Molecular Responses to Photooxidative Stress in ...

Plant Physiol. (1993) 103: 1385-1391

Molecular Responses to Photooxidative Stress in Pinus sylvestris (1.)' II. Differential Expression of CuZn-Superoxide Dismutases and Clutathione Reductase Stanislaw Karpinski*, Cunnar Wingsle, Barbara Karpinska, and Jan-Erik Hallgren

Department of Forest Genetics and Plant Physiology, Faculty of Forestry, Swedish University of Agricultural Science, S-901 83 Ume5, Sweden

oxygen. GR (EC 1.6.4.2) is a flavoprotein that catalyzes the reduction of GSSG to the GSH in the presence of NADPH. GR together with ascorbate peroxidase and -dehydroascorbate reductase constitutes the chloroplastic H202-scavengingsystem (Foyer and Halliwell, 1976; Asada and Takahashi, 1987). In the plant kingdom three basic forms of SODs are found: CuZn-SODs, active in cytosol and the chloroplast stroma; FeSOD, active in the chloroplast stroma; and Mn-SOD, active in mitochondrial matrix (Fridovich, 1986; Palma et al., 1986; Halliwel, 1987; Bowler et al., 1989). GR activity is localized mainly in the chloroplasts but has also been detected in a variety of nonphotosynthetic tissues and organelles (Young and Conn, 1956; Gillham and Dodge, 1986; Edwards et al., 1990). Sod can be induced by diverse stress conditions such as high light and/or low temperature, drought, air pollutants, and xenobiotic and fungal attack. Information so far indicates that Sod genes respond to increased .O2- concentration. However, regulatory and signaling mechanisms are not known. Matters and Scandalios (1986,1987) showed that the herbicide paraquat and a high O2level induced two cytosolic isozymes of Sod genes. Bowler et al. (1989) showed that MnSod responded to mitochondrial localized stress events. Similarly, Tsang et al. (1991) showed that Fe-Sod mRNAs increased with chloroplast-localizedoxidative stress. They also convincingly showed that CuZn-Sod, Mn-Sod, and Fe-Sod mRNA reacted differently to stress conditions and that appearance of the corresponding mRNA was not dependent on phytochrome action. The Sod genes are also induced by the photoactivated fungal toxin cercosporin (Williamson and Scandalios, 1992). Air pollutants such as SOz together with NO2 can also induce Sod gene expression (Karpinski et al., 1992a). The knowledge of the regulation and expression of Gor genes in plants is limited. Available data are based mainly on protein activity studies. Phytochrome regulation and light control of GR activity in Synapsis alba (L.) were described by Drumm-Herrel et al. (1989), who showed that the appearance of the chloroplastic isoform of GR requires phytochrome

l h e influence of photooxidative stress on genes expressing superoxide dismutase (Sod) and glutathione reductase (Cor) was analyzed in needles of top and side shoots of 3-year-old Pinus sylvestris (1.) seedlings. l h e study was carried out in the field during spring recovery. From mid-April the top shoots of seedlings protruded above the snow and thus were exposed to sunlight, whereas the side shoots were covered with snow until May 4. Needles were sampled from top and side shoots on five different occasions. At the beginning of May the mRNA levels for cytosolic CuZn-Sod were significantly higher in top-shoot needles than in side-shoot needles. Similar results were obtained for chloroplastic CuZn-Sod mRNA. After May 6 we could not detect any significant differences between top- and side-shoot needles for either CuZnSod mRNA level. lranscript accumulation for the chloroplastic CuZn-Sod was up to 4-fold higher than for cytosolic CuZn-Sod in both types of shoots. On lune 1 minimum transcript levels were observed for both CuZn-SOD isoforms. Protein activity analysis for CuZn-SOD isozymes did not reveal any significant differences between top- and side-shoot needles during the whole period of measurements. l h e mRNA level for chloroplastic Cor was similar in both types of shoots. However, the total CR activity was significantly higher in top-shoot needles than in side-shoot needles at the beginning of May. l h e analysis of mRNA accumulation for chloroplastic CuZn-Sod and Cor indicates that transcript levels were at least 5- to 20-fold higher for CuZn-Sod than for chloroplastic Cor. l h e differential expressions of Sod and Cor genes are discussed in relation to regulation of the enzymic scavenging system during photooxidative stress conditions.

Early spring in the boreal forest, with snow still covering the ground, high quantum flux densities, and large variations in day and night temperature, is a particularly stressful period for evergreen seedlings. The needles could be exposed to high light while the root zone is still frozen. Such conditions may generate photoinhibition and subsequent photooxidation by free radical reactions within plant cells (Oquist et al., 1987). SOD (EC 1.15.1.1) converts .O2- to H202 and constitutes the first link in the enzymic scavenging system of active

' This work was supported by The Swedish Council for Forestry and Agricultura1 Research, The Swedish Council for Natural Sciences, and The Swedish Environment Protection Board. * Corresponding author; fax 46-90-165901.

