Characterization of Transformed Arabidopsis with Altered Alternative

Characterization of Transformed Arabidopsis with Altered Alternative Oxidase Levels and Analysis of Effects on Reactive Oxygen Species in Tissue1[W] Ann L. Umbach*, Fabio Fiorani2, and James N. Siedow Developmental, Cell, and Molecular Biology Group, Biology Department, Duke University, Durham, North Carolina 27708–1000

The alternative oxidase (AOX) of plant mitochondria transfers electrons from the ubiquinone pool to oxygen without energy conservation. AOX can use reductant in excess of cytochrome pathway capacity, preventing reactive oxygen species (ROS) formation from an over-reduced ubiquinone pool, and thus may be involved in acclimation to oxidative stresses. The AOX connection with mitochondrial ROS has been investigated only in isolated mitochondria and suspension culture cells. To study ROS and AOX in whole plants, transformed lines of Arabidopsis (Arabidopsis thaliana) were generated: AtAOX1a overexpressors, AtAOX1a anti-sense plants, and overexpressors of a mutated, constitutively active AtAOX1a. In the presence of KCN, leaf tissue of either mutant or wild-type AOX overexpressors showed no increase in oxidative damage, whereas antisense lines had levels of damage greater than those observed for untransformed leaves. Similarly, ROS production increased markedly in anti-sense and untransformed, but not overexpressor, roots with KCN treatment. Thus, AOX functions in leaves and roots, as in suspension cells, to ameliorate ROS production when the cytochrome pathway is chemically inhibited. However, in contrast with suspension culture cells, no changes in leaf transcript levels of selected electron transport components or oxidative stress-related enzymes were detected under nonlimiting growth conditions, regardless of transformation type. Further, a microarray study using an anti-sense line showed AOX influences outside mitochondria, particularly in chloroplasts and on several carbon metabolism pathways. These results illustrate the value of expanding AOX transformant studies to whole tissues.

The cyanide-resistant alternative oxidase (AOX) of plant mitochondria accepts electrons from the ubiquinone (UQ) pool and uses them to reduce oxygen to water, with no conservation of energy through proton gradient formation. AOX can compete with the energyconserving cytochrome (Cyt) pathway for reductant (Millenaar and Lambers, 2003; Finnegan et al., 2004), and in vivo studies using oxygen isotope fractionation indicate that AOX is nearly always active (Robinson et al., 1995). This situation raises the question of why an apparently energy-wasteful pathway operates in plant mitochondria. Because AOX can use reductant in excess of either the Cyt pathway capacity or the rate of ATP use, AOX may act to reduce formation of reactive oxygen species

1 This work was supported by the National Science Foundation (grant no. MCB–0091080 to J.N.S. and A.L.U.). 2 Present address: VIB-Ghent University, Plant Systems Biology, Technologiepark 927, B–9052 Ghent, Belgium. * Corresponding author; e-mail [email protected]; fax 919– 613–8177. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Ann L. Umbach ([email protected]). [W] The online version of this article contains Web-only data. Article, publication date, and citation information can be found at www.plantphysiol.org/cgi/doi/10.1104/pp.105.070763.

1806

(ROS; Purvis and Shewfeldt, 1993; Millar et al., 2001a; Mittler, 2002). Over-reduction of the mitochondrial UQ pool is an important source of cellular ROS (Møller, 2001), and AOX, unrestricted by proton gradient size, can prevent such over-reduction. AOX function in this respect has been demonstrated in culture cells and isolated mitochondria. Tobacco (Nicotiana tabacum) culture cells anti-sense for AOX produce more ROS than wild-type cells, while AOX-overexpressing cells produce less (Maxwell et al., 1999). In isolated mitochondria, ROS production decreases when AOX is activated by a-keto acids (Pastore et al., 2001; Purvis, 2001), and exogenous ROS induce production of AOX transcript and/or protein, as first demonstrated in the yeast Pichia anomala (Minagawa et al., 1992). These findings point to ROS as a common element in the signaling pathways that induce AOX synthesis (McIntosh et al., 1998). However, whether decreasing mitochondrial ROS formation is an important AOX function in intact plants has yet to be demonstrated. AOX activity stabilizes the reduction state of the UQ pool in intact roots when the Cyt pathway is partially inhibited by KCN (Millenaar et al., 1998), with the likely corollary that it would also modulate ROS production under this condition. When transgenic tobacco plants overexpressing AOX were challenged with Tobacco mosaic virus (TMV), they developed smaller hypersensitive response lesions, indicating reduced programmed cell death (Ordog et al., 2002), perhaps by reduced ROS

Plant Physiology, December 2005, Vol. 139, pp. 1806–1820, www.plantphysiol.org Ó 2005 American Society of Plant Biologists

Alternative Oxidase and Oxidative Stress

production. Several abiotic stresses increase tissue ROS production and frequently result in a concomitant increase in AOX protein levels (Finnegan et al., 2004). Whether this correlation reflects either enhanced induction of AOX synthesis through ROS signaling or the ability of the resulting AOX to reduce ROS generation, or both, remains unclear. More generally, measurements of in vivo AOX activity using the oxygen isotope fractionation technique show no consistent connection between AOX activity and stress tolerance (Gonza`lez-Meler et al., 1999, 2001; Ribas-Carbo et al., 2000). To our knowledge, the rate of ROS production in unstressed or stressed intact plant tissues has not yet been unambiguously connected to changes in AOX levels and activity. One approach to the study of AOX function in whole plants is to use transgenics. Plants with altered AOX levels have been produced for potato (Solanum tuberosum; Hiser et al., 1996) and tobacco (Vanlerberghe et al., 1994, 1998; Millenaar, 2000; Gilliland et al., 2003) using the cauliflower mosaic virus (CaMV) 35S promoter and, for Arabidopsis (Arabidopsis thaliana), using a copper-inducible promoter (Potter et al., 2001). Limited characterization of detached leaves from these transformants has been reported (Vanlerberghe et al., 1995; Potter et al., 2001) and no growth phenotypes were noted. However, a striking decrease in pollen viability occurred when an AOX anti-sense construct was used with a tapetum-specific promoter in tobacco (Kitashiba et al., 1999). Only two studies have examined the effects of changes in AOX content on whole plants, both with respect to TMV resistance (Ordog et al., 2002; Gilliland et al., 2003). The tobacco CaMV 35S transformants have been used almost exclusively as suspension culture cells (Vanlerberghe et al., 1997; Parsons et al., 1999; Robson and Vanlerberghe, 2002), including the work examining ROS generation (Maxwell et al., 1999). These studies have offered insight into AOX function at the cellular level, but they also have limitations. Cells in culture may have a metabolic poise different from that of intact plant tissue, particularly with respect to oxidative stress (Antoniw et al., 1981; Halliwell, 2003; Navarro et al., 2004). Also, cells in culture may behave like some, but not all, plant tissue types. For example, tobacco culture cells grown at low inorganic phosphate have increased AOX protein (Parsons et al., 1999), but tobacco leaves grown at low inorganic phosphate do not (Gonza`lez-Meler et al., 2001). To extend the study of AOX function to whole plants, we used Arabidopsis transformed with self-AOX (AtAOX1a) under the control of the constitutive CaMV 35S promoter. The transformations made consisted of overexpression and anti-sense silencing of AOX1a and overexpression of AOX1a mutated to prevent its biochemical inactivation (Rhoads et al., 1998). In these transformed plants, as in cultured cells, AOX protein level did affect ROS formation when the Cyt pathway was chemically inhibited but, unlike in cultured cells, it did not affect the basal oxidative state of leaf tissue under nonlimiting growth conditions. Plant Physiol. Vol. 139, 2005

RESULTS AND DISCUSSION Transformation and Selection of Lines

The Arabidopsis AOX gene family consists of five members (AOX1a-d and AOX2). AOX1a was chosen to use for transformation because it is by far the most abundant transcript in all examined tissues (Saisho et al., 1997; Thirkettle-Watts et al., 2003) and its transcript levels increase when plants are treated with the respiratory inhibitor antimycin A (Saisho et al., 1997). Agrobacterium-mediated transformation was used to generate plants that were either overexpressing or anti-sense for AOX1a, or overexpressing a mutated AOX1a in which a regulatory Cys was changed to a Glu. This mutated form of AOX is not subject to inactivation by formation of the intersubunit disulfide bond and is highly active in the absence of a-keto acids normally required for activity by wild-type AOX (Rhoads et al., 1998). Transformants (T1) were isolated on kanamycincontaining medium. T2 plants were selected from T1 seed groups exhibiting 3:1 segregation for kanamycin sensitivity. PCR with genomic DNA from leaves verified the presence of the AOX transgene (data not shown). A third round of selection on kanamycin determined homozygous individuals (T3) and these and their progeny (T4) were used for further characterization. During the selection process, no morphological or fertility phenotypic differences were observed for any of the transformants, similar to other cases of transformation with AOX under CaMV 35S promoter control (Vanlerberghe et al., 1994; Millenaar, 2000; Gilliland et al., 2003). Line Characterization

Overexpression of AOX protein was verified by immunoblots of whole-leaf extracts from soil-grown plants of selected T3 lines (Fig. 1A). In untransformed or empty-vector transformed leaves, no AOX protein was detectable, in contrast to a report in which AOX was detected in blots of Arabidopsis leaf extracts (Simons et al., 1999). Variable, but considerable, amounts of AOX of the expected mature protein size (34 kD) were detected in the transformants expressing wild-type or mutated AOX (Fig. 1A). Due to lack of detectable AOX in leaf extracts of wild-type plants, decreased levels of AOX in antisense plants had to be verified by incubating leaf tissue with 3 mM KCN. Incubation alone was frequently sufficient to induce some AOX protein expression in control tissue (Fig. 1B, pBI lanes C), but treatment with KCN markedly enhanced the level of AOX (Fig. 1B, pBI lanes 1K). In lines considered good anti-sense lines, no AOX protein appeared under either condition (Fig. 1B, AS-12 and AS-11). One line cosuppressed in leaves, but not roots, was also identified by this method (Fig. 1A, ASE-10; data not shown), but only two out of 18 assayed potential anti-sense transformants passed this test for anti-sense suppression of 1807

Umbach et al.

Figure 1. Immunoblots of AOX in whole-leaf tissue extracts from soilgrown wild-type and transformed Arabidopsis plants. A, AOX levels in single T3 leaves from the overexpressor lines selected for further study: XX-1, X-3, X-6, and XX-2, Overexpressors of wild-type AOX; E-14, E-4, and E-9, overexpressors of mutated AOX; ASE-10, a cosuppressed line; V2, empty-vector control; wild type, Col-0. Note different levels of expression and the absence of any detectable AOX in Col-0 or V2. Faint bands in these and other samples are the result of nonspecific binding by the secondary antibody (data not shown). B, Suppression of AOX protein production in leaves of T2 anti-sense plants (AS-12 and AS-11) when incubated with KCN. Lanes are control leaf tissue incubated in buffer alone for 40 h (C) and leaf tissue incubated with 3 mM KCN for 40 h (1K). Two sets of leaves are shown for each line. The incubation conditions alone promoted some AOX accumulation in the vector control line (pBI, both C lanes). A and B, Molecular mass standard positions are indicated to the left.

