Planta +2001) 214: 57±66 DOI 10.1007/s004250100580

ORIGINAL ARTICLE

Tyler D.B. MacKenzie á Tara M. MacDonald Luc A. Dubois á Douglas A. Campbell Seasonal changes in temperature and light drive acclimation of photosynthetic physiology and macromolecular content in Lobaria pulmonaria

Received: 8 November 2000 / Accepted: 27 January2001 / Published online: 13 June 2001 Ó Springer-Verlag 2001

Abstract Lobaria pulmonaria +L.) Ho€m. is an epiphytic Keywords Acclimation +seasonal) á Electron lichen common to temperate deciduous forests where it transport +PSII) á Gene expression +rbcL) á Lobaria copes with large changes in temperature and light levels +acclimation) á Photosynthesis lichen á Ribulose-1,5- through repeated annual cycles. Samples of L. pulmo- bisphosphate carboxylase-oxygenase naria were taken from a deciduous forest in southeastern Canada at 35-dayintervals from February1999 to Abbreviations ETR: electron-transport rate through February2000 and also from a rare population in an PSII, á LSU: large subunit of RuBisCO, á NPQ: non- evergreen forest in March and August 1999. At ®eld- photochemical chlorophyll a ¯uorescence quench- ambient temperatures and light levels, the realised pho- ing, á PPFD: photosynthetic photon ¯ux density, á qp: tosystem II +PSII) electron transport was low both in the photochemical chlorophyll a ¯uorescence quench- summer and winter, with transient peaks in the spring ing, á RuBisCO: ribulose-1,5-bisphosphate carboxylase- and autumn. In contrast, the seasonal pattern of po- oxygenase tential electron transport measured at a ®xed 20°C peaked in winter, showing the importance of tempera- ture in driving photosynthesis to low levels in the winter Introduction despite an acclimation of electron-transport potential to exploit the high ambient light. Realised gross CO2 up- Lichens are unusual in that theyare long lived and take was correlated with PSII electron transport at persist through manyseasonal cyclesof environmental mechanisticallyplausible rates at all sampling sites in the change, but theyare onlyintermittentlyactive during summer but not in the winter, indicating electron di- these changes because theyare poikilohydric.The pho- version awayfrom CO 2 ®xation in the winter. Chloro- tosynthesis and growth of lichens are in¯uenced by phyll content was highest in the dark summer months. seasonal environmental change in the ambient light, The amount of ribulose-1,5-bisphosphate carboxylase- temperature and moisture. Acclimation of their photo- oxygenase +RuBisCO) large subunit +LSU) was highest synthetic physiology could help lichens to maintain in spring. Changes in the level of this hyperabundant carbon reduction in these changing environments. Al- protein and in the activityof PSII maintained a rela- ternatively, lichens might wait to exploit only those tivelyconstant rate of maximum CO 2 uptake per interludes of conditions that match a more ®xed physi- RuBisCO LSU from April through November, despite ological acclimation state. great changes in the seasonal light and temperature. Lobaria pulmonaria is a tripartite lichen association L. pulmonaria acclimates between light and temperature with a minor nitrogen-®xing cyanobacterial partner and stress in the winter months to light-limitation in the dark a primarygreen-algal photobiont +Jordan 1973), whose summer months. Transition intervals in the spring and photosynthesis provides organic carbon to sustain the autumn, with warm, bright and wet conditions, are association +Honegger 1991). It is a foliose, epiphytic likelythe most amenable times for growth. lichen with a circumpolar distribution and is common in undisturbed deciduous woodlands as the most conspic- uous member of the lobarion alliance of epiphyte species T.D.B. MacKenzie +&) á T.M. MacDonald á L.A. Dubois +Rose 1988). Several studies have used this lichen as an D.A. Campbell old-forest indicator species because it is sensitive to de- Department of Biology, Mount Allison University, forestation +Gauslaa and Solhaug 1996, 1999) and SO Sackville, New Brunswick, E4L 1G7, Canada 2 E-mail: [email protected] air pollution +Farmer et al. 1991). It is typically an epi- Fax: +1-506-3642505 phyte on maple +Acer spp.) trees but more rarelyit will 58 grow on spruce +Picea spp.). In deciduous woods there is Sackville station, and ®eld-ambient photosynthetic photon ¯ux a strong seasonal change in the light environment as- density+PPFD) was measured at the time and site of thallus col- lection at all sampling sites +Fig. 1). sociated with canopyclosure in spring and subsequent Intact, air-dry, individual thalli were removed from their sub- defoliation in autumn. Lichens in these forests are sub- strate trees and sealed in plastic bags for transport back to the jected to greater maximum solar radiation during the laboratoryin the dark at <5 °C. On 9 March 1999, 8 August 1999 cold winter months because of the open canopy, while and 17 October 1999 it was impossible to select dryindividuals. In those in coniferous forests experience a lower and rela- these cases, thalli were air-dried for up to 3 h at 25°C in the dark, immediately upon return to the laboratory. This drying wasdone as tivelyconsistent light environment throughout the year. rapidlyas possible at room temperature and in the dark to mini- Several other lichens have been shown to acclimate to mise excitation pressure within the desiccating photosynthetic ap- seasonal changes in both light and temperature. Periods paratus and to minimise acclimation that might be driven by of photosynthetic activity measured by chlorophyll ¯u- ambient light. We have tested +one-wayANOVA) for signi®cant e€ects of hydration status at the time of sampling on subsequent orescence varyseasonallyin Peltigera +Lange et al. physiological measures and found no meaningful e€ect. We have 1999). Seasonal changes in photosynthetic CO2 uptake thus discounted anystrong connection between hydrationstatus can be driven bylight and temperature +Coxson and- immediatelyprior to sample harvesting and the physiologicaland Kershaw 1984; Kershaw 1984) or the combination of macromolecular status of the lichens. All thalli were stored dry both +Larson and Kershaw 1975). Chlorophyll content and in darkness at ±20 °C until required for physiological can also change seasonally+MacFarlane et al. 1983; Kershaw and Webber 1984). L. pulmonaria is sensitive to high light and particularlyto seasonal changes in light +Gauslaa and Solhaug 1999). Photosynthetic light-response curves are a means to quantitativelyassess the light acclimation status of a photosynthetic organism by determining maximal rates and eciencies of photosynthetic processes. We there- fore measured light response curves of photosystem II +PSII) electron transport and CO2 exchange in popula- tions of L. pulmonaria from both deciduous and ever- green forests to detect seasonal changes in photosynthetic performance and acclimation status. Changes in chlorophyll and ribulose-1,5-bisphosphate carboxylase-oxygenase +RuBisCO) concentrations and in the level of transcription of the rbcL gene encoding the large subunit +LSU) of RuBisCO were also tracked through the year. RuBisCO is the hyperabundant en- zyme responsible for primary CO2 ®xation. The con- centration of this enzyme is one limit on the maximum rate of photosynthetic CO2 uptake. These measurements helped explain the molecular basis for the large changes observed in light capture, light energyallocation and CO2-®xation capacityas L. pulmonaria moved from a light-limited acclimation state in the summer to a light- saturated acclimation state in the winter.

