Differential response of antioxidant genes in maize leaves exposed to ...

dRR17

23/7/99

1:43 pm

Page 95

Research article

Differential response of antioxidant genes in maize leaves exposed to ozone Stephanie M. Ruzsa, Photini Mylona, John G. Scandalios Department of Genetics, North Carolina State University, Raleigh, North Carolina, USA

Antioxidant enzymes function to eliminate reactive oxygen species (ROS) produced as a consequence of normal metabolic functions as well as environmental stress. In these studies, the responses of catalase (Cat), superoxide dismutase (Sod) and glutathione S-transferase (Gst), as well as D-ribulose1,5-bisphosphate carboxylase/oxygenase (RbcS) genes were analyzed in 9- and 15-day postimbibition maize seedlings exposed to various ozone (O3) concentrations and time periods. After a single (acute) 6 h exposure, or 3, 6 and 10 consecutive days (chronic) exposure to O3, Cat1, Cat3, Gst1, Sod3, Sod4 and Sod4A transcript levels generally increased, while Cat2, RbcS and Sod1 levels decreased. Such changes in mRNA levels do not necessarily reflect parallel changes in the protein products of these genes. Changes in transcript levels seemed to be correlated with the spatial location of the isozymes encoded by the genes. The results are discussed with respect to gene regulation and expression, and the localization and function of these antioxidant enzymes during ozone-mediated oxidative stress.

INTRODUCTION Due to their immobility, plants are constantly exposed to a changing environment, requiring them to adapt in various ways in order to survive. Appropriate adjustments in gene expression must occur to maintain the biochemical and physiological state most conducive to growth and productivity. Many environmental stresses, both biotic and abiotic, are thought to affect plant growth and productivity by causing an over-accumulation of reactive oxygen species (ROS), such as the superoxide (O2•–) and hydroxyl (•OH) radicals, as well as hydrogen peroxide (H2O2). These stresses include pathogen attack, excessive cold or heat, drought, wounding, exposure to UV radiation or heavy metals, and air pollutants, such as sulfur dioxide, nitric oxide and ozone.1 Ozone (O3) is a very powerful oxidant, and the major component of air pollution in the form of smog. Consequently, its biological effects have long been studied.2 More recently, studies have been initiated to determine the Received 9 April 1999 Accepted 8 May 1999 Correspondence to: Dr John G. Scandalios, Department of Genetics, North Carolina State University, Raleigh, NC 27695-7614, USA Tel: +1 919 515 7079; Fax: +1 919 515 3355 E-mail: [email protected] Redox Report, Vol. 4, No. 3, 1999

biochemical and molecular mechanisms of ozone toxicity. The results of these studies have been reviewed periodically, and various modes of action and signal transduction pathways have been proposed.2–5 The detrimental effects of O3 are ultimately due to the production of excess quantities of ROS, especially O2•–, • OH and H2O2, through various chemical reactions.6 These ROS, and their products, can readily react with proteins, DNA and membrane lipids to cause reduced photosynthesis, increased respiration, reduced transpiration, electrolyte leakage, gene mutations and accelerated cell senescence.3,7 Plants have developed various defenses, both enzymatic and non-enzymatic, against the toxic effects of ROS. Non-enzymatic antioxidants include low molecular weight scavengers, such as polyamines, glutathione, ascorbate, phenolics and α-tocopherol.2,4 Enzymatic defenses include glutathione reductase (GR), superoxide dismutase (SOD), catalase (CAT), glutathione S-transferase (GST), and ascorbate peroxidase (APX).1,2,8 These defense systems act to prevent the formation of and/or to scavenge ROS. Breakdown of O3 to H2O2 and subsequent reactions involving H2O2 are thought to be the major source of ROS in organisms upon exposure to O3.9 SOD catalyzes the dismutation of O2•– to H2O2,10 further increasing the levels of H2O2. Removal of both H2O2 and O2•– is critical, since © W. S. Maney & Son Ltd

dRR17

23/7/99

96

1:43 pm

Page 96

Ruzsa, Mylona, Scandalios

they can interact to form the highly reactive •OH, which is capable of reacting with biomolecules to cause deleterious effects and, consequently, cell death.7,11 CAT prevents this interaction by catalyzing the breakdown of H2O2 to water and molecular oxygen.12 Thus the combined action of SOD and CAT is critical in modulating the oxidation state of the cell, and in mitigating the effects of oxidative stress. Glutathione S-transferase (GST) is another enzyme considered to play an important role in counteracting oxidative stress induced by O3. GST may be involved in the removal of the toxic products of O3-initiated lipid peroxidation by catalyzing the conjugation of the tripeptide glutathione (GSH) to electrophilic molecules, which are then transported out of the cytosol, or to the vacuole, via glutathione pumps.13,14 For organisms to be able to withstand periods of increased oxidative stress, an increase in antioxidant enzyme activities might be expected. A number of studies have documented increases in the activities of several of these enzymes after exposure to ozone, such as SOD, APX, dihydroascorbate reductase (DHAR), GR, CAT and GST.15,16 Other studies have reported contradictory or opposite results.4,17 These variable responses may be due to the diverse plant species and cultivars used, the developmental stage of the tissues examined, the specific enzymes or isozymes analyzed, and the concentration and duration of O3 exposure, as well as the timing of sampling and measurement. Recently, transgenic plants have also been used to study the responses and effects of antioxidant enzymes to O3, through overexpression of the corresponding genes. It was reported that transgenic tobacco plants, overexpressing the chloroplastic Cu/ZnSOD of petunia, did not exhibit any reduction in deleterious symptoms after O3 fumigation.18,19 Other studies, however, have shown increased tolerance to O3 with the overexpression of certain enzymes. Transgenic tobacco plants, with elevated levels of MnSOD targeted to the chloroplast, exhibited some increased protection,20 and tobacco plants overexpressing GR were also more resistant to O3 damage than non-transformed plants.21 Based on the results of these studies, overexpression of certain antioxidant genes seems to confer some increased O3 tolerance in some instances, and corroborates the importance of antioxidant systems in the removal of excess ROS generated upon exposure to O3. At present, the specific mechanisms that modulate the expression of antioxidant genes during times of oxidative stress are little understood, especially of those genes encoding various isozymes of the same protein. In an effort to increase our understanding of the regulation of these genes under stress, this study examines and analyzes the effects of O3 on young maize leaves. The CAT and SOD gene–enzyme systems in maize afford an ideal opportunity to study the effects of ozone on a well-characterized system of antioxidant genes encoding multiple