AOX. Vanlerberghe et al. (1994) also reported relatively low recovery of anti-sense lines (one from 27 tested transformants), perhaps indicating that a threshold concentration for AOX is necessary for plant viability. Plate-grown seedlings were used to examine AOX protein in roots (Fig. 2, A and B). AOX was present at high levels in both roots and shoots of wild-type overexpressors (Fig. 2, A [XX-2 and XX-1] and B [X-3 and X-6]). The disulfide-linked form of AOX, rarely detected in whole-tissue extracts (Millenaar and Lambers, 2003), was evident in both roots and shoots of these seedlings (Fig. 2, A and B). For the mutant AOX protein, while its expression levels in shoots were similar to that of wild-type overexpressors, expression in the roots was considerably less in all three characterized lines (Fig. 2, A [E-9 and E-4] and B [E-14]). This was the only examined characteristic for 1808

which a difference between wild-type AOX and mutant AOX overexpressors was observed. The reduced protein levels may be unrelated to the mutation. However, growing root tips, where respiration rates and ATP demand are likely to be high (Bidel et al., 2000) and where AOX protein is concentrated (Hilal et al., 1997), could represent a tissue in which AOX biochemical regulation is critical. Consistent with this possibility, more than one-half of the AOX protein in 4-d-old soybean (Glycine max) roots appeared to be in the oxidized, inactive form, whereas in older roots the oxidized form was not present (Millar et al., 1998). Apart from this feature, the absence of differences between the wild-type AOX overexpressors and the mutant AOX overexpressors indicates that, as the weight of the literature suggests (discussed in Millenaar and Lambers, 2003), AOX is usually in its activated state in tissue and expression of a mutated, constitutively active AOX is therefore a neutral change. To further verify the anti-sense lines and confirm proper AOX organellar targeting, mitochondria were isolated from leaves of selected transformant lines and wild-type plants. AOX protein levels in the mitochondria mirrored levels observed in whole-leaf extracts (Fig. 3A). The molecular masses of the transgene protein product and the native AOX are the same (Fig. 3A), indicating correct processing of the protein in overexpressor mitochondria. Mitochondria from

Figure 2. Immunoblot of AOX in shoot and root tissue of T3 plategrown transformant lines. Seeds were germinated on kanamycincontaining plates and seedlings were harvested after 2 weeks. Each lane is from a pooled sample of about 24 seedlings. Roots and shoots for each line are from the same pooled sample. A and B, Immunoblots from separate SDS-PAGE gels. Lines: pBI and V2, Vector controls; XX-2, XX-1, X-3, and X-6, overexpressors of wild-type AOX; E-9, E-4, and E-14, overexpressors of mutated AOX; AS-12 and AS-11, anti-sense for AOX. Molecular mass standard positions are indicated to the left. D and M at the right designate the disulfide-linked AOX dimer and the AOX monomer, respectively. The same levels of AOX were observed with T4 seedlings. Plant Physiol. Vol. 139, 2005

Alternative Oxidase and Oxidative Stress

Figure 3. Immunoblots of AOX in mitochondria isolated from T3 transformant and untransformed leaves. A, Samples of isolated mitochondria were prepared with reductant (DTT) in the sample buffer. For each lane, 20 mg of mitochondrial protein were loaded. Representative transformant lines: XX-2, Overexpressor of wild-type AOX; E-9, overexpressor of mutated AOX; AS-11 and AS-12, anti-sense for AOX; ASE10, a cosuppressed line; pBI, vector control; wild type, Col-0. B, Mitochondria samples were treated with DTT (left four lanes) or with DTT followed by diamide (right four lanes) before loading on the gel. The gel sample buffer did not contain reductant. For each lane, 25 mg of protein were loaded. Soybean (G. max) mitochondria were used as a positive oxidation control. D and M at the right designate the disulfide-linked AOX dimer and the AOX monomer, respectively. Labels as for A. E-133, Overexpressor of mutated AOX. For both A and B, the position of molecular mass standards is shown to the left.

anti-sense lines and the cosuppressed line (Fig. 3A) had only traces of AOX protein present. Cross-linking of overexpressed wild-type AOX with diamide (Fig. 3B, XX-2), which oxidizes disulfide bonds, demonstrated that the intersubunit disulfide bond could be formed (Umbach and Siedow, 1993). As expected, diamide cross-linking did not occur with the mutated AOX protein (Fig. 3B). The possibility existed that other AOX genes could be responsible for some portion of the AOX protein observed in the single band on immunoblots. To rule out changes in transcription of these other genes, realtime PCR was used to obtain measures of transcript levels of all five AOX genes (Table I) in leaves from soil-grown T4 plants of selected lines. Consistent with the observed AOX protein levels (Figs. 1A and 2, A and B), AOX1a transcripts were elevated over 10-fold in the overexpressor lines XX-1, XX-2, and E-14 (Table II) relative to control leaf tissue (Table II, wild type and pBI). All other AOX gene transcript levels in the transformants could not be distinguished from either untransformed or vector control (Table II) and were very low, consistent with previous observations (Saisho et al., 1997; Thirkettle-Watts et al., 2003). The low Plant Physiol. Vol. 139, 2005

expression levels (high crossing points) account for the frequent high variability between replicate measurements for these genes. Gene-specific primers were used for the anti-sense lines at the reverse transcription step to selectively transcribe either the sense or anti-sense AOX1a messages to determine their levels. The anti-sense transcript was present at the same level in both anti-sense lines, whereas the amount of sense AOX1a was reduced to 50% in AS-12 and to 6% in AS-11, relative to wild type (Table II). Regardless of the AOX1a transcript levels, AOX protein was essentially undetectable in mitochondria isolated from either anti-sense line (Fig. 3A). Respiration rates of whole-leaf tissue were measured using the inhibitors KCN and salicylhydroxamic acid (SHAM). These measurements are uninformative concerning in vivo alternative or Cyt pathway fluxes, which can only be determined using the oxygen isotope fractionation method (Robinson et al., 1995). This technique is not yet developed to work effectively with slowly respiring tissue, such as Arabidopsis leaves. However, inhibitor-based measurements can determine whether AOX activity (respiration rate in the presence of KCN) and protein levels correlate, with AOX overexpressors having the potential for greater AOX activity and anti-sense lines having reduced AOX activity. Total leaf respiration rates (Supplemental Table I) were variable and no significant differences (significance at P # 0.05) were detected among the control and transformant groups. Cyt pathway capacities (measured in the presence of SHAM; data not shown) were also variable. These results are similar to observations made with transformed tobacco cell lines (Vanlerberghe et al., 1994, 1997). For the wild-type AOX and mutant AOX overexpressor lines, adjusted respiration rates in the presence of KCN were frequently high (Supplemental Table I, rate B–C), but were not significantly different from the control group rates. However, these lines had a distinguishing feature: Upon addition of KCN, there was an increase in the rate of O2 uptake (Table III). Expressed as percentage of total rate, this increase achieved significance in the case of the wild-type AOX overexpressor group relative to the control group (P 5 0.05; see Table III for groups used in analysis). A similar, smaller stimulation was observed for one set of wild-type plants (Table III, planting 2). Stimulation of O2 uptake by antimycin A (Vanlerberghe et al., 1997) or KCN (Vanlerberghe et al., 1994) was also reported for a line of tobacco suspension culture cells that overexpressed AOX. When the potential for AOX activity is high in tissue and substrate is limited, KCN inhibition, by reducing ATP synthesis, can stimulate glycolysis leading to an increased supply of respiratory substrate and an increased respiratory rate through AOX (Atkin et al., 2002). Adjusted rates in the presence of KCN were lowest for the two anti-sense lines and for the cosuppressed line (Supplemental Table I), and, when 1809

Umbach et al.

Table I. Real-time PCR: gene loci, primers, and characterization and/or nomenclature sources Gene

Locus

Forward Primer (5# to 3#)

AOX1a AOX1b AOX1c AOX1d AOX2 UCP1 UCP2 NDC1 NDA1 NDA2 NDB1 NDB2 NDB4 COX6b Complex I, 75 kD

At3g22370 At3g22360 At3g27620 At1g32350 At5g64210 At3g54110 At5g58970 At5g08740 At1g07180 At2g29990 At4g28220 At4g05020 At2g20800 At1g22450 At5g37510

CTGGACCACGTTTGTTC GATGATGAGTCGTCGCT CACTACATTACTCCGTCG TACCGCACTCTTCGAC AAGCAGCTTTACGAGTTC ATTAGCGAGTTACGATCAAG TTAGTGCGCTATGGACTG TGGAATAGTGCAGTCCAT CCTTACGGTCCCTTAGT AGAGGCCAGGGATATTC CACGCGAGATGGTATT TATGCGTGGCATTGCAG CCAGCGAGATGGTATTG ATGACGAAAACGCTAGTG AGGTCGTGGTAGTGGA

ACACCCCAATAGCTCG CCACCGCTAGATCCTT CTTCACGCCCCAATAACTA GGCTGGTTATTCCCACT CCCATCCACCTCAAGTTA CCTGTTCGAGGGTCAA GCCGTGTAAAATTCGGGA GGACATTGGGACCCGA CTATTGGCCCGTCCTC ACACGTCTTGCACTGA ACCACGCCTTCTAACT GCGTAGACGGAGTACC CCATTCGTCGGTAGCC GAGGACCAGGGAATGT ATCATAGGGTCGCTCAG

MSD1(Mn-SOD) Peroxiredoxin II F Peroxiredoxin II C Glutathione reductase, cytosolic Glutathione reductase, organellar Ascorbate peroxidase, organellar Ubiquitin 5

At3g10920 At3g06050 At1g65970 At3g24170

TCAAATTCAACGGCGGAG CCGGATCGGAGTTTCA CCAATTACTGTCGGCG GTAGTGGGTCTCAGCG

AAGTAGTAGGCGTGCT AGCAAAGGGATCATTAACAG CCAAGAAGATGCGTGTAT AATGCGTCTGGTCACA

Horling et al. (2003) Horling et al. (2003) Chew et al. (2003)

At3g54660

GGCTACCCTTTCAGGAC

AGGACTTGCCTTTCCC

Chew et al. (2003)

At4g08390

TCTAGGCCAGAACGTAGT

TGGGCTACAGCGTAAT

Chew et al. (2003)

At3g62250

GACGCTTCATCTCGTCC

GTAAACGTAGGTGAGTCCA

analyzed collectively, were significantly different from those of the other groups (P 5 0.05, 0.03, and 0.03 for comparison with group 0, 1, and 2, respectively). These lines also exhibited significantly lower ratios of total rate to the rate in the presence of KCN (Table III; P 5 0.04, 0.01, and 0.02 for comparison with groups 0, 1, and 2, respectively). Overall, the respiration rates in the presence of KCN indicated increased potential for AOX activity in the overexpressor lines and, particularly, decreased activity in the anti-sense lines. The lack of impact of AOX overexpression on total respiration rates could be due to no concomitant increase in reductant availability or, in the case of the wild-type AOX overexpressors, inactivation of a pro-

Reverse Primer (5# to 3#)