Materials and methods Fig. 1 Mean dailymaximum temperature measured at the Envi- ronment Canada Sackville station +A), ®eld-ambient PPFD Sampling sites and thallus collection measured at the time and site of thallus collection +B) and total precipitation at Walker Road +C) from 7 February1999 to 18 Populations of L. pulmonaria +L.) Ho€m. were collected from four February2000. Mean dailymaximum temperature values are forested sites in southeastern New Brunswick and northern Nova means ‹ SE for the 13 days up to and including the date of Scotia, Canada. Three of the sites +Walker Road, N.B., 45° 56.65¢ sampling. Sample PPFD was determined once for each thallus at N, 64° 24.56¢ W; Fenwick Park, N.S., 45° 44.22¢ N, 64° 10.01¢ W; the point of collection and averaged to give the ambient PPFD. and Sugarloaf Brook, N.S., 45° 34.23¢ N, 63° 49.07¢ W) were in Open squares in B indicate data interpolated due to instrument maple +Acer sp.)-dominated hardwood forests, while the remaining failure; interpolated data were based on recorded values from site +EconomyMountain, N.S., 45 ° 24.33¢ N, 64° 01.79 W) was in a adjacent sampling dates and were used onlyas a guide for the dense white spruce +Picea glauca) coastal forest. The L. pulmonaria realised light levels described in Fig. 4. Total precipitation was population at Walker Road was sampled between 7 February1999 normalised to water-column height and is expressed as a total for and 6 April 2000 at approximately35-dayintervals. The remaining the 13 days up to and including the date of sampling, rather than sampling sites were sampled in March 1999 and August 1999. mean ‹ SE because of its discontinuous nature. Precipitation was Mean dailymaximum temperature and integrated precipitation dominated bysnow between mid December and earlyApril, thus over the 13 days preceding each site visit to Walker Road was liquid water available from precipitation was largelyrestricted to calculated from measurements at the nearbyEnvironment Canada the spring, summer and autumn months 59 experimentation, all of which occurred within 15 months of sample and Snel 1990). In addition to measures at ®eld-ambient tempera- collection. L. pulmonaria remains photosynthetically active upon ture, chlorophyll ¯uorescence was also measured at a ®xed 20 °Cto re-hydration after many years of dry freezer storage +Feige and distinguish between e€ects of seasonal changes in temperature and Jensen 1987). light intensityon PSII function. Upon their dayof use, tissue disks of ca. 1.25 cm 2 were punched from each individual thallus at the ®rst branch node behind the young growing tip of the thallus, a region broad enough to con- Fluorescence parameters sistentlyprovide full disk samples. The physiologicaland macro- molecular parameters associated with this location in the thallus are With the ¯uorescence values measured as described above, we similar to those of the marginal 1 cm +data not shown). Immedi- calculated PSII electron-transport rate +ETR), photochemical ¯u- atelyprior to use, each disk was allowed to re-hydratefor 90 min orescence quenching +qp) and non-photochemical ¯uorescence between sheets of tissue paper dampened with double-distilled quenching +NPQ) to track seasonal acclimation of PSII. The water. This re-hydration period was sucient to allow full activa- photochemical yield of PSII electron transport, /PSII, was calcu- tion of photosynthetic processes. Processing of the lichen samples- lated as +Gentyet al. 1989): from collection to physiological examination in the laboratory was F0 À F planned to minimise the length of metabolicallyactive +hydrated / ˆ m s 1† PSII F0 and warm) time for each thallus between collection and measure- m ment, and to minimise the potential for physiological or molecular and electron transport through PSII was estimated as: acclimation awayfrom the ®eld condition at the time of sampling. / I ETR ˆ PSII 2† PSII 2 Measurement of chlorophyll ¯uorescence where I is the PPFD in lmol photons m±2 s±1 and ETR is electron ±2 ±1 transport rate in lmol m s of thallus tissue. The product of /PSII After re-hydration, tissue disks were blotted of excess water and and I is halved, assuming an equal distribution of absorbed light to placed into a cuvette for chlorophyll ¯uorometry and CO2-ex- PSII and PSI +Gentyet al. 1989). The electron-transport results for change analysis in which the temperature, PPFD, ambient gas and the light-intensityseries were ®tted with a model curve of electron humiditywere controlled. For most experiments, measurement transport through PSII versus PPFD +modi®ed from Ricker 1975): temperature was set to match the ®eld-ambient temperature pre- vailing at the time of sample collection. Cold actinic illumination ETR ˆ aIe ÀbI † 3† from a DC-powered incandescent bulb +Phillips) was delivered to the cuvette via one branch of a ®bre-optic bundle +Walz) termi- where a is the initial slope of electron transport versus PPFD, I is nating at a glass window just above the lichen surface. The PPFD the PPFD and b is a curvature parameter. To assess photosynthetic reaching the lichen surface was continuouslymonitored within the acclimation, we extracted chlorophyll ¯uorescence parameters at cuvette byan S1133 silicon photodiode +Hamamatsu), which was the PPFD required to reach half-maximum electron-transport rate periodicallycalibrated with a LI-250 PAR Meter +Li-Cor). The on the light response curve, measured at ®eld-ambient temperatures atmosphere within the cuvette was isolated from the laboratoryby +see example in Fig. 2). To assess realised ®eld performance we a piping system, which drew fresh air from outside the building extracted chlorophyll ¯uorescence parameters at the ®eld-ambient light level at the time of collection. We also estimated the maximum +which is located near the Walker Road sampling site). The CO2 