Dynamic photo-inhibition and carbon gain in a C4 and a C3 grass native to high latitudes


David S. Kubien, (current address) Institute of Molecular Biosciences, Massey University, Private Bag 11 222, Palmerston North, New Zealand. Fax: +64 6350 5688; e-mail: dkubien@massey.ac.nz


C4 plants are rare in the cool climates characteristic of high latitudes and altitudes, perhaps because of an enhanced susceptibility to photo-inhibition at low temperatures relative to C3 species. In the present study we tested the hypothesis that low-temperature photo-inhibition is more detrimental to carbon gain in the C4 grass Muhlenbergia glomerata than the C3 species Calamogrostis Canadensis. These grasses occur together in boreal fens in northern Canada. Plants were grown under cool (14/10 °C day/night) and warm (26/22 °C) temperatures before measurement of the light responses of photosynthesis and chlorophyll fluorescence at different temperatures. Cool growth temperatures led to reduced rates of photosynthesis in M. glomerata at all measurement temperatures, but had a smaller effect on the C3 species. In both species the amount of xanthophyll cycle pigments increased when plants were grown at 14/10 °C, and in M. glomerata the xanthophyll epoxidation state was greatly reduced. The detrimental effect of low growth temperature on photosynthesis in M. glomerata was almost completely reversed by a 24-h exposure to the warm-temperature regime. These data indicate that reversible dynamic photo-inhibition is a strategy by which C4 species may tolerate cool climates and overcome the Rubisco limitation that is prevalent at low temperatures in C4 plants.


The high CO2 concentration in the bundle sheath gives C4 plants advantages over C3 species that enable them to dominate warm, high light habitats. By contrast, the C4 pathway is uncommon in the cool climates characteristic of high latitudes or high elevations (Teeri & Stowe 1976; Long 1983; Sage, Wedin & Li 1999). Globally, C4 plants are rare at latitudes and elevations at which the minimum growing season temperatures are less than 8 to 10 °C, and the average temperatures are less than approximately 16 °C (Long 1999; Sage et al. 1999). The reasons for the general lack of C4 success in cool climates have received considerable attention. Recent work indicates that the reduced amount of Rubisco associated with C4 photosynthesis leads to the bundle sheath reactions becoming an important limitation at low temperatures (Kubien et al. 2003). Alternatively, C3 plants have a higher maximum quantum yield (φCO2, the linear region of the light-response curve) than C4 plants when measured below 23 to 30 °C (Ehleringer & Björkman 1977; Ehleringer & Pearcy 1983), and this may allow C3 plants to have higher rates of carbon gain when light is limiting at low temperatures (Ehleringer 1978). Despite these fundamental limitations, some C4 species persist in high latitude or high elevation sites, indicating that C4 photosynthesis is not inherently excluded from such areas (Schwarz & Redmann 1988; Sage & Sage 2002; Kubien & Sage 2003).

It has also been proposed that their lack of success in cool climates reflects an inherently greater susceptibility to low-temperature photo-inhibition in C4 versus C3 species (Long 1983). This is a comprehensive hypothesis, because photo-inhibition at chilling temperatures may be manifest as a direct effect on the light-harvesting apparatus, or as an indirect effect from some lesion in carbon fixation and reduction. For example, chilling temperatures affect both the light-harvesting apparatus and the CO2-fixing enzymes of corn (Long, East & Baker 1983). The lack of alternative sinks for light energy and the slower induction of the C4 cycle have been suggested to render C4 plants more susceptible to chilling-induced photo-inhibition (Long 1983). Further, a Rubisco limitation at low temperatures in C4 plants (Kubien et al. 2003) represents a reduction in the strength of light energy sinks, which could promote photo-inhibition.

The definition of photo-inhibition is a vital consideration. Photodamage has been termed chronic photo-inhibition, which is a slowly relaxing (hours to days) phenomenon involving damage to reaction centres and pigments (Osmond 1994). Chronic photo-inhibition often occurs following exposure to high light, and is exacerbated by low temperatures (Long, Humphries & Falkowski 1994; Huner, Öquist & Sarhan 1998). Low temperatures slow the rate of de novo chlorophyll and protein synthesis, increasing the likelihood of chronic photo-inhibition by altering the balance between damage and repair (Huner et al. 1993). Conversely, dynamic photo-inhibition is a protective process characterized by short (minutes to hours) relaxation times (Osmond 1994). Dynamic photo-inhibition often involves an increase in the capacity for photoprotection, typically through the dissipation of light energy by photoprotective pigments such as the xanthophyll cycle carotenoids, and anthocyanins (Demmig-Adams 1990; Ort 2001; Pietrini, Iannelli & Massacci 2002).

