In past studies, it was hypothesized that reductions in chloroplast isoprene emissions at high atmospheric CO2 concentrations were caused by competition between cytosolic and mitochondrial processes for the same substrate, possibly phosphoenolpyruvate (PEP). We conducted field and laboratory experiments using leaves of white poplar (Populus alba L.) to identify whether an inverse relationship occurs between the dark respiration rate (a mitochondrial process) and the isoprene emission rate. Field experiments that were carried out in a free-air CO2-enriched (FACE) facility showed no clear effect of elevated CO2 on either isoprene emission rate or respiration rate by leaves. In young, not yet fully expanded leaves, low isoprene emission and high dark respiration rates were measured in both ambient and elevated CO2. In these leaves, isoprene emission was inversely correlated with dark respiration. It is possible to interpret from these results that, in young leaves, high rates of growth respiration compete with isoprene biosynthesis for the same substrate. However, it is also possible that the negative correlation reflects the contrasting reductions in growth respiration and increases in expression of the enzyme isoprene synthase at this final stage of leaf maturation. In contrast to our observations on young leaves, respiration rate and isoprene emission rate were positively correlated in older, fully expanded leaves (8 and 11 from apex). A positive correlation was also found between respiration rate and isoprene emission rate when these parameters were modulated using different ozone exposure, growth light intensity, growth temperature and exposure to different leaf temperatures in laboratory experiments. These data show that competition for substrate between isoprene biosynthesis and leaf respiration does not determine the rate of isoprene emission in most circumstances that affect both processes. A negative correlation was observed across all experiments between isoprene emission rate and the activity of phosphoenolpyruvate carboxylase (PEPc), a cytosolic enzyme that competes with isoprene biosynthesis for substrate. The cytosolic metabolite, PEP, occurs at a metabolic branch point from which substrate flows into three processes: (1) the production of pyruvate for mitochondrial respiration, (2) the production of oxaloacetate (OAA) by PEPc for anabolic support of mitochondrial respiration and (3) transport into the chloroplast to support chloroplastic demands for pyruvate, including isoprenoid biosynthesis. The results of our observations suggest that only the second process competes for substrate with isoprenoid synthesis, while the partitioning of PEP between mitochondrial respiration and chloroplast isoprenoid biosynthesis is controlled in a way that retains balance in substrate demand.
Isoprene (2-methyl 1,3 butadiene) is a hydrocarbon emitted by many plants at high rates. Plant emission of isoprene may account for about half of the total emission of biogenic hydrocarbons (Guenther et al. 1995). Once in the atmosphere, isoprene actively plays many roles. Especially in polluted environments, isoprene may contribute to ozone, aerosol and particulate formation (Andreae & Crutzen 1997; Monson 2002). Isoprene may also have a functional role in plants. Isoprene emission has been reported to improve plant resistance to heat (Singsaas et al. 1997) and oxidative stress (Loreto et al. 2001a), and may be involved in balancing the partitioning of metabolites between the cytosol and chloroplast (Rosenstiel et al. 2004).
It is important to know whether current and future environmental changes will affect isoprene emission by plants. CO2 concentration is rising at unprecedented fast rates (IPCC 2001) and this is likely to have a striking effect on plant biology. CO2 rise is generally beneficial for photosynthesis (Centritto, Lucas & Jarvis 2002; Long et al. 2004). Isoprene is made predominantly from photosynthetic carbon entering the methylerythritol 4-phosphate (MEP) pathway of chloroplastic isoprenoid formation (Lichtenthaler et al. 1997). Upon this discovery, it has been predicted that rising CO2 would increase isoprene emission rate as a result of a stimulation of photosynthesis. However,experiments of leaf exposure or whole plant growth at CO2 concentrations higher than ambient have shown that rising CO2 may inhibit isoprene emission. Populus species are all strong isoprene-emitting trees, and have been often used for these experiments. Isoprene is strongly inhibited by elevated CO2 in American Populus species, Populus tremuloides (Monson & Fall 1989) and Populus deltoides (Sharkey, Loreto & Delwiche 1991). Experiments with whole plants growing in large, controlled experimental facilities have later confirmed the negative impact of elevated CO2 on isoprene emission by Populus (e.g. Rosenstiel et al. 2003; Pegoraro et al. 2004, 2005). Not only is isoprene inhibited at rising CO2. Elevated CO2 is shown to inhibit monoterpenes, also formed from the MEP pathway, and emitted by Mediterranean oaks (Loreto et al. 2001b).
