JXB Advance Access published online on January 27, 2008
Journal of Experimental Botany, doi:10.1093/jxb/erm283
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
SPECIAL ISSUE REVIEW PAPER |
The biochemistry of Rubisco in Flaveria
1Department of Biology, The University of New Brunswick, Box 4400, Fredericton, New Brunswick, E3B 5A3, Canada
2Research School of Biological Sciences, The Australian National University, GPO Box 475, Canberra, ACT 2601, Australia
* To whom correspondence should be addressed. E-mail: kubien{at}unb.ca
Received 20 September 2007; Revised 16 October 2007 Accepted 22 October 2007
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
C4 plants have been reported to have Rubiscos with higher maximum carboxylation rates (kcatCO2) and Michaelis–Menten constants (Km) for CO2 (Kc) than the enzyme from C3 species, but variation in other kinetic parameters between the two photosynthetic pathways has not been extensively examined. The CO2/O2 specificity (SC/O), kcatCO2, Kc, and the Km for O2 (Ko) and RuBP (Km-RuBP), were measured at 25 °C, in Rubisco purified from 16 species of Flaveria (Asteraceae). Our analysis included two C3 species of Flaveria, four C4 species, and ten C3-C4 or C4-like species, in addition to other C4 (Zea mays and Amaranthus edulis) and C3 (Spinacea oleracea and Chenopodium album) plants. The SC/O of the C4 Flaveria species was about 77 mol mol–1, which was approximately 5% lower than the corresponding value in the C3 species. For Rubisco from the C4 Flaverias kcatCO2 and Kc were 23% and 45% higher, respectively, than for Rubisco from the C3 plants. Interestingly, it was found that the Ko for Rubisco from the C4 species F. bidentis and F. trinervia were similar to the C3 Flaveria Rubiscos (
650 µM) while the Ko for Rubisco in the C4 species F. kochiana, F. australasica, Z. mays, and A. edulis was reduced more than 2-fold. There were no pathway-related differences in Km-RuBP. In the C3-C4 species kcatCO2 and Kc were generally similar to the C3 Rubiscos, but the Ko values were more variable. The typical negative relationships were observed between SC/O and both kcatCO2 and Kc, and a strongly positive relationship was observed between kcatCO2 and Kc. However, the statistical significance of these relationships was influenced by the phylogenetic relatedness of the species. Key words: C3, C4, Flaveria, kinetics, Michaelis–Menten, phylogeny, Rubisco, specificity
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The most abundant enzyme on Earth is Rubisco (ribulose-1,5-bisphosphate carboxylase/oxygenase, EC 4.1.1.39 [EC] ; Ellis, 1979), which fixes most of the carbon entering the biosphere. Rubisco catalyses the addition of CO2 to ribulose-1,5-bisphosphate (RuBP), producing two molecules of 3-phosphoglycerate (3-PGA). The energy required to drive CO2 fixation in eukaryotic phototrophs is delivered by the light-harvesting reactions of the thylakoid; the associated water-splitting biochemistry has produced today's aerobic atmosphere. This atmosphere greatly reduces the efficiency of photosynthesis because Rubisco is unable to distinguish completely between CO2 and O2 as substrates for fixation to RuBP. The oxygenation reaction produces a 3-PGA and a 2-phosphoglycolate (2-PGO). The recovery of carbon from 2-PGO via the photorespiratory pathway is energetically expensive, and greatly reduces net CO2 fixation. This bi-functionality, coupled with slow catalysis, leads to Rubisco frequently being the principal determinant of the efficiency with which autotrophs use CO2, light, water, and mineral resources (Tcherkez et al., 2006).
The efficiency of photosynthesis is strongly dependent on the relative specificity of Rubisco for CO2 versus O2 (SC/O), which is simply the ratio of the specificities for CO2 and O2:
 | (1) |
and 
are the maximal turnover rates for the carboxylation and oxygenation reactions, and Kc and Ko are the Michaelis–Menten constants for CO2 and O2, respectively (Laing et al., 1974; Kane et al., 1994). The specificity values denote the proportional efficiencies of carboxylation (
) to oxygenation (
) of a Rubisco in the presence of equal amounts of CO2 and O2. While Form I Rubiscos (e.g. from plants and algae) have SC/O values ranging from 75–230 mol mol–1 (Jordan and Ogren, 1981; Kane et al., 1994; Tcherkez et al., 2006), which greatly favour CO2 fixation, the enzyme falls far short of the near perfect selectivity shown by other enzymes with alternative substrates (Kane et al., 1994). Moreover, the relative level of oxygen is about 500 times greater than that of CO2 in the modern atmosphere, meaning that carboxylation proceeds at only about four times the rate of oxygenation in C3 plants under optimal physiological conditions. A range of evidence indicates that modifying higher plant Rubiscos to emulate the high SC/O of the enzyme from some of the non-green algae could result in substantial increases in crop yield (Andrews and Whitney, 2003; Zhu et al., 2004). However, the folding and assembly requirements of the Rubiscos from non-green algae are not met by higher plant chloroplasts (Whitney et al., 2001), and our understanding of the subtle structural variations that afford the kinetic variability found amongst Rubiscos is still in its infancy. Consequently, improving yield by increasing SC/O is beyond our present capabilities.
