Helbling, E. Walter, Anita GJ Buma, Peter Boelen, Han J. van der Strate, M. Valeria Fiorda

13 Pages·2011·0.93 MB·German

Limnol.Oceanogr.,56(4),2011,1330–1342
E2011,bytheAmericanSocietyofLimnologyandOceanography,Inc.
doi:10.4319/lo.2011.56.4.1330
Increase in Rubisco activity and gene expression due to elevated temperature partially
counteracts ultraviolet radiation–induced photoinhibition in the marine diatom
Thalassiosira weissflogii
E. Walter Helbling,a,* Anita G. J. Buma,b Peter Boelen,b Han J. van der Strate,b
M. Valeria Fiorda Giordanino,a and Virginia E. Villafan˜ea
aEstacio´ndeFotobiolog´ıa PlayaUnio´nandConsejoNacional deInvestigacionesCient´ıficasy Te´cnicas, Casilla deCorreos No. 15
Rawson, Chubut,Argentina
bDepartmentofOceanEcosystems,EnergyandSustainabilityResearchInstituteGroningen,UniversityofGroningen,TheNetherlands
Abstract
Weperformedoutdoorexperimentstoevaluatetheeffectoftemperatureonphotoinhibitionpropertiesinthe
cosmopolitan diatom Thalassiosira weissflogii. Cultures were exposed to solar radiation with or without
ultravioletradiation(UVR,280–400nm),UV-A(320–400nm),andUV-B(280–320nm)atboth20uCand25uC.
FourpossiblecellularmechanismsinvolvedinUVRstressweresimultaneouslyaddressed:carbonincorporation,
chlorophyll a fluorescence of photosystem II, xanthophyll cycle activity, and ribulose-1,5-biphosphate
carboxylase:oxygenase (Rubisco) activity and gene expression. Experiments consisted of daily cycles (i.e., the
daylight period) and short-term incubations (i.e., 1 h centered on local noon). Samples incubated at 25uC had
significantly less UVR-induced inhibition of carbon fixation and effective photochemical quantum yield
comparedtothoseincubatedat20uC.At25uCRubiscoactivityandgeneexpressionweresignificantlyhigherthan
at20uC.ThehigherRubiscoactivityandgeneexpressionwerecorrelatedwithlessdissipationofexcessenergy,
evaluatedvianon-photochemicalquenching,andthede-epoxidationstateofthexanthophyllpigments,asmore
photons could be processed. An increase in temperature due to climate change would partially counteract the
negative effects of UVR by increasing the response of metabolic pathways, such as those involved in Rubisco.
This, in turn, may haveimportant consequences for the ecosystem, as higher production (dueto more Rubisco
activity) could beexpected undera scenario ofglobalwarming.
Marinephytoplanktonareresponsibleforabout40%of Most studies addressing the effects of UVR on
global primary productivity (Behrenfeld et al. 2006) phytoplankton photosynthesis have focused on carbon
through the process of photosynthesis. In addition to their incorporation or oxygen evolution, many of which
notable importance as primary producers, phytoplankton demonstrate the relative importance of UV-A (315–
alsohaveakeyroleintakinguptheexcesscarbonthathas 400 nm) in causing photoinhibition (reviewed by Villafan˜e
been released from anthropogenic activities since the time et al. 2003). Other studies have focused on the effects of
oftheIndustrialRevolution(Beardalletal.2009).Thereis, UVR on the reaction center of photosystem II (PSII).
therefore, an obvious interest in assessing the relative Measurementsofchlorophylla(Chla)fluorescenceofPSII
importance of the factors that affect physiological charac- inphytoplanktonhavedemonstratedaclassicalresponseof
teristics as well as the possible effects of climate change inhibition of the PSII yield under exposure to photosyn-
upon these organisms. thetic activeradiation(PAR, 400–700 nm),aresponse that
One of the factors that affect phytoplankton photosyn- is generally further reduced by UVR. These responses are
thesis is ultraviolet radiation (UVR, 280–400 nm). In- typicallyfollowedbypartialorcomplete recoveryoncethe
creased interest in UVR-driven effects has occurred since stress is removed (Villafan˜e et al. 2007; Roncarati et al.