Abbreviations: Gor, gene encoding GR; GR, glutathione reductase; poly(dT), polydeoxythymidylate; Sod, gene encoding SOD; SOD, superoxide dismutase. 1385

Karpinski et al.

i386

action, whereas the cytosolic isoform of GR was not dependent on the integrity of the plastid. Drought and photooxidative herbicide tolerances in maize were correlated with high levels of CuZn-SODs and GR activities (Malan et al., 1990). High light and magnesium deficiency enhance the activities of SOD, ascorbate peroxidase, and GR in bean leaves (Cakmak and Marschner, 1992). Air pollutants such as 0 3 can induce GR activity. Increased GR protein leve1 due to 0 3 fumigation in spinach and pea leaves has been reported (Tanaka et al., 1988; Madamanchi et al., 1992). All known Sod and Gor genes in plants are located in the nucleus, but it is not clear how Sod and Gor gene regulation in plants is controlled. By analogy with bacterial systems, a regulatory gene (or genes) may regulate the expression of Sod and Gor genes (Shaaltiel et al., 1988). Plants carrying mutations in regulatory genes controlling Sod expression have been reported (Baum and Scandalios, 1982). The regulation of antioxidant responses seems to be much more complex in plants than in prokaryotic systems. Responses specific to the cytosol, mitochondria, or chloroplasts may be superimposed on the general stress induction of Sod and Gor genes and have to involve signal transduction systems (Bowler et al., 1992). Previously, we showed higher mRNA levels in top-shoot needles than in side-shoot needles for LhcaZ*l, Lhcbl*2, and RbcS genes, whereas rbcL, psbA, nad3, and cox2 genes had constant and similar levels of mRNA in both types of shoots. This was correlated with reduced content of pigments and photochemical efficiency of PSII (reduced variable fluorescenceImaxima1 fluorescenke ratio of Chl a fluorescence) in top-shoot needles compared to side-shoot needles (Karpinski et al., 1993). In this paper we have, in addition, analyzed the expression and regulation of Sod and Gor genes using exactly the same plant material as above. MATERIALS and METHODS Plant Material a 4 Crowth Conditions

Three-year-old Pinus sylvestris (L.) seedlings growing in the field were used for studies of Sod and Gor gene expression during spring 1992. The mean height of the seedlings was about 40 cm. They had four to five side shoots 20 cm above ground and one top shoot. From mid-April the top shoots protruded above the snow and were exposed to sunlight and variations in day and night air temperatures. The side shoots were covered by snow until May 4. Additional climatic data during the period of the experiment will be presented (Karpinski et al., 1993). The mRNA levels and protein activities were analyzed for current-year needles of top and side shoots of six individual seedlings. Because only a limited number of needles could be sampled to avoid too much disturbance of the plants, pooled samples were used (three pairs of needles from each shoot and seedling, 18 pairs of needles per sample total), which were collected separately from top and side shoots. Additionally, samples were collected on May 4 and June 1 from the above six seedlings for individual variation analysis.

Plant Physiol. Vol. 103, 1993

Preparation of Total RNA and Poly(A+) RNA

An LiC1-based extraction method was used for RNA preparations (Whitmore and Kriebel, 1987) and modified by Karpinski et al. (1992b). The average yield of total RNA was 0.8 mg from 1 g of needles. Poly(A+) RNA was obtained from the total RNA by two sequential passages through an oligo(dT)-cellulose-spuncolumn (Pharmacia LKB, Uppsala, Sweden). Cene-Specific and Poly(dT) Probes

The homologous CuZn-Sod cDNA probes (300-bp cytosolic PS3 EcoRIIHincII fragment and 400-bp chloroplastic PST13 EcoRIIHaeIII fragment) described by Karpinski et al. (199213) and heterologous chloroplastic Gor cDNA probe from Pisum sativum (L.) (2051-bp pGR201 BamHI fragment) described by Creissen'et al. (1991) were used for the hybridization with total RNA and mRNA. Probes were labeled with [a-32P]dCTP by the random primer method using the kit and protocol of the supplier (Pharmacia LKB). Poly(dT) cDNA probes were synthesized as described by Hollander and Fomace (1990) and were hybridized with total RNA. The average length of the probes was 120 to 160 bp. The poly(dT) cDNA probes were labeled during synthesis with [a-32P]dTTP. RNA Slot Blot Hybridization

