expression, and data suggest that decreases observed in the levels of some amino acids were due to a higher flux to proteins and other nitrogen-containing compounds to support increased growth. Strikingly, SDH1-1/sdh1-1 seedlings grew considerably better in nitrogen-limiting conditions. Thus, a subtle metabolic alteration may lead to changes in important functions such as stomatal function and nitrogen assimilation.


INTRODUCTION
Succinate:ubiquinone oxidoreductase [succinate dehydrogenase (SDH), EC 1.3.5.1], commonly referred to as complex II, plays an essential role in mitochondrial metabolism both as a member of the electron transport chain (mETC) and the tricarboxylic acid (TCA) cycle. This membrane associated complex catalyzes the oxidation of succinate to fumarate and the reduction of ubiquinone to ubiquinol. SDH has been well characterized in bacteria and heterotrophic eukaryotes (Lemire and Oyedotun, 2002;Yankovskaya et al., 2003). In these organisms, complex II contains only four polypeptides: two peripheral membrane proteins, a flavoprotein (SDH1) which contains the succinate-binding site and an ironsulfur protein (SDH2), and two small integral membrane proteins (SDH3 and SDH4) anchoring the SDH1-SDH2 subcomplex to the matrix side of the inner membrane (Yankovskaya et al., 2003). Interestingly, it has been shown that plant complex II may contain additional subunits of unknown function along with the four classical subunits (Eubel et al., 2003;Millar et al., 2004).
Few gene functional analyses have been employed to evaluate in plants the physiological role of complex II and their constituent subunits, which are all encoded in the nuclear genome in Arabidopsis thaliana (Figueroa et al., 2001;Figueroa et al., 2002;Millar et al., 2004). For instance, the absence of SDH2-3, which is one of the three genes encoding the iron-sulfur subunit and is specifically expressed in the embryo during seed development (Elorza et al., 2006), slows down seed germination, pointing to a role of a SDH2-3containing complex II at an early step of germination (Roschzttardtz et al., 2009). The flavoprotein is encoded by two genes, designated SDH1-1 (At5g66760) and SDH1-2 (At2g18450). However, SDH1-2 is significantly expressed only in roots, albeit at a very low level (less than 10 % that of SDH1-1 in the same tissue), and its disruption has no effect on growth and development of homozygous mutant plants, leaving SDH1-1 as the only gene that encodes for a functional flavoprotein (León et al., 2007). Molecular and genetic characterization of heterozygous SDH1-1/sdh1-1 mutant plants demonstrate that SDH1-1 is essential for pollen development and important for normal embryo sac development, explaining the reduced seed set of the heterozygous mutant plants and why no homozygous mutant plants could be obtained (León et al., 2007). Although these studies have clearly demonstrated the essential role of complex II in gametophyte development, its importance in other tissues such as photosynthetically active leaves remains unknown.
In recent years, evidence has been accumulating in support of a major role for mitochondria in photosynthetic metabolism (Raghavendra and Padmasree, 2003;Noguchi and Yoshida, 2008;Nunes-Nesi et al., 2008;Nunes-Nesi et al., 2011). A growing body of evidence suggests an important role for both the TCA cycle and the mETC in the maintenance of optimal rates of photosynthesis. Indeed, photosynthetic performance can be modulated by modifications in these mitochondrial pathways. For instance, a wide range of mutant, antisense or silenced plants with deficient expression of enzymes from the TCA cycle have been analysed. Thus, tomato plants with reduced expression of aconitase or malate dehydrogenase showed an enhanced photosynthetic performance (Carrari et al., 2003;Nunes-Nesi et al., 2005). In contrast, tomato plants with less succinyl CoA ligase or citrate synthase, or Arabidopsis plants with less mitochondrial isocitrate dehydrogenase have no change in photosynthesis (Lemaitre et al., 2007;Studart-Guimaraes et al., 2007;Sienkiewicz-Porzucek et al., 2008), while tomato plants with less fumarase showed a decrease in photosynthesis (Nunes-Nesi et al., 2007). The reasons for these intriguing quite diverse effects remain somewhat unclear at a mechanistic level. However the reduced photosynthetic activity found in fumarase antisense plants was shown to be related to an impairment of stomatal function (Nunes-Nesi et al., 2007), and we have recently found that antisense repression of the iron-sulfur subunit of SDH in tomato results in a combined increase in stomatal conductance, photosynthetic rate and growth (Araújo et al., 2011a).
Few defects in the respiratory chain have been reported. The best-characterized respiratory chain mutant is a Nicotiana sylvestris mitochondrial mutant, CMSII, lacking a functional complex I (Gutierres et al., 1997). This leads to impaired photosynthesis and slower growth, however plants attain biomass similar to that of wild type plants and undergo reproductive development, although they are partially male sterile (Gutierres et al., 1997;Dutilleul et al., 2003a). Mutant plants were acclimated to the complex I defect and this acclimation includes enhanced activity of non-phosphorylating NAD(P)H dehydrogenases, which bypass complex I (Sabar et al., 2000;Dutilleul et al., 2003a;Dutilleul et al., 2003b).
An Arabidopsis mutant lacking the 18 kD Fe-S subunit of complex I has also been described and shown to be affected in cold acclimation (Lee et al., 2002). Homozygous mutant plants grow more slowly than wild type plants, nevertheless these plants eventually reach heights similar to those of wild type plants and were fully fertile. Altogether, these results indicate that a defect in complex I in plants is not lethal. In contrast, loss of complex II leads to a more severe phenotype, at least in gametophyte development (León et al., 2007), and this may be related to the dual role of SDH, in both the TCA cycle and the respiratory chain.
Here we analyse Arabidopsis SDH deficient plants, compromised in the expression of the flavoprotein subunit of SDH that were previously described by León et al. (2007).
We demonstrate that a mild reduction in mitochondrial SDH activity has a positive impact on photosynthetic performance. Our results suggest that this effect is fairly specific, with detailed characterization revealing that the major effect is in stomatal function.
Furthermore, mildly SDH deficient plants grew better under nitrogen-limiting conditions, suggesting improved nitrogen assimilation.

