CHARACTERIZATION AND EXPRESSION ANALYSIS OF THE Lhcf GENE FAMILY IN EMILIANIA HUXLEYI (HAPTOPHYTA) REVEALS DIFFERENTIAL RESPONSES TO LIGHT AND CO21

CHARACTERIZATION AND EXPRESSION ANALYSIS OF THE Lhcf GENE FAMILY INEMILIANIA HUXLEYI (HAPTOPHYTA) REVEALS DIFFERENTIAL RESPONSES TO

LIGHT AND CO21

Stephane C. Lefebvre2, Gayle Harris, Richard Webster, Nikos Leonardos, Richard J. Geider, Christine A. Raines3

Department of Biological Sciences, University of Essex, Colchester CO4 3SQ, Essex, UK

Betsy A. Read

Department of Biological Sciences, California State University San Marcos, San Marcos, California 92096, USA

and Jose L. Garrido

Instituto de Investigacions Marinas, C.S.I.C., Av. Eduardo Cabello 6, 36208 Vigo, Spain

Emiliania huxleyi (Lohmann) W. W. Hay et H.Mohler is the most abundant marine unicellular coc-colithophore in the ocean and belongs to the groupof organisms that have chl c and fucoxanthins aspigments in the photosynthetic light-harvesting com-plexes (LHCs). In this study, we report on the isola-tion and characterization of the mRNAs encodingsix light-harvesting complex fucoxanthin-bindingproteins (LHCFs) from E. huxleyi. Phylogenetic anal-ysis of these sequences has revealed that they formthree distinct subgroups: haptophyte, diatom ⁄ hapto-phyte, and LI818-like. Expression analysis of the sixLhcf genes showed a clear down-regulation at thetranscriptional level when the cultures were grownin high light (300 lmol Æ m)2 Æ s)1) when compared toequivalent samples in low light (30 lmol Æ m)2 Æ s)1).In contrast, little impact on transcript levels wasobserved between cultures grown in either low CO2

(180 ppm) or high CO2 (750 ppm) at either lightintensities. Using polyclonal antibodies to three ofthe LHCFs revealed a down-regulation in proteinlevels in response to increased light availability witha minor increase in two of the LHCFs in elevatedCO2. This study has provided an insight into thediversity of LHCFs and how changes in the levels ofthese proteins, together with altered pigment com-position, may contribute to the flexible response ofE. huxleyi to changes in the light environment.

Key index words: diatoms; gene expression; LHCFphylogeny; light-harvesting complex fucoxanthin-binding proteins (LHCF); light-harvesting com-plex proteins (LHCP); protein expression

Abbreviations: LHCF, light-harvesting complexfucoxanthin-binding proteins; LHCP, light-har-vesting complex proteins

E. huxleyi is the most abundant bloom-formingcoccolithophore in the modern ocean and contrib-utes to the export of both particulate organic car-bon and calcium carbonate from surface waters tothe deep sea (Buitenhuis et al. 2001). E. huxleyiforms regular (annual) large-scale blooms in sub-polar latitudes during summer and early autumn(Holligan et al. 1983, Brown and Yoder 1994,Iglesias-Rodriguez et al. 2002). Substantial labora-tory research has identified how single environmen-tal factors, including light (Nielsen 1997, Suggettet al. 2007), temperature (Stoll et al. 2002), andnutrient availability (Paasche 1998) as well as theinteraction of multiple environmental factors(Zondervan et al. 2002, Sciendra et al. 2003,Leonardos and Geider 2005) affect the growth andcalcification of E. huxleyi. As compared with otherglobally important phytoplankton, E. huxleyi is rela-tively well studied (Zondervan 2007). Nonetheless,we know very little about the molecular mechanismsthat enable E. huxleyi to modify its photosyntheticapparatus to facilitate acclimation to changes in itsenvironment.

In eukaryotic organisms, the photosynthetic pig-ments associate with proteins in the thylakoid, form-ing LHCs. In higher plants and the Chlorophyta(green algae), the light-harvesting pigments are chla ⁄ b and carotenoids, but in the Chromophyta algalgroup, chl a ⁄ c, carotenoids, and fucoxanthin arepresent. E. huxleyi, a member of the Haptophyta(part of the larger chromophyte lineage), has LHCscomposed of chl a ⁄ c, carotenoids, and fucoxanthin.Despite the difference in pigment composition,LHCs and LHCFs are phylogenetically and structur-ally related, and evidence for a common ancestorhas been presented (Green and Kuhlbrandt 1995,Durnford et al. 1999, Koziol et al. 2007). Knowledgeof LHCFs and the Lhcf genes has expanded signifi-cantly in recent years, and there is now sequenceinformation on several diatoms, including Cyclotellacryptica, Odontella sinensis, Skeletonema costatum, Tha-lassiosira pseudonana, Phaeodactylum tricornutum, andNitzschia closterium. A detailed study of the Lhcf gene

1Received 22 December 2008. Accepted 19 August 2009.2Present address: Department of Biology, Romberg Tiburon Cen-

ter, San Francisco State University, Tiburon, California 94920, USA.3Author for correspondence: e-mail [email protected].

J. Phycol. 46, 123–134 (2010)� 2009 Phycological Society of AmericaDOI: 10.1111/j.1529-8817.2009.00793.x

123

family of C. cryptica has revealed the presence ofbetween 12 and 20 members, which can be dividedinto two groups based on amino acid sequence simi-larity (Eppard and Rhiel 1998, 2000). Expressionanalysis of the C. cryptica Lhcf genes in response tolight has suggested that during reillumination, Lhcfgene expression increased for one group, while in asecond group, expression decreased, demonstratingdifferential expression of the Lhcf genes in thisorganism (Oeltjen et al. 2002). New information isnow available on the oligomeric structure of theP. tricornutum and Cyclotella meneghiniana LHCFcomplexes, revealing differences with higher plantLHCs and providing evidence for a diadinoxanthin-binding subcomplex (Guglielmi et al. 2005, Beeret al. 2006, Gundermann and Buchel 2008).However, despite the ecological importance of thehaptophyte E. huxleyi, there has been no analysis ofthe Lhcf gene family from this organism either atthe gene or protein level. The aim of the work pre-sented here was to isolate and characterize the Lhcfgene family from E. huxleyi, to explore the relation-ship of these Lhcf genes with Lhc genes of otherphotosynthetic organisms, and to determine theimpact of the environmental stimuli light and CO2

on transcript and protein levels.

MATERIALS AND METHODS

E. huxleyi culture conditions. Unialgal turbidostat-cyclostatcultures of E. huxleyi (calcifying strain PML–B11) were grown at15�C, using 0.2 lm filtered artificial seawater (ESAW, Bergeset al. 2001, 2004) with metals and vitamins added to achieve f ⁄ 2trace-element concentrations (Guillard and Ryther 1962),200 lL nitrate, and 40 lM phosphate. The continuous cultureswere operated as manual turbidostats. A cell density of 1,000cells Æ lL)1 (±10%) (i.e., about half of the maximum yield underhigh-light conditions for the inflow nutrient levels) was the targetdensity, and the inflow medium flow rate was adjusted once dailyto maintain a steady cell abundance. Cultures were illuminatedon a light:dark (L:D) cycle of 14:10 at photon flux densities(PFDs) of 300 lmol photons Æ m)2 Æ s)1 (High Light, HL) and30 lmol photons Æ m)2 Æ s)1 (Low Light, LL). Culture vesselswere cylindrical 3 L borosilicate glass, illuminated from the side.PFD was measured in the center of a similar vessel containing 3 Lof distilled water, using a 4p Biospherical Instruments probe,model QSL 100 (San Diego, CA, USA). Illumination was via cool-white fluorescent lamps (Lumilux CoolWhite, EdmundsonElectrical, Colchester, Essex). The cultures were incubated ateither 180 ppmv (low) or 750 ppmv (high) CO2. The pH of theculture was continuously monitored, to an accuracy of ±0.004 pHunits, using Thermo Orion Ross pH electrodes (model 8103 BN;Thermo Corp., Waltham, MA, USA) and a custom-built high-precision amplifier (John Bartington Associates, Colchester,UK) and logged in a computer system. The pH electrodes werecalibrated with seawater buffers (Dickson and Goyet 1994). ThepH was controlled within ± 0.01 pH units by aeration with CO2-enriched or CO2-free air as described in Leonardos (2008). Thetarget pH values corresponding to low (180 ppmv) and high(750 ppmv) CO2 conditions were obtained using a CO2 systemcalculator (v. 01.05, Lewis and Wallace 1998) from the measuredpH and seawater alkalinity. Alkalinity was measured dailyfollowing Bradshaw et al. (1981). The alkalinity of the culturemedium was adjusted to compensate for the depletion ofalkalinity due to calcification by E. huxleyi. Cell density was

monitored daily using improved Neubauer hemocytometers(Paul Marienfeld, Lauda-Koenigshofen, Germany).

