• Some coniferous forest ecosystems undergo complete photosynthetic down-regulation in winter. The present study examined the influence of several environmental parameters on intrinsic, needle-level photosynthesis and photoprotection during the spring reactivation of photosynthesis in subalpine conifers.
• Maximal photosystem II (PSII) efficiency, photosynthetic capacity, and amounts of zeaxanthin and early light-inducible protein (Elip) family members were assessed in three subalpine conifer species over 3 years, and intensively during the 2003 winter-to-spring transition.
• During summers, maximal PSII efficiency remained high while intrinsic photosynthetic capacity varied depending on precipitation. During winters and the winter-to-spring transition, photosynthetic capacity and PSII efficiency were highly correlated and (during the spring transition) strongly influenced by air and soil temperature and liquid water availability. Decreases in the amount of Elip family members from winter through spring paralleled disengagement of sustained zeaxanthin-dependent photoprotection, although one of four anti-Elip antibody-reactive bands increased during spring.
• Intrinsic photosynthetic capacity and maximal PSII efficiency were highly responsive to day-to-day environmental changes during spring, indicating that multiple environmental signals are integrated to orchestrate the reactivation of photosynthesis from the inactive winter state to the active summer state.
Over recent decades, there has been growing consensus among scientists that anthropogenic carbon dioxide emissions have caused and will continue to result in shifts in the global climate (Crawley, 2000). One hypothesized effect of this change, particularly in regions characterized by cold winters (e.g. high latitudes and altitudes), is an increase in the length of the growing season caused by warmer spring temperatures (Serreze et al., 2000), resulting in an earlier onset of photosynthesis. These changes could have pronounced impacts on the global carbon cycle.
Measurements of net ecosystem carbon dioxide exchange (NEE) of a high-altitude subalpine forest in the Colorado Front Range have shown the spring to be a particularly important time for net carbon gain (Monson et al., 2002, 2005). This is presumably caused by two factors. First, environmental conditions during the late spring – when air temperatures are warm but not hot, the soils have thawed, and there is a readily available supply of water from snowmelt – are ideal for photosynthesis. Second, the persistence of the snowpack and water-saturated soils presumably limit soil respiration (and thus carbon dioxide release) from increasing substantially.
During the winter, this subalpine ecosystem is no longer a carbon sink (Monson et al., 2002, 2005), and even the intrinsic photosynthetic capacity of these trees is strongly and persistently down-regulated, as shown in detail for lodgepole pine and subalpine fir, two of the dominant species of the subalpine forest of the Colorado Front Range (Zarter et al., 2006b). These trees employ a sustained form of photoprotection in which the dissipative pigments of the xanthophyll cycle (zeaxanthin and antheraxanthin) are retained and photosystem II (PSII) efficiency is maintained in a dissipative state throughout the winter, as has also been observed in other overwintering evergreens (Adams et al., 2002, 2004). Evergreen species in these forests also exhibited a strong up-regulation of multiple stress-induced proteins in the winter that cross-reacted with antibodies against early light-inducible proteins (Elips; Zarter et al., 2006b) and/or high light-inducible proteins (Hlips, Zarter et al., 2006a). It was proposed that these proteins may play a role in facilitating the persistent form of photoprotection employed by evergreens during the winter.
A down-regulation of intrinsic photosynthetic capacity in the winter in Scots pine (Pinus sylvestris) was also observed in a Siberian boreal forest (Ensminger et al., 2004), and for this system it was suggested that springtime photosynthetic recovery was strongly linked to air temperatures. In the same study, a strong up-regulation of a single Elip was also reported that occurred transiently in the spring. One may speculate that the absence of high levels of Elips in boreal Scots pine during the winter (Ensminger et al., 2004) could be caused by the low light levels characteristic of high northern latitudes during those months. These low light conditions contrast with the very high levels of solar radiation experienced by the subalpine forest in the Rocky Mountains of Colorado.