The reasons for the reduction of isoprene formation in poplars under elevated CO2 have been studied by Rosenstiel et al. (2003, 2004). A principal determinant appeared to be the activity of the cytosolic enzyme phosphoenolpyruvate carboxylase (PEPc), which was proposed to compete with the chloroplastic import of phosphoenolpyruvate (PEP) for the production of isoprenoid compounds. These authors also reasoned that the number of mitochondria and leaf respiration rate might both increase in leaves developed at high CO2, as previously reported (Griffin et al. 2001; Wang et al. 2001), potentially creating an even greater sink for cytosolic PEP following its conversion to pyruvate and transport into the mitochondria to support the tricarboxylic acid cycle.
Using these past studies as a foundation, we hypothesized that an inverse relationship may exist between leaf respiration rate and isoprene emission rate. If the relationship between these two functions is controlled completely by PEP substrate availability, then any experimental treatment that causes a change in the leaf respiration rate should cause a complementary and inverse change in the leaf isoprene emission rate. We tested this hypothesis by conducting measurements on leaf respiration rate and isoprene emission on leaves of Populus alba growing in a broad range of environmental conditions. The field experiment was carried out in a free-air CO2-enriched (FACE) structure, to study the relationship between isoprene and respiration in plants permanently exposed to elevated CO2, as described by Rosenstiel et al. (2003). Additional experiments were carried out in potted plants to investigate relationships when respiration is reduced (e.g. in shaded and ozonated plants) or stimulated (e.g. in plants grown or exposed to high temperature), and if it involves variations in the activity of PEPc.
MATERIALS AND METHODS
Field experiment at the FACE facility
The EUROFACE facility was located in Tuscania (central Italy, 42°20′N, altitude 150 m). The facility was described in detail elsewhere (e.g. Tricker et al. 2004), and a comprehensive description was also given at http://www.unitus.it/euroface/. Briefly, P. alba trees were exposed since 1999 to enriched CO2 concentration. The plantation was coppiced in 2001, and our experiments were therefore carried out on re-sprouts. Plants in the three CO2-fumigated rings were exposed to CO2 concentrations of 510 ± 60 ppmv, measured hourly at mid-canopy height with a Li-Cor 6400 (Li-Cor, Lincoln, NE, USA). Plants growing inside three more experimental rings were maintained at ambient (370 ppmv) CO2 concentration. Plants of both elevated and ambient CO2 rings were grown at two different N fertilization levels. Plants maintained at high N were supplied an extra amount of N of > 200 kg ha−1 year−1 during 2002 and 2003 (Calfapietra et al. 2005). Measurements were made during August on the top of the canopy (around 8 m from soil) of 2-year-old plants, during the morning hours of sunny days (average air temperature 31 ± 3 °C). Measurements were carried out on leaves from nodes 4, 8 and 11 counting from the apex of the trees. All leaves were from the main stem, and were exposed to full sunlight. Leaves 8 and 11 were fully expanded, while Leaf 4 was 40% of the size of fully expanded leaves. Leaves were considered as fully expanded when their size did not increase further for a week before measurements. Measurements were repeated on at least 5 different plants per treatment. Plants growing at enriched and natural levels of N showed no differences in isoprene emission rates, and the results collected in these two plots were therefore pooled for subsequent analysis.
Respiration rates were measured with the Li-Cor 6400 portable gas-exchange system on a 6 cm2 area of a leaf that had been dark adapted for 30 min following measurement of net photosynthesis rate and isoprene emission rate. The measurements of photosynthesis and isoprene emission were carried out at either under ambient temperature (30–35 °C) and light intensity (> 1500 µmol photons m−2 s−1). On a different batch of plants (n = 5), isoprene emission rate was measured at a controlled leaf temperature (25 or 35 °C). The CO2 concentration flowing over the leaf inside the gas-exchange cuvette was not controlled and therefore reflected the concentration of the different plots at high and ambient CO2 levels. Isoprene measurements were carried out as explained by Scholefield et al. (2004) connecting the outflow from the cuvette to a portable gas chromatograph (Syntech GC855; Syntech, Groningen, the Netherlands) with a Teflon line. Respiration rate and isoprene emission rate were measured on 3 different leaves per plant.
P. alba 1-year-old cuttings were grown outdoors in 5 L pots filled with a mixture of soil, sand and vermiculite (1:1:1). The plants were regularly watered and fertilized. All measurements were carried out during the summer on fully expanded leaves (Leaves 8–10 from the apex). Measurements of isoprene emission rate and respiration rate were carried out using the same types of instruments described for the field experiment.