concentration of the air ¯owing into the cuvette was periodically potential capacities of these parameters byextracting them at the monitored by one of two infrared gas analysers +IRGAs; Analytical PPFD required for maximal electron transport at a ®xed 20 °C. Development Corp.; or Qubit Systems). Light response curves of electron transport estimated bychloro- Chlorophyll ¯uorescence was measured from the thallus disk phyll ¯uorescence have been used to assess photosynthetic samples in the cuvette described above. The samples were kept in the dark in the cuvette for at least 14 min to allow for dark-accli- mation, a condition with the algal photosystems maximally able to accept light energyand release electrons to plastoquinone +Schreiber et al. 1994). Chlorophyll ¯uorescence from the samples was excited bya blue light-emitting diode +LED; peak emission 460 nm) modulated at 1.6 kHz for an average PPFD of ca. 0.02 lmol photons m±2 s±1. Fluorescence was measured via a branch of the common ®bre-optic bundle bya PAM-101 Chlorophyll¯uorometer +Walz). The intensityof the measuring LED was sucient to excite minimal chlorophyll ¯uorescence from the sample, but insucient to drive signi®cant photochemistry. Minimal +Fo) and maximal +Fm) ¯uorescence were determined from the dark-acclimated sam- ples using a light-saturation ¯ash of ca. 1,800 lmol photons m±2 s±1 for 0.8 s delivered bya Schott KL 1500 electronic ¯ash lamp +Walz) and controlled bya PAM103 Flash Trigger Control unit +Walz). This light intensitywas strong enough to excite maximal ¯uores- cence, but low and brief enough to avoid photoinhibitorydamage. Then the sample was illuminated bya range of cold actinic light intensities +PPFDs of 4.4, 5.1, 12, 29, 58, 100, 201 and 298 lmol photons m±2 s±1), in increasing order. During illuminated periods, the excitation LED modulation rate was increased to 100 kHz +providing an average PPFD of 1.23 lmol photons m±2 s±1 in ad- dition to the actinic illumination) to increase the signal/noise ratio of the ¯uorescence signal. Once the sample reached steady-state ¯uorescence yield +F s) at each new light intensity, typically after Fig. 2 Light response curves of electron transport, and gross CO2 100±300 s, a saturation ¯ash was delivered to measure the realised uptake light-response curves, showing representative summer +solid 0 maximal ¯uorescence +F m) at each light intensity, and the sample symbols and line) and winter +open symbols and dashed line) was then transientlydarkened to determine realised minimal ¯uo- samples, measured at ®eld-ambient temperatures from the time of 0 rescence +F o) +¯uorescence nomenclature following van Kooten collection of L. pulmonaria thallus 60 acclimation in algae +Masojõ dek et al. 1999) and other photosyn- ±80°C until use. Serial re-extraction of the ground tissue showed we thetic organisms +Schreiber et al. 1994). attained an average of 80% yield of RuBisCO using this protocol, The approximate proportion of open PSII centres, qp, was with little variation in extraction eciencybetween samples. calculated as +Gentyet al. 1989): The proteins were separated on linear 10% SDS-polyacryla-

0 mide ReadyGels +BioRad) using a Mini-Protean II +BioRad) cell. Fm À Fs Samples were loaded on the basis of equal thallus area, with extract qp ˆ 0 0 4† 2 Fm À Fo from 8 mm of thallus per sample lane, and with isolated denatured spinach RuBisCO loaded as a quantitystandard on all gels. The Non-photochemical quenching +NPQ), representing the absorbed gels were electrophoresed at 100 V for 100 min in 25 mM Tris base, light energydissipated as heat, was calculated by+Bilger and 192 mM glycine, 0.1% SDS. Schreiber 1986): The proteins were then transferred from the gel in 49 mM Tris, F ÀF0 39 mM glycine, 0.04% SDS, 20% methanol to polyvinylidenedi- NPQ ˆ m m 5† ¯uoride +PVDF) membranes in a Semi-DryTransfer Cell +BioRad) F0 m at 13 V for 25 min. The PVDF membrane was then washed in Tris- bu€ered saline +TBS-T: 20 mM Tris base, 500 mM NaCl, 0.05% Measurement of CO exchange Tween-20; pH 7.5) and soaked for 1 h at 25°C in 5% low-fat 2 powdered milk in TBS-T. The membrane was incubated with a primaryrabbit anti-spinach RuBisCO LSU antibody+kind gift of The gas mixture in the cuvette was sampled for CO2 analysis in the dark and during the actinic illumination at PPFDs of 5.1, 29, 100, GoÈ ran Samuelsson, UmeaÊ University, Sweden) diluted 1:8,000 in 201 and 298 lmol photons m±2 s±1, simultaneouslywith the chlo- 10 ml of blocking bu€er, for 1 h at 25°C. The membrane was then rophyll ¯uorescence measurements. The gas was sampled by rinsed with TBS-T and incubated with a secondarygoat anti-rabbit ¯ushing it into a 20-ml syringe after a measured time of incubation IgG antibodyconjugated to alkaline phosphatase +Sigma-Aldrich), in the static cuvette atmosphere +typically 2±4 min). The gas sam- diluted 1:4,000 in 10 ml blocking bu€er, for 1 h at 25°C. The ples were then injected through a Drierite desiccant cartridge into membrane was then incubated in 10 ml 100 mM Tris base, an IRGA, with responses standardised byknown mixtures of 100 mM glycine, 5 mM MgCl2, 0.028% 4-nitro blue tetrazolium chloride +NBT) and 0.014% 5-bromo-4-chloro-3-indolyl-phosphate 1,020 ppm CO2 in air and N2 +BOC Gases). Brown et al. +1983) and Sundberg et al. +1999) have shown that lichen respiration is ele- +BCIP) for 4±5 min in the dark at 25°C to quantitativelyindicate vated and variable soon after re-hydration +the respiration-burst the presence of RuBisCO LSU. The blots were imaged on a phenomenon). Thus all CO -exchange data here are expressed as UMAX Astra 1220S scanner for later image analysis with the 2 NIHImage analysis package. gross CO2 uptake, correcting net exchange byseveral measure- ments of dark respiration rate from each sample. Chlorophyll ¯uorescence, CO2 exchange, actinic light intensity Expression