the discovery of the ozone ‘hole’ over the Antarctic 2008;Halacetal.2009).Tominimizethedamageproduced
continent (Farman et al. 1985), with the concomitant on PSII, phytoplankton deploy a suite of mechanisms that
increase in the most energetic waveband (i.e., UV-B [280– can adjust photosynthesis on short time scales (seconds to
315 nm]). Since then, a vast body of literature has been minutes), such as the xanthophyll cycle (Olaizola and
produced with regard to the effects of UV-B on polar Yamamoto 1994; Dubinsky and Stambler 2009; Van De
phytoplankton photosynthesis (see review by Vernet and Poll et al. 2010), although the effectiveness of the
Smith [1997]). More recently, though, it has been recog- xanthophyll cycle under high UVR levels is under debate
nized that natural levels of UVR cause significant effects (VanDePollandBuma2009).Inaddition,statetransitions
with regard to phytoplankton photosynthesis around the (reversible phosphorylation-mediated antenna size reduc-
globe(seereviewsbyVillafan˜eetal.[2003]andbyHa¨deret tionofPSIIbyredistribution ofenergytoPSI)mayoccur,
al. [2007]). Yet phytoplankton may respond to UVR to thereby contributing to non-photochemical fluorescence
various degrees based on their specific sensitivity and their quenching (NPQ; Finazzi et al. 2006). Only a few UVR
potential to photoacclimate. studies have focused on particular key enzymes associated
with carbon acquisition, such as ribulose-1,5-biphosphate
*Corresponding author:[email protected] carboxylase:oxygenase (Rubisco). Lesser et al. (1996)
1330
Temperature, UVR affect T. weissflogii 1331
found a significant reduction in the total Rubisco pool of was grown in 4-liter Erlenmeyer flasks (UV opaque) in f/2
natural Antarctic phytoplankton when samples were medium (Guillardand Ryther1962) witha photoperiodof
exposed to UVR; such a reduction was also found for the 12:12-hlight:darkintwo growthchambers (Sanyomodel
dinoflagellate Prorocentrum micans (Lesser 1996a). Bou- ML 350 and Minicella) at different temperatures: Prior to
chard et al. (2008) found a reduced abundance of the large theoutdoorexposures,cultureswereacclimatedfor2weeks
subunitsofRubiscoinphytoplanktonsubjectedtonutrient to 20uC or 25uC at a saturating irradiance of 235 mmol
limitation and supplemental UV-B as well as exacerbated quanta m22 s21 (51 W m22) of PAR. We used this PAR
photoinhibition,ascomparedtophytoplanktonexposedto level during the acclimation period because it was
ambient surface irradiance. previously determined by us to be above the saturation
While there is a negative effect of UVR on the light value (I ) (V. E. Villafan˜e unpubl.). Nevertheless, in
k
photosyntheticprocess,theinteractionofthisvariablewith the water column, cells would experience irradiance above
otherscanaltertheoverallpicture,assomevariablescanact andbelowthatlevel,dependingnotonlyonthepositionof
synergisticallyorantagonistically(Dunne2010).Forexam- thecellsinthewatercolumnbutalsoonotherfactors,such
ple, in cultures exposed to UVR, nutrient addition resulted as the depth of the upper mixed layer and the attenuation
in higher effective photochemical quantum yield as com- coefficientinthewater column,amongothers.Thelightin
pared to non-enriched cultures (Marcoval et al. 2007). both chambers was provided by 10 cool-white fluorescent
Studies carried out by Gao et al. (2009) indicated that lamps(Philipsdaylight,1.2mlong,40W),andphotonflux
enhanced CO concentrations exacerbated UVR stress on density was measured with a spherical micro quantum
2
photosynthesis of calcified organisms; however, Sobrino sensor (Walz GmbH, model US-SQS/WB). During the
et al. (2005) found opposite results when evaluating the acclimation period, the cultures were diluted every other
photosynthesis of two species of Nannochloropsis. Temper- daytokeepthecellsintheexponentialgrowthphase(Chla
atureisalsoavariablethatcanconsiderablyinfluenceUVR between 15 and 40 mg L21). The growth rates (m) of T.
responsesinphytoplankton.Inparticular,elevatedtemper- weissflogii at 20uC and 25uC were 0.63 d21 (standard
atureseemstoameliorateUVR-inducedstress(Sobrinoand deviation[SD]50.04)and0.8d21(SD50.05),respectively.
Neale2007;Halacetal.2010)andissuggestedtoberelated Aftertheacclimationperiod,theculturesweretransferredto
to enhanced enzymatic conversion of xanthophyll cycle UVR-transparent polycarbonate (XT tubes, Rohm and
pigments(Demming-AdamsandAdams1992),thepotential Haas) or to quartz vessels and were exposed outdoors, as
enhancementofRubiscoactivity,andtoenhancedD1repair described below. The experiments were done in late spring
(Bouchard et al. 2006). A recent study (Halac et al. 2010) 2008 (24–28 November) at the EFPU (43u18942.380S,
showed the responses of two marine diatoms (Chaetoceros 65u02930.420W)inPatagonia,Argentina.
gracilis and Thalassiosira weissflogii) exposed to UVR and
elevated temperature using pulse amplitude modulated Experimentationandradiation treatments—Twotypes of
(PAM)fluorescencetechniques.Theauthorsconcludedthat experiments were carried out: full day–length exposures
increased temperature benefited C. gracilis by decreasing and short-term (i.e., 1-h) exposures. The daily cycle
photoinhibition via dissipation of excess energy (NPQ). experiments were carried out on 24 November, while the
However,thisprocess,althoughmeasurable,wasmuchless short-term exposures were done on 25 November (day of
significant in T. weissflogii, and therefore it did not fully the year [DOY] 329). The results obtained from the three
explainthevariability observedin theeffectivephotochem- daylight experiments were similar, so in this article we are
icalquantumyieldinthisspecies. only presenting the data derived from the daily cycle
The temperature dependence of physiological changes experimentdoneon24November(DOY238)andfromthe
underUVRstressisofspecialinterestinthecontextofglobal short-term experiment performed on 25 November (DOY
change.Forthisreason,theaimofthepresentstudywasto 239), as during these two days we had a complete data set
thoroughly evaluate the physiological background of the (i.e., more variables were measured).
