Proteomics 2007, 7, 0000–0000
1
DOI 10.1002/pmic.200700407
RESEARCH ARTICLE
Comparative quantitative proteomics to investigate the remodeling of bioenergetic pathways under iron deficiency in Chlamydomonas reinhardtii Bianca Naumann1, 2, Andreas Busch1, 2, Jens Allmer1, 2, Elisabeth Ostendorf 1, Martin Zeller3, Helmut Kirchhoff 4 and Michael Hippler1, 2 1
Institute of Plant Biochemistry and Biotechnology, University of Münster, Münster, Germany Plant Science Institute, Department of Biology, University of Pennsylvania, Philadelphia, PA, USA 3 Thermo Fisher Scientific, Bremen, Germany 4 Institute of Botany, University of Münster, Münster, Germany 2
The basic question addressed in this study is how energy metabolism is adjusted to cope with iron deficiency in Chlamydomonas reinhardtii. To investigate the impact of iron deficiency on bioenergetic pathways, comparative proteomics was combined with spectroscopic as well as voltametric oxygen measurements to assess protein dynamics linked to functional properties of respiratory and photosynthetic machineries. Although photosynthetic electron transfer is largely compromised under iron deficiency, our quantitative and spectroscopic data revealed that the functional antenna size of photosystem II (PSII) significantly increased. Concomitantly, stressrelated chloroplast polypeptides, like 2-cys peroxiredoxin and a stress-inducible light-harvesting protein, LhcSR3, as well as a novel light-harvesting protein and several proteins of unknown function were induced under iron-deprivation. Respiratory oxygen consumption did not decrease and accordingly, polypeptides of respiratory complexes, harboring numerous iron–sulfur clusters, were only slightly diminished or even increased under low iron. Consequently, iron-deprivation induces a transition from photoheterotrophic to primarily heterotrophic metabolism, indicating that a hierarchy for iron allocations within organelles of a single cell exists that is closely linked with the metabolic state of the cell.
Received: April 27, 2007 Revised: June 20, 2007 Accepted: July 25, 2007
Keywords: Bioenergetic pathways / Chlamydomonas reinhardtii / Comparative proteomics / Iron deficiency / Mass spectrometry
1
Introduction
According to the WHO iron deficiency is the most common nutritional disorder in the world (http://www.who.int/nut/ ida.htm). This is due to the fact, that despite the high abunCorrespondence: Professor Michael Hippler, Institute of Plant Biochemistry and Biotechnology, University of Münster, Hindenburgplatz 55, 48143 Münster, Germany E-mail:
[email protected] Fax: 149-251-8328371 Abbreviations: ETR, electron transfer rate; LHCI, light-harvesting complex of PSI; LHCII, light-harvesting complex of PSII; PSI, photosystem I; PSII, photosystem II
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
dance of iron on earth, its bioavailability is restricted. Iron is used as a cofactor in numerous biochemical pathways and especially its ability to donate and accept electrons makes iron crucial in enzymes (heme- and iron–sulfur proteins) that catalyze redox reactions. Since iron can also react with oxygen to generate cytotoxic agents, its accessibility within the cell has to be under tight homeostatic control, which requires complex regulatory mechanisms [1]. Heme- and iron–sulfur proteins are highly abundant in energy transducing pathways like respiration and photosynthesis, thus translating in a high demand for iron in the cell. In eukaryotic cells, the mitochondrion is a major ironutilizing compartment. It is well established that iron is transported to the mitochondria for hemesynthesis and www.proteomics-journal.com
