Plant Physiol. (1998) 116: 259-269

Manipulation of Catalase Levels Produces Altered Photosynthesis in Transgenic Tobacco Plants¹

Louise F. Brisson², Israel Zelitch^*, and Evelyn A. Havir

Department of Biochemistry and Genetics, The Connecticut Agricultural Experiment Station, P.O. Box 1106, New Haven, Connecticut 06504

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

Constructs containing the cDNAs encoding the primary leaf catalase in Nicotiana or subunit 1 of cottonseed (Gossypium hirsutum) catalase were introduced in the sense and antisense orientation into the Nicotiana tabacum genome. The N. tabacum leaf cDNA specifically overexpressed CAT-1, the high catalytic form, activity. Antisense constructs reduced leaf catalase specific activities from 0.20 to 0.75 times those of wild type (WT), and overexpression constructs increased catalase specific activities from 1.25 to more than 2.0 times those of WT. The NADH-hydroxypyruvate reductase specific activity in transgenic plants was similar to that in WT. The effect of antisense constructs on photorespiration was studied in transgenic plants by measuring the CO₂ compensation point () at a leaf temperature of 38°C. A significant linear increase was observed in  with decreasing catalase (at 50% lower catalase activity  increased 39%). There was a significant temperature-dependent linear decrease in  in transgenic leaves with elevated catalase compared with WT leaves (at 50% higher catalase  decreased 17%). At 29°C, also decreased with increasing catalase in transgenic leaves compared with WT leaves, but the trend was not statistically significant. Rates of dark respiration were the same in WT and transgenic leaves. Thus, photorespiratory losses of CO₂ were significantly reduced with increasing catalase activities at 38°C, indicating that the stoichiometry of photorespiratory CO₂ formation per glycolate oxidized normally increases at higher temperatures because of enhanced peroxidation.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

About 90% of the dry weight of plants is derived from CO₂ assimilated by the Rubisco reaction during photosynthesis. This enzyme also catalyzes a reaction with oxygen that leads to the formation of phosphoglycolate and glycolate. The latter is metabolized by the photorespiratory pathway with the production of CO₂ in C₃ plants (Tolbert, 1980). Photorespiration can be considered wasteful because it consumes ATP, and the CO₂ released must be fixed again within the leaf. Therefore, a number of laboratories have attempted to reduce photorespiration by genetically regulating critical biochemical reactions in leaves by altering the CO₂/O₂ specificity (Ogren, 1984; Chen et al., 1990), or by screening for photorespiratory mutants (Somerville and Ogren, 1979; Zelitch, 1989; Lea and Blackwell, 1990).

Based on enzymatic studies, it has been estimated that 25% of the glycolate metabolized during photorespiration is released as CO₂ at 25°C (Jordan and Ogren, 1983). There is evidence that the stoichiometry of the CO₂ produced per mole of glycolate oxidized increases, however, under conditions favoring rapid photorespiration, such as increased O₂ and temperature (Grodzinski and Butt, 1977; Hanson and Peterson, 1985, 1986). The stoichiometry could also change in leaves with insufficient catalase activity, because excess H₂O₂ may rapidly decarboxylate ketoacids such as hydroxypyruvate and glyoxylate to generate additional CO₂ (Zelitch, 1992a). This additional loss of assimilated CO₂ might be avoided with higher catalase (EC 1.11.1.6) activity, thereby reestablishing the stoichiometry closer to 25% and increasing net photosynthesis.

The relation of catalase activity to net photosynthesis was supported by studies with a tobacco (Nicotiana tabacum) mutant selected by screening for superior growth at elevated, near-lethal O₂ levels (Zelitch, 1992a), in which a correlation was obtained between elevated catalase and decreased photorespiration. The mutant had catalase activity 1.4 times that of WT, a higher level of catalase protein, and increased net photosynthesis when photorespiration was rapid, such as at elevated temperature or O₂ levels (Zelitch, 1989; Zelitch et al., 1991).

In addition to the catalatic reaction (2H₂O₂  O₂ + 2H₂O), catalase can use H₂O₂ to oxidize organic substrates such as ethanol to acetaldehyde (H₂O₂ + CH₃CH₂OH  CH₃CHO + 2H₂O). The latter represents the peroxidatic activity of catalase. Catalases of tobacco are encoded by a small gene family (Willekens et al., 1994). Catalases with high catalatic activity (CAT-1) are the major isoforms in leaves, and there is relatively less CAT-1 in stem and sepal tissue (Havir et al., 1996).