temperature dependence of natural UVR stress on photo-
synthesisinthecosmopolitandiatomT.weissflogii(Grunow) Dailycycles:CulturesofT.weissflogiiweredilutedtocell
G.Fryxell&Hasle.Tothisend,wesimultaneouslyaddressed densitiesmeasuringbetween83103cellsmL21and43104
four possible cellular mechanisms involved in UVR stress: cells mL21 just prior to the outdoor exposure and were
carbon incorporation, Chl a fluorescence of PSII, xantho- transferred to various vessels, as follows: (1) 800-mL UVR-
phyllcycleactivity,andRubiscoactivityandgeneexpression. transparent polycarbonate for Rubisco measurements; (2)
Tothebestofourknowledge,thisstudyrepresentsthefirstin 500-mLquartztubesforpigmentanalysis;(3)50-mLquartz
which four cellular ‘targets’ are simultaneously considered tubes for Chl a fluorescence measurements, and (4) 20-mL
under natural solar conditions in order to understand the quartztubesforcarbonincorporation.Forallmeasurements
UVR effect from the ‘light cycle’ (antennae) to the ‘dark (with the exception of Rubisco, due to volume constraints),
reactions’(Calvin–Bensoncycle). three radiation treatments were implemented: (1) full solar
radiation—PAB treatment: PAR + UV-A + UV-B (280–
Methods 700nm)—uncoveredvessels;(2)PAtreatment:PAR+UV-A
(320–700 nm)—vessels covered with UV cut-off filter foil
Culture conditions and study area—T. weissflogii (Gru- (Montagefolie, No. 10155099, Folex); and (3) P treatment:
now) G. Fryxell & Hasle (Microalgal Culture Collection, only PAR (400–700 nm)—vessels covered with Ultraphan
Estacio´ndeFotobiolog´ıaPlayaUnio´n[EFPU],Argentina) film (UV Opak, Digefra). For Rubisco measurements only
1332 Helbling et al.
the PAB and P treatments were done. The transmission immediately after sampling, without any dark adaptation.
spectraofthesefiltersandmaterialsarepublishedelsewhere The effective photochemical quantum yield (Y) was
(Figueroaetal.1997;Villafan˜eetal.2003). calculated using the equations of Genty et al. (1989) and
Thevesselswereplacedinanoutdoorthermostaticwater Weis and Berry (1987), as follows:
bath(Fr´ıo21)equippedwithtwoindependenttemperature
circuits. The experimental temperatures were set to 20uC Y~DF:Fm’ ~(Fm’ {Ft)=Fm’ ð1Þ
and25uC(61uC).Exposurestartedat08:00handendedat
where F’ is the maximum fluorescence induced by a
17:00h.Every60–90minsampleswerewithdrawnfromthe m
saturating light pulse (ca. 5300 mmol quanta m22 s21 in
water baths, as further described below.
0.8s)andF isthecurrentsteady-statefluorescenceinduced
t
byaweakactiniclight(82Wm22)inlight-adaptedcells.The
Short-term exposures: These 1-h incubations were
NPQofChlafluorescencewasdeterminedbymeasuringF
centeredonlocalnoon,withtreatmentsasexplainedabove m
at the start of the experiments in samples maintained in
fordailycycles.Theoutdoorexposuresstartedatca.11:30
darknessandF’ duringtheexposuretime,asfollows:
h and 13:30 h, and every 10–15 min samples were m
withdrawn, transported to the laboratory under dim light, NPQ~(F {F’ )=F’ ð2Þ
m m m
and immediately processed, as further described below.
Two short-term exposures were done during the study There were no significant differences between NPQ values
period, and radiation treatments were PAB and P only. calculated in this way and those obtained directly using the
PAM fluorometer. As a result, we used the values obtained
Sampling protocol, analysis, and measurements—Solar directlywiththeinstrument.
radiation: Incident solar radiation over the study area was
monitored continuously using a broadband European Pigment analysis: Every 90 min, duplicate 50-mL
Light Dosimeter Network (ELDONET) radiometer (Real aliquots were obtained for pigment analysis (in daily cycle
Time Computers) permanently installed on the roof of the experiments only). Samples were immediately filtered on
EFPU.TheinstrumentmeasuresUV-B(280–315nm),UV- GF/F filters (Whatman, 25-mm diameter) by mild vacuum
A (315–400 nm), and PAR (400–700 nm) with a frequency (, 10 mm Hg) and were then frozen and stored in liquid
of one reading per second and stores the minute-averaged nitrogen.AftertransportationtoTheNetherlandsinliquid
values for each channel. nitrogen, the samples were freeze-dried and pigments were
extracted in 90% acetone overnight in darkness at 4uC.