2
B. Naumann et al.
iron–sulfur cluster assembly. This is required for the formation of a functional respiratory electron transport machinery [2]. A number of studies [3–6] have indicated that in plants, fungi, and also mammalian muscle, respiratory chain components, particularly FeS containing NADH dehydrogenase and succinate dehydrogenase, are reduced greatly under iron-limitation while cytochrome containing complexes are less dramatically affected. In plants, the chloroplast is another major sink for iron. Changes in chloroplast structure, photosynthetic capacity, and the composition of thylakoid membranes have been described for plants that are challenged with iron malnutrition (for review see ref. [7, 8]). In all the photosynthetic organisms, iron deficiency leads to the activation of the iron uptake systems. For example, the accumulation of the ferroxidase in Chlamydomonas reinhardtii is enhanced very rapidly in iron limited conditions, before any chlorosis symptoms become visible [9]. On the other hand, inactivation of IRT1, the most prevalent Fe21 transporter in Arabidopsis thaliana, leads to a dramatic iron deficiency and therefore chlorosis [10–12]. Despite the evolution of elaborated iron-uptake mechanisms in plants, iron deficiency chlorosis remains an agricultural problem. This deficiency is even enforced in an alkaline environment by the low solubility of Fe31 oxides. The global impact of iron deficiency on photosynthetic productivity has also been shown in vast ocean regions that are severely limited in iron [13, 14]. In consequence, photosynthesis in the oceans and also on land occurs in an ironlimited environment. On the molecular level, photosystem I (PSI) is a prime target of iron deficiency, probably because of its high iron content (12 Fe per PSI). In the green alga C. reinhardtii iron deficiency leads not only to a pronounced degradation of PSI, but also to a remodeling of the PSI-associated lightharvesting antenna (LHCI), which precedes severe iron deficiency [9, 15]. These structural changes decrease the functional efficiency of excitation energy transfer between LHCI and PSI and minimize photo-oxidative stress to the thylakoid membrane. Interestingly, the same phenomenon is observed in red algae [16, 17]. Cyanobacteria, also respond to iron deficiency by degradation of light harvesting phycobilisomes [18]. Additionally, cyanobacteria express the “iron-stress-induced” gene isiA. The isiA protein has significant sequence similarities with CP43, a chlorophyll a-binding protein of photosystem II (PSII; [19, 20]) and forms a ring of 18 molecules around a PSI trimeric reaction center, as shown by electron microscopy [21, 22]. Interestingly, in the obligate eukaryotic photoautotrophic alga Dunaliella salina a chlorophyll a/b-binding protein homolog is induced by iron deficiency and functionally associated with PSI [23]. Overall, these results highlight the adaptive nature of the response to iron deficiency that is directed toward optimizing the photosynthetic architecture to the conditions in which iron is a limiting cofactor. The driving force towards adaptation to iron deficiency is fur 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Proteomics 2007, 7, 0000–0000
ther exemplified by constitutive differences in the photosynthetic architecture, as demonstrated for coastal and oceanic diatoms [24]. The basic question addressed in this study is how energy metabolism is adjusted to cope with iron deficiency. To study dynamic changes of the bioenergetic proteome of the green alga C. reinhardtii in response to iron deficiency, we employed a comparative proteomics approach using tandem mass spectrometric peptide profiling. In addition, spectroscopic and voltametric oxygen measurements were performed to assess the functional status of respiratory and photosynthetic electron transfer complexes. The data revealed a rather dynamic remodeling of bioenergetic pathways and indicated that Chlamydomonas cells change from photoheterotrophic to primarily heterotrophic metabolism under progressing iron deficiency. Interestingly, concomitant with the change in metabolic status is an enlargement of the functional PSII antenna size, which is thought to play a role as pigment buffer and storage allowing fast restoration of photosynthesis after iron becomes available again.
2
Materials and methods
2.1 Strains and cultures For all experiments the arginine auxothrophic Chlamydomonas strain CC424 mt- was used. Cells were grown under standard conditions [15] or (i) in the presence of isotopically labeled 13C6-L-arginine essentially as described in ref. [15], (ii) for 5 days under iron deficient conditions as described in Moseley et al. [9] and (iii) under either low light: 2.5, medium light: 50, or high light: 250 mE?m22 ?s21 for 24 and 48 h. 2.2 Isolation of intact chloroplasts and thylakoid membranes Intact chloroplasts and thylakoid membranes were isolated essentially as described in Naumann et al. [15]. 2.3 Protein analysis, immunodetection, and LC-MS/ MS analyses of proteins Protein analysis and immunodetection were done as previously described in ref. [25]. Antibodies against LhcSR3 and C_420064 proteins were peptide antibodies raised using the following peptides EDVFAYTKNLPGVTA and SAKKDDGYISEDEGL, respectively and were obtained from Eurogentec (Belgium). LC-MS/MS analysis was performed essentially as in ref. [15] with the following modifications. For nano-LC the Ultimate 3000 system was used. Buffers and flow rates were the same as in [15]. Quantitative experiments were carried out using samples from four independent cell harvests. www.proteomics-journal.com