Enhanced Growth of SDH Deficient Plants
We have previously described heterozygous SDH1-1/ sdh1-1 mutant plants and plants with down regulation of SDH1-1 expression by RNA interference (León et al., 2007). The heterozygous mutant plants showed consistent reductions in SDH1-1 mRNA (∼50% in flowers and ∼25% in seedlings) and SDH activity (28-34 % in seedlings). SDH1-1 expression was also only partially silenced in the RNAi plants, suggesting that a drastic reduction may lead to unviable plants. To favour vegetative growth and obtain large leaves suitable for measuring photosynthetic parameters, we grew SDH1-1/sdh1-1 mutant plants and RNAi plants alongside their respective wild type controls in short day conditions (8 h light / 16 h dark). A clear increase in the growth of the aerial part of mutant and RNAi plants was observed (Fig. 1A). Determination of shoot biomass confirmed that mutant and RNAi plants accumulate more shoot matter, 2.3-and 1.7-fold respectively, than their wild type controls (Fig. 1B). Similar results were obtained in two different experiments with silenced plants and four different experiments with mutant plants, with increases in biomass ranging from 60% to 350%. Increases in fresh weight were paralleled by similar increases in dry weight (Supplemental Fig. S1). This enhanced growth phenotype was observed throughout growth (Supplemental Fig. S2). Furthermore, mutant plants attain higher biomass when sown directly in soil (Supplemental Fig. S3). However, it is worth mentioning that, in spite of this enhanced growth, seed yield is compromised, as previously described (León et al., 2007). Indeed, about 60% of the embryo sacs carrying the mutated sdh1-1 allele failed to complete their development, explaining why siliques contain fewer seeds (~33% reduction) and are shorter than those of wild type plants (León et al., 2007).