The total RNA samples (20, 10, 5, and 2.5 r g ) were prepared for blotting according to the procedure of Sambrook et al. (1989). Blotting to Hybond N membrane was performed in a Minifold I1 apparatus (Schleicher & Schuell, Dassel, Germany) according to the protocol of the supplier. RNA was fixed to the membrane by baking at 8OoC for 2 h. Filters were prehybridized at 68OC (6OOC for Gor probe only) for 5 h in 6X SSC ( l x SSC is 0.15 M NaCl, 15 mM sodium citrate), 5X Denhardt's solution, 50 mM sodium phosphate, 0.1% SDS, and 100 r g mL-' of salmon sperm DNA. The filters were hybridized in the same solution and temperature plus 0.003 mg L-' (3.66 X 107 Bq) of radioactive probe for 24 h and washed four times: twice for 20 min in 0.1X SSC and 0.1% SDS (IX SSC and 0.5% SDS for Gor probe) at room temperature and twice for 10 min in 0.1 x SSC and 0.1% SDS at 58OC (55OC, 1.OX SSC and 0.5% SDS for Gor probe). Under this condition no cross-hybridization for chloroplastic and cytosolic CuZn-Sod was observed (Karpinski et al., 1992b). The filters were exposed to x-ray film up to 120 h. The cDNA inserts from recombinant plasmids PS3, PSTl3, and pGR201 were diluted to four different concentrations corresponding to 40,20,10, and 5 X 106molecules of doublestranded cDNA, mixed with 1 Pg of sonicated salmon sperm DNA, denatured, and blotted on Hybond N. Such prepared standards were used as references in the total RNA slot blot hybridization experiments. Slot blot hybridization to poly(dT) cDNA probes was performed according to the method of Hollander and Fomace (1990) and Karpinski et al. (1992a). The only difference was that we applied 0.030 mg L-' (40 MBq) of 32P-labeled poly(dT) cDNA and 4 mg L-' of unlabeled poly(dT) cDNA of hybridization mixture.

Expression of Genes Encoding Superoxide Dismutase and Glutathione Reductase Poly(A+) RNA Gels and Hybridization +

The poly(A ) RNA samples (5 Mg) were separated on 1.8% (w/v) agarose gels after glyoxylation in phosphate buffer (Sambrook et al., 1989). Vacuum blotting to Hybond N membranes was performed according to the protocol of the suppliers (Pharmacia LKB and Amersham [Aylesbury, UK]). RNA was fixed onto the membrane by baking at 80°C for 2 h. Prehybridization, hybridization, washing, and autoradiography were performed as described above for the Gor probe. Extraction of Protein and Analysis of SOD and GR Activity Needles (0.5 g) stored at —80°C were frozen in liquid nitrogen and ground twice for 1 min in a dismembrator (Retsch, Darmstadt, Germany). The still-frozen powder was suspended in 5 mL of 50 min Na-phosphate (pH 7.0) containing 4% (v/w) soluble PVP 360, 2 HIM EDTA, and 5 min DTT and homogenized for 1 min in the dismembrator. The extract was then sonicated for 30 s (Soniprep 150; MSB, UK) and shaken for 15 min at 2°C. The PVP in the extracts was precipitated by adding (NH4)2SO4 to a final concentration of 25% (v/w). After the extracts were centrifuged (15,000g for 5 min) the supernatant was divided into two parts. Four hundred units of CuZn-SOD from bovine liver (Sigma) were added to 3.25 mL, and quantitative analysis of different CuZn-SOD isozyme activities on native PAGE gradient gels (8-25%) was performed as described by Wingsle et al. (1991). The sample was concentrated with a Microsep concentrator (Filtron Technology Corp., Northborough, MA) and equilibrated with a phosphate buffer (10 ITIM Na2HPO4 [pH 7.0], 1 mM EDTA). SOD activity staining in the gels was performed according to the method of Beauchamp and Fridovich (1971). The gels were scanned at 540 nm (DU-8; Beckman, Irvine, CA). The relative activities of the P. sylvestris chloroplastic and cytosolic CuZn-SOD forms were calculated using bovine liver SOD as an internal standard. The other part of the supernatant was used for total SOD and GR activity measurements. Total SOD was determined by the direct KO2 method (Marklund, 1976). The total GR activity in the partially purified protein extract was measured at 25°C as the rate of oxidation of NADPH (170 ^M) by GSSG (1 mM) in Na2HPO4 buffer (25 mM, pH 7.8). The reaction was started by adding NADPH and monitored by the decrease in A340. All data were analyzed using analysis of variance. Further analysis was carried out using a two-way analysis of variance. RESULTS

Transcript Analysis Relative and absolute CuZn-Sod and Gor transcript levels in current-year needles of top and side shoots of P. sylvestris (L.) seedlings were analyzed by a total RNA slot blot and northern blot hybridization. The findings from hybridization of cytosolic and chloroplastic CuZn-Sod, chloroplastic Gor, and poly(dT) cDNA probes to total RNA are presented in Figure 1. Relative hybridization values after normalization to poly(A+) content for all probes (except Gor) obtained from