Carbonincorporation:Atotalof36quartztubes(20mL) Pigments were resolved by high-performance liquid chro-
per temperature level were placed in the water bath at the matography (HPLC; Waters) with a C18 5-mm Delta Pack
start of both short-term and daily cycle experiments, with reversed-phase column (Milford) following the technique
12tubessubjectedtoPAB,12tubestoPA,and12tubesto described in Van Leeuwe et al. (2006). Pigment identifica-
the P treatments. Each tube was inoculated with 0.1 mL tion was done by retention time and diode array
containing 5 mCi (0.185 MBq) of 14C-labeled sodium spectroscopy (Waters). Chl a, fucoxanthin, diadinoxanthin
bicarbonate (ICN Radiochemicals). After every 90 min (DD), and diatoxanthin (Dt) standards (DHI) were used
(dailycycles)or15min(shortterm)ofexposure,triplicates for quantification. Cellular pigment concentrations were
tubes from each radiation treatment were withdrawn and calculated from cell counts and extraction volume.
filtered immediately onto a Whatman GF/F filter (25-mm Chl a concentration was measured twice, once for
diameter).Thefilterswerethenplacedin7-mLscintillation pigment fingerprinting (using HPLC, as described above)
vials,exposedtoHClfumesovernight,anddried.Afterthis and once for calculations of carbon assimilation numbers,
step, scintillation cocktail (HiSafe’ 3, Perkin Elmer) was using fluorometry and spectrophotometry. HPLC-derived
added, and the samples were counted using a liquid Chl a concentration and fluorometer-derived Chl a data
scintillation counter (Holm-Hansen and Helbling 1995). werehighlycorrelated.Atthebeginningofeachexperiment
aliquots of 50–100 mL of sample were filtered onto
Chl a fluorescence parameters: A total of six quartz Whatman GF/F filters (25-mm diameter); the filters were
tubespertemperaturelevelwereexposedtosolarradiation then placed in centrifuge tubes (15 mL) with 5 mL of
under the three radiation treatments, giving duplicates per absolutemethanol(Holm-HansenandRiemann1978).The
treatment. Samples (5 mL) were taken every ca. 90 min tubes containing the methanolic extract and filters were
(daily cycles) or 10 min (short term), and in vivo Chl a placed in a sonicator for 20 min (20uC or 25uC, according
fluorescence parameters of the PSII were determined using to the experiment) and then in the dark (4uC) for at least
a portable PAM fluorometer (Walz, model Water-ED 1h.Aftertheextractionperiodthesamplewascentrifuged
PAM). At noon (only during the daily cycles) and at the (15 min at 1750 3 g) and scanned (250–750 nm) in a
end of the exposure period (in both daily cycles and short- Hewlett Packard diode-array spectrophotometer (model
term exposures) subsamples from each tube were placed HP 8453 E) using a 5-cm–path length cuvette; Chl a
inside the growth chambers (at 20uC and 25uC), and the concentration was calculated using the equation of Porra
recoveryoffluorescenceparametersindimlight(,10mmol (2002). The same sample was used to calculate Chl a
quanta m22 s21) was followed for 12 h, measuring each concentration from the fluorescence of the extract (Holm-
sampleeveryca.30min(indailycycles)or10min(inshort- Hansenetal.1965)beforeandafteracidification(1molL21
term exposures). Each sample was measured six times HCl)usingafluorometer(TurnerDesigns,modelTD700).
Temperature, UVR affect T. weissflogii 1333
The fluorometer was calibrated using purified Chl a from synthesizedbyBiolegioB.V.(Nijmegen).Amplificationwith
Anacystis nidulans (Sigma No. C 6144). theseprimersresultedintheexpected102–basepairfragment
Rubiscoactivity:Foreachradiationcondition,triplicate for T. weissflogii. Quantitative RT-PCR (qRT-PCR) was
samples (100 mL) were filtered onto 2.0-mm pore-size performedusinganiCycler1(BioRad).
polycarbonate filters (Osmonics), immediately frozen in cDNA amplification reactions were performed in tripli-
liquidnitrogen,andstoredat280uCuntilfurtheranalysis. cate for every sample in a final reaction mixture of 15 mL,
Rubiscoactivitywasdeterminedfollowingtheprocedureof containing 1 mL cDNA (10ng mL21), 7.5 mL Power SYBR
Gerard and Driscoll (1996). Crude enzyme extracts were Green Master Mix (Applied Biosystems), 0.5 mL of both
prepared by transferring the filters to test tubes containing forward and reverse primer (20 mmol), and 5.5 mL H O.
2
ice-cold extraction buffer (25 mmol KHCO , 20 mmol The amplification reaction was executed as follows: 2 min
3
MgCl ,0.2mmolethylenediaminetetraaceticacid[EDTA], at50uCand10minat95uC,followedby40cyclesof15sat
2
5 mmol dithiothreitol [DTT], 0.1% Triton X100, and 95uC and 60 s at 60uC. Melting curves for excluding
50mmol4-(2-hydroxyethyl)-1-piperazineethanesulfonicac- nonspecific PCR side products were acquired by heating
id [HEPES]-KOH, pH 8.0). The cells were disrupted by theproductsfor15sat95uCand1minat55uC,subsequent
sonication, after which the extracts were centrifuged for quick heating to 60uC, followed by a more steady heating
15min(20,0003g)at4uC.Fiftymicrolitersofsupernatant to 95uC with increments of 0.5uC:10 s21. PCR amplifica-
wasusedforproteindetermination(BioRadProteinAssay), tions without template (solely H O) and RNA samples
2
and 150 mL was used for the Rubisco activity assay. The without treatment with reverse transcriptase were used as
Rubiscoassaymixture(pH8.0)contained25mmolKHCO , controls. Four cDNA dilutions (1, 10, 100, and 1000 ng)
3
20 mmol MgCl , 0.2 mmol EDTA, 5 mmol DTT, 50 mmol were used to estimate the efficiency in a validation
2
HEPES-KOH,3mmoladenosine59-triphosphate,0.2mmol experiment. Individual PCR efficiency was calculated
nicotinamide adenine dinucleotide (NADH), 5 mmol phos- (Ramakers et al. 2003) to bypass the assumption that in
phocreatine, 22 units of creatine phosphokinase, 9 units of all samples the PCR efficiency is constant. The triplicate
glyceraldehyde-3-phosphate dehydrogenase, and 18 units of PCRreactionspersamplewereaveragedbeforeperforming
3-phosphoglycericphosphokinase. the 22DDCT method (Livak and Schmittgen 2001). Relative
The mixture was placed in a Cary 3E UV:Vis quantification was presented as the fold in change in
spectrophotometer, and after temperature equilibration to mRNA transcript level relative to the sample with the
20uC the background oxidation of NADH (absorbance at lowest number of mRNA transcripts present.