Stomatal Conductance
Given that mild SDH deficiency generated an increase in plant growth, we asked whether this phenomenon is associated with an increase in photosynthetic rates. First, gas exchange of fully expanded leaves was measured under photon flux densities (PFDs) that ranged from 100 to 1000 μmol m -2 s -1 . Compared to their respective wild type controls, mutant and RNAi plants showed significantly higher CO 2 assimilation rates ( Fig. 2A). At saturating light levels, CO 2 assimilation rates (A) were around 30% higher in the mutant and RNAi plants when compared to their respective controls. This was not due to an enhanced chloroplast electron transport since relative electron transport rates (ETRs) did not change significantly between wild type and mutant or RNAi plants (Fig. 2B).
Moreover, non photochemical quenching (NPQ) indicated no differences between lines (data not shown). Interestingly, mutant and RNAi plants were characterized by a clear increase in stomatal conductance (g s ) (Fig. 2C, increases of around 80% for mutant plants and 45% and 75% for the two RNAi lines), and there was a strong positive correlation between A and g s (Pearson correlation of 0.84, p<10 -4 , n=25). No significant differences were found in dark respiration between mutant, RNAi plants, and their respective controls (data not shown). We also plotted electron transport rate determined from fluorescence (Jf) against electron transport rate calculated from gas exchange (Jg) as described by von Caemmerer and Farquhar (1981) (Fig. 2D). Altogether, these results suggest that photosynthesis is enhanced in SDH-deficient plants by a mechanism that improved CO 2 uptake via the stomata.
To further characterize the photosynthetic response in the SDH-deficient plants we evaluated the response of A to the internal CO 2 concentration (C i ), at 700 μmol photons m -2 s -1 . These A/C i curves indicate that SDH1-1/sdh1-1 mutant plants and RNAi plants resemble wild type plants in their A responses (Supplemental Fig. S4). The maximum rate of carboxylation (Vcmax), rate of photosynthetic electron transport (Jmax), triose phosphate use (TPU) and Jmax/Vcmax were calculated for each genotype using the fitting model developed by Sharkey et al (2007). No significant differences were found between SDH-deficient and wild type plants (Table I). Gas exchange parameters were also evaluated under natural growth conditions inside the greenhouse (Supplemental Table S1). SDH deficient plants exhibited in these conditions A and g s that were significantly higher than those in the wild type and, importantly, these increases are associated with higher C i /C a ratios. When taken together our results are consistent with improved CO 2 uptake in SDHdeficient plants independently of the mesophyll photosynthetic capacity to fix CO 2 at a given C i . Interestingly, we have recently observed similar effects by decreasing the ironsulfur subunit of SDH in tomato (Araújo et al., 2011a). Further support for unaffected photosynthetic machinery was obtained by evaluation of CO 2 assimilation at saturating concentrations of 14 CO 2 . Under these conditions it is expected that CO 2 uptake via the stomata would not be limiting and, consistently, we did not found significant differences neither in total assimilation nor in radiolabel incorporation into starch, sugars, amino acids or organic acids (Supplemental Fig. S5).

Aperture and Number
In order to analyze whether increases in stomatal conductance are related to changes in stomatal aperture and/or density, we evaluated these parameters in our mutant and RNAi SDH deficient plants. Interestingly, we found an increase in stomata number which is statistically significant when comparing the mutant SDH1-1/sdh1-1 line and one of the RNAi lines (ih1.1) with their respective wild type controls (Fig. 3). Furthermore, significant increases in stomatal aperture were observed in mutant and RNAi plants ( Consistent with an important role of complex II in guard cells, we observed a high SDH1-1 promoter activity in these cells. Indeed, strong GUS staining in guard cells was revealed when 0.8 kb of the SDH1-1 promoter was fused to the β-glucuronidase (GUS) reporter gene (Fig. 4A). Furthermore, data in the existing large expression databases confirmed that SDH1-1 expression is higher in guard cells than in mesophyll cells, and that the same holds true for genes encoding other SDH subunits ( Fig. 4B; data from Leonhardt et al., 2004 andWinter et al., 2007 at http://bar.utoronto.ca).

Metabolic and Transcript Profiling of SDH Deficient Plants
We compared the levels of metabolites in the mutant SDH1-1/sdh1-1 plants with those in wild type plants grown in parallel, using 6-week-old leaves and an established gas chromatography-mass spectroscopy (GC-MS) protocol (Lisec et al., 2006). We were able to reliably identify 36 different metabolites (Supplemental Table S2). Those that are significantly different between mutant and wild type plants are shown in Figure 5. As expected for succinate dehydrogenase inhibition, we found increased levels of succinate. In contrast, no statistically significant changes were observable in the levels of other TCA cycle intermediates (malate, citrate), not even in fumarate despite of being a reaction product of SDH (Supplemental Table S2). This may be related to the fact that mutant plants have only a mild reduction in SDH activity or the fact that fumarate may be also generated from malate during operation of an anticlockwise phase of the TCA cycle (Sweetlove et al., 2010). Statistically significant decreased levels were, however, found in the mutant plants for seven amino acids: glutamate, aspartate, glutamine, asparagine, alanine, proline and 4hydroxyproline ( Fig. 5; Supplemental Table S2). These amino acids are key metabolites in nitrogen metabolism and are directly connected to TCA cycle intermediates or pyruvate (Glu, Asp, Ala), or derived from Glu (Gln, Pro) and Asp (Asn). Among the other identified metabolites, only erythritol level was found to be decreased significantly in mutant plants.
The observed decreases in amino acid levels could result from a slowing in their synthesis, which in turn may be the consequence of a slower functioning of the TCA cycle, source of the required carbon skeletons, and/or a deficiency in N assimilation. Our data argue against these possibilities since mutant plants grew better ( Fig. 1