Plant Proteomics
Proteomics 2007, 7, 0000–0000
For each preparation the mean and the SD of the Fe2/ Fe1 ratios from all detected y-ions for a peptide were calculated. In addition, the mean ratio (Fe2/Fe1) of a protein over all preparations was calculated. This was done by averaging the respective mean values from the different preparations. The SD of the mean is represented as the combined SD that was calculated by taking all y-ion pairs, measured over all preparations, into account. After careful revision of the data it was decided to discard values with extremely high SDs. It is of note, that from the four biological replicates, LhcSR3 was induced under iron deficiency in three biological harvests (H2–H4). The differential expression pattern within the biological replicates might be due to the fact that besides iron deficiency also light stress is an important factor for induction of the protein as demonstrated (see Fig. 4). Additional LC-MS/MS analysis (see Fig. 7) was carried out on a LTQ Orbitrap (Thermo Fisher Scientific, Bremen, Germany) using Tune instrument software version 2.0 Developers kit. Peptide digests were separated by nano-LC consisting of a Surveyor MS pump plus (Thermo Fisher Scientific, San Jose, CA, USA) and a MicroAS autosampler (Thermo Fisher Scientific). Peptides were loaded at high flow (5 mL/min) onto a trap column (100 mm internal diameter, biphasic, 30 mm C18 RP material 10R2 and 10 mm R18Aqua, all 5 mm particles, nanoseparations, The Netherlands) for 5 min with mobile phase A containing water and 0.1% formic acid (both HPLCgrade; Sigma–Aldrich, Taufkirchen, Germany). Injection method was partial loop and 1 mL digest was injected. Peptides were separated on a 10 cm long, 75 mm internal diameter C18 RP (Biosphere, 5 mm particles, nanoseparations, The Netherlands) at a flow of 250 nL/min. The gradient was from 3% mobile phase B containing ACN and 0.1% formic acid (both HPLC-grade, Sigma–Aldrich) to 35% B in 35 min, followed by a wash and re-equilibration step. Peptides were ionized by nano-ESI using a source voltage of 1.8 kV. Following target values were used: FT full scan, 1 000 000; FTMSn, 50 000; IT-MSn, 10 000. The maximum accumulation times and number of microscans were as follows: FT full scan, 1000 msec, 1; FT-MSn, 750 msec, 2; IT-MSn, 500 msec, 1. Peptides were fragmented by HCD in the C-trap. HCD stands for “higher collision energy dissociation”. The C-trap is a device to store and collisionally cool ions prior to their injection in the Orbitrap. The C-trap is situated between the linear IT of the LTQ and the Orbitrap. Dissociation can be achieved by increasing the voltage offset between the linear trap and the C-trap.
3
2.5 Chlorophyll a fluorescence induction kinetic Induction curves induced by broad green excitation were measured in a laboratory built fluorometer described in detail in ref. [26]. Chlamydomonas cells were incubated in 7 mM MgCl2, 40 mM KCl, 10 mM MES, pH 6.5 (KOH) at a chlorophyll concentration of 10 mmol?L21. After incubating the membranes in strict darkness for at least 5 min 50 mM DCMU was added and the induction kinetics were recorded after 1–2 min. 2.6 Oxygen evolution Oxygen evolution was measured with a Clark-type O2 electrode. Whole cells were centrifuged for 5 min, 5000 rpm at room temperature. The supernatant was removed by tilting and the cells were carefully resuspended in the remaining medium. The electrode was calibrated using dithionite. O2 evolution or consumption was measured in the amount of cells corresponding with 20 mg chlorophyll dissolved in 1 mL TAP medium and 1.5 mM 2,5-dimethyl-1,4-benzoquinone. Measurements were carried out in the dark or under various light conditions. 2.7 Isolation of total RNA and cDNA synthesis Total RNA of Chlamydomonas cells was isolated with TRIreagent (Sigma) according to the supplier’s manual. The subsequent cDNA synthesis was performed with the Reverse Transcription System (Promega) according to the manual. 2.8 Quantitative RT-PCR Quantitative RT-PCR was performed using isolated cDNA from iron deficient or sufficient cells and the gene specific primers 50 -tgc ctg cag ggc ttc cag g-30 and 50 -ggg atc cac tcg acg aag ta-30 for b-tubulin, 50 -gca atg aag act gca aca cc-30 and 50 - cca cgc tgc tct tgc tta g-30 for lhcbP1 and the primer pair 50 - ttc ccc ctg ttt ttc aac tg-30 and 50 -aca ggc tct tga ggt tgt cg -30 for lhcSR3. 2.9 Low temperature (77 K) fluorescence emission spectroscopy Whole cells (5 mg chlorophyll/mL) were resuspended in 60% glycerin, 10 mM HEPES pH 7.5 and frozen in liquid nitrogen. Low temperature fluorescence emission spectra were recorded with the FP-6500 spectrofluorometer (Jasco). The obtained data were normalized to the emission peak at 685 nm.
