5. Investigations on DDE overexpressing transformants 65
5.4.3. Chlorophyll fluorescence measurements
The xanthophyll cycle (XC) dependent non-photochemical chlorophyll fluorescence quench-ing (NPQ) was examined as a function of light intensity in the WT and the transformants.
As for the pigment analysis cells were either dark- or low light adapted and measurements were performed as described in the Material and Methods (chapter 5.3). After the cells were adapted to darkness for several hours, different patterns of NPQ development could be observed when comparing the WT and the ddeOE transformants (Fig. 5.4A). NPQ started to develop at an irradiance of about 60µmol·photons·m−2·s−1 in the WT and reached a maximal value of about 0.5 ±0.02 for the conditions used here. The development of NPQ occurred earlier (30 µmol·photons·m−2·s−1) in ddeOE1 and ddeOE4 and the maximal value of NPQ was also increased in comparison to the WT to 0.76 ±0.05 and 0.54± 0.03, respectively. Strikingly, the NPQ value measured at 30 µmol·photons·m−2·s−1 for those two transformants was already as high as for the WT at a light intensity which was 3 times higher (89µmol·photons·m−2·s−1). In contrast, the transformants ddeOE16 and ddeOE22 showed a maximal NPQ which was much lower than in the WT (about −44 %), although NPQ started to develop at the same irradiance as for the WT but to a much lower extent.
The presence of quite high amounts of DT already under dark conditions might explain the
Table 5.1.:Pigment analysis of the WT and the Dde overexpressing transformants of P. tricornutum. DD is diadinoxanthin and DT is diatoxanthin. Pigment content is shown in mol/100mol Chl a, cells were either adapted to low light (Low light, 45µmol·photons·m−2·s−1) or complete darkness (Dark) for several hours.
Values are average (±SD) of one to three measurements.
WT ddeOE1 ddeOE4 ddeOE16 ddeOE22
Dark
Chl c 12.33±0.13 12.27 ±0.01 12.28 ±0.03 12.16 ±0.21 12.23 ±0.02 Fucoxanthin 53.17±0.23 52.11 ±1.02 51.56 ±0.48 50.36 ±0.25 50.47 ±0.92
β-carotene 2.25 2.46 1.84 2.07 1.77
Diadinoxanthin 6.01±0.05 6.37±0.02 6.04±0.43 5.58 ±0.19 4.88 ±0.07 Diatoxanthin 0.14±0.11 0.38±0.25 0.74±0.25 0.55 ±0.14 0.6 ±0.14 DD+DT 6.15±0.16 6.75±0.27 6.78±0.68 6.14 ±0.33 5.48 ±0.21 DEP 2.23±1.81 5.48±3.45 10.71±2.6 8.93 ±1.84 10.92 ±2.4 Low light
Chl c 13.36 12.76 ±0.59 12.68±0.6 12.83 ±0.35 12.66 ±0.24 Fucoxanthin 48.69 48.2±0.47 48.22 ±0.74 46.87 ±0.59 46.95 ±1.08 β-carotene 1.54 2.6 2.28±0.02 2.62 ±0.24 2.15 ±0.35 Diadinoxanthin 5.78 4.33±1.07 3.73±0.8 3.43 ±0.58 3.48 ±0.87 Diatoxanthin 0.4 2.44±0.04 3.11±0.52 3.2±0.4 2.75 ±0.29 DD+DT 6.18 6.77±1.11 6.84±1.32 6.63 ±0.98 6.23 ±1.16 DEP 6.87 36.87 ±5.42 45.71±9.2 48.4 ±3.25 44.58 ±5.4
WT
ddeOE1
ddeOE4
ddeOE16
ddeOE22
DEP degree (%)
0 10 20 30 40 50
dark-adapted light-adapted
Figure 5.3.:De-epoxidation degree of the WT and theDdeoverexpressing transformants ofP. tricornutum. DEP (in %) = DT/(DD+DT)×100 where DD is diadinoxanthin and DT is diatoxanthin. Cells were either acclimated to darkness (dark-adapted) or low light (light-adapted) for several hours. Values are the average±SD of one to three measurements.