From leaf libraries we have cloned and characterized a partial cDNA encoding CAT-1 in Nicotiana sylvestris and a full-length clone in N. tabacum (Schultes et al., 1994). To modify catalase expression in tobacco leaves and investigate the physiological role of catalase on photosynthesis, these tobacco cDNA clones and those encoding cottonseed (Gossypium hirsutum) CAT-1 (Ni et al., 1990) were fused to the constitutive CaMV 35S promoter (Rodermel et al., 1988) in the sense and antisense orientation. A cDNA clone was also inserted in the sense orientation with a chimeric promoter that combined elements from the CaMV 35S and the mannopine synthase promoters (Comai et al., 1990). Transgenic plants were analyzed for catalase activities, seedling growth with kanamycin, and NPTII cDNA (kanamycin resistance) inserts. In the present study we used a population of transgenic plants with catalase specific activities ranging from 0.20 to more than 2.0 times those of WT plants to study further the function of catalase in photorespiration. We describe the generation of transgenic plants and the stability of catalase activity over generations, and we demonstrate that in transgenic plants with reduced catalase activity, there was a significant increase in . Conversely, in plants with increased catalase,  was significantly reduced in a temperature-dependent manner. Our findings add further support for a physiological role of catalase in photorespiration and net photosynthesis at higher temperatures.

	MATERIALS AND METHODS

Top Abstract Introduction Methods Results Discussion References

Seeds of WT tobacco (Nicotiana tabacum cv Havana) Seed used for transformations were surface sterilized by a 2-min treatment in 70% ethanol and 10 to 25 min in 1.3% sodium hypochlorite solution containing 0.025% Tween 20. The seeds were rinsed twice (5 min each time) in sterile distilled water and air dried. They were germinated on sterile Petri plates containing Murashige and Skoog medium (Murashige and Skoog, 1962) without Suc or vitamins in 0.25% Phytagel (Sigma) in a growth room in continuous light (100 µmol photons m² s¹) at 28°C. Transgenic plants were first grown in the growth room and then transferred to a commercial potting mix and grown in a greenhouse, ultimately in 10.6-L plastic pots. Greenhouse plants were fertilized with a complete nutrient solution and grown at a minimum air temperature of 18°C, whereas day temperatures often ranged between 25 and 41°C.

To determine kanamycin-resistant growth in seedling progeny of selfed transgenic plants, seeds were surface sterilized as described above, and about 25 seeds were allowed to germinate on sterile Petri plates containing Murashige and Skoog medium in 0.25% Phytagel (Sigma) with 1% Suc, vitamins, and 75 or 100 µg of kanamycin/mL. Plants were scored in comparison with untreated plants after 3 to 4 weeks in the growth room under the conditions described above.

Plasmid Constructs

Plasmid pBZ1 was constructed by cloning a 0.7-kb fragment of N. sylvestris catalase cDNA, cut with HindIII/EcoRI from pZZ1.4, into pAC1352L to obtain the catalase sequences in the antisense orientation relative to the CaMV 35S promoter. Plasmid pBZ2 was constructed by cloning a 1.9-kb N. tabacum cDNA fragment cut with SalI/SmaI from pZ2S and into pAC1352L to obtain catalase sequences in the sense orientation relative to the CaMV 35S promoter. Plasmid pBZ3 was obtained by replacing the SacI/SmaI GUS-encoding sequences in pCGN7366 with a SacI/HincII 1.9-kb N. tabacum catalase sequence from pZ2S to obtain catalase sequences in the sense orientation relative to the Big Mac promoter. The XhoI fragment from pBZ3 (containing the Big Mac and catalase sequences) was cloned into the SalI site of the binary vector pBIN19 (Clontech, Palo Alto, CA) to create pBZ8.

For plasmids pBZ4 and pBZ5, the 1.8-kb EcoRI cotton catalase sequence pC9 was cloned into the EcoRI site of pAC1352L. For plasmids pBZ6 and pBZ7, the 1.9-kb tobacco catalase fragment pZ2A was similarly cloned into the EcoRI site of pAC1352L. Five cotton catalase-containing plasmids and five tobacco catalase-containing plasmids in pAC1352L were introduced into A. tumefaciens. Each construct was independently transformed into plants. These ligations resulted in either sense or antisense constructs, and were identified by measuring leaf catalase specific activity in transgenic plants and comparing it with the specific activity in WT plants.

Plant Transformations

Catalase and NADH-Hydroxypyruvate Reductase Specific Activity

To identify transgenic plants with altered levels of catalase specific activity, assays were conducted on different leaves of each plant at intervals of at least 1 week, and specific activities (units per milligram of protein) were compared with assays made on at least four WT plants of similar size sampled at the same time. The mean catalase specific activities for WT leaves ranged from 200 to 350 units/mg protein in different experiments, and the values were lower in winter than in summer, as has previously been observed for greenhouse-grown plants (Zelitch, 1990b). However, on a given day the sd for WT plants was generally not greater than 15% from the mean. Thus, transgenic plants with altered catalase phenotypes were selected if they were consistently at least ± 30% from the WT mean (more than 2 sd, or 94% of all observations) on at least two sampling days 1 week apart.

