Low-temperature leaf photosynthesis of a Miscanthus germplasm collection correlates positively to shoot growth rate and specific leaf area.

BACKGROUND AND AIMS: The C4 perennial grass miscanthus has been found to be less sensitive to cold than most other C4 species, but still emerges later in spring than C3 species. Genotypic differences in miscanthus were investigated to identify genotypes with a high cold tolerance at low temperatures and quick recovery upon rising temperatures to enable them to exploit the early growing season in maritime cold climates. Suitable methods for field screening of cold tolerance in miscanthus were also identified.
METHODS: Fourteen genotypes of M. sacchariflorus, M. sinensis, M. tinctorius and M. × giganteus were selected and grown under warm (24 °C) and cold (14 °C) conditions in a controlled environment. Dark-adapted chlorophyll fluorescence, specific leaf area (SLA) and net photosynthetic rate at a photosynthetically active radiation (PAR) of 1000 μmol m(-2) s(-1) (A1000) were measured. Photosynthetic light and CO2 response curves were obtained from 11 of the genotypes, and shoot growth rate was measured under field conditions. KEY
RESULTS: A positive linear relationship was found between SLA and light-saturated photosynthesis (Asat) across genotypes, and also between shoot growth rate under cool field conditions and A1000 at 14 °C in a climate chamber. When lowering the temperature from 24 to 14 °C, one M. sacchariflorus exhibited significantly higher Asat and maximum photosynthetic rate in the CO2 response curve (Vmax) than other genotypes at 14 °C, except M × giganteus 'Hornum'. Several genotypes returned to their pre-chilling A1000 values when the temperature was increased to 24 °C after 24 d growth at 14 °C.
CONCLUSIONS: One M. sacchariflorus genotype had similar or higher photosynthetic capacity than M × giganteus, and may be used for cultivation together with M × giganteus or for breeding new interspecies hybrids with improved traits for temperate climates. Two easily measured variables, SLA and shoot growth rate, may be useful for genotype screening of productivity and cold tolerance.

INTRODUCTION

C4 plant species are considered to have a carbon assimilation advantage over C3 species because of their ability to concentrate CO2 around Rubisco in the bundle sheath cells, thus reducing the risk of photo-oxidation. C4 species dominate in warm and light-intensive environments, while they are rare in cool climates because of a marked loss of photosynthetic capacity at low temperatures (Long, 1999), and the distribution of most C4 species is limited by a mean minimum temperature of 8–10 °C during the period of active growth (Long, 1983). However, of all the C4 plants, the perennial rhizomatous C4 grass miscanthus (M. × giganteus) exhibits exceptional cold tolerance (Beale and Long, 1995; Dohleman and Long, 2009). Because of its high yield potential under cool climatic conditions and rapid accumulation of biomass, it has been proposed as a potential bioenergy crop in northern Europe and the USA (Heaton ; Zub and Brancourt-Hulmel, 2010; Long and Spence, 2013).

Previous studies have mainly focused on the sterile triploid hybrid M. × giganteus, which has been reported to have a high photosynthetic capacity (Wang , ), leaf expansion rate (Clifton-Brown and Jones, 1997) and biomass productivity in central and northern Europe at mean annual temperatures of 7·3–8·0 °C (Lewandowski ).

Miscanthus × giganteus is a natural sterile hybrid between M. sacchariflorus and M. sinensis, which was introduced into Europe in the 1930s and has been widely investigated and grown for bioenergy (Greef and Deuter, 1993). The original clone is now widely distributed across Europe and the USA, where many local varieties have been named. However, different varieties of M. × giganteus are genetically very similar (Głowacka ). The genus Miscanthus has its origin in the tropics and sub-tropics, but different species are found throughout a wide climatic range in East Asia (Greef and Deuter, 1993). Gauder found that M. × giganteus produced higher yields than clones of M. sacchariflorus, M. sinensis and M. sinensis hybrids in a field study in south-western Germany. However, in newly established miscanthus plantations, several M. sinensis genotypes are reported to have greater frost tolerance and winter survival than M. × giganteus and M. sacchariflorus (Clifton-Brown and Lewandowski, 2000). Investigations within different miscanthus species and environments of origin may therefore help to broaden the genetic base for the photosynthetic response to low temperatures which is useful when breeding for new high-yielding hybrids with improved cold tolerance (Głowacka ). The availability of a larger number of high-quality genotypes might also increase the resistance to pests and diseases in large-scale plantations.

Miscanthus × giganteus has previously been found to form photosynthetically competent leaves at lower temperatures than maize (Naidu, 2003) and to have a larger leaf area and longer growing season than maize (Dohleman and Long, 2009). The reason for the ability of M. × giganteus to maintain photosynthetic capacity at low temperatures is apparently related to a higher content and activity of pyruvate orthophosphate dikinase (PPDK) (Naidu ; Wang ) and Rubisco (Spence, 2012), and not of phosphoenolpyruvate carboxylase (PEPc) (Naidu and Long, 2004). A higher content of the carotenoid zeaxanthin may also help to protect photosystem II (PSII) from photoinhibition (Farage ). Although miscanthus emerges later in spring and thus has a delayed canopy development compared with C3 grass species, recent studies indicate that the limitations due to temperature are not an inherent C4 trait that cannot be circumvented (Long and Spence, 2013). This increases the feasibility of finding genotypes that are better suited to temperate areas and can better utilize the entire length of the growing season in cool climates.