The cultures were maintained in each condition for at least10 d prior to starting each experiment. One E. huxleyi chemo-stat culture for each condition was used. Cultures were sampledsix times over a 2-week period, with steady-state reestablishedbetween each sampling date. All samples were collected 6 hafter the light onset when the cultures were at steady state (celldensity variation was 10% or less for three consecutive days).Samples were collected for pigment characterization andquantification, gene expression, and Western blotting.

DNA isolation. Samples of culture (200 mL) in logarithmicphase were harvested by centrifugation in a 6 · 300 rotor (MSEHI-21, DJB Labcare, Buckinghamshire, England) at 3,000g for10 min at room temperature. Cells were resuspended gently infresh lysis buffer [500 lL, 0.5% (w ⁄ v) SDS (sodium dodecylsulfate), 20 lg Æ mL)1 proteinase K], preheated to 55�C, andincubated at 55�C for 30 min, with gentle mixing every 10 min.Following this, 100 lL of solution containing 5 M NaCl and80 lL CTAB (cetyl trimethylammonium bromide) were added,and samples were incubated at 65�C for 10 min. Chloro-form:isoamyl alcohol (24:1 v ⁄ v) was then added, and thesample vortexed prior to centrifugation at 13,000g for 5 min.DNA was precipitated using isopropanol and collected bycentrifugation for 10 min at 12,000g, washed with 70% ethanol,and dried in a Bachhofer vacuum concentrator (Bachofer,Reutlingen, Germany).

Genomic DNA analysis (Fig. S1 in the supplementary material).The restriction enzymes were purchased from Roche MolecularBiochemicals (Indianapolis, IN, USA) or New England Biolabs(Hitchin, UK). Genomic DNA (10 lg) was digested at 37�Covernight using 15–20 units of restriction enzyme. The DNAwas separated on a 1% agarose TAE (0.04 M Tris, 0.12 M aceticacid, 0.001 M EDTA, pH 7.8) gel. After electrophoresis, theDNA was denatured and neutralized by shaking the gels for30 min at room temperature in denaturing solution (1.5 MNaCl, 0.5 M NaOH) and neutralization solution (1 M Tris-HClpH 7.4, 3 M NaCl), respectively. The DNA was blotted ontonylon membrane (Hybond—N) (Boehringer Mannheim,Lewes East Sussex, UK) overnight at room temperature using10 X SSC-buffer (1.5 M NaCl, 0.15 M sodium citrate, pH 7.0).After transfer, DNA was cross-linked to the membrane using aUV stratalinkerTM 2400 (Stratagene, La Jolla, CA, USA). DNAwas labeled with [a-32P]-dATP using the Prime-a-Gene� label-ing system (Promega, Southampton, UK) according to themanufacturer’s instructions. Hybridization was carried out at62�C overnight. Washes were performed twice at 65�C, first in 2X SSC containing 0.1% SDS, and then in 0.5 X SSC containing0.1% SDS. The hybridization conditions were the same for thedifferent gene probes. On completion of the hybridizationsusing the Lhcf probes, the blots were placed between two piecesof plastic inside a Harmer X-ray cassette (Genetic ResearchInstrumentation [GRI], Felsted, Essex, UK). A Cronex plusintensifying screen (GRI) was placed on top of the Fuji X-rayfilm (Kompass, London, UK), and the cassette was left at )70�Cand exposed for either 3 or 16 h. The film was developed usingan Agfa Curix 60 automated developer (A Somerville Ltd.,Holmfirth, UK).

Sequencing and primers. DNA sequencing reactions werecarried out by the Genome laboratory, (Genome laboratoryJohn Innes Centre, Norwich, UK).The sequences were ana-lyzed using the Chromas software (http://www.technelysium.com.au/chromas.html), and the genes identified using eitherBLAST searches at National Center for Biotechnology Informa-tion (NCBI, Betheseda, MD, USA) (http://wwwncbi.nlm.nih.gov/BLAST/) or MUSCLE 3.7 (Edgar 2004) alignment(http://www.phylogeny.fr/version2_cgi/one_task.cgi?task_type=muscle). Primers were designed to have G + C content of 55%–65%, a melting temperature of 56�C–61�C, and no secondary

124 STEPHANE C. LEFEBVRE ET AL.

structure using the DNA calculator at (http://www.sigma-genosys.com/calc/DNACalc.asp) and synthesized by Invitrogen(Carlsbad, CA, USA).

RNA extraction, quantification, and integrity. Total RNA wasextracted by grinding the filters containing�4.5 · 108 E. huxleyicells (harvested from �50 mL of culture) in liquid nitrogenwith 1 mL Trizol reagent (Sigma Aldrich, St. Louis, MO, USA)in a prechilled mortar with liquid nitrogen. The following stepswere carried out according to the manufacturer’s instructions.RNA samples were cleaned up using RNAeasy mini spincolumn (Qiagen, Crawley, UK), total RNA was quantified usinga NanoDrop N-1000 spectrophotometer (NanoDrop Techno-logies Inc., Wilmington, DE, USA), and the integrity of RNAsamples was assessed using an Agilent 2100 bioanalyzer(Agilent Technologies Inc., Santa Clara, CA, USA).

Real-time RT-PCR. Real-time RT-PCR primers were designedusing the Primer3 software version 0.4.0 (http://frodo.wi.mit.edu/). Primers were designed to have G + C content of55%–65%, a melting temperature of 56�C–61�C, and to amplifya target sequence between 75 and 100 bp. Forward and reverseprimers were synthesized by Integrated DNA Technologies Inc.(Coralville, IA, USA). A total of at least four independent RNAsamples for each condition were pooled together and used forcDNA synthesis. cDNA synthesis was performed with Omni-script reverse transcriptase kit (Qiagen) using 1.5 lg templateRNA, 1X RT buffer, 0.5 mM deoxynucleoside triphosphate,1 lM oligo(dT) primer, 10 U of RNase inhibitor, and 4 U ofOmniscript reverse transcriptase. The total reaction volume wasadjusted to 20 lL, denatured at 65�C for 5 min, and incubatedat 37�C for 1 h, and the reverse transcriptase was inactivated at72�C for 15 min. Real-time PCR was performed with iCycler IQversion 3.0 (Bio-Rad, Richmond, CA, USA) using SYBR greenchemistry. The thermal profile was 95�C for 10 min followed by40 cycles of 95�C for 10 s, 60�C for 30 s, and 82�C for 30 s.Reactions were carried out in a 96-well plate in a 25 lL reactionvolume containing 7.1 lL SYBR green Supermix (Bio-Rad),12.3 lL dH2O, 0.3 lL each of forward and reverse primer(20 lM), and 5 lL of template cDNA (250 ng). The plateswere loaded using a Biomek� 2000 automation workstationrobot (Beckman Coulter Inc., Fullerton, CA, USA). Eachsample was run twice in triplicate using template cDNAsynthesized from RNA extracted from E. huxleyi cells grownunder each condition. A negative no-template control wasincluded on every plate for each set of primers. This wasrepeated with another independent cDNA synthesis. Actin type1 (ACT1) was chosen as reference gene, and specific primers,which did not match with the five other types of actin presentin E. huxleyi, were designed as describe above.