NADH-hydroxypyruvate reductase assays were carried on leaf extracts used for catalase determinations after each set of four catalase assays was completed (Zelitch, 1990a). The reaction mixture consisted of 167 mm potassium phosphate buffer (pH 6.4), 0.05 mm NADH, and enzyme extract. The rate of NADH oxidation was measured at 340 nm at 30°C for 1 min upon addition of 1 mm lithium hydroxypyruvate. One unit was defined as the activity catalyzing the reduction of 1 µmol of hydroxypyruvate per min.

DNA and RNA Analysis

Leaf samples were taken for RNA analysis at the same time or within 24 h after discs were taken for catalase activity. The RNA samples were frozen in liquid N₂ and stored at 70°C until analyzed. Total RNA was isolated from frozen material using the guanidinium thiocyanate method (Mehdy and Lamb, 1987), and RNAs were separated by electrophoresis in a 1% agarose gel containing 6.5% formaldehyde and transferred to a Nytran membrane. Filters were then UV cross-linked, heated at 80°C for 2 h, prehybridized in Church buffer (Church and Gilbert, 1984), hybridized overnight with randomly labeled cDNA coding for the N. tabacum catalase gene (Schultes et al., 1994), and washed as described above for DNA analysis. To assess RNA quality and quantity, a N. sylvestris SSu probe was used (a gift of Alice Cheung, Yale University). The quantitative intensity was determined by applying densitometry to video images of the blots.

Catalase Isozymes in Stem and Sepal Tissue

CO₂ Compensation Point and Dark Respiration

1

Two 1.6-cm leaf discs, excised in late morning on sunny days from the tip-half region of WT or transgenic plants in the T₂ generation, were placed upside down on 1.0 mL of water in a 50-mL syringe the tip of which was sealed with a rubber serum stopper. Syringes were placed in constant-temperature chambers in the light (100 µmol photons m² s¹) with the leaf temperatures inside the syringe maintained at 38 ± 0.5°C or 29 ± 0.5°C as determined with a thermistor thermometer placed on a leaf disc in a separate syringe. Gas samples (5 mL) were withdrawn from the syringe through the serum stopper. They were injected into a IR CO₂ gas analyzer (model 865, Beckman) after 90 min and at 30-min intervals thereafter to ensure that a steady-state CO₂ concentration was obtained (± 5%). This method of determining  showed clear differences between C₃ and C₄ leaf segments (Schultes et al., 1996). At the same time that discs were taken for  determination, or within 24 h, two 0.8-cm leaf discs were taken from nearby positions on the same leaves and their catalase specific activities were determined as described above.

Dark-respiration rates were measured in some experiments after determination of  at 38°C in transgenic and WT leaf tissue after turning off the light in the chambers. The leaf temperature in the dark was maintained at 36 ± 0.5°C, and rates of CO₂ evolution in transgenic and WT plant tissue were determined by measuring the CO₂ concentrations in the 50-mL syringes after 15 min of darkness (0 time), and for two 10-min intervals thereafter.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Catalase Activity in Transgenic Plants

Transgenic plants were first selected by their ability to grow on selective medium (Murashige and Skoog with kanamycin). Transgenic families with antisense constructs and low catalase activity grew poorly in the presence of kanamycin (100 µg/mL) in normal air, even though NPTII was present, but growth usually improved and plants recovered in 1% CO₂ when photorespiration was eliminated. Thus, a greatly reduced level of catalase activity appeared to increase the sensitivity of the plants to higher levels of kanamycin in normal air.

Leaf genomic DNAs of plants with kanamycin resistance and altered specific activity of leaf catalase were also screened for the presence of NPTII. Southern-blot analysis of EcoRI-digested DNA showed that plants with altered catalase activity contained one or more copies of NPTII, and the location of the bands hybridizing with a NPTII probe varied in different transformants. Transformed plants contained additional bands hybridizing with our leaf catalase probe that were not present in untransformed control plants (data not shown). Some plants containing NPTII had catalase activities similar to untransformed plants (see Table II).