Chronic photo-inhibition may impair the performance of warm-climate C4 species when they are exposed to cool temperatures. For example, maize is subject to reaction centre damage at low temperature, and prolonged depression of photosynthesis is related to the rate at which damaged centres can be repaired (Fryer et al. 1995; see Long et al. 1994). However, when exposed to high light at chilling temperatures, the cool-climate species Muhlenbergia glomerata (C4) and Calamogrostis canadensis (C3) do not differ in their susceptibility to photodamage (Kubien et al. 2001). When grown at low temperatures, C4 species often increase the pool size and epoxidation state of the xanthophyll cycle pigments (Haldimann, Fracheboud & Stamp 1995; Leipner, Francheboud & Stamp 1997; Kubien et al. 2001). This may result in a depression of the maximum efficiency of exciton transfer (Fv/Fm), with a concomitant increase in the induction rate of non-photochemical fluorescence quenching upon illumination (Leipner et al. 1997). Although a reversible effect, such dynamic photo-inhibition may play a role in the seemingly reduced performance of C4 plants in cool climates by reducing the potential carbon gain.

In this study, we examine photo-inhibition and carbon gain in the C4 grass M. glomerata and the C3 species C. canadensis. These grasses frequently co-occur in boreal fens. Both species have wide latitudinal ranges; M. glomerata occurs from mid-latitudes to above 60°N (Schwarz & Redmann 1988), and C. canadensis occurs from the mid-temperate zone to the Arctic coast (Johnson et al. 1995). This approach allows for direct comparisons between C3 and C4 species that occupy the same cool climates. Further, using M. glomerata as a representative C4 grass mitigates the effects of general chilling intolerance that are often associated with warm-climate C4 species (Sage & Kubien 2003). We address the following questions: (1) does growth at low temperature have a different impact on the photosynthetic performance of putatively chilling tolerant C4 and C3 grasses at low measurement temperatures; and (2) does dynamic photo-inhibition affect the potential carbon gain of these species?


Plant growth

Muhlenbergia glomerata (Willd.) Trin. (C4) and C. canadensis (Michx.) Beauv. (C3) were collected from a fen near Plevna, Ontario (45° N, 76°5′ W). Rhizomes were planted in 6 L pots containing 66% (v/v) Promix (Plant Products, Brampton, Canada), 17% sand, and 17% plant-compost. Plants were grown in controlled environment chambers (GC-20; Enconair, Winnipeg, Canada) and maintained under a 16 h photoperiod with a maximum photosynthetic photon flux density (PPFD) of 800 µmol m−2 s−1. Plants were grown at day/night temperatures and relative humidities of 14/10 °C and 50/80% (cool-grown) or 26/22 °C and 70/80% (warm-grown), respectively. Plants were grown for 10 weeks prior to the beginning of the experiment, and growth regimes were rotated between two chambers every 2 to 3 weeks to minimize between-chamber variation. Plants were watered daily and fertilized weekly with 0.5× Hoagland's solution supplemented with 3 m m NH4NO3.

Gas-exchange and chlorophyll a fluorescence measurements

Photosynthesis was measured in an open-type leaf gas-exchange system described elsewhere (Kubien et al. 2003). The light response of net CO2 assimilation was measured at 10, 20 and 30 °C, and at O2 and CO2 partial pressures of 200 ± 10 mbar and 370 ± 3 µbar, respectively. Leaf temperature was measured by placing three fine wire thermocouples (36 ga) in contact with the abaxial surface of the leaves. The leaf–air vapour pressure deficit (VPD) was maintained at 14 ± 2 mbar at the warmer temperatures, and 8 ± 2 mbar at 10 °C. Illumination was provided by a cool-light source (KL-2500; Schott, Mainz, Germany) to minimize interference with the fluorescence detector (see below). The PPFD in the cuvette was measured using a photodiode (G1738; Hamamatsu, Bridgewater, NJ, USA) calibrated against a quantum sensor (Li-190s: LiCor Inc., Lincoln, NE, USA). All gas-exchange calculations follow von Caemmerer & Farquhar (1981).