Even though M. × giganteus is relatively cold tolerant, it is likely that further cold tolerance exists within the Miscanthus genus (Głowacka ). To utilize the whole growing season fully, quick recovery of photosynthesis when exposed to higher temperatures is another important feature that plants should possess when grown in maritime climates. Thus, it will be helpful for future breeding work if miscanthus genotypes had a high cold tolerance at low temperatures in conjunction with a rapid restorative capacity of the photosynthesis when temperatures rise. Another essential variable to consider is the canopy development in early spring. Efficient capture of solar radiation in early spring is dependent not only on cold-tolerant photosynthetic processes, but also on fast leaf area development. At the canopy level, a large specific leaf area (SLA) allows a more extensive foliar display for a similar biomass investment in leaves, resulting in improved light absorption (Niinemets, 1998). In miscanthus, nitrogen fertilization did not increase the net photosynthetic rate (A), but increased SLA (Wang ), which means that thin leaves at high N-fertilization rates resulted in similar values of A but increased the photosynthetic rate per gram of leaf (Amass). Such a positive increase in Amass with N-fertilization has also been observed in willow (Salix viminalis and Salix dasyclados) (Merilo ). It would therefore be interesting to investigate whether the genetic variation within a species would also code for differences in resource investment in leaves for photosynthesis, and if these differences are reflected in different Amass.

In previous investigations, the leaf elongation rate has been used to compare miscanthus (Purdy ) and maize (Reymond ) genotypes at low temperatures. However, the elongation rate depends on the developmental stage of the leaf (Durand ), and it is difficult to compare between genotypes due to the requirement for exactly the same age of the leaf (Muller ). It may thus be more promising to compare shoot growth rates during early vegetative stages of miscanthus growth (Clifton-Brown and Jones, 1997).

Here we studied 14 miscanthus genotypes selected from a collection of 166 genotypes on the basis of shoot growth rate, and we tested for its correlation with assimilation rate. We compared the photosynthetic performance under warm and subsequent cool conditions, and also determined whether they were able to recover their photosynthetic capacity fully upon return to warm conditions. Our hypothesis was that we would be able to identify genotypes with improved photosynthetic performance under cold conditions and with a rapid recovery rate during a temperature rise. In addition, we expected that genotypes with higher SLA and a stable A could be selected.

MATERIALS AND METHODS

Plant material

A total of 14 genotypes were selected for a detailed climate chamber experiment. Four species of miscanthus, namely M. sinensis, M. sacchariflorus, M. tinctorius and M. × giganteus (M. sinensis × M. sacchariflorus), were represented (Table 1). They were selected from a pool of 166 genotypes screened under field conditions (Supplementary Data Fig. S1) based on their shoot growth rate and leaf gas exchange rate in 2012, as described below. Of the 166 screened genotypes, 27 were grown in Poland (Institute of Plant Genetics of the Polish Academy of Sciences, Poznań, 50°2'N, 21°59'E), 41 in Sweden (Department of Biosystems and Technology, Swedish University of Agricultural Sciences, Alnarp, 55°39'N, 13°05'E) and 98 in Denmark (Department of Agroecology, Aarhus University, Foulumgaard, 56°30'N, 9°35'E). Eleven genotypes were selected from the Danish collection, of which three originated from the European Miscanthus Improvement (EMI) project (Lewandowski ; Clifton-Brown ; Jørgensen ) and eight were collected on Honshu and Hokkaido Islands in Japan in October 1995 (Kjeldsen ). The reference M. × giganteus was the clone ‘Hornum’ (Głowacka ) which was grown for many years at Hornum in Jutland, Denmark (Jørgensen, 1997), and was confirmed as a sterile triploid (Linde-Laursen, 1993). The genotype Sin-7 was selected from the Swedish collection, and Sin-6 and Gig-14 (a hybrid bred at Tinplant, Germany) from the Polish collection. The share of the gene pool screened in Denmark was a part of the larger pool previously screened for cold tolerance by chlorophyll fluorescence (Fv/Fm) in a separate study (Głowacka ). Only three genotypes (Tin-1, Sin-3 and Sac-10) were used in both the study of Glowacka et al. and our study.

T

Genotypes measured in the climate chamber experiments

Identification	Species	Origin	Altitude (m)	Latitude (ºN)	Ploidy
Tin-1	M. tinctorius	Ainukura, Honshu*	350	36	2x
Sin-2	M. sinensis	Yoichi, Hokkaido; OUAV collection*		43	2x
Sin-3	M. sinensis	Kamiyoshino (5 km south-east of Kanazawa), Honshu*	280	36	2x
Sin-4	M. sinensis	North-east of Obihiro, Hokkaido*		43	2x
Sin-5	M. sinensis	South of Shirakawa, Honshu*	600	36	2x
Sin-6	M. sinensis	Tinplant GmbH, Klein Wanzleben, Germany			2x†
Sin-7	M. sinensis	Hokkaido		43	2x
Sin-H8	M. sinensis	Hybrid from Deuter, Germany‡			3x‡
Sac-9	M. sacchariflorus	Biratori (south-east of Sapporo), Hokkaido*		42	4x
Sac-10	M. sacchariflorus	Hakusan National Park, Honshu*	900	36	4x
Sac-11	M. sacchariflorus	South of Shirakawa, Honshu*	600	36	4x
Sac-12	M. sacchariflorus	Tinplant GmbH, Klein Wanzleben, Germany‡			4x‡
Gig-13	M.×giganteus	Cultivar Hornum§, Larsen, Denmark‡			3x‡
Gig-14	M.×giganteus	Tinplant, GmbH, Klein Wanzleben, Germany¶			2x**

*Seed plants collected in Japan 1995 and grown in Denmark since 1996 (Kjeldsen ).

