SEARCH

SEARCH BY CITATION

Keywords:

  • calorimetry;
  • growth capacity;
  • growth regulation;
  • respiration;
  • temperature response

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Respiratory acclimation to growth temperature differs between species, but underlying mechanisms are poorly understood. In the present study, we tested the hypothesis that respiratory acclimation of CO2 release is a consequence of growth regulation such that growth rates of young foliage of Eucalyptus spp. are similar at contrasting growth temperatures. Further, we tested whether such a response is affected by adaptation of Eucalyptus to different thermal environments via growth at different altitudes in the Australian Alps.
  • We employed calorimetric methods to relate rates of CO2 release (mainly from substrate oxidation) and rates of O2 reduction to conservation of energy. Temperature responses of these processes provided insight into mechanisms that control energy conservation and expenditure, and helped define ‘instantaneous enthalpic growth capacity’ (CapG).
  • CapG increased with altitude, but was counteracted by other factors in species adapted to highland habitats. The acclimation response was partly driven by changes in respiratory capacity (inline image), and partly by more pronounced dynamic responses of CO2 release inline image to measurement temperature. We observed enhanced temperature sensitivity of O2 reduction inline image at higher altitudes.
  • Adaptation to growth temperature included differences in respiration and growth capacities, but there was little evidence that Eucalyptus species vary in metabolic flexibility.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Current distributions of plant species reflect evolutionary adaptation to a variety of environmental factors. Amongst these, temperature and precipitation are pre-eminent in defining the climate of a region, and together with characters such as soil moisture retention define the duration of growth periods for plants. Geographic patterns of plant growth and productivity reflect the climate under which plant communities evolved (Schlesinger, 1997; Schulze et al., 2002). For instance, plant development requires favourable temperature conditions and therefore proceeds in accordance with thermal time (i.e. the classical temperature-sum concept of plant development; Thornley & Johnson, 1990; Granier et al., 2002). At high latitudes and/or in alpine areas, the time suitable for plant development varies with altitude, as does the species composition of plant communities (von Humboldt, 1845).

In alpine areas, plants not only have to cope with low average growth temperatures (or short thermal time), but also have to cope with comparatively large temporal temperature fluctuations. These recurrent and persistent patterns form the basis for adaptation of plant metabolism to climate (sensuCriddle et al., 2005; Lambers et al., 2008). We know that highland and lowland species differ in phenology (Mooney, 1963; Friend & Woodward, 1990), and it is widely presumed that such differences are reflected in adaptation of physiological traits, driving plant development and growth. Rates of growth generally co-vary with rates of respiration, but it remains difficult to distinguish between respiratory acclimation to seasonal or daily temperature variation and adaptation of metabolism to prevailing climate.

Studies over many decades suggest that plant respiration (defined as rate of release of CO2 per unit plant mass, unless otherwise stated) acclimates quickly to growth temperature (Rook, 1969; Collier & Cummins, 1990; Atkin et al., 2000; Lee et al., 2005; Campbell et al., 2007). Acclimation within days to weeks of a shift in temperature can result in respiration rates remaining similar at each growth temperature (provided they are measured at the respective growth temperature). Newly developed tissues seemingly display a ‘memory’ of temperatures experienced during development, providing a significant component of the seasonal acclimation of plant metabolism. For example, plants that develop under colder conditions have greater mitochondrial density (and/or increased leaf protein contents; Klikoff, 1966; Stitt & Hurry, 2002), and cold-developed mitochondria often show changes in ultrastructure (Armstrong et al., 2006b). Enhanced capacity for respiration helps underpin required rates of anabolic and catabolic reactions (i.e. Billings et al., 1971) as, under cold conditions, the maximum velocities (vmax) of these processes are heavily dependent on the total amount of enzymes present (Kruse et al., 2011). Homeostatic responses to cold temperatures have been widely interpreted as a means of maintaining rates of metabolism under adverse conditions (Amthor, 2000). Near-neighbour competition is common in early stages of vegetative growth and individual seedlings or plantlets may need to maximize rates of growth (in competition for light), if they are to survive (Grime, 1977). By contrast, if growth at the seedling stage is ‘maximized’ at high temperatures, plants may be at risk of carbohydrate starvation (Friend & Woodward, 1990; Way & Oren, 2010). We recently suggested (Kruse et al., 2011) that acclimation of respiration to seasonal, and even short-term fluctuations of temperature, might thus reflect plant species’ ability to stabilize rates of growth and development according to climate. In fact, growth rates of any given plant species are surprisingly constant across a wide range of growth temperatures (Atkin et al., 2006). Respiratory acclimation can be viewed as the cumulative effects or ‘memory’ of environmental conditions during development. Acclimation of this type ensures optimization of growth and maintenance processes such that plants avoid substrate depletion. Further, and on the basis that long-term average temperature is a primary driver of rates of growth, acclimation of metabolism is a mechanism that helps compensate for deviations from such average temperatures. It is presently not known if species that are adapted to short vegetation period (or low average temperature) differ in respiratory plasticity, that is, their ability to metabolically acclimate to temperature fluctuations. Such an analysis is complicated by additional factors that determine relative growth rates (RGRs; the increase in plant mass per unit starting mass and time).

RGR can be defined as a function of net assimilation rate (NAR; the increase in plant mass per unit leaf area and time), Specific leaf area (SLA; leaf area per unit mass) and biomass allocation (leaf mass ratio (LMR); leaf mass per unit plant mass) (Lambers & Poorter, 2004; Atkin et al., 2006):

  • image(Eqn 1)

The term NAR can be expressed as the difference between daily, leaf area-based net photosynthesis and respiration, divided by plant carbon concentrations – as affected by storage and remobilization processes (Atkin et al., 2006). While NAR is driven by growth processes subject to short-term stabilization, changes in leaf architecture and biomass allocation are more important for regulation of plant growth in the mid and long term.

In relation to acclimation of short-term processes relevant to NAR, Tjoelker et al. (1999) found a close correlation between RGR and rates of respiration in evergreen and broad-leafed seedlings. Evergreen species showed a greater degree of respiratory acclimation than deciduous species (Tjoelker et al., 1999). Physiological plasticity apparently played a greater role in the ‘stabilized’ development of evergreen perennials than plasticity of leaf structure and biomass allocation (also see Bruhn et al., 2007 for a study with Eucalyptus spp.). Evergreen species generally display lower SLA values than deciduous species, and slow-growing herbaceous species have lower SLA values than fast-growing species (Poorter et al., 2009). The slow growth of some alpine herbaceous species seems mostly a result of lower SLA when compared with lowland species (e.g. Atkin et al., 1996), and growth and respiration are not necessarily correlated (e.g. Atkin & Day, 1990). Larigauderie & Körner (1995) found no systematic differences in the extent of respiratory acclimation to contrasting growth temperatures between lowland and alpine herbaceous species.