isozymes, which are highly regulated and localized during plant development.1 Gene-specific probes are now available to accurately measure the responses of the genes at the transcript level. Various maize high and low CAT and SOD expression mutants are also available to further analyze individual gene responses. The response of GST genes to O3 has been examined in other species, but, to our knowledge, not yet in maize. In maize, three CAT isozymes are encoded by three unlinked structural genes Cat1, Cat2 and Cat3. CAT-1 and CAT-2 are found in peroxisomes and glyoxysomes, as well as the cytosol. CAT-3 is associated with mitochondria. The expression of each gene is tissue- and developmental stage-specific.12 Maize SOD consists of nine distinct isozymes encoded by Sod1, Sod2, Sod3.1, Sod3.2, Sod3.3, Sod3.4, Sod4, Sod4A and Sod5. SOD-1 is a Cu/Zn enzyme associated with chloroplasts. SOD-2, SOD-4, SOD-4A and SOD-5, also Cu/Zn enzymes, are cytosolic, while the MnSOD-3, encoded by the multigene Sod3 family, is localized in the mitochondria.10,22 Maize glutathione S-transferase is represented by four distinct enzymes, GST I, GST II, GST III, and GST IV, that are members of the type I GSTs, which respond to oxidative stress and the resulting lipid peroxidation.14 Like SOD-1, D-ribulose1,5-bisphosphate carboxylase/oxygenase (Rubisco) is a chloroplastic enzyme, which is involved in photosynthetic carbon fixation. Rubisco protein was inactivated by O3 in potato cultivars23 and RbcS mRNA levels have been found to decline in response to O3 exposure in Arabidopsis and potato.16,24 In early reports from this laboratory, no significant change in total CAT or SOD enzyme activity was observed after 8 h exposure to O3, with concentrations ranging from 200 to 600 ppb.25 With the availability of gene-specific probes to examine the responses at the transcript level, O3 experiments on maize seedlings have been repeated. In this two-part study, the effects of ozone on expression of Cat, Sod, Gst and RbcS genes are documented. In initial experiments, maize seedlings of two different ages, 9- and 15-day postimbibition, were exposed to various acute and chronic doses of O3. In the second part, 9-day postimbibition seedlings of six maize lines with variant CAT and SOD expression were treated with four O3 concentrations for a single (acute) exposure period. This study provides further insights relative to the differential regulation of these genes in response to O3mediated environmental stress in maize.

MATERIALS AND METHODS Plant materials and growth conditions Maize seeds were surface-sterilized in 20% Clorox for 10 min., rinsed, then soaked 20–24 h in de-ionized

dRR17

23/7/99

1:43 pm

Page 97

Response of antioxidant genes in maize leaves exposed to ozone 97 water. The seeds were planted in germination trays containing Metro-Mix 200 (Scotts-Sierra Horticultural Products, Marysville, OH, USA). Germination and growth of seedlings was carried out in the NCSU phytotron facility under a 12 h photoperiod: 8:00 am–8:00 pm (500 µm–2s–1 PAR), in controlled-growth chambers at 25°C and 70–80% relative humidity, for initial experiments, and in continuously stirring tank reactors (CSTRs) for subsequent treatments. Conditions in the CSTRs – light, temperature and humidity – were maintained at the levels of those in the growth chambers. An automated system providing de-ionized water was used to maintain soil moisture. The inbred maize line W64A, maintained by this laboratory, was used for initial tests of acute and chronic O3 doses, and as a control for subsequent experiments. Acute O3 exposures were repeated with five additional maize lines. CAT variant lines nearisogenic with W64A included WA10C (CAT-2 null), WI10D (CAT-3 null) and WDN7 (null for both CAT-2 and CAT-3). Two additional maize lines were also included: A350 (low in Cu/ZnSOD expression; normal CAT) and A352 (low Cu/ZnSOD; CAT-2 null).

