On the relationship between isoprene emission and thermotolerance in Phragmites australis leaves exposed to high temperatures and during the recovery from a heat stress

Authors


Dr V. Velikova. Fax: + 359 2 73 99 52; e-mail: violet@obzor.bio21.bas.bg

ABSTRACT

Thermotolerance induced by isoprene has been assessed during heat bursts but there is little information on the ability of endogenous isoprene to confer thermotolerance under naturally elevated temperature, on the interaction between isoprene-induced thermotolerance and light stress, and on the persistence of this protection in leaves recovering at lower temperatures. Moderately high temperature treatment (38 °C for 1.5 h) reduced photosynthesis, stomatal conductance, and photochemical efficiency of photosystem II in isoprene-emitting, but to a significantly lower extent than in isoprene-inhibited Phragmites australis leaves. Isoprene inhibition and high temperature independently, as well as together, induced lipid peroxidation, increased level of H2O2, and increased catalase and peroxidase activities. However, leaves in which isoprene emission was previously inhibited developed stronger oxidative stress under high temperature with respect to isoprene-emitting leaves. The heaviest photosynthetic stress was observed in isoprene-inhibited leaves exposed to the brightest illumination (1500 µmol m−2 s−1) and, in general, there was also a clear additive effect of light excess on the formation of reactive oxygen species, antioxidant enzymes, and membrane damage. The increased thermotolerance capability of isoprene-emitting leaves may be due to isoprene ability to stabilize membranes or to scavenge reactive oxygen species. Irrespective of the mechanism by which isoprene reduces thermal stress, isoprene-emitting leaves are able to quickly recover after the stress. This may be an important feature for plants coping with frequent and transient temperature changes in nature.

INTRODUCTION

Isoprene (2-methyl, 1,3,-butadiene) is emitted from mature, photosynthesizing leaves of many plant species (Kesselmeier & Staudt 1999). It is an important trace gas that affects the oxidative capacity of the atmosphere (Fehsenfeld et al. 1992). Isoprene emission is regulated by many environmental variables, the most important being temperature and light (Sanadze & Kursanov 1966; Monson & Fall 1989; Loreto & Sharkey 1990). Isoprene is emitted in the presence of sunlight and exhibits an exponential increase with temperature to a maximum level near 40 °C (Guenther et al. 1993). The rapid response of isoprene emission rate to light and moderate temperature is similar to that observed for foliar photosynthesis (Guenther, Monson & Fall 1991; Sharkey, Loreto & Delwiche 1991; Harley, Guenther & Zimmerman 1996), but the relationship between isoprene and photosynthesis is broken in many cases, such as under extremely high temperature (Loreto & Sharkey 1990), elevated CO2 (Rosenstiel et al. 2003), or stress events (Sharkey & Loreto 1993). Light controls isoprene emission through the production of photosynthetic metabolites and supply of ATP that are both used in the plastidic methylerythritol 4-phosphate (MEP) pathway for isoprene biosynthesis (Lichtenthaler 1999).

The question why plants emit isoprene has attracted the attention of many researchers. Sharkey & Singsaas (1995) first proposed that isoprene has a role in thermotolerance. However, the mechanism for this protection against high temperature damage has yet to be shown. Environmental temperature is a critical factor in the life of almost all organisms. High temperature affects the photosynthetic functions of plants by its effect on the rate of chemical reactions and on structural organization. It has been previously reported that high temperatures are responsible for changes in the thylakoid membranes, altering not only their physicochemical properties, but also their functional organization (Berry & Björkman 1980). Many physiological factors could be involved in the heat stress injury, including secondary oxidative stress. Oxidative stress arises from an imbalance in the generation and metabolism of reactive oxygen species. Reactive oxygen species include toxic oxygen species such as superoxide radicals (O2), hydrogen peroxide (H2O2), and hydroxyl radicals (OH) (Bowler, Van Montagu & Inze 1992; Foyer, Leandais & Kunert 1994; Inze & Van Montagu 1995). However, plants have evolved antioxidant mechanisms (enzymatic and non-enzymatic), by which reactive oxygen species are removed from the cell (Noctor & Foyer 1998). Catalase and peroxidase are two major systems for the enzymatic removal of H2O2 in plants (Willekens et al. 1995).