Photosynthesis was measured simultaneously on two to four leaves from a single plant, and different plants were used for each replicate. After 30 min in the dark at the measurement temperature the light intensity was set at 250 µmol m−2 s−1. The leaves were allowed to acclimate to this PPFD for a minimum of 45 min before measurement of steady-state CO2 assimilation and fluorescence parameters. Light intensity was subsequently decreased to the compensation point, and then increased to saturation; no hysteresis was detected. At each PPFD, the leaves were allowed to equilibrate for a minimum of 15 min before measurements were made. After measurements were completed, leaf area was measured and the leaves were frozen in liquid N2 for chlorophyll determination. Chlorophyll content was assayed spectrophotometrically in N,N-dimethylforamide, following Porra, Thompson & Kriedeman (1989).

Chlorophyll a fluorescence was determined simultaneously with gas exchange during the light-response measurements. We used a PAM-101 (Walz, Effeltrich, Germany) equipped with an emitter-detector unit (ED-101BL; Walz) that provides excitation light at 470 nm and detection in the 660–710 nm waveband. This enabled the isolation of the fluorescence signal originating from photosystem II (PSII) (Adams et al. 1990; Pfündel 1998). Leaves were allowed to dark-adapt at the measurement temperature for 30 min before the ratio of variable to maximal fluorescence (Fv/Fm) was determined. Reaction centre closure was achieved by applying a 0.8-s pulse of high light (approximately 4000 µmol m−2 s−1). Once net photosynthesis had reached steady state at each PPFD, the quantum yield of PSII (ΦPSII) was measured (Genty, Briantais & Baker 1989) by applying three saturating pulses at 90 s intervals; the mean fluorescence peak was used as the estimate of Fm′. There was no trend for Fm′ to decline with successive pulses. The minimum fluorescence emission of light-adapted leaves (Fo′) was assessed by rapidly darkening the leaf in the presence of far-red light, 30 s after the last saturation pulse. All fluorescence nomenclature and calculations follow van Kooten & Snel (1990).

Transfer experiment

The effect of a short exposure to warm temperatures on the photosynthetic light responses of cool-grown plants was assessed by transferring plants grown at 14/10 °C to the 26/22 °C chamber for 24 h. After 1 d in the warm temperature regime, the light responses of photosynthesis and fluorescence were measured at 30 °C, as described above. Plants were transferred at random points throughout the photoperiod, but all plants experienced an entire diurnal cycle at the warm temperatures.


In C. canadensis (C3) grown at 26/22 °C, the light-saturated rate of net CO2 assimilation (A) was approximately 14 µmol m−2 s−1 when measured at 20 or 30 °C, and 8 µmol m−2 s−1 at 10 °C (Fig. 1a). In warm-grown M. glomerata (C4), A was approximately 17 µmol m−2 s−1 at 20 °C, and 28 µmol m−2 s−1 at 30 °C (Fig. 1b). At 10 °C, A in the C4 species was less than 7 µmol m−2 s−1 at light saturation. Cool-grown (14/10 °C) C. canadensis had reduced photosynthetic rates relative to warm-grown plants when measured at 20 and 30 °C, with maximum CO2 assimilation rates of 10–11 µmol m−2 s−1 at these temperatures (Fig. 2a). At 10 °C the light-saturated A of cool-grown C. canadensis was 8 µmol m−2 s−1, as in the warm-grown plants. In M. glomerata low growth temperatures lead to greatly reduced rates of photosynthesis relative to warm-grown plants at all measurement temperatures. At 20 or 30 °C, A in cool-grown M. glomerata was 12–13 µmol m−2 s−1, and at 10 °C was less than 5 µmol m−2 s−1 (Fig. 2b).

Figure 1.

The photosynthetic response to light intensity (PPFD) at three temperatures in (a) Calamogrostis canadensis (C3) and (b) Muhlenbergia glomerata (C4). Plants were grown at 26/22 °C under a 16 h photoperiod. Measurements were made at ambient conditions of 370 µbar CO2 and 200 mbar O2, at leaf temperatures of 10 °C (squares), 20 °C (triangles), or 30 °C (circles). Each value represents the mean (± 1 SE) of measurements on sets of leaves from four individual plants.