nitrogen. To minimize plant-to-plant variation, 4–6 plants per treatment were sampled and mixed. Total RNA was isolated,26 separated on 1.6% denaturing agarose gels at 20 µg per sample, and transferred to nylon (Schleicher and Schuell Nytran Plus) membranes. The blots were sequentially hybridized in modified Church buffer,27 containing 7% SDS, 1 mM EDTA, 0.25 M NaH2PO4 and 1% BSA, with [α-32P]-labeled gene-specific DNA probes (gsp) for Cat1, Cat2, Cat3, Sod1, Sod4, Sod4A, Gst1 (isolated in this laboratory1) and RbcS (kindly provided by Tim Nelson, Yale University). Full-length (fl) probes were used for the multigene Sod3 family28 and for Cat2. The Cat2 fl probe detects the normal Cat2 gene transcript, as well as the shortened transcript of the CAT-2 null lines, whereas the Cat2 gsp detects only the normal Cat2 transcript.29 Hybridized membranes were exposed to Kodak X-omat XAR film, with intensifying screens, at –80°C to produce autoradiograms. After each analysis, probes were removed from the filters by repeated washes in boiling 0.1% SDS/0.1X SSC (0.15 M NaCl, 0.0015 M sodium citrate, pH 7.0). Blots were finally probed with a DNA fragment from clone pHA2, containing an 18S ribosomal sequence.30

O3 exposures RESULTS Initially, 9- and 15-day postimbibition seedlings of W64A were exposed to acute and chronic doses of O3, at various concentrations. Acute exposure consisted of a single 6 h fumigation, 9:00 am–3:00 pm, at concentrations of 0, 100, 200, 300, 500 or 1000 ppb O3. For chronic exposure, seedlings were treated 6 h per day, as above, for 3, 6 or 10 consecutive days, at 0, 100, 200 or 300 ppb O3. Fumigations were carried out in the CSTRs. O3 was produced from dry O2 by the electric discharger of a Griffith Technics Corporation (Lodi, NJ, USA) O3 generator and was delivered to the CSTR chambers via Teflon tubing. The concentrations in the chambers were monitored on a continuous basis using an Ultraviolet Photometric Ambient O3 Model 47 Analyzer (Thermo Environmental Instruments, Inc., Franklin, MA, USA) and a Keithley data acquisition system (Keithley Data Acquisition and Control, Cleveland, OH, USA), which compared actual O3 concentrations in the chambers to target concentrations. The mass flow controllers were adjusted accordingly. O3 levels were held to within ± 5% during fumigations. Acute exposures were repeated as above using the maize variants (at 9-day postimbibition), at concentrations of 0, 300, 600 or 950 ppb O3. RNA extraction and analysis Leaves were harvested between 3:30–4:00 pm, immediately following O3 treatments, and quick-frozen in liquid

For all of the genes examined, the responses to a given O3 exposure, whether acute or chronic, showed no significant differences in the 9- and 15-day old W64A seedlings treated initially. These results will, therefore, be presented and discussed without differentiating between the two developmental stages. Seedlings exhibited no visible damage from O3 levels below 600 ppb. Visible injury, in the form of chlorosis and/or lesions was noted in plants exposed to concentrations of 600–1000 ppb, with no differences noted among the maize variant lines. Plants exposed to chronic doses of 300 ppb O3 for 6 days or longer experienced some growth retardation (data not shown), but no other visible injury, while plants exposed to lower doses exhibited no physical changes.

Effects of acute ozone exposure Levels of gene transcripts for Cat1, Cat3, Gst1, Sod3, Sod4 and Sod4A generally increased, while Cat2 and RbcS levels decreased, in response to increasing levels of O3. Sod1 levels also decreased, but much less noticeably. None of the transcripts exhibited a significant change at low (< 200 ppb) O3 concentrations (Figs 1 & 2). As the concentration rose from 200 to 1000 ppb, Cat1, Gst1 and Sod4 levels increased steadily, while Cat3 and Sod3 levels rose more abruptly, at 200–300 ppb, with less marked changes thereafter (Figs 1 & 2).

dRR17

23/7/99

98

1:44 pm

Page 98


Fig. 1. Response to acute ozone exposure of the Cat, Gst1, and RbcS genes in maize line W64A. Plants were fumigated for a single 6 h period. Leaves were harvested immediately after fumigation. Total RNA was extracted, separated on 1.6% agarose gels, at 20 µg per lane, and transferred to membranes. Probing was done with [α-32P]-labeled gene-specific DNA fragments.

Fig. 2. Response of Sod genes to acute ozone exposure in maize line W64A. Plants and samples were handled as in Figure 1. The 18S rRNA was used as a loading control in all experiments reported, but is shown only in Figures 2, 4 and 6.

Sod4A transcript levels changed less in response to O3 levels of 100–500 ppb than Sod4, while increasing significantly at 600–1000 ppb (Fig. 2). Cat2 and RbcS transcript levels decreased at 300 ppb, and continued decreasing to O3 concentrations up to 1000 ppb (Fig. 1). Sod1 exhibited a very small, but steady, decrease from O3 concentrations of 300–1000 ppb. Its response to O3 was the least of all the transcripts examined (Fig. 2). Among the maize variant lines, Cat1 and Cat3 transcript levels were slightly lower, overall, in maize line A352, though also rising with increasing O3. In addition, a significant increase in Cat3 levels did not occur in this line until higher levels (950 ppb) of O3 were reached (Fig. 3). Levels of both the normal and shortened Cat2 transcripts decreased to equal extents in response to O3, except in A350 where the decrease in the Cat2 transcript is less marked. Control levels in all variant lines were higher than in W64A (Fig. 3). RbcS transcript levels were