The range of catalase activities in different transformed families relative to WT plants is shown in Table I. WT plants generally showed a sd within 15% of the mean (see ``Materials and Methods''). Specific activities from 0.20 to 0.75 times those of WT plants were obtained with the antisense constructs, and activities less than 0.5 times those of WT plants were observed only when the full-length clone and not the partial segment of the catalase gene was inserted in the antisense orientation. There was no correlation between the number of genes introduced and the level of inhibition or enhancement of activity. For example, transformant 4.7D (catalase activity, 0.62 times that of WT; Table I) contained at least two NPTII inserts in its genome, but a higher inhibition was obtained in transformed plant 5.13.13 (catalase activity, 0.24 times that of WT), which had only one introduced copy of NPTII (data not shown).

Greatly decreased catalase levels (about 0.30-0.40 times that of the specific activities found in WT) often slowed growth and caused the plants to become yellow in the greenhouse, as was previously observed with a barley mutant with catalase activity 0.05 times that of WT (Kendall et al., 1983) and in transgenic tobacco with an antisense catalase construct (Chamnongpol et al., 1996). Transformant plants 5.13.13 and 7.2.3 (Table I) were examples of plants that yellowed and grew poorly in the greenhouse. However, when they were transferred to a growth room with lower light and temperature and elevated CO₂ to slow photorespiration (continuous 100 µmol photons m² s¹; 450 µL CO₂ L¹; 27°C), they promptly regained their green color and later developed normal flowers and seed.

Overexpression of the cottonseed catalase gene produced only modest increases in catalase specific activity (about 1.36 times that of WT; Table I), whereas introduction of the tobacco leaf catalase gene with the CaMV 35S promoter yielded increases of about 1.5 times that of WT. The greatest increase in catalase activity (2.0 times that of WT and frequently greater) was obtained in transformants containing the Big Mac promoter, although some transformed plants with this promoter also showed increases about 1.5 times that of WT (Table I).

Transmission of Altered Catalase Phenotype

Catalase mRNA Levels

View larger version (58K): [in this window] [in a new window]   Figure 1. Northern-blot analysis of RNA from WT and transgenic plants (9.8, 2.2A, 2.3B; Table I) with altered catalase specific activities. The blots were obtained on Nytran membranes using total RNA (15 µg) from the tip region of leaves that was hybridized to radiolabeled N. tabacum catalase DNA and SSu DNA. Plant 9.8, mean catalase activity, 2.04 times that of WT, plant 2.2A; mean catalase activity, 0.62 times that of WT; and plant 2.3B, mean catalase activity, 0.66 times that of WT. The catalase signals relative to WT were: plant 9.8, 0.3×; plant 2.2A, 0.5×; plant 2.3B, 0.2×. The SSu signals relative to WT were: plant 9.8, 1.7×; plant 2.2A, 2.1×; plant 2.3B, 2.9×.

Catalase Isoform Profiles

View larger version (28K): [in this window] [in a new window]   Figure 2. Elution profile showing separation by chromatofocusing of catalase isozymes in stems and sepals of WT and transgenic tobacco in which catalase activity was overexpressed in leaves. The fractions encompassing the catalase isoform with enhanced peroxidatic activity are shown in the shaded areas. Recoveries of catalase activities of 79 to 90% were obtained from the chromatofocusing columns. A, An extract of stem tissue (18 g) of transformant plant 9.8 (Table I) with a leaf catalase specific activity 2.51 times that of WT yielded 3830 units of catalase activity and 30.5 mg of protein for chromatofocusing. B, An extract of WT stem tissue (31 g) yielded 6560 units of catalase activity and 44.8 mg of protein for chromatofocusing. C, An extract of sepal tissue (4.1 g) of transformant plant 9.11 (Table I) with a leaf catalase specific activity 2.50 times that of WT yielded 1800 units of catalase activity and 10.2 mg of protein for chromatofocusing. D, An extract of WT sepal tissue (5.0 g) yielded 4060 units of catalase activity and 16.5 mg of protein for chromatofocusing.

There were some striking differences, however, in the distribution of catalase isoforms in extracts of stem and sepal tissue between transformants with about 2.5-fold greater leaf catalase specific activity compared with WT plants (Fig. 2). In WT stem tissue CAT-3 was a major isoform (Fig. 2B), but in transgenic overexpressor plant 9.8 the forms with high-catalatic activity (CAT-1) were increased at least 3-fold (Fig. 2A) relative to CAT-3. CAT-3 was also a major isoform in WT sepals (Fig. 2D), but in transgenic overexpressor plant 9.11 the predominantly high-catalatic forms were greatly enhanced compared with CAT-3 (Fig. 2C). Thus, our cloned N. tabacum leaf cDNA (Schultes et al., 1994) was clearly shown to encode CAT-1-type activity.