†Ploidy level was analysed at the Institute of Plant Genetics, Poland.

‡Clones from the ‘European Miscanthus Improvement’ project; Sac-12 corresponds to EMI-5 and Sin-H8 to EMI Sin-H6 reported by Clifton-Brown .

§EMI-1 was equivalent to Hornum, since they are almost genetically similar (Głowacka ).

¶Gig-14 corresponds to M114 reported by Sampson , origin information is reported in Kim ).

**Martin Deuter, pers. comm.

Ploidy examination

The ploidy levels of genotypes Tin-1, Sin-2, Sin-3, Sin-4, Sin-5, Sac-9, Sac-10 and Sac-11 were not known from the literature and therefore were determined by flow cytometry. Flow cytometry was applied to leaves or rhizomes by a Partec PA II flow cytometer equipped with an HBO-100 mercury arc lamp (Partec GmbH, Germany) and a filter combination for 4',6-diaminino-2-phenylindole (DAPI) staining (Partec 06-03-310). Samples were chopped for 30 s in a Petri dish containing 0·6 mL of citric acid buffer and left for 5 min to allow release of nuclei. The nuclei were stained by adding 2·5 mL of fluorescent solution containing 5 μM DAPI and left for another 5 min (Otto, 1990). The suspension of nuclei was passed through a nylon filter with pore size of 50 μm to remove large debris. The relative fluorescence of the total DNA of single nuclei was analysed, and in each sample the DNA content of 5000 nuclei was checked. Samples of a diploid M. sinensis genotype (MS 88-110) were used as an internal standard.

Screening in the field by determination of shoot growth rate

The leaf elongation rate is difficult to compare between genotypes (Muller ), so instead we chose to measure the daily shoot growth rate, which is defined as the difference in the distance between the soil surface and the tip of the tallest leaf measured on two different days, divided by the number of days between the measurements.

In May 2012, five shoots from all genotypes were marked. Using a transparent ruler (IMPEGA) with a precision of 1 mm, shoot length was measured every second or third day from shoot emergence. From these measurements, the average daily growth rate was calculated for a cold and a warm period as described below. The temperature was measured 20 cm above the soil surface in the field. In Denmark a cold period lasted from 30 May until 1 June with a daily mean temperature of 9·8 ± 0·7°C. This cold period succeeded a warm period from the 24 to 26 May, with an average temperature of 18·8 ± 0·2°C. The cool and warm periods for Sweden, determined by the daily mean temperatures, were from 4 to 7 May (9·7 ± 0· 9°C) and from 9 to 11 May (14·8 ± 0·4°C), respectively, and for Poland from 14 to 16 May (10·5°C) and from 9 to 11 May (24°C), respectively.

Screening in the field by gas exchange measurements

A total of 37 genotypes that displayed the most vigorous growth in the first 2 weeks of May 2012 were selected for gas exchange measurements in the field with 2, 3 or 4 d intervals from 18 May until 4 June 2012. The measurements were performed using CIRAS-2 (PP Systems, Amesbury, MA, USA) equipment on the same leaves as those measured for shoot growth rate. The leaves were dry and were measured between the hours of 1000 and 1400, with the following cuvette chamber conditions: photosynthetic active radiation (PAR) of 1600 μmol m–2 s–1, CO2 concentration of 390 μmol mol–1 and relative humidity (RH) as close to ambient as possible. The leaf chamber temperature was set to the ambient temperature measured.

Detailed screening in the growth chamber

On 9 April 2013, pots (ø, 16 cm; height, 50 cm) were filled with peat (Pindstrup Substrate no. 4, 10–30 mm, pH 6·0).

Rhizomes of each of the 11 genotypes from Denmark and Sin-7 from Sweden were dug up from the field on 16 April 2013. Sin-6 and Gig-14 were dug up on 11 March 2013 in Poland and stored at 2 ºC. Young rhizome pieces with live buds were planted in individual pots on 18 April. They were grown in the greenhouse for 7 weeks. The temperature in the greenhouse was set at 22/15 ºC day/night, the daylength was ambient and no supplementary light was given.

The plants were irrigated once or twice a week when necessary. A nutrient solution with 1 % inorganic fertilizer (Prima Væksthusgødning, NPK 3-1-4) was added on 8 May 2013 (300 mL per pot).

On 30 May, all plants were moved into a climate chamber and randomly arranged on trolleys with four replicates of each genotype. The plants were re-arranged randomly every second or third day in order to avoid the impact of position effects within the chamber. They were initially grown in warm conditions of 24/20°C and 14/10 h day/night cycles under a photon flux of 670 μmol m–2 s–1. The RH was set at 85 % day/night. After 10 d of acclimatization in warm conditions, light and CO2 response curve measurements were initiated as described below on a selection of 11 genotypes comprising one M. tinctorius, four M. sinensis, four M. sacchariflorus and two M.×giganteus.

