Planta (2001) 213: 575±585 DOI 10.1007/s004250100522
O R I GI N A L A R T IC L E
A.G. Ivanov á P.V. Sane á Y. Zeinalov á G. Malmberg P. GardestroÈm á N.P.A. Huner á G. OÈquist
Photosynthetic electron transport adjustments in overwintering Scots pine (Pinus sylvestris L.) Received: 8 October 2000 / Accepted: 13 December 2000 / Published online: 14 March 2001 Ó Springer-Verlag 2001
Abstract As shown before [C. Ottander et al. (1995) Planta 197:176±183], there is a severe inhibition of the photosystem (PS) II photochemical eciency of Scots pine (Pinus sylvestris L.) during the winter. In contrast, the in vivo PSI photochemistry is less inhibited during winter as shown by in vivo measurements of DA820/A820 (P700+). There was also an enhanced cyclic electron transfer around PSI in winter-stressed needles as indicated by 4-fold faster reduction kinetics of P700+. The dierential functional stability of PSII and PSI was accompanied by a 3.7-fold higher intersystem electron pool size, and a 5-fold increase in the stromal electron pool available for P700+ reduction. There was also a strong reduction of the QB band in the thermoluminescence glow curve and markedly slower QA± re-oxidation in needles of winter pine, indicating an inhibition of electron transfer between QA and QB. The data presented indicate that the plastoquinone pool is largely reduced in winter pine, and that this reduced state is likely to be of metabolic rather than photochemical origin. The retention of PSI photochemistry, and the suggested metabolic reduction of the plastoquinone pool in winter stressed needles of Scots pine are discussed in terms of the need for enhanced photoprotection of the needles during the winter and the role of metabolically supplied energy for the recovery of photosynthesis from winter stress in evergreens.
A.G. Ivanov á P.V. Sane á G. Malmberg á P. GardestroÈm G. OÈquist (&) UPSC, Department of Plant Physiology, University of UmeaÊ, UmeaÊ 90187, Sweden E-mail:
[email protected] Fax: +46-90-7866676 A.G. Ivanov á N.P.A. Huner Department of Plant Sciences, University of Western Ontario, London, Ontario, N6A 5B7, Canada Y. Zeinalov Institute of Plant Physiology, Bulgarian Academy of Sciences, 1113 So®a, Bulgaria
Keywords Electron transport á P700 á Photosynthesis á Pinus (winter stress) á Thermoluminescence á Winter stress Abbreviations AL: actinic light á Chl: chlorophyll á CP: cold-acclimated pine á CS-SP: cold-stressed summer pine á DCMU: 3-(3¢,4¢-dichlorophenyl)-1,1-dimethylurea á Fo: minimum yield of chlorophyll ¯uorescence at open PSII centers in dark-adapted needles á Fv/Fm: maximal photochemical eciency of PSII á FR: far red light á MT: multiple-turnover ¯ash of actinic white light á P700: reaction-center pigment of PSI á P700+: oxidized form of the reaction center of PSI á PQ: plastoquinone á SP: summer pine á ST: single-turnover ¯ash of actinic white light á TL: thermoluminescence á WP: winter pine
Introduction Earlier studies have documented that evergreen conifers of cold climates experience great seasonal changes in photosynthetic activities, exhibiting a gradual decline during late summer and autumn, a strong inhibition during the winter, and a rapid recovery in early spring (OÈquist et al., in press). These seasonally induced changes in the rate of net photosynthesis are well correlated with electron transport studies performed in chloroplast thylakoids isolated from Scots pine (Pinus sylvestris L.) and Norway spruce (Picea abies L. Karst.) (Martin et al. 1978; Senser and Beck 1978). The autumnand winter-induced inhibition of PSII photochemistry of Scots pine in vivo has been reported to be due to lowtemperature-induced photoinhibition of PSII (Strand and OÈquist 1985a) accompanied by a loss of the reaction-center D1 protein (Ottander et al. 1995). Plastoquinone (PQ) has also been recognized as a primary site of winter-induced inhibition of electron transport (OÈquist and Martin 1980). However, studies by Vogg et al. (1998a, b) have revealed that the shortened
576