Large species-specific differences might be best related to different growth and life strategies. Some species control rates of growth primarily through acclimation of metabolism (i.e. Arnone & Körner, 1997), while others swiftly adjust structural traits. For example, plasticity of SLA is different in tropical and boreal species, and changes in SLA for a given change in temperature are larger in tropical species (Poorter et al., 2009). As another example, Loveys et al. (2002) demonstrated varied importance of physiological and morphological traits (particularly SLA) to differences in RGRs among species. The importance of those traits also varied among growth temperatures. For slow-growing species at moderately cold temperatures (18°C), variations in growth rates were better explained by variations in NAR than variation in SLA. Put differently, slow-growing species exhibited stronger acclimation of area-based respiration to temperature than fast-growing species (Loveys et al., 2002). This conclusion could not be confirmed in a subsequent study of mature leaves (Loveys et al., 2003) and, to reconcile the difference between studies, Loveys et al. (2003) proposed that the degree of acclimation of developing leaves may well differ from that of mature leaves. This hypothesis was confirmed for Arabidopsis by Armstrong et al. (2006a).

In the present study, we investigated physiological acclimation in young Eucalyptus seedlings, and explored variation in acclimation linked to adaptive differences between species. We studied 12 evergreen Eucalyptus species that are adapted to low, mid and high altitudes. Seedlings from each of these species were grown at four sites, along an altitude gradient in the Australian Alps. Physiological acclimation of young, developing foliage (e.g. < 1 cm2) is seldom determined because of the physical constraints of conventional CO2-exchange systems. Micro-calorimetric methods provide an excellent means of analysing small, but vigorously respiring and actively growing tissue and of simultaneously determining rates of CO2 release and O2 reduction. In the present study, we confined respiration measurements to small, newly emerged foliage. That is, respiratory acclimation was not confounded by adjustment of leaf structural traits at this early stage of development. Instantaneous rates of growth (or ‘enthalpic growth’) can be readily determined (on a per unit mass basis), as can the temperature response of respiration (Criddle et al., 1997, 2000). We show how a new parameter derived from the temperature response of instantaneous growth rates, the instantaneous growth capacity of young foliage, provides a useful tool for separating acclimation and adaptation of plant metabolism to temperature. This parameter helps to describe principal relationships between respiration and growth and helps to test the hypothesis that respiratory acclimation is a consequence of growth regulation. Further, we sought to understand the mechanistic basis for increased fitness of Eucalyptus species that are adapted to differing thermal environments.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Site description and plant material

Twelve different Eucalyptus species were studied. Species were grouped by ecotype according to their natural geographic distributions (see Supporting Information Table S1). The ‘low-altitude species’ comprised Eucalyptus botryoides Sm., Eucalyptus tereticornis ssp. tereticornis, Eucalyptus radiata Sieber ex Dc. ssp. radiata and Eucalyptus sideroxylon Wools. These species are common in the foothills of the Australian Alps, at altitudes from 200 to 500 m above sea level (asl). Eucalyptus cypellocarpa L. A. S. Johnson, Eucalyptus dives Schauer, Eucalyptus fastigata H. Deane & Maiden, Eucalyptus globulus Labill. ssp. bicostata, Eucalyptus nitens H. Deane & Maiden and Eucalyptus sieberi L. A. S. Johnson were classified as ‘mid-altitude species’ as they are most common at elevations of 400–900 m asl. The final two species, Eucalyptus delegatensis R. T. Baker ssp. delegatensis and Eucalyptus pauciflora Sieber ex Spreng. ssp. pauciflora, were classified as ‘high-altitude species’, given that they dominate areas above 900 m asl (E. pauciflora continues to the tree-line at altitudes varying from c. 1300 to 1700 m depending on physiography).

Seeds were obtained from the Australian Tree Seed Centre (Kingston, ACT, Australia). Seeds of those species requiring stratification (E. delegatensis and E. pauciflora) were cooled to 3°C for 48 h before germination. Germinants (defined as cotyledons plus one to three sets of leaves) were transferred to 9-l pots containing 60% standard potting mix, 30% sandy loam and 10% cocoa peat. Nutrients were provided via Osmocote® (Scotts Australia PTY Ltd., Baulkham Hills, Australia) slow-release fertilizer granules (N : P : K of 17 : 8.5 : 1.5 plus trace elements) and plants were watered to field capacity daily.

Ten seedlings per species were grown at four sites (spanning an 18-km transect) of contrasting altitude in north-east Victoria, Australia: Falls Creek (1600 m asl), Howman’s Gap (1200 m asl), Bogong Village (700 m asl) and Mount Beauty (400 m asl), from February to May 2007 (Table S2). Average midday growth temperatures declined with increasing altitude at a rate of 0.011°C m−1 in February, 0.009°C m−1 in March, 0.007°C m−1 in April and 0.005°C m−1 in May (Table S2). Respiration measurements were conducted in late May/early June 2007, covering a time period of 3 wk. There were generally four independent replicates per species and site (but two to four replicates per species at Falls Creek), and respiration measurements were conducted in an alternating fashion.

Respiration measurements and determination of respiration variables

Respiration measurements were conducted with the youngest, not fully expanded leaf, growing next to the apex. After excision, these small and intact leaves were put into 1-cm3 ampoules, and subsequently placed into multi-cell differential-scanning calorimeters (CSC 4100, MC-DSC; TA Instruments-Waters LLC, Lindon, Utah, USA) for measurement of respiration at 10, 15, 20 and 25°C, as described by Kruse et al. (2008). At each measurement temperature, two respiration variables were recorded: first, the heat rate of the sample (q) alone and secondly, the heat rate of the sample in the presence of NaOH (qNaOH). This second heat rate is generally greater than the first, as a consequence of the exothermic formation of carbonate from CO2 released from plant material. Rates of CO2 production (inline image) are calculated as the difference between the second and first heat rates, taking into account the enthalpy change for carbonate formation (−108.5 kJ mol−1) (Criddle & Hansen, 1999).

Two sources of CO2 release from plant tissue (inline image) in the dark can be identified. The bulk of CO2 released results from oxidative decarboxylation in the mitochondrial matrix. All of the reducing equivalents that are produced in citric acid cycle processes are channelled into the electron transport chain of the inner mitochondrial membrane, with O2 as the terminal electron acceptor. A second source is the comparatively large and variable fraction of CO2 that is related to decarboxylation processes outside mitochondria. Most of these processes are oxidative (e.g. those catalysed by glucose-6-phosphate dehydrogenase or cytosolic isocitrate dehydrogenase), although some CO2 may arise from direct substrate degradation (i.e. reduction by loss of CO2). Part of the reducing equivalents generated by oxidative processes outside plant mitochondria may enter the electron transport chain via external dehydrogenases, while some are needed for reductive anabolism.