Reference

Saisho et al. (1997) Saisho et al. (1997) Saisho et al. (1997) Thirkettle-Watts et al. (2003) Saisho et al. (1997) Maia et al. (1998) Wantanabe et al. (1999) Michalecka et al. (2003) Michalecka et al. (2003) Michalecka et al. (2003) Michalecka et al. (2003) Michalecka et al. (2003) Michalecka et al. (2003) Ohtsu et al. (2001) Kruft et al. (2001); Millar et al. (2001b)

portion of the protein by oxidation (e.g. Fig. 2). However, oxygen isotope fractionation analysis is needed to determine whether there was a change in proportional flux through the alternative pathway in vivo in the transformants. Effect of AOX Protein Level on ROS Production in Whole Tissue

A primary goal of this study was to determine whether AOX could function in whole tissue to decrease ROS formation from an over-reduced mitochondrial UQ pool. An indirect measure of ROS, thiobarbituric acid reactive substances (TBARS), which are primarily lipid

Table II. Transcript levelsa of the five Arabidopsis AOX genes in wild-type and transformed plant leaves Lineb

AOX1a

WT pBI XX-1 XX-2 E-14 AS-11 AS-12

20.67a (0.04) 20.33 (0.15) 16.19 (0.01) 16.93 (0.08) 16.86 (0.11) 26.51d (0.07) 23.50d (0.05)

AOX1b

34.93 32.41 33.42 32.61 33.67 33.77 33.49

(0.42) (0.16) (0.05) (0.10) (1.22) (0.07) (0.46)

AOX1c

33.81 32.71 33.33 33.36 33.19 33.32 33.03

(0.78) (0.22) (0.21) (0.40) (0.50) (0.62) (0.32)

AOX2c

AOX1d

32.64 30.65 30.93 31.45 30.91 31.47 30.56

(0.15) (0.16) (0.20) (0.13) (0.82) (0.24) (0.09)

.46/n.d. 35.52/.46 35.97/.46 .46/.46 36.00 (0.07) .46/34.84 35.95 (0.15)

a

The average crossing-point value determined from duplicate LightCycler reactions. Half the range of the measurements is given in b parentheses. Wild type is untransformed Col-0; pBI is a vector-transformed control; XX-1 and X-2 are wild-type AOX overexpressors; E-14 is c a mutated AOX overexpressor; AS-11 and AS-12 are anti-sense lines. Values for the individual reactions are given, with two exceptions where d the average is given; n.d., not detectable. Crossing points for the anti-sense lines were determined from cDNA synthesized with an AOX sense transcript-specific primer. Crossing-point averages from wild type and XX-1, using the sense-specific primer, were 22.36 (0.00) and 17.96 (0.05), respectively. 1810

Plant Physiol. Vol. 139, 2005

Alternative Oxidase and Oxidative Stress

Table III. Leaf respiration characteristics of wild-type Col-0 and transformed Arabidopsis Where dashes appear, the plant line was not included in a planting (% Total Rates) or the value given in the column refers to the group as a whole (Estimated Group Mean). % Total Rate in the Presence of KCNa Groupb

0

1

2

3

Lineb

WT V2 pBI XX-1 XX-2 X-6 X-3 E-4 E-9 E-14 AS-11 AS-12 ASE-10

Planting 1

Planting 2

89 – 95 – 331 – – – 192 – 22 24 –

108 77 – 147 – 127 166 112 – 117 – – –

Estimated c Planting 3 Group Mean

59 – 48 – 151 – – – 108 – 23 39 26

76 – – 169 – – – 130 – – 29 – –

a

Two replicate measurements for each line at each planting were made of the uninhibited rate and the subsequent rate established following addition of KCN (10 mM for planting 1, 2.5 mM for plantings 2 and 3). Residual O2 uptake rates, in the presence of KCN and SHAM, were subtracted from the initial (total) rate and the rate in the presence of KCN to obtain corrected rates. Percent of total rate is the ratio of the b corrected 1KCN rate to corrected initial rate. Group 0, Controls (Col-0 and the vector-transformed controls, V2 and pBI); group 1, wildtype AOX overexpressors (XX-1 to X-3); group 2, mutated AOX overexpressors (E-4 to E-14); group 3, anti-sense (AS-11 and AS-12) c and cosuppressed (in leaves; ASE-10) lines. Estimated group means are inverse-transformed values of estimates from a mixedmodel ANOVA using log2-transformed data.

peroxidation products, was used for leaf tissue, whereas for roots ROS were detected directly by confocal microscopy using 5-(and 6)-carboxy-2#7#-dichlorofluorescein diacetate (carboxy-H2DCFDA). In leaves, basal levels of TBARS were similar among wild-type and overexpressor transformants (Fig. 4). Both anti-sense lines had slightly elevated TBARS at the end of the 6-h incubation period (Fig. 4), but the difference was not statistically significant. Addition of 5 mM KCN to the incubation medium caused a marked increase in TBARS in untransformed tissue and increases to an even higher level in tissue of the anti-sense lines (Fig. 4). Conversely, none of the four overexpressor lines, either of wild-type or mutant AOX, showed any increase in TBARS during KCN incubation. This is consistent with the overexpressed AOX being functional in these lines. Incubation with 5 mM SHAM and 5 mM KCN together for 6 h increased TBARS in leaf tissue even of the overexpressors, although the increase in TBARS in the presence of both inhibitors was greatest in the anti-sense lines (data not shown). The results obtained in roots using carboxyH2DCFDA mirrored those from shoots. Without KCN incubation, only background fluorescence was Plant Physiol. Vol. 139, 2005

observed and was similar in untransformed and transformant root tips (Fig. 5, a–d). With the addition of 1 mM KCN, increased fluorescence was visible in the untransformed, wild-type root tip and was pronounced in the anti-sense root tip (Fig. 5, e and h). Neither wild-type nor mutant AOX overexpressors showed any increase in fluorescence following addition of KCN (Fig. 5, f and g). These results demonstrate that, in intact leaves and roots, as in suspension culture cells (Maxwell et al., 1999), increased alternative pathway activity was sufficient to oxidize the highly reduced UQ pool resulting from KCN inhibition of the Cyt pathway and prevent ROS-related oxidative damage, whereas a decrease in pathway activity led to greater damage. During this short-term and specific chemical inhibition of the Cyt pathway, other ROS-dissipating mechanisms in tissues did not appear to be involved because the level of ROS formation was tied to the amount of AOX present (Figs. 4 and 5). Consequently, under short-term conditions of Cyt pathway impairment or saturation, the degree to which AOX can prevent mitochondrial ROS formation may be dependent upon how much excess alternative pathway capacity is present before the imposition of the stress. Survey of Electron Transport Components and Oxidative Stress-Related Transcripts by Real-Time PCR

To examine how changes in AOX protein content might affect potential interacting components in the mitochondrial electron transport chain, and how the

Figure 4. Oxidative damage in leaf tissue incubated with or without KCN. TBARS were measured as an indirect assay of ROS production. Leaves from untransformed and transformed plants grown under standard conditions were incubated with agitation in buffer either with or without 5 mM KCN at 80 mmol m22 s21 PAR for 6 h. TBARS were assayed immediately following the incubation period. For each measurement there are four replicate leaf samples, each consisting of about 10 fully expanded leaves. Error bars indicate the SD of the mean. Wild type, Col-0, untransformed; AS-11 and AS-12, anti-sense lines; E-4 and E-14, mutated AOX overexpressors; XX-1 and XX-2, wild-type AOX overexpressors. 1811

Umbach et al.

Figure 5. Confocal microscope images of Arabidopsis root tips showing ROS production in the presence of KCN. Root tips from plate-grown plants were incubated in buffer with (or without) 1 mM KCN for 10 min and then with 10 mM carboxy-H2DCFDA for 10 min before its fluorescence (brighter white against a background of endogenous fluorescence) was observed. a to d, Without KCN; e to h, after incubation with KCN. a and e, Untransformed Col-0; b and f, wild-type AOX overexpressor XX-2; c and g, mutant AOX overexpressor E-9; d and h, anti-sense line AS-12.

overall redox state of the tissue might be affected, expression levels of selected genes were analyzed in leaf tissue using real-time PCR (Table I). To distinguish changes in transcripts due to differences in AOX protein levels from those that were transformation related, the two anti-sense lines, two AOX wild-type overexpressors (XX-1 and XX-2), and a single mutant overexpressor line (E-14) were compared with ecotype Columbia (Col-0) and a vector control line (pBI). The selected genes included a complex I 75-kD protein, a complex IV subunit (COX6b), plant uncoupling proteins (UCP) UCP1 and UCP2, several of the nonphosphorylating NAD(P)H dehydrogenases (NDC1, NDA1 and 2, NDB1, 2, and 4), and enzymes commonly associated with mitochondrial or cellular oxidative stress—Mn-superoxide dismutase (Mn-SOD), mitochondrial and cytosolic peroxiredoxins, cytosolic and organellar glutathione reductase, and organellar ascorbate peroxidase (Table I). Transcript levels varied little. The maximum crossing-point difference among the seven genotypes was 1 (equivalent to a 2-fold change) or usually much less for all genes except two, NDB4 and organellar ascorbate peroxidase (1.43 and 1.56 crossing-point ranges, respectively). Most important, across all the genes surveyed, none of these modest changes were consistent between any of the duplicate genotypes used, and therefore we concluded that none were related to AOX protein levels. The lack of change in transcript levels of genes representative of mitochondrial electron transport in the AOX-transformed plants, while consistent with no detectable change in total respiration rates, is in contrast to converse situations reported in the literature. Impairment of either complex I (in maize [Zea mays; Karpova et al., 2002]; in tobacco [Sabar et al., 2000; Dutilleul et al., 2003]) or complex IV (in maize [Karpova et al., 2002]) led to marked AOX transcript and protein level increases. Similarly, AOX content and levels of the nonphosphorylating NAD(P)H dehydrogenases were found to be coordinated (Svensson and Rasmusson, 2001; Michalecka et al., 2004). Thus, modification of the alternative respiratory output path1812

way appears to have a smaller impact at the transcriptional level on the electron transport chain than modification of either the electron input pathways or the Cyt pathway. The other genes whose transcripts were measured, related to tolerance or survival of oxidative stress, could be functionally redundant with AOX or complement its role in tissues. Particularly, UCPs can dissipate the mitochondrial proton gradient, decreasing production of mitochondrial ROS (Kowaltowski et al., 1998; Pastore et al., 2000; Smith et al., 2004), and UCP overexpression made tobacco leaf tissue more resistant to hydrogen peroxide oxidative stress (Brandalise et al., 2003). In addition, while mitochondrial Mn-SOD (Kliebenstein et al., 1998) and mitochondrial peroxiredoxin II F (Horling et al., 2003) generally maintain steady levels under certain oxidative stress conditions, cytosolic peroxiredoxin II C (Horling et al., 2003) and organellar ascorbate peroxidase and glutathione reductase (Chew et al., 2003) transcript levels are known to respond to oxidative stresses. Nevertheless, none of these transcript amounts changed in any of the transformants. AOX-transformed tobacco suspension culture cells, on the other hand, did show signs of a changed oxidative status at the transcript level (Maxwell et al., 1999). Mitochondrial Mn-SOD transcript was decreased in the AOX-overexpressing cell line, while the anti-sense cell line showed induction of several stress-related genes compared to wild type. The difference in response at the transcript level between cells and leaves may reflect differences between a green tissue and a heterotrophic one, may result from cells in culture existing at a different redox set point with respect to stress, or may be due to cells lacking the larger physiological flexibility of tissues in an intact plant. Transcriptome Comparison between Wild-Type and Anti-Sense Leaves