higher overall in both low SOD lines, especially in A350, than in the normal SOD lines, but also exhibited the same decline with higher O3 concentrations as in the other lines (Fig. 3). Sod 4 and Sod4A RNA levels were much lower in both A350 and A352 at all concentrations of O3, although the trend of increasing RNA levels with rising O3 concentration is still evident in A352 for Sod4 (Fig. 4). In A350, Sod4 and Sod4A transcripts increased to 600 ppb, then slightly decreased at 950 ppb (Fig. 4). Sod3 levels increased the least in line A352, and also to a lesser extent in WI10D and A350 than in other lines (Fig. 4). Sod3 represents a multigene family,22 where four different transcripts, corresponding to Sod3.1, Sod3.2, Sod3.3 and Sod3.4 are detected by the full-length Sod3 probe. The Sod3 bands, while comprising four individual transcripts, can only be separated into two distinct doublet bands by the techniques used here. The upper and lower doublet bands reacted to O3 somewhat differently in each of the

Fig. 3. Response of Cat and RbcS genes to acute ozone exposure in six variant maize lines. Plants and samples were handled as in Figure 1. W64A, standard maize CAT line; WA10C, CAT-2 null; WI10D, CAT-3 null; WDN7, CAT-2 & CAT-3 double null; A350, low SOD/normal CAT; A352, low SOD/CAT-2 null. Cat2 probe used was full-length which can also identify the aberrant shortened Cat2 transcript in the CAT-2 nulls.

dRR17

23/7/99

1:44 pm

Page 99

Response of antioxidant genes in maize leaves exposed to ozone 99

Fig. 4. Response of Sod genes to acute ozone exposure in six variant maize lines (as in Figure 3). Plants and samples were handled as in Figure 1. 18S rRNA is loading control.

maize variant lines (Fig. 4). Sod1 mRNA showed no discernible differences between W64A and the CAT or SOD variant lines.

Changes in gene expression in response to chronic ozone exposure Cat1, Cat3, Gst1, Sod3, Sod4 and Sod4A transcript levels generally increased, while Cat2, RbcS, and Sod1 levels again decreased, with increasing concentrations of O3, whether after 3, 6 or 10 days of repeated exposure (Figs 5 & 6). Cat1 and Sod4A gene transcript levels were slightly, if at all, affected at 100 ppb O3, with their respective responses becoming more significant at 200 and 300 ppb (Figs 5 & 6). After 10 consecutive days of O3 exposure, the increase over control levels of Cat1, Cat3, Gst1 and Sod3 was less marked, whereas Sod4 and Sod4A maintained a gradual increase in response to rising O3 levels throughout the 10-day period. The decrease in Sod1 mRNA is almost undetectable after 6- or 10days’ treatment, while Cat 2 and RbcS levels continued decreasing over this time period (Figs 5 & 6).

Fig. 5. Response of Cat, Gst1, and RbcS genes to chronic ozone exposure in maize line W64A. Plants were fumigated for 6 h per day for 3, 6, or 10 consecutive days. Tissues and samples were handled as in Figure 1.

Transcript accumulation increased over time for Cat2, Cat3, Gst1, Sod3, and Sod4 and to a lesser extent for Cat1 and Sod4A, as seen in the increasing levels of these transcripts at 0 ppb O3 from 3–10 days of the treatments (Figs 5 & 6). The amount of RbcS RNA decreased over the same time period (Fig. 5). Sod1 RNA accumulation did not vary significantly over the exposure periods (Fig. 6).

DISCUSSION Differential expression of maize catalase genes in response to ozone In maize, Cat1, Cat2 and Cat3 are nuclear genes which code for the CAT-1, CAT-2 and CAT-3 isozymes, respectively.12 CAT-1 and CAT-2 are located in peroxisomes, glyoxysomes and the cytosol, while CAT-3 is associated with mitochondria. In leaves, CAT-1 and CAT-3 are found in mesophyll cells, while CAT-2 is expressed in bundle sheath cells. The expression of each Cat gene is tissue- and developmental stage-specific. The three maize catalase

Fig. 6. Response of Sod genes to chronic ozone exposure in maize line W64A. Plants and samples were handled as in Figure 1. 18S rRNA is loading control.