5.4. Results and Discussion
fast and strong increase in NPQ in the transformants ddeOE1 and ddeOE4. Although the same amount of DT was detected in WT under low light conditions, the development of NPQ for low light adapted cells (Fig. 5.4B) was still not as fast as for the transformants ddeOE1 and ddeOE4 after dark adaptation. We think that the fast development of NPQ in those transformants is due to a higher amount of enzyme which enhances the conversion of DD into DT already at low irradiances. Obviously a basic level of DT alone is not sufficient to enhance NPQ, as the NPQ pattern of the transformants ddeOE16 and ddeOE22 shows.
Even if the amount of DT in the dark was the same as in the other two transformants, NPQ did not develop as fast and strongly. The reason might be lacking protonated LHC-sites, which activate the DT molecules dissipating excess energy [134, 74]. Different NPQ phenotypes are reported for dark-adapted leaves of aba-1 mutants of A. thaliana, which are unable to epoxidize zeaxanthin and therefore accumulate this xanthophyll constitutively.
While Tardy and Havaux observed that NPQ induction was faster with the same final am-plitude [214], a lower amam-plitude was measured by Pogson and co-workers [195]. Still, they also observed a faster induction in NPQ and suggested, that in the presence of zeaxanthin in the cell, the induction of NPQ solely depends on the generation of a transthylakoid proton gradient upon illumination. When NPQ and de-epoxidation as a function of time were mea-sured in dark-adapted leaves of tobacco plants overexpressing the violaxanthin de-epoxidase (VDE) ofA. thaliana, a faster induction of both parameters was observed and no differences in amplitude in comparison to tobacco WT plants [101].
Figure 5.4B shows the development of NPQ for low light adapted cells. As for dark-adapted conditions, the WT started to develop NPQ at a light intensity of about 60µmol· photons·m−2·s−1, but reached a higher maximal value of 0.96±0.06. While only one mu-tant showed a higher maximal NPQ value of 1.13±0.44 (ddeOE1) under these conditions, all other transformants had a lower NPQ amplitude varying from 0.66 to 0.72. Interestingly, in contrast to ddeOE1, NPQ started to develop much earlier in these transformants than in the WT (already at an irradiance of 30µmol·photons·m−2·s−1) and partially with high values corresponding to 30 % of the maximal amplitude for the respective strain (ddeOE4 and ddeOE22). Similar observations were made for cells of the npq2 mutant of C. rein-hardtiiwhich accumulates zeaxanthin. Cells were also low light adapted and showed a lower but accelerated pattern of NPQ [174]. In contrast, npq2 mutants of Arabidosis showed an NPQ amplitude similar as WT. But an accelerated NPQ induction was also observed, not only as a function of time but also as a function of light intensity [174]. Accumulation of zeaxanthin does not necessarily lead to altered NPQ patterns, as shown in zea1 mutants ofDunaliella salina where no differences in the overall kinetics of NPQ induction (vs time) were observed for low light adapted cells [216]. No differences in growth in comparison to the WT could be observed for the ddeOE transformants of P. tricornutum, similar to the zea1 mutant of Dunaliella and the npq2 mutants of Arabidosis [114, 174]. Assuming that
A B
Figure 5.4.:NPQ measurements of the WT and theDde overexpressing transformants ofP. tricornutum. Non-photochemical Chl a fluorescence quenching (NPQ) development in the WT and the Dde overexpressing transformants ofP. tricornutumcells as a function of light intensity. Cells were either adapted to the dark (A) or to low light (B). Values are average±SD of one to two measurements.
the number of NPQ loci in the LHC antenna systems is similar in the WT and the ddeOE transformants, the reason for the different NPQ patterns could be a disequilibrium between the DT synthesis and the protonation of LHC sites. This could disturb the activation of DT molecules and ultimately lead to a disturbance in NPQ development as a function of light intensity [134].
