Arabidopsis thaliana overexpressing glycolate oxidase in chloroplasts

Abstract

Reactive oxygen species (ROS) represent both toxic by-products of aerobic metabolism as well as signaling molecules in processes like growth regulation and defense pathways. The study of signaling and oxidative-damage effects can be separated in plants expressing glycolate oxidase in the plastids (GO plants), where the production of H₂O₂ in the chloroplasts is inducible and sustained perturbations can reproducibly be provoked by exposing the plants to different ambient conditions. Thus, GO plants represent an ideal non-invasive model to study events related to the perception and responses to H₂O₂ accumulation. Metabolic profiling of GO plants indicated that under high light a sustained production of H₂O₂ imposes coordinate changes on central metabolic pathways. The overall metabolic scenario is consistent with decreased carbon assimilation, which results in lower abundance of glycolytic and tricarboxylic acid cycle intermediates, while simultaneously amino acid metabolism routes are specifically modulated. The GO plants, although retarded in growth and flowering, can complete their life cycle indicating that the reconfiguration of the central metabolic pathways is part of a response to survive and thus, to adapt to stress conditions imposed by the accumulation of H₂O₂ during the light period.

Reactive oxygen species (ROS) are key molecules in the regulation of plant development, stress responses and programmed cell death. Depending on the identity of ROS species or its subcellular production site, different cellular responses are provoked.¹ To assess the effects of metabolically generated H₂O₂ in chloroplasts, we have recently generated Arabidopsis plants in which the peroxisomal GO was targeted to chloroplasts.² The GO overexpressing plants (GO plants) show retardation in growth and flowering time, features also observed in catalase, ascorbate peroxidase and MnSOD deficient mutants.^3–5 The analysis of GO plants indicated that H₂O₂ is responsible for the observed phenotype. GO plants represent an ideal non-invasive model system to study the effects of H₂O₂ directly in the chloroplasts because H₂O₂ accumulation can be modulated by growing the plants under different ambient conditions. By this, growth under low light or high CO₂ concentrations minimizes the oxygenase activity of RubisCO and thus the flux through GO whereas the exposition to high light intensities enhances photorespiration and thus the flux through GO.

Here, we explored the impact of H₂O₂ production on the primary metabolism of GO plants by assessing the relative levels of various metabolites by gas chromatography coupled to mass spectrometry (GC-MS)⁶ in rosettes of plants grown at low light (30 µmol quanta m⁻² s⁻¹) and after exposing the plants for 7 h to high light (600 µmol quanta m⁻² s⁻¹). The results obtained for the GO5 line are shown in Table 1 as representative of three independent lines (GO5, −16 and −20).

Table 1.

Relative metabolite levels determined by GC-MS in rosette of plants grown at normal atmospheric CO₂ concentration (380 ppm)

	After 1 h at 30 µE	After 7 h at 600 µE
Alanine	0.88 ± 0.05	2.83 ± 0.68
Asparagine	1.39 ± 0.12	3.64 ± 0.21
Aspartate	0.88 ± 0.03	1.65 ± 0.10
GABA	1.14 ± 0.05	1.13 ± 0.05
Glutamate	0.97 ± 0.04	1.51 ± 0.07
Glutamine	1.06 ± 0.11	1.87 ± 0.06
Glycine	1.23 ± 0.07	0.30 ± 0.02
Isoleucine	3.52 ± 0.40	3.00 ± 0.15
Leucine	1.36 ± 0.22	0.57 ± 0.06
Lysine	1.49 ± 0.13	0.38 ± 0.02
Methionine	0.96 ± 0.05	4.54 ± 0.51
Phenylalanine	0.95 ± 0.03	0.94 ± 0.04
Proline	1.32 ± 0.22	1.60 ± 0.13
Serine	1.05 ± 0.04	1.49 ± 0.15
Threonine	4.74 ± 0.17	5.51 ± 0.34
Valine	0.91 ± 0.13	0.29 ± 0.02
Citrate/Isocitrate	0.65 ± 0.02	0.64 ± 0.02
2-oxoglutarate	0.95 ± 0.11	0.76 ± 0.05
Succinate	0.78 ± 0.04	0.72 ± 0.02
Fumarate	0.64 ± 0.03	0.31 ± 0.01
Malate	0.74 ± 0.03	0.60 ± 0.02
Pyruvate	1.19 ± 0.28	0.79 ± 0.04
Ascorbate	1.13 ± 0.14	2.44 ± 0.45
Galactonate-γ-lactone	1.81 ± 0.40	1.62 ± 0.28
Fructose	1.20 ± 0.13	0.37 ± 0.01
Glucose	1.38 ± 0.17	0.30 ± 0.01
Mannose	0.90 ± 0.27	1.34 ± 0.28
Sucrose	1.04 ± 0.07	0.49 ± 0.02
Fructose-6P	0.82 ± 0.15	1.20 ± 0.15
Glucose-6P	0.87 ± 0.06	1.25 ± 0.18
3-PGA	1.13 ± 0.11	0.35 ± 0.02
DHAP	1.38 ± 0.09	1.26 ± 0.08
Glycerate	0.99 ± 0.04	0.67 ± 0.01
Glycerol	1.07 ± 0.04	1.12 ± 0.05
Shikimate	1.18 ± 0.04	0.35 ± 0.01
Salicylic acid	1.04 ± 0.18	0.66 ± 0.18