, Supplemental
Figures S1 to S3), and no significant differences were found between wild type and mutant plants for dark respiration, 14 CO 2 incorporation into starch, sugars, amino acids and organic acids (at saturating [CO 2 ], Supplemental Fig. S5), and TCA cycle metabolite levels (malate, citrate and fumarate, Supplemental Table S2). Furthermore, neither the protein content nor the chlorophyll content were significantly altered in mutant plants (Supplemental Fig. S6), and analysis of leaf mRNA levels for genes in the nitrate assimilation pathway by real-time quantitative RT-PCR (qRT-PCR) showed that expression of these genes did not decrease in mutant plants (Fig. 6). While the levels of the transcripts for the plastid enzymes glutamine synthetase (GLN2), ferredoxin-dependent glutamate synthase (Fd-GOGAT, GLU1) and nitrite reductase (NiR), and for the cytosolic asparagine synthetase (ASN1) did not differ significantly between wild type and SDH1-1/sdh1-1mutant plants, transcript levels for one of the two NIA genes encoding nitrate reductase (NR) showed a modest increase in leaves from 6-week-old mutant plants when compared to wild type plants (Fig. 6). However, this increase in NIA2 mRNA did not result in an increase in NR activity (Supplemental Figure   S7).
Altogether, these results strongly suggest that nitrogen metabolism was not impaired in mutant plants and that the decrease in amino acid levels may be due to a higher flux to proteins and other nitrogen-containing compounds (e.g. chlorophyll) to support increased growth.  Table S2), very limited effects on gene expression were found in the mutant plants (Table II). Only 16 genes were affected, 5 genes being up-regulated and 11 genes being down-regulated, and regulation of three genes (At2g15040, At2g18440 and At4g15630) was confirmed by qRT-PCR (Supplemental Fig. S8).