Reductant flow is directed towards O2 as well as anabolic products, and the relative rates of these flows vary during plant development (i.e. production of reduced organic compounds is much greater in the early stages of leaf growth, dominated by cell division). A large amount of the energy made available during ‘downhill’ transport of electrons in the inner mitochondrial membrane (with O2 as the terminal acceptor) is lost as heat – at least 60–65%, but more if some fraction of O2 is reduced by the alternative oxidase (AOX). The remainder is transiently stored in ATP, but little of the energy stored in ATP is retained in anabolic products (MacFarlane et al., 2002). In fact, adenylate pools are small, but exhibit fast turnover. Consequently, heat rates are directly proportional to inline image (Hopkin 1991). Indeed q = inline image. The term inline image varies only slightly with the class of compound used as substrate for respiration, and is usually substituted by Thornton’s ‘constant’ or the oxycaloric equivalent (−455 ± 15 J mmol−1).

After determination of inline image and inline image at each measurement temperature, the temperature response of respiration can be described by three parameters (compare Kruse & Adams, 2008a). The temperature response of both respiratory O2 reduction and CO2 release is given by:

  • image(Eqn 2)

where inline image, RREF is the respiration rate (either inline image or inline image) at the (low) reference temperature (TREF; 283 K or 10°C in the present study), Eo is the overall activation energy of inline image or inline image (given in kJ mol−1), and δ (given in 103 K2) describes the deviation from strictly linear Arrhenius kinetics. The three parameters RREF, Eo and δ are determined from the linearization of Eqn 1, as shown in Fig. 1(a). This figure shows that Eo determines the slope (or steepness) of the exponential function, whereas δ determines the dynamic response of respiration to the measurement temperature, that is, the deviation from strictly linear Arrhenius kinetics. If we substitute the TTERM by x (compare with the equation in Fig. 1), the linearization of Eqn 2 becomes:

image

Figure 1. Determination of respiration and growth variables. The temperature response of respiration was measured in 5°C steps ranging from 10 to 25°C. To obtain the respiration variables RREF (respiration rate at the reference temperature, 10°C), Eo (overall activation energy, kJ mol−1) and δ (dynamic response, 103 K2) at the reference temperature, the temperature response of respiration was linearized as shown in (a) (also see Eqn 3). The respiratory capacity (CapR) was calculated as the definite integral of the linearized temperature response (see Eqn 4), ranging from x = 0 to x = b. (b) Exemplary temperature responses of inline image and inline image. (c) Calculation of enthalpic growth capacity. The instantaneous growth capacity was estimated from the definite integral of enthalpic growth rates, ranging from Tmin to Tmax (c). At these temperatures, the difference between inline image and inline image becomes zero, assuming that inline image (compare b). Maximum growth rates (RSGΔHB) and enthalpy conversion efficiency (ηH) at optimal temperature (Topt) were determined as shown in (c).

Download figure to PowerPoint

  • image(Eqn 3)

This second-order polynomial function can be differentiated or integrated. Respiratory capacity, that is, the sum of active enzymes involved in either oxygen reduction or decarboxylation reactions, is defined as the respiration rate integrated over a range of measurement temperatures. As respiration rates (R) are proportional to logeR, such a dimensionless parameter to describe the respiratory capacity can also be obtained from a definitive integral of the function given by Eqn 1:

  • image(Eqn 4)

where CapR is the respiratory capacity, Eo, δ and the gas constant are dimensionless, and b (dimensionless, the upper integration limit) is defined by TREF and the highest measurement temperature.

In the present study inline image.

The integral thus represents the temperature interval ranging from 10 to 25°C. Calculated capacities ranged from 0.2 to 0.6. Note that, in addition to the three respiration parameters RREF, Eo and δ, these capacities also depend on the upper integration limit, b.

Enthalpic growth rates and the instantaneous growth capacity: measurement and meaning

It is possible to obtain an estimate of growth, following the enthalpy balance approach outlined by Hansen et al. (1994):

  • image(Eqn 5)

where RSG is the specific rate of conversion of substrate carbon to biomass carbon (nmol g−1 s−1) and ΔHB is the total enthalpy change for incorporation of one mole of substrate carbon (carbohydrates) into structural carbon (‘biomass’), including enthalpy effects from all elements (kJ nmol−1). Note that ΔHB is identical to the difference between the heats of combustion of anabolic products and that of substrate, depending on their respective degree of reduction. For example, the enthalpy change for combustion of one mole of substrate carbon to CO2 (inline image) is highly dependent on the degree of reduction of substrate. It can be as low as −350 kJ Cmol−1 for some highly oxidized compounds (i.e. organic acids), but may exceed −700 kJ Cmol−1 for some lipids. By contrast, the degree of reduction of substrate carbon has little effect on inline image (the oxycaloric equivalent is near-constant for aerobic respiration; Kemp, 2000).

Intuitively more comprehensible, the enthalpy balance model can be expressed in terms of CO2-release and O2-uptake rates, and the degree of reduction of substrate (γS) and anabolic products (γp), taking account of Thornton’s rule (inline image) (compare with Eqn 5):

  • image(Eqn 6)

Eqn 6 implies that four electrons are associated with each O2 consumed in respiration. The flow of reductant towards O2 depends upon rates of substrate oxidation and the degree of reduction of substrate (the equivalent of electrons that are available per unit C) and the flow of reductant towards anabolic products (thereby determining γp of anabolic products). In the current study we assumed that carbohydrates were the only substrates for respiration so that inline image and inline image are of the same magnitude (i.e. −469 kJ mol−1). This assumption is probably valid for the very young, rapidly expanding tissue of the present study that depends almost entirely on carbohydrate supplied through the phloem. It is more questionable for older, senescing tissue and plants coping with environmental stresses (i.e. heat stress; Tcherkez et al., 2003). While cold-grown plants show accumulation of carbohydrates (Usadel et al., 2008), this does not invalidate our assumption. It is important to note that enthalpic rates of growth (RSGΔHB) can be determined more accurately given precise information about the substrate(s) of respiration. Clearly, enthalpic rates of growth (and enthalpy conversion efficiencies; see Eqns 8 and 9) calculated in the present study ought be properly viewed as estimates.

Measuring CO2 release and O2 uptake as a means of noninvasive characterization of biosynthetic processes is not new (see Willms et al., 1999; Cen et al., 2001), but has been obstructed by difficulties in measuring O2 uptake against high background O2 concentrations (c. 21%). Difficulties are particularly apparent at low measurement temperatures, when metabolic rates are slow. These problems are circumvented by micro-calorimetry that is capable of quantifying rates of heat loss (as a surrogate for O2 uptake) even at low measurement temperatures, so that simultaneous determination of temperature responses, of both CO2 release and O2 uptake, are readily achieved with calorimetry.

From temperature responses of inline image and inline image, we determined minimum and maximum temperature at which RSGΔHB becomes zero. Calculating the difference between the two temperature response functions, and integration of enthalpic growth rates from Tmin to Tmax, yields the instantaneous ‘enthalpic growth capacity’ (CapG):

  • image(Eqn 7)

T min, Tmax (and Topt) and the integral were computed using Origin 6.1 (Microsoft, Redmond, Washington, USA).