of rbcL and cuvette temperature were recorded bya Universal Lab Inter- face and LoggerPro software +Vernier Software), running on a Expression of rbcL, which encodes the protein RuBisCO LSU, was Power Macintosh G3 Model M4787 computer +Apple Computer, determined qualitativelyusing reverse transcription polymerase Inc.). The CO2 signal was recorded bya Serial Box Interface and chain reaction +RT-PCR) ina Biometra T-Gradient thermocycler Logger software +Vernier Software), running on a Macintosh IIsi +Whatmann). RNA was extracted from ca. 100-mg samples of computer +Apple Computer, Inc.). L. pulmonaria tissue taken from the same thallus region as the After physiological analysis, the sample disks were removed samples used for protein measures, using the Trizol reagent and from the cuvette and their images were digitised on a UMAX Astra procedure +Life Technologies). A 50-ng sample of extracted RNA 1220S scanner and analysed with the NIHImage +National Insti- was then mixed in 50 ll of DEPC water with Ready-To-Go RT- tutes of Health, USA) analysis software package to determine their PCR beads +Amersham Pharmacia) with 20 pmol of each forward area. Sample dryweight was measured after baking the samples +5¢-AGGYGTTCCWSCWGAAG-3¢) and reverse +5¢-CACGRCT- over a bed of Drierite desiccant at 65 °C for 24 h. The samples were ACRATCTTTTTC-3¢) DNA primer speci®c to conserved regions then stored at ±20 °C until use for pigment and protein analyses. of chlorophyte rbcL, bracketing a 946-bp segment of the gene. These primers were designed to be speci®c for chlorophyte rbcL to Chlorophyll determination exclude contamination by rbcL expressed bycephalodial cyano- bacteria. Reverse transcription of rbcL transcripts proceeded for Chlorophyll was extracted from thallus disk samples of ca. 30 min at 42°C, followed by2 min at 94 °C to denature reverse 1.25 cm2 taken from the same point of the thallus at which the transcriptase. PCR cycling then proceeded directly in the same tube: physiological measures were recorded. The tissue disks were ground 94°C for 0.5 min, 48.6°C for 0.5 min, and 72°C for 1 min. The PCR and pigments extracted with 90% acetone in water saturated with was cycled either 30 or 36 times to gauge the rate of ampli®cation and hence gain a qualitative estimate of initial rbcL template con- MgCO3 as described in Barnes et al. +1992). Tissue pellets were reserved from the acetone extractions for protein extraction. We centration. Controls for DNA contamination were prepared for used acetone extraction rather than the popular dimethyl sulfoxide each RT-PCR process byomitting reverse transcriptase in the re- extraction also described byBarnes et al. +1992) because the latter action. In no case did the DNA controls generate ampli®cation was incompatible with the following protein extraction protocol. products, showing the absence of DNA contamination of the RNA Serial re-extraction of the ground tissue showed we attained >95% preparations. RT-PCR products from each thallus were run in 1% yield of chlorophyll using this acetone extraction. agarose gels in Tris-borate-EDTA bu€er at 70 V for 100 min. Gels were stained with GelStar nucleic acid stain +VWR, Canada). Gel images were captured with NIH Image v1.62 running on a Power Determination of RuBisCO LSU Macintosh 7600/120 via a Sonydigital video camera. The reserved tissue pellets from the acetone extractions +see above) were dried to completion in a stream of nitrogen, then proteins were solubilised in 250 ll of 100 mM Tris base, 160 mM sucrose, Results 1 mM EDTA, 1% sodium dodecyl sulfate, 0.5% dithiothreitol. The samples were frozen in liquid nitrogen, thawed while sonicated bya Photosynthetic optima at ®eld-ambient temperatures Sonic 300 Dismembrator +Artek Systems), then heated at 95°C for 5 min in a water bath to extract, denature and solubilise proteins. The samples were centrifuged at 12,000 g for 5 min and the To track photosynthetic performance as a function of supernatants containing the solubilised proteins were stored at the seasonallychanging temperature +Fig. 1) we present 61 physiological parameters measured at ®eld-ambient qp, the proportion of open PSII units measured at temperatures. First, we present data measured at the ETR0.5 Max, was high throughout earlyspring to light level required to reach ETR0.5 Max as an approxi- lateautumn, but was depressed in the winter at the mation of the optimal light level for the thalli +Fig. 2). Walker Road and Fenwick Park deciduous sites. Sam- Both the PPFD required to reach ETR0.5 Max and ples from the evergreen EconomyMountain population ETR0.5 Max rose from low levels in the winter to highs in and from the deciduous Sugarloaf Brook showed nearly earlysummer when measured at the ®eld-ambient tem- steady qp in winter and summer +Fig. 3C). The Sugarloaf perature in the Walker Road samples +Fig. 3A, B). The Brook site is located on a north-facing hillside with low seasonal changes in these two parameters were similar, incident light on the site in the winter which maymake but not as large, in the March and August samples from Sugarloaf Brook more similar to the evergreen Economy the other two deciduous forest sites, while the evergreen Mountain site than the other deciduous forest sites. site with the permanentlyclosed canopyat Economy The non-photochemical dissipation of light energy Mountain showed less change between March and Au- +NPQ) at ETR0.5 Max was low throughout the year gust +Fig. 3A, B). Ambient light was lower than the except for two transient peaks in the spring and autumn PPFD required to reach ETR0.5 Max in the closed- in the Walker Road samples, coinciding with periods canopysummer months, but ambient light greatly when ambient light exceeded the PPFD required for exceeded the PPFD required to reach ETR0.5 Max in the ETR0.5 Max +Fig. 3D). winter months.