Conclusion
Our results show, that overexpressing the diadinoxanthin de-epoxidase in P. tricornutum leads to a higher de-epoxidation degree and to an altered NPQ pattern. It seems that the de-epoxidation degree and the NPQ kinetics are independent of the strength of the overexpression. The generated mutants represent a good opportunity to study, whether diatoxanthin is helping the organism to protect itself from photo-oxidative damage not only by dissipating excess energy as heat but also by preventing lipid peroxidation as proposed for zeaxanthin inC. reinhardtii [17].
5.4. Results and Discussion
Acknowledgements
We like to thank D. Ballert for help with the transformation and cultivation of the cells and I.
Adamska for access to the Imaging PAM and HPLC instrument. This work was supported by the University of Konstanz and grants from the Deutsche Forschungsgemeinschaft (project LA 2368/2-1) to JL.
6. Evolution, cellular localization and light-dependent transcription of members of the light-harvesting complex (LHC) protein superfamily in the diatom Phaeodactylum tricornutum
Sabine Sturm*,†, Johannes Engelken*, 1, Ansgar Gruber*, Sascha Vugrinec, Iwona Adamska, Peter G. Kroth and Johann Lavaud2
Fachbereich Biologie, Universität Konstanz, 78457 Konstanz, Germany
*These authors contributed equally to this work.
†Author for correspondence. E-mail: sabine.sturm@uni-konstanz.de
1present address: Institute of Evolutionary Biology (CSIC-UPF), Pompeu Fabra University, Dr.
Aiguader 88, 08003 Barcelona, Spain.
2present address: UMR CNRS 6250 ‘LIENSs’, Institute for Coastal and Environmental Research, University of La Rochelle, 2 rue Olympe de Gouges, 17042 La Rochelle Cedex, France.
6.1. Abstract
The LHC-like superfamily of proteins comprises different families of chlorophyll binding pro-teins with one to four transmembrane helices. We identified members of this superfamily in the diatoms Phaeodactylum tricornutum and Thalassiosira pseudonana and found one helix protein 1 -like (OHP1-like), one helix protein 2 (OHP2) and stress enhanced Protein (SEP, two transmembrane helixes) to be encoded on the nuclear genomes of the investi-gated diatoms. In addition, we found so far uncharacterized three helix LHC-like proteins.
Phylogenetic analyses revealed that these proteins are not related to the three helix ELIPS (early light induced proteins) found in green algae and higher plants. Instead they form a distinct protein family that is exclusively found in red algae and algae with secondary plastids of red algal origin. Therefore we termed this protein family REDCAP (Red lin-eage CAB-like Proteins). Via presequence analyses and GFP fusion poteins we found out that these proteins are plastid targeted in diatoms. Transcription patterns of the REDCAP genes ressemble those of FCP genes rather than those of ELIPS in higher plants as shown by quantitative PCR. This indicates that REDCAPS might have a different function in dia-toms than ELIPS in higher plants. Also, the other investigated OHP1-like, OHP2 and SEP genes showed different transcription patterns compared to their respective plant homologues.
One of the investigated OHP1-like genes shows a similar transcription pattern as the HLIP (high light induced proteins) genes in cyanobacteria. Taken together our results show that LHC-like genes in diatoms are distributed and transcribed in a different way than in green algae and plants, which might reflect the differences in high light protection between these groups.
Keywords
Chloroplast· Light stress ·Early light-induced proteins ·Light-harvesting chlorophyll a/b-binding (LHC) proteins
Abbreviations
CAB: chlorophyll a/b-binding protein; CB: chlorophyll-binding; Chl: chlorophyll; ELIP:
early light-induced protein; FCP: fucoxanthin-chlorophyll a/c-binding protein; HL: high light; HLIP: high light inducible protein; LHC: light-harvesting complex; LL: low light; NPQ:
non-photochemical chlorophyll fluorescence quenching; OHP: one-helix protein; PS: photo-system; REDCAP: red lineage chlorophyll a/b-binding like protein; SEP: stress-enhanced protein; TM: transmembrane; TMH: transmembrane helix; WT: wildtype.