Note also that, in addition to inherent growth capacity, instantaneous growth capacity partly depends on respiratory acclimation to environmental conditions (e.g. at four different sites). We determined the enthalpic growth rate at optimum temperature (compare Fig. 1c), and enthalpy conversion efficiency at Topt:

  • image(Eqn 8)

Combining Eqn 5 with Eqn 8, the enthalpy conversion efficiency can also be expressed as follows (compare Kruse et al., 2008):

  • image(Eqn 9)

If only carbohydrates are respired, enthalpic efficiency (ηH) is proportional to the respiratory quotient CO2/O2, and a respiratory quotient of 1 is equivalent to ηH = 0. Assuming that carbohydrates are the principal substrate for respiration, the enthalpy conversion efficiency mainly reflects the degree of reduction of anabolic products and not that of substrate carbon (Willms et al., 1999). A further note of caution needs to be considered. It is not strictly true that respiratory quotients solely depend on the substrate class used for respiration. While this might be true during remobilization of reserve compounds (i.e. proteins and oils that are stored in seeds or tubers), respiratory quotients are not constant but change with temperature because of the differing temperature sensitivities of CO2-producing reactions (inline image and inline image) and O2-consuming reactions (inline image and δ(inline image)). In the present study, these variables are determined from short-term experiments, where sudden switches between substrate classes (with significantly different γs) are unlikely. Hence, we determined ηH as a proxy for CO2/O2, only at Topt.

Statistical analysis

Statistical analysis was performed with the results of 48 independent replicates for sites 1, 2 and 3 (Mt Beauty, Bogong and Howman’s Gap), but only 33 replicates for site 4 (Falls Creek) where between two and four independent replicates (individuals) per species were available. As a consequence, there were 58 independent replicates for lowland species (group 1: E. botryoides, E. tereticornis, E. radiata and E. sideroxylon), 87 independent replicates for montane species (group 2: E. cypellocarpa, E. dives, E. fastigata, E. globulus, E. nitens and E. sieberi) and 32 independent replicates for high-altitude species (group 3: E. delegatensis and E. pauciflora). For the respiratory variables RREF, Eo, δ and CapR, and growth variables CapG, RSGΔHB and ηH at 25°C, the main effects (site and group) and their interaction were assessed by ANOVA (Statistica, version 6.0; StatSoft, Inc, Tulsa, OK, USA). Categorical predictor variables were coded according to sigma-restricted parameterization, and sums of squares were calculated according to Type VI. Statistical significance of main effects and their interactions is indicated by: *, P < 0.05; **, P < 0.01; ***, P < 0.001.

A general linear model, as a mixture of continuous and categorical predictor variables, was employed to explain variation in dependent variables inline image, inline image and δ(inline image). A general linear model facilitates evaluation of the strength of different correlations, and allows estimation of the sensitivity of these correlations to site and group effects. Further, we used ANCOVA to assess interdependence between respiratory capacities and individual respiration parameters that determine temperature responses of respiration.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Acclimation of respiration variables to growth temperature, and species-specific differences between respiration variables (adaptation)

In the present study we interpret respiratory acclimation to growth temperature as the average response of all species (low-altitude, mid-altitude and highland species) growing at differing sites (which are characterized by different growth temperatures). We also interpret the species (group) effects (low-, mid- and high-altitude species averaged across the four sites) as adaptations to their thermal origin, as there were no interactive effects between site and species (Table 2).

Rates of O2 reduction at a low reference temperature (10°C) and capacities for O2 reduction were lower in high- than low-altitude species. These parameters were not consistently affected by site (Table 1).

Table 1.   Respiration variables as affected by site and Eucalyptus species (group)
 Site effectSpecies (group) effect
Mt BeautyBogongHowman’s GapFalls CreekLow-altitudeMid-altitudeHigh-altitude
  1. R REF (nmol g−1 s−1) is the respiration rate at the reference temperature (10°C). Eo (kJ mol−1) is the overall activation energy of either O2 reduction or CO2 release and determines the ‘slope’ of the exponential function. The parameter δ (103 K2) describes the dynamic response of respiration to the measurement temperature, that is the deviation from strictly linear Arrhenius kinetics.

  2. 1Note that the respiratory capacity was calculated as a definite integral, as described in Fig. 1(a), and is a dimensionless measure of capacity. Values shown are site and species data ± SD (shown in parentheses). A summary of statistical results is given in Table 2.

R10(O2) (nmolg−1 s−1)3.22 (0.81)2.32 (1.1)3.25 (1.17)2.42 (0.79)3.07 (1.18)2.91 (0.95)2.07 (0.85)
inline image (kJ mol−1)67.1 (14.2)76.8 (42.3)86.0 (28.8)100 (26.9)83.4 (32)79.3 (30.2)85.9 (36.5)
δ(inline image) (kK2)− 9.5 (8.7)− 14.8 (18.9)− 20.7 (13.2)− 23.3 (15.1)− 17.3 (14.5)− 15.8 (15.2)− 18.9 (17.1)
inline image (without dimension)10.31 (0.04)0.25 (0.06)0.32 (0.05)0.30 (0.04)0.31 (0.05)0.30 (0.05)0.24 (0.05)
R10(CO2) (nmol g−1 s−1)6.8 (1.8)7.4 (3.2)9.9 (3.1)10.5 (3.1)9.8 (3.1)8.6 (3.3)6.4 (2.2)
inline image (kJ mol−1)71.1 (23.2)63.8 (28.5)72.3 (31.3)71.7 (38.0)63.6 (27.8)72.1 (31.2)73 (28.1)
inline image (kK2)− 29.8 (14.0)− 37.4 (17.2)− 44.2 (18.4)− 47.1 (23.1)− 37.0 (19.1)− 38.8 (18.4)− 41.6 (18.9)
inline image (without dimension)10.41 (0.05)0.39 (0.06)0.45 (0.05)0.45 (0.05)0.45 (0.05)0.43 (0.05)0.37 (0.05)

However, there were significant effects of site on inline image and δ(inline image), but these parameters were similar across the three species groups (Tables 1,2). On average, inline image increased from 67.1 kJ mol−1 at Mt Beauty to 76.8 kJ mol−1 at Bogong. It further increased to 86 kJ mol−1 at Howman’s Gap and 100 kJ mol−1 at Falls Creek. Concomitantly, δ(inline image) decreased with increasing altitude. These two parameters were closely correlated (Fig. 2c). Approximately 86% of the variation in δ(inline image) could be explained by inline image.

Table 2.   Results of two-way ANOVA, showing significance of main effects (site and Eucalyptus species) for RREF, Eo, δ and Cap(R), both for inline image and inline image, and for the enthalpic growth (and instantaneous growth capacity) and efficiency at the optimum temperature
 SiteSpecies (group)Site × species
FPFPFP
  1. Significance: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant. Dependent variables were determined as described in Fig. 1(a).

  2. R REF, respiration at the reference temperature, i.e. 10°C; Eo, overall activation energy; δ, the dynamic response of respiration to measurement temperatures; Cap(R), the respiratory capacity.