Figure 2.

The photosynthetic response to light intensity (PPFD) at three temperatures in (a) Calamogrostis canadensis (C3) and (b) Muhlenbergia glomerata (C4). Plants were grown at 14/10 °C under a 16 h photoperiod. Measurements were made at ambient conditions of 370 µbar CO2 and 200 mbar O2, at leaf temperatures of 10 °C (squares), 20 °C (triangles), or 30 °C (circles). Each value represents the mean (± 1 SE) of measurements on sets of leaves from four individual plants.

In both species, the maximum quantum yield of photosynthesis (φCO2, the initial slope of the light-response curve) was higher in warm- versus cool-grown plants at all measurement temperatures (Fig. 3, circles and squares only; Table 1). In C. canadensis φCO2 was inversely related to measurement temperature, regardless of the growth condition (Fig. 3a). In M. glomerata, φCO2 was insensitive to measurement temperature (Fig. 3b). The maximum quantum yield of the C3 plants exceeds that of the C4 species at temperatures below 15 or 20 °C, when both species are grown under the warm and cool regimes, respectively.

Figure 3.

The temperature effect on the maximum quantum yield of net CO2 assimilation (φCO2) of (a) Calamogrostis canadensis (C3) and (b) Muhlenbergia glomerata (C4) grown at 14/10 °C (▪) or 26/22 °C (○) Quantum yield values were obtained from the regression of the relationship between net CO2 assimilation and absorbed PPFD, at incident light intensities of less than 150 µmol m−2 s−1 (mean ± 1 SE, n = 4). Leaf absorbance between 380 and 720 nm was measured in an integrating sphere (SD2000; Ocean Optics, Dunedin, FL, USA). In each panel, the filled-triangle indicates the results of the transfer experiment, where cool-grown M. glomerata and C. canadensis were exposed to the warm-growth condition for 24 h before measurement; these symbols have been shifted +1 °C for clarity (mean ± SE, n = 3).

Table 1.  Analysis of variance of the maximum quantum yield of CO2 assimilation (φCO2) in Calamogrostis canadensis (C3) and Muhlenbergia glomerata (C4)
  1. The main effects are growth temperature (G; 14/10 °C or 26/22 °C), photosynthetic pathway (P; C3 or C4), and measurement temperature (M; 10, 20, or 30 °C). The three-way interaction was not significant (P > 0.5). The anova was run in S-Plus (MathSoft 1994).

φCO2G 14.42 e−30.000
P 16.00 e−50.399
M 23.35 e−40.025
G × P 13.90 e−50.495
G × M 23.00 e−50.698
P × M 22.74 e−40.046
Residuals368.21 e−5 

Low growth temperatures led to reduced Fv/Fm in both species, but particularly in M. glomerata in which the Fv/Fm of the cool-grown plants was 20% lower than in the plants grown at 26/22 °C (Table 2). In both species, low growth temperatures led to increases in the pool size of the xanthophyll-cycle pigments, and reductions in its epoxidation state (EPS) (Table 2; Kubien et al. 2001). These effects were most pronounced in M. glomerata; the xanthophyll pigment pool size (V + A + Z) was 124 mmol inline image in the cool-grown plants, and 60 mmol inline image in plants grown at warm temperatures. Further, the EPS of the cool-grown C4 is 41% less than in warm-grown leaves, but in C. canadensis grown at 14/10 °C the EPS of the xanthophyll cycle pigments was only 18% less than in warm-grown leaves (Table 2).

Table 2.  Light harvesting and photoprotective characteristics of Muhlenbergia glomerata (C4) and Calamogrostis canadensis (C3)
temp. (°C)
(µmol m−2)
V + A + Z
(mmol mol1 Chl)
(V + 0.5 A)/(V + A + Z)
  1. The Fv/Fm measurements were obtained following a 30-min dark period. Chlorophyll determinations follow Porra et al. (1989); each value represents the mean (± 1 SE) of 10–12 measurements. Data on the pool size of the xanthophyll pigments (V + A + Z) is from Kubien et al. (2001) (mean ± 1 SE, n = 5). Xanthophyll pigments were determined by HPLC following Thayer & Björkman (1990); V, A and Z represent violaxanthin, antheraxanthin, and zeaxanthin, respectively. The epoxidation state (EPS) of the xanthophyll pool was calculated following Thayer & Björkman (1990). Values with different superscripts are significantly different from each other (P < 0.05, Tukey's test). 1Values in parenthesis from plants that had been transferred to 26/22 °C for 24 h (mean ± SE, n = 3). See text for details.