Conditions
We analyzed growth of SDH1-1/sdh1-1 mutant and wild type seedlings in nitrogenlimiting conditions. A modified MS medium without nitrogen was supplemented with variable KNO 3 concentrations, ranging from 0.3 to 60 mM ( Fig. 7A  Consistently, biomass of mutant SDH1-1/sdh1-1 seedlings grown on 3 mM KNO 3 was 1.7 to 2.7-fold higher than that of wild type plants (eleven experiments in which either fresh weight and/or dry weight were determined). Better growth of SDH1-1/sdh1-1 plants was also observed following provision of 3 mM KNO 3 in the absence of sucrose (Supplemental Fig. S9B) and additionally, when KNO 3 was replaced by 1.5 mM NH 4 NO 3 (Fig. 7C).
Furthermore, enhanced growth was significant from 7 days onwards (Fig. 7D) and differences in size were observed throughout growth, with plants grown hydroponically in the presence of 3 mM KNO 3 ( Figure 7E). In some experiments, both shoot (rosette) and root weights of plants grown on 3 mM KNO 3 were separately determined: similar increases in biomass were observed for roots and shoots of the mutant SDH1-1/sdh1-1 plants (Supplemental Figures S10 A and S10B), whilst no significant differences in the rosette/root ratio were observed between genotypes. Increases in root biomass were mainly due to higher numbers of lateral roots, i.e. root branching appears to be significantly increased in the mutant plants (Supplemental Figures S10 C to S10 E). Therefore, our results demonstrate that mild SDH deficient plants have a better performance in nitrogenlimiting conditions, consistent with improved nitrogen assimilation and use.
In order to understand the enhanced growth of mutant seedlings in nitrogen-limiting conditions, we analyzed the expression of key genes. We used qRT-PCR to measure the root transcript levels of NIA1 and NIA2, which encode the two isoforms of NR, and NRT1.1 and NRT2.1, encoding nitrate transporters involved in nitrate uptake by roots (Forde, 2000;Tsay et al., 2007). Interestingly, the transcript levels of NRT1.1 and NRT2.1 were 2.5-and 2-fold higher in the roots of 15-day-old seedlings grown on 3 mM KNO 3 , and NIA1 and NIA2 expression was also slightly increased (Fig. 8). Furthermore, nitrate uptake was evaluated using K 15 NO 3 and two-week-old seedlings grown on 3mM KNO 3 . Nitrate uptake by SDH1-1/sdh1-1 mutant seedlings was higher than in the wild type (Fig. 9). Thus, the increase in biomass occurring in mutant seedlings grown in nitrogen-limiting conditions when compared to wild type plants correlated with higher nitrogen uptake and assimilation. SDH1-1 is the gene encoding the functional flavoprotein of Arabidopsis mitochondrial complex II or succinate dehydrogenase, and is an interesting target for reverse genetics analysis to gain insight into the role of complex II in plants. We have previously established that sdh1-1 is a gametophytic mutation and that complex II is essential for gametophyte development (León et al., 2007). Therefore, only heterozygous SDH1-1/sdh1-1 and partially silenced SDH1 plants, with modest reductions in SDH1-1 expression, were viable. Here we have characterized how this mild deficiency of mitochondrial complex II affects photosynthetic tissue, where the mitochondrial role is far to be fully characterized, and plant growth. On the one hand, SDH deficient plants showed higher CO 2 assimilation rates which correlated very well with higher stomatal conductance (Fig. 2). Furthermore, these changes in CO 2 assimilation and stomatal conductance were caused by increases in both stomatal aperture and density (Fig. 3). On the other hand, very few additional significant alterations were detected in the mutant and silenced plants. The lack of differences in chloroplast ETR (and NPQ), dark respiration, biochemical parameters such as the maximum rates of carboxylation by Rubisco and of photosynthetic electron transport (Supplemental Fig. S4 and Table I), CO 2 assimilation and distribution into starch, sugars, amino acids and organic acids at saturating concentrations of CO 2 (Supplemental Fig. S5), and TCA cycle metabolite levels (Supplemental Table S2), reveal that neither the photosynthetic machinery nor the TCA cycle are compromised. Moreover, few metabolites (mainly amino acids, Supplemental Table S2) and very few transcripts (Table II) were found to be different between mutant and wild type plants, confirming that only subtle changes occur in the heterozygous mutant plants. Altogether, our results suggest that enhanced photosynthesis and growth in SDH deficient plants is specifically mediated by a mechanism that improved CO 2 uptake via the stomata. It is important to point out that our experiments were performed under controlled conditions that preclude water stress. Indeed, water loss from excised leaves from SDH1-1/sdh1-1 mutant and RNAi plants resulted in 27-28 % fresh weight loss after 180 min, whereas in leaves from wild type plants (Ws and Col0) fresh weight loss was only 22-23 % (data not shown). Thus SDH deficiency may have a negative impact on water loss by transpiration and water use efficiency, however further work is required to analyse for instance the sensitivity of mutant stomata to low humidity and other environmental factors.

DISCUSSION
Quite diverse effects on photosynthetic performance have been observed on downregulation of the various steps of the TCA cycle in tomato (Carrari et al., 2003;Nunes-Nesi et al., 2005;Nunes-Nesi et al., 2007;Studart-Guimaraes et al., 2007;Sienkiewicz-Porzucek et al., 2008;Araújo et al., 2011a). While the reasons for these differences are currently unclear in several cases, results obtained by down-regulation of fumarase and succinate dehydrogenase in tomato (Nunes-Nesi et al., 2007;Araújo et al., 2011a)