In the spring, Monson et al. (2002) observed a rapid activation of carbon uptake by the subalpine forest, with NEE switching from positive values (representing a net respiratory CO2 loss by the forest ecosystem) to negative values (indicating a net photosynthetic CO2 gain by the forest) over a short time course. In the present study, the reactivation of intrinsic photosynthetic capacity and disengagement of sustained photoprotection are characterized during the transition from winter through spring in lodgepole pine and subalpine fir. In addition, the pattern of changes in intrinsic photosynthetic capacity and efficiency from spring through summer into autumn for these two species, as well as Engelmann spruce (the third dominant conifer of this forest), are documented for three successive growing seasons with different precipitation patterns.
Materials and Methods
Field site and sampling dates
The capacity for photosynthetic oxygen evolution and maximal PSII efficiency in subalpine fir (Abies lasiocarpa (Hook.) Nutt.), Engelmann spruce (Picea engelmannii Parry ex Engelm.), and lodgepole pine (Pinus contorta Dougl. ex Loud.) were monitored over the course of 3 years (spring 2001 through fall 2003) in trees growing near the C1 climate station (3022 m; 40°02′N, 105°32′W) at Niwot Ridge located in the Roosevelt National Forest 15 km north of Nederland, CO, USA. A detailed characterization of photosynthetic up-regulation of subalpine fir and lodgepole pine was carried out during the spring of 2003. For the latter, intermittent sampling was begun on 15 March 2003, with regular sampling beginning on 2 April. Between 28 April and 27 May, a 3 d wk−1 sampling regimen was implemented.
Photosynthetic capacity and chlorophyll fluorescence measurements
Needles were collected before sunrise from several trees of each species that experienced direct sun exposure for at least part of the day. For a detailed description of the handling of these samples and the measurement of dark-adapted (maximal) PSII efficiency (Fv/Fm) at room temperature and the light- and CO2-saturated rates of photosynthetic oxygen evolution at 25°C, see Zarter et al. (2006b).
Protein extraction and characterization
Amounts of proteins that reacted with anti-Elip and anti-PsbS (PSII subunit S) antibodies were determined via immunoblotting subsequent to thylakoid isolation and protein extraction (after Ottander et al., 1995) as described in Zarter et al. (2006b). A polyclonal antibody raised against Elip from pea (Adamska et al., 1993; Lindahl et al., 1997) and chicken anti-PsbS (Agrisera AB, Vännäs, Sweden) were used. Samples from four dates corresponding to the extremes between the beginning of the transition at the end of winter and the completion of the transition at the end of spring, as well as two intermediate dates (representing partial recovery in response to a warming period and partial reversion to the winter state during a cooling period), were chosen for each species for protein analysis. For subalpine fir, these dates were 17 March, 15 and 19 April, and 27 May, whereas for lodgepole pine, they were 17 March, 15 and 21 April, and 27 May. Relative protein amounts were determined by laser densitometry and relative optical density of each band quantified as described in Zarter et al. (2006b). All replicates from seasons intended for direct comparison were run on the same gel. For each species, the highest mean relative optical density (from winter samples) was arbitrarily set to 1.0.
Average monthly air temperature and monthly precipitation over the course of the study period, as well as daily average air temperature, soil temperature, and relative soil moisture data from the spring of 2003 were obtained from the C1 climate station at the Niwot Ridge site. For analysis, soil temperature data were used to create the categorical variable ‘soil category’, based on whether the average soil temperature for a given date was above or below 0°C. The temperatures during the 24 h period immediately before sampling were used for analysis.
All statistical analyses were carried out using JMP In (SAS Institute, Cary, NC, USA). Multivariate regression analysis of photosynthetic capacity and PSII efficiency vs average air temperature, average soil moisture, and soil category during the spring of 2003 was conducted using the least-squares model. Photosynthetic capacities and PSII efficiencies between years/seasons and from the Elip collection dates were compared using the Tukey-Kramer test for honestly significant differences.