6.2. Introduction
6.2. Introduction
Light energy is absorbed by higher plants, algae and cyanobacteria to drive oxygenic pho-tosynthesis [81]. The light harvesting antennae of higher plants are composed of several proteins that are members of the LHC (light harvesting complex) superfamily, which can be divided into four groups, 1. the LHC family, consisting of CAB (chlorophyll (Chl) a/b-binding proteins), FCP (fucoxanthin Chl a/c-binding proteins) and LI818 proteins, 2. the LHC-like family, consisting of ELIP (early-light induced proteins), SEP (stress enhanced proteins), OHP (one-helix proteins) and LHL4 (LHC-like protein 4), 3. the REDCAP fam-ily and 4. PSBS (photosystem II subunit S) [54]. While cyanobacteria and red algae use phycobilisomes as their major LHC [82], in higher plants and green algae the light-harvesting antennae are composed of several homologous light-harvesting CAB proteins that are en-coded by the Lhc gene family and contain three transmembrane (TM) α-helices [110, 40].
They also share the chlorophyll-binding (CB) fold designated as the LHC motif [111]. Differ-ent environmDiffer-ental parameters such as light and nutriDiffer-ents regulate the synthesis of the LHC proteins [111], which are translated in the cytosol and posttranslationally directed to the chloroplasts, where they associate with pigments and insert into the thylakoid membrane.
LHC-type polypeptides that bind Chl c and fucoxanthin, FCPs, are found in brown algae, diatoms, dinoflagellates and chrysophytes [81].
While some researchers have proposed that a four-helix intermediate might be the ances-tor of the LHC-like, PSBS and LHC families [79, 75, 48, 164], a new hypothesis suggests that PSBS and LHCs originated from a group of two-helix proteins, while LHC-like proteins evolved independently [54]. In higher plants, most algae and cyanobacteria, the LHC-like protein family consists of three helix ELIPs [2, 164], two helix SEPs [100] and OHPs [112, 11], which are called high light induced proteins (HLIPs) or small CAB-like proteins (SCPs) in cyanobacteria [45, 69]. While the main function of the LHC proteins is the absorption of light by excitation of Chl molecules and the subsequent transfer of energy to the reaction centers of the photosystems, members of the LHC-like protein family were shown to be involved in stress protection. Especially their role in photoprotection is discussed widely as the expression levels of some proteins belonging to this family were shown to be increased under excess light conditions. In cyanobacteria for example, the HLIPs encoded by the hli or scp genes were shown to be crucial for the survival of the cells under high light (HL) conditions [99]. Mutants lacking HLIPs were unable to compete with wild-type (WT) cells under HL and mutants lacking all four HLIPs inSynechocystis sp. PCC6803 gradually lose photosynthetic function, a process called photoinhibition, and die following exposure to HL.
Different functions are proposed for the HLIPs, it may serve as Chl carriers [235] or sup-port dissipation of light energy absorbed in excess [97]. However, it is widely accepted that HLIPs might perform multiple functions and contribute to specific stress responses, thereby helping cyanobacteria to cope with various stress conditions [22, 34]. Wang and co-workers
found that all four HLIPs are associated with photosystem I (PSI) and thathli single mu-tants lose more than 30 % of the PSI trimers and show reduced PSI activity after exposure to HL for 12 h. From their results they suggest that HLIPs stabilize PSI trimers under HL and thereby help to protect the cells under light stress [228]. However, there are also reports that HLIPs are associated with PSII [199, 237, 131]. ELIPs were first described as polypeptides that are transiently expressed during greening in etiolated seedlings of pea and barley [160, 80]. Later, it was shown that ELIPs also accumulate under a variety of stress conditions that would cause photoinhibition [3, 4, 196]. All ELIPs contain conserved amino acid residues that could potentially bind Chl [78]. It has been suggested that they could function as pigment carriers during thylakoid membrane development, thereby protecting the photosynthetic apparatus from oxidative damage [164, 97, 2] or that they could influence Chl biosynthesis [1, 220]. Nevertheless, the exact role of ELIPs under light stress conditions remains unclear. The presence of two helix SEPs in Arabidopsis thaliana was also shown and their transcript level was increased under HL stress [100], however their physiological function has not yet been investigated.