Light manipulation and exposure to elevated CO2
Five plants were grown since spring under a 25 m2 shade structure made with white cloth netting. The shade reduced the photosynthetic photon flux density (PPFD) to 75% of the value outside the cloth. The daily temperature inside the shade was reduced on average by 1.5 °C with respect to the temperature recorded outside cloth. Five more plants were grown outside, under ambient conditions (day/night temperature = 28/20 °C, peak light intensity = 1800 µmol photons m−2 s−1). Measurements were carried out on one fully expanded leaf of each individual. Isoprene emission was measured at four different CO2 concentrations (300, 400, 600 and 1000 ppmv) while maintaining leaf temperature at 30 °C and light intensity at 1000 µmol photons m−2 s−1. Respiration rate was measured on dark-adapted leaves as explained earlier. Respiration rate in the light was also measured using the procedure outlined by Pinelli & Loreto (2003). After isoprene measurements at ambient CO2 concentration, the leaves were immediately frozen by clamping them into two metal plates pre-chilled in liquid nitrogen, as explained by Nogués et al. (2006). These samples were then used for PEPc activity measurements.
Another batch of five plants was grown in a greenhouse under ozone fumigation. Ozone was enriched using the fumigation system described by Fares et al. (2006). This system allowed us to fumigate selected leaves enclosed in Teflon branch cuvettes for 11 h d−1 under natural daylight with an ozone concentration of 150 ppbv. The ozone concentration in the cuvettes was monitored with a photometric ozone analyser (1008; Dasibi Environmental Corp., Glendale, CA, USA).
Isoprene emission rate and dark respiration rate were measured on two fully expanded leaves of each plant after 7, 15 and 21 d of fumigation. The leaves were immediately frozen after gas-exchange measurements for determination of PEPc activity.
Two groups of five plants were grown at two different temperatures (25 and 35 °C) for 2 months in two growth chambers (Sanyo-Gallenpkamp, Loughborough, UK) under a light intensity of 1000 µmol photons m−2 s−1 (daylight period 11 h). Isoprene emission rate and dark respiration rate were measured at growth temperatures and, in plants grown at 25 °C, also after exposing the entire plant to treatments for 60 min at 20, 25, 30 and 35 °C. PEPc activity was measured immediately after gas-exchange measurements following the procedure outlined earlier, only at growth temperatures. Measurements were carried out on 2 leaves per plant.
PEPc activity measurements
Phosphoenolypyruvate carboxylase activity was determined from frozen leaves free of the midrib and sampled as previously explained. The leaf tissue (100–150 mg) was pulverized in a pre-chilled mortar with 1 mL extraction buffer containing 100 mM HEPES-KOH (pH 7.2), 10 mM DTT, 0.3% (w/v) Triton X-100 (Roche, Mannheim, Germany), PVP-40 and 5 mM MgCl2, then centrifuged for 5 min at 11.000 g at 4 °C. The supernatant fraction was immediately assayed for PEPc activity in an assay buffer containing 25 mM tricine-KOH (pH 8.1), 5 mM MgSO4, 5 mM NaHCO3, 5 mM DTT, 0.2 mM NADH and 5 U malic dehydrogenase (MDH) (from porcine heart). The reaction was initiated by the addition of 2 mM PEP, at room temperature, and absorbance changes of NADH were registered on a time-based curve for 5 min at A340, with a sampling frequency of 0.2 min (Rosenstiel et al. 2004). The protein content of each sample was determined by Bradford staining (Bradford 1976), at A595 nm at room temperature. For calibration, BSA was used as standard.
Mean values + SEs are shown, with the exception of data, which report single measurements. Mean separation between CO2 levels was assessed statistically with a t-test in the field experiment and with a Tukey's test in the laboratory experiment. In both cases, the level of confidence was P < 0.05. Linear regressions and regression coefficients between respiration rate and isoprene emission rate, and between isoprene emission rate and PEPc activity, were generated by the Sigmaplot 2002 software (Systat, San Jose, USA).
In the field experiment at the FACE facility, the mean emission rate of isoprene was low in the leaf that was not completely expanded (Leaf 4), compared with older, fully expanded leaves (Leaves 8 and 11) both when sampled at 25 and 35 °C (Fig. 1). The divergence in isoprene emission rate as a function of leaf developmental state was particularly evident when the leaves were sampled at 35 °C. The CO2 treatment did not cause any statistically significant change in the emission rates of the three leaf age classes, at both sampling temperatures. However, a moderate reduction of isoprene emission rate was consistently observed in leaves grown and assayed at elevated CO2 when compared to leaves of the same age class and sampled at the same temperature, but grown and assayed at ambient CO2.