Expression levels were determined as the number of ampli-fication cycles needed to reach a fixed threshold in theexponential phase of the PCR reaction (Ct). All amplificationplots were analyzed with a selected threshold of 30.0 to obtainCt values.

Pigment sample collection, extraction, and analysis. Samples forpigment analysis were collected by filtering 50 mL of cultureon Whatman GF ⁄ F glass fiber filters (Whatman, Maidstone,UK) under low vacuum. The filters were placed in a plasticcentrifuge tube (1.5 mL), flash frozen in liquid nitrogen, andstored at )80�C.

Pigments were analyzed by HPLC as described by Garridoand Zapata (1998) and Garrido et al. (2000) using a WatersAlliance HPLC System with a 2,695 separations module, aWaters 996 photodiode array detector (350–750 nm; 1.2 nmoptical resolution) interfaced to a Waters 474 scanningfluorometer (Selectscience Ltd., Bath, UK). Frozen filters wereextracted on ice under low light in methanol and then filteredthrough 25 mm diameter polypropylene syringe filters (MFSHP020, 0.2 lm pore size; Advantec MFS Inc., Dublin, CA, USA)to remove cell and filter debris. An aliquot (0.5 mL) of the

methanol extract was mixed with 0.2 mL of water, and 200 lLwas injected immediately onto a C8 column (Waters Sun-Fire,150 · 4.6 mm, 3.5 lm particle size; SelectScience Ltd., Bath,UK) thermostated at 25�C. The mobile phases were: A = meth-anol:acetonitrile:aqueous pyridine solution (0.025 M pyridine,pH adjusted to 5.0 with acetic acid) in the proportions 50:25:25(v ⁄ v ⁄ v), and B = acetonitrile:acetone (80:20 v ⁄ v). A segmentedlinear gradient was (time, min, % B): 0 min, 0%; 18 min, 40%;22 min, 100%; 38 min, 100%. Initial conditions were reestab-lished by reversed linear gradient (4 min). Flow rate was1 mL Æ min)1. Pigments were identified by cochromatographywith authentic standards obtained from supply-chain opera-tions reference (SCOR) cultures and diode-array spectroscopyand quantified using external standards.

Protein extraction, quantification, and Western blot analysis.Samples for protein isolation (450 mL) were collected bygentle filtration on 45 mm diameter Whatman polycarbonatefilters. Total proteins were extracted by grinding the filterscontaining 450 mL of E. huxleyi culture (� 4.5 · 108 cells) inliquid nitrogen with 1 mL of extraction buffer [50 mM HEPES,[pH 8.2]; 5 mM MgCl2; 1 mM EDTA; 1 mM EGTA; 10%glycerol; 0.1% Triton X-100; 2 mM benzamidine; 2 mMaminocaproic acid; 0.5 mM phenylmethylsulphonyl fluoride(PMSF); 10 mM dithiothreitol (DTT)] in a prechilled mortarat liquid nitrogen temperature. The resulting frozen powderwas transferred to a centrifuge tube, allowed to thaw com-pletely, and then centrifuged for 1 min at 14,000g at 4�C. Thesupernatant (1) was aliquoted and stored at )80�C until furtheranalysis. Protein quantification was determined according toBradford (1976). For Western blot analysis, 150 lL DMSH5a(313 mM Tris-HCl [pH 6.8], 10% SDS, 25% glycerol and 25%2-mercaptoethanol, 0.004% bromophenol blue) was added to500 lL of supernatant (1) and boiled for 5 min. Samples wereloaded on an equal protein basis, separated using 12% (w ⁄ v)SDS-PAGE, transferred to PVDF membrane, and probed usingantibodies raised against RUBISCO and LHCFs. Proteins weredetected using horseradish peroxidase conjugated to thesecondary antibody and ECL chemiluminescence detectionreagent (Amersham PLC, Little Chalfont, UK). RUBISCOpolyclonal antibodies were raised in rabbit against a purifiedRUBISCO from Porphyridium cruentum (HTI antibody, Ramona,CA, USA). Three LHCF antibodies were produced; the purifiedpolyclonal antibodies KG13, SE11, and DE10 (GenScriptCorporation, Piscataway, NJ, USA) were raised in rabbitinjected with peptides KAPSPLRPDYYPG, SNPDNNEDDGE,and DYDQWSQGQE corresponding to position 151 through163 of E. huxleyi LHCF17H08, position 146 through 156 ofE. huxleyi LHCF16B09, and position 57 through 66 of E. huxleyiLHCF31G05, respectively.

Phylogenetic and sequence analysis. Amino acid and nucleotideBLAST searches were performed using the NCBI Blast software(http://www.ncbi.nlm.nih.gov/blast/Blast.chi), and open read-ing frames (ORF) were determined using the ORF findersoftware (http://www.ncbi.nlm.nih.gov/projects/gorf/).

Sequences alignments and phylogeny programs were runfrom the ‘‘Phylogeny.fr: Robust Phylogenetic Analysis for theNon-Specialist’’ platform accessible at http://phylogeny.lirmm.fr/phylo_cgi/index.cgi (Dereeper et al. 2008). Sequencealignments were performed using the MUSCLE 3.7 (Edgar2004) run in full mode with the number of iterations equal to16. The ProtTest program (Abascal et al. 2005) was usedto select the best amino acid substitution model with theAkaike information criterion (AIC) framework (Akaike 1974).Maximum-likelihood analyses were performed using thePhylML3.0 program (Guindon and Gascuel 2003, Guindonet al. 2005) with the LG substitution model (Le and Gascuel2008); the proportion of invariable sites was 0.033, andthe gamma shape distribution (a) equal to 1.333. The numberof substitution rate categories was four, and the number of

Lhcf GENE FAMILY IN EMILIANIA HUXLEYI 125

bootstraps was 100. The branch lengths of the tree wereoptimized, and the tree topology improved with the NearestNeighbor Interchange (NNI). Bayesian posterior probabilitieswere obtained using the MrBayes v3.1.2 (Huelsenbeck andRonquist 2001) accessible at http://mrbayes.csit.fsu.edu/.Blossum62 with four gamma distribution categories (a) wasused as the substitution matrix. Five million generations wereperformed with a sampling frequency of 100 and a 25% burn-in, and the consensus type was all-compatible. In this analysis, aequals 1.81, the proportion of invariant sites was 0.093, thetotal estimated marginal likelihood harmonic mean was12,251.54, and the average SD of the split frequencies was0.009284. The distance-based phylogeny analysis was per-formed using the BioNJ application (Gascuel 1997) with theJones-Taylor-Thornton (JTT) amino acid substitution model,the number of boostraps was 1,000, and the gamma distribu-tion parameter was 1.333. The maximum-parsimony phylogenyanalysis was performed using TNT1.1 (Goloboff et al. 2000)using the new technology search with sectorial search (RSS andCSS), tree fusing, and a standard bootstrap resampling with1,000 replicates. Acetabularia acetabulum PSBS1 was used as anoutgroup to root the trees. All methods yielded comparableresults, and only the graphical PHYML tree is displayed usingthe TreeDyn (Chevenet et al. 2006) and incorporating thesupport values and Bayesian posterior probabilities of the otheranalyses.

Statistical analysis. All statistical analyses were performed bycomparing analysis of variance (ANOVA), using Systats, andthe differences between mean values were tested using the posthoc Tukey’s test (SPSS, Chicago, IL, USA).

RESULTS

Characterization of E. huxleyi Lhcf cDNA clones.Analysis of 3,000 expressed sequence tags from acDNA library prepared from calcifying cells of E.huxleyi (Wahlund et al. 2004) revealed the presenceof six Lhcf cDNA sequences (Lhcf 16H08, Lhcf 16B09,Lhcf 28G02, Lhcf 31G05, Lhcf 17H08, and Lhcf 19E10).Sequencing of these six cDNA clones showed thatthey were full length, and the deduced E. huxleyiLHCF protein sequences ranged from 206 to 281amino acids with predicted pIs of between 4.81 and6.52, and molecular masses comprised between21.9 kDa and 29.5 kDa (Table 1). By comparisonwith LHCFs from other organisms, a bipartite prese-quence region of �40 amino acid residues was iden-tified at the N-terminal region of the six LHCFs; thefirst 14 amino acids constitute the signal sequencefor passage across the endomembrane system,which is followed by a hydrophobic region of the

transit peptide to target to the chloroplast (Fig. 1).No consensus sequence indicating the cleavage sitefor the signal peptide found in diatoms and the redalgae was evident in the E huxleyi LHCFs, with theexception of Lhcf16B09, which has the sequenceAQAFSAPA containing the highly conserved phenyl-alanine (Gruber et al. 2007).