Nitrogen assimilation in SDH deficient plants
Our metabolic profiling analysis revealed that the levels of organic acids were unaffected in mutant SDH1-1/sdh1-1 whole leaves, with the exception of the SDH substrate, succinate (Fig. 5, Supplemental Table S2). As outlined above this may be due, at least in part, to the fact that mutant plants have only a mild reduction in SDH activity.
Additionally in the case of fumarate, the product of the SDH reaction, unaltered levels may not be unexpected since this organic acid is present at high concentration in Arabidopsis leaves (Chia et al., 2000), where it may represent a transient storage form of fixed carbon, and it has been recently reported that it is likely produced from malate by a cytosolic fumarase (Pracharoenwattana et al., 2010). In contrast, decreases in key amino acids of nitrogen metabolism were observed (Fig. 5). Although nitrogen deficiency typically results in decreased levels of Gln, which is the first amino acid formed during ammonium assimilation, and of many other amino acids, it also leads to decreased levels of protein and other N-containing structural components like chlorophyll, decreased nitrate reductase activity, inhibition of growth and changes in plant architecture, including preferential root growth (Tschoep et al., 2009 and references therein). Therefore, the decreases in amino acids observed in our mutant plants do not appear to be the result of an impairment in nitrogen assimilation since these plants grew better (Fig. 1, Supplemental Fig. S1 to S3) and neither the protein content nor the chlorophyll content were altered (Supplemental Fig.   S6). Moreover, leaf mRNA levels for genes in the nitrate assimilation pathway did not decrease in mutant plants (Fig. 6), and nitrate reductase activity did not differ between mutant and wild type plants (Supplemental Fig. S7). It has to be pointed out that such an inverse relationship between the amino acid levels and growth or the N status is not without precedent, since Tschoep et al. (2009) found, using a soil-based growth system that allows a mild but sustained restriction of growth by N, that this condition did not alter protein or chlorophyll content and concomitantly led to decreases in growth and increases in amino acids. Thus, we interpreted the decrease in amino acid levels as the result of higher synthesis of proteins and other nitrogen-containing compounds to support increased growth.
To test if nitrogen assimilation is altered in the mutant SDH1-1/sdh1-1we analyzed growth of wild type and mutant seedlings in nitrogen-limiting conditions. Notably, mutant plants grew better, showing increased size and biomass (Fig. 7, Supplemental Fig. S9 and S10). Consistently, transcript levels of the nitrate transporters NRT1.1 and NRT2.1 were significantly elevated in roots of mutant seedlings grown under N-limiting conditions (Fig.   8), and nitrate uptake by the SDH1-1/sdh1-1 mutant was higher than in the wild type ( Fig.   9). Altogether, our results indicate that a mild reduction in SDH led to enhanced nitrate uptake and assimilation, and better growth under limiting nitrogen, an agronomical important trait. Although not directly comparable to our results, data obtained by growing plants at elevated CO 2 concentrations may be relevant (Stitt and Krapp, 1999;Matt et al., 2001). Within this context, plants growing on nitrate under elevated CO 2 led to an increase in nitrate uptake and assimilation and improved nitrogen use efficiency, which paralleled the increased rate of photosynthesis. Moreover, slight increases in NIA transcripts, minor effects on NR activity, and no effects on GLN transcripts or activity were found at elevated www.plantphysiol.org on August 20, 2017 -Published by Downloaded from Copyright © 2011 American Society of Plant Biologists. All rights reserved. CO 2 , resembling our results on the SDH1-1/sdh1-1 plants (Fig. 6, Supplemental Fig. S7).

However, these results contrast those recently published by Bloom et al. (2010) who
showed that enhanced CO 2 inhibits nitrate assimilation in wheat and Arabidopsis, hence this interaction remains far from fully understood. Further work is required to understand the relationship between the effects of SDH deficiency on stomatal function and nitrogen assimilation, a plausible hypothesis being that increased supply of C has altered the utilization of N to maintain the C/N balance.
Since carbon and nitrogen assimilation has a large influence on plant growth and crop yields, attempts to increase the efficiency of these processes are manifold.
Unfortunately, as yet in only a few cases significant improvements in C and N assimilation have been achieved (Miyagawa et al., 2001;Sinclair et al., 2004;Yanagisawa et al., 2004;Nunes-Nesi et al., 2010). The results demonstrating that mild SDH deficient plants grew

CONCLUSION
We have characterized Arabidopsis plants with a mild deficiency in mitochondrial complex II (succinate dehydrogenase) and found that these plants grew better than their wild type counterparts. CO 2 assimilation is improved through the enhancement of stomatal aperture and density, and this higher photosynthetic performance appears to lead to better nitrogen assimilation, maintaining the C/N balance to support biosynthesis of macromolecules and growth. Although the detailed molecular mechanisms linking SDH deficiency with stomatal biogenesis and function and nitrogen metabolism are far from fully understood and deserve further work, it is striking that such a subtle metabolic alteration led to changes in very important functions associated with plant growth potential.

