Isoprene is a biogenic hydrocarbon that has significant effects on tropospheric chemistry. It is emitted by a number of plant species, including kudzu, a leguminous vine that grows profusely in the south-eastern United States. This study investigated development of the capacity for isoprene emission in kudzu. Previous studies examined isoprene emission during leaf development, but a molecular explanation for the observed developmental delay in emission was lacking. This study found that kudzu leaves grown at a high temperature could emit isoprene at least a week before they were fully expanded and 1 d after becoming photosynthetically competent. When grown at low temperature, however, leaves did not emit isoprene until 1 week after they became fully expanded and 2 weeks after the onset of photosynthetic competence. Levels of mRNA and protein for isoprene synthase, which catalyses the final step in isoprene biosynthesis, were investigated; it was found that transcription and translation of this gene began at the same developmental stage as onset of emission in both growth conditions. Therefore, plant growth conditions, not leaf developmental stage, have primary control over expression of isoprene synthase and onset of kudzu isoprene emission. This finding may be useful in modelling early season isoprene emission rates.
Isoprene (C5H8) is an abundant biogenic hydrocarbon in the atmosphere (Fehsenfeld et al. 1992). Some plant species, including many trees and the leguminous vine kudzu, can emit large amounts of isoprene; others emit none (Sharkey & Yeh 2001). As global isoprene emission rates are very large [400 Tg per year (Guenther et al. 1995; Fuentes et al. 2000)] and isoprene is quite reactive with hydroxyl radicals and nitrogen oxides in the atmosphere, isoprene plays a significant role in atmospheric chemistry (Thompson 1992; Monson & Holland 2001). As such, the ability to predict isoprene emission rates is quite important, and for such predictions, an understanding of the mechanisms that control isoprene emission is necessary.
Isoprene is synthesized from dimethylallyl diphosphate (DMAPP) made by the chloroplastic methylerythritol 4-phosphate pathway (Schwender et al. 1997). Isoprene synthase (IspS) converts DMAPP to isoprene (Silver & Fall 1991). The activity of isoprene synthase can control isoprene emission rate (Kuzma & Fall 1993; Schnitzler et al. 1997; Sharkey et al. 2005). IspS activity can, in turn, depend on transcriptional, translational, or post-translational regulatory mechanisms. This study was conducted to elucidate the influence of leaf developmental stage on the capacity for isoprene emission from kudzu. Isoprene emission rates and IspS mRNA and protein quantities were measured for leaves developing in 20 or 30 °C greenhouse rooms. IspS protein was found in both soluble and insoluble leaf fractions, so these were measured separately and correlated with isoprene emission rates throughout leaf development.
MATERIALS AND METHODS
Plant growth conditions
Kudzu plants (Pueraria montana var. lobata (Willd.) Maesen & S. Almeida) were grown from stem cuttings in 10-L pots containing a vermiculite/peat moss-based growth medium (Metro-Mix 360, The Scotts Company, Marysville, OH, USA). The plants were grown in temperature-controlled greenhouses at the Biotron facility of the University of Wisconsin-Madison. Three plants were grown at 20/16 °C day/night temperature with no extended day-length, whereas two were grown at 30/20 °C day/night temperature and day-length extended to 16 h with high pressure sodium vapour lamps. Plants were watered with varying strength Hoagland's solution (Hoagland & Arnon 1938) and fertilized with 14-14-14 N-P-K Osmocote according to the manufacturer's instructions (The Scotts Company).
Leaf measurement and sample collection
Leaves near the tops of the plants were tagged when they were 1–2 cm in length, and were measured daily thereafter until they were fully expanded. Leaf length was measured from the junction of the terminal leaflet blade and petiole, along the midrib, to the tip of the blade. Samples were collected from mid-July up to the end of August 2004, and leaf measurements and collections were carried out between 1030 and 1230 h. All tissue samples were collected close to the tips of the lateral leaflet blades and were frozen immediately in dry ice and then stored at −80 °C until use.