dRR17

23/7/99

100

1:44 pm

Page 100


genes share some homology at the nucleotide level, but their promoter regions are unique, each containing a number of transcription factors, which are thought to play a role in their expression and regulation. In contrast to earlier studies,31 in which no detectable changes were reported in Arabidopsis Cat RNA in response to O3, all three maize Cat genes exhibited marked reactions upon exposure to O3. Cat1 transcript levels increased in response to increasing concentrations of O3, after both acute and chronic exposures. As Cat1 is constitutively expressed, the CAT-1 isozyme it encodes may play a key role in the defense against O3-mediated stress, as well as other forms of oxidative stress. The amount of Cat1 RNA has been found to rise in response to salicylic acid.32 Upon treatment with cercosporin, a fungal toxin, Cat1 levels decreased at low levels of the toxin, then increased with exposure to higher concentrations.33 Cat1 levels might have been expected to show enhanced reaction to O3 in the maize lines null for one or two of the other CAT isozymes as compensation, but did not. This may indicate that steady-state levels of the CAT-1 isozyme are already at an optimum for plant survival under the conditions presented here. Cat3 transcript levels also increased upon exposure to O3. Cat3 mRNA levels in maize leaves are regulated by a circadian rhythm,34 possibly in response to fluctuating H2O2 and other ROS levels, in conjunction with normal daily metabolism. A number of cis-acting elements have been identified in the promoter region of the Cat3 gene that are thought to play a role in circadian regulation.35 However, their specific function in this process has not yet been elucidated. Whether or not O3 has any effect on, or whether its effects are dependent on, the circadian rhythm is not clear. This would need to be investigated with O3 fumigation, as well as sampling, done at various times during the light cycle. Leaf Cat3 transcript levels are also increased by wounding in plants grown in constant light, which eliminates the circadian rhythm of Cat3 (unpublished data, this laboratory). This would indicate that at least some of the response by the Cat3 gene to abiotic stresses is independent of the circadian cycle. FSD1, an iron SOD found in the chloroplasts of Arabidopsis, is also under control of a circadian rhythm, but its mRNA levels were reported to decrease in response to O3.36 Cat2 transcripts decreased steadily in response to O3 exposure. One interesting result is that the shortened Cat2 transcript of the CAT-2 null plants29 reacted to this stress, as well as to wounding, H2O2 and paraquat (unpublished data, this laboratory), in the same manner as the normal transcript. This indicates that regulation of the Cat2 responses to these conditions involves the 5′ region of the gene, which is intact in the CAT-2 null lines. Cat1 and Cat3 transcripts reacted similarly in response to O3, while Cat2 exhibited the opposite reaction. As O3 enters the mesophyll through the

stomata,16 increasing Cat1 and Cat3 transcript levels may be part of the plant’s initial defense against rising amounts of ROS. The reactions of Cat1, Cat3 and Cat2 mRNAs in leaves were the same, respectively, in response to wounding (unpublished data, this laboratory) as to O3. The responses to O3 of the maize Cat2 and Cat3 are similar to their respective responses to the fungal toxin cercosporin.33 The similar responses of the maize Cat genes to O3, wounding, and cercosporin, supports the hypothesis that the same signal transduction pathway is triggered by these stresses.2,4,37 The responses to O3 by all three Cat genes are similar, though not identical, to their respective responses to jasmonic acid (JA) (unpublished data, this laboratory). This supports the theory that JA, released by O3-induced lipid peroxidation in cell membranes, may mediate the reactions of antioxidant defenses in response to O3.4 The maize Cat genes, however, have different responses, respectively, to salicylic acid (SA),32 which has also been hypothesized as triggering defense reactions in plants exposed to O3.17 The decrease in Cat2 mRNA and the increases in Cat1 and Cat3 mRNA might account for the lack of change noted by Matters and Scandalios,25 at the enzyme level, as assays done previously had only measured total CAT activity, rather than individual isozyme expression.

Ozone activation of glutathione S-transferase genes In this study, Gst1 transcript levels increased in response to increasing concentrations of O3 in young maize seedlings. Similar results were observed in Arabidopsis thaliana plants.16 Glutathione S-transferases are key enzymes of the detoxification mechanism of the plant cell. GST detoxifies the byproducts of lipid peroxidation by the conjugation of glutathione to activated alkenals.38 As lipid peroxidation is thought to be a major source of O3 injury, by propagating ROS, elevation of Gst transcript levels in response to O3 is likely to be an important part of an organism’s defense. The mechanism by which electrophiles induce Gst genes has been extensively investigated in animals. Recent evidence39 suggests that antioxidant responsive element (ARE) motifs in the promoter region of mammalian Gst genes regulate expression of these genes in the presence of electrophiles and under other conditions that generate oxidative stress. Plant Gst promoters have not been found to contain functional ARE motifs.14 However, we recently demonstrated that the maize Cat genes contain elements that share homology with the ARE elements.12 Further studies may help to elucidate the role these elements play upon exposure to O3.

dRR17

23/7/99

1:44 pm

Page 101

Response of antioxidant genes in maize leaves exposed to ozone 101 Response of maize superoxide dismutase and Rubisco genes to ozone treatments Superoxide dismutases represent another major class of antioxidant enzymes that play an important role in eliminating ROS during oxidative stress. Sod1, encoding the chloroplast-associated Cu/ZnSOD-1 isozyme that is most abundant in leaves, would seem the logical gene to react in defense of the plant in response to O3-induced oxidative stress. However, its mRNA levels decreased, though to a very modest extent, with increasing O3, as did RbcS transcript levels. This is similar to the response of genes encoding various chloroplast-associated antioxidant enzymes and photosynthesis-related proteins in other plant species, such as FeSOD, Cu/ZnSOD, GR, Chl a/b binding protein and Rubisco in Arabidopsis16,36 and tobacco.40 In potato, RbcS and chloroplastic glyceraldehyde-3-phosphate dehydrogenase (gapA and gapB) levels decreased with O3 exposure as well.24 The amount of Sod4 transcript increased steadily with acute exposure to rising O3 concentrations, while Sod4A mRNA did not respond as markedly until higher levels of O3 had been reached (Figs 2 & 4). Previous work has shown that maize Sod4 and Sod4A, encoding almost identical cytosolic Cu/Zn isozymes, share similar gene structure, although the promoter regions of the two genes are very different, with almost no sequence homology.41 This implies that Sod4 and Sod4A may be regulated differently and may, thus, respond differentially to environmental signals. It has already been shown that these two genes respond differentially to light, H2O2, cercosporin, ethephon and SA.41 In contrast to the results of acute O3 exposure, Sod4 and Sod4A showed more similar responses after chronic exposure to lower levels of O3, with only a slightly greater response exhibited by Sod4. These results suggest that Sod4A may play a role in ozone-induced stress chiefly after chronic exposure, possibly as a secondary response, or in response to higher levels of acute stress. This study is the first to report the differential expression, within the same species, of genes encoding two cytosolic Cu/ZnSODs in response to ozone fumigation. Previous studies in Arabidopsis have reported an increase in the transcript levels of cytosolic Cu/ZnSOD with rising levels of O3, before injury to the plant was apparent,16,31 similar to the response of maize Sod4 reported here. The reaction of Sod4A in maize was more like that of the cytosolic Cu/ZnSOD mRNA in tobacco, which rose only with the onset of visible plant injury.40 In light of these reports, a comparison of the promoter regions of genes encoding cytosolic Cu/ZnSODs in these, and other, plant species may provide insights into the differential regulation of these genes, which encode, in many respects, very similar enzymes. Levels of gene transcripts encoding the maize mitochondrial MnSOD, represented by the Sod3 multigene