RREF (CO2) (nmol g−1 s−1)15.2***18.1***1.6ns
inline image (kJ mol−1)1.6ns1.7ns1.0ns
inline image (103 × K2)5.5**0.2ns0.5ns
inline image14.7***21.2***2.4*
RREF (O2) (nmol g−1 s−1)12.3***12.2***0.8ns
Eo (inline image) (kJ mol−1)8.7***0.7ns0.6ns
δ (inline image) (103 × K2)7.1***0.6ns0.6ns
inline image19.7***23.3***1.2ns
Enthalpic growth at Topt17.5***12.5***1.5ns
Enthalpic efficiency at Topt23.8***1.6ns1.8ns
CapG15.7***11.5***1.1ns
image

Figure 2. Correlation between different variables that describe temperature sensitivities and capacities of inline image and inline image. (a) Correlation between capacities for inline image and inline image. (b) Correlation between overall activation energy and the dynamic response of inline image. (c) Correlation between overall activation energy and the dynamic response of inline image. While the correlation between inline image and δ(inline image) explained most of the variability observed for the two parameters (c), the relation between inline image and inline image, and between inline image and inline image also depended on site conditions (a, b; also compare Fig. 3).

Download figure to PowerPoint

Rates of CO2 respiration at a low reference temperature (10°C) increased from 6.8 nmol g−1 s−1 at Mt Beauty to 7.4 nmol g−1 s−1 at Bogong; and further to 9.9 nmol g−1 s−1 at Howman’s Gap and 10.5 nmol g−1 s−1 at Falls Creek. Conversely, high-altitude species exhibited lower rates of basal CO2 respiration than low-altitude species (Table 1). The temperature sensitivity of CO2 respiration (inline image) was not affected by acclimation and adaptation responses.

There was, however, a significant site effect (but no species effect) on the dynamic response of CO2 respiration. On average, inline image amounted to −29.8 kK2 at Mt Beauty and −37.4 kK2 at Bogong. It further decreased to −44.2 kK2 at Howman’s gap and −47.1 kK2 at Falls Creek (Table 1). Similar to inline image and δ(inline image), we found a significant correlation between inline image and inline image. This correlation was a little weaker (Fig. 2). To separate influences on inline image, we subjected the data to a general linear model (GLM), with: inline image as a continuous predictor variable, and site and species (group) as categorical predictor variables. This model (results not shown) explained 79% of observed variation in inline image, and confirmed a significant site effect (compare Fig. 3e).There were no additional effects on δ(inline image), other than those related to inline image (Fig. 3g–i). As a consequence of additional influences on inline image, respiratory capacity (inline image) was affected less by site than RREF(CO2), whereas the species effect on these parameters was similar (Table 1). Not surprisingly, inline image was significantly correlated with inline image (Fig. 2a). However, GLM (explaining 75% of the observed variation; not shown) revealed additional and opposing site and species effects on inline image (Fig. 3a–c).

image

Figure 3. Sensitivity of three respiration variables inline image, inline image and δ(inline image) to different factors. To separate the effects of continuous predictor variables (also compare Fig. 2) from those of categorical predictor variables (site and species), the data were analysed using a general linear model (GLM). (a) Dependency of inline image on inline image, and additional influences of site (b) and species (c). (d) Dependency of the dynamic response of inline image on the overall activation energy of inline image, and additional influences of site (e) and species (f). (g) Dependency of the dynamic response of inline image on the overall activation energy of inline image, and additional influences of site (h) and species (i). The values shown for dependent variables inline image, inline image, δ(inline image) are predicted responses at each level of each factor (continuous for inline image, inline image and inline image; categorical for site and species effects), holding all other factors (including the block factor) constant. Red dashed lines, predicted values for the average estimate, if all other factors are held constant. The bars indicate the 95% confidence interval.

Download figure to PowerPoint

Enthalpic growth rates and the instantaneous growth capacity

Instantaneous enthalpic growth capacity was calculated from the difference between inline image and inline image (Fig. 1b), and integrated between Tmin and Tmax (Fig. 1c). While there were slight differences among sites in minimum temperatures for positive enthalpic growth (Fig. 4a), optimum temperatures for instantaneous rates of growth were little affected by site conditions. This result reflects two contrasting trends: growth at high altitudes increased respiratory capacity inline image, but reduced the dynamic response of respiration (inline image). As a result, optimum temperatures for instantaneous rates of growth were similar across sites, and growth rates at this temperature were generally much greater at higher altitudes (Figs 4, 5). This finding tells us little about biomass growth, because average midday growth temperatures were 6–13°C lower at Falls Creek than at Mt Beauty over the duration of the experiment.

image

Figure 4. Instantaneous enthalpic growth capacities as affected by site conditions and species (group). Growth capacity was determined from the integral of enthapic growth rates between Tmin and Tmax, at which RSGΔHB was zero (compare Fig. 1b,c). Growth at high elevations increased instantaneous growth capacity (a), but inherent growth capacities were lower for species that originated from high altitudes (b). In (a) the site effect is singled out, that is, it combines all species grown at a respective site. Values shown in (b) are averages of site and species data ± SD: circles, species group effect; squares, site effect.

Download figure to PowerPoint

image

Figure 5. Instantaneous rates of enthalpic growth (grey bars) and enthalpy conversion efficiency (red squares) at the optimum temperature (Topt). The optimum temperature was determined as described in Fig. 1(b,c). Each bar and data point represents the average ± SD of four independent replicates (two to four independent replicates at Falls Creek). The order of bars for each species is, from left to right: Mt Beauty, Bogong, Howman’s gap, Falls Creek. To assist comparisons among species, the red horizontal line indicates overall average enthalpic efficiencies. Low-altitude species: Eucalyptus botryoides, Eucalyptus tereticornis, Eucalyptus radiata and Eucalyptus sideroxylon. Mid-altitude species: Eucalyptus cypellocarpa, Eucalyptus dives, Eucalyptus fastigata, Eucalyptus globulus, Eucalyptus nitens and Eucalyptus sieberi. High-altitude species: Eucalyptus delegatensis and Eucalyptus pauciflora.

Download figure to PowerPoint

Our measure of instantaneous growth capacity aggregates all respiration variables in one parameter, alleviating the need to test individual parameter hypotheses. From Fig. 4(b) we conclude that the increase in growth capacity with altitude (acclimation) was much less pronounced in species originating from highland habitats (adaptation). As a consequence, instantaneous rates of growth at optimum temperature were lower for ‘high-altitude’ than ‘low-altitude’ species, across all sites (Fig. 5). RSGΔHB averaged 4.2 μW mg−1 across all sites for ‘low-altitude’ species. Within this group, enthalpic rates of growth of E. tereticornis (3.2 μW mg−1) were somewhat lower than the group average (Fig. 5). Similarly, while average enthalpic growth was 3.7 μW mg−1 for ‘mid-altitude’ species, there were significant departures (E. dives, 3.1 μW mg−1; E. globulus 4.6 μW mg−1; compare Fig. 5). These modest outliers do, however, reinforce our earlier comment that the classification of species used here is necessarily arbitrary and that the range of some study species is not easily captured. For example, the geographic distribution of E. tereticornis is such that there is a basis for arguing that it be classed as ‘mid-altitude’ (‘low-altitude’ in this study). Likewise, E. globulus grows naturally close to sea level on some sites and at elevations of up to 800 m at others, while E. dives is perhaps more common at elevations > 500 m than at those < 500 m. Enthalpic growth rates for ‘high-altitude’ species averaged 2.7 μW mg−1 (Fig. 5).