Plants were grown at 30 µmol m⁻² sec⁻¹ (30 µE). The samples were collected 1 h after the onset of the light period and after 7 h of exposure to 600 µmol m⁻² sec⁻¹ (600 µE), respectively. The values are relative to the respective wild-type (each metabolite = 1) and represent means ± SE of four determinations of eight plants. (*) indicates the value is significantly different from the respective wild-type as determined by the Student's t test (p < 0.05).

At the beginning of the light period in low light conditions, some significant deviations in the levels of metabolites tested were observed in GO plants when compared to the wild-type (Table 1). This indicated that although GO plants did not show an aberrant phenotype in this condition,² the transgenic GO activity is sufficient to induce a characteristic metabolic phenotype (Fig. 1). The levels of the tricarboxylic acid (TCA) cycle intermediates, citrate/isocitrate, succinate, fumarate and malate were lower in the GO plants (Table 1). This depletion of TCA cycle intermediates could be due to sustained inhibition of the TCA cycle enzyme aconitase, which has been shown to undergo inactivation by H₂O₂.⁷ In consequence, OAA might not freely enter the TCA cycle and is redirected to the synthesis of Lys, Thr and Ile, which accumulate in the GO plants (Table 1).

Figure 1.

Simplified scheme of the primary metabolism showing the qualitative variations in metabolite abundance in GO plants obtained by GC-MS analysis (Table 1). The results of starch content were taken from Fahnenstich et al.² Blue boxes indicate a significant increase in the content of the particular metabolite compared to the wild-type, while red boxes indicate a significant decrease. Metabolites without boxes have not been determined. The arrows do not always indicate single steps. Adapted from Baxter et al., 2007.

High light treatment induced massive changes in the metabolic profile of GO plants (Table 1 and Fig. 1). The OAA-derived amino acids Asp, Asn, Thr, Ile and Met as well as the 2-oxoglutarate-derived amino acids Glu and Gln accumulated. On the contrary, the levels of the Pyr-derived amino acids Val and Leu and the OAA-derived amino acid Lys decreased. A rational explanation for these metabolic changes is difficult to assess, but these changes could be a consequence of a metabolic reconfiguration in response to high light leading to required physiological functions and thus ensuring continued cellular function and survival, e.g., production of secondary metabolites to mitigate photooxidative damage. The higher levels of Glu observed in the GO plants could be attributed to alternative pathways of glyoxylate metabolism that may occur during photorespiration.⁸ It has been shown earlier that isocitrate derived from glyoxylate and succinate is decarboxylated by cytosolic isocitrate dehydrogenase producing 2-oxoglutarate and further glutamate.⁸

In GO plants grown under low light conditions (minimized photorespiratory conditions), the levels of Gly were similar to those of the wild-type whereas, after exposure to high light (photorespiratory conditions), the Gly levels were extremely low, indicating that the GO activity diverts a significant portion of flux from the photorespiratory pathway (Table 1).