family, increased in response to acute and chronic O3 exposure. Previous studies25 have shown that MnSOD (cyanide-resistant) enzyme activity was not affected by O3. MnSOD protein in O3-treated Arabidopsis was also unaffected by O3.17 These data, taken together with the results presented here suggest that, while O3 induces higher mRNA levels for the genes encoding MnSODs, this induction does not correlate with an increase in the corresponding isozyme activity. The responses of maize Sod3, Sod4 and Sod4A to O3 are almost identical to their responses to cercosporin,33 again lending support for a similar signal transduction pathway in reaction to these stresses.2,4,37

General observations With the exception of Sod4A, changes in the gene transcript levels examined in this study were noticed at levels of O3 below those needed to produce visible plant injury. This was also the case in Arabidopsis.16 As noted above, Cat1 transcript levels increased steadily with O3 levels. Though CAT-1 is primarily associated with glyoxysomes and peroxisomes, it is also found in the cytosol.42 The increase in mRNA for this isozyme, as well as for GST1, SOD-4 and SOD-4A, is similar to transcript increases reported for other cytosolic antioxidant enzymes, such as APX, GST and Cu/ZnSODs in Arabidopsis.16,31,36 This seems to implicate the cytosol as a major site of ROS removal during O3-mediated stress. The responses of Cat3 and Sod3 were very similar upon O3 exposure. This was also found to be the case in wounding, and in paraquat treatments of post-pollination scutella (unpublished data, this laboratory). Since both genes encode mitochondrial enzymes, this may suggest a common reaction of the mitochondria to these oxidative stresses. These genes have, however, responded differently in post-pollination scutella treated with H2O2 or cercosporin. As no differences have been noted between the response of Cat3 and Sod3 to cercosporin, wounding, and H2O2 in leaf tissue, this may indicate a different role for mitochondrial antioxidant enzymes depending on the plant tissue involved. Other possibilities include varying signal transduction pathways in each tissue, as well as differing sites of primary damage depending on the stress involved.4 Further studies would be needed, possibly imposing the same stresses on a variety of tissues and developmental stages, in order to clarify this issue. It has been suggested that Sod gene transcript levels might not rise in response to stress until the levels of Cat and peroxidase mRNAs were sufficient to cope with the increased levels of H2O2 produced by increased SOD activity.16 No trend of this type was observed upon acute O3 exposure, as levels of Sod4 and Sod3 rose concomitantly with Cat1 and Cat3 levels. However, the increases in

dRR17

23/7/99

102

1:44 pm

Page 102


Cat1 and Cat3 mRNAs did slow after 10 consecutive days’ treatment, while the levels of Sod4 and Sod4A transcripts continued to rise, suggesting that levels of the former transcripts necessary for plant protection against chronic O3 stress are reached before optimum levels of the latter. Cytosolic Gst levels have been reported to increase to a greater extent than other cytosolic antioxidant genes in Arabidopsis.31 This does not seem to be true in this study, as levels of Sod4 and Cat1, as well as Cat3, exhibited increases at least as great as those of Gst1 over control levels of each transcript.

Responses to O3 by the CAT and SOD variant lines The maize lines null for one or two CAT isozymes exhibited no differences from W64A in their responses to O3 exposure, either physiologically or in mRNA levels. As noted above, this may indicate that levels of the constitutively expressed CAT-1 isozyme are sufficient for plant survival under the conditions presented here. The reduction of SOD activity in maize lines A350 and A352, as opposed to the complete absence of one or more CAT isozymes, resulted in varying responses among the genes. However, though the genes’ responses were somewhat different in these lines, the plants themselves did not suffer greater visible injury than the plants with normal levels of SOD. The loss of Rubisco has been associated with senescence, as well as with O3-mediated stress.5 The maize lines with low SOD expression exhibited consistently higher levels of RbcS transcripts at all concentrations of O3. Higher levels of RbcS were also found in transgenic tomato plants that had lower levels of ethylene. Higher ethylene levels after O3 exposure have been correlated with greater foliar damage.37 The release of stress ethylene has also been suspected of inducing antioxidant defenses in response to O3.4 If the presence of higher levels of RbcS in lines A350 and A352 are also an indication of lower ethylene levels, this could explain the comparative health of these plants under O3 stress, i.e. no greater visible injury than to any of the other maize genotypes. While these low SOD lines were no more susceptible to O3-mediated stress than normal lines, this was not found to be the case for another form of oxidative stress: constant light (unpublished data, this laboratory), suggesting that the molecular bases for these stresses are dissimilar. Maize variant lines lacking or low in CAT or SOD isozyme activity were no more O3-sensitive than the normal maize line. Naturally occurring variant lines also exist which have higher than normal expression of CAT or SOD. It remains to be seen whether plants with these phenotypes would be more resistant to O3 stress than the lines examined here.