It is possible to assess the influence of individual respiration parameters on respiratory capacities (which, in turn, determine the growth capacity; Fig. 6), which helps to disentangle acclimation and adaptation effects on individual parameters. As the capacities for CO2 respiration and O2 reduction were strongly positively correlated, their effects on the growth capacity were analysed by ANCOVA. This analysis showed that CapG was more strongly affected by inline image (F = 285***, Fig. 6) than by CapO2 (F = 45***). In this analysis, the additional effect of inline image (separated from that related to inline image) on CapG was negative. The capacity for O2 reduction was strongly correlated with R10(O2), but not with inline image and δ(inline image). Nonetheless, as these three parameters were strongly interrelated, ANCOVA revealed that inline image had an additional effect on inline image that was not related to concomitant changes of R10(O2) in the opposite direction. Similar to inline image, inline image was also strongly related to R10(CO2). In the case of inline image, however, inline image and inline image had a more substantial influence.

image

Figure 6. Correlations among respiration variables. From the bottom up, the capacities of CO2 respiration inline image and O2 reduction inline image are dependent on respective respiration variables (R10, Eo and δ) that were determined from instantaneous temperature responses of respiration (see Eqn 1). As many of these parameters co-vary, we estimated their effects on respiratory capacities by ANCOVA. Growth capacity (CapG) can be approximated by the difference between these capacities, and thereby encompasses variable effects on different respiration parameters (compare Fig. 4). Effects of strongly correlated capacities of O2 reduction and CO2 release on the growth capacity were also estimated by ANCOVA. The hierarchical organization of the figure provides a perspective on site and group effects on individual respiration parameters (compare Tables 1,2). r, regression coefficient between a pair of variables; ns, not significant; ***, significant at P ≤ 0.001.

Download figure to PowerPoint

Interestingly, inline image was correlated with the growth capacity, and high activation energies (at low reference temperature) were associated with large growth capacities (r = 0.4, P ≤ 0.001; not shown in Fig. 6). At a lower level, inline image was positively related to inline image (F = 27.8*** for ANCOVA; not shown in Fig. 6), and to inline image (F = 15.7***, multiple r = 0.51), but not to R10(CO2) (F = 1.9ns).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Respiratory acclimation to cool temperatures

The present study sought to characterize temperature acclimation of respiratory processes, as opposed to species-specific adaptation to cold temperatures. First, we discuss respiratory acclimation in phenomenological terms, as expressed in alterations of the instantaneous temperature response of respiration. Secondly, we assess the hypothesis that, in young Eucalyptus, respiratory acclimation is a consequence of balanced growth at contrasting temperatures. Finally, we discuss physiological mechanisms that may allow flexible regulation of growth, and if these mechanisms differ between species adapted to contrasting growth temperatures.

In simple terms, acclimation of respiration is reflected in similar respiration rates for plants grown at contrasting temperatures, if measured at the respective growth temperature. In general, the degree of respiratory acclimation seems to differ widely among plant species (Larigauderie & Körner, 1995; Loveys et al., 2002, 2003). Most studies refer to rates of respiratory CO2 release (inline image) in order to characterize respiratory acclimation. Two kinds of acclimation are frequently proposed (sensuAtkin & Tjoelker, 2003): ‘Type II’ as defined by shifts in the intercept of the instantaneous temperature response curve, and ‘Type I’ as defined by shifts in the slope of this exponential function. Type II acclimation is believed to reflect changes in respiratory capacity, that is, mitochondrial density or the amount of active enzymes involved in inline image. Type I acclimation is equated with changes in activation energy of overall CO2 respiration (i.e. inline image), indicating variation in carbon flux through different decarboxylating pathways (Kruse et al., 2011). We might thus presume that an increase in respiration at low measurement temperatures (i.e. R10CO2) from 6.8 nmol g−1 s−1 at Mt Beauty to 10.5 nmol−1 s−1 at Falls Creek was driven by an enhanced decarboxylation capacity. However, this conclusion is not certain, as R10(CO2) and inline image are correlated (also see Xu & Griffin, 2006, Searle et al., 2010), such that shifts in R10(CO2) are at least partly dependent on relative contributions of distinct decarboxylating pathways.

We defined ‘respiratory capacity’ as a definite integral of the instantaneous temperature response (Kruse et al., 2011; see the Materials and Methods section). This approach incorporates the effect of variable activation energy on respiration over measurement temperatures and yields a dimensionless measure of respiratory capacity, instead of describing this capacity by respiration measured at a single, low measurement temperature. High activation energy results in low rates of respiration at low temperatures, but high rates of respiration at higher temperatures. We argue that changes in respiratory capacity can be better judged by integrating respiration rates over a broad range of measurement temperatures.

The site-dependent increase of inline image was less pronounced than that of R10(CO2), and we conclude that acclimation of respiration to cold temperatures is only partly a result of increased amounts of respiratory enzymes – it is also a result of flexible control of carbon fluxes through differing pathways. This flexibility is not apparent in site-dependent alterations of inline image, but is revealed in changes in inline image– this is discussed in the later section ‘Evidence for acclimatory shifts between decarboxylating pathways?’. However, adaptation to cool environments was mostly associated with reduced respiratory capacity sensu stricto, as high-altitude species clearly exhibited lower inline image than low-altitude species, albeit there were no species (group) effects on inline image and inline image.

To what extent were these differences related to relative rates of growth of differing species; and could homeostasis of growth at contrasting temperatures help to explain respiratory acclimation?

Enthalpic rates of growth and the instantaneous growth capacity – understanding respiratory acclimation and species-specific differences of acclimation responses

It is increasingly apparent that we require information about both CO2 release (mainly resulting from substrate oxidation) and O2 reduction if we are to understand respiratory plasticity in response to environmental (and developmental) cues.

The enthalpy balance model developed by Hansen et al. (1994) and Criddle & Hansen (1999) provides a measure of instantaneous rates of growth from rates of CO2 respiration and O2 reduction (or, alternatively, rates of heat loss). This measure is frequently termed ‘enthalpic rate of growth’ and describes the increase in enthalpy (roughly equivalent to ‘energy’) stored in the chemical bonds of anabolic products (MacFarlane et al., 2002; Kruse & Adams, 2008b; see the Materials and Methods section).

In this study we have distinguished between energy made available by mitochondrial respiration (ATP) and that stored in anabolic products and driven by processes outside plant mitochondria. As applied in this study, ‘enthalpic growth’ refers to young leaves. However, their respiratory metabolism may be required to meet demands beyond the cell and organ level.