Plant material and growth conditions
Heterozygous mutant SDH1-1/sdh1-1 plants and plants with partial down-regulation of SDH1-1 expression by RNA interference (ih1.1 and ih1.3) have been previously described (León et et al., 2007). In all experiments these mutant and silenced plants were grown in parallel with their control wild type Arabidopsis (Arabidopsis thaliana) ecotypes, Wassilewskija (Ws) and Columbia (Col0) respectively. Seeds were cold treated for 48 h at 4ºC in darkness and then sterilized for 10 min with a solution of 10% (v/v) commercial bleach. After rinsing thoroughly with sterile distilled water, seeds were sown on one-half-  (Gibeaut et al., 1997). Growth medium was slightly modified to contain 3 mM KNO 3 and 1.5 mM CaCl 2 (replacing 1.5 mM Ca(NO 3 ) 2 ).
Construct structure was verified by DNA sequencing and introduced into Agrobacterium tumefaciens GV3101 by electroporation. A. tumefaciens-mediated transformation of Arabidopsis plants was accomplished using the floral dip protocol (Clough and Bent, 1998).

Measurements of Photosynthetic Parameters
Fluorescence emission and gas-exchange measurements were made on intact, fully expanded leaves of 6-week-old plants grown as described (two weeks under axenic conditions and then 4 weeks in soil), with an open-flow gas exchange system (LI-COR, model LI-6400; http://www.licor.com/). Dark respiration was determined using the same gas exchange system. Measurements were performed at a leaf chamber temperature of 22 °C and a vapor pressure deficit of 1.55 ± 0.27 (SD) kPa. In the experiment of Figure 2, irradiances ranged from 100 to 1000 μmol photons m -2 sec -1 , and the leaf chamber CO 2 concentration was 400 μmol mol -1 . In the experiment of Supplemental Fig. S4, A/C i curves were performed at a Photon Flux Density (PFD) of 700 μmol photons m -2 s -1 . Measurements started at 350 μmol CO 2 mol -1 , and once the steady state was reached (within 3 to 5 min), CO 2 concentration was gradually lowered to 50 μmol mol -1 and then increased stepwise up to 2000 μmol mol -1 , exactly as described by Long and Bernacchi (2003). Estimates of maximum carboxylation rate (Vcmax), photosynthetic electron transport rate (Jmax) and triose phosphate use (TPU) were calculated for each A/C i curve using the fitting model of Sharkey et al. (2007).
The 14 C-labelling pattern of sucrose, starch, amino acids and organic acids was performed by illuminating leaf discs (10-mm diameter) from the same plants in an oxygen electrode chamber (Hansatech, Norfolk, UK, www.hansatech.co.uk) containing saturated level of 14 CO 2 at a PFD of 700 μmol photons m -2 sec -1 of photosynthetically active radiation at 22°C for 30 min, and subsequent fractionation was performed exactly as described by Lytovchenko et al. (2002)

Analysis of Stomatal Numbers and Apertures
Stomatal numbers and apertures were determined from the same or similar leaves as used for gas-exchange analysis. Negative impressions were taken from the abaxial surface of 6-week-old leaves 2 hours after the beginning of the light period, with dental silicone.
Positive images were then obtained using nail polish and viewed with a light microscope (Olympus IX81,Olympus, Germany) equipped for differential interference contrast (DIC).
To quantify stomata frequency, 16 to 18 samples were analyzed for each genotype (around 0.04 μ m 2 each sample). To evaluate stomatal aperture, 78-82 stomata were scored for each genotype.

Metabolic Profiling
Plants were grown for two weeks under axenic conditions and then for 4 weeks in soil, as described. Leaf samples from these 6-week-old plants were taken at the middle of the day, immediately frozen in liquid nitrogen, and stored at -80ºC until further analysis.
Extraction was performed by rapid grinding of tissue in liquid nitrogen and immediate addition of the extraction buffer, as described by Lisec et al. (2006). The levels of metabolites were quantified by GC-MS as described by Roessner et al. (2001). Data analysis was performed using the TagFinder program and included recent additions to the Max Planck Institute mass spectral libraries (Schauer et al., 2005).
Chlorophyll content was determined as described (Arnon, 1949) using pigment extracts from around 0.1 g of leaf tissue. Protein extracts were prepared as described (Roschzttardtz et al., 2009) and total protein content was assessed by the method of Bradford (1976).

Determination of NR Activity
The NR activity was measured following the protocol described by Scheible et al. (1997), in leaves from plants grown for two weeks in plates and 4 weeks in soil.