Intrinsic photosynthetic capacity, maximal PSII efficiency, and climate
In 2001, intrinsic photosynthetic capacity (assessed as the light- and CO2-saturated rates of photosynthetic oxygen evolution of warmed needles) in both subalpine fir (Fig. 1a) and Engelmann spruce (Fig. 1b) increased progressively throughout the growing season, peaking at the end of September. During 2001, monthly precipitation was relatively regular with greater rainfall in mid-summer compared to the following 2 years (Fig. 1d). In 2002, when summer precipitation was regular but considerably less than in 2001 (Fig. 1d), intrinsic photosynthetic capacity in both species (Fig. 1a,b) remained at an intermediate level until the autumn when precipitation levels increased, temperatures decreased (presumably reducing evapotranspirative demand), and photosynthetic capacities peaked again in September. In 2003, a large input of precipitation occurred in February and March, but was followed by low and declining amounts of precipitation through most of the summer (Fig. 1d). In contrast to the previous two summers, photosynthetic capacities in both subalpine fir (Fig. 1a) and lodgepole pine (Fig. 1c) peaked at the beginning of June 2003 and then declined steadily and significantly throughout the summer. However, as temperatures declined and a small spike in additional precipitation occurred in August (Fig. 1d), a significant increase in intrinsic photosynthetic capacity was observed in both species in September 2003 (Fig. 1a,c).
All three species showed strong, significant (P < 0.05) decreases in the capacity for photosynthetic oxygen evolution and maximal PSII efficiency during the winter of each year (Fig. 1a–c). In general, changes in maximal PSII efficiency were similar to those of intrinsic photosynthetic capacity throughout the year, with the exception of the summer of 2003 when photosynthetic capacity declined but maximal PSII efficiency remained high until September 2003 (Fig. 1a,c).
During the growing season of 2002, intrinsic photosynthetic capacity varied among the three conifers; it was lowest in subalpine fir (Fig. 1a), intermediate in Engelmann spruce (Fig. 1b), and highest in lodgepole pine (Fig. 1c). The two species with the lowest and highest photosynthetic capacities, subalpine fir and lodgepole pine, were therefore selected for further characterization during the winter-to-spring transition in 2003 (Fig. 2). As spring progressed, the capacity of photosynthetic oxygen evolution and the maximal efficiency of light utilization by PSII increased, with arrests in the upward trajectory and/or decreases following periods of decreased temperatures, for example, following days 86–88, days 94–99, days 106–110, day 139, and during the period of relatively stable temperatures from the end of April to the beginning of May from days 118 to 132 (Fig. 2).
In contrast to the summer period, where intrinsic photosynthetic capacity varied greatly (i.e. between 15 and 45 µmol O2 m−2 s−1 in lodgepole pine during the summer of 2003; Fig. 1) but maximal PSII efficiency remained high and stable, concurrent changes in intrinsic photosynthetic capacity and maximal PSII efficiency occurred during the spring of 2003 in both conifer species (Fig. 2). A strong positive correlation existed between the two parameters in both subalpine fir (R2 = 0.64, P < 0.0001; Fig. 3a) and lodgepole pine (R2 = 0.82, P < 0.0001; Fig. 3b). Furthermore, multivariate regression analysis was conducted (Table 1) for either intrinsic photosynthetic capacity or maximal PSII efficiency vs a combination of climatic parameters, including average daily air temperature and soil moisture as well as the categorical variable ‘soil category’ based on whether the daily average soil temperature was above or below freezing (for the latter, see Monson et al., 2002). Covariation among the three environmental parameters was examined and found to be minimal, thus allowing inclusion of all three in the model for analysis. This multivariate regression analysis yielded strong correlations between either intrinsic photosynthetic capacity or maximal PSII efficiency and the combined climatic parameters, each of which contributed significantly to these correlations (Table 1).
Table 1. Results of multivariate regression analysis of climatic variables (average daily air temperature, average daily soil moisture, and the soil temperature category (above or below freezing)) vs either the light- and CO2-saturated capacity for photosynthetic oxygen evolution (PS capacity) or photosystem II (PSII) efficiency in subalpine fir (Abies lasiocarpa) and lodgepole pine (Pinus contorta). Shown are the R2-value and the significance of the whole model multivariate analyses, as well as the significance of each parameter included in the analyses.