2.4 Chlorophyll fluorescence
3
Chlorophyll fluorescence from whole cells was measured with a MAXI-IMAGING-PAM Chlorophyll Fluorometer (Walz, Effeltrich, Germany).
To enumerate quantitative changes in the thylakoid proteome in response to iron deficiency, a comparative quantitative proteomics approach was employed. On that account,
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Results
www.proteomics-journal.com
4
B. Naumann et al.
we took advantage of stable isotope labeling of amino acids in cell cultures by using arginine auxotrophic Chlamydomonas cells [15]. In a first round of analyses, sequence identity of tryptic peptides was established using SEQUEST for database correlation of tandem mass spectra and PEAKS for de novo amino acid sequencing in conjunction with the GenomicPeptideFinder (GPF) algorithm [27]. The concerted action of SEQUEST and PEAKS/GPF allowed identification of 2622 distinct peptides, including 98 intron-split peptides, and resulted in the recognition of 233 nonredundant proteins [27]. In a second round of analyses, Arg containing peptides were analyzed quantitatively. The presented quantitative data stem from four independent biological samples and numerous SDS-PAGE fractionations and comparative quantitative analyses. For the presentation of the data we decided to show, when possible, the combined mean and SD for each protein stemming from y-type ions of one or several peptides measured from the different independent biological samples (Tables 1–3 of the Supporting Information). 3.1 Lhcb proteins remain stable under iron-deficient conditions – in particular, specific Lhcb polypeptides increase under iron-deprivation We first projected to enumerate relative changes of the Lhcb polypeptides, which form the light-harvesting complex of PSII (LHCII), in response to iron deficiency. In a former study we demonstrated that Lhca polypeptides, light-harvesting proteins associated with PSI (LHCI), were affected to varying degrees under iron deficiency, in particular, the following responses were established: (i) down-regulation of subunits, (ii) N-terminal processing of Lhca3, and (iii) upregulation of Lhca4 and Lhca9 in respect to the amount of PSI [15]. In contrast to Lhca polypeptides, most Lhcb proteins remain stable and some even increase in abundance under iron deficiency (Fig. 1A). Similar results have also been obtained by immunoblot experiments using polyclonal anti-Lhcb antibodies in a previous study [9]. Moreover, the comparative quantitative data clearly show that Lhcbm1 and Lhcbm3, which represent subunits of the major light-harvesting complex, increase almost fourfold in response to iron deficiency, whereas subunits Lhcbm4/6 and Lhcbm5 remain unchanged. The same holds true for the core subunits of PSII, D2, and CP43, as well as the peripheral lumenal subunits PsbO and PsbP, which also remain stable. The D1 polypeptide however is slightly diminished to 65% under low iron in respect to iron-sufficient conditions. 3.2 Comparative quantitative proteomics reveals induction of stress response proteins and proteins of unknown function under iron deficiency The LC-MS/MS analyses of SDS-PAGE fractionated proteins from isotopically labeled and unlabeled thylakoids also identified numerous peptides and proteins that are not “house 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Proteomics 2007, 7, 0000–0000
keeping” components of the photosynthetic apparatus, but are involved in other chloroplast functions and/or regulation. For 25 of these thylakoid proteins, comparative quantitative data could be obtained. From the data shown in Fig. 1B, it becomes visible that proteins involved in redox regulation, oxidative defense, and stress response are up-regulated under iron deficiency. Particularly, 2-cys peroxiredoxin and a stress-induced light-harvesting protein LhcSR3 increase about fivefold and represent the strongest responses among all proteins analyzed. Gene expression studies and mircoarray analyses demonstrated that expression of LhcSR genes, a gene family consisting of three genes in Chlamydomonas, are also induced under high light stress, phosphorus and sulfur deficiency [28–30]. Our quantitative data show that besides LhcSR3, protein expression of LhcSR1 is also threefold elevated under iron deficiency. The function of LhcSR proteins is currently unknown; it is possible that they play a direct or indirect role in regulation of light energy dissipation. The role of peroxiredoxins is more