We formerly observed that isoprene effectively reduces the accumulation of reactive oxygen species and protects membranes from denaturation under elevated temperatures (Velikova, Pinelli & Loreto 2004b). However, in that experiment we were not able to induce thermotolerance by manipulating isoprene synthesis. It was found that by increasing temperature from 30 to 48 °C, photosynthesis decreased both in isoprene-emitting and isoprene-inhibited leaves when leaf discs were exposed to short-term temperature treatment (30, 38, 44 and 48 °C for 15 min each) in the light (840 µmol m−2 s−1). We speculated that the temperature used exceeded the temperature at which isoprene might have exerted a protective effect, or that isoprene removal might have caused a stress prior to the temperature stress. This activated per se a protective antioxidant mechanisms and this cross-protection mechanism efficiently worked when the temperature stress was subsequently imposed. We also thought that this mechanism could have been less efficient if leaves were also exposed to bright light rather than to the moderate light of our experiment. Those results induced us to further explore the role of isoprene in thermotolerance. We investigated the direct effect of exposure to high temperature stress (38 °C) at different light intensities, and for a period reasonably similar to that experienced in nature (1.5 h), and the recovery from this stress in isoprene-emitting and isoprene-inhibited leaves. Our objectives were to determine whether isoprene protection is mediated by light, and whether isoprene presence also induces a faster or stronger recovery from stress.

MATERIALS AND METHODS

Plant material

Reed (Phragmites australis L.) plants were grown in a controlled-environment cabinet (Fitotron; Sanyo-Gallenkampt, Loughborough, UK) under a 14/10 h light/dark regime at 30/25 °C day/night temperature, and under photon flux density at the level of the leaves 700 µmol photons m−2 s−1. The relative humidity was 60%.

Protocol, gas exchange and fluorescence measurements

Measurements were made on detached leaves. The basal part of the leaf was immersed in a beaker containing water while the central part was clamped in the gas-exchange cuvette and exposed to a synthetic air made mixing O2, N2, and CO2 (20%, 80%, 370 p.p.m., respectively) but devoid of  contaminants  and  isoprenoids.  A  flow  rate of 300 µmol s−1 was used. Relative humidity was set at 35–40%; leaf temperature was set at 30 °C. The leaf was exposed to either 500, 1000 or 1500 µmol photons m−2 s−1 of incident light intensity. When photosynthesis and isoprene emission were stable, the temperature of the cuvette was raised to 38 °C. The temperature increase was quickly achieved (within 5 min). All environmental parameters were controlled with the control devices of the gas-exchange system LI-6400 (LI-COR, Lincoln, NE, USA). The elevated temperature treatment lasted 90 min, after which the original temperature (30 °C) was restored. Gas exchange and chlorophyll fluorescence measurements before, during and after the high temperature treatment were made using the same LI-6400, which allows simultaneous measuring of gas exchange and fluorescence over the same leaf area. A set of measurements at 1000 µmol photons m−2 s−1 were carried out under non-photorespiratory conditions (1.5% O2). Electron transport rate was calculated from measurements of variable and maximal fluorescence under the different light intensities, as shown by Loreto et al. (1994). Isoprene emission was measured on-line by diverting the air at the exit of the gas-exchange cuvette into a portable gas chromatograph (Syntech Spectras BTX Analyser GC 855; Syntech, Groningen, The Netherlands) as detailed elsewhere (Loreto & Velikova 2001). At the end of each measurement the leaf disc was frozen in liquid nitrogen and then used for the biochemical assays.

Treatments with fosmidomycin

Isoprene emission was inhibited by feeding 5 µm fosmidomycin, a compound that specifically blocks the pathway of isoprenoid synthesis in chloroplasts (Zeidler et al. 1998), through the transpiration stream. Isoprene emission was inhibited to a constant level (< 10% of the original emission) within 40–60 min after fosmidomycin feeding. In our previous experiments we tested whether fosmidomycin has any effect on photosynthesis. No changes in photosynthesis and electron transport rate after the end of the 8-h treatment were observed (Loreto & Velikova 2001).