An interesting feature of Rubisco is that forms with higher SC/O have slower , so that as the fraction of oxygenation reactions declines fewer reactions occur overall (Bainbridge et al., 1995; Tcherkez et al., 2006). The molecular basis for this trade-off is uncertain, although it appears to be a consequence of the Rubisco reaction mechanism (Tcherkez et al., 2006). There is measurable habitat-dependent variation in SC/O between C3 species, with species from arid habitats having more specific Rubiscos (Delgado et al., 1995; Galmés et al., 2005). However, much of the kinetic variability between Rubiscos is associated with the presence or absence of CO2-concentrating mechanisms (Badger and Andrews, 1987; Badger et al., 1998). Under high CO2 conditions, maximum rates of carboxylation are primarily determined by (Badger and Andrews, 1987; Sage, 2002). For example, in C4 leaves Rubisco is exposed to CO2 concentrations several times above ambient (von Caemmerer and Furbank, 1999); Rubiscos from C4 plants have 25–50% higher kcat than the Rubiscos from C3 species (Wessinger et al., 1989; Sage, 2002). Associated with their faster turnover, the Kc values of C4 Rubiscos are 1.5–3 times higher than that of the C3 enzymes (Yeoh et al., 1980, 1981; Seemann et al., 1984; Wessinger et al., 1989). Despite these kinetic variations, the few SC/O measurements made in C4 Rubiscos appear comparable with the many values determined in C3 plants (Jordan and Ogren, 1981; Kane et al., 1994). There is also a lack of measured Ko values reported for Rubisco from C4 plants (Badger et al., 1974; Jordan and Ogren, 1981), which impedes efforts to model accurately C4 photosynthesis (von Caemmerer and Furbank, 1999). Knowledge of Ko is requisite for comparing the Rubiscos of C3 and C4 plants.
In order to examine the functional differences in Rubisco from C3 and C4 plants it is necessary to have plants that are closely related phylogenetically. Flaveria (Asteraceae) has long been recognized as a powerful system in which to examine the evolution of C4 photosynthesis, particularly because of the presence of C3-C4 intermediate species. Recent molecular evidence suggests that C4 photosynthesis may have arisen between two and four times within Flaveria (McKown et al., 2005; Sudderth et al., 2007). The and Kc of Rubisco from 11 species of Flaveria have been determined previously (Wessinger et al., 1989). However, the SC/O and Ko parameters have not been examined, and there is no clear understanding of what, if any, changes to Ko might occur during the transition from C3 to C4 photosynthesis.
The goal of this study was comprehensively to compare the kinetics of Rubisco from C3, C4, and C3-C4 intermediate plant species and to clarify whether or not there is variation in SC/O and Ko between the two groups. SC/O, , and the Michaelis–Menten constants for CO2, O2, and RuBP were measured in 16 species of Flaveria. Our analysis includes two C3 species (F. pringlei and F. cronquistii), and four C4 plants (F. australasica, F. bidentis, F. kochiana, and F. trinervia), as well as 10 photosynthetically intermediate Flaveria species. For comparison, the kinetics of Rubisco from the C3 plants Spinacia oleracea and Chenopodium album, and the C4 species Zea mays and Amaranthus edulis, were also measured. The existing molecular phylogeny of Flaveria was used to test for differences in the kinetics of Rubisco between photosynthetic pathways, after correcting for the evolutionary relationships between species.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Plant material and growth
Seed or cuttings of 16 Flaveria species were obtained from Professor Rowan Sage (University of Toronto). Seed was germinated in soil, and cuttings were rooted in vermiculite on a misting bench. Once seedlings were sufficiently robust, or cuttings had well-developed roots, plants were transplanted in 10 l pots containing 80% Promix (Plant Products, Brampton, Canada) and 20% sand. Plants were grown in a naturally illuminated glasshouse at the University of New Brunswick (45°56' N, 66°33' W). Daytime temperatures in the greenhouse were typically 27–33 °C. Plants were watered as needed, and fertilized monthly.
Rubisco extraction
Leaves were harvested when plants were 3–4-months-old. Actively photosynthesizing leaves (approximately 100 g FW) from four or five plants of each species were detached, pooled, frozen in LN2, and stored at –80 °C.
Rubisco extraction was based on the procedures described by Paech and Dybing (1986) and Wessinger et al. (1991). Leaf tissue was ground for 60 s in a blender containing 500 ml of ice-cold buffer (20 mM Na2HPO4, 5 mM MgCl2, 1 mM Na2-EDTA, 40 mM 
-amino caproic acid, 8 mM benzamide, 100 mM β-mercaptoethanol, 40 mg ml–1 PVP, and 4 mg ml–1 BSA, pH 6.1). The homogenate was filtered through three layers of cheese-cloth and three layers of Miracloth. The sample was made to 5 mM ATP (from a 500 mM, pH 7 stock), heated to 58 °C in a microwave (approximately 130 s), held at that temperature in a water bath for 9 min, and then rapidly cooled to 4 °C (8–10 min in a –80 °C freezer). The extract was centrifuged for 25 min at 15 300 g at 4 °C, and the supernatant was slowly mixed with an equal volume of ice-cold saturated (NH4)2SO4 (pH 7) at 4 °C. After 20 min the sample was centrifuged as above, and the resulting pellet containing Rubisco was dissolved in a minimal amount of a storage buffer (10 mM KH2PO4, 50 mM NaCl, 1 mM Na2-EDTA, 5 mM DTT, pH 7.6), made to 20% glycerol (v/v), and stored at –80 °C.
CO2/O2 specificity assay
The purified Rubiscos were used to measure SC/O, using the method of Kane et al. (1994), at 25 °C in an atmosphere of 500 ppm CO2 in O2, controlled by a set of three Wösthoff precision gas-mixing pumps. The reaction was initiated by the addition of 1 nmole of 1-3H-RuBP (final assay volume 1 ml), and was terminated after 60 min by the addition of alkaline phosphatase. The 3H-glycerate and 3H-glycolate were separated on a HPX-87H column (Bio-Rad, Gladesville, NSW, Australia) using HPLC (Prominence, Shimadzu, Rydalmere, NSW), and their ratio quantified using an on-line, continuous flow scintillation analyser (505TR, Perkin-Elmer, Melbourne, Vic., Australia).