After 22 d in warm conditions, the temperature was reduced to 14/10°C day/night on 24 June (designated Day 1 in cold conditions) with the same day/night cycle and light conditions as for the warm conditions. The RH under the cold conditions was 75/85 % day/night. All other conditions were set as described above. The measurement of light and CO2 response curves on the above-mentioned 11 genotypes in cold conditions was started on 30 June, which was Day 7 after the temperature had been decreased.

During the period of measurements in the climate chamber, the plants were irrigated every evening, and fertilizer was added every second or third day. ‘Nitaman 235’, which contains manganese at a concentration of 235 g L–1 and nitrate N at 120 g L–1, was sprayed on the leaves on 27 June due to manganese deficiency symptoms on leaves (leaves turned pale green in color).

Gas exchange measurements and photosynthetic response curves

The carbon assimilation rate at a PAR of 1000 μmol m–2 s–1 (A1000) was measured once at the end of the warm period (designated Day 0), then 7 h after the temperature was decreased to 14°C (Day 1), and thereafter on Day 2, 3, 4, 5, 6, 9, 11, 15 and 18. A recovery measurement of A1000 was made on Day 24, equivalent to 36 h after the temperature was increased to 24/20°C day/night.

Gas exchange was always measured on the youngest fully developed leaf (ligule present) of all plants from 0830 to 1500 h, using an open-flow gas exchange system (CIRAS-2). The environment in the leaf cuvette was set in accordance with the climate chamber conditions, i.e. the CO2 concentration was set to 400 μmol mol–1 and mean leaf temperature was maintained at the experimental temperature (24 or 14°C). The vapour pressure deficit (VPD) in the leaf cuvette was kept at 1·2 kPa (warm conditions) and 1·0 kPa (cold conditions) by controlling the relative humidity, and the airflow through the chamber was 250 mL min–1.

For the light response curve measurements, the leaves were acclimatized in the leaf cuvette to a PAR of 2000 μmol m–2 s–1 until the photosynthetic rate stabilized (Wang ). Then the PAR was decreased from 2000 to 20 μmol m–2 s–1 in 14 steps (2000, 1800, 1500, 1200, 1000, 800, 500, 300, 200, 150, 100, 80, 50 and 20). The measurements were logged after photosynthetic rates had reached steady state. The rate of photosynthesis at a PAR of 1500 μmol m–2 s–1 (A1500) was defined as the saturated net photosynthetic rate (Asat).

For the determination of A–C curves, leaves were acclimatized in the cuvette for 10 min at a CO2 concentration of 400 μmol mol–1 and a of PAR 1500 μmol m–2 s–1 until a steady state of A was reached. The CO2 concentration in the cuvette was decreased in eight steps (300, 250, 200, 150, 100, 80, 50 and 20 μmol mol–1 CO2) with around 5 min for each step. The CO2 concentration was then returned to 400 μmol mol–1 and kept at this level for about 10 min for stabilization. Thereafter it was increased in three steps (600, 1000 and 1200 μmol mol–1 CO2) with approx. 5 min for acclimatization at each step.

Chlorophyll fluorescence measurements

Chlorophyll fluorescence was determined on Day 6 after the temperature had been reduced to 14°C, using a Mini-PAM fluorometer (Walz, Germany). The leaves were dark-adapted with dark leaf clips (Walz, Germany) for 30 min before the measurements were performed. The minimal fluorescence (F0) was determined at very low PAR, where the PSII reaction centres are in the ‘open’ state. The maximal fluorescence (Fm) was measured by applying a 0·8 s pulse at a high light level of approx. 4000 μmol m–2 s–1, which drives the reaction centre to close (Krause and Somersalo, 1989). The maximum quantum yield of PSII (Fv/Fm) was calculated from F0 and Fm as Fv/Fm = (Fm – F0)/Fm.

Specific leaf area (SLA)

Ten leaf discs with an area of 0·44 cm2 were cut from the first two fully expanded leaves (normally the third and fourth leaf counting from the top) using a cork borer with a diameter of 7·5 mm from the upper canopy of each plant after the gas-exchange measurements had finished. Leaf discs were oven-dried at 80°C for 3 d, weighed, and the SLA, which describes the ratio of leaf area to leaf dry matter (DM), was calculated as m2 kg–1. In addition, Amass was calculated as Asat multiplied by SLA.

Photosynthetic response curves and calculation

Data from the light response experiment were fitted to a non-rectangular hyperbola model (Marshall and Biscoe, 1980) by means of the non-linear least squares curve-fitting procedure of R Studio for Windows (R Core Team, 2005).

The apparent quantum yield (AQY) values were obtained from linear regression of the relationship between net CO2 assimilation and PAR across five points at incident light intensities from 20 to 150 μmol m–2 s–1 (Long ). The dark respiration (Rd) was estimated from the y-intercept of the initial linear line of five measured data points (Herrick and Thomas, 1999). Asat was the rate of photosynthesis at a PAR of 1500 μmol m–2 s–1 (A1500).

In order to describe the response of A to Ci, a linear regression was fitted for the initial linear part of the A–Ci curve (Ci <100 μmol mol–1) with the slope representing the carboxylation efficiency (CE) of PEPc. The CO2-saturated photosynthetic rate (Vmax) was calculated from the horizontal asymptote of the A–Ci curve (von Caemmerer, 2000). Stomatal limitation (Ls) was calculated from the A–Ci curves [Ls = (Ai – Aa)/Ai; Ci,Ca = 400 μmol mol–1] giving the photosynthetic reduction due to stomatal closure (Long and Bernacchi, 2003).