Because the PCR measurements showed no AOXrelated changes in any of the selected gene transcript levels, a microarray experiment was carried out to Plant Physiol. Vol. 139, 2005

Alternative Oxidase and Oxidative Stress

examine more broadly where in the transcriptome changes in gene expression might have taken place. The Affymetrix full-genome ATH1 array was used to compare the anti-sense line AS-12 with untransformed Col-0 under nonlimiting growth conditions. Transcript levels varied significantly for 203 genes (Supplemental Table II). Of these, 118 transcripts increased in AS-12, with all but one (2.23-fold) having increases of 1.5-fold or less. Eighty-five transcripts decreased, with generally larger fold changes: 10 decreased to between 64% and 50% of wild-type levels (equivalent to a 1.5- to 2.0fold change) and 13 decreased from 50% to 31% of wild type (equivalent to 2.0-fold or more). AOX1a transcript was at 33% of the wild-type level (Table IV) in this experiment, comparable with the 50% level measured for this line by real-time PCR (Table II). About 20% of the 203 genes were classified as being involved in transcription or signal transduction, and about 25% were expressed or genes with unknown function (The Arabidopsis Information Resource [TAIR] analysis, biological process; Affymetrix annotation). All genes predicted to encode mitochondrial components whose transcript levels changed are listed in Table IV. In keeping with the real-time PCR results, no electron transport components or oxidative stressrelated genes showed altered transcript levels. In addition, no members of the TCA cycle were affected. A gene for a putative protein import-related protein had the largest transcript decrease in the group (apart from AOX1a; Table IV). Changes in mitochondrial import characteristics occur following stress (Taylor et al.,

2003) and this transcript change could alter mitochondrial protein composition posttranslationally. The upregulated putative carboxy-terminal peptidase could also affect mitochondrial protein composition, and the up-regulated permease and substrate carrier protein could moderate enzyme substrate or cofactor availability (Table IV). Only 14 potential stress-related genes showed transcript level changes (Table V). None are mitochondrial components. Eight are typically associated with response to pathogens or herbivores (At4g31470, At1g65690, At2g02100, At2g27080, At3g28910, At3g51660, At5g44420, and At5g50200) and all but one of these (At4g31470) exhibited down-regulation. Only three genes were directly related to oxidative stress: two glutaredoxin family proteins and APX1 (Table V). Of these, the glutaredoxin family protein, At5g58530, is predicted to be chloroplastic, and APX1, although cytoplasmic, has been found to be essential for protection of chloroplasts from ROS originating from photosynthesis (Davletova et al., 2005). Thus, consistent with the real-time PCR results, there is little evidence of oxidative stress being chronically sustained by the anti-sense plants and, if any, the stress appears to be focused on chloroplasts rather than mitochondria. Consideration of genes beyond those associated directly with stresses or mitochondria suggested that a decrease in the amount of active AOX resulted in subtle modifications of carbon metabolism (Table VI). For chloroplast-related genes (31 affected; Supplemental

Table IV. AOX anti-sense plant transcript levels of predicted mitochondrial components: fold change relative to wild type (TAIR cellular component functional category) Gene Locus

Probe Set ID

Description

At1g29440 At1g55630 At1g66540 At1g70550

257506_at 264540_at 256386_at 260363_at

At2g37630 At2g47490 At3g08670 At3g45210 At4g22130 At4g38840 At5g03555

267157_at 245152_at 258663_at 252619_at 254341_at 252972_at 250926_at

At5g54640 At1g17970 At1g66160 At2g28900

248175_at 255899_at 256522_at 266225_at

At3g23170 At3g22370 At3g24503

257925_at 258452_at 258140_at

At3g49790 At3g62720 At5g43190

252250_at 251192_at 249140_at

Auxin-responsive family protein Pentatricopeptide repeat (PPR)-containing protein Cytochrome P450, putative Expressed protein, similar to putative carboxyl-terminal peptidase MYB-domain protein (MYB91), ASYMMETRIC LEAVES1 Mitochondrial substrate carrier protein Expressed protein Expressed protein Protein kinase family protein Auxin-responsive protein, putative Permease, cytosine/purines, uracil, thiamine, allantoin family protein Histone H2A Zinc finger (C3HC4-type RING finger) family protein U-box domain-containing protein Mitochondrial import inner membrane translocase subunit Tim17/Tim22/Tim23 family protein Expressed protein AOX1a AtALDH1a, aldehyde dehydrogenase (this protein has recently been classified as cytoplasmic; see Table V) Expressed protein Galactosyl transferase GMA12/MNN10 family protein F-box family protein (FBX6)

Plant Physiol. Vol. 139, 2005

q Value

Transcript Level

0.0022 0.0429 0.0320 0.0045

1.18 1.09 1.12 1.16

0.0033 0.0110 0.0057 0.0035 0.0369 0.0119 0.0053

1.15 1.18 1.10 1.35 1.09 1.39 1.30

0.0346 0.0133 0.0047 1.56E-07

1.10 0.89 0.78 0.67

0.0146 0.0056 0.0285

0.85 0.33 0.90

0.0018 0.0014 0.0149

0.82 0.84 0.80 1813

Umbach et al.

Table V. AOX anti-sense plant transcript levels of predicted stress-related genes: fold change relative to wild type (TAIR biological process functional category, except where noted) Gene Locus

Probe Set ID

Description

At1g07890 At1g10360

261412_at 264435_at

At1g48600

261309_at

At3g20200 At4g31470a At5g58530a At1g06830a At1g65690a At2g02100

257120_at 253527_at 247830_at 260831_at 262930_at 266119_at

At2g27080a At3g28910 At3g51660

266316_at 257140_at 252076_at

At5g44420

249052_at

At5g50200

248551_at

Ascorbate peroxidase I (APX1) Glutathione S-transferase, putative (ATGSTU18) Phosphoethanolamine N-methyl transferase, putative (NMT2) Protein kinase family protein Pathogenesis-related protein, putative Glutaredoxin family protein Glutaredoxin family protein Harpin-induced protein-related Plant defensin-fusion protein, putative (PDF2.2) Harpin-induced protein-related MYB30; myb family transcription factor Macrophage migration inhibitory factor family protein Ethylene and jasmonate-responsive plant defensin (PDF 1.3-2a) Expressed protein, wound signal transduction (WR3)

Location

q Value

Transcript Level

Cytosol Cytosol

0.0070 0.0001

1.10 1.18

Cytosol

0.0345

1.22

– Endomembrane Chloroplast Endomembrane Chloroplast Endomembrane

0.0189 0.0312 0.0066 4.55E-07 0.0180 0.0159

1.09 1.13 1.12 0.81 0.88 0.73

– Nucleus –

0.0343 0.0230 0.0075

0.91 0.86 0.74

Endomembrane, cell wall Endomembrane

3.00E-17

0.46

0.0024

0.87

a

Added by authors from the complete list of changed transcripts.

Table II), transcripts for eight genes associated with the photosynthetic light reactions were affected (Table VI). Increases in the transcript for b-carotene hydroxylase 2, which catalyzes protective zeaxanthin synthesis (Table VI), and for transcripts of two enzymes involved in chlorophyll turnover (chlorophyllase [Benedetti and Arruda, 2002]; possible protochlorophyllide reductase; Table VI), occurred. These changes, together with the two chloroplast stress-related genes discussed above (Table V), suggest the existence of some pressure on the light-harvesting apparatus. No transcripts associated with the dark reactions changed, although downregulation of a gene for a thioredoxin family protein (Table VI) could have a regulatory effect on Calvin cycle enzymes. Another group of changed transcripts occurred for enzymes functioning at the point of sugar partitioning into glycolysis and the pentose phosphate pathway (PPP). Two of these, identified as trehalose-6-P (T6P) synthases, decreased to 64% and 51% of wild-type levels (ATTPS8, ATTPS11; Table VI). T6P has been identified as a critical regulator of sugar flux through glycolysis and, consequently, of carbohydrate metabolism (Eastmond and Graham, 2003; Schluepmann et al., 2003). Although ATTPS8 and ATTPS11 may be noncatalytic, they may play a role in the formation of a T6P synthase complex by analogy with yeast (Saccharomyces cerevisiae; Eastmond and Graham, 2003) and thus regulate synthase activity. Consistent with altered sugar flux, transcripts for a hexokinase and a probable 6-phosphogluconolactonase, an enzyme in the oxidizing arm of the PPP (Kruger and von Schaewen, 2003), both increased in a manner suggestive of greater partitioning of carbohydrate to the PPP (Table VI). 1814

Galactose is a component of the oligosaccharide raffinose, which, together with Suc, is exported by Arabidopsis source leaves (Haritatos et al., 2000). A number of transcript changes suggest that the anti-sense plants shifted to greater raffinose export. In addition to two b-galactosidases, transcripts of galactinol synthase, which catalyzes the committed step in raffinose synthesis (Table VI, sequence identical to AtGol1; Taji et al., 2002), increased. Conversely, UDP-Gal 4-epimerase transcripts decreased (Table VI). This enzyme converts UDP-Gal to UDP-Glc and decreasing its activity could promote a greater flux of Gal into raffinose. Consistent with reduced Suc transport, which occurs primarily via apoplastic mechanisms (Haritatos et al., 2000), transcripts of two sugar transporters decreased while transcripts of a possible invertase inhibitor increased (Table VI). Raffinose transport can take a primarily symplastic path through plasmodesmata (Haritatos et al., 2000) and a number of the transcripts differing between anti-sense and wild type may reflect the altered cell wall properties needed to accommodate this process (listed in Table VII). Additional transcript level changes were associated with events near end points of carbon metabolism. ALDH2C4 (Table VI) catalyzes the last step in ferulic and sinapic acid synthesis (Nair et al., 2004) and the three UDP-glucosyltransferase family proteins (Table VI) belong to glycosyltransferase family 1, enzymes that use a variety of substrates including sinapate (they can also be involved in anthocyanin synthesis; http://afmb.cnrs-mrs.fr/CAZY). These changes could lower the concentration of free phenylpropanoids, possibly affecting cell wall composition. Last, the two 2-oxoglutarate-dependent dioxygenases whose transcripts increased (Table VI) catalyze the final steps Plant Physiol. Vol. 139, 2005

Alternative Oxidase and Oxidative Stress

Table VI. AOX anti-sense plant transcript levels of predicted photosynthesis- and carbohydrate metabolism-related genes: fold change relative to wild type (Affymetrix annotation, with additions from TAIR, including AraCyc) Gene Locus

Probe Set ID

Description

At1g03130 At1g06680 At1g19670

263114_at 262632_at 255786_at

At2g46820 At3g63410

266716_at 251118_at

At3g59710

251480_at

At5g04260 At5g52570 At1g70290 At2g18700 At1g50460 At5g24420

245693_at 248311_at 264339_at 266072_at 261851_at 249732_at

At4g36360 At2g28470 At2g47180 At1g12780

253090_at 264078_at 263320_at 261211_at

At3g17130 At1g11260 At1g77210 At3g24503 At1g06000 At2g31790 At4g15480 At4g03060 At4g03070 At1g52000