Microarray Hybridization and Data Analysis
Rosette leaves from 6-week-old (4 weeks in soil) Arabidopsis plants were collected in the middle of the light period and frozen in liquid nitrogen. Three independent biological replicates were performed for microarray experiments on the Arabidopsis ATH1 Affymetrix gene chip (Affymetrix, Santa Clara, CA). Total RNA was extracted by the TRIzol method and treated with DNase I. Concentration and purity of RNA samples were determined with a NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, DE). First-and second-strand cDNA syntheses (from 0.5 μ g of RNA), and biotin labeling of antisense cRNA were performed using the GeneChip® 3' IVT Express Kit, following the protocols of the manufacturer (Affymetrix). cRNA was quantified with the NanoDrop spectrophotometer and 15 μg were fragmented and used to hybridize the ATH1 genome Array for 16 h at 45ºC. Hybridization, washing and staining of the chips were carried out in the Affymetrix EukGE-WS2v5-450 Fluidics Station.
Arrays were scanned with the Affymetrix Genechip Scanner 3000 7G and their quality assessed using the Affymetrix quality controls. CEL files were imported into the R software for statistical computing (http://www.r-project.org) and data quality was analyzed using the simpleaffy package from the Bioconductor website

Statistical Analysis
The Student t tests were performed using the algorithm embedded into Microsoft Excel (Microsoft, Seattle). The term "significant change" is employed in the text when it is supported by a p value of less than 0.05.

Supplemental Data
The following materials are available in the online version of this article. Table S1. Gas exchange parameters of wild type and SDH-deficient plants. Table S2. Relative metabolite content of wild type and sdh1-1/SDH1-1 leaves. Figure S1. Fresh weight and dry weight of six week-old shoots. Figure S2. Improved growth of SDH deficient plants. Figure S3. Biomass of rosette leaves from six week-old plants grown in soil. Figure S4. Rate of net CO 2 assimilation as a function of internal CO 2 concentration.

Supplemental
Supplemental Figure S5. Effect of SDH deficiency on photosynthetic carbon partitioning at the onset of illumination of leaves from 6-week-old plants.   Asterisks indicate values that were determined by the t-test to be significantly different from the wild type (three asterisks, p<0.01; one asterisk, p<0.05). In A to C, asterisks indicate values that were determined by the t-test to be significantly different (p < 0.05) from the wild type. Plants were grown for two weeks under axenic conditions and then for four weeks in soil.

Supplemental
Asterisks indicate values that were determined by the t test to be significantly different from the wild type. One asterisk, p < 0.05; three asterisks, p < 0.01.  The full data set from this metabolic profiling experiment is available as Supplemental Table S2. Values are presented as means ± SE of determinations on six individual plants grown for two weeks in plates and then for 4 weeks in soil. The metabolites shown are those for which values in the mutant (grey bars) were determined by the t test to be significantly different (p<0.05) from the wild type (Ws, black bars). An asterisk indicates a value that was determined by the t test to be significantly different from the wild type, at p < 0.05.  Asterisks indicate values that were determined by the t test to be significantly different from the wild type. One asterisk, p < 0.05; three asterisks, p < 0.01. Transcript levels for genes encoding nitrate transporters (NRT1.1 and NRT2.1) and nitrate reductase (NIA1 and NIA2) were determined by qRT-PCR in the roots of 15-day-old seedlings grown under axenic conditions on 3 mM KNO 3 . Expression in mutant SDH1-1/sdh1-1 seedlings (black bars) is given relative to wild type seedlings (grey bars, set to 1).
Values were normalized using clathrin as an internal control and are shown as means ± SE from six biological replicates. Asterisks indicate values that were determined by the t test to be significantly different from the wild type, at p < 0.01. Figure 9. Higher nitrate uptake in SDH1-1/sdh1-1 mutant plants. Wild type (grey bar) and SDH1-1/sdh1-1 mutant (black bar) seedlings were grown for two weeks on solid medium containing 3 mM and 1% sucrose. Nitrate uptake was then measured by incubating for two hours seedlings in a solution with the same medium (3 mM KNO 3 ) containing 20 % K 15 NO 3 . Values are means ± SE of six replicates, each containing 15-31 seedlings.

Table I. Parameters derived from A/C i curves.
Maximum carboxylation rate (V cmax ), photosynthetic electron transport rate (J max ), triose phosphate use (TPU) and J max /V cmax ratio were computed using the fitting model developed by Sharkey et al. (2007). Values presented are means ± SE of five individual plants per genotype, and no significant differences were found by the t-test between mutant SDH1-1/sdh1-1 and RNAi (ih1.1, ih1.3) plants and their respective wild type controls (Ws and Col0).