Samples were collected in leaf lengths of: 3–4 cm, 4–5 cm, 5–6 cm, 6–7 cm, and 7–8 cm; also 24 h past full expansion, 1 week past full expansion and 2 weeks past full expansion. These categories were selected because healthy kudzu leaves can have a final length ranging from 8 to 20 cm, and rarely stop growing before reaching 8 cm (unpublished observation). Collected leaf punches were 93.3 mm2 in area.
Gas exchange and isoprene emission measurement
Gas exchange measurements were conducted as described by Wolfertz et al. (2003); all measurements were made at 30 °C and 1000 µmol m−2 s−1. The air source for the Li-Cor 6400 (Li-Cor Inc. Lincoln, NE, USA) was a cylinder of compressed air. Ten millilitres of the air exiting the leaf cuvette were collected by syringe and analysed for isoprene content by gas chromatography as described by Loreto & Sharkey (1993) with the following modifications: the column was maintained at 52 °C, and the liquid isoprene standard was serially diluted to 128 nmol mol−1 in N2. Leaf isoprene emission rates were calculated as described by Singsaas et al. (1997).
RNA extraction and quantitative polymerase chain reaction analysis
Total RNA was extracted from leaf samples as follows: A leaf punch was ground in liquid nitrogen with a mortar and pestle. Extraction buffer [0.1 m Tris-HCl (Fisher Scientific, Pittsburgh, PA), pH 8.0; 0.1 m LiCl; 0.2 m m ethylenediaminetetraacetic acid(EDTA); 1% w/v sodium dodecyl shlphate (SDS); 50% v/v phenol], heated to 80 °C, was added (500 µL per 10 mg tissue) and then homogenized. The mixture was transferred to a polypropylene tube and shaken vigorously; 100 µL 24 : 1 chloroform : isoamyl alcohol were added and the tube was shaken again and then centrifuged at 12 000 × g at 4 °C for 5 min. The upper phase solution was collected and precipitated with an equal volume of 4 m LiCl at 4 °C for 2 h. This solution was centrifuged at 12000 × g at 4 °C for 10 min, the supernatant was removed, and the pellet was resuspended in 700 µL diethylpyrocarbonate (DEPC)-treated H2O by pipetting. This was extracted with 25 : 24 : 1 phenol : chloroform : isoamyl alcohol twice and with 24 : 1 chloroform : isoamyl alcohol once. Three molar sodium acetate (1/10 volume) and isopropanol (1 volume) were added to the resulting solution and RNA was precipitated overnight at −20 °C. This solution was centrifuged at 12000 × g for 10 min, the supernatant was removed, and the pellet was washed once with 70% ethanol and once with 100% ethanol. The pellet was air-dried and redissolved in 30–50 µL DEPC-treated H2O. All chemicals were obtained from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise noted.
Total soluble protein was extracted from tissue samples as described by Heck et al. (1995). The extract was centrifuged for 5 min at 12 000 × g. The supernatant from this extraction was separated from the pellet, and the pellet was mixed with a volume of SDS sample buffer (1 m Tris-HCl, pH 6.8; 10% SDS; 10%α-mercaptoethanol; 20% glycerol; 0.004% bromphenol blue) equal to the volume of the pellet, vortexed, and heated at 70 °C for 20 min. This was centrifuged briefly at 12 000 × g to bring down cell wall fragments, leaving membrane-associated proteins in the supernatant. Soluble and membrane protein fractions were separated on NuPAGETM. NOVEX 4–12% Bis-Tris gels (Invitrogen Corp., Carlsbad, CA, USA) and transferred to Hybond-P PVDF membranes (Amersham Biosciences, Piscataway, NJ, USA) using the NOVEX XCellIITM. Mini-Cell system as directed by the manufacturer (Invitrogen Corp.). Immunoreactive protein was detected using the ECL western blotting system according to manufacturer's instructions (Amersham Biosciences). A polyclonal primary antibody for IspS was generated using kudzu IspS cDNA expressed in a pET15b vector in Escherichia coli and subjected to 10% denaturing PAGE (Sharkey et al. 2005). The primary antibody was diluted 1 : 1250, and the secondary donkey antirabbit antibody coupled to horseradish peroxidase was diluted 1 : 5000. Blots were exposed to Kodak (Rochester, NY, USA) Biomax MS X-ray film for 5–20 min. Membranes were stained with Coomassie blue to visualize all of the protein and check for equal loading. The isoprene synthase used as a standard was purified from E. coli as previously described (Sharkey et al. 2005) and its concentration determined by Bradford assay (Bradford 1976). Western blot films were scanned and bands were quantitated using ImageJ (U.S. National Institutes of Health, http://rsb.info.nih.gov/ij/).