Realised photosynthetic parameters at ®eld-ambient light level and temperatures

To approximate ®eld photosynthetic performance we present physiological parameters measured at the ®eld- ambient light level and temperatures. The electron transport rate at the ®eld-ambient light and temperature level was low in the winter months despite the high light levels at that time. Electron transport was relativelyhigh in spring and autumn, outside the period ofclosed canopyat Walker Road, but dropped in the summer when the leaf canopywas closed and light levels were low +Fig. 4A). This pattern con- trasts with the pattern of higher levels of ETR0.5 Max in the summer than in the spring and autumn +Fig. 3B), and suggests that low temperature limits realised PSII electron transport in spring and autumn while low light limits PSII electron transport in summer. Thus, the other sampling sites, measured onlyin March and Au- gust, showed low and consistent values for both dates. The seasonal pattern of qp determined at ®eld-ambi- ent light and ®eld-ambient temperatures +Fig. 4B) was similar to the pattern in qp at ETR0.5 Max. These patterns were probablydriven mainlybytemperature. NPQ determined at ®eld-ambient light was high throughout the winter but dropped in the summer to lows similar to that of the NPQ determined at ETR0.5 Max +Fig. 4C).

Coupling of gross CO2 uptake and PSII Fig. 3A±D PSII parameters of L. pulmonaria thallus measured at electron transport ®eld-ambient temperatures and near-optimal PPFD. PPFD at ETR0.5 Max +dashed line shows ®eld-ambient light level at Walker Road) +A), ETR +B), q at ETR +C) and NPQ at Maximum gross photosynthesis at ®eld-ambient tem- 0.5 Max p 0.5 Max peratures peaked in the late spring and earlysummer, ETR0.5 Max +D) between 7 February1999 and 18 February2000 for the Walker Road sampling site +open circles), and in March and and was lowest in the winter +Fig. 5), suggesting that August 1999 for the Fenwick Park +open triangles), Sugarloaf seasonal changes in gross CO2 uptake were dominated Brook +open diamonds) and EconomyMountain + ®lled squares) bythe changing measurement temperatures. Gross po- sampling sites. Electron-transport estimates assume 100% absor- bance of photosynthetically active photons by photosynthetic tential CO2 uptake measured at 20 °C, using a slightly pigments. Values are means ‹ SE with n=3 di€erent method, was lower in winter than in summer 62

Fig. 5 Maximum CO2 ¯uxes of L. pulmonaria thallus between 7 February1999 and 18 February2000 for the Walker Road sampling site +open circles), and in March and August 1999 for the Fenwick Park +open triangles), Sugarloaf Brook +open diamonds) and EconomyMountain + ®lled squares) sampling sites. Gross photosynthetic CO2 uptake was measured at ®eld-ambient tem- peratures. Values are means ‹ SE with n=3

metabolism, and demonstrates that this coupling breaks down at ambient winter temperatures and light intensi- ties. That the slope is actuallyslightlynegative in the March samples is not surprising. In these samples, even Fig. 4A±C Realised PSII parameters of L. pulmonaria thallus under the lowest light level for both electron transport measured at ®eld-ambient temperatures and ®eld-ambient light ±2 ±1 and gross CO2 uptake +5 lmol m s ), there was con- levels. Electron transport through PSII +A), qp +B) and NPQ +C) between 7 February1999 and 18 February2000 for the Walker siderable light stress and non-photochemical dissipation Road sampling site +open circles), and in March and August 1999 of excess excitation energy. Several photosynthetic or- for the Fenwick Park +open triangles), Sugarloaf Brook +open ganisms regularlyperform non-assimilatoryelectron diamonds) and EconomyMountain + ®lled squares) sampling sites. ¯ow, and Green et al. +1998) described it as a likely Electron-transport estimate assumes 100% absorbance of photo- synthetically active photons by photosynthetic pigments. Values photoprotective strategyexplaining the poor overall are means ‹ SE with n=3 coupling of electron transport and CO2 uptake observed in lichens. This or other dissipatorymechanisms would likelyincrease with increasing light stress, possibly causing the negative correlations observed it the winter +data not shown). CO2 uptake at 10 °C was only20± 35% of that at 20 °C. Only6% of daysbetween the samples in this study. November and March sampling dates had a daily maximum temperature exceeding 10 °C and on no date Potential photosynthesis at a constant, was 20 °C reached. The suppression of gross CO2 uptake at low temperature and the paucityof warm moderate temperature winter days imply little winter CO2 uptake occurs in this lichen population. Measured at 20°C, the PPFD for maximum potential The reliabilityof PSII electron transport as a pre- electron transport +Fig. 6A) and the potential for dictor of CO2 uptake in lichens has been questioned elsewhere +Green et al. 1998). The mechanistic minimum Table 1 Seasonal changes in the coupling of PSII electron trans- rate of electrons transported through PSII per CO2 ®xed port and gross CO2 uptake of L. pulmonaria thallus measured at ±1 is 4 electrons CO2 ®xed. Our summer data agreed well ®eld-ambient temperatures over a range of light levels with this, ranging from 3.82‹1.11 to 8.74‹1.05 elec- ±1 Site Month PSII e±/CO2 P trons CO2 with strong correlations for all sites. This rate dropped to zero or below in the winter, showing Walker Road +deciduous) August 3.82‹1.11 0.0064a that PSII electron transport was onlya good predictor March ±1.59‹1.48b 0.3082 a of CO uptake in the darker summer months +Table 1), Fenwick Park +deciduous) August 6.69‹1.21 0.0002 2 March ±0.69‹0.88b 0.4491 and that few PSII-generated electrons were being used Sugarloaf Brook +deciduous) August 8.74‹1.05 <0.0001a a for assimilating CO2 in the winter. We did not arti®cially March 2.09‹0.91 0.0429 ®x the regression equations