defined. Peroxiredoxins are nonheme-containing peroxidases that detoxify H2O2 (hydrogen peroxide), alkyl hydroxides and peroxynitrite [31–34]. Peroxiredoxins play a role in the antioxidative protection system of bioenergetic pathways, in particular photosynthesis and respiration as well as in general stress response. In the light of these functions it is very likely that chloroplast 2cys-peroxiredoxin is involved in protection of the photosynthetic apparatus from oxidative stress during progressing iron deficiency. It is of note that a peptidyl-prolyl-cis-trans isomerase (PPI) of cyclophilin type (CYN37, C_90033) also increased about fourfold under iron-deficient conditions. In Arabidopsis, it was shown that a cyclophilin CYP20-3 interacts with chloroplast peroxiredoxins and that it has the potential to reduce disulfide-bonds of 2-cys-peroxiredoxins [35]. It is possible that the Chlamydomonas cyclophilin CYN37 has a similar function in respect to the chloroplast 2Cys peroxiredoxin, thereby helping to regenerate the peroxide-detoxifying activity of the nonheme-containing peroxidase under iron deficiency. Other stress response proteins that were up-regulated under iron deficiency are three metalloproteases, FTSH1, FTSH2, and a putative metalloprotease (C_770040). The latter protease was about three to fourfold, whereas the two FTSH proteases were about twofold induced. The Arabidopsis ortholog EGY1 of the Chlamydomonas gene product C_770040 was shown to be important for chloroplast development [36], since egy1 loss-offunction mutants had reduced granal thylakoids, a poorly developed lamellae network and a significantly lower amount of Lhca and Lhcb polypeptides. Among the four quantified enzymes of terpenoid and tetrapyrrole metabolism, protoporphyrinogen oxidase (C_330078), phytoene desaturase (C_490019), and 3,8-divinyl protochlorophyllide a 8-vinyl reductase (C_1330031) remained stable, whereas geranylgeranyl reductase (C_180047) is significantly decreased under iron deficiency. Another protein that is largely reduced under iron-deprivation is the retinal-binding protein chlamyopsin 2. In contrast, the Stt7 kinase, which is www.proteomics-journal.com
Proteomics 2007, 7, 0000–0000
Plant Proteomics
5
Figure 1. Comparative quantitation of thylakoid membrane proteins from ironsufficient and -deficient conditions. Quantitiation of proteins from the PSII/ LHCII complex (A) and selected thylakoid proteins not involved in primary photosynthesis (B). The y-axis represents the change in abundance of the respective proteins (log of the ratio of relative abundance Fe2/relative abundance Fe1) under iron deficiency. The dotted line visualizes a ratio of one, indicating no changes in protein expression. The solid line visualizes a ratio of two, which was regarded as the significance level for an increase in protein expression under iron-deficient conditions.
required for light-harvesting phosphorylation in the process of state transitions [37], remained unchanged under iron deficiency. Pyruvate formate lyase, a polypeptide involved in carbon metabolism and one of three pyruvate decarboxylating enzymes in Chlamydomonas [38], did not change in response to iron deficiency. Interestingly, among the identified proteins, several hypothetical proteins with unknown function could be quantified (Fig. 1B). Three of these proteins were induced more than twofold by iron deficiency and may represent important players in plastid iron-homeostasis. All three proteins were also identified to the thylakoid membrane in Arabidopsis [39, 40]. The gene product of gene model C_1390010 is about two- to threefold induced under iron 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
deficiency. This protein possesses an ATP-binding domain, a PDZ/DHR/GLGF domain, and carbamoyl-phosphate synthase domain. The second protein of unknown function that increases about threefold in abundance is the gene product of gene model C_420064. It is of particular interest since it possesses a zinc/iron binding motif and may be involved in plastid iron-trafficking. The third gene product that is induced about fourfold under iron-deprivation is a putative calcium sensor. The Arabidopsis ortholog, which was named CAS, has been described as cell surface receptor that mediates extracellular Ca21 sensing in guard cells [41]. However, as mentioned above, this protein has also been identified in Arabidopsis thylakoid membranes [39]. Proteomic data also www.proteomics-journal.com