Determination of kcat
The kcat of carboxylation () was determined on leaf protein extracts, following Kubien et al. (2003). Leaf samples (1.0–1.6 cm2) were harvested from photosynthesizing leaves, frozen in LN2, and stored at –80 °C until being assayed (<72 h). Samples were ground on ice in 2.0–2.5 ml of extraction buffer [100 mM HEPES-KOH, 1 mM Na2-EDTA, 20 mM MgCl2, pH 8, 5 mM DTT, 12 mM -amino caproic acid, 2.4 mM benzamide, 10 mg ml–1 PVPP, 2 mg ml–1 BSA, 2 mg ml–1 PEG, 2% (v/v) Tween-80, 2 mM NaH2PO4, and 2% (v/v) protease inhibitor cocktail (Sigma, St Louis, MO, USA)]. The extracts were briefly centrifuged, 900 µl of the supernatant was added to 100 µl of an activating solution (100 mM Bicine–NaOH, 200 mM MgCl2, 100 mM NaHCO3, pH 8), and incubated at 25 °C for 30 min to carbamylate Rubisco. The activity of the enzyme at 25 °C was determined by the incorporation of 14C into acid-stable products, as described by Kubien et al. (2003). The concentration of Rubisco catalytic sites was determined by the 14CABP binding assay (Ruuska et al., 1998).
Determination of Michaelis–Menten constants
Measurements of Kc and Km-RUBP were conducted as described by Paul et al. (1991), and Ko was determined by measuring Kc at 0%, 10%, 20% or 30% O2 (mixed v/v with N2, using a Wösthoff pump). All assays were conducted at 25 °C and used highly purified RuBP, synthesized as described by Kane et al. (1998). Seven RuBP concentrations (0–100 µM) were used to measure Km-RuBP at 10 mM NaH14CO3. Measurements of Kc and Ko were carried out in septum-sealed, N2-sparged vials containing assay buffer (100 mM Bicine, 10 mM MgCl2, 1 mM Na2-EDTA, pH 8.1), 500 µM RuBP and 10 µg ml–1 carbonic anhydrase; seven different concentrations of NaH14CO3 (0.15–6.75 mM) were used for each Kc measurement. A pKa of 6.25 was used for the 
equilibrium (Tcherkez et al., 2006). Purified Rubisco was diluted to 1 µM catalytic sites with assay buffer containing 10 mM NaH12CO3, and activated for 30 min at 30 °C before adding 20 µl of the fully-activated enzyme to initiate the assays (final assay volume 500 µl). After 30 s the assays were terminated with 250 µl 20% (v/v) HCOOH.
Statistical analysis
To calculate Kc and Km-RuBP, CO2 or RuBP response curves were fit to the Michaelis–Menten first-order rate equation. To calculate Ko the apparent 
, measured at a range of different oxygen mixtures, was regressed against oxygen; the regression was forced through the ordinate at Kc. The regression was used to calculate at 20% O2, from which Ko was calculated:
 | (2) |
Because closely related species may share similar trait values, they probably do not represent statistically independent data points, and thus violate the assumptions of conventional statistical methods (Felsenstein, 1985). To examine this possibility, we rested the effect of the photosynthetic pathway on Rubisco biochemistry without any phylogenetic correction, treating each species as independent (TIPs analysis), and using equivalent phylogenetically independent contrasts (PICs; Felsenstein, 1985) based on a three-gene consensus phylogeny (i.e. Fig. 5 in McKown et al., 2005). In this more conservative approach the difference in trait values between the descendant species is calculated for each node of the phylogeny, resulting in n–1 contrasts where n is the number of species. PIC analyses assume a Brownian model of character evolution, and weights differences by the length of branches between descendant values. As a result, most of the observed variation is allocated to phylogenetic, rather than physiological, processes (Westoby et al., 1995). The statistical adequacy of four types of branch-lengths was examined: branch lengths from the Flaveria phylogeny (AD McKown, personal communication), branch lengths set to unity, and the branch lengths in Grafen (1989) and Pagel (1992). Based on the plots of absolute values of standardized contrasts versus their standard deviation, uniform branch lengths were adequate and were used for the PIC analysis.
To test for correlations between the continuous biochemical characters (SC/O, Kc, kcat, and Ko) Pearson's correlations (TIPs analysis) and Felsenstein's (1985) independent contrasts (PICs analysis) were used, as implemented in the PDAP module (Midford et al., 2005) of Mesquite version 1.12 (Maddison and Maddison, 2006). We examined whether each biochemical character varied between the continuous characters and photosynthetic pathway (C3, C4, C3-C4 intermediate, or C4-like) using analysis of variance (ANOVA). Analysis of covariance (ANCOVA) was used to test for differences in the slopes of the relationship between SC/O and the other biochemical characters, and between Kc and kcat, among the different pathways. For the TIPs analysis, differences between C3 and C4 species were examined using Tukey's test. For the PICs analyses, P-values returned from the ANOVA or ANCOVA were adjusted using simulations of random-walk evolution of each character state, as described below.
Both TIPS and PICs ANOVAs and ANCOVAs were conducted using the module PDSINGLE in PDAP version 6.0 (Garland et al., 2005a). PDSINGLE returns values from a conventional analysis of variance or covariance, which may inflate type I error if there is strong phylogenetic signal. For the PICs analysis, the F-statistic of each ANOVA or ANCOVA was compared to a null distribution of F-statistics generated by computer simulation. The module PDSIMUL in PDAP version 6.0 was used to generate 1000 data sets of simulated trait evolution along the Flaveria phylogeny, assuming a Brownian model of character evolution, and the module PDANOVA to conduct ANOVA and ANCOVA on these simulated datasets. If the F-value of the real data set was greater than the 95% percentile of the distribution of the simulated datasets it was considered that photosynthetic pathway had a significant effect on either the parameter (for ANOVA), or on the slope between two parameters (for ANCOVA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
SC/O and kcat
The C3 species of Flaveria (F. pringlei and F. cronquistii) had SC/O values near 81 mol mol–1 (Fig. 1). Slightly lower values were measured in the other C3 species spinach (79.8 mol mol–1) and Chenopodium album (78.7 mol mol–1). By contrast, in the C4 Flaveria species SC/O was between 75.5 mol mol–1 and 77.2 mol mol–1, similar to the specificity measured for Rubisco from corn (74.9 mol mol–1) and Amaranthus edulis (77.5 mol mol–1), but about 5% lower than SC/O in the C3 Flaveria (TIPs, P <0.001; Table 1). The C3-C4 intermediate and C4-like Flaveria species had SC/O ranging between 84.5 mol mol–1 (F. floridana) and 77.9 mol mol–1 (F. anamola).