1387

Figure 1 are shown in Table I. Significant differences in specific RNA levels between top and side shoots for cytosolic CuZn-Sod were detected on May 4 (2-fold higher in top shoots) and on May 6 (4-fold higher in top shoots). Similar differences were observed for chloroplastic CuZnSod transcript levels during the same period. After May 6, the relative levels in top- and side-shoot needles for both cytosolic and chloroplastic CuZn-Sod mRNA were similar (Fig. 1 and Table I). Chloroplastic Gor transcript was not detected in total RNA samples, and the detection limit was estimated at approximately 2 X 106 mRNA molecules for chloroplastic Gor in 20 jig of total RNA. However, chloroplastic Gor transcript was detected when purified poly(A+) RNA samples were analyzed in the northern blot experiment (Fig. 2). In this experiment, specific mRNA levels for chloroplastic Gor were found to be similar in both types of shoots (Fig. 2 and Table II). A relatively higher mRNA level for chloroplastic Gor was observed on May 15 and May 19 in top- and side-shoot needles, twice as high as on May 4 and June 1 (Fig. 2 and Table II). Figure 3 presents an estimation of mRNA copy numbers for cytosolic and chloroplastic CuZn-Sod derived from hybridization data (cf. Fig. 1). Significant differences in cytosolic CuZn-Sod mRNA levels were observed on May 4 (10 X 106 molecules 20 /xg"1 of RNA for top-shoot needles and 5 x 106

cyt Sod

cp Sod

cp Cor

poly dT

May 4. only T shools exposed to light

pGR201

I t

40 20 10 5

40 20 10 5

I I I 40 20 10 5

Figure 1. Slot blot hybridization of homologous cytosolic CuZnSod (cyt Sod), homologous chloroplastic CuZn-Sod (cp Sod), heterologous (pea) chloroplastic Cor, and poly(dT) cDNA probes to different amounts (20, 10, 5, and 2.5 jtg) of total RNA isolated on different occasions (May 4, 6, 15, and 19 and June 1) from needles of top and side shoots of P. sylvestris (L.) seedlings. Strips containing 40, 20, 10, and 5 x 106 molecules of cDNA inserts of recombinant plasmids PS3, PST13, and pCR201 were used as external molecule number references as described in "Materials and Methods."

Karpinski et al.

1388

Table I. Relative values for hybridization of cytosolic and chloroplastic (cp) CuZn-Sod cDNA and poly(dT) probes to P. sylvestris (L.) total RNA from needles of top shoot on May 4, 6, 15, and ?9 and tune 1 Hybridization to top-shoot needles RNA (Fig. 1) was normalized to side-shoot needles of the same date (value set to 1). Hybridization was normalized to poly(A+) RNA content for each sample by dividing the relative hybridization to the cDNA probes by the relative hybridization to the poly(dT) probe. Differences were considered to be significant when they were equal to or higher than 2fold. Results are ratios of relative mRNA levels (top/side).


Table II. Relative values for northern hybridization of heterologous (pea) chloroplastic Cor cDNA to P. sylvestris (L.) poly(A+) RNA from needles of top and side shoot on May 4, 6, 15, and 19 and /une 1 (cf. Fig. 2) Differences were considered to be significant when they were equal to or higher than 2-fold. Results are relative mRNA levels. Shoot

May 4

May 6 May 15 May 19 June 1

Probe

May 4 May 6

May 15 May 19 June 1

Cytosolic Sod

cpSod

poly(dT)

2

2 4 1 1 1

1 1 1 1 1

4 1 1 1

molecules 20 jtg * of RNA for side-shoot needles) and on May 6 (40 X 106 molecules 20 ^g"1 of RNA for top-shoot needles and 10 X 106 molecules 20 ^g"1 of RNA for sideshoot needles). For chloroplastic CuZn-Sod-specific transcript levels, differences were observed also on May 4 (20 x 106 molecules 20 jig"1 of RNA for top-shoot needles and 10 X 106 molecules 20 jig"1 of RNA for side-shoot needles) and on May 6 (60 x 106 molecules 20 jig"1 of RNA for top-shoot needles and 15 x 106 molecules 20 jtg'1 of RNA for sideshoot needles). These results confirm the relative estimations (cf. Table I). Quantitatively, the mRNA level was significantly higher (up to 4-fold) for chloroplastic CuZn-Sod than for cytosolic CuZn-Sod in top- and side-shoot needles (Fig. 3). The lowest transcript level for cytosolic CuZn-Sod was observed on May 4 in side shoots (5 x 106 molecules 20 ^g"1 of RNA) and on June 1 in top and side shoots (5 X 106 molecules 20 ng~l of RNA). In the case of chloroplastic CuZnSod the lowest mRNA level was observed on June 1 (5 X 106 molecules 20 jig"1 of RNA) in top and side shoots. Assuming that total RNA extract from P. sylvestris (L.) contains approximately 1 to 2% of poly(A+) RNA (results not shown) and a detection limit for chloroplastic Gor transcript estimated at approximately 2 X 106 mRNA molecules 20 ^g"1 of total RNA, we were able to estimate the highest possible