340nm)wasfollowedforseveralminutes.Thereactionwas
started by adding 2 mmol (final concentration) ribulose- Data treatment—The data are reported either as mean
1,5-bisphosphate,andthetimecourseofNADHoxidation and half mean range (when duplicate samples or measure-
wasrecordedbythedecreaseinabsorbanceat340nm.The mentsweredone)orasmeanandSDwhenmorereplicates
background oxidation of NADH was subtracted from the were obtained. The inhibition by UV-B, UV-A, and UVR,
measured activity, which was expressed as the declining respectively, was calculated as follows:
absorbance per milligram of total protein per second
UVB (%)~100(P {P ):P ð3Þ
(mAbs340 mg protein21 s21). inh PA PAB P
Rubisco gene expression: For each radiation condition, UVA (%)~100½(P {P )(cid:2):P ð4Þ
inh P PA P
triplicate samples (100 mL) were filtered over 2.0-mm pore-
sizepolycarbonate filters(Osmonics),immediately frozen in UVR (%)~100½(P {P )(cid:2):P ð5Þ
liquid nitrogen, and stored at 280uC until later analysis. inh P PAB P
Rubisco gene expression was determined by quantitative where P , P , and P represent either the carbon
P PA PAB
real-time polymerase chain reaction (RT-PCR) of the large fixation or the effective photochemical quantum yield of
subunit of Rubisco. RNA was isolated using TRIzolR samples in the P, PA, and PAB treatments, respectively.
Reagent (1 mL per sample), according to the manufacturer A one-way repeated-measurements ANOVA test was
(Invitrogen Life Technologies). The isolated RNA was used to determine differences among radiation treatments
subsequently treated with DNase I (Amersham Bioscience). and temperatures. A two-way ANOVA test was used to
RNA concentrations and purity were determined on a determine interactions between factors (i.e., irradiance and
Nanodrop ND-1000 spectrophotometer (NanoDrop Tech- temperature), both using a 95% confidence limit (Zar
nologies). First strand complementary DNA (cDNA) was 1999). When the data did not follow homoscedasticity, the
synthesizedfrom65ngofDNA-freeRNAusing500ngoligo non-parametric Kruskal–Wallis test was used to assess
(dT) 12, 18 primer, and Superscript III H reverse transcrip- significant differences between the samples.
tase (Invitrogen Life Technologies). cDNA concentrations
were also determined on the Nanodrop ND-1000 spectro- Results
photometer.BasedonanalignmentofknownalgalRubisco
large subunit sequences, primers for T. weissflogii were Solar radiation was variable during the experimental
designed using the program Primer3 (Rozen and Skaletsky period,withamaximumPARlevel(Fig. 1A)ofalmost500
2000). The designed Rubisco forward (59-AAAATGGGT- Wm22,whileUV-A(Fig. 1B)andUV-B(Fig. 1C)reached
TACTGGGATGCTTC-39) and Rubisco reverse (59-CAG- maximum values of 70 W m22 and 2 W m22, respectively.
CAGCTTCTACTGGATCTACACC-39) primers were Mostofthe day-to-day variabilitywas due tovariations in
1334 Helbling et al.
Fig. 1. Irradiancelevels(inWm22) during theexperimentalperiod(24–28 November2008;
DOY328to332)for(A)PAR(400–700nm),(B)UV-A(315–400nm),and(C)UV-B(280–315nm).
cloud cover, ranging from an almost completely covered fixation of samples incubated at 20uC was immediately
day (DOY 332) to an almost cloudless day (DOY 329). inhibited after exposure, and the differences between the
Thedailycourseofcarbonincorporationmeasuredduring PAB and P treatments significantly increased as the
acloudyday(DOY328)showedsignificantdifferences(p, incubation progressed (Fig. 2C). At 25uC (Fig. 2D)—and
0.05)betweenradiationtreatmentsat20uCtowardtheendof despite the fact that UVR inhibition was significant (p ,
the incubation (Fig. 2A). A much higher rate of carbon 0.05)—the differences were smaller than those observed at
incorporation was attained at 25uC, and no significant 20uC; again, a higher rate of carbon incorporation was
inhibition was observed among radiation treatments at this attainedatthehightemperature.