257876_at 262456_at 264482_at 258140_at 260955_at 263477_at 245560_at 255437_at 255438_at 265053_at

PSI reaction center subunit II, chloroplast, putative PSII oxygen-evolving complex 23 (OEC23) Chlorophyllase (AraCyc); coronatine-induced protein 1 (CORI1; Affymetrix) Thylakoid membrane phosphoprotein of 14 kD (TMP14) Methyltransferase involved in plastoquinone and vitamin E biosynthesis (APG1) Short-chain dehydrogenase/reductase (SDR) family protein; possible protochlorophyllide reductase activity Thioredoxin family protein b-Carotene hydroxylase 2 T-6-P synthase, putative (ATTPS8) Trehalose phosphatase family protein (ATTPS11) Hexokinase, putative 6-Phosphogluconolactonase (TAIR); glucosamine/ galactosamine 6-P isomerase-related (Affymetrix) b-Galactosidase, putative/laccase, putative b-Galactosidase, putative/laccase, putative Galactinol synthase UDP-Glc-4-epimerase/UDP-Gal 4-epimerase/ galactowaldenase Invertase/pectin methylesterase inhibitor family Glu transporter (STP1) Sugar transporter Aldehyde dehydrogenase (ALDH2C4) UDP-glucoronosyl/UDP-glucosyl transferase family protein UDP-glucoronosyl/UDP-glucosyl transferase family protein UDP-glucoronosyl/UDP-glucosyl transferase family protein 2-Oxoglutarate-dependent dioxygenase, putative (AOP2) 2-Oxoglutarate-dependent dioxygenase (AOP1.2) Jacalin lectin family protein; myrosinase-binding protein homolog/myrosinase (AraCyc)

in biosynthesis of glucosinolates, the largest group of secondary metabolites in Arabidopsis (Kliebenstein et al., 2001). All the detected transcript changes were relatively small, in agreement with a transcriptome analysis of mutants in the phenylpropanoid pathway, where the reported changes were all under 4-fold, with most under 3-fold (Rohde et al., 2004), suggesting that plants perturbed metabolically (versus by mutations

Location

q Value

Transcript Level

Chloroplast Chloroplast –

0.0428 0.0239 0.0003

1.08 1.08 1.37

Chloroplast Chloroplast

0.0232 0.0199

1.09 1.16

Chloroplast

0.0294

1.15

Chloroplast Chloroplast Endomembrane Cytosol Endomembrane –

0.0440 0.0395 0.0012 0.0001 0.0189 0.0075

0.90 1.25 0.64 0.51 1.24 1.28

Endomembrane Endomembrane – –

0.0412 0.0019 0.0321 5.23E-5

1.13 1.14 1.51 0.82

Endomembrane Membrane Membrane Cytosol (TAIR) – – – – – –

0.0053 5.66E-8 0.0011 0.0285 0.0276 0.0133 0.0212 0.0298 0.0006 0.0017

1.13 0.79 0.85 0.90 1.14 1.16 1.09 1.30 1.18 1.37

in signal transduction or following acute environmental challenges) undergo small, but pervasive, adjustments. Although speculation concerning individual genes has to be limited until additional lines are analyzed, our results do point to AOX having clear effects on metabolism outside the mitochondrion. Changes in the chloroplast-related transcripts, in particular, are consistent with the growing body of evidence that mitochondrial function is necessary for chloroplast

Table VII. AOX anti-sense plant transcript levels of predicted cell wall-associated genes: fold change relative to wild type (Affymetrix annotation) Gene Locus

Probe Set ID

Description

q Value

Transcript Level

At1g69530 At1g35230 At2g18660 At2g40610 At3g29030 At5g19810 At1g11580 At4g02330 At5g57560

256299_at 259550_at 266070_at 255822_at 258003_at 245968_at 261826_at 255524_at 247925_at

Expansin, putative (EXP1) Arabinogalactan-protein (AGP5) Expansin family protein (EXPR3) Expansin, putative (EXP8) Expansin, putative (EXP5) Pro-rich extensin-like family protein Pectin methylesterase, putative Pectin esterase family protein Endoxyloglucan transferase (TCH4)

0.0353 0.0004 0.0126 0.0002 0.0047 0.0390 0.0068 4.69E-05 0.0190

1.25 0.86 0.86 1.28 1.22 1.12 1.15 0.57 0.39

Plant Physiol. Vol. 139, 2005

1815

Umbach et al.

function (e.g. Cardol et al., 2003; Fernie et al., 2004). AOX has been linked to the use of excess reductant arising specifically from the light reactions of photosynthesis (in addition to that generated during photorespiration; Padmasree and Raghavendra, 1999; Raghavendra and Padmasree, 2003; Fernie et al., 2004). The changes in the light reaction-associated gene transcript levels are consistent with this role as is the apparent increase in raffinose transport, a process correlated with higher light levels (Haritatos et al., 2000).

SUMMARY

Transformation of Arabidopsis produced stable lines with altered levels of functional AOX protein that did affect KCN-stimulated mitochondrial ROS production in leaves and roots. However, there was little evidence of transformation effects on mitochondrial electron transport components or the oxidative state of the tissue as assessed by transcript levels of selected genes. In contrast, AOX-transformed tobacco cultured cells showed evidence of changes in oxidative state at the transcript level (Maxwell et al., 1999). The different transcriptional responses of leaves and cultured cells indicate that effects of altered AOX levels need to be examined in intact tissues to achieve a better understanding of AOX function and its interaction with other metabolic systems in whole plants. Despite the stability of the transcripts that were examined by real-time PCR, the microarray results from a comparison of wild-type and AOX anti-sense plants showed that transcriptome changes did occur to maintain homeostasis in the anti-sense plants. These changes were largely chloroplast and carbohydrate metabolism related, suggesting that alterations in alternative pathway activity can have effects well upstream of the mitochondrial electron transport chain, and that AOX function in whole plants has repercussions beyond moderating ROS generated at the mitochondrial UQ pool (Palmer, 1976; Fernie et al., 2004). Although transcriptional analysis demonstrated that leaves and cultured cells differ, it gives an incomplete picture of the physiological status because it cannot account for posttranscriptional and posttranslational regulation of protein amount and activity (Gibon et al., 2004). For example, while preliminary results from blue native-PAGE studies are in agreement with our transcript data, showing no major differences in respiratory protein complex amounts among the different types of transformants, gel quantification suggests that some small differences may be present (M. Kolber and A. Umbach, personal communication), which, if confirmed, could have metabolic consequences not revealed at the transcript level. As a further example, AOX anti-sense tobacco cultured cells have reduced mitochondrial NADP1-isocitrate dehydrogenase protein and activity levels relative to those of wild-type cells, yet the transcript level for this enzyme is un1816

changed (Gray et al., 2004). Such posttranscriptional differences would be missed in a transcriptional analysis but could affect considerably cell oxidative status because NADPH is required for ROS detoxification via glutathione and ascorbate in mitochondria (Møller, 2001). A variety of such potentially deleterious and corresponding compensatory posttranscriptional changes related to oxidative status or to other metabolic systems is presumably present in the Arabidopsis transformants. These, added to the measured transcriptional changes, result in a normal morphology and oxidative status under standard growth conditions. Consequently, we screened for phenotypic differences among the transformants under stress conditions, in part because of the proposed role of AOX in stress tolerance and in part because limiting growth conditions might cause the plant’s ability to maintain homeostasis in the face of changed AOX protein level to be exceeded, revealing a phenotypic effect. Stressful treatments, such as growth for 3 d at 32°C or on 100 mM NaCl, when applied to seedlings grown in plates on agar-based medium, did not produce root or shoot phenotypes unique to any particular transformant type. However, when plants were grown in soil under a high-light and low-temperature regime approximating field conditions, a phenotypic effect of AOX expression level was revealed, as described in the accompanying paper (Fiorani et al., 2005).

MATERIALS AND METHODS Plant Growth Conditions For soil-grown plants, Arabidopsis (Arabidopsis thaliana L. [Heynh]) ecotype Col-0 (wild-type) and transformed plants were grown in flats of soil-less potting mix under 80 mmol photons m22 s21 photosynthetically active radiation (PAR) with a 16-h photoperiod in a growth room kept at 23°C. No fertilizer was applied.

Generation and Identification of Transformants The full-length cDNA clone of Arabidopsis Col-0 AOX1a in the expressed sequence tag (EST) clone pZL1:127M17 was used. The Glu mutation at C127 in the translated cDNA sequence was made using the QuikChange site-directed mutagenesis kit (Stratagene Cloning Systems), according to the manufacturer’s instructions with the primers 5#-GGTTCTGAATGGAAGTGGAACGAATTCAGGCCATGGG (forward) and 5#-CCCATGGCCTGAATTCGTTCCACTTCCATTCAGAACC (reverse), where the mutagenic codon is underlined. Presence of the mutation was checked by the appearance of a new EcoRI site and confirmed by sequencing the entire cDNA, which also showed no nonspecific mutations. The plasmid with the mutated AOX1a is termed pZL1:127M17:C127E. The plant expression vector pBI1.4t (Mindrinos et al., 1994) was used for transformation with AOX1a under the control of the CaMV 35S promoter. For sense constructs of the wild-type and mutated AOX1a, pBI1.4t, pZL1:127M17 and pZL1:127M17:C127E were digested with SalI and BamHI. The cDNA inserts were recovered and ligated into the digested pBI1.4t. For the anti-sense construct, pBI1.4t and pZL1:127M17 were digested with SacI and SalI, respectively, blunt ended with Klenow fragment, then digested with BamHI. pBI1.4t was further treated with calf alkaline phosphatase. The cDNA insert released from pZL1:127M17 was recovered and ligated into the digested pBI1.4t. All the ligated plasmids were introduced into XL1-Blue Escherichia coli and reisolated to confirm successful ligation and to yield sufficient plasmid DNA for Agrobacterium tumefaciens transformation by electroporation.

Plant Physiol. Vol. 139, 2005

Alternative Oxidase and Oxidative Stress

Plant transformation with Agrobacterium was by the floral-dip method (Clough and Bent, 1998). Transformants (T1) were selected on kanamycincontaining medium.

boiling for 5 min. Other samples were treated with 3 mM diamide on ice for 30 min before addition of sample buffer and boiling.

Leaf Respiration Measurements PCR for Identification of Transformants To verify cDNA insertion in the plant genome, PCR analysis was performed on genomic DNA isolated from leaves of putative transformants. Forward and reverse primers were AOX.for (5#-GATGATAACTCGCGGTGGAGCCAA) and AOX.rev (5#-GCCGAATCCAAGTATGGCTTAAGC), which are specific for the AOX1a sequence, and CMV.for (5#-CGAAAGGCTCAGTCGAAAGACTGG) and NOS.rev (5#-GACACCGCGCGCGATAATTTATCC), which are specific for the vector sequence. The AOX1a cDNA PCR product had a lower Mr than the product of the native genomic AOX1a. The vectorspecific primers verified empty-vector control lines while AOX.rev and NOS.rev detected anti-sense cDNA orientation. Amplification was by standard methods. PCR products were visualized on agarose gels with ethidium bromide staining.