|
) at 210 mbar O2, used to model the RuBP-saturated photosynthetic rate of C3 and C4 photosynthesis (
|
Rubisco from the two C3 species of Flaveria had carboxylation turnover rates (
) of 3.1 s–1, which were comparable to that measured for the spinach and C. album Rubiscos (Fig. 1). The of the C4 Flaveria Rubiscos ranged between 4.4 s–1 (F. trinervia) and 3.7 s–1 (F. kochiana), nearly 25% higher than the C3 species (TIPs, P <0.05; Table 1), but similar to the values measured for the Z. mays and A. edulis enzymes. The C3-C4 intermediate and C4-like Flaveria species showed high variation in . Rubisco from F. vaginata (C4-like) and F. anamola (C3-C4) had within the range of the C4 Flaveria Rubiscos (Fig. 1). By contrast, Rubisco from F. brownii (C4-like) and F. ramosissima (C3-C4) had of 2.6 s–1 and 2.7 s–1, respectively, which were nearly 15% lower than for the enzyme from the C3 Flaverias.
There was no overall effect of photosynthetic type on either SC/O or kcat, when all four photosynthetic types were considered and the phylogenetic relationships between the species were accounted for (e.g. PICs; Table 2). There was a significant negative relationship between SC/O and (Fig. 2a; Table 3). The slope of this relationship was not significantly different between the four different photosynthetic types (P >0.9, ANCOVA; Table 4).
|
|
|
|
Michaelis–Menten constants
In Flaveria, Kc was highest for the enzyme from the C4 species and the C4-like plant F. vaginata, with values exceeding 20 µM in most cases (Fig. 1; Table 1). Similarly, high Kc were measured for Rubisco from Z. mays and A. edulis. By contrast, Rubisco from the C3, C3-C4, and other C4-like Flaverias had Kc values between 13.5 µM (F. palmeri) and 10.2 µM (F. sonorensis), which were consistent with those determined in spinach (12.1 µM) and C. album (11.2 µM) Rubiscos. Statistically, the Kc for Rubiscos from C4 species of Flaveria were significantly higher than for the enzyme from the C3 species (P <0.001; Tables 1, 2). There was a strong negative relationship between Kc and SC/O in Flaveria (Fig. 2b; Table 3). However, the correlation between Kc and SC/O was not significant when analysed by independent contrasts (Table 3), probably because the slopes differed between photosynthetic pathway (Table 4).
Unlike Kc, there was considerable variation in Ko for the different Rubiscos both within and between photosynthetic pathway, and no significant differences were detected (Fig. 1; Tables 1, 2). For Rubiscos from the C3 species of Flaveria, Ko was about 660 µM, which was 13–37% higher than the values determined for the enzyme from spinach and C. album. In Rubisco isolated from the C4 plants F. bidentis and F. trinervia, Ko was the same as the enzyme from the C3 Flaverias. By contrast, Rubisco from the C4 species F. kochiana and F. australasica had Ko values of 150 µM and 309 µM, respectively, similar to the low Ko values for the Rubiscos from the C4 plants Z. mays (157 µM) and A. edulis (289 µM). Notably there was considerable variation in the Ko of Rubisco from C3-C4 and C4-like Flaveria species, even between species that were on adjacent branches of the phylogeny. For example, in clade A the C4-like F. palmeri had a Rubisco with a Ko of 193 µM, while the enzyme from the C3-C4 intermediate F. ramosissima had a Ko of 722 µM. By contrast, there was less variability in Ko in Rubisco from clade B C3-C4 and C4-like species (Fig. 1). No significant relationship was detected between SC/O and Ko overall (Fig. 2c; Tables 3, 4), although it was noted that the Ko values from F. brownii and F. palmeri Rubiscos have a strong effect on the analysis.
No consistent relationship was detected between the photosynthetic pathway and the Michaelis–Menten constant for RuBP (Km-RuBP) (Fig. 1). For the Flaveria Rubiscos the Km-RuBP ranged between 9–23 µM, while the enzyme from the C3 and C4 control species had Km-RuBP 7–26 µM.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
This study presents an analysis of the biochemical characteristics of Rubisco from closely related angiosperm taxa with differing photosynthetic pathways. The data include the most extensive measurements of the Michaelis–Menten constant for oxygen (Ko) for C4 species published to date, and highlights that the oxygenase function of Rubisco may be more variable than previously realized. The expected negative relationship was observed between the enzyme's SC/O and
, but this is not significantly altered by photosynthetic pathway. By contrast, the negative correlation between SC/O and Kc was affected by the photosynthetic pathway and phylogenetic relationship between species. Our data are consistent with previous studies that have shown the kinetics of Rubisco to differ between C3 and C4 plants, and it is suggested that this represents an evolutionary reversion of the enzyme's efficiency. Last, it is shown that conclusions drawn from observed variation in the kinetic properties of Rubisco from distinct photosynthetic pathways may change if the phylogenetic relatedness of the species is considered.