temperature (Fig. 2B). During the short-term incubations The differences between exposures at both temperatures
(Fig. 2C,D),whichweredoneduringaclearday(DOY329, were further evident when considering the assimilation
see Fig. 1), the cultures of T. weissflogii were much more numbers at the end of the incubation period (Fig. 3): The
inhibited than during the previous day (DOY 328): Carbon assimilationnumbersat25uCweresignificantlyhigher(p,
Temperature, UVR affect T. weissflogii 1335
Fig. 2. Carbonfixation(inmgCL21)ofThalassiosiraweissflogiiduringexperimentscarried
out during spring 2008 in Patagonia. Results of daily cycle (DOY 328) are expressed as (A)
carbon fixation of samples incubated at 20uC and (B) carbon fixation of samples incubated at
25uC.Representativeresultsofashort-termincubation(DOY329)areexpressedas(C)carbon
fixationofsamplesincubatedat20uCand(D)carbonfixationofsamplesincubatedat25uC.The
symbolsrepresentthemeanoftriplicatesamplesunderthedifferentradiationtreatments,while
the vertical lines are the SD. Solid symbols indicate samples exposed at 20uC, whereas open
symbols indicate samples exposedat25uC.
0.05)thanthoseat20uCforallradiationtreatmentsandfor P treatment. Recovery of Y was not evident during the
both types of incubations. During the daily cycle (Fig. 3A), exposureperiod(i.e.,duringtheafternoon),butitdidoccur
carbon fixation in the PAB treatment at 20uC was when samples were transferred to dim light, either at noon
significantly reduced (p , 0.05) by UVR (i.e., 22%, as or at the end of the exposure period. When T. weissflogii
compared to the P treatment), with UV-B and UV-A samples were incubated at 20uC during short-term expo-
accountingfor38%and62%,respectively,oftotalinhibition. sures, but at higher irradiances (mean UVR of 52.3 6 1.1
Basedonassimilationnumbers,theinhibitionduetoUVRat W m22, Fig. 4B), the observed inhibition of Y was
theendoftheshort-termincubations(Fig. 3B)was70%and significantly higher, and recovery was not complete even
36%forculturesat20uCand25uC,respectively. after 1 h under dim light. At 25uC and during both daily
Effective photochemical quantum yield (Y) of T. and short-term exposures (Fig. 4C,D), Y values were
weissflogii samples measured during the daily cycle significantly higher than those at 20uC. The recovery of
(Fig. 4A,C,E) and short-term exposures (Fig. 4B,D,F) samples during the daily cycle was complete when the
showedsignificantdifferences(p,0.05)betweenradiation samples were transferred to dim light (Fig. 4C); however,
treatments as well as between incubation temperatures. during the short-term exposure the recovery was not
During the daily cycle, and at all times during exposure complete, with Y values at the end being significantly
(mean UVR of 26.1 6 8.9 W m22), samples incubated at lower than at the beginning of incubation (Fig. 4D). The
20uC (Fig. 4A) under UVR (i.e., PAB and PA treatments) calculated UVR inhibition of Y for both the daily cycle
showed a significantly lower Y compared to those in the (Fig. 4E) and the short-term exposures (Fig. 4F) showed
1336 Helbling et al.
different temperatures were maintained throughout the
exposure. There were no significant differences in DEPS
among radiation treatmentsat 20uC, but DEPS in samples
underthePABat25uCweresignificantlyhigherthanthose
in the P treatment (Fig. 5A).
Photoprotective capacity, expressed as the ratio of
xanthophyll to photosynthetic pigments, was significantly
higherat25uCthanat20uC(Fig. 5B)beforeandduringthe
outdoor exposures (Table 1). At both temperatures, the
ratiochangedduringtheoutdoorexposures,infavorofthe
xanthophyll pigment pool. However, as was the case with
the DEPS responses, there were no differences among
radiation treatments. Finally, heat dissipation mechanisms
via NPQ were much more evident at 20uC than at 25uC
(Fig. 5C). Moreover, at 20uC NPQ increased as the
experiment progressed, with samples under the PAB
treatment having the highest NPQ, followed by those
under the PA treatment (Fig. 5C).
During the daily cycle, Rubisco activity and gene
expression were significantly affected by temperature
(Fig. 6).Ingeneral,Rubiscoactivitywasrelativelyconstant
during the daily cycle, but cultures incubated at 25uC
showed significantly higher activity compared to those
incubated at 20uC (Fig. 6A). However, there were no
significant differences among radiation treatments. Ru-
bisco gene expression was significantly higher in samples
incubated at 25uC compared to those incubated at 20uC
(Fig. 6B). At 20uC, Rubisco gene expression was rather
similar throughout the exposure; however, at 25uC, gene
expression showed an optimum around noon.