CONCLUSIONS The maize CAT and SOD gene-enzyme systems provide an ideal opportunity to examine the expression and regulation of genes encoding multiple forms of highly regulated antioxidant enzymes, which are also differentially expressed and localized in various tissues, cell types and subcellular organelles. Similar to responses reported in other plant species, the genes encoding cytosolic and mitochondrial isozymes reacted with raised transcript levels upon exposure to O3, while those encoding the plastidic SOD-1 and peroxisomal/glyoxisomal CAT-2 responded with decreased mRNA levels. The increase of mRNA levels of cytosolic- and mesophyll-targeted isozymes implicates these locations as primary sites for ROS removal upon exposure to O3-mediated stress. The similarities in the responses of the Cat and Sod genes to O3, wounding and cercosporin reinforce the hypothesis of shared signal transduction pathways among these stresses, possibly in response to signaling by JA. The antioxidant Gst1 and the photosynthesis-related Rubisco gene also reacted to O3 in maize as they have been reported to respond in other plant species. The differential responses of the very similar genes, Sod4 and Sod4A, may provide insights into gene regulation upon further examination of the promoter regions of these and possibly other related genes. Since maize variants lacking one or two functional CAT isozymes and/or having low activity of the Cu/ZnSOD isozymes showed no greater physical distress, and few differences in transcript responses, it may be assumed that steady-state levels of the remaining isozymes are adequate for the plants’ survival under these conditions, or that other antioxidant defenses may be compensating for these mutations. These results underscore the complexity of, and highly balanced coordination among, the antioxidant defenses that allow plants to adapt to and survive many forms of environmental oxidative stress. ACKNOWLEDGEMENTS We thank John Dunning and Jeff Barton for assistance with ozone treatments in the NCSU-Phytotron Facility. Research was supported by research grants from the US Environmental Protection Agency to JGS.

REFERENCES 1. Scandalios JG. ed. Oxidative Stress and the Molecular Biology of Antioxidant Defenses. Plainview, NY: Cold Spring Harbor Laboratory Press, 1997. 2. Mustafa MG. Biochemical basis of ozone toxicity. Free Radic Biol Med 1990; 9: 245–265. 3. Runeckles VC, Chevone BI. Crop responses to ozone. In: Lefohn AS, ed. Surface Level Ozone Exposures and their Effects on Vegetation. Boca Raton, FL: CRC Press, 1992; 185–266.

dRR17

23/7/99

1:44 pm

Page 103

Response of antioxidant genes in maize leaves exposed to ozone 103 4. Kangasjärvi J, Talvinen J, Utriainen M, Karjalainen R. Plant defense systems induced by ozone. Plant Cell Environ 1994; 17: 783–794. 5. Pell EJ, Schlagnhaufer CD, Arteca RN. Ozone induced oxidative stress:mechnisms of action and reaction. Physiol Plant 1997; 100: 264–273. 6. Mehlhorn H, Wellburn AR. Man-induced causes of free radical damage: O3 and other gaseous pollutants. In: Foyer CH, Mullineaux PH, eds. Causes of Photooxidative Stress and Amelioration of Defense Systems in Plants. Boca Raton, FL: CRC Press, 1994; 155–197. 7. Scandalios JG. Oxygen stress and superoxide dismutases. Plant Physiol 1993; 101: 7–12. 8. Scandalios JG. Response of plant antioxidant defense genes to environmental stress. Adv Genet 1990; 28: 1–41. 9. Heath RL. Alterations of plant metabolism by O3 exposure. In: Alscher RG, Wellburn AR, eds. Plant Responses to the Gaseous Environment. New York, NY: Chapman and Hall, 1994; 121–145. 10. Scandalios JG. Molecular genetics of superoxide dismutase in plants. In: Scandalios JG, ed. Oxidative Stress and the Molecular Biology of Antioxidant Defenses. Plainview, NY: Cold Spring Harbor Laboratory Press, 1997; 527–568. 11. Halliwell B. Oxidative damage, lipid peroxidation and antioxidant protection in chloroplasts. Chem Phys Lipids 1987; 44: 327–340. 12. Scandalios JG, Guan LM, Polidoros AN. Catalases in plants: gene structure, properties, regulation and expression. In Scandalios JG, ed. Oxidative Stress and the Molecular Biology of Antioxidant Defenses. Plainview, NY: Cold Spring Harbor Laboratory Press, 1997; 343–406. 13. Hausladen A, Alscher GR. Glutathione: synthesis in oxidative stress. In: Packer L, Cadenas E, eds. Biothiols in Health and Disease. New York, NY: Marcel-Dekker, 1993; 1–30. 14. Marrs KA. The functions and regulation of glutathione Stransferase in plants. Annu Rev Plant Physiol Plant Mol Biol 1996; 47: 127–158. 15. Price A, Lucas P, Lea P. Age dependent damage and glutathione metabolism in ozone fumigated barley: a leaf section approach. J Exp Bot 1990; 41: 1309–1317. 16. Conklin PL, Last RL. Differential accumulation of antioxidant mRNAs in Arabidopsis thaliana exposed to ozone. Plant Physiol 1995; 109: 203–212. 17. Sharma YK, Davis KR. The effects of ozone on antioxidant responses in plants. Free Radic Biol Med 1997; 23: 480–488. 18. Tepperman JM, Dunsmuir P. Transformed plants with elevated levels of chloroplastic SOD are not more resistant to superoxide toxicity. Plant Mol Biol 1990; 14: 501–511. 19. Pitcher LH, Brennan E, Hurley A, Dunsmuir P, Tepperman JM, Zillinskas BA. Overproduction of petunia chloroplastic copper/zinc superoxide dismutase does not confer ozone tolerance in transgenic tobacco. Plant Physiol 1991; 97: 452–455. 20. Bowler C, Slooten L, Vandenbranden S et al. Manganese superoxide dismutase can reduce cellular damage mediated by oxygen radicals in transgenic plants. EMBO J 1991; 10: 1723–1732. 21. Broadbent P, Creisson GP, Kular B, Wellburn AR, Mullineaux PM. Oxidative stress responses in transgenic tobacco containing altered levels of glutathione reductase activity. Plant J 1995; 8: 247–255. 22. Zhu D, Scandalios JG. Maize mitochondrial manganese superoxide dismutases are encoded by a differentially expressed multigene family. Proc Natl Acad Sci USA 1993; 90: 9310–9314.