Statistical analysis

Data for each temperature treatment were analysed separately using analysis of variance. If a significant effect was observed, multiple comparisons were performed using Turkey’s HSD (honest significant difference) test. All the analyses and tests were done in R version 3.1.2 (R Core Team, 2005). The R package ‘lsmeans’ (Lenth and Hervé, 2005) was used for multiple comparisons.

RESULTS

Ploidy

All M. sacchariflorus genotypes in this study were tetraploid, while all M. sinensis genotypes were diploid, and the two M. × giganteus genotypes were triploid and diploid, respectively (Table 1).

Shoot growth and gas exchange in the field

For the 166 genotypes tested under field conditions in 2012, the daily shoot growth rate varied from 1·8 to 6·7 cm d–1 at high temperatures and from 0·7 to 3·6 cm d–1 at low temperatures (Fig. S1). The preliminary measurements of net photosynthesis in the field indicated a positive correlation between this variable and shoot growth rate at both high (P = 0·053) and low temperatures (P = 0·167) (Fig. S1) On the basis of these results, a total of 14 genotypes were selected for further investigation in the climate chamber in 2013. They were selected based on one or more of the following traits: high daily shoot growth rate during the cool period, high carbon assimilation rate during the cool period or late onset of flowering and senescence as observed in the field (data not shown). As we wanted to cover a broad variation in productivity, we also included genotypes with low daily shoot growth rates in cool conditions. Four miscanthus species were represented by the selection.

Photosynthetic performance under warm and cold conditions

When the 14 selected genotypes were grown at the high temperature (24/20°C) they all displayed higher photosynthetic rates at all the measured light levels than they did at 14/10°C (data not shown). Furthermore, there were significant differences between genotypes under light-saturated conditions at both temperatures (Fig. 1C).

F

(A) Apparent quantum yield of CO2 uptake (AQY), (B) net photosynthetic rate at a PAR of 1500 μmol m–2 s–1 (A1500) of 11 miscanthus genotypes under warm (24 °C) and cold (14 °C) conditions; (C) carboxylation efficiency (CE) of phosphoenolpyruvate carboxylase (PEPc) and (D) maximum photosynthetic rate on the A–Ci curve (Vmax) of 11 miscanthus genotypes under warm (24 °C) and cold (14 °C) conditions. The CE was calculated as the slope of the net photosynthetic rate (A) vs. the intercellular CO2 concentration (Ci) between 10 and 100 μmol m–2 s–1, and Vmax as the asymptote of the ACi curve. Error bars represent the s.e. (n = 3 or 4). Values with the same letter are not significantly different at the P = 0·05 level; upper case letters show the comparison for the warm treatment and lower case letters for the cold treatment.

Most genotypes had higher AQY values when they were grown in warm conditions than under cold conditions (Fig. 1A). Under warm conditions, the genotypes Gig-14, Sac-12, Gig-13 and Sin-5 showed values of 0·045, 0·043, 0·043 and 0·042 mol mol–1, respectively, which were significantly higher than the values for Sac-10, Tin-1 and Sin-7 of 0·033, 0·029 and 0·023 mol mol–1, respectively. When the temperature was lowered to 14/10°C, Sac-9 and Sac-11 had significantly higher AQY values than Sin-2 and Sin-7, and the reduction in AQY was only 18 and 19 % in the two M. sacchariflorus genotypes; the reduction in Tin-1 and Sac-10 was even less.

The Rd was significantly higher under high temperatures, with the mean value of Rd declining from 1·94 to 0·91 μmol m–2 s–1 under cold conditions (data not shown). The genotypes Gig-14, Sac-12, Sac-11 had significantly higher rates of Rd than Sin-7 under warm conditions. Sac-9 showed a significantly higher Rd rate than Tin-1 at 14°C (data not shown). For Sin-7 and Tin-1, both AQY and Rd remained relatively stable when the temperature was lowered from 24 to 14°C. The rate of Rd accounted for 5·2–9·5 % of gross photosynthesis under warm conditions and 6·5–8·9 % under cold conditions, thus contributing only a minor share to gross photosynthesis. No significant difference was observed between genotypes concerning the light compensation point (LCP), with a mean value of 51 mol mol–1 at 24°C and 33 mol mol–1 at 14°C (data not shown).

The low growth temperature led to a significant reduction in Asat (Fig. 1B). Sac-12 had the highest Asat rate of the tested genotypes at 24°C at 29·5 μmol m–2 s–1, which fell to 11·9 μmol m–2 s–1 at 14°C. However, at this temperature Sac-11 had a significantly higher Asat value (13·8 μmol m–2 s–1) than Sin-2, Sin-5, Sin-7 and Tin-1 (Fig. 1B). The largest reduction in Asat with the temperature decrease was observed for Sin-5 (67·2 %) with a fall from 28·6 to 9·3 μmol m–2 s–1. Sin-4, Sac-9, Sac-11 and Gig-13 had the smallest reduction in Asat (Fig. 1B). At 1500 μmol m–2 s–1 PAR, Ci increased for all genotypes when the temperature was decreased (data not shown), and Ls only accounted for 4–5 % of total limitation at 14°C, which indicates that non-stomatal limitation was the main reason for photosynthetic reduction.