6
B. Naumann et al.
indicated a thylakoid location in Chlamydomonas [27]. Moreover, the protein was identified in enriched eyespot preparations from Chlamydomonas [42]. In contrast to the quantitation of thylakoid proteins presented in Fig. 1, numerous comparative quantifications of other proteins were not successful. For example, quantitation of a recently identified novel light harvesting protein LhcbP1, which shares homology with Lhca and Lhcbb polypeptides, (C_20371) [27] was not effective. The LhcbP1 gene product has a molecular weight of about 38.2 kDa. The gene product possesses a putative chloroplast transit peptide, where the cleavage site is unknown, but it is known that such a light-harvesting protein transit peptide can consist of only a few amino acids in Chlamydomonas [25, 43, 44]. The protein sequence is conserved from Ostreococcus tauri to A. thaliana [45, 46]. In the proteomics approach, the SDS-PAGE fractionated protein was identified with an apparent molecular mass of about 40 kDa [27]. Immunoblot analyses using polyclonal LHCII antibodies recognized a protein at this molecular mass, normally absent from iron-sufficient cells, which is induced after 5 days of iron deficiency (Fig. 2A) as well as after growth in iron-deficient medium (1 or 0.1 mM) (Fig. 2C). Quantitative RT-PCR analyses verified that lhcbp1 mRNA expression is indeed up-regulated under iron-deprivation (Fig. 2B). Therefore, we propose that the low ironinducible 40 kDa protein, which was recognized by the polyclonal anti-LHCII antibodies, is LhcbP1. Thus, LhcbP1 is another thylakoid membrane protein that is induced under iron-deprivation. To verify and further investigate proteins that were recognized by the comparative proteomics approach to be inducible under low iron, we selected LhcSR3 and the puta-
Figure 2. LhcbP1 is induced under iron deficiency. (A) Western blot analysis of thylakoid membranes extracted from Fe1 and Fe2 cells, loaded on equal protein (10 mg), using antibodies against LhcbP1. (B) Quantitative RT-PCR of isolated cDNA from Fe1 and Fe2 cells, b-tubulin was used as a loading control. (C) Western blot analysis of whole cells grown in different iron concentrations, loaded on equal chlorophyll (10 mg), using antibodies against LhcbP1 and PsaD.
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Proteomics 2007, 7, 0000–0000
Figure 3. Induction of the stress-related light harvesting protein LhcSR3 under iron-deficient as well as high light conditions. (A) Quantitiative RT-PCR of isolated cDNA from Fe1 and Fe2 cells, the signal of b-tubulin was used as a loading control. (B) Western blot analysis of thylakoid membranes extracted from Fe1 and Fe2 cells, loaded on equal protein (10 mg). Antibodies used were raised against LhcSR3 and Lhcb proteins, the latter was used as a loading control. (C) Elution profile, representing the relative abundance, of the y6-ion of the peptide EAILELDDIER from LhcSR3 measured in thylakoid membrane preparations from Fe1 and Fe2 cells. (D) Western blot analysis of whole cells grown for 24 and 48 h under various light conditions, loaded on equal chlorophyll (5 mg). Antibodies against LhcSR3 and PsaD were used.
tive iron/zinc binding protein, encoded by gene model C_420064, for additional analysis (Figs. 3 and 4). Quantitative RT-PCR analyses clearly demonstrate that lhcSR3 mRNA expression is up-regulated under iron-deprivation (Fig. 3A). Immunoblot analyses using anti-LhcSR3 antibodies show that protein levels of LhcSR3 are markedly more elevated under low iron than under iron-sufficient conditions in isolated thylakoid membranes (Fig. 3B). Equal loading was proven by using anti-Lhcb antibodies. The Western blot results verify the comparative proteomic experimental data (Fig. 1B). The proteomic data are further exemplified by www.proteomics-journal.com
Proteomics 2007, 7, 0000–0000
Plant Proteomics
7
48 h, fractionated by SDS-PAGE and analyzed by immunoblotting using anti-LhcSR3 and anti-PsaD antibodies (Fig. 3D). The immunoblot analyses show that, whereas PsaD levels remain constant, LhcSR3 is clearly induced by high light. Thus, confirming that LhcSR3 protein expression is responsive to stresses other than iron. To analyze the induction of gene product C_420064 by iron-deprivation, isolated thylakoid membranes from iron-sufficient and -deficient cells were separated by SDS-PAGE and analyzed by immunoblotting using anti-C_420064 gene product antibodies (Fig. 4A). The immunoblot data confirmed the induction of the iron/zinc protein under iron deficiency as already demonstrated by the comparative quantitative data (Fig. 1B). The proteomics data are again further exemplified by quantitative comparison of y6 fragment ions of labeled and unlabeled C_420064 gene product sister peptides (Fig. 4B), where the unlabeled y-type ion is about threefold more abundant than the unlabeled ion. The use of antiLhca3 and anti-PsaD antibodies as protein marker (Fig. 4A) proves the iron-sufficient and -deficient status of the thylakoids. The immunoblot was also decorated with anti-Lhcbm6 antibodies to validate equal protein loading. In contrast to LhcSR3, the iron/zinc binding protein is not induced under high light stress as evidenced by immunoblot analyses of thylakoid membranes isolated from cells that were shifted to three different light conditions, 2?5, 50, and 250 mE?m22 ?s21 for 2 days (Fig. 4C, see also Fig. 3D). Expression of the iron/zinc binding protein is apparent under all three light conditions, implicating that this protein of unknown function is a protein response factor more specific to iron deficiency.