Recently, four non-LHC members of this protein superfamily (with one, two and three pu-tative membrane helices) were identified in the genome of Phaeodactylum tricornutum [54].
In the present study, we characterized these novel proteins concerning their evolution, cellu-lar localization and light dependent transcription. As other LHC-like proteins were shown to be involved in photoprotection in higher plants and cyanobacteria, we were especially interested in the regulation of these genes under diverse light conditions.
6.3. Materials and Methods
6.3. Materials and Methods
6.3.1. Cell cultivation and light experiments
Phaeodactylum tricornutum(University of Texas Culture Collection, strain 646) was grown at 22◦Cat a light intensity of 45µmol·photons·m−2·s−1with a light/dark cycle of 16h/8h.
Cells were cultured in f/2 seawater medium [87] prepared with ‘Tropic Marinr’ artificial seawater (Dr. Biener GmbH, Wartenberg/Angersbach, Germany) at a final concentration of 50 % compared to natural seawater and continuously bubbled with sterile air. For the light experiments, cultures at mid-logarithmic phase were transferred to moderate light (750µmol·photons·m−2·s−1), darkness or maintained at LL (45µmol·photons·m−2·s−1) for 33h. Photosynthetic active radiation was measured using a hand-held quantum photometer (Model LI-185A, Li-Cor Inc., Lincoln, NE, USA). With the beginning of the dark phase cells were harvested every 3 h by centrifugation at 3000 g for 1 min. Resulting cell pellets were frozen in liquid nitrogen and stored at -80◦Cuntil use. For the HL experiment, cells adapted to LL for 6hwere transferred to a light intensity of about 1500–2000µmol·photons·m−2·s−1 for 2h. Samples were taken as described above at 0, 15, 30, 45, 60 and 120 min after the HL treatment. Cultures were transferred back to LL after the HL exposure and a sample was taken after 4 hof recovery. Four independent experiments were performed.
6.3.2. Sequence search and annotation
Sequences from the genomes ofP. tricornutumv2.0 (http://genome.jgi-psf.org/Phatr2/
Phatr2.home.html) [25] and Thalassiosira pseudonana v3.0 (http://genome.jgi-psf.
org/Thaps3/Thaps3.home.html) [14] were identified online at the United States Depart-ment of Energy Joint Genome Institute (JGI) (http://www.jgi.doe.gov/) using TBLASTN and BLASTP [9]. Additional sequence data was collected from public databases including NCBI (http://www.ncbi.nlm.nih.gov), UniProt (http://www.expasy.uniprot.org/) and JGI. Diatom genes were annotated manually from DNA sequence using Genewise [23], and 6-frame translation at the JGI genome browser. The newly identified diatom sequences of the LHC protein superfamily were classified according to their predicted secondary structures as well as sequence similarity to known CB proteins as described [54]. Prediction of TM alpha-helices in the new sequences was done with the DAS algorithm [35], which is optimized for prokaryotic membrane proteins and therefore is well suited for chloroplast targeted pro-teins. DAS plots for the diatom sequences are given (figure A.13, page 134). Signal peptides were identified with the help of SignalP (http://www.cbs.dtu.dk/services/SignalP/)’s Neuronal Networks (NN) and Hidden Markov Models (HMM) [171, 172] Transit peptides were predicted with the program ChloroP [53], for a detailed description of protein lo-calisation prediction see also [52]. Bipartite signal peptides were manually predicted by their characteristic N-terminal sequence motif [84, 119]; additionaly the HECTAR (http:
//www.sb-roscoff.fr/hectar/) prediction server [85] was used to predict Heterokont plas-tid targeting signals.