NADH-Hydroxypyruvate Reductase Activities in Transgenic Plants with Altered Catalase Activities

View larger version (16K): [in this window] [in a new window]   Figure 3. Comparison of NADH-hydroxypyruvate reductase and catalase specific activities in transgenic plants in the T2 generation with decreased and increased catalase specific activities relative to WT. The ratio of transgenic-to-WT activities of NADH-hydroxypyruvate reductase and catalase were determined on the same leaf extracts. The solid line is a least-squares fit with a slope of 0.009 and r2 = 0.001 (P = 0.97) over a range of catalase activities that varied from 0.2 to 2.0 times that of WT.

The CO₂ Compensation Point

1

View larger version (16K): [in this window] [in a new window]   Figure 4. The effect of antisense catalase constructs in the T2 generation on  at a leaf temperature of 38°C. The results are expressed as a ratio of transgenic-to-WT activities. Transgenic leaves were from the progeny of three different self-pollinated transformants containing the pBZ1 construct (Tables I and II). The mean  for WT (± se) was 133 ± 6.3 µL CO2 L1 (n = 6). Each point represents experiments done on different leaves on different days. The solid line is a least-squares fit with a slope of 0.87 ± 0.41 and r2 = 0.36 (P < 0.05).

Photorespiration increases greatly relative to net photosynthesis with increasing temperature (Hanson and Peterson, 1985, 1986; Zelitch, 1992a). WT leaves had a mean  of 132 µL CO₂ L¹ at leaf temperatures of 38°C, and 74 µL CO₂ L¹ at 29°C (Fig. 5). Elevated catalase levels decreased the ratio of  in transgenic relative to WT plants at 38°C and the data gave a linear fit that was highly significant (r² = 0.67). The effect at 29°C was smaller, and the slope of the line was not significant. When leaf catalase was 50% greater than that of WT, the mean  was 17% lower than that of WT at 38°C, and about 9% lower than that of WT at 29°C.

View larger version (19K): [in this window] [in a new window]   Figure 5. The effect of enhanced catalase activity in transgenic plants of the T2 generation on  at leaf temperatures of 38°C () and 29°C (). The results are expressed as a ratio of transgenic-to-WT activities. Transgenic leaves were used from progeny of two different self-pollinated transformants containing the pBZ8 construct (Tables I and II). The mean  for WT (± se) at 38°C was 132 ± 4.7 µL CO2 L1 (n = 10), and at 29°C was 74 ± 1.6 µL CO2 L1 (n = 10). For each experiment the points at leaf temperatures of 38 and 29°C represent results with transgenic leaf samples taken at the same time from the same leaf and compared with WT samples also taken at the same time. The solid lines are least-square fits with slopes of 0.21 ± 0.053 and 0.13 ± 0.071 for 38 and 29°C, respectively. The r2 = 0.67 (P < 0.01) at 38°C, and r2 = 0.32 (P > 0.10) at 29°C.

Changes in  can be caused by changes in photorespiration as well as "dark respiration." The rates of dark respiration in transgenic and WT leaves were therefore measured at leaf temperatures of 36°C after completing  determinations at 38°C (Fig. 5). The rates of dark respiration were the same in transgenic and WT leaves, with a mean ratio of transgenic-to-WT plants (± se) of 1.04 ± 0.057 (n = 8). The mean increase in CO₂ concentration (µL L¹) in the syringes with WT discs (± se) was 378 ± 4.9 after 10 min, and 759 ± 17 after 20 min. The increase in CO₂ concentration was linear and extrapolated to 0 at 0 time, consistent with a lack of effect of CO₂ concentration on the rate of dark respiration. Dark respiration undoubtedly occurs in the light, although it is uncertain that its magnitude is exactly the same as in darkness, but the results indicate that the changes in the values for shown in Figures 4 and 5 in transgenic relative to WT leaves did not result from differences in rates of dark respiration between transgenic and WT leaves.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Elevated and Reduced Catalase Activities in Transformants with Different Constructs

1

1

Increases in catalase observed with the above mutant could have been a consequence of increased photosynthesis rather than its cause, and uncertainty remained regarding whether altered catalase was fully responsible for altered photosynthesis. Therefore, we decided on a more specific transgenic approach to changing catalase levels that would specifically alter catalase levels (see Fig. 3) and especially CAT-1 (Fig. 2).

When the CaMV 35S promoter was used in constructs with cottonseed catalase DNA in the sense orientation, transformant tobacco plants showed increases in leaf catalase activity to about 1.36 times that of WT (Table I). Insertions with the full-length N. tabacum cDNA enhanced leaf catalase activity of transformants to about 1.5 times that of WT, an enhancement similar to that obtained previously with a tobacco mutant selected for O₂-resistant growth (Zelitch, 1989, 1992a). Among the transformed plants with the pBZ2 construct that was designed to obtain overexpression, there were two plants that showed consistent underexpression of catalase activity by 0.50 times or more (Table I). These plants may represent examples in which the insertion of a transgene produces co-suppression, a phenomenon sometimes encountered in transformants (Flavell, 1994).