R2 = 0.62
R2 = 0.71
R2 = 0.78
R2 = 0.77
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
Average air temperature
P = 0.001
P < 0.0001
P = 0.0002
P < 0.0001
Average soil moisture
P < 0.0001
P < 0.0001
P < 0.0001
P < 0.0001
Soil temperature category
P < 0.0001
P = 0.0323
P < 0.0001
P = 0.0019
Photosynthesis, Elip family members, and the xanthophyll cycle on selected dates
Intrinsic photosynthetic capacities for four dates during the spring transition selected for protein analysis are shown in Fig. 4(a) for subalpine fir and lodgepole pine. In both species, photosynthetic capacities were low at the end of winter and at near maximal levels at the end of the spring recovery period, but after an initial increase, showed intermittent decreases to winter-like values on 19 and 21 April 2003 following the return of more winter-like conditions (Fig. 4a). The fact that subalpine fir showed a less pronounced decrease in photosynthetic capacity may be a result of the fact that capacities in subalpine fir were measured after two fewer days of exposure to the winter conditions compared with lodgepole pine (Fig. 4a).
Maximal pre-dawn PSII efficiencies for each species determined upon warming of needles exhibited similar patterns to those of intrinsic photosynthetic capacity (Fig. 4b). Furthermore, changes in the amounts of nocturnally retained zeaxanthin and antheraxanthin mirrored the changes in photosynthetic capacities and maximal PSII efficiency as spring progressed, with high amounts at the end of winter and during the intermittently colder period on 19 and 21 April, and lower amounts following the warming period at the beginning of April and especially at the end of May (Fig. 4c). The sum of all anti-Elip-reactive protein bands generally decreased during the spring (Fig. 4d). On the two dates where both species displayed decreases in photosynthetic capacity and maximal PSII efficiency (21 April for lodgepole pine and 19 April for subalpine fir), neither species showed an increase in the sum of anti-Elip-reactive proteins (Fig. 4d). While the overall trend was for anti-Elip-reactive proteins to decrease from mid-March to the end of May 2003, the anti-Elip-reactive band with the highest apparent MW showed a somewhat different trend, especially in subalpine fir (Fig. 5), with apparent increases during the spring transition. Mean optical densities for the top Elip-reactive band of subalpine fir were 1 ± 0.31 on 17 March, 1.80 ± 0.14 on 15 April, 1.13 ± 0.23 on 19 April, and 0.03 ± 0.05 on 27 May. Amounts of proteins responsive to the anti-PsbS antibodies, on the other hand, did not exhibit decreases during the spring transition, except for the lower of the two bands on 27 May in lodgepole pine (Fig. 5).
In all of the subalpine conifers characterized in this study, two different types of adjustment of photosynthetic capacity were seen. One adjustment altered maximal intrinsic photosynthesis rates (as maximal rates of photosynthetic oxygen evolution) in the summer, apparently largely in response to water availability, while leaving light energy conversion in PSII unaltered and at maximum efficiency. Intrinsic capacities of photosynthesis were as high in the fall as in the spring, particularly in years with greater amounts of precipitation. On the other hand, the subalpine forest ecosystem as a whole serves as a carbon sink only during the spring, with carbon uptake decreasing during the drier summer months and soil respiration becoming more dominant (Monson et al., 2002, 2005). Similar effects of mid-summer drought were seen in Douglas fir forests in the Pacific Northwest (Chen et al., 2004; Paw et al., 2004). Additionally, interannual variability in precipitation has been shown to affect carbon assimilation rates in evergreen forests (Man & Lieffers, 1997; Monson et al., 2002, 2005; Paw et al., 2004) as well as in deciduous forests (Chen et al., 1999). By contrast, photosynthetic capacity and NEE of European boreal and subalpine forests have been found to be influenced primarily by prevailing temperatures during the growing season (Mäkeläet al., 2004; Wieser & Stohr, 2005).