to transit the origin; thus the EconomyMountain +evergreen) August 4.99‹0.94 0.0003 a slopes were free to follow the trend of the data regardless March ±1.91‹0.76b 0.0309a of background levels of electron transport or CO2 ex- aP-values indicate signi®cant correlation at 0.05 probabilitylevel change, or errors in either. This slope shows the coupling bRegression equations were not forced through zero, permitting between the extreme ends of photosynthetic energy negative slopes 63 maximum electron transport through PSII +ETRMax) pate light energyin the bright winter months, while the +Fig. 6B) were higher in the bright winter months than evergreen forest EconomyMountain samples showed in the dark summer months. Comparing PPFD at little seasonal change in their low NPQ capacity ETRMax, ETRMax and the actual ®eld-ambient PPFD +Fig. 6D). These results show that seasonal patterns in show similar seasonal patterns in all three. The other the potential photosynthetic performance of PSII at deciduous sites, sampled in March and August, were 20°C were distinct from the realised patterns achieved at similar to the Walker Road samples. In contrast, the ®eld-ambient temperatures. seasonal amplitude of change in the evergreen Economy Mountain samples was lower. Under saturating light at 20°C, qp also showed relativelyhigh values in the winter Seasonal changes in macromolecular composition and low values in the summer and autumn, indicating a greater potential to keep PSII open during the bright Chlorophyll content rose in the Walker Road popula- winter months at Walker Road +Fig. 6C). Under satu- tion +Fig. 7A) from basal levels of 85±130 lmol m±2 to a ±2 rating light at the other deciduous sites, qp also dropped peak of about 180 lmol m in the dark, late summer in the summer compared to the winter, but qp under months. The chlorophyll a/b ratio +Fig. 7B) averaged saturating light was consistent in winter and summer at about 4.9 throughout the year, but was generally higher the evergreen EconomyMountain site. Finally,changes +>5.5) under high-light conditions in the spring and fall in maximum NPQ at 20°C showed that the samples and showed a sustained low +<4.5) in the dark, late from the deciduous sites had a higher capacityto dissi- summer months, suggesting a higher proportion of an- tenna chlorophyll under the closed canopy. There were also changes in the concentration of RuBisCO LSU throughout the year in the Walker Road population. RuBisCO LSU remained at about 0.25 lmol m±2 from October through April, doubled in May, then decreased over the summer months to a minimum in September +Fig. 8A). There was thus a pattern of rapid accumulation in the late spring, at about +7.7 nmol m±2 day±1, followed bya sustained decayof RuBisCO LSU through the closed canopy summer months, at about ±3.8 nmol m±2 day±1,with slow re-accumulation in the fall after defoliation, at about +1.9 nmol m±2 day±1. The concentrations we

Fig. 6A±D PSII parameters of L. pulmonaria thallus measured at a ®xed 20°C. PPFD at ETRMax +dashed line shows ®eld-ambient light level) +A), ETRMax +B), maximum qp +C) and maximum NPQ +D) between 7 February1999 and 18 February2000 for the Walker Road sampling site +open circles), and in March and August 1999 for the Fenwick Park +open triangles), Sugarloaf Brook +open Fig. 7 Chlorophyll content of L. pulmonaria thalli estimated by diamonds) and EconomyMountain + ®lled squares) sampling sites. spectrophotometryof acetone extracts of thallus samples + A), and Electron-transport estimate assumes 100% absorbance of photo- chlorophyll a/b ratios +B) from 7 February1999 to 18 February synthetically active photons by photosynthetic pigments. Values 2000 for the Walker Road sampling site. Values are means ‹ SE are means‹ SE with n=3 with n=3 64 transport from energydissipation in the open-canopy cold months to light-limitation in the closed-canopy summer months. In the open-canopymonths chloro- phyll content and realised electron transport at ambient temperatures were low despite abundant ambient light. Realised electron transport peaked in spring and au- tumn when high light and moderate temperatures coin- cided. In the closed-canopywarm months, when available light was at a minimum the chlorophyll content rose but realised electron transport again fell. Ambient light was in excess of the PPFD to drive ETR0.5 Max for most of the year, except in the dark summer months. Kershaw and Webber +1984) have previouslyshown chlorophyllcontent and CO 2- exchange changes in Peltigera that were interpreted as causing higher light-harvesting eciencyin closed- canopypopulations. The photochemical performance of the photosystems and associated electron-transport chains are sensitive to their redox status, which in turn responds to the inte- grated light and temperature environment +Huner et al. 1998). We separated seasonal light and temperature ef- fects on electron transport and energydissipation by holding measurement temperature constant at 20°C. These data clearlyshowed that the potential for both electron transport and excitation-dissipation capacity Fig. 8 RuBisCO LSU content of L. pulmonaria thalli estimated were high in the bright winter months in the deciduous forest populations, and lower in the darker summer from immunoblots of tissue extracts +A), maximum gross CO2 uptake at ®eld-ambient temperatures per RuBisCO LSU +B) and months. Furthermore, the seasonal patterns of ambient RT-PCR products from 30- and 36-cycle reactions showing PPFD, PPFD for potential ETR and potential changes in the rbcL transcript pool +C) from 7 February1999 to Max 18 February2000. RT-PCR bands +marked by arrows) correspond ETRMax at 20°C were all similar, showing that the li- to sampling dates in graphs +A, B) and band migration was chens activelytracked ambient PPFD byacclimating downward. The lane at extreme right is a molecular weight their potential