Amino acid sequence comparisons of the matureproteins revealed that the E. huxleyi LHCFs are quitedivergent with similarity values ranging from 10% to84% across members of this gene family (Table 2).Hydrophobicity plots of E. huxleyi’s LHCF proteinsequences predicted three membrane-spanningregions, MSR 1, MSR 2, and MSR 3 (Fig. 1A). Ofthe 16 residues strictly conserved across all themature LHCF sequences, 14 are located in the puta-tive membrane-spanning regions. Alignment ofE. huxleyi’s LHCF protein sequences indicates thepresence of seven well-conserved amino acids acrossthe heterokont algae (Caron et al. 1996) known tobind chl a ⁄ c molecules (Fig. 1B). Four of theseseven amino acids are involved in ion pairingbetween MSR 1 and MSR 3 and are bound to chl amolecules (Fig. 1B). The binding motifs E(X2)HGRin MSR 1 and E(X2)(H ⁄ N)GR in MSR 3 are con-served in all E. huxleyi’s LHCF sequences analyzed,while the motif Q(X8)E in MSR 2 is only conservedin LHCF16B09, LHCF16H08, LHCF31G05, andLHCF28G02. In the two sequences LHCF17H08 andLHCF19E10 (part of the LI818 and LI818-likegroup, see below), the MSR 2 region is quite diver-gent from the four other LHCs, and although theglutamate residue (E) is conserved, the glutamineresidue (Q) was replaced by valine in LHCF17H08and leucine in LHCF19E10. However, theLHCF17H08 and LHCF19E10 MSR2s do contain anumber of residues that are highly conserved inother LI818-type proteins, and these are shown in(Fig. 1B).

Genomic analysis of Lhcf gene numbers. The numberof Lhcf genes likely to be present in the E. huxleyigenome was investigated using the six Lhcf cDNAclones as probes on Southern blots of genomicDNA digested with restriction endonucleases(Fig. S1). The number and pattern of DNA frag-ments varied dependent on the Lhcf cDNA beingused as a probe, and taking these results togethersuggested the presence of at least 10 Lhcf genes inthe E. huxleyi genome. However, it is highly likelythat this number is an underestimate as it does nottake into account the presence of tandem repeatsor the presence of more divergent sequences.Indeed, preliminary BLAST searches of E. huxleyi’sgenome (http://www.jgi.doe.gov/Ehux/) wouldsuggest that there could be as many as 50 distinctLhcf genes (data not shown).

Comparison of E. huxleyi Lhcf genes with diatom,mosses, and red and green algae Lhcs. To explore theevolutionary relationships of the Lhcf genes fromE. huxleyi, we constructed a phylogenetic tree based

Table 1. Pair-wise comparison of Emiliania huxleyi LHCFdeduced amino acid sequences.

LHCF 16B09 16H08 31G05 28G02 17H08 19E10

16B09 – 84 27 14 11 1016H08 – – 23 22 10 2031G05 – – – 51 16 2028G02 – – – – 16 1717H08 – – – – – 5419E10 – – – – – –

LHCF, light-harvesting complex fucoxanthin-binding proteins.


Fig. 1. Hydropathy plot andalignment of Emiliania huxleyideduced amino acid LHCF seque-nces. (A) Representative hydropa-thy plot given for LHCF16B09.Hydrophobicity values on they-axis range from )4 (most hydro-philic) to 4 (most hydrophobic)and were calculated with a windowsize of 16. The amino acid resi-due numbers are on the x-axis.Transit peptide location (TP)and membrane-spanning regions(MSR 1, MSR 2, and MSR 3) loca-tion are indicated by hatchedrectangles. (B) Deduced aminoacid sequence alignment of sixE. huxleyi LHCFs compared usingMUSCLE 3.7. Dashes indicategaps introduced to provide theoptimal alignment. Asterisksdenote putative ligands to chlmolecules, and dots indicateresidues involved in ion pairing.The putative transit peptidesand membrane-spanning regions(MSR 1, MSR 2, and MSR 3) areindicated by a broken line abovethe sequence. Highly conservedresidues are in bold, and + indi-cates residues conserved in theMSR2 region of the LI818-likeproteins. Conserved residues out-side of the MSRs found in higherplant LHCPs are underlined inbold. LHCF, light-harvesting com-plex fucoxanthin-binding proteins;LHCP, light-harvesting complexproteins.


on amino acid sequence comparisons of LHCFsfrom E. huxleyi, other available chl a ⁄ c sequences[including the red alga (Galderia sulphuraria)and organisms that contained secondary plastids(the haptophyte Isochrysis galbana) and diatoms(C. cryptica, T. pseudonana, and P. tricornutum)], anda number of members of the green algal group(Micromonas sp., Ostreococcus tauri, Bigelowiella natans,A. acetabulum, and Chlamydomonas reinhardtii)(Fig. 2). Sequence alignment scores clearly showthat the E. huxleyi LHCFs occur in three distinctgroups (Fig. 2). One clade comprising LHCF16B09and LHCF16H08 contains only LHCFs from E. hux-leyi and I. galbana. The resolution within this groupis robust (1.00; 1.00; 0.976; 0.64), and we named ithaptophyte-specific LHCF. In the second group,E. huxleyi LHCF28G02 and LHCF31G05 associatedspecifically with LHCF from I. galbana and T. pseudo-nana with good support (0.82; 0.98; 0.916; 0.78)and was called diatom ⁄ haptophyte-specific LHCF.These two groups are part of the fucoxanthin–chl a ⁄ c division and comprise LHCFs of variousdiatoms and haptophytes. LHCF19E10 andLHCF17H08 are located in a third clade, which isstrongly supported (1.00; 1.00; 0.926; 0.92) andbelongs to the LI818 and LI818-like branch, com-prising LHC homologs believed to have appearedearly during the evolution before the green and redalgal lineages diverged (Koziol et al. 2007). The res-olution within this group was generally good, andamong the many LHC homologs from green algae,diatoms, haptophytes, and bryophytes, the hapto-phyte LHCFs formed a distinct cluster (Fig. 2). Withthe exception of the two LI818-like proteins, noneof the E. huxleyi LHCF proteins analyzed here clus-tered with the red algal LHCs (LHCR) or the greenalgal LHCs associated with PSII (LHCII, CP26 andCP29) or PSI (LHCI) (Fig. 2).

E. huxleyi Lhcf gene expression analysis. Theexpression of the Lhcf genes in E. huxleyi was investi-gated using quantitative RT-PCR in order to investi-gate the changes that occur in response to changesin light and CO2. With the exception of Lhcf19E10,which showed a small but significant increase inexpression under high light, the other five Lhcf

genes responded similarly and were significantlylower under high-light conditions (Fig. 3). Theimpact of growth in high CO2 on LHCF transcriptlevels was minimum. In low light, CO2 concentra-tions had no significant effect on the level ofexpression of Lhcf17H08, Lhcf16H08, andLhcf28G02, and only a small decrease in expressionof the Lhcf31G05, Lhcf17H08, and Lhcf19E10 tran-script levels was evident under high-CO2 conditions.In high-light conditions, the Lhcf16B09, Lhcf16H08,and Lhcf17H08 transcript levels were lower in cul-tures grown in high-CO2 levels (Fig. 3).