Immunoblotting of Whole Tissue and Isolated Mitochondria Leaves or roots were rapidly harvested, weighed, and frozen in liquid N2. The frozen tissue was ground to a powder and double-strength SDS-PAGE sample buffer, with reductant omitted, was added (300 mL to 0.1 g tissue). Samples were boiled for 5 min and centrifuged at 14,000g for 10 min. Thirty microliters of supernatant were loaded on and separated by a Laemmli-type 10% to 13% gradient SDS-PAGE gel. Proteins from the gel were subsequently blotted onto nitrocellulose by standard methods. To probe for AOX, the blot was blocked in Tris-buffered saline plus Tween (TBST; 10 mM Tris, 120 mM NaCl, 5% Tween) with 25 g/L nonfat dry milk, rinsed in distilled water, and then incubated for 1.5 h with AOA antibody (Elthon et al., 1989) in antibody buffer (TBST with 1% bovine serum albumin) at a dilution of 1:400. The blot was extensively washed with blocking buffer, rinsed with water, and then incubated with horseradish peroxidase-conjugated anti-mouse antibody (Pierce or Santa Cruz Biotechnology) at 1:25,000 in antibody buffer for 45 min to 1 h. Following additional blocking buffer and water rinses, the blot was saturated with SuperSignal West Dura (Pierce) luminol reagent and exposed to film. For identification of anti-sense lines, two leaves from a plant were divided down their midribs. Two halves, one from each leaf, were incubated in buffer in dim light; the remaining halves were incubated with 3 mM KCN for 40 h. At the end of the incubation period, the leaves were collected and subjected to SDS-PAGE and immunoblotting as described. For immunoblots of isolated mitochondria (see below), 20 mg mitochondrial protein were used per gel lane. Gel and protein transfer conditions were the same as for whole tissue samples. Mitochondrial protein immunoblots were probed for AOX as described above, except that the AOA antibody was used at a dilution of 1:1,000 and the incubation time was extended to 2.5 h with AOA and to 1.5 h with the secondary antibody.

Isolation of Mitochondria and AOX Cross-Linking by Diamide Mitochondria were isolated from 4- to 5-week-old prebolting rosette leaves. A scaled-down version of the continuous Percoll gradient procedure (Day et al., 1985; Umbach and Siedow, 1993) was used. Briefly, 1.3 to 1.5 g fresh leaf tissue were homogenized with a mortar and pestle in 10 mL of grinding buffer, filtered through four layers of cheesecloth, and centrifuged until 3,000g was reached. The resulting supernatant was centrifuged at 27,000g for 10 min. The pellets were resuspended in 1 mL wash buffer and layered on 8.5-mL continuous Percoll gradients. These were centrifuged at 27,000g for 30 min. The mitochondrial layer toward the bottom of the gradient was collected, diluted with wash buffer, and centrifuged for 10 min at 27,000g. This wash step was repeated up to four times. The final pellet was resuspended in wash buffer from which bovine serum albumin was omitted. Mitochondrial protein was assayed by the method of Lowry et al. (1951). The isolated mitochondria were first treated with 20 mM dithiothreitol (DTT) for 1 h on ice to reduce any disulfide-linked AOX dimers, then pelleted by centrifugation and washed twice with wash buffer. Mitochondrial samples were prepared for SDS-PAGE by adding double-strength sample buffer and

Plant Physiol. Vol. 139, 2005

Leaves were harvested from rosettes in the fourth week, sliced into approximately 1-mm strips to yield between 70 and 80 mg, placed at once in assay buffer (15 mM TES, 0.2 mM CaCl2, pH 7.0), and incubated in the dark for 10 min to allow wound respiration to subside. Oxygen uptake was then monitored using a Hansatech oxygen electrode with 4.0 mL of fresh assay buffer and a black cloth cover to block out light. Additions made to the assays were 2.5 or 10 mM KCN to inhibit the Cyt pathway and 15 mM SHAM to inhibit the alternative pathway. The concentrations of KCN and SHAM to be used for complete inhibition were determined by titration experiments. Initial titrations indicated 10 mM KCN was needed for complete Cyt pathway inhibition, but, subsequently, 2.5 mM was determined to be sufficient. About 10 min were required for a steady rate to develop. We analyzed log2-transformed values of both respiration rates and percent of total rates using mixed-model ANOVA techniques (SAS PROC MIXED; SAS Institute). The logarithmic transformation was used to better meet needed assumptions. For the mixed model, we regarded ‘‘group’’ as a fixed effect and included random effects for ‘‘planting’’, ‘‘line’’, interaction terms involving those effects, and ‘‘replicate’’. The reported estimated group means are derived from the analysis (SAS least-squares means) of the log2-transformed data via the inverse transformation.

TBARS Measurements TBARS were measured according to Hodges et al. (1999) and Taylor et al. (2002). One-hundred milligrams of intact, fully expanded leaves were harvested from 3-week-old plants and incubated for 6 h at 23°C and at 80 mmol photons m22 s22 PAR in 20 mL of 15 mM TES buffer, pH 7 (control), or in the same buffer containing either 5 mM KCN or both 5 mM KCN and 5 mM SHAM. After incubation, leaves were blotted on filter paper, ground in a mortar with sand in 2.5 mL of 80% ethanol, and centrifuged at 3,000g for 10 min. The supernatant was divided into two aliquots of equal volume (typically 1 mL). One aliquot was incubated for 20 min at 95°C with an equal volume of a 20% TCA (w/v) solution including 0.01% (w/v) butylated hydroxytoluene and 0.65% thiobarbituric acid (w/v). The other aliquot was treated in the same way, but thiobarbituric acid was omitted from the TCA solution for assessment of nonspecific absorption. Samples were subsequently cooled on ice for 5 min and centrifuged at 3,000g for 10 min. The absorbance of the supernatant was measured at 532 nm, and at 440 and 600 nm (for correction for anthocyanin and sugar absorbance, respectively). TBARS levels were calculated according to Hodges et al. (1999) using the extinction coefficient for malondialdehyde at 532 nm (0.157 mol L21). Values were assessed for differences among the lines using one-way ANOVA with P , 0.05 required for significance.

Detection of ROS in Roots Seeds of wild-type Col-0, XX-2, E-9 (AOX1a wild-type and mutant overexpressors), and AS-12 (AOX1a anti-sense) transformed genotypes were surface sterilized and grown in plates on 50-mL Murashige and Skoog medium containing 1% (w/v) Suc. Five days after germination, the primary roots of four seedlings of each line were excised and placed in 5 mL of 15 mM TES buffer, pH 7, in the dark at 23°C and shaken gently for 30 min to allow recovery. Subsequently, the buffer was replaced and KCN (freshly prepared water stock solution, stabilized with 5N HCl) was added to a final concentration of 1 mM. Control samples were treated in the same way without the addition of KCN. To evaluate ROS formation, after 10 min roots were rinsed in fresh buffer and incubated for 10 min in TES buffer containing 10 mM carboxyH2DCFDA (Molecular Probes). Carboxy-H2DCFDA permeates membranes and is retained by cells after cleavage of the acetate moiety by cellular esterases. Fluorescence develops upon oxidation of the dye by hydrogen peroxide, peroxyl radical, and also peroxynitrite anion (Tarpey and Fridovich, 2001). Stock solutions of the dye were prepared in dimethyl sulfoxide and kept in the dark at 280°C. Before microscopy, samples were briefly rinsed in TES buffer to remove the dye that had not penetrated into the tissue. Roots were imaged by a Zeiss LSM410 confocal microscope outfitted with Kr-Ar, He-Ne, and UV lasers, using a 20 3 Zeiss plan-Neofluar with 0.50 numerical aperture objective and optical filters set to a maximum absorption wavelength of

1817

Umbach et al.

488 nm and maximum emission wavelength of 530 nm. Images of optical sections corresponding to the root cortex were acquired digitally 30 s after excitation for all roots. The experiment was repeated twice (n 5 8 for each genotype) with similar results.

Semiquantitative Real-Time PCR Transcript levels of selected genes were assessed in leaves from mature plants that were not yet bolting (4 weeks from planting). The RNeasy plant mini kit (Qiagen), with on-column DNase treatment (Qiagen), was used to isolate RNA, according to the manufacturer’s instructions. cDNA was synthesized from 2 mg RNA per reaction using SuperScript II RNase H2 reverse transcriptase (Invitrogen) with oligo(dTs) as primers. For gene-specific cDNA synthesis, the following primers were used: 5#-TAAGCTATTCTATCAAGA (sense AOX1a), 5#-CATCACTTATAGTCTACG (AOX1a anti-sense), and 5#-CATACATTCTTCAGATAC for the Mn-SOD transcript that was included in reactions as a control. Following synthesis, the cDNA was diluted with water at 1:20 for use in subsequent steps. Real-time PCR was performed with a Roche LightCycler (Roche Applied Science). Primers (Table I) were developed using the software LightCycler probe design, version 1.0. Most primer pairs were constructed to span intron-exon junctions. Genomic DNA contamination was detectable only for the NDB4 primer pair; NDB4 was a very low-abundance transcript. Each primer pair was assessed for efficiency, 1.5 to 2.1 being the manufacturer’s recommended range, using a five-point dilution series of cDNA. Most primer pairs had efficiencies of 1.9 or greater. Five had efficiencies of 1.7 or greater. For three genes (AOX1d, AOX2, and NDB4), transcript levels were too dilute to accurately assess efficiency. For all reactions, the QuantiTect SYBR Green PCR kit (Qiagen) was used with a scaled-down volume of 10 mL. Cycle conditions were denaturation for 15 min at 95°C; 50 amplification cycles of melting at 95°C for 15 s, annealing at 55°C for 25 s, followed by extension at 72°C for 30 s; final melting from 65°C to 95°C followed by cooling to 35°C. The final melting cycle generated a curve whose analysis helped to rule out the presence of spurious products. Reactions for each transformant line or wild type were set up in duplicate for the gene of interest and for ubiquitin in each LightCycler run, and all genotypes were included in each run. All PCR product-expected sizes were verified on agarose gels. Data were evaluated based on crossing points determined by the LightCycler software (version 3.5). Crossing point is the amplification cycle number at which the rate of change in the rate of increase of fluorescent product formation (second derivative) is maximal. Ubiquitin 5, the housekeeping gene standard, gave highly repeatable crossing points run to run and its transcripts were at comparable levels in all the lines.