Ko of Rubisco from C4 plants
The Michaelis–Menten constant for oxygenation (Ko) is clearly the least reported Rubisco kinetic parameter, and the technical complexity associated with its determination has resulted in considerable variation in the Ko values reported in the literature. In C3 species generally, Ko values between 200 and 650 µM have been reported (see von Caemmerer, 2000), with values of 500 µM (Jordan and Ogren, 1984) and 354 µM (Badger and Andrews, 1974) being reported for spinach Rubisco. We obtaind slightly higher Ko values in Rubisco from spinach and the C3 Flaverias, however, they were within the range of variation reported for Rubisco from other C3 species. There have been only two studies that report Ko values for C4 Rubiscos. Badger et al. (1974) reported Ko values of 550 and 610 µM for corn Rubisco, and 710 µM for the Atriplex spongiosa enzyme, while Jordan and Ogren (1981) reported Ko values of 640 µM and 810 µM in Amaranthus hybridus and corn Rubiscos, respectively. We obtained similar Ko results for Rubisco from the C4 species Flaveria bidentis and F. trinervia, but across all of the C4 species examined here Ko varied more than 4-fold, and in many cases was considerably lower than was observed in C3 species. The effect of O2 on the measured Kc for Rubisco from four different Flaveria species is shown in Fig. 3. Consistent with O2 being a competitive inhibitor of CO2 there was a linear relationship between O2 and the measured Kc for all species examined. The reproducibility of these data raises questions as to the possible reasons for the discrepancies between our Ko measurements and those reported previously. For example, our Ko for the maize enzyme was more than 4-fold lower than that measured previously (Badger et al., 1974; Jordan and Ogren, 1981).
|
The variability in Ko is evident in the large range of apparent
values, which represent Kc in the presence of atmospheric levels of O2 (Fig. 1). The enzyme from C3 Flaveria species is half-saturated for CO2 at 450–500 µbar, slightly higher than the present atmospheric pCO2. By contrast, in the C4 F. kochiana half-saturation requires CO2 pressures more than four-times the current ambient level. To attain the same Rubisco efficiency of the C3 Flaverias the CO2 pump in Flaveria kochiana must be very efficient to suppress the oxygenase function of Rubisco; recent gas-exchange analysis supports this conclusion (Sudderth et al., 2007). However, even in the presence of a CO2-concentrating mechanism there is no benefit to having a low Ko, as any oxygenation would require that resources are invested in the photorespiratory pathway. Does the occurrence of extremely low Ko values suggest that this parameter is not under strong selection in C4 plants? If Ko was not under selection then it should evolve randomly, and so would change in both directions. Further, it is difficult to envision a mechanism by which selection on the carboxylation reaction (which is clearly occurring, see below), could become uncoupled from that of the oxygenation function of Rubisco. Regardless, the expectation that Rubisco from C4 species will have a higher Ko than the enzyme from C3 plants clearly warrants further study (von Caemmerer, 2000; von Caemmerer and Quick, 2000).
Rubisco in C3 and C4 plants
Our results confirm previous reports that Rubisco from C4 plants has higher and Kc than the enzyme from C3 species (Yeoh et al., 1980, 1981; Seemann et al., 1984; Wessinger et al., 1989; Sage, 2002). Wessinger et al. (1989) suggested that C3-C4 intermediate species of Flaveria had Rubisco kinetics more like that of the C3 species. Our results support this assertion in general, although there are differences in the way the photosynthetic pathway of species were classed in the two studies. Overall, the Kc values determined here agree reasonably well with those reported by Wessinger et al. (1989), although they reported slightly higher values near 14 µM for the C3 Flaveria Rubiscos. The reported by Wessinger et al. (1989) for Flaveria were considerably higher than those measured here, particularly for the Rubiscos from C4 species. The reasons for this are unclear, but do not appear to be related to the particular assay conditions as these were similar. Kubien et al. (2003) reported a of 3.9 s–1 for F. bidentis, while Sage (2002) reported a of 4.9 s–1 for seven C4 species at 28 °C. Assuming an activation energy of 56.1 kJ mol–1 (Kubien et al., 2003) this converts to 3.9 s–1 at 25 °C, which is very similar to the reported for C4 Flaverias in the present study (Fig. 1).
The structural basis for the catalytic differences between the Flaveria Rubiscos is uncertain. Hudson et al. (1990) investigated the Rubisco large (L) subunit sequences from pairs of C3 and C4 species of Flaveria, Atriplex, and Neurachne. Only one conserved amino acid difference (Met-309 to Ile) occurred across all three comparisons. However, because this position is occupied by Ile in other C3 Rubiscos (e.g. Nicotiana tabacum, Chlamydomonas reinhardtii), Hudson et al. (1990) concluded that this could not explain the altered kinetic phenotype, at least by itself. The comparison of C3 (F. pringlei) and C4 (F. trinervia and F. bidentis) L-subunit sequences identified three amino acid differences in total, none of which were suitability positioned to effect the observed variation in kinetic traits (Hudson et al., 1990). Curiously, the Rubisco small (S) units from F. pringlei (Genbank AAB67851 [GenBank] ) and F. bidentis (AAP31054 [GenBank] ) show significant (>93%) homology, so it is difficult to conclude the extent to which the kinetic differences in Flaveria Rubiscos are influenced by the S-subunit. Notably, although the precise role of the S-subunit is uncertain (Spreitzer, 2003) it is essential for maximum catalytic activity (Andrews and Ballment, 1983) and sequence modification to S-subunit residues, far from the catalytic sites on the L-subunit, have a pervasive effect on the kinetic properties of the enzyme (Spreitzer, 2003). The influence of the S-subunit on Rubisco catalysis suggests that further study on the multiple nuclear-encoded rbcS sequences may identify residues, or structural motifs, that influence kinetic variability amongst Rubiscos, particularly in closely related species where the L-subunit is almost totally conserved.