Enzyme extraction and assays

The enzymes were extracted from frozen leaf tissue (0.1 g) ground in liquid nitrogen into a fine powder. The powder was transferred to a pre-cooled (4 °C) mortar and pestle with 1 mL of 10 mm potassium phosphate buffer (pH 7.0), containing 4% (w/v) polyvinylpyrrolidone (Mr 25 000). The resultant suspension was centrifuged at 12 000 × g for 30 min and the supernatant obtained was used as enzyme extract.

All spectrophotometric analyses were conducted at room temperature on an UV/visible spectrophotometer (Perkin Elmer Lambda Bio20; Perkin Elmer, Norwalk, CT, USA). Catalase was assayed by monitoring the consumption of H2O2 at 240 nm. The reaction mixture (1 mL final volume) contained 10 mm potassium phosphate buffer (pH 7.0), 0.2 mL enzyme extract and 10 mm H2O2. The activity was calculated using the extinction coefficient 40 mm−1 cm−1 for H2O2 (Kato & Shimizu 1987). Guaiacol peroxidase was determined by measuring the oxidation of guaiacol at 470 nm. The assay mixture (1 mL) contained 10 mm potassium phosphate buffer (pH 7.0), 0.1 mL enzyme extract, 0.6 mL guaiacol 1% (w/v) aqueous solution and 0.05 mL of 100 mm H2O2. The linear initial reaction was used to estimate the activity, which was expressed in µmol of guaiacol dehydrogenation product (GDHP) formed, using the extinction coefficient of 26.6 mm−1cm−1 (Dias & Costa 1983 with some modifications).

Determination of H2O2 content

Hydrogen peroxide levels were determined as described by Velikova, Yordanov & Edreva (2000). Leaf tissues (0.1 g) were homogenized in an ice bath with 1 mL 0.1% (w/v) trichloroacetic acid (TCA). The homogenate was centrifuged at 12 000 × g for 15 min and 0.5 mL of the supernatant was added to 0.5 mL 10 mm potassium phosphate buffer (pH 7.0) and 1 mL 1 m KI. The absorbance of the supernatant was read at 390 nm. The content of H2O2 was given on a standard curve.

Malonyldialdehyde estimation

For the measurements of lipid peroxidation in leaves, the thiobarbituric acid (TBA) test, which determines malonyldialdehyde (MDA) as an end product of lipid peroxidation (Heath & Parker 1968), was used. Leaf material (0.1 g) was homogenized in 1 mL 0.1% (w/v) TCA solution. The homogenate was centrifuged at 12 000 × g for 15 min and 0.5 mL of the supernatant was added to 1 mL 0.5% (w/v) TBA in 20% TCA. The mixture was incubated in boiling water for 30 min, and the reaction stopped by transferring the reaction tubes to an ice bath. Then the samples were centrifuged at 10 000 × g for 5 min, and the absorbance of supernatant was read at 532 nm. The value of non-specific absorption at 600 nm was subtracted. The results were recorded as thiobarbituric acid reactive substances (TBARS), which represent MDA equivalents. The amount of MDA was calculated from the extinction coefficient 155 mm−1 cm−1.

Statistical analysis

Analysis of variance was performed to test for the significance of the difference between treatments of all measured parameters, using anova. Means ± SE (n = 6) were statistically separated by Tukey's test, P > 95%.

RESULTS

At 30 °C, isoprene emission increased with increasing light intensity (Fig. 1). When the temperature of the leaves was increased to 38 °C, isoprene emission, measured at the end of the high-temperature treatment, increased approximately two-fold over the initial rate. The emission decreased again to its original rate when the 30 °C temperature was restored. Residual isoprene emission after feeding fosmidomycin also increased during the high temperature treatment in leaves illuminated with 1000 and 1500 µmol m−2 s−1 but to a lesser extent than isoprene emission rate in the leaves that were not fed fosmidomycin. Isoprene emission did not change considerably in isoprene-inhibited leaves illuminated with 500 µmol m−2 s−1, because of the high-temperature treatment and recovery (Fig. 1).