photoperiod during the autumn also induces partial inhibition of PSII in Scots pine. This observation is supported by the partial loss of the D1 protein of PSII prior to the onset of frost in the autumn (Ottander et al. 1995). In contrast to PSII, no systematic in vivo studies have been performed in conifers to elucidate the seasonal changes in structure and function of PSI. Earlier in vitro studies have demonstrated: (i) increased Chl/P700 ratios of Scots pine during the winter (Martin et al. 1978), (ii) increased amounts of Cyt b-563 in cold-acclimated Scots pine (OÈquist and Hellgren 1976), and (iii) an increased capacity for phenazine methosulfate-mediated PSI cyclic electron transport in chloroplasts isolated from winterstressed Norway spruce (Senser and Beck 1977). Recently, it has also been hypothesized that a substantial preservation of PSI function throughout the winter in evergreen conifers like Scots pine might be of vital importance to maintain a modi®ed yet functional photosynthetic apparatus during winter conditions (OÈquist et al., in press). In addition, this might have a signi®cant impact during the early stages of photosynthetic recovery during spring. This notion is further supported by the observations that cold-acclimated plants exhibit an increased capacity for PSI electron transport, as reported for isolated thylakoids from overwintering periwinkle (Huner et al. 1988) and winter rye (Huner and Reynolds 1989). In addition, cold-acclimated barley exhibits an increased electron donor pool size to PSI, and an increased resistance to photoinactivation of PSI (Gray et al. 1998; Ivanov et al. 1998). It has been suggested that PSI may be protected from photoinactivation by a cyclic electron ¯ow around PSI (Havaux 1996). In this report, we examine the in vivo function of PSI in winter-stressed Scots pine, as well as during the recovery of photosynthesis from winter stress. We show that, in contrast to the severe inhibition of PSII, PSIdependent electron transport is much less sensitive to winter stress and exhibits an increased capacity for cyclic electron transport. Furthermore, winter stress causes a reduction of the intersystem electron transport chain, which appears to be of non-photochemical origin.
Materials and methods Plant material Current-year needles from exposed branches of a 35-year-old Scots pine (Pinus sylvestris L.), growing on an open stand on the campus of UmeaÊ university (63°50¢N, 20°20¢E), were collected regularly at midday during the winter (January±March) or summer (July± August) over a 2-year period (1998±1999). In February±March, branches were brought indoors and the recovery of photosynthesis from winter inhibition was followed under standardized laboratory conditions. The shoots were thawed at 4 °C in darkness, re-cut and placed in water at 20 °C and a photosynthetic photon ¯ux density of 100 lmol photons m±2 s±1 for recovery (Ottander et al. 1995). Seedlings of Scots pine obtained from the Forest Tree Nursery of PiparboÈle, UmeaÊ, Sweden were grown and cold-acclimated in controlled-environment chambers under the following standardized conditions: photoperiod 8 h (200 lmol m±2 s±1) and 3 weeks at a
day/night temperature regime of 15/10 °C followed by 5±10 weeks at a temperature regime of 5/5 °C but with a photon ¯ux density decreased to 50 lmol m±2 s±1 to avoid photoinhibition (Strand and OÈquist 1985a). Current secondary needles were used. Chlorophyll ¯uorescence Maximal photochemical eciency of PSII (Fv/Fm), and the kinetics of the ¯uorescence transient from Fo to Fm were measured in Scots pine needles after 30 min dark adaptation at room temperature using a Plant Stress Meter (PSM Chlorophyll Fluorimeter, Biomonitor S.C.I. AB, UmeaÊ, Sweden). The photon ¯ux density of the actinic light was 400 lmol m±2 s±1 and the time of excitation was 5 s. The chlorophyll ¯uorescence nomenclature of van Kooten and Snel (1990) was used for the calculation of maximal photochemical eciency of PSII. Measurements of P700 Unless speci®ed otherwise, the redox state of P700 was determined in vivo under ambient O2 and CO2 conditions using a PAM-101 modulated ¯uorimeter equipped with an ED-800T emitter-detector and PAM-102 units, following the procedure of Schreiber et al. (1988) as described in detail by Ivanov et al. (1998). Far red light (FR; kmax=715 nm, 10 W m±2, Schott ®lter RG 715) was provided by an FL-101 light source. The FR photon ¯ux density was measured by a Techtum spectroradiometer (Model QSM-2500; Techtum Instrument, UmeaÊ, Sweden). Multiple-turnover (MT ± 50 ms) and single-turnover (ST ± half-peak width 14 ls) saturating ¯ashes of white light were applied with XMT-103 and XST-103 power/control units, respectively. The redox state of P700 was evaluated as the absorbance change around 820 nm (DA820/A820) in a custom-designed cuvette. The measurements were performed at 20 °C for summer pine (SP) or 5 °C for winter pine (WP) and coldacclimated pine (CP) needles, respectively. The temperature was maintained by a refrigerated circulating-water bath (RMT6; Lauda Dr. R. Wobser, Lauda-KoÈnigshofen, Germany). The signals were recorded using an oscilloscope card (PC-SCOPE T6420, Version 2.43´; Intelligente Messtechnik, Backnang, Germany) installed in an IBM-compatible personal computer. The complementary areas between the oxidation curve of P700 after ST and MT excitation, and the stationary level of P700+ under FR represent the ST and MT areas, respectively, which were used to estimate the intersystem electron donor pool size for P700+ (Asada et al. 1992): e =P700 MT area/ST area: For estimation of the functional pool size of electrons that can be donated to P700+ from stromal sources, the complementary areas between the stationary level of P700+ under illumination with FR light and the oxidation curves of P700 by FR after an MT ¯ash (MT area), and after the illumination with actinic light (AL; 150 lmol photons m±2 s±1; AL area) were determined as described by Asada et al. (1992) using the following expression: Stromal e pool/P700
MT area/ST area) f(AL area/MT area)
1g:
Chlorophyll a ¯uorescence decays The re-oxidation kinetics of QA were measured as the decay of chlorophyll a ¯uorescence using a PAM ¯uorimeter as described in Schreiber (1986). Saturating ST ¯ashes obtained from an XST 103 xenon discharge lamp connected to a PAM 103 unit (Heinz Walz, Eeltrich, Germany), were used to convert all QA to QA . The lowintensity measuring beam, provided by a pulsed light-emitting diode (660 nm), was modulated at 100 kHz. The variable ¯uorescence decay, re¯ecting the re-oxidation of QA , was detected at 20-ls resolution. The signals were recorded using the oscilloscope card mentioned above. Data from at least 10 recordings were averaged. In order to avoid possible gating artifacts, the ®rst data point used
577 for analysis was at 150 ls after the actinic ¯ash reached its maximum intensity. Final curve ®tting was performed by a non-linear data analysis using a Microcal Origin Version 6.0 software package (Microcal Software Inc., Northampton, Mass., USA). Assay for ATP and ADP Preparation of needle extracts for determination of ATP and ADP contents was performed as described in Stitt et al. (1989). ATP and ADP were determined by the ®re¯y luciferase method (GardestroÈm and Wigge 1988) in 70 mM Tris (pH 7.75), 1.4 mM EDTA, 0.2 mM phosphoenolpyruvate (PEP) using an ATPmonitoring kit (BioThema AB, DalaroÈ, Sweden) and an LKB 1250 luminometer. A 7-ll sample was added to a ®nal assay volume of 285 ll. ADP was converted to ATP using pyruvate kinase (Serva). Each measurement was calibrated with an addition of an ATP standard. Thermoluminescence measurements Thermoluminescence (TL) measurements of intact needles were performed on a personal-computer-based TL data acquisition and analysis system similar to that described earlier (Zeinalov and Maslenkova 1996). A photomultiplier tube (R943±02; Hamamatsu Photonics K.K., Shizuoka-ken, Japan) equipped with a photomultiplier power supply (Model PS-302; EG&G Electro Optics, Salem, Mass., USA) and a preampli®er (Model C1556±03) was used as a radiation measuring set. The temperature was monitored by a microthermoresistor (FMS 2101; Umwelt Sensor Technik, Germany) placed on the surface of the sample holder and a bridge ampli®er. The output signal from the photomultiplier, which is proportional to the TL photon emission, and the output signal from the bridge ampli®er, which is proportional to the temperature of the sample holder, were connected to two dierent channels of an A/D converter data acquisition card and an IBM-compatible personal computer. Decomposition analysis of the TL glow curves in terms of Gaussian bands was carried out by a non-linear, leastsquares algorithm that minimises the chi-square function, using the Microcal software package mentioned above. Illumination treatment of the samples during cooling was performed with continuous white light as described earlier (Sane et al. 1983). The nomenclature of Vass and Govindjee (1996) was used for characterisation of the TL glow peaks. Statistical analysis Dierences between the variants were assessed by one-way analysis of variance (ANOVA) with all pairwise multiple comparisons using the Student-Newman-Keuls method. The SigmaStat, Version 1 (Jandel Corporation, San Rafael, Calif., USA) software package was used for all statistical calculations.