The response of E. huxleyi LHCF protein levels tochanges in light and CO2 levels. The derived aminoacid sequence obtained from the Lhcf cDNA cloneswas used to produce antibodies to three of theLHCF proteins, each belonging to one of thethree different groups (LHCF17H08 [LI818-like],LHCF16B09 [haptophyte-specific chl a ⁄ c], andLHCF31G05 [chl a ⁄ c]). Western blot analysis of pro-teins extracted from samples used for the quantita-tive RT-PCR studies showed the three LHCFsanalyzed migrating �22–24 kDa on SDS-PAGE gel(data not shown), and they responded differentiallyto the growth conditions. All three LHCF proteinswere reduced significantly in high light as comparedto low light, irrespective of CO2 conditions. In bothhigh- and low-light cultures, a small but noticeableincrease in the levels of the LHCF16B09 andLHCF31G05 proteins was observed in high CO2, butno change in the protein level of LHCF17H08 wasobserved (Fig. 4). Interestingly, light and CO2 hadno effect on the amount of RUBISCO protein (52–54 kDa on SDS-PAGE gel), which exhibited similarlevels across all conditions tested (Fig. 4).

Influence of light and CO2 on E. huxleyi pigmentcompositions. In order to explore further theacclimatory response of E. huxleyi to changes in theenvironment, photosynthetic pigments wereextracted and analyzed from the same samples usedfor quantitative RT-PCR. Detailed pigment analysisis presented in Table S1 (in the supplementarymaterial). Total chl and carotenoid levels did notchange significantly under the different growthenvironments. In contrast, total xanthophyll levels

Table 2. Chl and carotenoid contents of Emiliania huxleyi grown in two light and CO2 conditions and relevant ratios.

Treatment Schl (fg Æ cell)1) Chl a ⁄ c Tcar (M:Chl a) Txan (M:Chl a) Fx ⁄ Txan (Dd + Dt) ⁄ Txan

LL LCO2 231.2 (a) ± 21.33 1.29 (a) ± 0.011 0.06 (a) ± 0.004 2.02 (a) ± 0.040 0.49 (a) ± 0.004 0.07 (a) ± 0.005LL HCO2 275.9 (b) ± 13.71 1.23 (a) ± 0.016 0.06 (a) ± 0.002 2.16 (a) ± 0.036 0.59 (b) ± 0.002 0.06 (a) ± 0.002HL LCO2 188.4 (a) ± 15.98 1.61 (b) ± 0.039 0.09 (a) ± 0.004 1.49 (b) ± 0.039 0.38 (c) ± 0.006 0.39 (b) ± 0.011HL HCO2 190.1 (a) ± 21.13 1.71 (b) ± 0.077 0.09 (a) ± 0.018 1.41 (b) ± 0.049 0.34 (c) ± 0.016 0.42 (b) ± 0.023

Chl and carotenoid pigments were determined by HPLC and expressed as molar ratio to chl a (M:Chl a) or in fg Æ cell)1.Values represent mean ± SE for at least four independent samples. Tchl, total chlorophylls; Chl a ⁄ c, chl a ⁄ c ratio; Tcar, total car-otenoids; Txan, total xanthophylls; Fx ⁄ Txan, ratio fucoxanthin ⁄ total xanthophylls; (Dd + Dd) ⁄ Txan, ratio (diadinoxanthin +diatoxanthin) ⁄ total xanthophylls. Letters in parentheses indicate significant differences (P < 0.05), after analysis of variance(ANOVA), Tukey’s post hoc test, SPSS analysis. LL, low light (30 lmol photons Æ m)2 Æ s)1); HL, high light (300 lmolphotons Æ m)2 Æ s)1); LCO2, low CO2 (180 ppm); HCO2, high CO2 (750 ppm).


Fig. 2. Phylogenetic analysis of the Emiliania huxleyi LHCF amino acid sequences. A maximum-likelihood (PhyML) tree is shown withsupport values and posterior probabilities >50% for the four phylogeny programs indicated at the specific nodes (PhyML 3.0, MrBayes3.1.2, BioNJ, TNT1.1). A total of 152 amino acid positions and 56 sequences were used in this analysis. The tree was rooted using Acetabu-laria acetabulum PSBS1 sequence as the outgroup. The LG substitution model was chosen with the proportion of invariable sites being0.033 and the gamma shape distribution (a) equal to 1.333. The branch length is proportional to the number of substitutions per site.E. huxleyi LHCF sequences can be accessed via the JGI portal under protein ID: 419207, 416733, 356694, 438462, 443878, and 358662. Allother sequences were derived from information contained in Koziol et al. (2007). LHCF, light-harvesting complex fucoxanthin-bindingproteins.


and the ratio of fucoxanthin ⁄ total xanthophyllswere significantly decreased in high light comparedto low light, while CO2 had no significant effect(Table 2). More interestingly, the chl a ⁄ c ratio andthe ratio of diadinoxanthin and diatoxanthin to totalxanthophylls showed significant 1.3-fold and 6-foldincreases in high light compared to low light,

respectively. Again, CO2 had no significant effect onthese two ratios.

DISCUSSION

Phylogenetic analysis of the six E. huxleyi LHCFsequences, characterized in this study, with that of a

Fig. 3. Influence of environ-ment on Emiliania huxleyi Lhcf,RUBISCO, and transketolase steady-state transcripts. Quantitative RT-PCR analysis of mRNA levels forLhcf genes. Cycle thresholds arerepresented for each individualgene as 1 ⁄ cycle threshold (Ct)· 10)2 for the four differentgrowth conditions as described inthe Materials and Methods. Valuesrepresent the mean of four repli-cates for each growth condi-tion ± SE. Letters above barsindicate significant differences(P < 0.01), after analysis of vari-ance (ANOVA), Tukey’s post hoctest. LL, low light (30 lmol pho-tons Æ m)2 Æ s)1); HL, high light(300 lmol photons Æ m)2 Æ s)1);LCO2, low CO2 (180 ppm); HCO2,high CO2 (750 ppm).


diverse range of chl a ⁄ c–containing organisms,including a number of LHCFs from the haptophyteI. galbana, has identified two distinct homologygroups. In the first of these groups, the E. huxleyiLHCF16B09, LHCF6H08, LHCF28G02, andLHCF31G05 proteins clearly belong to the majorchl a ⁄ c clade within which the LHCF16B09 andLHCF16H08 form a haptophyte-specific group. TheE. huxleyi LHCF17H08 and LHCF19E10 proteinsclustered with the LI818 ⁄ LI818-like group shownpreviously to contain homologs from a diverse rangeof algae, including the green algae, diatoms, hapto-phytes, and chloroarachniophytes (Koziol et al.2007). Identification of a separate haptophyte chla ⁄ c clade and the fact that a haptophyte subgroupwas also seen within the LI818 ⁄ LI818-like cladereflect the evolutionary divergence between haptophy-tes and diatoms. Although none of the six E. huxleyiLHCF proteins were found clustered within the redalgal LHCR and ⁄ or the LHCZ ⁄ LHCZ-like groups,this might be due to the restricted number of LHCFsequences available at the time of this analysis. OurSouthern blot analysis indicates the presence of atleast 10 additional LHCFs, and the genome miningsuggests even higher numbers approaching 50. It istherefore entirely possible that these additionalE. huxleyi LHCF proteins would fall into thesegroupings. Future analysis of these sequences willprovide further insight into the relationshipsbetween these complex LH protein families andmay begin to shed light on the puzzling question ofwhat the functional requirements are for the largenumber of closely related LH proteins.