Figure 1.

Effect of a high temperature treatment (38 °C for 1 h 30 min) at increasing light intensities (panels a, b, c) on isoprene emission of Phragmites australis leaves. Pair of bars represent isoprene emission of control (white) and fosmidomycin-fed (striped) leaves sampled at 30 °C before the heat treatment (first pair), at 38 °C, during the heat treatment (second pair) and after a 6-h recovery from the heat treatment again at 30 °C (third pair). Results are means ± SE (n = 6). Means were separated by Tukey's test and different letters indicate means statistically different with P > 95%.

Unlike isoprene emission, photosynthesis dropped when the leaf temperature was increased from 30 to 38 °C. Photosynthesis inhibition was 24, 38 and 47% of the rates measured at 30 °C at 500, 1000 and 1500 µmol m−2 s−1, respectively (Fig. 2). Six hours after returning the leaf temperature to 30 °C, however, photosynthesis recovered completely in leaves illuminated with 500 µmol m−2 s−1, whereas recovery was 73 and 65% at 1000 and 1500 µmol m−2 s−1, respectively. In fosmidomycin-fed, isoprene-inhibited leaves, the inhibition of photosynthesis at 38 °C was stronger than in isoprene-emitting leaves, at each light intensity. The recovery was absent or very limited at the three light intensities of the experiment.

Figure 2.

Photosynthesis of Phragmites australis leaves in response to a high temperature treatment (38 °C for 1 h 30 min) at increasing light intensities (panels a, b, c) and after recovery from heat treatment at 30 °C. Bar assignment and treatments were as in Fig. 1, but photosynthesis was measured three times during recovery (after 1.5, 3 and 6 h). Results are means ± SE (n = 6). Means were separated by Tukey's test and different letters indicate means statistically different with P > 95%.

The decline in photosynthesis with increasing temperature was associated with a reduced stomatal conductance in both isoprene-emitting and isoprene-inhibited leaves (Fig. 3). However, while reduction of stomatal conductance was similar in the two treatments, the inhibition of photosynthesis was more suppressed in isoprene-inhibited than in isoprene-emitting leaves, suggesting that changes in stomatal conductance do not account for the more negative effect of heat on photosynthesis in isoprene-inhibited leaves.

Figure 3.

Stomatal conductance of Phragmites australis leaves in response to a high temperature treatment (38 °C for 1 h 30 min) at increasing light intensities (panels a, b, c) and after recovery from heat treatment at 30 °C. Bar assignment, treatments, replications and statistical treatment as in Fig. 2.

A small drop in electron transport rate estimated by chlorophyll fluorescence measurements was observed when the temperature was increased to 38 °C in isoprene-emitting leaves exposed to 500 and 1000 µmol m−2 s−1 of light (Fig. 4a & b). The negative effect was stronger in leaves exposed to 1500 µmol m−2 s−1 of light (Fig. 4c). After returning to the temperature of 30 °C electron transport rate further decreased and remained nearly constant during the recovery (Fig. 4c). In isoprene-inhibited leaves, the electron transport rate was significantly reduced by the high-temperature treatment in comparison with isoprene-emitting leaves. The effect was more pronounced at the highest light intensity, at which the electron transport rate continued to drop, also during the recovery (Fig. 4c).

Figure 4.

Electron transport rate of Phragmites australis leaves in response to a high temperature treatment (38 °C for 1 h 30 min) at increasing light intensities (panels a, b, c) and after recovery from heat treatment at 30 °C. Bar assignment, treatments, replications and statistical treatment as in Fig. 2.