Top

Side

1.0 1.3 2.0

1.0 1.7 2.0 2.0 1.0

2.0 1.0

number of mRNA molecules of Gor in total RNA extracts (1.0-3.0 X 106 molecules 20 jig"1 of RNA). These data indicate that transcript levels for chloroplastic CuZn-Sod were at least 5- to 20-fold higher than for chloroplastic Gor. SOD and GR Protein Activity Measurement

Total SOD activity in the top and side shoots did not differ significantly between May 4 and June 1. The values for May 4 were 2151 ± 234 and 1777 ± 166 units g~' fresh weight (mean ± SE, n = 6) for top- and side-shoot needles, respectively. On June 1, values were 2070 ± 180 and 1911 ± 296 units g"1 fresh weight (mean ± SE, n = 6) for needles of top and side shoots, respectively. A comparison of CuZn-SOD isozymes did not reveal any significant differences in activity for the corresponding isozymes between the top- and side-shoot needles (Fig. 4). However, the relative activity of chloroplastic CuZn-SOD

• cyt Sod T

0 cyt Sod S B cpSodT O cpSodS

3

S 40

June 1

20 z at

e

I May 4

Figure 2. Northern blot hybridization analysis of chloroplastic Cor mRNA levels in poly(A+) RNA. Poly(A+) RNA (5 ng per line) from top (T) and side (S) shoot needles of P. sylvestris (L.) seedlings (May 4, 6, 15, and 19 and June 1) were separated by gel electrophoresis, transferred to a filter, and hybridized with heterologous (pea) Cor cDNA probe as described in "Materials and Methods."

May 6

May 15

May 19

June 1

Figure 3. The estimated number of mRNA molecules of cytosolic CuZn-Sod (cyt Sod) and chloroplastic CuZn-Sod (cp Sod) in top (T) and side (S) shoot needles of P. sylvestris (L.) seedlings (May 4, 6, 15, and 19 and June 1). The CuZn-Sod mRNA molecules were estimated at 20 Mg of total RNA and normalized to poly(A+) RNA content (cf. Fig. 1 and Table I).

Expression of Genes Encoding Superoxide Dismutase and Glutathione Reductase

May

Figure 4. Activity staining of P. sylvestris (L.) SOD isozymes after native PACE. The SODs were isolated from top (T) and side (S) shoot needles on different occasions (May 4, 6, 15, and 19 and June 1). The bovine liver SODs were added to all samples (except T*) as an internal standard. Relative density was calculated after normalization to bovine liver SOD content. Chloroplastic SOD (SOD 3): O, in top shoot; •, in side shoot. Cytosolic SOD—SOD 1: A, in top shoot; A, in side shoot; SOD 2: D, in top shoot; •, in side shoot; SOD 4: O, in top shoot; *, in side shoot.

(SOD 3) was 3- to 4-fold higher than that of other isoforms in both types of shoots (Fig. 4). The total GR activity was significantly higher in top shoots than in side shoots only at the beginning of May (Fig. 5, P = 0.0003, two-way analysis of variance). Reduced activity with minimum activity occurring on May 19 was found. DISCUSSION

Previously, we showed that pigment levels as well as the photochemical efficiency of PSII were reduced in top-shoot needles (light-exposed) compared with side-shoot needles (covered by snow). During spring, top shoots suffered from photoinhibition, whereas side shoots were less affected. After May 4 a rapid recovery of pigments and photosynthesis was observed in both types of shoots. On May 4 and 6 significantly higher mRNA levels for LhcaVl, Lhcbl*2, and RbcS genes were detected in top-shoot needles compared with side-shoot needles. However, the mRNAs for chloroplastencoded psbA and rbcL genes as well as mitochondria-encoded cox2 and nad3 genes showed similar and constant levels in both types of shoots (S. Karpinski, unpublished results). For this paper, we analyzed Sod and Gor gene expression and regulation during photooxidative stress using the same plant material as above. The mRNA levels for cytosolic and chloroplastic CuZnSODs were significantly higher in top-shoot needles than in side-shoot needles only at the beginning of May. These changes in transcript levels were not reflected by a corresponding increase in protein activities. The disproportion between mRNA levels and protein activity for CuZn-SODs as a response to higher oxidative stress has been observed before and was suggested to be a result of higher turnover rates of CuZn-SODs in stressed plants (Karpinski et al.,