6.3.3. Isolation of RNA, reverse transcription and quantitative PCR (qPCR) Cells were crushed under liquid nitrogen with mortar and pestle and total RNA was iso-lated using a combination of Phenol/Chloroform extraction with Trizolr reagent (Invitro-gen, Carlsbad, CA, USA) and the RNeasyr Kit (Qiagen, Hilden, Germany). Genomic DNA (gDNA) contaminations were removed using TurboTM DNase (Ambion, Woodward, TX, USA) according to the manufacturer’s instructions. 350 ng gDNA free RNA was re-verse transcribed with the QuantiTectrreverse transcription kit (Qiagen, Hilden, Germany).
Resulting cDNA was diluted 4-fold in RNase/DNase free water and one µL of this cDNA template was used in a 20 µL qPCR reaction containing primers and DNA polymerase master mix with SYBRr Green (MESA GREEN qPCR MasterMix Plus for SYBRrAssay Low ROX, Eurogentec Deutschland GmbH, Cologne, Germany). The reaction was heated to 95◦Cfollowed by 40 cycles for 15sat 95◦Cand 1minat 60◦C. The amount of amplified DNA was monitored by measuring fluorescence at the end of each cycle using the Real-Time PCR System 7500 (Applied Biosystems, Lincoln, CA, USA). Each gene was analyzed using the primers given in table A.7, page 130. Relative transcript levels were calculated using RESTras described in [192] using the first sample of each light condition as calibrator and 18S rDNAas endogenous control.
6.3.4. PCR and construction of plasmids
Standard cloning procedures were used [206]. Polymerase chain reaction (PCR) was per-formed in a Master Cycler Gradient (Eppendorf, Hamburg, Germany) using recombinant Pfu polymerase (Fermentas, St. Leon Rot, Germany) according to the manufacturer’s in-structions. TheP. tricornutumtransformation vector pPha-T1 (GenBank AF219942, [239]), was used as described in [84, 119]. All constructs were sequenced (GATC Biotech AG, Kon-stanz, Germany) from their 5’ end, to ensure correct cloning.
6.3.5. Nuclear transformation
Nuclear transformation of P. tricornutum has been performed using a Bio Rad Biolistic PDS-1000/He Particle Delivery System (Bio-Rad, Hercules, CA, USA) fitted with 1350 psi rupture discs as described previously [12, 130]. For the selection and cultivation of P. tri-cornutum transformants 75 µg·ml−1 Zeocin (Invitrogen, Carlsbad, CA, USA) was added to the solid medium. The plates were incubated at 20◦C under continuous illumination (50µmol·photons·m−2·s−1) for three weeks.
6.3. Materials and Methods
6.3.6. Microscopy
Transformed cell lines were screened for the expression of GFP and using an Olympus BX51 epifluorescence microscope (Olympus Europe, Hamburg, Germany). Nomarski’s differential interference contrast (DIC) illumination was used to view transmitted light images (100×
UplanFL objective, Olympus). Chlorophyll autofluorescence and GFP fluorescence of the transformants were dissected using the mirror unit U-MWSG2 (Olympus) and the filter set 41020 (Chroma Technology Corp, Rockingham, VT, USA) respectively. Images were acquired with a confocal laser scanning microscope LSM 510 META (Carl Zeiss MicroImag-ing GmbH, GöttMicroImag-ingen, Germany) usMicroImag-ing a Plan-Apochromat 63×/1.4 Oil DIC objective.
GFP and Chl fluorescence was excited at 488nm, and detected by the meta detector with spetral resolution (lambda mode) at 16 bit dynamic range. GFP and Chl fluorescences were separated via linear unmixing. Reference spetra were beforehand aquired from wild type cells (for chlorophyll autofluorescence) and a transformed cell line expressing cytosolic GFP with spatial separation from the plastidic chlorophyll autofluorescence (for GFP fluo-rescence). Maximum intensity z-projections were calculated from slices of image stacks to ensure complete detection of fluorochromes within a cell. Transmitted light images (488nm wavelength) were recorded separately after the fluorescent image stacks were recorded.
6.3.7. Phylogenetic analysis
6.3.7. Phylogenetic analysis