Similar to all previously known LHC proteins, theE. huxleyi Lhcf genes analyzed here are nuclearencoded. The LHCF primary polypeptides aresynthesized in the cytosol and are predicted to belarger than the mature polypeptides observed onSDS-PAGE gels, suggesting that they have a prese-quence that is cleaved during posttranslationalmaturation of the protein. By analogy with theLHCFs from M. pyrifera, L. saccharina, L. digitata,and P. tricornutum (Grossman et al. 1990, Bhaya andGrossman 1993, Apt et al. 1995, De Martino et al.2000), it was possible to infer a bipartite signal

peptide of �40 amino acids. The first 14–16 aminoacids correspond to the endoplasmic reticulumsignal sequence, and the remainder to the chloro-plast transit peptide (Table 1). In secondary-derived,plastid-containing organisms, site-directed mutagen-esis experiments have shown Phe and other aromaticor bulky residues (Tyr, Trp, Leu) at the N-terminusto be essential for the correct targeting of proteinsto the plastid lumen (Killian and Kroth 2005, Gouldet al. 2006, Gruber et al. 2007, Patron and Waller2007). Although the presence of the highlyconserved Phe residue was found in only one of theE huxleyi LHCFs, in equivalent regions of the prese-quence in the remaining five LHCFs, there was apredominance of hydrophobic residues (Ala, Val,Pro, Gly) in keeping with previous analysis of hapto-phyte signal peptide cleavage sites. This result isperhaps not surprising as it has been shown thatonly 30% of haptophyte presequences analyzed sofar contain the consensus Phe residue (Patron andWaller 2007). Determining the transport sequencesof haptophyte species is not straightforward, andfunctional analysis will be required to confirm theabove predictions, which were based on sequencecomparisons.

The mature amino acid sequences of the sixLHCFs from E. huxleyi are quite divergent withbetween 10% and 84% similarity. The variability inthe amino acid sequence of the E. huxleyi LHCFs isstriking, and it is tempting to speculate that thisvariability reflects different functions. The secondarystructure of the LHCF family from E. huxleyi issimilar to that of other LHCs in that all membersexhibit the three characteristic transmembranealpha helices. The highly conserved chl-bindingmotifs identified previously in MSR 1 and MSR 3were present in all six E. huxleyi LHCF sequencesanalyzed. However, the residues in MSR 2 are notas highly conserved between the six LHCFsequences. Specifically, the two sequences belongingto the LI818 ⁄ LI818-like group (LHCF17H08 andLHCF19E10) share little homology with the otherfour proteins. Comparing the E. huxleyi LI818-likesequences with those of other photosyntheticeukaryotes showed that a number of residues were

Fig. 4. Differential expression of the Emiliania huxleyi LHCF proteins in response to light and CO2. Western blot analysis of proteinextracts from E. huxleyi cultures grown under four different environmental conditions. Total proteins (LHCF [10 lg]; RUBISCO [5 lg])were loaded in each lane, and blots probed using polyclonal antibodies raised against RUBISCO, LHCF16B09, LHCF31G05, andLHCF17H08. Each lane represents an independent sample taken from each growth condition. LHCF, light-harvesting complex fucoxan-thin-binding proteins; LL, low light (30 lmol photons Æ m)2 Æ s)1); HL, high light (300 lmol photons Æ m)2 Æ s)1); LCO2, low CO2

(180 ppm); HCO2, high CO2 (750 ppm).


highly conserved within this group. One of thefirst roles proposed for the LI818-like protein inC. reinhardtii was a protective function. Recently, inthe diatom C. cryptica, a role for the LI818-like pro-tein in the binding of diadinoxanthin and diatoxan-thin pigments was suggested (Richard et al. 2000,Beer et al. 2006, Gundermann and Buchel 2008). Itis possible that this LI818-like MSR2 forms the sitebinding these pigments.

Outside of the MSRs, the level of homologybetween the E. huxleyi LHCFs and those of otherorganisms is not high. For example, a highly con-served motif known to be essential for the forma-tion LHC proteins of higher plants, algae, anddiatoms into trimeric units is absent in the sixE. huxleyi sequences presented in this study (Buchel2003, Beer et al. 2006, Gundermann and Buchel2008). Very little is known about the oligomericstatus of the functional light-harvesting units in thehaptophytes, including E. huxleyi. The absence ofthis trimerization motif may suggest either thatthe structure of the E. huxleyi antenna complexes isdifferent, or that the assembly into trimers does notrequire the presence of this conserved motif.Conversely, a motif (F ⁄ YDPL ⁄ FGL) similar to theFDPLGL motif occurring in higher plants is presentin four out of the six E. huxleyi LHCFs. Interestingly,this motif has not been found in other chl a ⁄ c–con-taining organisms, with the exception of thehaptophyte I. galbana (Durnford et al. 1996). Thechimeric nature of the E. huxleyi LHCFs revealed byour analysis provides support for the suggestion thatthe diversity of LHC sequences arose from genefusions of smaller one-helix proteins and that theevolution of the LHC family of proteins is highlycomplex (Koziol et al. 2007).

Significantly lower steady-state mRNA levels wereobserved under high light for five out of the six Lhcftranscripts. In contrast, the Lhcf19E10 transcriptlevel was significantly higher in high-light cultures. Asimilar situation has also been observed in thediatom C. cryptica where transcript levels for fivedifferent fcp genes decreased under high light, buttranscript levels for three others increased underhigh-light conditions (Oeltjen et al. 2002). In thebrown alga M. pyrifera, a decrease in steady-statemRNA levels of five different fcp genes was observedin high light compared to low light, one transcriptbeing as much as 5-fold lower in high light(Apt et al. 1995). In contrast to the observed lightdependence of transcript levels, no distinct trendsin expression were detected in response to varyingCO2 levels.

Interestingly, the E. huxleyi Lhcf19E10; C. crypticafcp6, fcp7, fcp12; and C. reinhardtii light-induciblecab-related LI818 belong to the same LI818 ⁄ LI818-like group. Oeltjen et al. (2002) proposed thatC. cryptica fcp6, fcp7, and fcp12 might play a photo-protective role similar to early light-inducibleproteins (ELIPS) in green plants or light-inducible

cab-related protein in C. reinhardtii. ELIPS accumu-late transiently before the assembly of the LHCsand are thought to protect the newly synthesizedreaction centers against photooxidation. The E. hux-leyi Lhcf19E10 was more highly expressed underlong-term high-light treatment than low light in ourexperiment (Fig. 3). Here we propose that E. huxleyiLhcf19E10 could play a role in photoprotection bybeing associated specifically with pigments involvedin light dissipation, such as the xanthophyll-cyclepigments diadinoxanthin and diatoxanthin, whichaccumulated in high light and are involved in light-energy dissipation. A role in photoprotectionappeared to be a general feature associated with theLI818 ⁄ LI818-like protein family across organisms.

Transcript levels of rbcL and Lhcfs in differentlight conditions were not always reflected inchanges in the abundance of the correspondingproteins. In high light, significantly lower levels ofLHCF16B09 and LHCF31G05 polypeptides were evi-dent compared to low light, qualitatively consistentwith the reductions in transcript levels. In contrast,despite increased Lhcf17H08 transcript levels inresponse to high light, LHCF17H08 protein levelsdeclined in high light, following the patternobserved for the other LHCFs. Small increases inthe amount of LHCF16B09 and LHCF31G05 pro-teins were observed in response to high CO2 underboth high- and low-light treatments, despite tran-script levels being unaffected or reduced inresponse to high CO2. In addition, despite a signifi-cantly lower level of rbcL transcript levels in highlight, the RUBISCO protein levels were equal acrossall conditions tested. This discrepancy betweenmRNA levels and protein levels is not unexpected.E. huxleyi cultures under examination in this studywere long-term acclimated to the environmentalconditions, and under steady state, it is known inC. reinhardtii that a combination of transcriptionaland posttranscriptional mechanisms contribute tothe level of protein (Mussgnug et al. 2005, McKimand Durnford 2006). Similar mechanisms are alsolikely to take place in E. huxleyi.

Increases in the chl a ⁄ c ratio and decreases in theratio of xanthophyll-cycle pigments to chl a betweenlow- and high-light grown E. huxleyi cultures indi-cated a modification in the antennae size and com-position, with increased levels of pigments involvedin light-energy dissipation under high-light condi-tions. These data are in agreement with the chla ⁄ b–ratio variations observed in Dunaliella salina andChlorella vulgaris in response to changes to lightintensity (Maxwell et al. 1994, Webb and Melis1995) but contrast with the small changes observedin chl a ⁄ b ratio in C. reinhardtii (Neale and Melis1986). In high-light conditions, PSII reaction cen-ters in E. huxleyi are often highly reduced, and ifexcess light energy is not dissipated, radical oxygenspecies are produced, which in turn have the poten-tial to damage the photosystems.