By plotting the electron transport rate of each leaf versus the corresponding photosynthetic rate we were able to estimate the number of electrons required to fix a mole of CO2(Fig. 5). This number increased with light intensity and, at each light intensity, increased during the exposure to high temperature. However, after the recovery, the ETR/A ratios were similar to those observed before the heat treatment. The ETR/A ratio was generally unaffected by isoprene inhibition, except under elevated light intensity. In this case a highly significant increase of the ETR/A ratio was observed in isoprene-inhibited leaves during the heat treatment and during the early stages of recovery, with respect to isoprene-emitting leaves.

Figure 5.

Ratio between rates of electron transport and photosynthesis in Phragmites australis leaves in response to a high temperature treatment (38 °C for 1 h 30 min) at increasing light intensities (panels a, b, c) and after recovery from heat treatment at 30 °C. Bar assignment, treatments, and replications as in Fig. 2. Pair means (isoprene-emitting and isoprene-inhibited leaves) were separated by Tukey's test and were not statistically different except than during and immediately after the heat treatment at the brightest light intensity (panel c) when statistical difference with P > 99% is shown by ***.

The levels of H2O2 and MDA increased with increasing light intensity (Fig. 6). When isoprene synthesis was prevented by fosmidomycin, a further increase in the level of H2O2 and MDA was observed in comparison with isoprene-emitting leaves sampled at the same light intensity. High temperature also induced an increase in both H2O2 and MDA content, at the three light intensities. These changes were observed in isoprene-emitting as well as in isoprene-inhibited leaves, but the content of H2O2 and MDA remained significantly higher in the latter. During the recovery both products of oxidative stress decreased but in isoprene-inhibited leaves they remained significantly higher than in isoprene-emitting leaves.

Figure 6.

Contents of hydrogen peroxide (H2O2, left sides of panels) and malonyldialdehyde (MDA, right side of panels) of Phragmites australis leaves in response to a high temperature treatment (38 °C for 1 h 30 min) at increasing light intensities (panels a, b, c) and after a 6-h recovery from heat treatment at 30 °C. Bar assignment, treatments, replications and statistical treatment as in Fig. 1.

Both catalase and peroxidase activities measured at 30 °C were higher in isoprene-inhibited than in isoprene-emitting leaves (Fig. 7). These changes were more pronounced under high light intensity (1000 and 1500 µmol m−2 s−1). Heat stress also caused an induction of catalase and peroxidase activities but this induction was more pronounced in isoprene-inhibited than in isoprene-emitting leaves, especially at low or moderate light intensities. At the highest light intensity no further increase of peroxidase activity during the heat stress and after the recovery was observed in isoprene-inhibited leaves, and this was coincident with a peak induction of H2O2 and MDA.

Figure 7.

Catalase (left sides of panels) and peroxidase (right side of panels) activities of Phragmites australis leaves in response to a high temperature treatment (38 °C for 1 h 30 min) at increasing light intensities (panels a, b, c) and after a 6-h recovery from heat treatment at 30 °C. Bar assignment, treatments, replications and statistical treatment as in Fig. 1.

DISCUSSION

The present study confirmed that thermotolerance is reduced when isoprene synthesis is inhibited (Sharkey & Singsaas 1995, Singsaas et al. 1997) and revealed that recovery from high temperature stress is also restrained in the absence of isoprene. Whereas it has been shown that thermotolerance may be increased by isoprene at temperatures higher than 40 °C (Singsaas et al. 1997; Sharkey, Chen & Yeh 2001), in our experiments we show that a better thermal protection may be induced by isoprene also at temperatures below 40 °C. Perhaps this underlies two different mechanisms of thermal protection associated with isoprene production. In any case, the incomplete recovery of photosynthesis in isoprene-inhibited leaves indicates that the photosynthetic structures are permanently damaged in these leaves.