Results Photochemistry of PSII The recovery of photosynthesis from winter stress in Scots pine needles was measured by following the recovery of the maximum photochemical eciency of PSII expressed as Fv/Fm after branches were brought into the laboratory and kept at 20 °C and 100 lmol m±2 s±1 (Fig. 1A). Consistent with previous reports (Ottander et al. 1995), the maximum photochemical eciency of outdoor pine needles during February (dark-adapted and measured at 5 °C) was severely reduced, exhibiting
Fig. 1 The time course of recovery of PSII photochemical eciency from winter stress in Scots pine (Pinus sylvestris) needles measured as Fv/Fm (A) and half-time rise of chlorophyll ¯orescence from F0 to Fm (t1/2FLUO) (B). The shoots were kept at 20 °C and light irradiance of 100 lmol photons m±2 s±1
Fv/Fm values of 0.310.01, as compared to the values of Fv/Fm registered during the summer (0.820.01). After the initial lag phase of 1±3 h, recovery of Fv/Fm was rapid and after 26 h the photochemical eciency had recovered to 87.6% of Fv/Fm values registered in control summer needles. This increase in the maximal photochemical eciency of PSII was accompanied by a concomitant increase in the t1/2 for the chlorophyll (Chl) ¯uorescence rise from Fo to Fm (Fig. 2B); the t1/2 values of winter-stressed and summer pine were 2 and 42 ms, respectively. After 26 h of recovery, the t1/2 of winterstressed needles had recovered to 72% of values exhibited by summer needles. Photochemistry of PSI Figure 2 and the data summarized in Table 1 show that the relative amount of FR-oxidized P700+ (DA820/A820) was 40% lower in winter-stressed pine (WP) than in summer pine (SP), while in cold-acclimated pine (CP) the P700+ signal was only 20% lower than for SP. Exposure of SP needles to either a low measuring temperature (SP at 5 °C) or to a short-term cold stress (CS-SP, 18 h at 5 °C), had minimal eects on the P700 oxidation (Table 1). This indicates that the lower level of P700 oxidation of WP could not be due to the low
578
Fig. 2 In vivo measurements of the redox state of P700 in Scots pine needles collected during the summer (A), during the winter (B) or fully developed at cold (5 °C) temperature (C). The measurements were performed at 20 °C (A) and at 5 °C (B, C). After reaching a steady-state level of P700+ by FR illumination, ST and MT pulses of white light were applied
measuring temperature, but is rather associated with alterations at the level of PSI photochemistry induced by the long-term winter stress conditions. This is consistent with the fact that long-term acclimation of pine seedlings to low growth temperature (5 °C) alone under controlled environmental conditions (CP needles) caused a much lower decrease in P700+ than seen in WP needles Table 1 Steady-state oxidation of P700 (DA820/A820), intersystem electron pool size (e±/P700), stromal electron pool size and halftimes of P700+ reduction (t1/2P700red) in needles of Scots pine (Pinus sylvestris) during the summer (SP), winter (WP) and in cold-acclimated pine seedlings (CP). SP was measured at 20 °C and 5 °C. In some experiments SP was subjected to cold stress at 5 °C for 18 h
(Table 1, Fig. 2C). Monitoring the changes in DA820/ A820 during the recovery of photosynthesis, we observed that the apparent oxidation state of P700 increased in a manner similar to that observed for the recovery of Fv/ Fm (compare Fig. 1A and Fig. 3A). After 26 h recovery at room temperature and moderate light, the WP needles exhibited values of DA820/A820 close to those observed in SP needles (Fig. 3A). However, the 40% lower DA820/A820 seen under FR excitation of WP needles (Table 1) appears to underestimate the potential for PSI photochemistry in WP. Normally, when white actinic light is added to FRexposed leaves, there is a reduction in the P700+ signal due to the induced electron ¯ow from PSII (Schreiber et al. 1988; Ivanov et al. 1998). This eect was clearly seen in SP needles where DA820/A820 decreased by 30% when white actinic light (AL; 150 lmol m±2 s±1) was added in the presence or absence of O2 (Fig. 4A, B). In contrast, AL excitation of WP caused an additional 40% increase of DA820/A820 (Fig. 4C). Apparently, FR light is unable to oxidize all PSI in WP. Furthermore, the FRinduced steady-state DA820/A820 signal increased in the absence of O2, and excitation by AL caused a transient P700+ reduction followed by a further oxidation of P700 (Fig. 4D). From these observations we estimate that winter stress of Scots pine only causes approximately a 20% reduction of the potential for PSI photochemistry as compared with 60% inhibition observed for PSII (Fig. 1). We also measured the kinetics of the dark re-reduction of P700+ after turning o the FR light (Fig. 2; Table 1, P700red). High rates of dark re-reduction of P700+ are ascribed primarily to the extent of cyclic electron ¯ow around PSI (Maxwell and Biggins 1976; Ravenel et al. 1994). We found that the rate of P700+ re-reduction in darkness was much faster in WP than in SP and CP needles. Furthermore, shifting SP plants to 5 °C for 18 h (CS-SP), or simply exposing SP needles to a measuring temperature of 5 °C, resulted in only minor changes in the rate of P700+ re-reduction. This indicates that the lower t1/2 values of P700+ re-reduction in WP needles could not be due to the lower measuring temperature per se but, rather, may re¯ect an increased capacity for cyclic electron ¯ow around PSI as a consequence of exposure to winter stress. Furthermore, the (CS-SP). For WP, CP and CS-SP, DA820/A820 was measured at 5 °C. Mean values SE were calculated from 6±25 measurements in 3±9 independent experiments. Statistically signi®cant dierences between SP at 20 °C and CP and WP are marked by asterisks (* P800 53.7 29.6 16.7 0.097
n.d 4,580228 >800 n.d. 80.3 16.7 0.029
and C luminescence (Table 3). The contribution of the main A, B and C bands were 27.6%, 22.8% and 28.3%, respectively. In the presence of 3-(3¢,4¢ dichlorophenyl)-
581 Fig. 6 Thermoluminescence glow curves and mathematical resolution of glow curves in Gaussian sub-bands in Scots pine needles collected during summer (A, B) and in winter (C, D). A, C Control needles; B, D needles treated with DCMU (50 lM). Experimental curves represent averages of 3±5 scans
1,1-dimethylurea (DCMU), the overall yield of TL emission decreased and the major TL band was centered at ±10 °C, representing QA recombining with the S states. A distinct dierence was observed in the TL emission curve of WP needles (Fig. 6C). Apart from the much lower yield of TL emission, the WP glow curve was best ®tted with only three TL bands with the following peak positions: 28.4 °C (Zv band), 1.6 °C (Q or A band) and 29.1 °C (B band). It is important to note that the band centered around ±10 °C is missing and treatment with
DCMU to block the electron transfer to QB did not cause any signi®cant changes in either TL yield, peak positions or their relative contribution to the overall glow curve (Fig. 6D). The relatively strong Zv band in WP accounting for almost 60% of the total luminescence is indicative of a high probability of recombination of primary donors with primary acceptors of PSII. The presence of a 0 °C band in place of a ±10 °C band in WP and its persistence even after DCMU addition suggests that the A band has shifted to a higher temperature. Over 93% of the TL in WP is accounted for by
Table 3 Emission temperatures (T °C) and contribution of dierent glow peaks to the total thermoluminescence in Scots pine needles during the summer (SP) and winter (WP). The contribution of glow peaks represented by characteristic Gaussian sub-bands
was estimated by decomposition of the total glow curves and the data are represented as a percentage of the total area (A, %). Mean values SE were calculated from 3±5 independent experiments. n.d. Not detectable
Peak
Zv band (P+ 680 QA ) A band (S3QA ) Q band (S2QA ) B1 band (S2QB ) B2 band (S3QB ) C band (TyrD+QA )
SP
WP
T °C
A (%)
T °C
A (%)
±32.00.1 ±10.70.0 4.10.1 30.20.1 43.40.4 61.30.4
14.30.1 27.60.2 3.90.1 3.50.3 22.82.2 28.32.8
±28.40.1 n.d. 1.60.1 29.10.1 n.d. n.d.
57.80.1 n.d. 35.30.1 6.80.1 n.d. n.d.