IspS activity assays
A number of different extraction protocols were attempted for extraction of isoprene synthase activity from kudzu leaf samples. In general, leaf punches were ground in extraction buffer and centrifuged briefly at 12 000 × g to bring down cell wall fragments, and the supernatants were assayed. Several buffers were tested: EB [50 m m Tris-HCl pH 8.0, 20 m m MgCl2, 5% glycerol; with 10% polyvinyl polypyrrolidone, 20 m m dithiothreitol (DTT), 1 m m phenylmethylsulphonyl fluoride (PMSF), and 1 m m benzamidine-HCl added immediately before use (Silver & Fall 1991)]; EBT [EB with 0.5% Triton X-100 added (Schnitzler et al. 1996)]; and EBC [extraction buffer with 15 m m 3-[(3-cholamidopropyl) dimethyl amino]-l-propanesulfate (CHAPS) added]. In some cases, the supernatants were concentrated with a 30–60% (NH4)2SO4 precipitation; the pellet from the 60% cut was re-suspended in an equal volume of assay buffer and then assayed. The assay buffer was 50 m m bicine pH 8.0, 50 m m MgCl2, 5 m m KCl, 2 m m NaF, and 5% glycerol. Assay reaction mixtures were: 20 µL assay buffer; 1 µL 1 m DTT; 10 µL 50 m m DMAPP; 1 µL (2.2024 µg) IspS purified from E. coli; and 9 µL extract, extraction buffer, or water. DMAPP was synthesized as described by Davisson, Woodside & Poulter (1985). These mixtures were incubated at 35 °C for 15 min in 5.5-mL sealed vials, after which 3 mL of headspace were removed by syringe, with water added to the vial simultaneously to equalize pressure, and analysed for isoprene in the manner described for plant gas samples.
Development of isoprene emission and photosynthesis capacities
Leaves grown in inductive conditions (30 °C daytime temperature) emitted isoprene about a week before they became fully expanded, while leaves grown in non-inductive conditions (20 °C daytime temperature) did not emit isoprene until after full expansion, when they emitted substantially less than their 30 °C counterparts (Fig. 1). Leaves grown at 30 °C began to emit when they were 5–6 cm long; since kudzu leaves grown in these conditions typically lengthen about 1 cm per day until they near their final size, and then grow more slowly for about 3 d before growth stops (unpublished observation), 5–6 cm leaves were at least a week away from reaching their final size of ≥ 8 cm. Emission rates for 30 °C-grown leaves increased during the first week past full expansion, and then nearly doubled at 2 weeks past full expansion. Leaves grown at 20 °C, however, did not emit significant amounts of isoprene (≥ 5 nmol m−2 s−1; lower quantities were indistinguishable from noise) until after they became fully expanded; even then, their emission rates were only about 13 nmol m−2 s−1, which is less than the 30 °C-grown leaves emitted when they were 6–7 cm long and at least 5 d away from full expansion.
Photosynthesis rates were substantially greater than zero for 30 °C-grown leaves 4–5 cm long and larger, and for 20 °C-grown leaves 6–7 cm long and larger (Fig. 2). In 30 °C-grown leaves, the development of photosynthetic competence preceded the onset of isoprene emission by about 1 d, whereas in 20 °C-grown leaves, photosynthetic competence preceded isoprene emission by about 2 weeks.