The same small effect of CO2 was again found in the experiment on potted plants grown at different light intensities (Fig. 2). The emission rates of shade- and light-adapted leaves were different, but in both cases, only a moderate inhibition was observed as a function of intercellular CO2 concentration increasing up to around 400 ppmv, which corresponded to external CO2 concentrations of 600 ppmv. At CO2 concentrations even higher than 600 ppmv, the negative impact of CO2 on isoprene emission rate became more evident and statistically significant in leaves subjected to both light levels.
In the FACE experiment, the dark respiration rate was relatively high in leaves with low isoprene emission rates (Fig. 3). In leaves grown and assayed at high and ambient CO2, an inverse relationship was observed between the dark respiration rate and isoprene emission rate when emission rates were lower than 8–11 nmol m−2 s−1. It should be noted, however, that the low isoprene emission rates and high respiration rates in this analysis also coincided with the youngest leaves (Leaf 4; see also Fig. 4). For emission rates above 10–11 nmol m−2 s−1, a concurrent increase of respiration rate and isoprene emission rate was observed, irrespective of the CO2 concentration. Once again, the leaves that exhibited this relationship were also leaves from the oldest cohorts measured (Leaves 8 and 11). The respiration rate in the light, estimated from gas-exchange measurements according to Kok (1948), was 10–30% lower than the dark respiration rate, but the shape of the relationship between respiration rate and isoprene emission rate did not change (data not shown).
Dark respiration and isoprene emission rates were measured with the same procedure used for the field experiment during several laboratory experiments on potted plants. As shown by the data in Fig. 4, dark respiration rate and isoprene emission rate were positively correlated in leaves exposed to or grown at different light intensities, ozone and CO2 levels, and temperatures. Leaves that were measured in the FACE experiment fit the same general regression relationship as those measured in the laboratory experiments, with the already noted exception of those leaves closest to the apex (Leaf node 4). Only in the latter case, as was previously noted, were high rates of respiration found when isoprene emission rates were still low.
The activity of PEPc was inversely correlated with isoprene emission rates (Fig. 5) when measured across all experiments. Independent of the experimental treatment that was used to manipulate isoprene emission rate, high PEPc activities were found when isoprene emission rates were low.
Growth at high CO2 in the FACE experiment did not cause a significant decrease in the isoprene emission rate by P. alba leaves (Fig. 1). This was an unexpected result because a substantial inhibition of isoprene emission in Populus species grown at elevated CO2 has been reported previously (Sharkey et al. 1991; Rosenstiel et al. 2003; Centritto et al. 2004; Pegoraro et al. 2004). However, there are also indications in other studies that within the context of this FACE experiment, the inhibition of isoprene emission is not as strongly expressed as when plants are grown under highly controlled growth conditions (Loreto et al. 2001c; Centritto et al. 2004). Perhaps the CO2 enrichment to which this species of poplar is exposed in this FACE experiment is not sufficient to elicit the same response as observed in past controlled-growth studies (which also tend to use higher growth CO2 concentrations). In our laboratory experiment, P. alba leaves exposed to a concentration of CO2 similar to that experienced in the FACE structure also did not show a significant reduction of isoprene emission rate, consistent with the field observations. It is also possible that a real trend exists towards lower isoprene emission rates in the elevated CO2, but leaf-to-leaf variability in emission was so great as to obscure the trend statistically given our limited sample size. It is a frequent challenge to obtain a sufficient sample size to statistically demonstrate clear trends under field conditions where environmental variation can be high.
Rosenstiel et al. (2003) reasoned that chloroplastic isoprene biosynthesis may compete with mitochondrial respiration for PEP, the compound that can either be converted to pyruvate through glycolysis and transported into the mitochondrion to be used as a substrate for respiration or transported via a specific PEP/Pi translocator into the chloroplast for conversion to pyruvate and use in isoprenoid biosynthesis (Flugge 1999). Indirect evidence for the incorporation in the MEP pathway of a C2 fragment, derived from pyruvate and PEP that originated in the cytosol, came by 13C labelling experiments showing incomplete labelling of isoprene and a characteristic intramolecular labelling when 13C-labelled sugars were supplied (Lichtenthaler et al. 1997; Schnitzler et al. 2004).