M. glomerata14/10 0.61 ± 0.02a
(0.74 ± 0.02)1
286 ± 23a123.8 ± 6.3c0.49 ± 0.04c
M. glomerata26/22 0.77 ± 0.01b363 ± 55b 60.4 ± 5.3b0.83 ± 0.03ab
C. canadensis14/10 0.74 ± 0.02c
(0.80 ± 0.01)1
254 ± 25a 91.1 ± 3.8a0.75 ± 0.02a
C. canadensis26/22 0.77 ± 0.01b302 ± 46a 62.1 ± 1.6b0.91 ± 0.03b

The quantum yield of PSII (ΦPSII) declined with increasing light intensity at all measurement temperatures, in both species (Fig. 4). The decrease in ΦPSII was more abrupt when measured at 10 °C, particularly in the warm-grown plants. There was no difference in ΦPSII between warm- and cool-grown C. canadensis measured at 20 or 30 °C (Fig. 4a & c). In cool-grown M. glomerata, ΦPSII at low light intensities was lower than in the other plants. In both species, photochemical quenching (qP) declined with increasing light intensity at all measurement temperatures (Fig. 5). In M. glomerata, qP was not affected by growth temperature. At light saturation, qP was higher in cool-grown C. canadensis than in warm-grown plants at 30 °C. Under saturating illumination at 10 °C, qP was severely reduced in both species. The decline in qP was accompanied by an increase in non-photochemical quenching (qN) in both species (Fig. 5). Growth temperature had no effect on the magnitude of qN in C. canadensis. In M. glomerata, qN was higher in the cool-grown than in the warm-grown plants at low light (< 150 µmol m−2 s−1) at 10 °C, but at 30 °C there was no difference (Fig. 5b & d).

Figure 4.

The relationship between light intensity and the quantum yield of PSII (ΦPSII) in Calamogrostis canadensis (C3) and Muhlenbergia glomerata (C4). Plants were grown at 14/10 °C (a, b) or 26/22 °C (c, d). Chlorophyll a fluorescence was measured concurrently with gas exchange, at leaf temperatures of 10 °C (▪), 20 °C (▵), or 30 °C (•). The ambient partial pressures of CO2 and O2 were 370 µbar and 200 mbar, respectively. Each value represents the mean of measurements on sets of leaves from four individual plants. For clarity, error bars are included on the 30 °C data only; the standard error was similar at each measurement temperature.

Figure 5.

The effect of light intensity on the photochemical (qP, circles) and non-photochemical quenching (qN, squares) of chlorophyll a fluorescence in Calamogrostis canadensis (C3) and Muhlenbergia glomerata (C4). Plants were grown at 14/10 °C (a, b) or 26/22 °C (c, d). Chlorophyll a fluorescence was measured concurrently with gas exchange, at leaf temperatures of 10 °C (open symbols) or 30 °C (filled symbols). The partial pressures of CO2 and O2 entering the cuvette were 370 µbar and 200 mbar, respectively. Each value represents the mean (± 1 SE) of measurements on sets of leaves from four individual plants.

The transfer of cool-grown plants of either species to 26/22 °C for 24 h led to an increase in photosynthesis relative to that measured in plants not exposed to the warm temperatures (Fig. 6a). In C. canadensis, the transferred plants had CO2 assimilation rates that were 40% higher than was measured in the warm-grown plants at 30 °C. Photosynthesis in transferred M. glomerata did not match the rate attained by the warm-grown plants, but was nearly doubled in comparison with cool-grown plants that were measured before the warm-temperature exposure. The transfer to 26/22 °C resulted in an increase in Fv/Fm in both species (Table 2), and the quantum yield of CO2 assimilation (φCO2) increased by 67 and 63% in M. glomerata and C. canadensis, respectively (Fig. 3a & b, triangles). In both species the increase in photosynthesis was accompanied by higher electron transport rates at all light intensities, relative to the plants that were not transferred to the warm conditions (Fig. 6b).

Figure 6.