IspS mRNA copy numbers
When quantitative RT-PCR was performed on samples expected to have low IspS copy numbers [based on lack of IspS amplification in non-quantitative RT-PCR (data not shown)], the QPCR software calculated very large copy numbers. The dissociation curves from these reactions showed peaks at slightly higher and lower temperatures than 77.5 °C, the melting point of this amplicon (Fig. 3), so it appeared that the SYBR® green dye being used for detection was binding to amplified primer dimers and these were being reported as copies of the IspS amplicon. This was observed only in samples with few or no IspS copies, so it seemed that the lack of sufficient target template was allowing dimer formation. All reactions were therefore spiked with 1000 copies of the amplicon, eliminating the formation of primer dimers (Fig. 3) and allowing accurate template quantification.
IspS mRNA first appeared in 30 °C-grown leaves when they were 5–6 cm long, the same stage at which they began to emit. With the exception of the samples taken at 1 week past full expansion, mRNA levels increased from 5–6 up to 7–8 cm leaves and then levelled off (Fig. 4). The 20 °C-grown leaves, however, did not have significant IspS mRNA present until 2 weeks after full expansion, at which time they had as much message as the 30 °C-grown leaves (Fig. 4).
IspS protein quantities
For 30 °C-grown leaves, small amounts of IspS protein first appeared in the membrane fraction of 5–6 cm leaves; in leaves 7–8 cm and larger, it was detected in both membrane and soluble fractions (Fig. 5). Both membrane-bound and soluble IspS levels increased throughout leaf development (Table 1). The IspS from the soluble fraction ran slightly further on the gel than that of the membrane fraction (Fig. 5). In 20 °C-grown leaves, IspS protein appeared in the membrane and soluble fractions of leaves that were 1 and 2 weeks past full expansion. As was observed for 30 °C-grown leaves, the soluble IspS ran to a slightly smaller molecular weight than did the membrane-bound IspS (Fig. 5). For all leaves, levels of membrane-bound, soluble, and total IspS correlated well with isoprene emission rates (Fig. 6).
Table 1. IspS protein found in emitting leaves
20 °C membrane
20 °C soluble
30 °C membrane
30 °C soluble
Values are in mg protein per square metre of leaf ± standard error; n = 3.
0.28 ± 0.09
0.50 ± 0.07
1.63 ± 0.30
0.24 ± 0.11
1.13 ± 0.46
0.48 ± 0.26
2.23 ± 0.71
0.04 ± 0.00
0.11 ± 0.00
1.05 ± 0.06
4.39 ± 0.46
0.07 ± 0.04
0.27 ± 0.09
1.35 ± 0.16
5.52 ± 0.14
IspS enzyme activity
Attempts were made to extract IspS activity in several buffers, but it was found that some compound present in kudzu whole-leaf extracts inhibited IspS activity. Reactions run with purified IspS added showed a decrease in IspS activity when leaf extracts were added (Fig. 7), and reactions run without added IspS showed no significant IspS activity (data not shown). EB itself did not inhibit added IspS activity, but EBT did inhibit. EBC, on the other hand, increased the activity of added IspS. Kudzu crude extracts in EB and EBC both inhibited the added IspS activity, as did ammonium sulphate precipitates of these crude extracts (Fig. 7)
Kudzu leaves are capable of emitting isoprene well before they reach full expansion when they are grown in inducing conditions, whereas growth in non-inducing conditions delays the onset of emission by about 2 weeks (Fig. 1). Emission from 30 °C-grown leaves begins about 1 d after the leaves become photosynthetically competent, whereas emission from 20 °C-grown leaves does not begin until about 2 weeks after photosynthesis rates rise above zero (Figs 1 & 2). Grinspoon et al. (1991) found that velvet bean isoprene emission begins about 3 d after photosynthetic competence in plants grown at 28 °C, at about the same time as leaves become fully expanded; these findings were supported by Harley et al. (1994) with velvet bean plants grown at 30 °C. Thus, the onset of isoprene emission follows the onset of photosynthetic competence, but the amount of time between the two events varies. This is also consistent with the observations of Monson et al. (1994), who found that poplar leaves that developed in cool spring temperatures had a delay of about 2 weeks between the onsets of photosynthetic competence and isoprene emission, whereas for leaves that developed in the summer the delay was about 2 d. In that study, data was collected based on days since leaf emergence, so it was unclear whether the more rapid onset of emission in the summer was due to direct induction of emission by high temperature or to a heat-induced increase in the rate of leaf development, with a concomitant decrease in the delay between photosynthetic competence and emission. The present study demonstrates that the former is the case in kudzu. Kudzu leaves do develop more rapidly when grown at higher temperatures (unpublished observation), but when leaves are analysed based on developmental stage rather than time past emergence, it is clear that the onset of isoprene emission is governed more by plant growth conditions than by leaf developmental stage.