Rubisco's biochemical trade-offs
It can be shown mathematically that SC/O is a function of the rate constants for the addition of CO2 or O2 to the 2,3-enediol of RuBP, and that the kcats are not a principal determinant (Farquhar, 1979; Tcherkez et al., 2006). However, in all cases examined there is an inverse relationship between SC/O and (Bainbridge et al., 1995; Badger et al., 1998; Tcherkez et al., 2006). Tcherkez et al. (2006) showed that SC/O is also inversely related to Kc across an evolutionarily broad range of Rubiscos. Our data showed similar relationships between SC/O and , where the slope of the relationship was not affected by photosynthetic pathway, and between SC/O versus Kc, where a pathway-related difference in the slope was detected (Fig. 2b; Table 4). Rubisco from the C4 plants (including ‘control’ species) and F. vaginata had a much different relationship between SC/O and Kc than the other enzymes examined (Table 4).
To explore the evolution of Rubisco further, we plotted the efficiencies of carboxylation (i.e. /Kc) versus oxygenation () for the Flaveria Rubiscos, and others from the literature (Fig. 4). The slope of this plot is equal to 1/SC/O (Badger and Andrews, 1987). Two strategies have been proposed for the evolution of Rubisco in response to the decline in atmospheric CO2 levels that has occurred since the enzyme evolved (Badger and Andrews, 1987). First, the specificity for CO2 can increase relative to that for O2, increasing the rate of carboxylation relative to oxygenation. Second, the two kcat/Km terms can increase in proportion, leaving SC/O unchanged but allowing a constant rate of carboxylation in the falling CO2. These strategies are indicated in the inset to Fig. 4. In the Flaveria and ‘control’ species measured here there is a strong linear relationship between /Kc and (Fig. 4). The enzyme from higher plant sources, including Flaveria, tend to cluster and show rather little variation in SC/O, when compared to Rubisco from wide evolutionary lineages. Our data support the conclusion of Badger and Andrews (1987) that higher plants have employed strategy 1 in response to declining ratios of CO2/O2 in the atmosphere; the higher plant Rubiscos are clearly on a lower slope than the Form II enzyme of Rhodospirillum rubrum (Rr) or the Form I Rubisco of the Chromatium vinosum (Cv). However, during the transition from C3 to C4 plants it appears that Rubisco underwent what amounts to an evolutionary reversion, where the specificities for both CO2 and O2 declined as the CO2 experienced by the enzyme increases. This can be seen by comparing the C3 (black squares) with the C4 enzyme (white squares) in Fig. 4, and is essentially the second strategy proposed by Badger and Andrews (1987).
|
Sage (2004) suggested that ‘optimization’ of enzyme characteristics is the final step in the evolution of completely expressed C4 photosynthesis. However, these data suggest that this might not be a universal pattern, as F. vaginata clearly has a Rubisco that kinetically resembles the enzyme from the recognized C4 Flaverias (Figs 1, 2). We are not suggesting that F. vaginata is a C4 species; although the leaf anatomy is very similar to the C4 Flaverias (McKown and Dengler, 2007) the O2-sensitivity of photosynthesis in considerably higher than the other C4 plants (Ku et al., 1991). Flaveria vaginata may represent a case where the evolution of a C4-type Rubisco preceded the development of full C4 photosynthesis. von Caemmerer and Quick (2000) suggested that if C4 species evolved from C3 ancestors then high CO2 affinity must be a readily labile characteristic of Rubisco. Significant changes to the biochemistry of Rubisco are readily apparent in C4 species, and high CO2 affinity must be a labile trait for this to have happened in the relatively short time since the C4 and C3 Flaverias diverged. Similarly, differences in SC/O among Limonium sp. (C3) may be the result of rapid selection on Rubisco kinetics to maximize fitness in arid habitats (Galmés et al., 2005).
Phylogenetic analysis of continuous enzymatic properties
Previous studies have concluded that there is little variation in SC/O between C3 and C4 plants (Jordan and Ogren, 1981; Kane et al., 1994). If analysed without consideration of phylogeny it was possible to detect a small but significant reduction in the SC/O of Rubiscos from C4 Flaverias compared with the enzyme from the C3 species, and our measurements on four non-Flaveria species are consistent with this. If most of the variation in SC/O is assumed to be due to phylogeny rather than photosynthetic pathway (e.g. PIC analysis) then the differences between pathways are non-significant. Differences in the results between PICs and TIPs analyses may indicate a loss of statistical power, as closely related taxa were likely to have similar photosynthetic pathways. Standardized phylogenetically independent contrasts are calculated as the difference in phenotypic values between sister taxa or node for the character of interest, divided by the total branch length between the taxa, and assumes a certain model of trait evolution (generally Brownian). Thus, these analyses are sensitive to the topology of the phylogeny, the assumption of branch lengths, and the model of character evolution (Felsenstein, 1985; Donoghue and Ackerly, 1996; Garland et al., 2005b).
Statistical issues aside, differences in analyses using TIPs versus PIC may provide some indication of the evolutionary lability of the traits of interest. For example, large changes that occur early in an adaptive radiation will show the greatest difference between the two analyses (Price, 1997). The severe structural, anatomical, and biochemical modifications necessary for transitions between C3 and C4 photosynthesis suggest strong stabilizing selection for the maintenance of certain adaptations, such as the localization of glycine decarboxylase to the bundle sheath, that seem to be necessary for the evolution of C4 photosynthesis (Monson, 1999; Sage, 2004). Indeed, the occurrence of one documented case of a reversion from C4 to C3 (Ellis, 1984) suggest that fully-expressed C4 photosynthesis is strongly maintained, and represents an essentially irreversible evolutionary transition (sensu Bull and Charnov, 1985). Because the evolutionary transitions between C3 and C4 photosynthesis are highly constrained, and because few genera contain both pathways, the problems associated with using phylogenetically-independent contrasts will be found in studies of other groups below the family level. Larger analyses of photosynthetic pathway and biochemical traits across families may help resolve some of these issues.
|
Acknowledgements |
|---|
DSK wishes to thank the Molecular Plant Physiology group at RSBS for his enjoyable stay in Canberra, and in particular Heather Kane for her patient guidance. Andrew McMurtrie assisted with Rubisco isolation. This work was supported by a Natural Science and Engineering Research Council of Canada Discovery Grant (327103-06) to DSK, and an Australian Research Council Discovery Grant (DP0450564) to SMW.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Andrews TJ, Ballment B. The function of the small subunits of ribulose-1,5-bisphosphate carboxylase oxygenase. Journal of Biological Chemistry (1983) 258:7514–7518.