At the cellular level, mitochondrial respiration provides the ATP needed to: maintain existing cell structures and ion gradients across membranes; fuel long-distance transport of nutrients and assimilates; and drive anabolic reactions that are thermodynamically unfavourable (Kruse et al., 2011). The speed of ATP turnover is proportional to the capacity for O2 reduction. Nevertheless, a large capacity to reduce O2 cannot be equated with rapid growth of the studied tissue because the fraction of ATP needed for growth, as opposed to transport and maintenance processes, is highly variable. For example, in the present study the site effect on inline image was not consistent, whereas we observed a clear increase of growth capacity CapG with altitude. Growth capacity can be approximated by the difference between inline image and inline image. These capacities are highly correlated, but not identical. Not all reducing power produced during oxidation of substrates is linked to O2 reduction; some is needed for reductive anabolism (e.g. reduction and assimilation of inline image) or synthesis of membrane lipids. This fraction is what we previously described as ‘enthalpic rate of growth’. Clearly, though, enthalpic rates of growth (in units of energy per unit time) are not the same as the rate of growth on a per unit mass basis, as these relations are confounded by oxidation states of produced compounds (Amthor, 2010). Plant carbon is typically much more reduced than the substrate carbon (typically carbohydrates) from which it is formed (Lambers et al., 2008). Accordingly, the capacity for substrate oxidation is generally greater than that for O2 reduction.

Both acclimation and adaptation influenced the relation between inline image and inline image of young Eucalyptus foliage, and produced opposing effects on enthalpic growth capacity. Growth capacity was greater at higher than lower altitudes and enthalpic rates of growth at high altitudes were comparatively fast at low measurement temperatures, so that actual growth rates of the youngest leafs were similar (but not exactly the same) across sites. Nonetheless, acclimation of growth and respiration hardly affected the optimum temperatures for growth. This finding stands in contrast to the general acclimation response of photosynthesis, in which optimum temperature shifts in accordance with alterations of growth temperature, while photosynthetic rates at each optimum temperature are hardly affected (Gunderson et al., 2010).

In summary, increased instantaneous growth capacity at low temperatures seems to be a means by which young eucalypts maintain similar rates of growth of young foliage under cold and warm conditions. That said, there were clear adaptive differences between low- and high-altitude species in their growth capacities. Acclimation and adaptation produced contrasting effects on the temperature responses of CO2 release and O2 reduction. These contrasting effects are the focus of the next section.

Acclimation and adaptation effects on instantaneous temperature responses of respiration

Briefly revisited mechanistic interpretation of the temperature response of O2 reduction  The integrative measure ‘enthalpic growth capacity’ is useful for pinpointing principal acclimation and adaptive effects on foliage growth. The present study confirmed our hypothesis that respiratory acclimation is closely related to growth regulation.

Fig. 6 provides a useful framework for discussion of site- and species-specific effects on gas exchange parameters, and of mechanistic interpretations of observed phenomena. While site had no consistent effect on inline image, we found a clear increase of inline image with altitude. Experiments with specific inhibitors of cytochrome oxidase (COX; inhibited by CN) and alternative oxidase (AOX; inhibited by salicylhydroxamic acid (SHAM)) in the 1980s and 1990s suggested different temperature sensitivities of the two pathways (for references, see Kruse et al., 2011). More recently, these results were called into question by studies employing the 18O-fractionation technique, but the issue of temperature sensitivity of COX vs AOX could not be resolved unequivocally (Armstrong et al., 2008; MacFarlane et al., 2009). We recently argued that general statements on putative temperature sensitivities of the two pathways can be misleading (Kruse et al., 2011), as these temperature sensitivities appear to change with measurement temperature. More explicitly, COX could be more temperature-sensitive than AOX at low measurement temperatures, with the reverse being true at high measurement temperatures (and at moderate temperatures, i.e. 18–26°C, sensitivities might be rather similar). The dynamic response of O2 reduction to a broader temperature range seems a more useful way of drawing conclusions about electron partitioning between the two pathways. We previously demonstrated that inhibition of one or other pathway resulted in strong (and smooth) dynamic responses (Kruse & Adams, 2008a,b), which was negative in the case of AOX inhibition and positive for COX-inhibited respiration (i.e. when oxygen was mainly reduced by AOX). With COX inhibited, the proportional decrease in overall O2 reduction was greater than that of substrate oxidation (CO2 release) (Kruse et al., 2008). Such conditions significantly increase the redox poise of the ubiqinone pool in the inner mitochondrial membrane, and activate AOX. In a recent review we showed how this effect could be amplified at high temperature (Kruse et al., 2011), whereby there is a majority of ‘transition state’ ubiquinole molecules able to overcome the activation barrier to oxygen reduction, resulting in a strong, positive dynamic response of O2 reduction to measurement temperature. Positive dynamic responses can also be observed for noninhibited, normal respiration: under in vivo conditions, we find a large spectrum of dynamic responses, ranging from –80 to +20 kK2, suggesting variable contributions of AOX and COX to overall O2 reduction in response to environmental and developmental factors (compare Fig. 2; also see Kruse & Adams, 2008a,b). Further, COX and AOX activities seem mutually dependent: the close correlation between inline image and δ(inline image) suggests continuous coordination of mitochondrial electron flow through cytochrome and the alternative pathway, in order to optimize upstream carbohydrate metabolism and downstream electron flow according to the demand for ATP, TCA-cycle intermediates and anabolic reducing power (see in-depth discussion by Kruse et al., 2011). For example, Vidal et al. (2007) convincingly demonstrated that electron partitioning between cytochrome and the alternative pathway is finely adjusted in two different tobacco (Nicotiana tabacum) genotypes (with one genotype constitutively overexpressing AOX1), so that ATP production in response to a cell death elicitor was the same in both genotypes.

Acclimation of temperature sensitivity of O2 reduction to measurement temperature  In the present study, we observed that plant growth at high altitudes increased the (negative) dynamic response that was associated with large temperature sensitivity of O2 reduction at low reference temperature (inline image). By contrast, there were no adaptive differences between Eucalyptus species in terms of the temperature sensitivity of O2 reduction. This acclimation response is different from adaptation of inline image in some high-elevation and high-latitude woody perennials found by Criddle et al. (1994). The acclimation response observed in the present study suggests that COX activity increased with altitude.

This result is a little surprising because numerous authors have associated temperature stress (chilling) with increased activity of AOX and protection against the formation of reactive oxygen species (ROS; Rychter et al., 1988; Ribas-Carbo et al., 2000). Increased activity of AOX may be a transient response (e.g. in the first 1–2 d after cold treatment; Armstrong et al., 2008). Under field conditions, the effect of growth temperature on alternative oxidase is more variable and cannot be readily distinguished from other environmental, and especially plant developmental, effects (Searle et al., 2011). Hence, while AOX is involved in mitigation of oxidative stress and mitochondrial signalling (Van Aken et al., 2009; Vanlerberghe et al., 2009), it could play a more central role in the regulation of primary plant metabolism (Moore et al., 2002; Gomez-Casanovas et al., 2007). As the temperature sensitivity of oxidative phosphorylation in plant mitochondria probably differs from the multitude of energy-demanding processes in the cell, flexible control of mitochondrial electron partitioning is essential for proper coupling of ATP production and consumption (Hansen et al., 2002). In addition, ATP production must be coordinated with cellular requirements of respiratory intermediates, and the demand for anabolic reducing power (Kruse et al., 2011).