Sac-10 and Sac-12 had significantly higher CE values than Gig-13, Sin-7, Tin-1, Sac-9 and Sin-2 at 24°C (Fig. 1C), and Sac-11 and Sac-12 had significantly higher Vmax values than Sin-2, Tin-1 and Sin-7. When the temperature was decreased, the CE of PEPc decreased for all the genotypes, except for Gig-13 (0·14 at 24°C and 0·13 at 14°C). The maximum photosynthetic rate at saturated light and CO2 conditions was decreased by cold conditions in all genotypes (Fig. 1D). At 14°C, Sac-11 showed the highest values of CE and Vmax, but those of Gig-13 were not significantly lower (Fig. 1C, D).

Dark-adapted chlorophyll fluorescence (Fv/Fm) ranged from 0·52 to 0·75 at 14°C. Tin-1 showed significantly lower values than the other genotypes, except Sin-4, Sac-10, Sac-12 and Gig-14 (Supplementary Data Table S1).

Development of A1000 in cold conditions and recovery of A1000 after 23 d in cold conditions

A sharp reduction in the net photosynthetic rate was observed at 1000 μmol m–2 s–1 (A1000) for all genotypes 7 h after the temperature had been lowered from 24 to 14°C (Fig. 2). When compared with Day 1 at 14°C, A1000 was further reduced by < 20 % on Day 2 for most genotypes (Fig. 2). After Day 4, A1000 stabilized at a reduced level for most genotypes, while a few genotypes continued to decrease. For genotypes Tin-1, Sin-4, Sac-11 and Gig-1, A1000 fell by < 20 % after the initial drop measured after 7 h. The A1000 rates were significantly higher for Sac-11 and Gig-13 for most of the period (Fig. 2).

F

Development of the net photosynthetic rate at a PAR of 1000 μmol m–2 s–1 (A1000) of all genotypes before and after transfer of plants from warm conditions (24 °C) to cold conditions (14 °C), and 1 d after the temperature was returned to 24 °C. Error bars represent the s.e. (n = 3). Significant differences between genotypes on Day 0 before the temperature was lowered and on Day 24 after the temperature was increased were examined using the Student’s t-test. Values with the same letter are not significantly different at the P = 0·05 level.

On Day 24, 1 d after the temperature had again been raised, A1000 increased for all genotypes. However, Sin-2, Sin-3, Sin-5, Sin-6, Sin-H8, Sac-10 and Sac-12 had significantly lower net assimilation rates than before the temperature drop, while the rest of the genotypes had values that were not significantly different from their pre-chilling values (Fig. 2). The Tin-1 and Sin-7 genotypes had the lowest values for all measurement dates.

Specific leaf area

Overall, the M. sacchariflorus species had higher SLAs than M. sinensis and M. tinctorius (Fig. 3). The Sac-11 genotype had the highest SLA at 30·7 m2 kg–1 and Sin-2 the lowest at 21·3 m2 kg–1. Sac-11, Sac-12 and Sac-9 had significantly higher SLAs than Tin-1, Gig-14 and all the sinensis genotypes, except for Sin-4 (Fig. 3). Asat showed a positive linear correlation with SLA at the low temperature (Fig. 4), and the relationship between SLA and Amass was also linear and positive, with an R2 value of 0·68 (Supplementary Data Fig. S2).

F

Specific leaf area (SLA) of the selected genotypes. Error bars represent the s.e. (n = 6). Values with the same letter are not significantly different at the P = 0·05 level.

F

The linear relationship between the net photosynthetic rate at a PAR of 1500 μmol m–2 s–1 (A1500) and specific leaf area (SLA) under cold conditions. Each point is the mean of SLA from each plant and its correlated A1500 from the same replication.

Relationship between daily shoot growth rate and A1000 under cold conditions

Of the genotypes selected for the detailed study, Sac-12 and Sac-11 had the highest shoot growth rates measured at low temperatures under field conditions in Denmark in 2012 (data not shown). A positive linear relationship with an R2 value of 0·502 (P = 0·01) was found between daily shoot growth rate in the field and average values of A1000 measured on M. sinensis, M. sacchariflorus and M. × giganteus (Tin-1 was excluded from this analysis as it did not fit the same relationship as the other species) 3 d after the temperature decrease in the growth chambers in 2013 (Fig. 5).

F

Relationship between shoot growth rate measured under cool 9·8/15·5 °C (mean/maximum) conditions in the field in 2012 and mean photosynthetic rate at 1000 μmol m–2 s–1 (mean A1000) measured for 13 genotypes (Tin-1 is excluded in the relationship but shown on the graph) 3 d after the temperature was decreased from 24 to 14 °C in the climate chamber in 2013.