The other type of photosynthetic adjustment consisted of a down-regulation of both maximal PSII efficiency and maximal electron transport rates (and thus intrinsic photosynthetic capacity) in the winter and a concomitant reactivation of both processes during spring. Strong down-regulation of intrinsic photosynthetic capacity was also seen during the winter in a Siberian boreal forest (Ensminger et al., 2004), even though in these regions the intensity of incident light during the winter is significantly lower than in the present study. Strong down-regulation of intrinsic photosynthetic capacity in these boreal ecosystems is associated with cessation of carbon uptake by the forest for the entire winter season (Lloyd et al., 2002; Öquist & Huner, 2003; Tanja et al., 2003; Ensminger et al., 2004). Similar winter down-regulation of photosynthetic capacity (Stecher et al., 1999), CO2 exchange by branches (Wieser, 1997), and net carbon uptake by the entire forest (Monson et al., 2002) has been observed in subalpine ecosystems. By contrast, other temperate forest ecosystems can show reduced rates of carbon uptake during cold winter periods but resume carbon uptake during intermittent milder periods in the winter. While Eastern Hemlock (Tsuga canadensis) exhibited greatly reduced CO2 exchange rates during the winter in central Massachusetts, and red spruce (Picea rubens) showed CO2 exchange rates that were negative or close to 0 µmol CO2 m−2 s−1 during the winter in Vermont, carbon uptake rates increased in both species during warmer winter periods (Hadley, 2000; Schaberg, 2000). Furthermore, Douglas fir (Pseudotsuga menziesii) and ponderosa pine (Pinus ponderosa) stands in the Pacific Northwest and Loblolly pine (Pinus taeda) in the southeastern USA exhibited little indication of photosynthetic dormancy during the winter (Anthoni et al., 1999; Ellsworth, 2000; Paw et al., 2004).
During the spring, both intrinsic photosynthetic capacity and maximal PSII efficiency underwent changes that closely tracked changes in environmental conditions. In previous studies, needles or branches were transferred from winter conditions to laboratory conditions of low light and room temperature. Upon such a transfer, full recovery of maximal PSII efficiency (Fv/Fm of 0.8) in Douglas fir and ponderosa pine needles was not seen even after 100 h (Adams et al., 1995; Verhoeven et al., 1996). In a later study, however, ponderosa pine experienced full recovery of maximal PSII efficiency c. 120 h after transfer from the field (Verhoeven et al., 1999). In branch-level experiments with lodgepole pine and subalpine fir, recovery of PSII efficiency showed different results over two different years (Monson et al., 2005). In the first year (1999), lodgepole pine did not exhibit full recovery during the time course of the experiment (100 h). In the second year (2004), both lodgepole pine and subalpine fir apparently showed full recovery of maximal PSII efficiency 1 d after transfer.
Multivariate regression analysis in the present study yielded strong correlations for both intrinsic photosynthetic capacity and maximal PSII efficiency with multiple environmental parameters, including air and soil temperature and soil moisture. These results suggest that different signals are integrated during the transition from the down-regulated and continually photoprotected state in the winter to the active state at the end of the spring. Air temperature is likely to be sensed at the canopy level and changes in soil temperature or moisture availability below ground. The multivariate analyses presented here suggest that both atmospheric and soil conditions contribute to a signaling network that orchestrates the spring reactivation of intrinsic photosynthetic capacity and deactivation of sustained photoprotection.
Monson et al. (2002) observed rapid activation of net carbon uptake by the subalpine forest over 1–2 wk. The initiation of springtime carbon uptake occurred within a few days of night-time soil temperatures rising above 0°C. In a follow-up study (Monson et al., 2005), springtime photosynthetic recovery was found to be dependent upon the interplay of warming air temperatures, increasing soil temperatures, and increasing soil moisture availability. They proposed a model in which resumption of carbon uptake is initiated by warming air temperatures but constrained by soil temperature and moisture availability. Full recovery of carbon uptake was not seen until soil temperatures warmed enough for water to become available – an event that was dependent upon the snowpack in the forest becoming isothermal. Monson et al. (2005) also suggested that several separate initiations of the spring recovery of forest carbon uptake might take place and be reversed with the intermittent return of colder temperatures. Only once soil moisture becomes available through sufficient warming does complete recovery of carbon uptake occur. It thus appears that air and soil temperature and water availability determine the time at which spring reactivation of intrinsic photosynthetic capacity as well as carbon uptake occurs, and that there may be considerable variation from year to year as to when this important event takes place (Monson et al., 2005). Photoperiod (Dodd et al., 2005) does not therefore appear to play a major role in signaling this event.