Figure 4. Induction of the putative iron/zinc binding protein (C_420064) under iron-deficient conditions. (A) Western blot analysis of thylakoid membranes extracted from Fe1 and Fe2 cells, loaded on equal protein (10 mg) using Lhca3, PsaD, and C_420064 antibodies as well as Lhcbm6 antibodies as a loading control. (B) Elution profile, representing the relative abundance, of the y6-ion of the peptide MYADAEPDYLR from C_420064 measured in thylakoid membrane preparations from Fe1 and Fe2 cells. (C) Western blot analysis of whole cells grown in for 24 and 48 h under various light conditions, loaded on equal chlorophyll (5 mg). Antibodies raised against C_420064 and PsaD were used.
quantitative comparison of y6 fragment ions of labeled and unlabeled LhcSR3 sister peptides (Fig. 3C), where the unlabeled y-type fragment ion is about tenfold more abundant than the labeled one. As suggested by literature [28–30, 47] LhcSR3 is a more general stress response factor. Therefore, we aimed to validate this further and investigated the induction of LhcSR3 by light-stress as an additional stressor. Chlamydomonas cells were grown under moderate light intensity of 60 mE?m22 ?s21 and shifted at defined cell density to three different light conditions, 2?5, 50, and 250 mE?m22 ?s21 for 2 days. Thylakoid membranes were isolated after 24 and 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3.3 Iron-deficient cells are more prone to photoinhibition than iron-sufficient cells The fact that several stress responsive proteins were induced and the LHCII antenna increased under iron-deprivation prompted us to investigate the photoinhibitory status of the iron-sufficient and -deficient cells using fluorescence videoimaging (Fig. 5). Three independent Chlamydomonas cell cultures for each condition were analyzed on the basis of equal cell number. From the fluorescence video images of the examined cell cultures it is obvious that the fluorescence emission of iron-deficient cells is more pronounced than that of the iron-sufficient cells (Fig. 5A). Measurement of photosynthetic electron transfer rates (ETR), that were based on fluorescence analysis (ETR = 0.56yield60.846PAR mE?m22 ?s21) revealed clear differences between iron-sufficient and -deficient cells (Fig. 5B). The ETR of iron-sufficient cells started to saturate at a light intensity of about 181 mE?m22 ?s21 and yielded maximal values between 30 and 40 mmol electrons m22s21, whereas ETR of iron-deficient cells saturated already at 56 mE?m22 ?s21 and showed maximal values of about 5 mmol electrons m22s21. For the irondeficient ETR rate, the low amount of PSI has to be taken into account, so that the actual rate is probably up to twofold www.proteomics-journal.com
8
B. Naumann et al.
Proteomics 2007, 7, 0000–0000
Figure 5. Chlorophyll fluorescence properties of iron-sufficient and -deficient cells. Chlorophyll fluorescence from iron-sufficient and -deficient cells measured with a chlorophyll fluorometer on the basis of same cell number (approximately 2.56106 cells/mL). The arrow indicates an increase in chlorophyll fluorescence in the color-scale (A); Electron transport rate (B) and nonphotochemical quenching (C) were measured for increasing light intensities. Yield (D) was measured for cells prior to, during, and after incubation in very high light for 75 min.