DISCUSSION

Cold tolerance under cold conditions

We found that the photosynthetic parameters of M. × giganteus ‘Hornum’ (Gig-13) were very competitive compared with those of other miscanthus genotypes selected from the germplasm collection of 166 individuals, and it showed the lowest reduction in Asat when the temperature was decreased from 24 to 14°C. This is in agreement with previous findings that, of the C4 grasses, the widely investigated triploid M. × giganteus is exceptionally productive in cold climates and has the ability to maintain high photosynthetic activity at 14°C (Naidu ). However, one M. sacchariflorus (genotype Sac-11) showed slightly higher values of AQY, Asat, CE and Vmax than Gig-13, even though it was not significantly different at 14°C. At 24°C, two M. sacchariflorus genotypes (Sac-10 and Sac-12) showed significantly higher CE values than M. × giganteus. Głowacka ) screened the plants of the same germplasm collection with the least frost damage in the field with a different procedure (chlorophyll fluorescence). Sac-10 was the only genotype that could be compared directly between our study and that of Głowacka ) (where it was named Msa ‘44/1’). We measured the same level of A1500 in sac-10 in the two studies. However, Głowacka ) found two M. sacchariflorus genotypes that had significantly higher photosynthetic capacity than M. × giganteus at 15°C, with a higher A1500 level than any genotype selected in our study. This may indicate that the two-stage procedure (first a field frost event and then Fv/Fm screening of the surviving genotypes) is a particularly efficient method for field screening for cold tolerance, which, however, is difficult to replicate on another population. The linear part of the light response curve reflects the maximum efficiency of light harvesting, and a reduction in the AQY is associated with photoinhibition (Long ). This can be due to increased non-photochemical energy dissipation as a photoprotective mechanism or to photodamage to the reaction centre of PSII (Osmond, 1994).

The genotypes Sac-9 and Sac-11 had the highest AQY values at 14°C and the lowest reduction in AQY when the temperature was decreased from 24 to 14°C. In northern Europe, saturating light is difficult to reach in the field, and light-limited photosynthesis contributes significantly to canopy carbon gain (Baker ). Therefore, a high AQY has greater influence on photosynthetic capacity than a high Asat under light-limited conditions and/or within a closed canopy (Long, 1983). Even though values of AQY in Sac-9 and Sac-11 were not significantly higher than the values of Gig-13, the Fv/Fm values in Sac-9 and Sac-11 were slightly higher (not significant) than that of Gig-13. Fv/Fm is a measure of the maximum quantum efficiency of PSII, and a decrease is also indicative of photoinhibition (Long ; Maxwell and Johnson, 2000). Furthermore, the slight decrease in photosynthetic performance of Sac-9 and Sac-11 as well as of Gig-13 was reversible within a day, as confirmed by the recovery of A1000 after increasing the temperature from 14 to 24°C (Fig. 2). All these results indicate that it is possible to find suitable genotypes for breeding of new improved triploid M. × giganteus varieties, and the two tetraploid M. sacchariflorus genotypes seem to be good candidates for achieving high productivity in light-limited environments at low temperature.

The lower photosynthetic activity at low temperatures has previously been found in M. × giganteus not to be due to stomatal limitations (Naidu and Long, 2004; Farage ), and our results of the Ls of the 11 genotypes confirmed this.

The key enzymes associated with carbon fixation, namely PPDK (Naidu , Wang ) and Rubisco (Spence, 2012), are sensitive to low-temperature damage in M. × giganteus and are thus central to maintaining high photosynthetic activity at low temperature. Vmax is controlled by a variety of processes, mainly PEP regeneration via the key enzymes mentioned above (Collatz ; von Caemmerer, 2000). Sac-11 and Gig-13, with their significantly higher Asat, both had higher Fv/Fm values and a lower reduction in AQY than several other genotypes at 14°C. The significantly higher Vmax values in Sac-11 and Gig-13 than in several other genotypes are indicative of relatively high amounts or activity of the key enzymes involved in carbon fixation at the low temperature.

At low Ci, the CE of PEPc controls the photosynthetic rate (Collatz ; von Caemmerer, 2000). There was a slight but insignificant decrease in CE values in Gig-13 when the temperature was decreased, which is similar to previous findings (Naidu and Long, 2004). Sac-11 had slightly higher CE values at both temperatures than Gig-13 (but not significant at 14°C), whereas in all the other genotypes CE decreased significantly when the temperature was lowered from 24 to 14°C. This suggests that the amount and activity of PEPc might be a limiting factor at low temperatures for most miscanthus genotypes, with M. × giganteus and Sac-11 as exceptions.

It is interesting to note that most M. sacchariflorus genotypes had higher Asat values than the M. sinensis genotypes at 14°C (Fig. 1B). This is in agreement with previous findings that M. sacchariflorus typically has a higher tolerance to chilling than M. sinensis (Głowacka ). This is also evident from the natural geographical distribution of M. sacchariflorus which extends further north in Asia than M. sinensis (Clifton-Brown ).

Development of A1000 under cold conditions and recovery after re-increase of the temperatures

The development patterns differed within the miscanthus genotypes and they indicate that there are different reactions during cold temperature exposure.

The reduction in A1000 for Gig-13 and Sac-11 ranged from 12 to 20 % compared with the value on Day 1 during the 18 d of cold treatment. Previous studies detected a drop in A1000 in M. × giganteus immediately after the change from 24 to 14°C, but after 3 d a recovery was observed and the reduction was only 12 % on the ninth day (Wang ). However, in most other genotypes, we observed a continued decrease in A1000 after Day 4, which could be due to a further decrease in the amount and activity of PPDK and/or Rubisco (as observed in maize during prolonged cold conditions; Wang ), but perhaps also caused by an accumulation of xanthophyll (violaxanthin, antherananthin and zeaxanthin) cycle pigments, which apparently is a cold tolerance strategy for C4 species whereby they dissipate surplus energy (Kubien and Sage, 2004).