for electron transport, even though low standard showing bands of 1,000 bp and 750 bp, as indicated. temperature depressed their realised electron transport RuBisCO LSU content in Mayis signi®cantlyhigher than April, in winter. The dependence of these seasonal patterns in August, September, October, January2000 and February2000, potential photosynthetic performance on the light envi- and CO2/RuBisCO LSU values in April and Julyare signi®cantly higher than in February, March, January 2000 and February 2000 ronment was clear because onlysmall seasonal changes at the a = 0.05 level +one-wayANOVA). Values in A and B are were observed in samples from the permanentlyclosed- means ‹ SE with n=3 canopysite at EconomyMountain. The reversal of the seasonal patterns of realised electron transport mea- found were similar to values in lichens derived from sured at the ®eld-ambient temperature, however, showed previouslypublished data +Palmqvist et al. 1998), al- the greater and overwhelming importance of tempera- though rates of accumulation and decayhave not pre- ture in determining the realised performance. viouslybeen calculated in lichens. The rates of CO 2 Muir et al. +1998) have shown greatest growth rates uptake per RuBisCO LSU were in a narrow range from of L. pulmonaria during the winter months in western ±1 ±1 4.4 to 6.8 CO2 RuBisCO LSU s from April through Oregon USA. Unlike southeastern Canada +this study), November +Fig. 8B). CO2 uptake per RuBisCO LSU Oregon has wet, mild winters and relativelydrysum- was low in the winter because of the depressed rate of mers, and dailymaximum temperatures are rarelybelow CO2 uptake in the cold. Our qualitative data for rbcL 0°C. Thus, the lichens were exposed to more liquid water transcript levels +Fig. 8C) show a temporal pattern and metabolicallyactive temperatures in winter in the of transcript levels closelymatching RuBisCO LSU Muir et al. +1998) studythan our lichen population, concentration. which is rarelyexposed to drought conditions at meta- bolicallypermissive temperatures. In summer theyexist in the dark and moist understoreyof a swampymaple Discussion forest. In winter theyare deprived of liquid water for periods, but onlyin the extreme cold of dryarctic air In the deciduous forest populations of L. pulmonaria, masses, when theyare not metabolicallyactive due to the physiological and macromolecular data suggest there low temperature. We contend that in this population was a strategyshift in light harvesting and electron the seasonal changes in water availabilityare minor 65 compared to changes in light and temperature and the temperature caused great changes in measured photo- latter's strong e€ects on metabolic rates. synthetic rates, changes in the light environment also The CO2-exchange data supported the hypothesis of drove striking changes in the photosynthetic potential a seasonal strategyshift in photosynthesis.Green et al. throughout the seasons in these lichens. The photosyn- +1998) suggested lichens have a capacityto divert excess thetic apparatus responds to the interaction of light and electrons from PSII awayfrom CO 2 ®xation. In the temperature, however, so measurements of realised bright, cold March samples there was no meaningful acclimation must account for both variables. Measure- correlation between PSII electron transport and CO2 ment at the ®eld-ambient temperature and light is one uptake at anysite,suggesting extensive diversion and means to simplifythis interaction and is a physiologi- dissipation of electrons from PSII. In the warm, light- callyrelevant approach, given support bythe good limited August samples, however, the correlation be- correlations found between gross photosynthesis and tween PSII electron transport and CO2 uptake was electron transport. This approach is also faster and less strong, and mechanisticallyplausible, indicating that laborious than the matrix-response approach extolled by electrons from PSII were channelled ecientlyto CO 2. Kershaw +1984). Sundberg et al. +1997), however, have shown that thallus This studyencompassed one annual environmental growth in L. pulmonaria is not correlated with common cycle while these lichens cope with both open- and productivityindices such as net CO 2 uptake or chloro- closed-canopyenvironments byprofoundlyswitching phyll content. Overwhelming and variable fungal respi- photosyntheticstrategy throughout the year over many ration, especiallythe high respiration rates of years of life. This active acclimation allows them to L. pulmonaria +Sundberg et al. 1999), maya€ect thallus maintain photosynthetic performance under a wide growth rate more than does algal productivity+Sund- range of conditions. berg et al. 1997). The seasonal changes in CO2 uptake in terms of Acknowledgements This studywas supported bythe Mount Allison RuBisCO content also showed a strong di€erence be- UniversityRice Memorial Graduate Fellowship +T.D.B.M.), NSERC of Canada operating and equipment grants +D.A.C.), an tween winter and summer months. Gross CO2 uptake at NSERC Summer Undergraduate Scholarship +T.M.M.) and a ®eld-ambient temperatures per RuBisCO LSU was near Mount Allison UniversitySummer Research Award +L.A.D.). We zero in winter but then remained high and fairlycon- thank S. Patterson for preliminarywork on RuBisCO extraction stant from late April to late November, through a 4-fold from L. pulmonaria, and K. Palmqvist for discussions. seasonal change in RuBisCO content and through great changes in the light and temperature environment. Furthermore, the calculated rates of gross CO2 uptake References per RuBisCO LSU in the period between April and