The levels of the intermediates of the TCA cycle were consistently lower in GO plants (Table 1). The oxidative inactivation of aconitase can explain this results⁷ and also the levels of the lipoic acid-containing subunits of the pyruvate- and 2-oxoglutarate dehydrogenases were shown to be significantly reduced under oxidative stress conditions.^9,10 Similarly, the contents of the soluble sugars sucrose, fructose and glucose and those of 3-PGA and glycerate were lower. In addition, the GO plants showed an impairment in the accumulation of starch under high light conditions, a feature that was not observed if the plants were grown under non-photorespiratory conditions.²

Together, these results indicate that the low photosynthetic carbon assimilation in the GO plants exposed to high light is most probably due to enhanced photoinhibition,² the repression of genes encoding photosynthetic components by H₂O₂,^11–13 and the direct damage or inhibition of enzyme activities involved in CO₂ assimilation and energy metabolism by H₂O₂.^7,10,14,15 Moreover, Scarpeci and Valle¹³ showed that in plants treated with the superoxid anion radical producing methylviologen (MV) most of the genes involved in phosphorylytic starch degradation, e.g., the trioseP/Pi translocator and genes involved in starch and sucrose synthesis were repressed, while genes involved in hydrolytic starch breakdown and those involved in sucrose degradation were induced. In line with this, the contents of carbohydrates were also lower in MV-treated plants. Together, these observations can also explain the lower growth rates of the GO plants in conditions where the oxygenase activity of RubisCO becomes important and thus, the flux through GO increases.²

The levels of shikimate were lower in GO plants (Table 1). Like all other flavonoids, anthocyanins are assembled from two different metabolic branches involving on the one hand the shikimate pathway producing the aromatic amino acid phenylalanine and further p-cumaric acid. The other branch produces three molecules of malonyl-CoA. Both pathways meet and are coupled together by the enzyme chalcone synthase. In this way, the extremely retarded production of anthocyanins observed in the GO plants could be due to the combination of the repression of anthocyanin biosynthetic genes^2,16 and the low levels of substrates available, as anthocyanins are ultimately synthesized from photosynthates and the GO plants showed a diminished photosynthetic performance.²

As expected, the levels of ascorbate and its precursor, galactonate-γ-lactone, were enhanced in the GO plants clearly showing the activation of the cellular antioxidant machinery (Table 1).

The fact that the GO plants, although retarded in growth, can complete their life cycle indicates that the reconfiguration of the central metabolic pathways is part of the response to adapt and thereby survive the stress situation imposed by the accummulation of H₂O₂ during the light period. Recently, Baxter et al.,¹⁰ described the metabolic response to oxidative stress of heterotrophic Arabidopsis cells treated with menadione, which also generates superoxide anion radicals. This oxidative stress was shown to induce metabolic inhibition of flux through the TCA cycle and sectors of amino acid metabolism together with a diversion of carbon into the oxidative pentose phosphate pathway.

Signaling and oxidative-damage effects are difficult to separate by manipulating the enzymes of antioxidant systems. In this regard, the GO plants represent a challenging inducible model that avoid acclimatory and adaptative effects. Moreover, it is possible to control the H₂O₂ production in the chloroplasts of GO plants without inducing oxidative damage by changing the conditions of growth.² Further exploration of metabolic changes imposed by different ROS at the cellular and whole organ levels will allow to address many intriguing questions on how plants can rearrange metabolism to cope with oxidative stresses.

Abbreviations

GC-MS

gas chromatography-mass spectrometry

GO

glycolate dehydrogenase

H₂O₂

hydrogen peroxide

ROS

reactive oxygen species

TCA

tricarboxylic acid

Addendum to: Fahnenstich H, Scarpeci TE, Valle EM, Flügge U-I, Maurino VG. Generation of H2O2 in chloroplasts of Arabidopsis thaliana overexpressing glycolate oxidase as an inducible system to study oxidative stress. Plant Physiol. 2008;148:719–729. doi: 10.1104/pp.108.126789.