The hypothesis that electron partitioning shifts towards the cytochrome path under (sustained) cool conditions at higher altitudes was confirmed by the study of Armstrong et al. (2006b). They observed that increased COX activity in cold-developed Arabidopsis was reflected in greater temperature sensitivity of O2 uptake. However, this response to development in cold conditions is also variable (Armstrong et al., 2006a). In an exhaustive study of responses of transcripts, enzyme activities and metabolites to cold treatment, Usadel et al. (2008) found a pronounced up-regulation of COX and F1-ATPase expression in Arabidopsis. Such changes facilitate maintenance processes by preventing shortages of ATP that might otherwise be induced by cold temperatures. Perhaps more importantly, growth of young foliage seems mostly homeostatic at variable temperatures (Kurimoto et al., 2004). Logically, it follows that respiration can acclimate quickly after temperatures change (e.g. Bolstad et al., 2003).

Linkage between mitochondrial respiration (catabolism) and anabolic processes, as affected by acclimation  From the standpoint of O2 reduction, acclimation may not need significant alteration of oxidation capacity (i.e. mitochondrial density), as it seems dependent on swift regulation of mitochondrial electron flow and ATP production. Statistically, inline image had a clear influence on growth capacity (CapG). Large inline image– indicating comparatively high COX activity – was associated with large CapG. This finding accords with that of Florez-Sarasa et al. (2007), who demonstrated enhanced COX activity in rapidly growing Arabidopsis, but contradicts the results obtained by Kruse & Adams (2008a,b) for Pinus radiata. In the latter study, rapid growth of young needles appeared to be associated with a large capacity of O2 reduction and AOX activity during the spring. Under optimal growth conditions, enhanced rates of O2 reduction through enhanced AOX activity can help to maximize the supply of TCA-cycle intermediates and rates of growth before environmental conditions deteriorate, albeit at a cost to respiratory efficiency (i.e. the P : O ratio). For example, Searle & Turnbull (2011) showed that, for Populus canadensis, electron flow through the alternative pathway was faster during the growing season than in the autumn. The slowing of respiration towards autumn was accompanied by an over-proportional reduction in concentrations of AOX protein, when compared with the simultaneous reduction in COX protein (Searle & Turnbull, 2011). In the present study, growth of Eucalyptus at high altitudes increased instantaneous growth capacity, but actual rates of growth at the relevant cool growth temperatures were still somewhat slower than at low altitudes. Under these conditions, respiratory efficiency was increased. Taken together, the findings suggest that the contribution of COX and AOX to mitochondrial O2 reduction is highly variable. This is partly a consequence of matching differences in ATP production and consumption under fluctuating environmental conditions, especially fluctuating temperatures. Further, mitochondrial respiration must be coordinated with outer-mitochondrial demands and processes that drive ‘enthalpic growth’. Active growth requires a sufficient supply of metabolic intermediates, including TCA-cycle intermediates. Demand for intermediates may sometimes even exceed that of ATP, so that AOX activity keeps the TCA cycle operative (Vanlerberghe & McIntosh, 1997). In addition, anabolic reducing power (mainly NADPH) is mostly provided by operation of the oxidative pentose phosphate (OPP) pathway (although other outer-mitochondrial pathways may assist, e.g. cytosolic isocitrate dehydrogenase). The OPP pathway competes with glycolysis for available substrate. While most substrate carbon is ultimately oxidized in the mitochondrial matrix and coupled to mitochondrial oxygen reduction, a variable fraction of (outer-mitochondrial) reducing power is needed for anabolism. These processes also require some coordination, possibly mediated by phosphoenolpyruvate carboxylase (see Supporting Information S2 in Kruse et al., 2011).

Evidence for acclimatory shifts between decarboxylating pathways?  In the present study, acclimation not only affected inline image and thereby enthalpic growth capacity, but also inline image. More precisely, site conditions had a significant impact on the correlation between inline image and inline image (see Fig. 2b and compare with Fig. 2c). As a consequence, the intercept of this correlation with the x-axis varied with growth conditions, whereas it was stable for the correlation between inline image and δ(inline image) (also see Kruse & Adams, 2008a,b). These findings mirror the greater complexity of carbohydrate decarboxylation when compared with oxygen reduction. While switching between (only) two electron transport pathways in the inner mitochondrial membrane alters inline image which is closely linked to changes of δ(inline image), regulation of substrate oxidation is more complex and involves a network of interconnected pathways. Different growth conditions can alter TCA-cycle turnover in relation to outer-mitochondrial cycles (the OPP pathway) and pathways, blurring the correlation between inline image and inline image. It is also worth noting that inline image was related to inline image and inline image, indicating that oxidative and reductive pathways are co-regulated to optimize the production of ATP, anabolic intermediates and reducing power. Likewise, Searle et al. (2011) found that the AOX : COX protein ratio had some influence over rates of CO2 release, and high AOX protein contents were related to faster rates of CO2 release (also see Searle & Turnbull, 2011).

Conclusions

The enthalpy balance approach outlined in the present study highlights the consequences of respiratory acclimation and adaptation for the energy status of Eucalyptus. We demonstrated that information on both substrate decarboxylation (mostly coupled to oxidation) and O2 reduction is required to accurately describe changes of a plant’s energy status. The instantaneous temperature responses of these processes are a basis for obtaining further insights into the physiological mechanisms that govern energy conservation and expenditure during respiration. Acclimation to cool growth temperatures was only partly driven by an increased capacity of respiration in young foliage. Maintenance of growth rates under cold conditions was also achieved by flexible control of ATP production, accompanied by switches in upstream carbohydrate metabolism. By contrast, inherently slow growth of highland species as compared with lowland species seemed mainly associated with lower respiratory capacity. Criddle et al. (2005) proposed that adaptive differences between plants could be linked to both metabolic rate (i.e. inline image, adapted to average growth temperatures) and metabolic efficiency (i.e. carbon conversion efficiency εc, adapted to temperature variability). While we could confirm adaptation in terms of metabolic capacity (for both inline image and inline image), differences in metabolic efficiency were a consequence of acclimation to growth temperature, which conferred metabolic flexibility to all Eucalyptus species of the present study. The link between respiratory acclimation/adaptation and control of energy conservation may help the development of process-based models of respiration (Buckley & Adams, 2011). In the face of climate change (Hughes, 2003), there is a pressing need to better understand the feedback of global warming on plant growth and respiration (Friedlingstein et al., 2003).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

This work was supported financially by the Australian Research Council (ARC). We are grateful to Prof. Owen Atkin and three anonymous reviewers for many valuable comments that helped improve an earlier draft of this manuscript.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Table S1 Grouping of Eucalyptus spp. according to their natural distribution

Table S2 Thermal variation between study sites

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

FilenameFormatSizeDescription
NPH_4155_sm_TableS1-S2.doc150KSupporting info item