Discussion
Photosynthesisisacomplexprocessthatdependsonthe
action of a large number of enzymes, as well as on the
availability of substrates (i.e., CO and H O). Therefore, it
2 2
is highly probable that a particular stressor will affect one
Fig. 3. Assimilation numbers (in mg C [mg Chl a]21 h21) of
or more targets simultaneously. The picture is further
Thalassiosira weissflogii at the end of the incubation period for
samples incubated at 20uC and25uC.(A) Daily cycle (DOY328); complicated when considering a multi-factor experimental
(B)short-termincubation(DOY329).Thebarsrepresentthemean approach, withthecombined effecteitherbeingsynergistic
oftriplicatesamplesunderthedifferentradiationtreatments,while or antagonistic (Dunne 2010). Climate change resulting
theverticallinesaretheSD.Blackbarsindicatesamplesexposedat from anthropogenic activities (IPCC 2007) simultaneously
20uC,whereasopenbarsindicatesamplesexposedat25uC. modifies several variables that in turn affect photosynthe-
sis. One such stressor of photosynthesis is UVR, which is
significantly lower values in samples incubated at 25uC known to reduce photosynthetic rates of phytoplankton
compared to those incubated at 20uC. The mean UVR organisms (see reviews by Villafan˜e et al. [2003] and
inhibitions during the daily cycle were 11.9% 6 6.9% and Harrison and Smith [2009]). Another interfering factor is
41.7% 6 6.6% for incubations at 25uC and 20uC, the temperature increase, and the evaluation of its effects
respectively. During the short-term incubations (Fig. 4F), deservesspecialinterestnotonlyintheaquaticenvironment
themeanUVRinhibitiondecreasedfromavalueof72.4% butwithregardtotheEarthasawhole(IPCC2007).Recent
6 8.9% at 20uC to a value of 39.6% 6 5.7% at 25uC. reports (Mckenzie et al. 2011) indicate that solar UV-B
PhotoprotectiveactivityofT.weissflogiiduringthedaily radiation reaching the Earth’s surface might not increase
cycle was assessed through the evaluation of the de- further inthe futureand thatthe ozonelayer might start to
epoxidation state (DEPS: Dt:(Dt + DD), as well as from recover. Nevertheless, aquatic organisms such as phyto-
heat dissipation mechanisms (i.e., NPQ) (Fig. 5C). The planktonwouldexperienceanincreaseinsolarradiationasa
initialDEPSwaslowearlyinthemorning,butitincreased feedbackmechanismduetoclimatechange:Increasedwater
significantly when samples were exposed to solar radiation temperature (due to climate change) would produce a
(Fig. 5A). By noon, samples incubated at 20uC had a shallower thermocline and thus reduce the depth of the
higher DEPS than those incubated at 25uC, and these upper mixed layer, thus exposing phytoplankton to higher
differences among the same radiation treatment but at radiationconditionsaswellastohighertemperatures.
Temperature, UVR affect T. weissflogii 1337
Fig. 4. Effectivephotochemical quantumyieldof Thalassiosiraweissflogii samplesexposed
tosolarradiationunderthethreeradiationtreatments(PAB,PA,andP)during(A)dailycycle
(DOY328)at20uC;(B)representativeshort-termexposureat20uC(DOY329);(C)dailycycleat
25uC;(D)representativeshort-termexposuresat25uC.UVR-inducedinhibitionofYat20uCand
25uCduring(E)dailycyclesand(F)short-termexposure.Circles,triangles,andsquaresrepresent
the mean of duplicate samples, while the vertical lines are the half-mean range. Solid symbols
indicatesamplesexposedat20uC,whereasopensymbolsindicatesamplesexposedat25uC.Half-
filled symbolsafterthe arrows indicate recovery indim light in thelaboratory.
1338 Helbling et al.
While most investigations dealing with the effects of
UVR on phytoplankton photosynthesis have evaluated
carbon fixation occurring during the ‘dark’ reactions
(Calvin–Benson cycle: i.e., through radiocarbon incorpo-
ration measurements), some others have focused on more
specifictargets,suchasRubisco(Lesser1996a;Bouchardet
al. 2008), xanthophyll pigments (Van De Poll and Buma
2009),carbonicanhydraseactivity(WuandGao2009),and
the electron transport rate of PSII (Bergmann et al. 2002;
Villafan˜e et al. 2007; Bouchard et al. 2008). On the other
hand, increased temperatures have been known to affect
phenology and biodiversity (Wrona et al. 2006) as well as
plant–herbivore dynamics and food-web length (Beisner et
al. 1997), among many other effects. Still, and at the
individual level, increased temperatures seem to directly
and indirectly favor different physiological processes in
aquatic organisms (Sobrino and Neale 2007; Halac et al.
2010; Herna´ndez Moresino and Helbling 2010).
Previous studies addressing the interactive effect of
temperature and UVR on T. pseudonana concluded that
temperatureaffectedthesensitivitytoUVRmainlybecauseof
changes in the rates of repair for similar rates of damage
(Sobrino and Neale 2007). The study of Halac et al. (2010)
obtained a similar conclusion for Chaetoceros gracilis, but
even though an increase in temperature resulted in a better
photosynthetic response of T. weissflogii, the underlying
physiological mechanisms were not clear, and the authors
speculated that the differences in responses between the two
species were probably related to their size or to their
distributionwithintheoceanicrealm(i.e.,coastalvs.oceanic).