23. Eckardt NJ, Pell EJ. Oxidative modification of Rubisco from potato foliage in response to ozone. Plant Physiol Biochem 1995; 33: 273–282. 24. Glick RE, Schlagenhaufer CD, Arteca RN, Pell EJ. Ozone-induced ethylene emission accelerates the loss of ribulose-1,5-bisphosphate carboxylase/oxygenase and nuclear-encoded mRNAs in senescing potato leaves. Plant Physiol 1995; 109: 891–898. 25. Matters GL, Scandalios JG. Synthesis of isozymes of superoxide dismutase in maize leaves in response to ozone, sulfur dioxide and elevated oxygen. J Exp Bot 1987; 38: 842–852. 26. Beachy RN, Chen ZL, Horsch RB, Rogers SG, Hoffman NL, Fraley RT. Accumulation and assembly of soybean β-conglycin in seeds of transformed petunia plants. EMBO J 1985; 4: 3047–3053. 27. Church GM, Gilbert W. Genomic sequencing. Proc Natl Acad Sci USA 1984; 81: 1991–1995. 28. White JA, Scandalios JG. Isolation and characterization of a complementary DNA for mitochondrial manganese superoxide dismutase (SOD-3) of maize and its relation to other manganese superoxide dismutases. Biochim Biophys Acta 1988; 951: 61-70. 29. Bethards LA, Scandalios JG. Molecular basis for the CAT-2 null phenotype in maize. Genetics 1988; 118: 149–153. 30. Jorgensen RA, Cuellar RE, Thompson WF, Kavanagh TA. Structure and variation in the ribosomal RNA gene of pea. Characterization of a cloned rDNA repeat and chromosomal rDNA variants. Plant Mol Biol 1987; 8: 3–12. 31. Sharma YK, Davis KR. Ozone-induced expression of stressrelated genes in Arabidopsis thaliana. Plant Physiol 1994; 105: 1089–1096. 32. Guan L, Scandalios JG. Developmentally related responses of maize catalase genes to salicylic acid. Proc Natl Acad Sci USA 1995; 92: 5930–5934. 33. Williamson JD, Scandalios JG. Differential response of maize catalases and superoxide dismutases to the photoactivated fungal toxin cercosporin. Plant J 1992; 2: 351–358. 34. Redinbaugh MG, Sabre M, Scandalios JG. Expression of the maize Cat3 catalase gene is under the influence of a circadian rhythm. Proc Natl Acad Sci USA 1990; 87: 6853–6857. 35. Polidoros AN, Scandalios JG. Circadian expression of the maize catalase Cat3 gene is highly conserved among diverse maize genotypes with structurally different promoters. Genetics 1998; 149: 405–415. 36. Kliebenstein DJ, Monde R-A, Last RL. Superoxide dismutase in Arabidopsis: an eclectic enzyme family with disparate regulation and protein localization. Plant Physiol 1998; 118: 637–650. 37. Langebartels C, Kerner K, Leonardi S et al. Biochemical plant responses to ozone. Plant Physiol 1991; 95: 882–889. 38. Melhorn H, Tabner B, Wellburn AR. Electron spin resonance evidence for the formation of free radicals in plants exposed to O3. Plant Physiol 1990; 79: 377–383. 39. Prestera T, Talalay P. Electrophile and antioxidant regulation of enzymes that detoxify carcinogens. Proc Natl Acad Sci USA 1995; 92:8965-8969. 40. Van Camp W, Willikens H, Bowler C et al. Elevated levels of superoxide dismutase protect transgenic plants against ozone damage. Bio/Tech 1994; 12: 165-168. 41. Kernodle SP, Scandalios JG. A comparison of the structure and function of the highly homologous maize antioxidant Cu/Zn superoxide dismutase genes, Sod4 and Sod4A. Genetics 1996; 144: 317–328. 42. Scandalios JG. Subcellular localization of catalase variants coded by two genetic loci during maize development. J Hered 1974; 65: 28–32.