Microarray Experiment A transcriptome comparison using the full Arabidopsis nuclear genome Affymetrix GeneChip ATH1 array (catalog no. 900385; www.affymetrix.com) was made between anti-sense (AS-12) and untransformed Col-0 Arabidopsis plants grown in the Duke Phytotron. Growth temperature was 23°C, light level was 350 mmol photons m22 s21 PAR with a 14-h photoperiod, and soilless potting mix, without addition of fertilizer, was used. Two bioreplicates were sampled for a total of four microarray chips—Col-0 and AS-12 leaf tissue from a first planting and from a second planting made 1 week later. At 21 d after seed sowing, leaves from the third or fourth pair were collected between 11:30 AM and noon. For each array, RNA was extracted (RNeasy plant mini kit; Qiagen.) and pooled from seven to nine individuals. Subsequent cDNA synthesis and labeled cRNA synthesis were conducted according to procedures described in the Affymetrix online literature with Affymetrix components. Arrays and cRNA were submitted to the Duke Microarray Core Facility for hybridization, signal intensity data collection, and analysis of GeneChip internal control genes using Affymetrix protocols and MicroArray Suite, version 5.0. The Affymetrix software-derived dataset from this experiment has been deposited with Gene Expression Omnibus (accession no. [series entry] GSE2406). To distinguish differences in transcript levels, a two-step mixedmodel ANOVA procedure (Wolfinger et al., 2001; see discussion in Watkinson et al., 2003) adapted specifically for Affymetrix arrays (Chu et al., 2002) was used that allows determination of meaningful changes in gene expression based on statistical significance rather than using cutoff values of expression level fold changes. Data from the Affymetrix software .CEL files (signal intensity for each probe cell) for each array were log2-transformed, and then, in the first ANOVA step of the mixed model, were scaled and normalized to each other. In the second ANOVA step, the normalized data were used for

1818

assessing statistically significant differences in individual gene transcript levels between AS-12 and Col-0. A false discovery rate parameter, called the q value (Storey, 2003), was calculated to correct for the multitesting problem and to determine differentially expressed genes. A q-value cutoff of 0.05 was chosen to maximize the likelihood of finding informative gene groups, yet minimize the probability of incorrect selections. Operationally, this means that, of 100 transcripts identified at that significance level, five are expected to be incorrect. Identified genes with q , 0.05 were further evaluated using Affymetrix notation (Liu et al., 2003) and tools available at TAIR, including AraCyc (http://arabidopsis.org). Reported fold changes are derived from the differences between the least-squares means obtained in the second ANOVA step of the analysis.

ACKNOWLEDGMENTS The Arabidopsis Biological Resource Center provided the AOX1a EST clone 127M17. We are grateful to Mikio Nakazono for the EST clone sequence and for help in early stages of this project, to Xinnian Dong’s laboratory for providing pBI1.4t, Agrobacterium for plant transformation, and primer sequences for ubiquitin 5, and to Tom Elthon for the AOA antibody. We thank David Umbach for the statistical analysis of the respiration data and Jeremy Erickson for statistical analysis of the microarray data. We also thank Vicki Ng and Marisa Puente-Jobaggy for valuable help with screening and establishing the transformed plant lines, and Julia Grammatikopolou for assistance with the confocal microscope experiments. Received August 30, 2005; revised October 18, 2005; accepted October 19, 2005; published November 18, 2005.

LITERATURE CITED Antoniw JF, Kueh JSH, Walkey DGA, White RF (1981) The presence of pathogenesis-related proteins in callus of xanthi-nc tobacco. Phytopathol Z 101: 179–184 Atkin OK, Zhang Q, Wiskich JT (2002) Effect of temperature on rates of alternative and cytochrome pathway respiration and their relationship with the redox poise of the quinone pool. Plant Physiol 128: 212–222 Benedetti CE, Arruda P (2002) Altering the expression of the chlorophyllase gene ATHCOR1 in transgenic Arabidopsis caused changes in the chlorophyll-to-chlorophyllide ratio. Plant Physiol 128: 1255–1263 Bidel LPR, Renault P, Page`s L, Rivie`re LM (2000) Mapping meristem respiration of Prunus persica (L.) Batsch seedlings: potential respiration of the meristems, O2 diffusional constraints and combined effects on root growth. J Exp Bot 51: 755–768 Brandalise M, Maia IG, Borecky´ J, Vercesi AE, Arruda P (2003) Overexpression of plant uncoupling mitochondrial protein in transgenic tobacco increases tolerance to oxidative stress. J Bioenerg Biomembr 35: 203–209 Cardol P, Gloire G, Haveux M, Remacle C, Matagne R, Franck F (2003) Photosynthesis and state transitions in mitochondrial mutants of Chlamydomonas reinhardtii affected in respiration. Plant Physiol 133: 2010–2020 Chew O, Whelan J, Millar AH (2003) Molecular definition of the ascorbateglutathione cycle in Arabidopsis mitochondria reveals dual targeting of antioxidant defenses in plants. J Biol Chem 278: 46869–46877 Chu TM, Weir B, Wolfinger R (2002) A systematic statistical linear modeling approach to oligonucleotide array experiments. Math Biosci 176: 35–51 Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16: 735–743 Davletova S, Rizhsky L, Liang H, Shengqiang Z, Oliver DJ, Coutu J, Shulaev V, Schlauch K, Mittler R (2005) Cytosolic ascorbate peroxidase 1 is a central component of the reactive oxygen gene network of Arabidopsis. Plant Cell 17: 268–281 Day DA, Neuburger M, Douce R (1985) Biochemical characterization of chlorophyll-free mitochondria from pea leaves. Aust J Plant Physiol 12: 219–228 Dutilleul C, Garmier M, Noctor G, Mathieu C, Che´trit P, Foyer CH, de Paepe R (2003) Leaf mitochondria modulate whole cell redox homeostasis, set antioxidant capacity, and determine stress resistance

Plant Physiol. Vol. 139, 2005

Alternative Oxidase and Oxidative Stress

through altered signaling and diurnal regulation. Plant Cell 15: 1212–1226 Eastmond PJ, Graham IA (2003) Trehalose metabolism: a regulatory role for trehalose-6-phosphate? Curr Opin Plant Biol 6: 231–235 Elthon TE, Nickels RL, McIntosh L (1989) Monoclonal antibodies to the alternative oxidase of higher plant mitochondria. Plant Physiol 89: 1311–1317 Fernie AR, Carrari F, Sweetlove LJ (2004) Respiratory metabolism: glycolysis, the TCA cycle and mitochondrial electron transport. Curr Opin Plant Biol 7: 254–261 Finnegan PM, Soole KL, Umbach AL (2004) Alternative mitochondrial electron transport proteins in higher plants. In DA Day, AH Millar, J Whelan, eds, Plant Mitochondria: From Genome to Function, Vol 17, Advances in Photosynthesis and Respiration. Kluwer Academic Press, Dordrecht, The Netherlands, pp 163–230 Fiorani F, Umbach AL, Siedow JN (2005) The alternative oxidase of plant mitochondria is involved in the acclimation of shoot growth at low temperature: a study of Arabidopsis AOX1a transgenic plants. Plant Physiol 139: 1795–1805 Gibon Y, Blaesing OE, Hannemann J, Carillo P, Ho¨hne M, Hendriks JHM, Palacios N, Cross H, Selbib J, Stitt M (2004) A robot-based platform to measure multiple enzyme activities in Arabidopsis using a set of cycling assays: comparison of changes of enzyme activities and transcript levels during diurnal cycles and in prolonged darkness. Plant Cell 16: 3304–3325 Gilliland A, Singh DP, Hayward JM, Moore CA, Murphy AM, York CJ, Slator J, Carr JP (2003) Genetic modification of alternative respiration has differential effects on antimycin A-induced versus salicylic acidinduced resistance to Tobacco mosaic virus. Plant Physiol 132: 1518–1528 Gonza`lez-Meler MA, Giles L, Thomas RB, Siedow JN (2001) Metabolic regulation of leaf respiration and alternative pathway activity in response to phosphate supply. Plant Cell Environ 24: 205–215 Gonza`lez-Meler MA, Ribas-Carbo M, Giles L, Siedow JN (1999) The effect of growth and measurement temperature on the activity of the alternative respiratory pathway. Plant Physiol 120: 765–772 Gray GR, Villarimo AR, Whitehead CL, McIntosh L (2004) Transgenic tobacco (Nicotiana tabacum L.) plants with increased expression levels of mitochondrial NADP1-dependent isocitrate dehydrogenase: evidence implicating this enzyme in the redox activation of the alternative oxidase. Plant Cell Physiol 45: 1413–1425 Halliwell B (2003) Oxidative stress in cell culture: an under-appreciated problem? FEBS Lett 540: 3–6 Haritatos E, Medville R, Turgeon R (2000) Minor vein structure and sugar transport in Arabidopsis thaliana. Planta 211: 105–111 Hilal M, Castagnaro A, Moreno H, Massa EM (1997) Specific localization of the respiratory alternative oxidase in meristematic and xylematic tissues from developing soybean roots and hypocotyls. Plant Physiol 115: 1499–1503 Hiser C, Kapranov P, McIntosh L (1996) Genetic modification of respiratory capacity in potato. Plant Physiol 110: 277–286 Hodges DM, DeLong JM, Forney CF, Prange RK (1999) Improving the thiobarbituric acid-reactive-substances assay for estimating lipid peroxidation in plant tissues containing anthocyanin and other interfering compounds. Planta 207: 604–611 Horling F, Lamkemeyer P, Ko¨nig J, Finkemeier I, Kandbinder A, Baier M, Dietz K-J (2003) Divergent light-, ascorbate-, and oxidative stressdependent regulation of expression of the peroxiredoxin gene family in Arabidopsis. Plant Physiol 131: 317–325 Karpova OV, Kuzmin EV, Elthon TE, Newton KJ (2002) Differential expression of alternative oxidase genes in maize mitochondrial mutants. Plant Cell 14: 3271–3284 Kitashiba H, Kitazawa E, Kishitani S, Toriyama K (1999) Partial male sterility in transgenic tobacco carrying an antisense gene for alternative oxidase under the control of a tapetum-specific promoter. Mol Breed 5: 209–218 Kliebenstein DJ, Lambrix VM, Reichelt M, Gershenzon J, Mitchell-Olds T (2001) Gene duplication in the diversification of secondary metabolism: tandem 2-oxoglutarate-dependent dioxygenases control glucosinolate biosynthesis in Arabidopsis. Plant Cell 13: 681–693 Kliebenstein DJ, Monde R-A, Last RL (1998) Superoxide dismutase in Arabidopsis: an eclectic enzyme family with disparate regulation and protein localization. Plant Physiol 118: 637–650