Comai et al. (1990) demonstrated that the plasmid vector containing the Big Mac promoter could often drive the expression of the GUS gene in tomato and tobacco about 10 times higher than the CaMV 35S promoter alone. Use of the Big Mac promoter in a construct containing the N. tabacum catalase cDNA insert produced some transformants with increases of about 1.5 times that of WT, and others with an enhancement of leaf catalase specific activity that was 2.0 or more times that of WT (Table I). The variation in overexpression of catalase activity with Big Mac is consistent with the finding of Comai et al. (1990) that T₁ plants transformed with this promoter produced plants with a broad range of expression levels.

All antisense constructs used the CaMV 35S promoter (Table I). The partial N. sylvestris cDNA insert produced transformants with a mean catalase underexpression of 0.63 times that of WT, and transformed plants containing the full-length N. tabacum DNA averaged 0.46 times that of WT catalase activity. Antisense cotton transformants averaged 0.43 times that of normal catalase activity, although plant 5.13.13 had 0.24 times the activity of WT. Two transgenic plants from a sense construct also showed strongly underexpressed catalase activities, presumably because of co-suppression (Table I), and in the T₂ generation of plant 10.33 selfed, a small proportion of the progeny had decreased catalase activity (Table II).

Characteristics of Some T₂ Plants Obtained by Selfing Transgenic Plants with Altered Catalase Activity

Because not every kanamycin-resistant plant can be depended on to over- or underexpresses catalase activity, at the present time the most rapid and reliable method of detecting plants with altered catalase expression is to conduct leaf-catalase assays. Our screening assays were carried out on different leaves of each plant at intervals of at least 1 week.

Relation of mRNA Transcripts to Catalase Activity in Transformants

A large number of northern-blot hybridizations were conducted with the catalase probe on transgenic plants that underexpress and overexpress catalase specific activity compared with WT, and no correlation was usually observed between the strength of RNA transcripts and catalase specific activities in transformants that overexpress catalase (Fig. 1). It is clear that the steady-state level of mRNA is not always related to mRNA turnover, and other posttranscriptional factors including light intensity (Zelitch, 1992b) and temperature may also contribute to a lack of correlation between catalase activity and transcript level.

Nature of the Catalase Isoform Overexpressed in Transgenic Plants

Results in Figure 2, A and C, show that transformed plants greatly overexpressing catalase in N. tabacum leaves also have large activity increases in stem and sepal tissue associated with CAT-1 (eluting from the chromatofocusing column at about pH 7.5) and CAT-1-like forms (eluting between pH 7.2 and 6.4) compared with CAT-3. Thus, all of these forms with a similar biochemical activity were enhanced by constructs that overexpress our leaf catalase cDNA. Understanding of the relation of these CAT-1 and CAT-1-like forms to the clones obtained from N. plumbaginifolia (Willekens et al., 1994) must await further comparative biochemical and protein-binding studies.

Because NADH-hydroxypyruvate reductase, like catalase, is present in leaf peroxisomes, both activities were assayed in the same extracts to determine whether transgenic alteration of catalase also affected another peroxisomal enzyme (Fig. 3). A 10-fold change in catalase activity produced no significant change in NADH-hydroxypyruvate reductase activity, providing further evidence for the specificity of the cloned CAT-1.

Effect of Altered Catalase Activity on Photorespiratory CO₂ Formation and the Role of Leaf Temperature

For WT tobacco leaves at a leaf temperature of 38°C the mean  (± se) was 133 ± 5.3 µL CO₂ L¹ (Fig. 4) and 132 ± 4.7 µL CO₂ L¹ (Fig. 5). At 29°C,  was 74 ± 1.6 µL CO₂ L¹ (Fig. 5). These values for  are well within the range of published values (Tichá and Catský, 1981). Figure 4 summarizes experiments conducted to examine the role of catalase on photorespiration by measuring  at a leaf temperature of 38°C using transformed plants with reduced levels of leaf catalase. The finding of a significant linear relation between reduced catalase and increased  relative to WT (with 50% lower catalase  increased by 39%) demonstrates that unless H₂O₂ is rapidly broken down, photorespiratory CO₂ will be enhanced at higher temperatures.