The responses of (a) net photosynthesis, and (b) the linear electron transport rate to light intensity in Calamogrostis canadensis (C3, □) and Muhlenbergia glomerata (C4, •). Electron transport rate was determined from φPSII, as described previously (Kubien & Sage 2004). Plants were grown at 14/10 °C; before measurement, plants were transferred to a 26/22 °C chamber for 24 h. Measurements were conducted at 30 °C. Each value represents the mean (± 1 SE) of measurements on leaves from three different individuals. In both panels, lines indicate cool- (—) or warm-grown (- - - -) M. glomerata, and cool- (— ·· —) or warm-grown (— · —) C. canadensis measured at 30 °C, as shown in Figs 1, 2 and 4.


There is little evidence that M. glomerata and C. canadensis differ in their susceptibility to chronic photo-inhibition. Recovery from chronic photo-inhibition may take several days (Osmond 1994), but the high rates of photosynthesis in cool-grown plants of both species following a 24 h exposure to warm temperatures (26/22 °C) indicate that the photo-inhibition detected here is reversible. Low growth temperatures do not improve the low-temperature photosynthetic performance of this putatively cold-tolerant C4 species. Muhlenbergia glomerata grown at 14/10 °C exhibits a significant reduction in the quantum yield of CO2 assimilation, associated with large increases in the pool size of the xanthophyll pigments and non-photochemical energy dissipation. Protective down-regulation of PSII during cold acclimation serves to compensate for an imbalance between the production of NADPH and ATP and the energy requirements of the stromal reactions of CO2 fixation; this avoids a build-up of potentially damaging excitation pressure (Huner et al. 1993, 1998). In the case of M. glomerata, an increase in photoprotective capacity may compensate for the Rubisco limitation of C4 photosynthesis at low temperatures (Kubien & Sage 2004), thereby enabling C4 plants to readily utilize the warm episodes that are required in order for them to persist in cool climates.

The effect of growth and measurement temperature on φCO2 and ΦPSII

The linear portion of the light-response curve reflects the maximum efficiency of light harvesting, and reductions in φCO2 are associated with photo-inhibition (Long et al. 1994). The qualitative temperature responses of φCO2 determined here are consistent with previous observations; in C3 species φCO2 declines with increasing measurement temperature, whereas in C4 plants φCO2 is largely insensitive to measurement temperature (Ehleringer & Björkman 1977; Ehleringer & Pearcy 1983). In the present study, low growth temperature led to reduced φCO2 in both the C3 and C4 species. Decreases in φCO2 and/or Fv/Fm following chilling have been noted previously for a range of C4 and C3 species (Baker, East & Long 1983; Demmig et al. 1987). However, this can indicate either chronic or dynamic photo-inhibition.

In both M. glomerata and C. canadensis the decline in ΦPSII with increasing PPFD was more pronounced at 10 °C relative to 20 or 30 °C. Similar findings were reported in the C4 sedge Cyperus longus (Blowers & Baker 1995). Non-photochemical fluorescence quenching (qN) is inversely related to ΦPSII, and qN reaches a maximum at a lower PPFD at 10 °C than at 30 °C. This is consistent with a lower EPS of the xanthophyll cycle pigments at low temperatures (Table 2); low EPS is associated with non-photochemical quenching and dynamic photo-inhibition (Demmig et al. 1987; Demmig-Adams & Adams 1992; Osmond 1994). The reduced φCO2 in cool-grown M. glomerata relative to its warm-grown counterpart is due to the higher non-photochemical quenching that is associated with the lower epoxidation state of the xanthophyll cycle even at low light intensities. Non-photochemical energy dissipation will depress both φCO2 and ΦPSII. Similar results are apparent in cool-grown C. canadensis, although both φCO2 and EPS are higher in this C3 species than in the cool-grown C4 plant at low measurement temperatures.