IspS mRNA and protein appear in kudzu leaves at the same developmental stage as the onset of emission (Figs 4 & 5); gene expression does not significantly precede enzyme activity. This indicates that much control over IspS activity is exerted at the level of IspS transcription. There does not seem to be any specific developmental stage at which IspS transcription is always turned on in kudzu leaves. However, this does not rule out post-translational regulation of IspS activity. Both membrane-bound and soluble forms of IspS were found in kudzu, as have been found in willow (Wildermuth & Fall 1998). Wildermuth (1997) suggested that the membrane-bound willow IspS may be palmitoylated, associating it with the thylakoid membrane and increasing its activity. The membrane-bound form in kudzu ran to a slightly higher molecular weight than the soluble form in SDS-PAGE analysis, indicating a possible post-translational modification. However, membrane-bound and soluble protein levels both correlated well with isoprene emission rates (Fig. 6), so it does not seem that any post-translational modification that might occur has an effect on IspS activity. The high correlation between isoprene emission rate and IspS protein levels could provide a basis for predictions of isoprene emissions in mixed forests. In such forests, measuring the isoprene emission rates of leaves at the top of the canopy is often physically impossible, but leaf samples could be obtained and their IspS content determined and used to predict emission rate, if an antibody that cross-reacted with the IspS of many species were available. The range of cross reactivity of our antibody has not yet been determined.
Numerous attempts were made to measure IspS activity in kudzu leaf extracts, but none was successful. First, it was found that Triton X-100, a component of some IspS extraction buffers (e.g. Schnitzler et al. 1996), inhibits the activity of kudzu IspS. Other buffer components did not themselves inhibit activity, but whenever kudzu leaf extract was added to reaction mixtures containing purified IspS, the activity of that IspS decreased by 29–99%. It therefore seems that kudzu whole-leaf extracts contain some compound that inhibits kudzu IspS activity. This compound could not be removed by ammonium sulphate precipitation; indeed, suppression of activity became more pronounced after precipitation. It would probably be possible to obtain IspS activity from isolated kudzu chloroplasts, since if the inhibitor were present in chloroplasts kudzu would probably not emit as much isoprene as it does, but activity obtained from this isolation would not be sufficiently quantitative to fit with the aims of the present study. Therefore, it was not possible within this study to determine how in vitro IspS activity changed over the course of leaf development, or whether the membrane-bound or soluble form was more active.
The observation that heat-grown leaves develop the capacity for isoprene emission much sooner than their cool-grown counterparts is consistent with the thermotolerance hypothesis for isoprene emission. This hypothesis states that plants emit isoprene to protect against damage caused by brief high-temperature episodes (Sharkey & Yeh 2001). Such episodes are more likely to be experienced when a plant is already exposed to high temperatures, as were the 30 °C-grown plants in this study. Plants grown in such conditions commonly have a higher isoprene emission capacity than cool-grown plants (Sharkey & Loreto 1993). It has been observed that temperatures experienced during the 2 d preceding measurement of a leaf's isoprene emission capacity exert considerable influence over the emission capacity (Sharkey et al. 1999); this study indicates that this is the case in developing as well as full-grown leaves.
The results of this study show that early season isoprene emission depends on growth conditions, mediated by transcriptional control of IspS. Kudzu leaves grown at high temperature express IspS and emit much sooner than leaves grown at low temperature. This information may be useful in predictions of early season isoprene emission. Work is currently underway to determine how variation of isoprene emission in the middle of a growing season is regulated.
This research was supported by the National Science Foundation grant IBN-0212204.