In transgenic plants exhibiting catalase overexpression (Fig. 5) at a leaf temperature of 38°C, there was a highly significant linear decrease in  relative to WT with increasing catalase (at 50% higher catalase  decreased 17%), showing that WT catalase levels are insufficient to remove all of the H₂O₂ produced under high photorespiratory conditions. At leaf temperatures of 29°C (Fig. 5), the effect of enhanced catalase on decreasing  was too small to attain statistical significance, presumably because the changes were smaller and the sum of the plant-to-plant variability and measurement errors became relatively larger. However, reduced catalase levels increased photorespiration and elevated catalase levels decreased photorespiration at higher temperatures, with the relative effect being greater with reduced catalase levels.

Relation of Altered Photorespiratory CO₂ Formation by Catalase Levels and the Stoichiometry of CO₂ Produced per Glycolate Oxidized

Similarly, the highly significant effect of increasing elevated catalase levels on decreasing  at a leaf temperature of 38°C (Fig. 5) was demonstrated. It would be expected that  would increase at 38°C because of the temperature dependence of the CO₂/O₂ specificity of Rubisco. But when transformant leaves had a 50% increase in catalase activity,  was decreased by 17%, suggesting that a portion of the increase in  in WT at higher temperature was caused by peroxidation. Again, if the stoichiometry did not change in these experiments it would require a 20% increase in the CO₂/O₂ specificity factor for Rubisco (Jordan and Ogren, 1983). Changes of such a magnitude in the specificity factor are highly unlikely.

Dark respiration was not changed in transformants, although it is uncertain whether the magnitude of dark respiration is exactly the same in light as in darkness. In any event, respiratory CO₂ must be a small fraction of the total CO₂ production in  compared with the contribution of photorespiratory CO₂, or one would not observe the well-known large effect of O₂ concentration on . We are left with the inescapable conclusion that catalase levels can strongly affect photorespiratory CO₂ formation by regulating the stoichiometry of the photorespiratory pathway at higher temperatures.

Thus, models that assume a fixed specificity factor and a fixed stoichiometry at a given temperature (Jordan and Ogren, 1983) may be correct for plants grown and tested at 25°C. However, models (Hanson and Peterson, 1985, 1986) that support a fixed specificity factor and a changing stoichiometry under conditions of high photorespiration are consistent with the experimental results presented here. Our experiments suggest a novel method of regulating photorespiration by maintaining the stoichiometry of CO₂ formation closer to the expected 25% per mol of glycolate oxidized, and thereby increasing the efficiency of photosynthetic net CO₂ assimilation.

	FOOTNOTES

   Received June 30, 1997; accepted October 13, 1997.

	ABBREVIATIONS

Abbreviations: CaMV, cauliflower mosaic virus. CAT-1, the high catalytic form of catalase. , CO₂ compensation point. NPTII, gene encoding neomycin phosphotransferase. SSu, gene encoding Rubisco small subunit. WT, wild type.

	ACKNOWLEDGMENTS

We appreciate the generous counsel, vectors, and plasmids (including pACL1352L and pACL1352S) given to us by Alice Cheung (Yale University). Plasmid pC9 was a gift from Richard N. Trelease (Arizona State University); pYU179 was a gift from Stephen L. Dellaporta (Yale University); and pCGN7366 containing the Big Mac promoter was a gift from J.C. Williams (Calgene, Davis, CA). We gratefully acknowledge the technical assistance of Carol Clark, Regan Huntley, and Emelyn Solivan, and helpful discussions with colleagues Francine M. Carland, Neil P. Schultes, and Richard B. Peterson (who also conducted video image densitometry on northern blots).

	LITERATURE  CITED

Top Abstract Introduction Methods Results Discussion References

Chamnongpol S, Willekens H, Langebartels C, Van Montagu M, Inze D, Van Camp W (1996) Transgenic tobacco with a reduced catalase activity develops necrotic lesions and induces pathogenesis-related expression under high light. Plant J 10: 491-503 [CrossRef]

Chen Z, Green D, Westhoff C, Spreitzer RJ (1990) Nuclear mutation restores the reduced CO₂/O₂ specificity of ribulose bisphosphate carboxylase/oxygenase in a temperature-conditional chloroplast mutant of Chlamydomonas reinhardtii. Arch Biochem Biophys 283: 60-67 [Medline]

Church GM, Gilbert W (1984) Genomic sequencing. Proc Natl Acad Sci USA 81: 1991-1995 [Abstract/Free Full Text]

Comai L, Moran P, Maslyar D (1990) Novel and useful properties of a chimeric plant promoter combining CaMV 35S and MAS elements. Plant Mol Biol 15: 373-381 [CrossRef][ISI][Medline]

Flavell RB (1994) Inactivation of gene expression in plants as a consequence of specific sequence duplication. Proc Natl Acad Sci USA 91: 3490-3496 [Abstract/Free Full Text]