The adaptive flexibility of E. huxleyi photosyntheticapparatus was demonstrated by a study in which itwas shown that two strains of E. huxleyi, including thestrain employed in our investigation, modified theantenna to reaction center ratio, thereby allowingphotoacclimation to the prevailing light environment(Suggett et al. 2007). Our molecular analysis of theLHCF proteins together with the changes in the dia-dinoxanthin and diatoxanthin pigments suggested acoordinated regulation of the light-harvesting anten-nae, allowing for flexibility to respond to changes inlight availability and offering enhanced photoprotec-tion against excess light energy.

This project was supported by NERC grant NER ⁄ A ⁄ S ⁄2003 ⁄ 00441 and the Leverhulme Foundation InternationalCollaborative Grant, F ⁄ 00 213 ⁄ M. NSF Grant BIO-OCE0723908 provided support for S. C. L. to undertake phylo-genetic analysis.

Abascal, F., Zardoya, R. & Psada, D. 2005. ProtTest: selection ofbest-fit models of protein evolution. Bioinformatics 21:2104–5.

Akaike, H. 1974. New look at the statistical-model identification.IEEE Trans. Automatic Control 19:716–23.

Apt, K. E., Clendennen, S. K., Powers, D. A. & Grossman, A. R.1995. The gene family encoding the fucoxanthin chlorophyllproteins from the brown alga Macrocystis pyrifera. Mol. Gen.Genet. 246:455–64.

Beer, A., Gundermann, K., Beckmann, J. & Buchel, C. 2006. Sub-unit composition and pigmentation of fucoxanthin-chloro-phyll proteins in diatoms: evidence for a subunit involved indiadinoxanthin and diatoxanthin binding. Biochemistry45:13046–53.

Berges, J. A., Franklin, D. J. & Harrison, P. J. 2001. Evolution of anartificial seawater medium: improvements in enriched sea water,artificial water over the last two decades. J. Phycol. 37:1138–45.

Berges, J. A., Franklin, D. J. & Harrison, P. J. 2004. Corrigendum:evolution of an artificial seawater medium: improvements inenriched seawater, artificial water over the last two decades.J. Phycol. 40:619.

Bhaya, D. & Grossman, A. R. 1993. Characterization of gene clustersencoding the fucoxanthin chlorophyll proteins of the diatomPhaeodactilum tricornutum. Nucleic Acids Res. 21:4458–66.

Bradford, M. M. 1976. A rapid and sensitive method for thequantification of microgram quantities of protein utilisingthe principle of protein dye binding. Anal. Biochem. 72:1423–31.

Bradshaw, A. L., Brewer, P. G., Shafer, D. K. & Williams, R. T. 1981.Measurements of total carbon dioxide and alkalinity bypotentiometric titration in the GEOSEGS program. Earth PlantSci. Lett. 55:99–115.

Brown, C. W. & Yoder, J. A. 1994. Coccolithophorid blooms in theglobal ocean. J. Geophys. Res. 99:7467–82.

Buchel, C. 2003. Fucoxanthin-chlorophyll proteins in diatoms: 18and 19 kDa subunits assemble into different oligomeric states.Biochemistry 42:13027–34.

Buitenhuis, E. T., van der Wal, P. & de Baar, H. J. W. 2001. Bloomsof Emiliania huxleyi are sinks of atmospheric carbon dioxide: afield and mesocosm study derived simulation. Glob. Biogeochem.Cycles 15:577–87.

Caron, L., Douady, D., Quinet-Szely, M., de Goer, S. & Berkaloff, C.1996. Gene structure of a chlorophyll a ⁄ c binding proteinfrom brown alga: presence of an intron and phylogeneticimplications. J. Mol. Evol. 43:270–80.

Chevenet, F., Brun, C., Banuls, A. L., Jacq, B. & Christen, R. 2006.TreeDyn: towards dynamic graphics and annotations foranalyses of trees. BMC Bioinformatics 7:439.

De Martino, A., Douady, D., Quinet-Szely, M., Rousseau, B.,Crepineau, F., Apt, K. & Caron, L. 2000. The light-harvestingantenna of brown algae. Eur. J. Biochem. 267:5540–9.

Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S., Cheve-net, F., Dufayard, J. F., et al. 2008. Phylogeny.fr: robustphylogenetic analysis for the non-specialist. Nucleic Acids Res.36:W465–9 (doi: 10.1093/nar/gkn180).

DOE. 1994. SOP 6: Determination of pH of sea water using aglass ⁄ reference electrode cell. In Dickson, A. G. & Goyet, C.[Eds.] Handbook of Methods for the Analysis of the VariousParameters of the Carbon Dioxide System in Sea Water, Version 2.DOE, ORNL ⁄ CDIAC-74, pp. 4–8.

Durnford, D. G., Aebersold, R. & Green, B. R. 1996. The fuco-xanthin-chlorophyll proteins from the chromophyte alga arepart of a large multigene family: structural and evolutionaryrelationships to other light harvesting antennae. Mol. Gen.Genet. 253:377–86.

Durnford, D. G., Deane, J. A., Tan, S., McFadden, G. I., Gantt, E. &Green, B. R. 1999. A phylogenetic assessment of the eukaryoticlight-harvesting antenna proteins, with implications for plastidevolution. J. Mol. Evol. 48:59–68.

Edgar, R. C. 2004. MUSCLE: multiple sequence alignment withhigh accuracy and high throughput. Nucleic Acids Res. 32:1792–7.

Eppard, M. & Rhiel, E. 1998. The genes encoding light-harvestingsubunits of Cyclotella cryptica (Bacillariophyceae) constitute acomplex and heterogeneous family. Mol. Gen. Genet. 260:335–45.

Eppard, M. & Rhiel, E. 2000. Investigations on gene copy number,introns and chromosomal arrangement of genes encoding thefucoxanthin chlorophyll a ⁄ c-binding proteins of the centricdiatom Cyclotella cryptica. Protist 151:27–39.

Garrido, J. L., Otero, J., Maestro, M. A. & Zapata, M. 2000. Themain non-polar chlorophyll c from Emiliania huxleyi (Prymne-siophyceae) is a chlorophyll c monogalactosyldiacylglycerideester: a mass spectrometry study. J. Phycol. 36:497–505.

Garrido, J. L. & Zapata, M. 1998. Detection of new pigments from(Emiliania huxleyi) (Prymnesiophyceae) by high-performanceliquid chromatography-mass spectrometry, visible spectros-copy, and fast atom bombardment spectrometry. J. Phycol.34:70–8.

Gascuel, O. 1997. BIONJ: an improved version of the NJ algorithmbased on a simple model of sequence data. Mol. Biol. Evol.14:685–95.

Goloboff, P., Farris, S. & Nixon, K. 2000. TNT: Tree Analysis UsingNew Technology (BETA). Published by the authors, Tucuman,Argentina.

Gould, S. B., Sommer, M. S., Kroth, P. G., Gile, G. H., Keeling, P. J.& Maier, U. G. 2006. Nucleus-to-nucleus gene transfer andprotein retargeting into a remnant cytoplasm of cryptophytesand diatoms. Mol. Biol. Evol. 23:2413–22.

Green, B. R. & Kuhlbrandt, W. 1995. Sequence conservation oflight-harvesting and stress-response proteins in relation to thethree dimensional molecular structure of LHCII. Photosynth.Res. 44:139–48.

Grossman, A., Manodori, A. & Snyder, D. 1990. Light-harvestingproteins of diatoms; their relationship to the chlorophyll a ⁄ bbinding proteins of higher plants and their mode of transportinto plastids. Mol. Gen. Genet. 224:91–100.

Gruber, A., Vugrinec, S., Hempel, F., Gould, S. B., Maier, U. G. &Kroth, P. G. 2007. Protein targeting into complex diatomplastids: functional characterisation of a specific targetingmotif. Plant Mol. Biol. 64:519–30.

Guglielmi, G., Lavaud, J., Rousseau, B., Etienne, A. L., Houmard, J.& Ruban, A. V. 2005. The light-harvesting antenna of thediatom Phaeodactylum tricornutum – evidence for a diadino-xanthin-binding subcomplex. FEBS J. 272:4339–48.