Plant Physiol. Vol. 139, 2005

Kowaltowski A, Costa ADT, Vercesi AE (1998) Activation of the potato plant uncoupling mitochondrial protein inhibits reactive oxygen species generation by the respiratory chain. FEBS Lett 425: 213–216 Kruft V, Eubel H, Ja¨nsch L, Werhahn W, Braun H-P (2001) Proteomic approach to identify novel mitochondrial proteins in Arabidopsis. Plant Physiol 127: 1694–1710 Kruger NJ, von Schaewen A (2003) The oxidative pentose phosphate pathway: structure and organisation. Curr Opin Plant Biol 6: 236–246 Liu G, Loraine AE, Shigeta R, Cline M, Cheng J, Valmeekam V, Sun S, Kulp D, Siani-Rose MA (2003) NetAffx: Affymetrix probesets and annotations. Nucleic Acids Res 31: 82–86 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265–275 Maia IG, Benedetti CE, Leite A, Turcinelli SR, Vercesi AE, Arruda P (1998) AtPUMP: an Arabidopsis gene encoding a plant uncoupling mitochondrial protein. FEBS Lett 429: 403–406 Maxwell DP, Wang Y, McIntosh L (1999) The alternative oxidase lowers mitochondrial reactive oxygen production in plant cells. Proc Natl Acad Sci USA 96: 8271–8276 McIntosh L, Eichler T, Gray G, Maxwell D, Nickels R, Wang Y (1998) Biochemical and genetic controls exerted by plant mitochondria. Biochim Biophys Acta 1365: 278–284 Michalecka AM, Agius SC, Møller IM, Rasmusson AG (2004) Identification of a mitochondrial external NADPH dehydrogenase by overexpression in transgenic Nicotiana sylvestris. Plant J 37: 415–425 ˚ S, Johansson FJ, Agius SC, Johanson U, Michalecka AM, Svensson A Brennicke A, Binder S, Rasmusson AG (2003) Arabidopsis genes encoding mitochondrial type II NAD(P)H dehydrogenases have different evolutionary origin and show distinct responses to light. Plant Physiol 133: 642–652 Millar AH, Atkin OK, Menz RI, Henry B, Farquhar G, Day DA (1998) Analysis of respiratory chain regulation in roots of soybean seedlings. Plant Physiol 117: 1083–1093 Millar AH, Considine MJ, Day DA, Whelan J (2001a) Unraveling the role of mitochondria during oxidative stress in plants. IUBMB Life 51: 201–205 Millar AH, Sweetlove LJ, Giege´ P, Leaver CJ (2001b) Analysis of the Arabidopsis mitochondrial proteome. Plant Physiol 127: 1711–1727 Millenaar FF (2000) The alternative oxidase of roots in Poa species: in vivo regulation and function. PhD thesis. Utrecht University, Utrecht, The Netherlands Millenaar FF, Benschop JJ, Wagner AM, Lambers H (1998) The role of the alternative oxidase in stabilizing the in vivo reduction state of the ubiquinone pool and the activation state of the alternative oxidase. Plant Physiol 118: 599–607 Millenaar FF, Lambers H (2003) The alternative oxidase: in vivo regulation and function. Plant Biol 5: 2–15 Minagawa N, Koga S, Nakano M, Sakajo S, Yoshimoto A (1992) Possible involvement of superoxide anion in the induction of cyanide-resistant respiration in Hansenula anomala. FEBS Lett 302: 217–219 Mindrinos M, Katagiri F, Yu G-L, Ausubel FM (1994) The A. thaliana disease resistance gene RPS2 encodes a protein containing a nucleotidebinding site and leucine-rich repeats. Cell 78: 1089–1099 Mittler R (2002) Oxidative stress, antioxidants and stress tolerance. Trends Plant Sci 7: 405–410 Møller IM (2001) Plant mitochondria and oxidative stress: electron transport, NADPH turnover, and metabolism of reactive oxygen species. Annu Rev Plant Physiol Plant Mol Biol 52: 561–591 Nair RB, Bastress KL, Ruegger MO, Denault JW, Chapple C (2004) The Arabidopsis thaliana REDUCED EPIDERMAL FLUORESCENCE1 gene encodes an aldehyde dehydrogenase involved in ferulic acid and sinapic acid biosynthesis. Plant Cell 16: 544–554 Navarro L, Zipfel C, Rowland O, Keller I, Robatzek S, Boller T, Jones JDG (2004) The transcriptional innate immune response to flg22. Interplay and overlap with Avr gene-dependent defense responses and bacterial pathogenesis. Plant Physiol 135: 1113–1128 Ohtsu K, Nakazono M, Tsutsumi N, Hirai A (2001) Characterization and expression of the genes for cytochrome c oxidase subunit VIb (COX6b) from rice and Arabidopsis thaliana. Gene 264: 233–239 Ordog SH, Higgins VJ, Vanlerberghe GC (2002) Mitochondrial alternative oxidase is not a critical component of plant viral resistance but may play a role in the hypersensitive response. Plant Physiol 129: 1858–1865

1819

Umbach et al.

Padmasree K, Raghavendra AS (1999) Response of photosynthetic carbon assimilation in mesophyll protoplasts to restriction on mitochondrial oxidative metabolism: metabolites related to the redox status and sucrose biosynthesis. Photosynth Res 62: 231–239 Palmer JM (1976) The organization and regulation of electron transport in plant mitochondria. Annu Rev Plant Physiol 27: 133–157 Parsons HL, Yip JYH, Vanlerberghe GC (1999) Increased respiratory restriction during phosphate-limited growth in transgenic tobacco cells lacking alternative oxidase. Plant Physiol 121: 1309–1320 Pastore D, Fratianni A, Di Pede D, Passarella S (2000) Effects of fatty acids, nucleotides and reactive oxygen species on durum wheat mitochondria. FEBS Lett 470: 88–92 Pastore D, Trono D, Laus MN, Di Fonzo N, Passarella S (2001) Alternative oxidase in durum wheat mitochondria: activation by pyruvate, hydroxypyruvate and glyoxylate and physiological role. Plant Cell Physiol 42: 1373–1382 Potter FJ, Wiskich JT, Dry IB (2001) The production of an inducible antisense alternative oxidase (Aox1a) plant. Planta 212: 215–221 Purvis AC (2001) Reduction of superoxide production by mitochondria oxidizing NADH in the presence of organic acids. J Plant Physiol 158: 159–165 Purvis AC, Shewfeldt RL (1993) Does the alternative pathway ameliorate chilling injury in sensitive plant tissues? Physiol Plant 88: 712–718 Raghavendra AS, Padmasree K (2003) Beneficial interactions of mitochondrial metabolism with photosynthetic carbon assimilation. Trends Plant Sci 8: 546–553 Rhoads DM, Umbach AL, Sweet CR, Lennon AM, Rauch GS, Siedow JN (1998) Regulation of the cyanide-resistant alternative oxidase of plant mitochondria: identification of the cysteine involved in a-keto acid stimulation and intersubunit disulfide bond formation. J Biol Chem 273: 30750–30756 Ribas-Carbo M, Aroca R, Gonza`lez-Meler MA, Irigoyen JJ, Sanchez-Diaz M (2000) The electron partitioning between the cytochrome and alternative respiratory pathways during chilling recovery in two cultivars of maize differing in chilling sensitivity. Plant Physiol 122: 199–204 Robinson SA, Ribas-Carbo M, Yakir D, Giles L, Reuveni Y, Berry JA (1995) Beyond SHAM and cyanide: opportunities for studying the alternative oxidase in plant respiration using oxygen isotope discrimination. Aust J Plant Physiol 22: 487–496 Robson CA, Vanlerberghe GC (2002) Transgenic plant cells lacking mitochondrial alternative oxidase have increased susceptibility to mitochondria-dependent and -independent pathways of programmed cell death. Plant Physiol 129: 1908–1920 Rohde A, Morreel K, Ralph J, Goeminne G, Hostyn V, De Rycke R, Kushnir S, Van Doorsselaere J, Joseleau J-P, Vuylsteke M, et al (2004) Molecular phenotyping of the pal1 and pal2 mutants of Arabidopsis thaliana reveals far-reaching consequences on phenypropanoid, amino acid, and carbohydrate metabolism. Plant Cell 16: 2749–2771 Sabar M, De Paepe R, de Kouchkovsky Y (2000) Complex I impairment, respiratory compensations, and photosynthetic decrease in nuclear and mitochondrial male sterile mutants of Nicotiana sylvestris. Plant Physiol 124: 1239–1249 Saisho D, Nambara E, Naito S, Tsutsumi N, Hirai A, Nakazono M (1997) Characterization of the gene family for alternative oxidase from Arabidopsis thaliana. Plant Mol Biol 35: 585–596 Schluepmann H, Pellny T, van Dijken A, Smeekens S, Paul M (2003)

1820

Trehalose 6-phosphate is indispensable for carbohydrate utilization and growth in Arabidopsis thaliana. Proc Natl Acad Sci USA 100: 6849–6854 Simons BH, Millenaar FF, Mulder L, Van Loon LC, Lambers H (1999) Enhanced expression and activation of the alternative oxidase during infection of Arabidopsis with Pseudomonas syringae pv tomato. Plant Physiol 120: 529–538 Smith AMO, Ratcliffe RG, Sweetlove LJ (2004) Activation and function of mitochondrial uncoupling protein in plants. J Biol Chem 279: 51944–51952 Storey JD (2003) The positive false discovery rate: a Bayesian interpretation and the q-value. Ann Statist 31: 2013–2035 ˚ S, Rasmusson AG (2001) Light-dependent gene expression for Svensson A proteins in the respiratory chain of potato leaves. Plant J 28: 73–82 Taji T, Ohsumi C, Iuchi S, Seki M, Kasuga M, Kobayashi M, YamaguchiShinozaki K, Shinozaki K (2002) Important roles of drought- and coldinducible genes for galactinol synthase in stress tolerance in Arabidopsis thaliana. Plant J 29: 417–426 Tarpey MM, Fridovich I (2001) Methods of detection of vascular reactive species: nitric oxide, superoxide, hydrogen peroxide, and peroxynitrite. Circ Res 89: 224–236 Taylor NL, Day DA, Millar AH (2002) Environmental stress causes oxidative damage to plant mitochondria leading to inhibition of glycine decarboxylase. J Biol Chem 277: 42663–42668 Taylor NL, Rudhe C, Hulett JM, Lithgow T, Glaser E, Day DA, Millar AH, Whelan J (2003) Environmental stresses inhibit and stimulate different protein import pathways in plant mitochondria. FEBS Lett 547: 125–130 Thirkettle-Watts D, McCabe TC, Clifton R, Moore C, Finnegan PM, Day DA, Whelan J (2003) Analysis of the alternative oxidase promoters from soybean. Plant Physiol 133: 1158–1169 Umbach AL, Siedow JN (1993) Covalent and noncovalent dimers of the cyanide-resistant alternative oxidase protein in higher plant mitochondria and their relationship to enzyme activity. Plant Physiol 103: 845–854 Vanlerberghe GC, Day DA, Wiskich JT, Vanlerberghe AE, McIntosh L (1995) Alternative oxidase activity in tobacco leaf mitochondria: dependence on tricarboxylic acid cycle-mediated redox regulation and pyruvate activation. Plant Physiol 109: 353–361 Vanlerberghe GC, McIntosh L, Yip JYH (1998) Molecular localization of a redox-modulated process regulating plant mitochondrial electron transport. Plant Cell 10: 1551–1560 Vanlerberghe GC, Vanlerberghe AE, McIntosh L (1994) Molecular genetic alteration of plant respiration: silencing and overexpression of alternative oxidase in transgenic tobacco. Plant Physiol 106: 1503–1510 Vanlerberghe GC, Vanlerberghe AE, McIntosh L (1997) Molecular genetic evidence of the ability of alternative oxidase to support respiratory carbon metabolism. Plant Physiol 113: 657–661 Wantanabe A, Nakazono M, Tsutsumi N, Hirai A (1999) AtUCP2: a novel isoform of the mitochondrial uncoupling protein of Arabidopsis thaliana. Plant Cell Physiol 40: 1160–1166 Watkinson JI, Sioson AA, Vasquez-Robinet C, Shukla M, Kumar D, Ellis M, Heath LS, Ramakrishnan N, Chevone B, Watson LT, et al (2003) Photosynthetic acclimation is reflected in specific patterns of gene expression in drought-stressed loblolly pine. Plant Physiol 133: 1702–1716 Wolfinger RD, Gibson G, Wolfinger ED, Bennett L, Hamadeh H, Bushnel P, Afshari C, Paules RS (2001) Assessing gene significance from cDNA microarray expression data via mixed models. J Comput Biol 8: 625–637

Plant Physiol. Vol. 139, 2005