Implications for models of C3 and C4 distribution based on φCO2

Models based on the maximum quantum yield of CO2 assimilation have been used to explain the latitudinal transition from C4- to C3-dominated landscapes and the relative advantages of C4 versus C3 species through evolutionary time (Ehleringer 1978; Ehleringer, Cerling & Helliker 1997). These models assume that the φCO2 of C4 plants is insensitive to temperature, which has been demonstrated in a variety of species during short-term evaluations of plants grown at favourable temperatures (20 to 30 °C) (Ehleringer & Björkman 1977; Ehleringer & Pearcy 1983). However, if φCO2 in C4 species is affected by growth temperature then such models would not reflect field conditions. Our data indicate that φCO2 is sensitive to low growth temperature in M. glomerata, and therefore predictions regarding the distribution of the two pathways based solely on φCO2 might be over-simplified. However, maximum quantum yield in C3 plants is directly proportional to the photorespiratory inhibition of A (Sage & Kubien 2003) and thus is a valuable index of the relative efficiency of A in C3 and C4 species regardless of light intensity. It is the relationship between φCO2 and photorespiratory inhibition that gives the quantum yield model its predictive power, not the performance of A in low light conditions as has often been assumed.

We have examined the effect of varying photosynthetic quantum yield over the course of a growing season by modelling the carbon gain of M. glomerata and C. canadensis(Fig. 7). If φCO2 in M. glomerata is constrained to the value observed in the cool-grown plant (0.042 mol CO2 inline image, Fig. 3b) then this species may fix less carbon than its C3 competitor over the growing season (Fig. 7). If there is no dynamic photo-inhibition and the quantum yield of M. glomerata is 0.062 mol CO2 inline image (Fig. 3b), then the C4 species would be able to acquire more carbon than the C3 plant, particularly if φCO2 in C. canadensis is not allowed to vary over the season. The actual amount of net photosynthesis should lie between the two extremes shown here. This simulation ignores differences in growth and allocation between these species, and simply reflects potential carbon gain of the C4 and C3 pathways in a cool southern boreal climate. As in temperate grasslands, the amount of carbon that the C3 species is able to acquire in early spring may very important in boreal fens (Sage et al. 1999). At the southern extremes of the boreal zone, C. canadensis generally emerges very soon after snowmelt in early April and senesces by mid to late August, whereas M. glomerata does not typically emerge until late May and remains active until late September (D.S.K., personal observation). Following severe winters, high water levels in the fens occupied by C. canadensis and M. glomerata can delay formation of the C3 canopy until relatively late into the spring (Kubien & Sage 2003). In the simulation shown in Fig. 7, delaying the growth of C. canadensis until day 120 would reduce seasonal carbon gain by approximately 20% in either scenario modelled here. By shortening the growing season for the C3 species but not the C4 plant, high water levels may contribute to the occurrence of the C4 species in these boreal ecosystems, which are atypical for plants with the C4 pathway.

Figure 7.

Modelled photosynthetic carbon gain in Muhlenbergia glomerata (C4) and Calamogrostis canadensis (C3) over a growing season. Values in parenthesis are the area under each curve, expressed in grams C. Curves were generated using Wimovac (Humphries & Long 1995), with the seasonal air temperature and radiation profiles at Renfrew ON (45°29′ N 76°42′ W), where the mean annual temperature is 4.9 °C (Canadian Climate Normals, http://www.climate.weatheroffice.ec.gc.ca/climate_normals/index_e.html). For M. glomerata, we specified φCO2 as 0.062 or 0.042 mol CO2 inline image, representative of warm- or cool-grown plants (Fig. 3b). For C. canadensis we specified φCO2 = 0.040 mol CO2 inline image, or allowed φCO2 allowed to vary with ambient temperature, as in Fig. 3a.