582
low-temperature peaks, re¯ecting a preferred back reaction of QA with primary donors and a low probability of transfer of electrons from QA to QB. Adenylate content The total adenylate content (ATP + ADP) in SP needles was about 66 nmol (mg Chl)±1 and the estimated ATP/ ADP ratio was 7.4 (Table 4). CP needles exhibited slightly higher adenylate content and ATP/ADP ratio (122%) as compared to SP controls. This appeared to be due to the elevated amounts of ATP (139%). As distinct from CP, ATP + ADP content observed in needles from WP was 3-fold higher than in SP. Interestingly, however, the relative increase of ADP was considerably higher (6.3-fold) than that of ATP (2.5-fold). This resulted in a much lower (35.7%) ATP/ADP ratio in WP needles compared to that in SP as well as in CP needles.
Discussion In vivo measurements of the FR-induced absorbance decrease at 820 nm revealed a 40% inhibition of P700 photooxidation (Fig. 2A, B; Table 1) as compared to 60% inhibition of PSII photochemistry (Fig. 1A). However, if one considers that actinic white light can further oxidize P700 in WP needles (Fig. 4C), the reduction of PSI photochemistry is only in the range of 20% during the winter. Clearly, the PSI reaction centers are much better preserved during the winter than the PSII reaction centers. This ®nding is in contrast to earlier reports based on electron transport studies of chloroplast thylakoids isolated from Scots pine showing almost similar degrees of winter stress inhibition of PSI and PSII (Martin et al. 1978), a 2- to 4-fold increase of the Chl/P700 ratio (Martin et al. 1978; OÈquist and Martin 1980), and a relatively strong reduction in the amount of the detergent-solubilized and electrophoretically separated PSI chlorophyll-protein complex during the winter (OÈquist et al. 1978). These comparisons between PSI detection under in vivo and in vitro conditions suggest organizational and functional modi®cations of PSI in WP making it more vulnerable to damage during isolation. Consistent with such an assumption is that WP needles have an increased capacity for cyclic electron
transfer as based on the increased rate of dark re-reduction of FR-oxidized P700 (Table 1). This observation coincides with the elevated absolute levels of ATP + ADP with still 73% of the adenylate pool being in the form of ATP in WP as compared to 87% in SP (Table 4). Verhoeven et al. (1999) have also reported a winter-induced increase of the adenylate pool of Pinus ponderosa grown in the ®eld. This suggests a role for cyclic electron transport in supplying ATP in WP upon thawing, and is consistent with reports on an increased capacity for phenazine methosulfate-mediated cyclic electron transport in chloroplast thylakoids isolated from Picea abies during the winter (Senser et al. 1975). In view of the strong dependence of the xanthophyll cycle on the magnitude of the pH gradient across the chloroplast membranes generated by electron transport (Demmig-Adams and Adams 1992), it is tempting to suggest that the enhanced capacity for PSI-mediated cyclic electron transport coupled with proton pumping would enhance the potential for non-photochemical, xanthophyll cycle-dependent light-energy dissipation under conditions of strongly inhibited PSII photochemistry during winter conditions. In fact, considerably lower values of the epoxidation state, corresponding to better performance of the xanthophyll cycle and accumulation of carotenoids involved in the xanthophyllcycle pool, have been reported for Scots pine needles during the winter (Ottander et al. 1995). Similar data have been published for other overwintering conifer species (Adams and Demmig-Adams 1994). We suggest that PSI-mediated cyclic electron transport is a prerequisite for maintaining functional integrity of the needles, including a sustained high-energy quenching state of the xanthophyll cycle, during periods of thawing of the needles during the winter. Frost causes inhibition of CO2 uptake in cold-acclimated Scots pine (Strand and OÈquist 1985a). This most likely results in a PSI-acceptor side limitation of photosynthesis and leads to photoinhibition of PSII (Strand and OÈquist 1985b), while at the same time PSI photochemistry is largely retained (Figs. 2B, 4C). It appears as if the winter-induced PSI acceptor-side limitation stimulates cyclic PSI electron transport in Scots pine. PSI acceptor-side limitation may also at least partly explain why stromal and intersystem electron pools available for P700+ reduction increase strongly during the winter in Scots pine (Table 1) despite a very strong reduction of PSII photochemistry (Fig. 1A). Indeed, Heber et al.
Table 4 ATP and ADP contents in needles of Scots pine during the summer (SP), winter (WP) and in cold-acclimated (CP) plants. Mean values SE were calculated from 3±6 independent measurements. Statistically signi®cant dierences between SP and CP and WP at P