Andrews TJ, Whitney SM. Manipulating ribulose bisphosphate carboxylase/oxygenase in the chloroplasts of higher plants. Archives of Biochemistry and Biophysics (2003) 414:159–169.[Web of Science][Medline]
Badger MR, Andrews TJ. Effects of CO2, O2, and temperature on a high-affinity form of ribulose disphosphate carboxylase-oxygenase from spinach. Biochemical and Biophysical Research Communications (1974) 60:204–210.[CrossRef][Web of Science][Medline]
Badger MR, Andrews TJ. Co-evolution of Rubisco and CO2 concentrating mechanisms. In: Progress in photosynthesis research—Biggens J, ed. (1987) Vol. III. The Netherlands: Kluwer. 601–609.
Badger MR, Andrews TJ, Osmond CB. Detection in C3, C4 and CAM plant leaves of a low-Km(CO2) form of RuDP carboxylase, having high RuDP oxygenase activity at physiological pH. In: Proceedings of the third international congress on photosynthesis—Avron M, ed. (1974) The Netherlands: Elsevier. 1421–1428.
Badger MR, Andrews TJ, Whitney SM, Ludwig M, Yellowlees DC, Leggat W, Price GE. The diversity and co-evolution of Rubisco, plastids, pyrenoids, and chloroplast-based CO2-concentrating mechanisms in algae. Canadian Journal of Botany (1998) 76:1052–1071.
Bainbridge G, Madgwick P, Parmar S, Mitchell R, Paul M, Pitts J, Keys AJ, Parry MAJ. Engineering Rubisco to change its catalytic properties. Journal of Experimental Botany (1995) 46:1269–1276.[Web of Science]
Bull JJ, Charnov EL. On irreversible evolution. Evolution (1985) 39:1149–1155.[CrossRef][Web of Science]
Delgado E, Medrano H, Keys AJ, Parry MAJ. Species variation in Rubisco specificity factor. Journal of Experimental Botany (1995) 46:1775–1777.
Donoghue MJ, Ackerly DD. Phylogenetic uncertainties and sensitivity analyses in comparative biology. Philosophical Transactions of the Royal Society, Series B, Biological Sciences (1996) 351:1241–1249.[CrossRef]
Ellis RJ. The most abundant protein in the world. Trends in Biochemical Sciences (1979) 4:241–244.[CrossRef][Web of Science]
Ellis RP. Eragrostis walteri: a first record of non-Kranz leaf anatomy in the sub-family Chloridiodeae (Poaceae). South African Journal of Botany (1984) 3:380–386.[Web of Science]
Farquhar GD. Models describing the kinetics of ribulose bisphosphate carboxylase oxygenase. Archives of Biochemistry and Biophysics (1979) 193:456–468.[CrossRef][Web of Science][Medline]
Felsenstein J. Phylogenies and the comparative method. American Naturalist (1985) 125:1–15.[CrossRef][Web of Science]
Galmés J, Flexas J, Keys AJ, Cifre J, Mitchell RAC, Madgwick PJ, Haslam RP, Medrano H, Parry MAJ. Rubisco specificity factor tends to be larger in plant species from drier habitats and in species with persistent leaves. Plant, Cell and Environment (2005) 28:571–579.[CrossRef]
Garland T, Bennett AF, Rezende EL. Phylogenetic approaches in comparative physiology. Journal of Experimental Biology (2005) 208:3015–3035.
Grafen A. The phylogenetic regression. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences (1989) 326:119–157.[CrossRef][Web of Science]
Garland T Jr, Midford PE, Jones JA, Dickerman AW, Diaz-Uriarte R. PDAP: Phenotypic Diversity Analysis Programs. Version 6 (2005).
Hudson GS, Mahon JD, Anderson PA, Gibbs MJ, Badger MR, Andrews TJ, Whitfeld PR. Comparisons of rbcL genes for the large subunit of ribulose 1,5-bisphosphate carboxylase from closely related C3 and C4 plant species. Journal of Biological Chemistry (1990) 265:808–814.
Jordan DB, Ogren WL. Species variation in the specificity of ribulose 1,5-bisphosphate carboxylase/oxygenase. Nature (1981) 291:513–515.[CrossRef]
Jordan DB, Ogren WL. The CO2/O2 specificity of ribulose-1,5-bisphosphate carboxylase/oxygenase. Planta (1984) 161:308–313.[CrossRef][Web of Science]
Kane HJ, Viil J, Entsch B, Paul K, Morell MK, Andrews TJ. An improved method for measuring the CO2/O2 specificity of ribulose-bisphosphate carboxylase-oxygenase. Australian Journal of Plant Physiology (1994) 21:449–461.[Web of Science]
Kane HK, Wilkin JM, Portis AR, Andrews TJ. Potent inhibition of ribulose-bisphosphate carboxylase by an oxidized impurity in ribulose-1,5-bisphosphate. Plant Physiology (1998) 117:1059–1069.
Ku MSB, Wu J, Dai Z, Scott RA, Chu C, Edwards GE. Photosynthetic and photorespiratory characteristics of Flaveria species. Plant Physiology (1991) 96:518–528.
Kubien DS, von Caemmerer S, Furbank RT, Sage RF. C4 photosynthesis at low temperature: a study using transgenic plants with reduced amounts of Rubisco. Plant Physiology (2003) 132:1577–1585.
Laing WA, Ogren WL, Hageman RH. Regulation of soybean net photosynthetic CO2 fixation by the interaction of CO2, O2 and ribulose-1,5-diphosphate carboxylase. Plant Physiology (1974) 54:678–685.