Growth temperature, dynamic photo-inhibition and carbon gain

When grown at 26/22 °C, C. canadensis has greater light-saturated and light-limited photosynthesis rates than M. glomerata when measured at 10 °C. In these cool-climate species, acclimation to low temperature exacerbated the C3 advantage, particularly at light saturation. The up-regulation of enzymes associated with sucrose metabolism during low-temperature acclimation enables C. canadensis to overcome low-temperature photosynthetic limitations and maintain higher rates of photosynthesis at lower measurement temperatures than M. glomerata or the warm-grown C. canadensis (Kubien & Sage 2004). Rubisco capacity places a low-ceiling carbon-gain in C4 species at low measurement temperatures, and low growth temperatures do not alter this result in M. glomerata (Kubien & Sage 2004). If C4 plants are to persist in cold climates, they need to be able to rapidly recover from chilling and exploit warm episodes (Sage & Sage 2002; Kubien & Sage 2003). In M. glomerata, this recovery probably involves a reduction in the high levels of non-photochemical quenching that appear to be necessary for this C4 plant to survive when grown at 14/10 °C. An alternative strategy is shown by Miscanthus × giganteus. When this C4 grass is grown at 14/11 °C (day/night) there is no decline in photosynthesis relative to warm-grown (25/20 °C) plants, when measured between 5 and 30 °C (Naidu et al. 2003). Further, field-grown M. giganteus does not show reductions in φCO2 at the start of the growing season in southern England (Beale, Bint & Long 1996). Miscanthus × giganteus does not appear to engage high levels of photoprotection in order to tolerate low growth temperatures, presumably because the high rates of photosynthesis maintain the consumption of light energy. It appears to maintain high rates of photosynthesis at low temperatures because levels of Rubisco and PPDK are not reduced, in contrast to the situation with maize (Naidu et al. 2003). However, growth temperature does not affect the amount of Rubisco or PEPCase in M. glomerata (Kubien & Sage 2004), indicating that multiple strategies are employed by those C4 species that inhabit cool climates.

Recovery of photosynthetic capacity from low temperature exposure is typically on the order of days to weeks in C4 species of tropical origin. Exposure to low growth temperatures does not improve the photosynthetic performance of maize, and carbon gain remains depressed when plants are returned to favourable temperatures (Nie, Long & Baker 1992). When grown at 14 °C, maize showed only a small increase (from approx. 5–7 µmol m−2 s−1) in photosynthesis following 24 h at 25 °C, despite a doubling of chlorophyll content and a 25% increase in the rate of electron transport through PSII (Nie et al. 1995). In a chilling-sensitive maize genotype, the recovery of photosynthetic capacity following growth at 16/12 °C was incomplete even after 4 weeks at 25/20 °C, relative to plants grown at the warmer temperatures (Pietrini et al. 1999). Slow recovery from low-temperature exposure may preclude tropical C4 species from cool climates, but this cannot explain the general rarity of the C4 pathway from such habitats. The C4 species that do occur in cool climates are consistently resistant to chilling injury, but they exhibit variable patterns of photosynthetic acclimation to low temperatures. Cool growth temperatures enhance low-temperature photosynthesis in high elevation populations of Bouteloua gracilis, but not in low elevation ecotypes (Pittermann & Sage 2000). In the high altitude C4 grass Muhlenbergia montanum, low night temperatures enhance photosynthesis at the thermal optimum, but not at low measurement temperatures (Pittermann & Sage 2001).

Photoprotection and C4 plants in cool climates

An important question is why M. glomerata should have a greater need for photoprotection under cool temperatures. The slower induction rate of C4 photosynthesis, particularly at low temperatures, may necessitate greater photoprotection in order to prevent photodamage during periods when enzyme activation and substrate availability are limiting for photosynthesis. The lack of alternative sinks for light energy, such as photorespiration, is a possible explanation (Long 1983). Photorespiration can alleviate photo-inhibition by acting as a sink for light energy in C3 plants (Björkman & Demmig-Adams 1995; Kozaki & Takeba 1996), but the potentially high carboxylation efficiency of C4 plants should reduce the need for energy dissipation through this pathway. An alternative explanation is that the enhanced photoprotective capacity observed in M. glomerata grown at 14/10 °C compensates for the relative lack of Rubisco. C4 plants have considerably less Rubisco than C3 species, and recent evidence shows that Rubisco capacity is an important control over C4 photosynthesis at temperatures below about 20 °C (Pittermann & Sage 2000, 2001; Kubien et al. 2003; Kubien & Sage 2004). Rubisco capacity places a low ceiling on photochemical light energy dissipation and the potential photosynthesis of C4 plants at low temperatures. By preventing photodamage, dynamic photo-inhibition may be a means by which C4 species may tolerate, rather than avoid, the cooler temperatures that prevail during the early part of the growing season in high latitude climates. This strategy may then allow these plants to rapidly respond to the warmer temperatures that occur during the middle of the growing season, enabling C4 species to persist in high latitude and altitude environments.


The authors wish to thank Alex Ivanov and Norm Huner for access to the HPLC and particularly for the many helpful discussions, and George Espie for access to the PAM. The comments of two anonymous reviewers have led to a greatly improved manuscript. This work was funded by an NSERC grant (OGP0154273) to R.F.S.