November are near maximum published turnover rates Badger MR, Andrews TJ +1987) Co-evolution of RuBisCO and CO2 for this enzyme +Badger and Andrews 1987), suggesting concentrating mechanisms. In: Biggens J +ed) Progress in pho- that the RuBisCO pool was nearlyfullyactivated. Ex- tosynthesis research, vol III. Nijho€, Dordrecht, pp 601±609 Barnes J, Balaguer L, Manrique E, Elvira S, Davison A +1992) A amination of the RuBisCO LSU and rbcL transcript reappraisal of the use of DMSO for the extraction and deter- patterns shows there was a variable allocation of re- mination of chlorophylls a and b in lichens and higher plants. sources to RuBisCO that maintained and maximised the Environ Exp Biol 32:85±100 rate of CO ®xed per RuBisCO LSU. Through the Bilger W, Schreiber U +1986) Energy-dependent quenching of dark- 2 level chlorophyll ¯uorescence in intact leaves. Photosynth Res spring, there was abundant light, warming temperatures 10:303±308 and increasing liquid water availabilityas snow melted Brown D, MacFarlane JD, Kershaw KA +1983) Physiological-en- and drier winter air gave wayto spring, leading to a vironmental interactions in lichens. XVI. A re-examination of resaturation respiration phenomena. New Phytol 93:237±246 great potential for CO2 ®xation supported byaccumu- lation of rbcL transcript and rapid accumulation of Coxson D, Kershaw K +1984) Low-temperature acclimation of net photosynthesis in the crustaceous lichen Caloplaca trachyphylla. RuBisCO LSU. After canopyclosure, the rbcL tran- Can J Bot 62:86±95 script concentration dropped, and RuBisCO LSU fell Farmer AM, Bates JW, Bell JNB +1991) Comparisons of three steadilythrough the summer. After the leaf canopy woodland sites in NW Britain di€ering in richness of the opened again in the autumn, rbcL transcript and epiphytic Lobarion pulmonariae community and levels of wet acidic deposition. Holarct Ecol 14:85±91 RuBisCO LSU re-accumulated to a second peak in the Feige G, Jensen M +1987) Photosynthetic properties of lichens earlywinter. This suggests that RuBisCO LSU content stored at ±25°C for several years. Bibl Lichenol 25:319±323 is dynamically modulated in part by transcript concen- Gauslaa Y, Solhaug K +1996) Di€erences in the susceptibilityto tration and in response to seasonal environmental light stress between epiphytic lichens of ancient and young boreal forest stands. Funct Ecol 10:344±354 changes. Gauslaa Y, Solhaug K +1999) High-light-intensitydamage to the The large seasonal shifts in photosynthetic physiolo- foliose lichen Lobaria pulmonaria within a natural forest: the gy, macromolecular composition and rbcL transcription applicabilityof chlorophyll¯uorescence methods. Lichenologist in the photobiont of L. pulmonaria occurred despite the 32:271±289 discontinuous nature of lichen physiology, which inter- GentyB, Briantais J-M, Baker N +1989) The relationship between the quantum yield of photosynthetic electron transport and acts with the environment onlyduring hydration epi- quenching of chlorophyll ¯uorescence. Biochim Biophys Acta sodes of variable duration and frequency. Although 990:87±92 66

Green T, Schroeter B, Kappen L, Seppelt R, Maseyk K +1998) An of the Chlorophyta +Scenedesmus and Chlorella) in laboratory assessment of the relationship between chlorophyll a ¯uores- and outdoor cultures: changes in chlorophyll ¯uorescence cence and CO2 gas exchange from ®eld measurements on a quenching and the xanthophyll cycle. Planta 209:126±135 moss and lichen. Planta 206:611±618 Muir P, Shirazi A, Patrie J +1998) Seasonal growth dynamics in the Honegger R +1991) Functional aspects of the lichen symbiosis. lichen Lobaria pulmonaria. Bryologist 100:458±464 Annu Rev Plant Physiol Plant Mol Biol 42:553±578 Palmqvist K, Campbell D, Ekblad A, Johansson H +1998) Photo- Huner NPA, OquÈ ist G, Sarhan F +1998) Energy-balance and acc- synthetic capacity in relation to nitrogen content and its parti- limation to light and cold. Trends Plant Sci 6:224±230 tioning in lichens with di€erent photobionts. Plant Cell Environ Jordan W +1973) The genus Lobaria in North America north of 21:361±372 Mexico. Bryologist 76:225±251 Ricker W +1975) Computation and interpretation of biological Kershaw K +1984) Seasonal photosynthetic capacity changes in statistics of ®sh populations. Bulletin 191. Fisheries Research lichens: a provisional mechanistic interpretation. Lichenologist Board of Canada, Ottawa 16:145±171 Rose F +1988) Phytogeographical and ecological aspects of Loba- Kershaw K, Webber M +1984) Photosynthetic capacity changes in rion communities in Europe. Bot J Linn Soc 96:69±79 Peltigera I. The synthesis of additional photosynthetic units in Schreiber U, Bilger W, Neubauer C +1994) Chlorophyll ¯uores- P. praetextata. New Phytol 96:437±466 cence as a nonintrusive indicator for rapid assessment of in vivo Lange OL, Leisner JMR, Bilger W +1999) Chlorophyll ¯uorescence photosynthesis. In: Schultz E-D, Caldwell MM +eds) Ecological characteristics of the cyanobacterial lichen Peltigera rufescens studies 100: ecophysiology of photosynthesis. Springer, Berlin, under ®eld conditions II. Diel and annual distribution of met- Heidelberg, New York pp 49±70 abolic activityand possible mechanisms to avoid photoinhibi- Sundberg B, Palmqvist K, Esseen P-A, Renhorn K-E +1997) tion. Flora 194:413±430 Growth and vitalityof epiphyticlichens. II. Modeling of carbon Larson D, Kershaw K +1975) Acclimation in arctic lichens. Nature gain using ®eld and laboratorydata. Oecologia 109:10±18 254:421±423 Sundberg B, Ekblad A, NaÈ sholm T, Palmqvist K +1999) Lichen MacFarlane J, Kershaw K, Webber M +1983) Physiological-envi- respiration in relation to active time, temperature, nitrogen and ronmental interactions in lichens. XVII. Phenotypic di€erences ergosterol concentrations. Funct Ecol 13:119±125 in the seasonal pattern of net photosynthesis in Cladonia van Kooten O, Snel J +1990) The use of chlorophyll ¯uorescence rangiferina. New Phytol 94:217±233 nomenclature in plant stress physiology. Photosynth Res Masojõ dek J, Torzillo G, Koblõ zek M, KopeckyJ, Bernardini P, 25:147±150 Sacchi A, Komenda J +1999) Photoadaptation of two members