Maddison WP, Maddison DR. Mesquite: a modular system for evolutionary analysis. Version 1.12 (2006) http://mesquiteproject.org.
McKown A, Moncalvo J-M, Dengler NG. Phylogeny of Flaveria (Asteraceae) and inference of C4 photosynthesis evolution. American Journal of Botany (2005) 92:1911–1928.
McKown A, Dengler NG. Key innovations in the evolution of Kranz anatomy and C4vein pattern in Flaveria (Asteraceae). American Journal of Botany (2007) 94:382–399.
Midford PE, Garland TJ, Maddison WP. PDAP package of Mesquite. Version 1.07 (2005) Website http://mesquiteproject.org/pdap_mesquite/index.html.
Monson RK. The origins of C4 genes and evolutionary pattern in the C4 metabolic phenotype. In: C4 plant biology—Sage RF, Monson RK, eds. (1999) Toronto: Academic Press. 377–410.
Paech C, Dybing CD. Purification and degradation of ribulose bisphosphate carboxylase from soybean leaves. Plant Physiology (1986) 81:97–102.
Pagel MD. A method for the analysis of comparative data. Journal of Theoretical Biology (1992) 156:431–442.[CrossRef][Web of Science]
Paul K, Morell MK, Andrews TJ. Mutations in the small subunit of ribulose bisphosphate carboxylase affect subunit binding and catalysis. Biochemistry (1991) 30:10019–10026.[CrossRef][Web of Science][Medline]
Price T. Correlated evolution and independent contrasts. Philosophical Transactions of the Royal Society, Series B (1997) 352:519–529.[CrossRef]
Read BA, Tabita FR. High substrate specificity factor ribulose bisphosphate carboxylase/oxygenase from eukaryotic marine algae and properties of recombinant cyanobacterial Rubisco containing ‘algal’ residue modifications. Archives of Biochemistry and Biophysics (1994) 312:210–218.[CrossRef][Web of Science][Medline]
Ruuska S, Andrews TJ, Badger MR, Hudson GS, Laisk A, Price GD, von Caemmerer S. The interplay between limiting processes in C3 photosynthesis studied by rapid-response gas exchange using transgenic tobacco impaired in photosynthesis. Australian Journal of Plant Physiology (1998) 25:859–870.[Web of Science]
Sage RF. Variation in the kcat of Rubisco in C3 and C4 plants and some implications for photosynthetic performance at high and low temperature. Journal of Experimental Botany (2002) 53:609–620.
Sage RF. The evolution of C4 photosynthesis. New Phytologist (2004) 161:341–370.[CrossRef][Web of Science]
Seemann JR, Badger MR, Berry JA. Variations in the specific activity of ribulose-1,5-bisphosphate carboxylase between species utilizing different photosynthetic pathways. Plant Physiology (1984) 74:791–794.
Spreitzer RJ. Role of the small subunit in ribulose-1,5-bisphosphate carboxylase/oxygenase. Archives of Biochemistry and Biophysics (2003) 414:141–149.[CrossRef][Web of Science][Medline]
Sudderth EA, Muhaidat RM, McKown AD, Kocacinar F, Sage RF. Leaf anatomy, gas exchange and photosynthetic enzyme activity in Flaveria kochiana. Functional Plant Biology (2007) 34:118–129.[CrossRef][Web of Science]
Tcherkez GGB, Farquhar GD, Andrews TJ. Despite slow catalysis and confused substrate specificity, all ribulose bisphosphate carboxylases may be nearly perfectly optimized. Proceedings of the National Academy of Sciences, USA (2006) 103:7246–7251.
von Caemmerer S. Biochemical models of leaf photosynthesis (2000) Canberra: CSIRO Publishing.
von Caemmerer S, Furbank RT. Modelling C4 photosynthesis. In: C4 plant biology—Sage RF, Monson RK, eds. (1999) Toronto: Academic Press. 173–211.
von Caemmerer S, Quick WP. Rubisco: physiology in vivo. In: Photosynthesis: physiology and metabolism—Leegood RC, Sharkey TD, von Caemmerer S, eds. (2000) The Netherlands: Kluwer Academic Publishers. 85–113.
Wessinger ME, Edwards GE, Ku MSB. Quantity and kinetic properties of ribulose-1,5-bisphosphate carboxylase in C3, C4, and C3-C4 intermediate species of Flaveria (Asteraceae). Plant and Cell Physiology (1989) 30:665–671.
Westoby M, Leishman MR, Lord JM. On misinterpreting the ‘phylogenetic correction’. Journal of Ecology (1995) 83:531–534.[CrossRef][Web of Science]
Whitney SM, Baldet P, Hudson GS, Andrews TJ. Form I Rubiscos from non-green algae are expressed abundantly but not assembled in tobacco chloroplasts. The Plant Journal (2001) 26:535–547.[CrossRef][Web of Science][Medline]
Yeoh H-H, Badger MR, Watson L. Variations in Km (CO2) of ribulose 1,5-bisphosphate carboxylase among grasses. Plant Physiology (1980) 66:1110–1112.
Yeoh H-H, Badger MR, Watson L. Variations in kinetic properties of ribulose 1,5-bisphosphate carboxylase among plants. Plant Physiology (1981) 67:1151–1155.
Zhu XG, Portis AR, Long SP. Would transformation of C3 crop plants with foreign Rubisco increase productivity? A computational analysis extrapolating from kinetic properties to canopy photosynthesis. Plant, Cell and Environment (2004) 27:155–165.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P.-A. Christin, N. Salamin, A. M. Muasya, E. H. Roalson, F. Russier, and G. Besnard Evolutionary Switch and Genetic Convergence on rbcL following the Evolution of C4 Photosynthesis Mol. Biol. Evol., November 1, 2008; 25(11): 2361 - 2368. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||