A previous study has shown that the rapid recovery of M. × giganteus photosynthesis at 14/11°C due to the temperature response of light-saturated photosynthesis is very similar for M. × giganteus grown in warm (25/20°C) and cold (14/11°C) conditions (Naidu ). In our study, genotypes Sin-2, Sin-3, Sin-5, Sin-6, Sin-H8, Sac-10 and Sac-12 did not have the capacity to recover their photosynthesis rapidly after a long cold period. Only the genotype Tin-1 showed a significantly lower Fv/Fm than the other genotypes, indicating a loss of PSII capacity, while Sin-3, Sin-5, Sin-6 and Sin-H8 showed relatively high Fv/Fm values at 14°C (Table S1), so the failure of a quick recovery may be explained rather by the high xanthophyll levels that continued to dissipate absorbed light and/or to reduced contents of PPDK and/or Rubisco. For the genotypes Sac-11and Gig-13, A1000 recovered to a similar level to that before the cold treatment after a 24 h exposure to the warm temperature, which indicates that a temperature of 14°C did not cause irreversible photoinhibition in PSII, and that these two genotypes are more suited to variable, cool, temperate conditions because of their high net photosynthetic rate under both light-limited and light-saturated conditions. Such reversible dynamic photoinhibition is apparently a strategy by which C4 species may tolerate the low temperatures that prevail in temperate climate zones (Kubien and Sage, 2004).

Correlation between photosynthetic capacity and two morphological characters: SLA and shoot growth rate

In our study we observed that genotypes with a high SLA were capable of attaining even higher photosynthetic efficiency than genotypes with a low SLA under cold conditions (Fig. 4). This indicates that within the Miscanthus genus there is significant variation in the photosynthetic return of investment in leaf biomass, which is an important trait for high productivity early in the season. The SLA of miscanthus has in previous studies been significantly higher than for switchgrass (Panicum virgatum) and had significantly higher A values than for switchgrass (Dohleman ). This shows that having thinner leaves is not necessarily at the expense of a low A. The significantly higher SLA in Sac-11 and Gig-13 combined with their high cold tolerance hints that these two genotypes have the ability to develop a large leaf canopy quickly during the early spring in maritime climates.

The relationship between SLA and net photosynthesis per unit leaf area has been rarely investigated (Reich ). In our study, a positive correlation between SLA and A was confirmed. However, whether the leaf thickness is an adaptation to the cold conditions or an inherent characteristic within each genotype is not clear, and this should be further investigated.

Of the 14 genotypes tested in this study, we found significantly higher shoot growth rates for genotypes Sac-11 and Gig-13 in the field than for most other genotypes investigated in detail in climate chambers (data not shown). The genotypes Sac-11 and Gig-13 also had relatively high A1000 values during the first 3 d at 14°C. The good linear relationship between shoot growth rate and photosynthesis of the investigated genotypes (except M. tinctorius) provided another simple procedure for selecting genotypes from large-scale field collections of M. sinensis, M. sacchariflorus and M. × giganteus for further detailed investigation. The Tin-1 genotype had the lowest photosynthetic performance under both temperature treatments in this study. This result was also achieved by Głowacka ), where M. tinctorius had the lowest cold tolerance with the lowest photosynthetic rate compared with M. sinensis, M. sacchariflorus and M. × giganteus. The M. tinctorius genotypes grow quite well in the field in Denmark, and this species seems to have a different physiological response to cope with cold temperatures compared with the other miscanthus species.

Conclusion

No severe cold stress was detected at 14°C in any of the genotypes selected. Although not significant, the genotype Sac-11 seemed to perform slightly better in terms of cold tolerance and quick recovery than the hitherto most widely used triploid M. × giganteus (Gig-13), plus its high SLA and shoot growth rate gave a fast leaf canopy development in the early growing season. Therefore, it may be used for large plantations together with M. × giganteus or for breeding new interspecies hybrids with improved traits for temperate climates. Sac-12 performed significantly better than M. × giganteus (Gig-13) at 24°C and high light intensities, and may be used for breeding improved varieties for warm climates. Interestingly, we found that two phenotypic variables that are relatively easy to measure, i.e. daily shoot growth rate and SLA, correlated well with leaf photosynthesis. These variables may be used for the selection of individuals from a large gene pool for more detailed analysis or for screening new hybrids. The ability of some genotypes to fix significantly more carbon per gram of leaf biomass than others and the discovery of a generally positive correlation between SLA and Amax were surprising compared with our hypothesis that Amax would remain unchanged at increased SLA. This leaf variable will be important for further studies and may be used for the selection of ideotype biomass crops with the most efficient solar capture and carbon fixation. Another interesting area for future investigation would be whether miscanthus genotypes become more disparate in their photosynthetic capacity at even lower temperatures.

SUPPLEMENTARY DATA

Supplementary data are available online at www.aob.oxfordjournals.org and consist of the following. Table S1: maximum quantum yield of PSII (Fv/Fm) measured at 14°C. Figure S1: correlation between daily shoot growth rate and net photosynthesis in 37 miscanthus genotypes measured in field trials in Denmark in 2012 during a warm and a cold period. Figure S2: the linear relationship between Amass and SLA in cold conditions.