Inthisworkwethoroughlyevaluatedthephotosynthetic
responses of the cosmopolitan diatom T. weissflogii when
exposed to different radiation and temperature treatments,
targeting both the photosystem and the Calvin–Benson
cyclelevels.Ourdatashowadegreeofresponsesinrelation
to the exposure to different radiation treatments (i.e.,
ranging from the characteristic inhibitory response [e.g.,
Figs. 2–4] to no significant effects [e.g., Fig. 5]). These
differences with regard to UVR exposure were observed
not only when comparing different time scales of experi-
mentation (i.e., short-term exposures vs. daily cycles) but
also different stages of the photosynthesis process, which
vary within a range of seconds (or fractions) to hours
(Falkowski 1984; Laney 2006). Additionally, differences in
responses were affected by incubation temperature, with
higher values generally benefiting the species, by reducing
UVR-inducedphotoinhibitiononPSII(Fig. 4)aswellasin
the Calvin–Benson cycle (Figs. 2, 3). In addition, and
althoughthexanthophyllcyclepigments(percell)werenot
differentbetweenthetwotemperatures(Table 1),therewas
a higher protective capacity at the higher temperature due
Fig. 5. Xanthophyll pigments in Thalassiosira weissflogii to a higher xanthophyll:photosynthetic pigment ratio
under different radiation treatments (circles: PAB; squares: P) (Fig. 5). Thisdifference was mainly due to a lower content
and temperatures (solid symbols: 20uC; open symbols: 25uC) of Chl a per cell at higher temperatures (Table 1). The
throughout the daily cycle (DOY 328). (A) DEPS, calculated as growth rate (see above) was significantly higher at 25uC
Dt:(Dt + DD); (B) ratio of xanthophyll to photosynthetic than at 20uC, indicating that as a result of the faster
(xan:phot)pigments; (C)NPQ. The symbolsrepresentthe mean growth,cellsgrowingat25uCwereslightlysmallerandthus
of duplicate samples, while the vertical lines are the half-
hadasmalleramountofChlapercell.Finally,therewasa
mean range.
significantincreaseinRubiscoactivityandgeneexpression
atthe highertemperature (Fig. 6). All of these dataclearly
Temperature, UVR affect T. weissflogii 1339
Table 1. Chlorophyll and xanthophyll concentration per cell (pg cell21) in the three
radiation treatments (PAB, PA, and P) at 20uC and 25uC at various times during the daily
cycle experiment.
Concentration 20uC 25uC
(pg cell21)
and time(h) PAB PA P PAB PA P
Chlorophyll
0 5.34 5.34 5.34 4.33 4.33 4.33
1.5 4.4160.38 5.2060.13 4.6460.25 3.7060.36 3.9360.13 3.8060.13
6 4.7960.29 4.40 4.7260.38 3.2360.09 3.2660.17 3.1960.31
Xanthophyll
0 0.69 0.69 0.69 0.74 0.74 0.75
1.5 0.7160.04 0.8160.03 0.77 0.8360.06 0.8960.05 0.8560.02
6 1.0260.12 0.94 0.9760.11 0.9160.03 0.9360.09 0.9560.08
point out that the physiological ‘strategy’ of T. weissflogii
involves multiple adjustments that are related to different
targets, as discussed below.
The samples exposed to solar radiation at 20uC had a
significant effect related to UVR reduction of carbon
fixation, which was noticeable in both the daily cycle and
the short-term incubations (Figs. 2, 3). There was a
characteristicresponsewhencellswereexposedtodifferent
wave bands, with UV-A accounting for the bulk of
inhibition, as seen in many environments of the world
(Villafan˜e et al. 2003). During our study, however, larger
differences among radiation treatments of samples incu-
bated at 20uC were observed during the short-term
incubation, which could be related mainly to the higher
radiation levels on DOY 329 (Fig. 1). Similarly, higher
inhibition due to UVR at both temperatures was observed
forYduringshort-termincubations(Fig. 4F),ascompared
to the daily cycle (Fig. 4E). Still, there was a clear trend of
recovery of Y, which was complete in the daily cycle, as
compared to the short-term incubation. However, the
magnitudeofUVR-induced effects(as well asthose dueto
PAR)oncarbonfixationandPSIIfluorescencewasclearly
reducedby exposure atthe higher temperature (Figs. 3,4),
butthisbeneficialeffectoftemperaturedoesnotseemtobe
universal, as there is a significant species-specificity
component. For example, high temperatures might lead
toenhancedsuperoxideradicalproduction,asobservedfor
symbiotic dinoflagellates (Lesser 1996b). Furthermore, the
study of Halac et al. (2010) found that although high
temperature (23uC, i.e., either during acclimation or
exposure) generally benefited photosynthetic performance
of T. weissflogii and C. gracilis, the latter seemed to adjust
bettertothechangefrom18uCto23uC.Thiswasprobably
relatedtoamoreefficientmechanismofenergydissipation,
such as NPQ, which was less evident in T. weissflogii. In
contrasttothisfinding,inourstudy theenergy dissipation
(evaluated via DEPS, calculated as Dt/[Dt + DD]) was
Fig. 6. (A) Rubisco activity (mAbs340 mg protein21 s21)
higher at the lower temperature (Fig. 5A). On the other
and (B) relative Rubisco gene expression (arbitrary units [a.u.])
hand, the xanthophyll pigment pool relative to photosyn-
duringthedailycycle(DOY328)atdifferentradiationtreatments
(circles:PAB;squares:P)andtemperatures(solidsymbols:20uC; theticpigmentswashigherat25uCthroughouttheoutdoor
opensymbols:25uC).Thesymbolsrepresentthemeanofduplicate exposure period (Fig. 5B), and, therefore, less conversion
samples, whilethe vertical lines arethe half-mean range. to Dt might be required to give the same level of heat