The negative effect of heat on photosynthesis was exacerbated, especially in isoprene-inhibited leaves, under bright illumination, which underlies a possible interaction between isoprene emission and the photochemistry of photosynthesis. Heat damage arises from inactivation of the highly sensitive water-splitting reaction, disconnection of PSII centres from the bulk pigments, thermal uncoupling of photophosphorylation, and biomembrane lesions (Berry & Björkman 1980). Emani et al. (1994) have suggested that the release of the manganese-stabilizing 33 kDaA protein of PSII occurs first, and then liberation of the manganese atoms and the loss of oxygen-evolving complex take place. Hideg et al. (2000) have shown that singlet oxygen production in inactivated PSII reaction centres is a unique characteristic of photosynthetic inhibition by excess light, and a very efficient singlet oxygen-scavenging action has been attributed to isoprene (Affek & Yakir 2002; Velikova, Edreva & Loreto 2004a). There was no evidence, however, of a photoinhibitory change in photosynthetic efficiency at high temperature in isoprene-emitting and isoprene-inhibited leaves, as indicated by fluorescence yield in the dark (data not shown).

The decrease of photosynthesis at high temperature has also been attributed to an increase in the ratio of oxygenation/carboxylation activities of Rubisco. As temperature increases, the proportion of dissolved O2/CO2 and the specificity of Rubisco for O2 increases, thus favouring oxygenase activity (Sage & Sharkey 1987; Crafts-Brander & Salvucci 2002). This probably explains the clear increase of ETR/A ratio during the heat treatment, slowly returning to the original values during the recovery at moderate temperature (Fig. 5). Isoprene inhibition increased the ETR/A ratio but the increase was statistically significant only in leaves exposed to the heat treatment at the highest light intensity. Even in the absence of irreversible effects (photoinhibition), the electron transport rate has been reported to limit photosynthesis under high temperature (Wise et al. 2004). However, the increase of ETR/A ratio when photosynthesis was most severely affected by heat makes electron transport rate limitations to photosynthesis because of isoprene-inhibition unlikely. The extra electron transport rate in isoprene-inhibited leaves might have fed photorespiration or might have been used to directly reduce oxygen to toxic reactive oxygen species. The latter fate would explain why thermal damage is exacerbated by high light in isoprene-inhibited leaves, and why, in fact, a certain damage of photosynthesis can also be observed in these leaves even prior to the high temperature treatment (Fig. 2).

Increased H2O2 and MDA level and higher catalase and peroxidase activities in isoprene-inhibited leaves as compared to the isoprene-emitting ones also indicated that the inhibition of isoprene biosynthesis in Phragmites leaves provoked an oxidative stress which may be related to membrane damage, confirming the powerful antioxidant role of isoprene (Loreto & Velikova 2001). The reductive cleavage of H2O2 yields the hydroxyl radical, an extremely reactive species capable of oxidizing protein groups, mutagenizing DNA and initiating lipid peroxide chain reactions, thereby leading to membrane disruption (Noctor, Veljovic-Jovanovic & Foyer 2000). It is now apparent that H2O2 acts as a signal molecule to induce a range of molecular, biochemical and physiological responses within cells and plants (Neill et al. 2002). Accumulation of H2O2 as a consequence of isoprene inhibition may serve as a signal for activating other antioxidant systems, such as the Halliwell–Asada pathway scavenging reactive oxygen species. Several studies have demonstrated that H2O2 modulates gene expression during defence responses (Levine et al. 1994; Marrs 1996; Foyer et al. 1997; Desican et al. 1998; Etienne et al. 2000). However, a dual role of H2O2 in plants is well recognized at low concentrations, it acts as a messenger molecule involved in acclamatory signalling, triggering tolerance against various abiotic stresses, and, at high concentrations it orchestrates programmed cell death (Dat et al. 2000). Our results indicate that accumulation of H2O2 is associated with increasing reduction of photosynthesis and that isoprene is a very effective molecule in reducing both H2O2 accumulation and photosynthesis reduction, especially under high temperature and light intensity.