In contrast to the interplay of several environmental factors influencing the reactivation of photosynthesis and disengagement of sustained photoprotection in Rocky Mountain subalpine conifers during spring, conifers of the boreal forests of Scandinavia and Siberia appear to respond primarily to air temperature alone. Although manipulative experiments with trees suggested that soil temperature can influence the timing of the vernal reactivation of photosynthesis in Scots pine in Sweden (Strand et al., 2002), natural studies have found that air temperature alone is the primary factor driving the up-regulation of photosynthesis in boreal coniferous forests during the transition from winter to summer (Tanja et al., 2003; Ensminger et al., 2004).
The two primary conifer species used in the present study expressed high levels of proteins in the winter that cross-reacted with antibodies against Elips (see also Zarter et al., 2006b). Starting from high amounts of several anti-Elip-reactive proteins in both species at the end of the winter in the present study, the overall amount decreased in both species as spring progressed without intermittent increases after colder days when photosynthetic capacity and maximal PSII efficiency decreased again intermittently for both species. However, one of the anti-Elip-reactive bands (particularly in subalpine fir) showed a trend for an accumulation specifically during the spring transition. In addition, amounts of anti-PsbS-reactive proteins that were similar to those present in the summer were maintained throughout the winter, and included an additional anti-PsbS-reactive band present exclusively during the winter (Zarter et al., 2006b) that was furthermore maintained throughout the transition to spring (Fig. 5).
These results for subalpine conifers in the Rocky Mountains are in stark contrast to those for conifers in a Siberian boreal ecosystem. In a study of Scots pine, Ensminger et al. (2004) reported very low amounts of PsbS during the winter with only one protein band that reacted with an anti-Elip antibody that was present in relatively low amounts during the winter and early spring, but increased transiently to maximal levels in the spring before decreasing again. The same pattern, with greatest accumulation in spring rather than winter, was observed for the amounts of zeaxanthin in boreal Scots pine (Ensminger et al., 2004), whereas both zeaxanthin and overall anti-Elip-reactive protein amounts were maximal in the winter in the subalpine forest characterized in the present study. It is therefore tempting to speculate that the strong up-regulation of three of the anti-Elip-reactive bands in the subalpine conifers in the winter may be related to the differences in light environment between the subalpine and the boreal environments.
During the winter and early spring in the boreal forest, light intensities are much lower than in the Colorado subalpine forest, and the boreal forest trees are consequently experiencing less excess light simply as a result of there being less incident light. As the spring progresses, trees in the boreal forest become exposed to more and more excess light (Ensminger et al., 2004), until climatic conditions become favorable for the resumption of photosynthesis – at which point the intensities of excess light begin to decrease significantly. In the subalpine forest of Colorado, however, incident light intensities remain quite high throughout the year, including the winter. This means that trees growing in the Rocky Mountains experience highly excessive light intensities throughout the winter and into early spring. Then, as spring progresses and intrinsic photosynthetic capacity is gradually up-regulated, the intensity of excess light decreases, and sustained photoprotection becomes disengaged. However, since one of the Elip-reactive bands accumulated specifically during the spring transition in the present study, it is conceivable that the different Elip-antibody-reactive proteins have different functions. The one anti-Elip-reactive band seen in Scots pine (Ensminger et al., 2004) showed a similar trend to this particular band (one of the four anti-Elip-reactive bands obtained in the two subalpine conifers examined here). Differences in the accumulation pattern were also observed for two Elips from Arabidopsis grown under controlled growth chamber conditions, where a higher photon flux density was required for the accumulation of Elip2 than of Elip1 (M. Heddad et al., unpublished). It has been suggested that Elips function in Chl turnover and/or zeaxanthin-associated photoprotection (Adamska, 2001; see also Montané & Kloppstech, 2000; Ensminger et al., 2004), and the correlations between Elip up-regulation and sustained zeaxanthin-associated photoprotection are consistent with these suggestions.
This work has been supported by grants from the Andrew W. Mellon Foundation (award no. 20200747) and the United States Department of Agriculture (award no. 00-35100-9564) to the Adams/Demmig-Adams laboratory, and Deutsche Forschungsgemeinschaft grants (AD 92/7-1; AD 92/7–2) and a Konstanz University grant to the Adamska laboratory. We would like to thank the Mountain Research Station, University of Colorado, for permission to work on the trees at Niwot Ridge. Daniel Liptzin's assistance with the multivariate analysis is also greatly appreciated.