Guillard, R. R. L. & Ryther, J. H. 1962. Studies of marine planktondiatoms. I. Cyclotella nana Hustedt and Detonula confervacea(Cleve) Gran. Can. J. Microbiol. 8:229–39.

Guindon, S. & Gascuel, O. 2003. A simple, fast, and accuratealgorithm to estimate large phylogenies by maximum likeli-hood. Syst. Biol. 52:696–704.

Guindon, S., Lethiec, F., Duroux, P. & Gascuel, O. 2005. PHYMLOnline – a web server for fast maximum likelihood-basedphylogenetic inference. Nucleic Acids Res. 33:W557–9.


Gundermann, K. & Buchel, C. 2008. The fluorescence yield of thetrimeric fucoxanthin-chlorophyll-protein FCPa in the diatomCyclotella meneghiniana is dependent on the amount of bounddiatoxanthin. Photosynth. Res. 95:229–35.

Holligan, P. M., Viollier, M., Harbour, D. S., Camus, P. & Cham-pagne-Philippe, M. 1983. Satellite and ship studies of cocco-lithophore production along a continental shelf edge. Nature304:339–42.

Huelsenbeck, J. P. & Ronquist, F. 2001. MRBAYES: Bayesianinference of phylogenetic trees. Bioinformatics 17:754–5.

Iglesias-Rodriguez, M. D., Brown, C. W., Doney, S. C., Kleypas, J.,Kolber, D., Kolber, Z., Hayes, P. K. & Falkowski, P. G. 2002.Representing key phytoplankton functional groups in oceancarbon cycle models: coccolithophorids. Glob. Biogeochem.Cycles 16:1100. doi: 10.1029/2001GB001454.

Killian, O. & Kroth, P. G. 2005. Identification and characterisationof a new conserved motif within the presequence of proteinstargeted into complex diatom plastids. Plant J. 41:175–83.

Koziol, A. G., Borza, T., Ishida, K. I., Keeling, P., Lee, R. W. &Durnford, D. G. 2007. Tracing the evolution of light-harvest-ing antennae in chlorophyll a ⁄ b-containing organisms. PlantPhysiol. 143:1802–16.

Le, S. Q. & Gascuel, O. 2008. LG: an improved, general amino-acidreplacement matrix. Mol. Biol. Evol. 25:1307–20.

Leonardos, N. 2008. Physiological steady state of phytoplankton inthe field? An example based on pigment profile of Emilianiahuxleyi (Haptophyta) during a light shift. Limnol. Oceanogr.53:306–11.

Leonardos, N. & Geider, R. J. 2005. Elevated atmospheric carbondioxide increases organic carbon fixation by Emiliania huxleyi(Haptophyta), under nutrient-limited high-light conditions.J. Phycol. 41:1196–203.

Lewis, E. & Wallace, D. W. R. 1998. Program Developed for CO2 SystemCalculations. ORNL ⁄ CDIAC-105. Carbon Dioxide Inf. Anal.Cent., Oak Ridge Natl. Lab., Oak Ridge, Tennessee, 21 pp.

Maxwell, D. P., Falk, S., Trick, C. G. & Huner, N. P. A. 1994. Growtha low temperature mimics high-light acclimation in Chlorellavulgaris. Plant Physiol. 105:535–43.

McKim, S. M. & Durnford, D. G. 2006. Translational regulation oflight-harvesting complex expression during photoacclimationto high-light in Chlamydomonas reinhardtii. Plant Physiol. Bio-chem. 44:857–65.

Mussgnug, J. H., Wobbe, L., Elles, I., Claus, C., Hamilton, M., Fink,A., Kahmann, U., et al. 2005. NAB1 is an RNA binding proteininvolved in the light-regulated differential expression of thelight-harvesting antenna of Chlamydomonas reinhardtii. Plant Cell17:3409–21.

Neale, P. J. & Melis, A. 1986. Algal photosynthetic membranecomplexes and the photosynthesis-irradiance curve: a com-parison of light-adaptation responses in Chlamydomonas rein-hardtii (Chlorophyta). J. Phycol. 22:531–8.

Nielsen, M. V. 1997. Growth, dark respiration and photosyntheticparameters of the coccolithophorid Emiliania huxleyi (Prym-nesiophyceae) acclimated to different day length-irradiancecombinations. J. Phycol. 33:818–22.

Oeltjen, A., Krumbein, W. E. & Rhiel, E. 2002. Investigation ontranscript sizes, steady state mRNA concentrations and diurnalexpression of genes encoding fucoxanthin chlorophyll a ⁄ clight harvesting polypeptides in the centric diatom Cyclotellacryptica. Plant Biol. 4:250–7.

Paasche, E. 1998. Roles of nitrogen and phosphorus in coccolithformation in Emiliania huxleyi (Prymnesiophyceae). Eur. J.Phycol. 33:33–42.

Patron, N. J. & Waller, R. F. 2007. Transit peptide diversity anddivergence: a global analysis of plastid targeting signals.BioEssays 29:1048–58.

Richard, C., Ouellet, H. & Guertin, M. 2000. Characterization ofthe LI818 polypeptide from the green unicellular alga Chla-mydomonas reinhardtii. Plant Mol. Biol. 42:303–16.

Sciendra, A., Harlay, J., Lefevre, D., Lemee, R., Rimmelin, P., Denis,M. & Gattuso, J. P. 2003. Response of coccolithophorid Emil-iania huxleyi to elevated partial pressure of CO2 under nitro-gen limitation. Mar. Ecol. Prog. Ser. 261:111–22.

Stoll, H. M., Klaas, C. M., Probert, I., Encinar, J. R. & Alonso, J. I. G.2002. Calcification rate and temperature effects on Sr parti-tioning in coccoliths of multiple species of coccolithophoridsin culture. Glob. Planet Change 34:153–71.

Suggett, D., Le Floc’, H. E., Harris, G. N., Leonardos, N. &Geider, R. J. 2007. Different strategies of photoacclimationby two strains of Emiliania huxleyi (Haptophyta). J. Phycol.43:1209–22.

Wahlund, T. M., Hadaegh, A. R., Clark, R., Nguyen, B., Fanelli, M.& Read, B. A. 2004. Analysis of expressed sequence tags fromcalcifying cells of the marine coccolithophorid, Emilianiahuxleyi. Mar. Biotechnol. 6:278–90.

Webb, M. R. & Melis, A. 1995. Chloroplast response in Dunaliellasalina to irradiance stress: effect on thylakoid mem-brane protein assembly and function. Plant Physiol. 107:885–93.

Zondervan, I. 2007. The effect of light, macronutrients, tracemetals and CO2 on the production of calcium carbonate andorganic carbon in coccolithophores – a review. Deep-Sea Res.54:521–37.

Zondervan, I., Rost, B. & Riebesell, U. 2002. Effect of CO2

concentration on the PIC ⁄ POC ratio in the coccolitho-phore Emiliania huxleyi grown under light-limiting conditionsand different daylengths. J. Exp. Mar. Biol. Ecol. 272:55–70.

Supplementary Material

The following supplementary material is avail-able for this article:

Table S1. Pigment contents of Emiliania huxleyigrown in two light and CO2 conditions.

Fig. S1. Southern blot analysis of Emiliania hux-leyi genomic DNA reveals the presence of at least10 Lhcf genes. Southern blot of E. huxleyi geno-mic DNA digested with restriction endonucleasesnamed above the lanes and probed with radiola-bel PCR-amplified FCP cDNAs from E. huxleyi.LHCF, light-harvesting complex fucoxanthin-binding proteins.

This material is available as part of the onlinearticle.

Please note: Wiley-Blackwell are not responsi-ble for the content or functionality of any supple-mentary materials supplied by the authors. Anyqueries (other than missing material) shouldbe directed to the corresponding author for thearticle.


CHARACTERIZATION AND EXPRESSION ANALYSIS OF THE Lhcf GENE FAMILY IN EMILIANIA HUXLEYI (HAPTOPHYTA) REVEALS DIFFERENTIAL RESPONSES TO LIGHT AND CO21

Documents