When Phragmites leaves were subjected to high temperature a further enhancement of oxidative species were observed (Figs 6 & 7). It is known that thylakoid membranes become leaky at moderately high temperature (Pastenes & Horton 1996; Bukhov et al. 1999). Increased MDA content indicates that heat-induced damage affects membranes but they were less pronounced in isoprene-emitting leaves (Fig. 6). The low MDA accumulation in isoprene-emitting leaves under heat stress as well as during the recovery suggested that less lipid peroxidation developed in these leaves. Sharkey & Yeh (2001) speculated that isoprene could reside in the thylakoid membrane and enhance hydrophobic interactions. Moreover, as a small lipophilic molecule isoprene might facilitate hydrophobic interactions within either membranes or protein complexes when high temperature allows large membrane-bound protein complexes (e.g. PSII) to fragment (Sharkey & Yeh 2001). It was also proposed that isoprene may stabilize lipid–lipid, lipid–protein, or protein–protein interactions (Singsaas et al. 1997).

The significant increase in catalase and peroxidase activity is also an indirect indication of the formation of free oxygen radicals in isoprene-inhibited leaves and this tendency was exacerbated by the high temperature treatment (Fig. 7). Gene disruption analysis revealed that a high activity of catalases and peroxidase was not critical for normal growth but became critical when H2O2 was added at sublethal concentrations in the growth medium (Tichy & Vermaas 1999). We speculate that the increase of H2O2 in isoprene-inhibited leaves (Fig. 6) may enhance gene expression, in turn stimulating the activities of catalases and peroxidases.

Isoprene enhancement of thermotolerance may also confirm isoprene capacity to scavenge reactive oxygen species, whose presence is exacerbated by low photosynthesis under high temperature and light. In support of this assumption it has been demonstrated that in isoprene-emitting leaves of Phragmites, singlet oxygen was not able to produce damage when produced at low to moderate concentrations (Velikova et al. 2004a). In isoprene, similarly to carotenoids, the presence of conjugated double bonds makes likely a direct reaction with reactive oxygen species and with toxic peroxynitrites, particularly if considering the chloroplast localization of its synthesis (Logan, Monson & Potosnak 2000).

An interesting question is whether the protective action observed in these experiments should only be attributed to isoprene or also to other chloroplastic isoprenoids, namely carotenoids, whose biosynthesis is also inhibited by fosmidomycin. We have previously demonstrated that carotenoid content remains similar for hours after monoterpene inhibition by fosmidomycin, which probably indicates a much slower turn-over of the more complex carotenoids with respect to volatile isoprenoids (Loreto et al. 2004). However, it may be possible that carotenoid turn-over is more rapid under high light conditions, when they more relevantly contribute to photochemical protection (Verhoeven et al. 1999). Further experiments should investigate the interaction between volatile and non-volatile isoprenoids in mutually determining photo and thermal protection. It should also be investigated whether exogenous isoprene may replace endogenous isoprene in inducing thermal protection. We maintain this likely, as this has been shown in leaves exposed to temperatures higher than 40 °C (Sharkey et al. 2001), and exogenous and endogenous isoprene also induced a similar protection against oxidative stress (Loreto et al. 2001; Loreto & Velikova 2001).

In conclusion, these experiments suggest that endogenous isoprene is not only able to protect leaves against heat but also helps leaves recover when the heat stress is alleviated. Inhibition of isoprene biosynthesis caused oxidative stress resulting in increasing levels of H2O2 and MDA, and enhanced catalase and peroxidase activities. A signalling role for H2O2 is now firmly established, but the relationship between isoprene inhibition and H2O2 accumulation remains to be elucidated. The increased thermotolerance capability of isoprene-emitting leaves might be explained with isoprene capacity to stabilize membranes as well as to scavenge reactive oxygen species. This may help plants recovering fast after transient and frequent temperature stress, which certainly is an important feature in many environments.

ACKNOWLEDGMENTS

This research was supported by the NATO Collaborative Linkage Grant LST.CLG. 978838 and by a CNR-NATO Outreach fellowship to V.V. F.L. was also supported by the European Science Foundation programme ‘Volatile Organic Compounds in the Biosphere-Atmosphere System (VOCBAS)’ and by the Italian Ministry of Environment programme ‘OZONIT’. Part of the research was also supported by the bilateral project within the framework agreement between Italian National Research Council and Bulgarian Academy of Sciences.

Ancillary