1389

1992a; Williamson and Scandalios, 1992). In addition, it was suggested that high turnover rates of CuZn-SODs in stressed plants could be caused by the inhibitory effect of H2O2 (Karpinski et al., 1992b). In vitro inhibition of CuZn-SOD by H2O2 has, in fact, been reported (Hodgson and Fridovich, 1975). We suggest that an increased rate of turnover of CuZnSOD might be a result of higher oxidative stress and sensitivity of the enzymes to H2O2. The reversible or irreversible inhibition may be crucial for the regulation of CuZn-Sod gene expression. A comparison of chloroplastic and cytosolic CuZn-Sod mRNAs showed a higher transcript level for the chloroplastic form until mid-May. This higher transcript level was also associated with a higher chloroplastic CuZn-SOD activity. Thereafter, transcript levels were reduced for both chloroplastic and cytosolic CuZn-Sod mRNAs, which reached similar low levels of expression on June 1. In general, the chloroplastic transcripts were reduced 8- to 15-fold, and the cytosolic ones were reduced 2- to 8-fold for the side and top shoots, respectively, during this period. This was not correlated with any significant change in the corresponding protein activity. The different pattern in mRNA levels for chloroplastic and cytosolic CuZn-Sod during the recovery indicates a chloroplast-dependent specific regulation of CuZn-Sod gene expression. This type of regulation for Sod genes was observed previously (Perl-Treves and Galun, 1991; Tsang et al., 1991; Karpinski et al., 1992b). In addition, our results also showed lowered turnover rates of the CuZn-Sod mRNA after the repair process of the photosynthetic apparatus was completed and photosynthetic capacity was fully recovered (Karpinski et al., 1993). The total GR enzyme activity was significantly higher in top-shoot needles then in side-shoot needles at the beginning of May and might indicate a higher capacity of the H2O2 scavenging system. A decrease in total GR enzyme activity in both top- and side-shoot needles in the middle of May with minimum activity reported on May 19 was detected. However, the transcript levels of chloroplastic Gor were similar in both types of shoots during the whole period. At the

-2? IX

Q

2,0

o

e 1,0

10

20

30

May Figure 5. Total GR activity in P. sylvestris (L.) needles measured on May 4, 6, 15, and 19 and June 1 in top-shoot needles (O) and sideshoot needles (•). Values from May 4 and June 1 represent means ± SE (n = 6).

1390

Karpinski et al.

beginning and at the end of the period the transcript levels were 2-fold lower than in mid-May. It should be noted that we measured total GR activity but mRNA for chloroplastic Gor only. Different levels of accumulation of GR isoforms in top- and side-shoot needles before or during winter could possibly have affected results. Severa1 isoforms of GR in plants and seasonal changes of GR activity (higher activity during winter and spring and lower activity during summer) have been observed and reported (Esterbauer and Grill, 1978; Anderson et al., 1991; Wingsle and Hallgren, 1993). The decrease in transcript level detected on June 1 might be the first signal for the reduction of GR activity that appeared later during June. An estimation of the number of mRNA molecules for chloroplastic CuZn-Sod and Cor showed that transcript levels were at least 20-fold higher for CuZn-Sod than for Gor. However, the protein activity level for CuZn-SODs was approximately 4-fold higher than GR (result not shown, calculated from the obtained activities per g fresh weight and the specific activities of purified GR and CuZn-SODs; Wingsle, 1989; Wingsle et al., 1991). These results suggest higher tumover rates for CuZn-SODs than for GR in the P. sylvestris (L.) needles during photooxidative stress. The data presented here indicate complex regulatory mechanisms for Sod and Gor gene expression, which has been confirmed by previous studies. Specific responses addressed to cytosol, chloroplasts, or mitochondria have to involve signaling systems, which might be superimposed on the overall regulation of Sod and Gor gene expression (Bowler et al., 1992). It has also been suggested that the search for a stress response signal could start in the chloroplast with the GSH biosynthesis pathway (Dinscha 1987; Alscher et al., 1991). Additionally, we suggested that H 2 0 2 might play a crucial role in the regulatory mechanism of Sod and Gor gene expression (Bowler et al., 1991, 1992; Karpinski et al., 1992a). The production rates of . 0 2 - and H202were estimated to be M in chloroplasts under normal photoabout 80 to 160 ~ L s-l synthetic conditions (Asada and Takahasi, 1987). The concentration and production rates of . 0 2 - and H202 in the needles during these spring conditions with impaired photosynthetic apparatus were not known. However, an increased capacity of the scavenging system was reported for hardened evergreens (Esterbauer and Grill, 1978; Anderson et al., 1991; Wingsle and Hallgren, 1993) and indicates that a high proportion of excitation energy might be scavenged by this system in chloroplasts during stressful spring conditions. This altemative pseudocyclic electron flow was proposed to be important for regulation of PSII protection against light excess, especially with limited C 0 2 supply or inhibited Calvin cycle (Schreiber and Neubauer, 1990). In summary, our data indicate that P. sylvestris (L.) is well adapted to photoinhibitory stress conditions. This adaptability is multidirectional and involves systems of enhanced reparation and synthesis of the photosynthetic apparatus (Karpinski et al., 1993) and enhanced protection against free radicals. Consequently, it leads to rapid recovery of photosynthetic efficiency. Some of these naturally established capacities of P. sylvestris (L.) have recently been confirmed by studies of transgenic tobacco plants that overexpress chloroplastic Gor or CuZn-Sod (Aono et al., 1993; Gupta et al.,