Gates DM (1980) Biophysical Ecology. Springer-Verlag, New York, pp 38-46

Grodzinski B, Butt VS (1977) The effect of temperature on glycollate decarboxylation in leaf peroxisomes. Planta 133: 261-266

Hanson KR, Peterson RB (1985) The stoichiometry of photorespiration during C₃-photosynthesis is not fixed: evidence from combined physical and stereochemical methods. Arch Biochem Biophys 237: 300-313 [Medline]

Hanson KR, Peterson RB (1986) Regulation of photorespiration in leaves: evidence that the fraction of ribulose bisphosphate oxygenated is conserved and stoichiometry fluctuates. Arch Biochem Biophys 246: 332-346 [Medline]

Havir EA, Brisson LF, Zelitch I (1996) Distribution of catalase isoforms in Nicotiana tabacum. Phytochemistry 41: 699-702 [CrossRef]

Havir EA, McHale NA (1987) Biochemical and developmental characterization of multiple forms of catalase in tobacco leaves. Plant Physiol 84: 450-455 [Abstract/Free Full Text]

Havir EA, McHale NA (1990) Purification and characterization of an isozyme of catalase with enhanced-peroxidatic activity from leaves of Nicotiana sylvestris. Arch Biochem Biophys 283: 491-495 [Medline]

Horsch RB, Fry JT, Hoffmann NI, Eicholtz D, Rogers SG, Fraley RT (1985) A simple and general method for transforming genes into plants. Science 227: 1229-1231 [Abstract/Free Full Text]

Jordan DB, Ogren WL (1983) Species variation in kinetic properties of ribulose 1,5-bisphosphate carboxylase/oxygenase. Arch Biochem Biophys 227: 425-433 [CrossRef][ISI][Medline]

Kendall AC, Keys AJ, Turner JC, Lea PJ, Mifflin BJ (1983) The isolation and characterization of a catalase-deficient mutant of barley (Hordeum vulgare L.). Planta 159: 505-511 [CrossRef][ISI]

Kilby NJ, Leyser HMO, Furner IJ (1992) Promoter methylation and progressive transgene inactivation in Arabidopsis. Plant Mol Biol 20: 103-112 [CrossRef][ISI][Medline]

Lea PJ, Blackwell RD (1990) Genetic regulation of the photorespiratory pathway. In I Zelitch, eds, Perspectives in Biochemical and Genetic Regulation of Photosynthesis. Wiley-Liss, New York, pp 301-318

Matzke MA, Matzke AJM (1995) How and why do plants inactivate homologous (trans)genes? Plant Physiol 107: 679-685 [ISI][Medline]

Mehdy MC, Lamb CJ (1987) Chalcone isomerase cDNA cloning and mRNA induction by fungal elicitor, wounding, and infection. EMBO J 6: 1527-1533 [ISI][Medline]

Murashige T, Skoog F (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol Plant 15: 473-497 [CrossRef]

Ni W, Trelease RN (1991) Post-transcriptional regulation of catalase isozyme expression in cotton seeds. Plant Cell 3: 737-744 [Abstract/Free Full Text]

Ni W, Turley RB, Trelease RN (1990) Characterization of a cDNA encoding cottonseed catalase. Biochim Biophys Acta 1049: 219-222 [Medline]

Ogren WL (1984) Photorespiration: pathways, regulation, and modification. Annu Rev Plant Physiol 35: 415-442 [CrossRef][ISI]

Redinbaugh M, Sabre M, Scandalios JG (1990) The distribution of catalase activity, isozyme protein, and transcript in the tissues of the developing maize seedling. Plant Physiol 92: 375-380 [Abstract/Free Full Text]

Rodermel SR, Abbott MS, Bogorad L (1988) Nuclear-organelle interactions: nuclear antisense gene inhibits ribulose bisphosphate carboxylase enzyme levels in transformed tobacco plants. Cell 55: 673-681 [CrossRef][ISI][Medline]

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY

Schultes NP, Brutnell TP, Allen A, Dellaporta SL, Nelson T, Chen J (1996) Leaf permease1 gene of maize is required for chloroplast development. Plant Cell 8: 463-475 [Abstract]

Schultes NP, Zelitch I, McGonigle B, Nelson T (1994) The primary leaf catalase gene from Nicotiana tabacum and Nicotiana sylvestris. Plant Physiol 106: 399-400 [Medline]

Somerville CR, Ogren WL (1979) A phosphoglycolate phosphatase deficient mutant of Arabidopsis. Nature 280: 833-836 [CrossRef]

Tai TH, Tanksley SD (1990) A rap