The small effect of CO2 on isoprene emission rate that we observed in the FACE and laboratory experiments did not allow us to directly examine whether there is a competition between respiration and isoprene biosynthesis, presumably caused by limited PEP substrate at high CO2 (Rosenstiel et al. 2003). In fact, in our experiments, respiration rate in the dark was also unaffected by the CO2 level; even the rate of respiration in the light, estimated by the extrapolation of the light response, did not respond to CO2 (data not shown). However, a large variation of isoprene emission rate and respiration rate was observed across all CO2 treatments. Within the overall context of these experiments, we found a correlated scaling of respiration rate and isoprene emission rate; this is similar to the correlated scaling observed previously between photosynthesis rate and isoprene emission rate (Monson & Fall 1989; Harley et al. 1994; Litvak et al. 1996). The only exception to this relationship was found in young leaves in the FACE experiment. It is well known that young, photosynthetically competent leaves do not emit isoprene (Kuzma & Fall 1993; Monson et al. 1994; Centritto et al. 2004), and it is likely that this is due to low expression levels of the enzyme isoprene synthase (Wiberley et al. 2005). It is also known that a high level of growth respiration is present in young leaves, supplying carbon for new tissues (Thomas et al. 1993). The inverse relationship between respiration rate and isoprene emission rate in the young leaves may be due to a progressive and rapid decrease in respiration rate with leaf development, accompanied by a progressive and rapid increase in isoprene emission with leaf development. We appear to have caught the crossover point of these two trends in the sampling of Leaf node 4. It may be hypothesized that in young leaves, high rates of respiration requires carbon otherwise allocated to isoprene biosynthesis. This would be similar to the competition that has been postulated to explain isoprene inhibition at high CO2 (Rosenstiel et al. 2003). However, we assume that these trends are due to independent controls, as we have currently no evidence to suggest that there is a causal link between them.
The positive scaling between respiration rate and isoprene emission rate in older leaves, however, does fit the general trend that was previously observed with photosynthesis – leaves that are more metabolically active and have developed the capacity for isoprene emission tend to emit isoprene at higher rates. These results are not consistent with the hypothesis of Rosenstiel et al. (2003), that respiration competes with isoprene for common substrates in mature leaves. Rather, there appears to be a developmentally and genetically controlled scaling of respiration and isoprene biosynthesis that balances their expression relative to one another. Increases in the demand for respiratory substrate does not appear to negatively influence the demand for MEP pathway substrate within the context of plant growth in differential environments.
Rosenstiel et al. (2003, 2004) provided evidence that isoprene emission rate is stimulated upon chemical inhibition or induced changes in the genetic expression of PEPc, a cytosolic enzyme that uses PEP to provide carbon skeletons for respiration or nitrate assimilation. Across all of our experiments, we observed a significant inverse relationship between isoprene emission rate and PEPc activity, confirming that the capacity for cytosolic processes, not associated with mitochondrial respiration, to use PEP may be a key controller of isoprene synthesis. We hypothesize the existence of independent controls over the glycolytic partitioning of PEP into respiration, the partitioning of PEP into oxaloacetate (OAA) production for anabolic metabolism and the import of PEP into the chloroplast for isoprenoid biosynthesis. Our data suggest that only PEP that is partitioned into the PEPc-catalysed transformation into OAA may also be available for import into chloroplasts for isoprenoid biosynthesis (Fig. 6). This would explain why isoprene is inversely related to PEPc activity while it scales positively with respiration rate, under conditions in which carbon flow through the glycolytic pathway and partitioning to mitochondrial processes would increase. PEP regulation may also influence the formation of other isoprenoids with important functional roles, such as carotenoids, which are also formed in the chloroplasts from the MEP pathway (Lichtenthaler 1999). More experiments are needed to test this possibility.
The finding that a positive relationship between isoprene emission rates and dark respiration rates exists may also be important to refine process-based isoprene emission models at leaf and canopy level (Niinemets et al. 1999; Zimmer et al. 2000).
We thank G. Alessio, F. Brilli and S. Wahbi for helping with field and laboratory measurements, and L. Ederli for helping with measurements of PEPc activity. Field measurements were made possible by the cooperation of the EUROFACE steering committee coordinated by the University of Viterbo, Italy.
This work was supported by the European Commission Marie Curie project ‘Ecological and Physiological Functions of Biogenic Isoprenoids and Their Impact on the Environment’ (ISONET, MRTNCT–2003–504720), by theEuropean Science Foundation scientific programme Volatile Organic Compounds in the Biosphere – Atmosphere System and an Italia-USA bilateral project supported by the Italian Ministry of Environment. R.K. Monson was supported by a short-term mobility fellowship of the Italian National Research Council (CNR), and C. Calfapietra was supported by a Marie Curie international fellowship (GLOBAL VOC, MOIF-CT-2005-007692).