1993). They indicate rapid recovery of photosynthesis after oxidative or photooxidative stress in transgenic plants compared to wild types. More detailed studies will be necessary to better understand the mechanisms of photoinhibitory stress response in plants. Furthermore, cloning of regulatory genes for oxidative stress and identification of putative signal transducers will be required. ACKNOWLEDCMENT We are grateful to Dr. Gary P. Creissen from John Innes Institute, Norwich, UK, for kindly providing recombinant plasmid pGR201, and to Dr. Olle Olsson at our department for stimulating discussion and valuable comments regarding the manuscript. Received May 21, 1993; accepted August 12, 1993. Copyright Clearance Center: 0032-0889/93/103/1385/07.

LITERATURE ClTED Alscher R, Madamanchi NR, Cramer CL (1991) Protective mechanisms in the chloroplast stroma. rn EJ Pell, KL Steffen, eds, Active Oxygen/Oxidative Stress and Plant Metabolism. American Society of Plant Physiologists, Rockville, MD, pp 145-155 Anderson JV, Chevone BI, Hess JL (1991) Seasonal variation in the antioxidant system of Eastem white pine needles. Evidence for thermal dependence. Plant Physiol98 501-508 Aono M, Kubo K, Saji H, Tanaka K,d Kondo N (1993) Enhanced tolerance to photooxidative stress of transgenic Nicotiana tabacum with high chloroplastic glutathione reductase activity. Plant Cell Physiol34 129-135 Asada K, 'I'akahashi M (1987) Production and scavenging of active oxygen in photosynthesis. In DJ Kyle, CB Osmond, CJ Amtzen, eds, Photoinhibition. Elsevier, Amsterdam, pp 227-287 Baum JA, Scandalios JG (1982) Multiple genes controlling superoxide dismutase expression in maize. J Hered 7 3 95-100 Beauchamp CO, Fridovich I(1971) Superoxide dismutase: improved assay and an assay applicable to airylamide gels. Ana1 Biochem 44: 276-287 Bowler C, Alliotte T, De Loose M, Van Montagu M, Inze D (1989) The induction of manganese superoxide dismutase in response to stress in Nicotiana plumbaginifolia. EMBO J 8: 31-38 Bowler C, Slooten L, Vandenbranden S, De Rycke R, Botterman J, Sybesma C, Van Montagu M, Inze D (1991) Manganese superoxide dismutase can reduce cellular damage mediated by oxygen radicals in transgenic plants. EMBO J 10: 1723-1732 Bowler C, Van Montagu M, Inzé D (1992) Superoxide dismutase and stress tolerance. Annu Rev Plant Physiol Plant Mo1 Biol 43: 83-116 Cakmak I, Marschner H (1992) Magnesium deficiency and high light intensity enhance activities of superoxide dismutase, ascorbate peroxidase, and glutathione reductase in bean leaves. Plant Physiol98: 1222-1227 Creissen G, Edwards AE, Enard C, Wellburn A, Mullineaux P (1991) Molecular characterisation of glutathione reductase cDNA from pea (Pisum sativum L.). Plant J 2: 128-131 Dinscha R (1987) Glutathione status and protein synthesis during drought and subsequent rehydration in Tortura ruralis. Plant Physi0183 816-819 Drumm-Herrel H, GerhaMer U, Mohr H (1989) Differential regulation by phytochrome of the appearance of plastidic and cytoplasmatic isoforms of glutathione reductase in mustard (Synapsis alba L.) cotyledons. Planta 178 103-109 Edwards EA, Rawsthorne S, Mullineaux PM (1990) Subcellular distribution of multiple isoforms of glutathione reductase in leaves of pea (Pisum sativum L.).Planta 1 8 0 278-284 Esterbauer H, Grill D (1978) Seasonal variation of glutathione and glutathione reductase in needles of Picea abies. Plant Physiol 61: 119-121 Foyer CH, Halliwell B (1976) The presence of glutathione and glutathione reductase in chloroplasts: a proposed role in ascorbic

Expression of Genes Encoding Superoxide Dismutase and Glutathione Reductase acid metabolism. Planta 133: 21-25 Fridovich I (1986) Superoxide dismutases. Adv Enzymol58 62-97 Gillham DJ, Dodge AD (1986) Hydrogen-peroxide-scavenging system within pea chloroplast. A quantitative study. Planta 167: 246-251 Gupta AS, Heinen JL, Holaday SA, Burke JJ, Allen RD (1993) Increased resistance to oxidative stress in transgenic plants that overexpress chloroplastic Cu/Zn superoxide dismutase. Proc Natl Acad Sci USA 90: 1629-1633 Halliwell B (1987) Oxidative damage, lipid peroxidation and antioxidant protection in chloroplasts. Chem Phys Lipids 4 4 327-340 Hodgson EK, Fridovich I(1975) The interaction of bovine erythrocyte superoxide dismutase with hydrogen peroxide: inactivation of the enzyme. Biochemistry 1 4 5294-5298 Hollander CM, Fornace Jr AJ (1990) Estimation of relative mRNA content by filter hybridization to polythimidine probe. Biotechniaues 9 174-179 Karpinski S, Karpinska 8,Wingsle G, Hallgren J-E (1993) Molecular responses to photooxidative stress in Pinus sylvestris (L.),I . Differential expression of nuclear and plastid genes in relation to recovery from winter stress. Physiol Plant (in press) Karpinski S, Wingsle G, Karpinska B, Hallgren J-E (1992a) Differential expression of CuZn-superoxide dismutases in Pinus sylvestris needles exposed to S 0 2 and N02. Physiol Plant 8 5 689-696 Karpinski Si Wingsle G, Olsson O, Hallgren J-E (1992b) Characterisation of cDNA encoding CuZn-superoxide dismutases in Scots pine. Plant Mo1 Biol18: 545-555 Madamanchi NR, Anderson JV,Alscher RG, Cramer CL, Hess JL (1992) Purification and multiple forms of glutathione reductase from pea (Pisum sativum L.) seedlings and enzyme levels in ozonefumigated pea leaves. Plant Physiol 100 138-145 Malan C, Greyling MM, Gressel J (1990) Correlation between CuZn superoxide dismutase and glutathione reductase and environmental and xenobiotic stress tolerance in maize inbreeds. Plant Sci 6 9 157-166 Marklund SL (1976) Spectrophotometric study of spontaneous disproportionation of superoxide anion radical and sensitive direct assay for superoxide dismutase. J Biol Chem 251: 7504-7507 Matters GL, Scandalios JG (1986) Effect of the free radical-generating herbicide paraquat on the expression of the superoxide dismutase ( S o d ) genes in maize. Biochim Biophys Acta 882 29-38 Matters GL, Scandalios JG (1987) Synthesis of isozymes of super-

1391

oxide dismutase in maize leaves in response to 0 3 , SO2 and elevated 0 2 . J Exp Bot 38: 842-852 Oquist G, Greer DH, Ogren E (1987) Light stress at low temperatures. Topics in Photosynthesis and Photoinhibition 9 67-87 Palma JM, Sandalio LM, De1 Rio LA (1986) Manganese superoxide dismutase and higher plant chloroplasts: a reappraisal of a controverted cellular localization. J Plant Physiol 1 2 5 427-439 Perl-Treves R, Galun E (1991) The tomato Cu,Zn superoxide dismutase genes are developmentally regulated and respond to light and stress. Plant Mo1 Biol17: 745-760 Sambrook J, Fritsch T, Maniatis F (1989) Molecular Cloning: A Laboratory Manual, Ed 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Schreiber U, Neubauer C (1990) 02-dependentelectron flow, membrane energization and the mechanism of non-photochemical quenching of chlorophyll fluorescence. Photosynth Res 2 5 279-293 Shaaltiel Y, Chua N-H, Gepstain S, Gressel J (1988) Dominant pleiotropy controls enzymes co-segregating with paraquat resistance in Conyza bonariensis. Theor Appl Genet 7 5 850-856 Tanaka K, Saji H, Kondo N (1988) Immunological properties of Spinach glutathione reductase and inductive biosynthesis of the enzyme with ozone. Plant Cell Physiol29 637-642 Tsang EWT, Bowler C, Herouart D, Van Camp W, Villarroel, R, Genetello C, Van Montagu M, Inze D (1991) Differential regulation of superoxide dismutases in plants exposed to environmental stress. Plant Cell 3: 783-792 Whitmore FW, Kriebel HB (1987) Expression of a gene in Pinus strobus ovules associated with fertilization and early embryo development. Can J For Res 17: 408-412 Williamson JD, Scandalios JG (1992) Differentialresponse of maize catalases and superoxide dismutases to the photoactivated funga1 toxin cercosporin. Plant J 2: 351-358 Wingsle G (1989) Purification and characterisation of glutathione reductase from Scots pine needles. Physiol Plant 76: 24-30 Wingsle G, Gardestrom P, Hallgren J-E, Karpinski S (1991) Isolation, purification, subcellular localization of isozymes of superoxide dismutase from Scots pine (Pinus sylvestris L.) needles. Plant Physiol95 21-28 Wingsle G, Hallgren J-E (1993) Influence of SOt and NO2 exposure on glutathione, superoxide dismutase and glutathione reductase activities in Scots pine needles. J Exp Bot 4 4 463-470 Young LCT, Conn EE (1956) The reduction and oxidation of glutathione by plant mitochondria. Plant Physiol 31: 205-211