higher, because the ratio between PSII and PSI increases about five- to tenfold (according to immunoblot experiments using PsaD antibodies, data not shown, in line with previous results [9, 15]). Furthermore, nonphotochemical quenching analysis, measured as NPQ (NPQ = (Fm 2 Fm0 )/Fm0 ) (Fig. 5C), demonstrated that maximal NPQ levels for irondeficient cells with values of about 0.025 were significantly lower in respect to iron-sufficient cells with maximal NPQ values between 0.06 and 0.1. It is of note that in iron-deficient cells, NPQ saturated similar to ETR already at a light intensity of about 56 mE?m22 ?s21, consequently indicating that the capacity of the cells to dissipate excess light energy via NPQ is already at its limit at this low light intensity. Thus suggesting that photoinhibition should be inherently pronounced in cells grown under iron-deprivation at such low light intensities. To validate this assumption, we measured the effective photochemical quantum yield of photosystem II, expressed as PSII yield (Y(II) = (Fm0 2 F)/Fm0 ), at increasing light intensities (Fig. 5D) in cells grown under iron-sufficient and -deficient conditions. One minute of light at an intensity of 21 mE?m22 ?s21 had already a great impact on the effective PSII quantum yield of iron-deficient cells. The yield dropped by 40% from 0.57 to 0.34. In iron-sufficient cells, the yield decreased at the same time by only 10% from 0.7 to 0.63. Further incubation of iron-sufficient and -deficient cells at a light intensity of 531 mE?m22 ?s21 for 75 min caused a strong decrease in the effective quantum yield of PSII. To recover from this photoinhibition, cells were 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
allowed to recuperate at a light intensity of 21 mE?m22 ?s21 for another 75 min. Most interestingly, iron-sufficient as well as iron-deficient cells recovered to values close to those measured at 21 mE?m22 ?s21 at the beginning of the experiment, indicating that although iron-deficient cells were more prone to photoinhibition, cells grown under both conditions possess a very similar capacity to recover from light stress. This leads directly to the question why the iron-deprived cells are more prone to photoinhibition? One explanation could be that the enlarged LHCII antenna is functionally coupled to PSII under iron-deficient conditions and thereby causing an increased excitation pressure as evidenced by the effective PSII quantum yield measurements. 3.4 The functional coupled PSII antenna is significantly enlarged under iron deficiency To measure the functional antenna size of PSII, the time resolved reduction of QA from iron-sufficient and -deficient cells was assessed from light-induced induction of the variable chlorophyll a fluorescence (Fv) in the presence of DCMU (Fig. 6). The rate of Fv induction is directly proportional to the antenna size and can be employed to relatively quantify the antenna sizes of PSII from different samples. The half-lives of Fv induction of iron-sufficient and -deficient cells were measured to be 0.093 and 0.042 s, respectively, demonstrating that the functional PSII antenna size of irondeficient cells is more than twofold larger than the PSII www.proteomics-journal.com
Plant Proteomics
Proteomics 2007, 7, 0000–0000
9
in the dark is about 60% of the oxygen consumption rate of iron-deficient cells. Again taking the chlorophyll content of cells from both conditions into account, it appears that the oxygen consumption rates per cell are rather similar. Therefore, indicating that under low iron conditions respiration is not as compromised as photosynthesis. This is rather astonishing since the respiratory machinery is also a large sink for iron. 3.5 Comparative quantitative proteomics of mitochondrial proteins reveals that respiratory electron transfer protein complexes are not compromised by iron deficiency Figure 6. The functional PSII antenna size increases under irondeprivation. Chlorophyll a fluorescence induction kinetic of whole cells grown in either iron-sufficient or iron-deficient conditions. T1/2 indicating the half-life of QA reduction.
antenna of iron-sufficient cells. To assess whether the larger antenna size resulted in enhanced oxygen evolution rates, oxygen evolution measurements from iron-sufficient and-deficient cells were performed in the presence of p-benzoquinone (PBQ) as an artificial electron acceptor for PSII (Table 1). Interestingly, oxygen evolution rates, measured as mmol O2/ mg Chl h at light intensities of 500 and 1800 mE?m22 ?s21, appeared to be similar. Taking into account that the chlorophyll content of iron-deficient cells is about 54% of that of iron-sufficient cells (54% 6 10, from three independent samples) and that the amount of PSII is comparable (Fig. 1B), the oxygen evolution per cell is about half in irondeprived relative to iron-sufficient cells, which is in line with the moderate decrease in the D1 protein under iron deficiency. Therefore, the larger antenna size of iron-deficient cells does not lead to higher oxygen evolution rates as compared to iron-sufficient cells. It is obvious that the differences in oxygen evolution rates cannot explain the significantly smaller ETR found under iron deficiency as compared to that under iron-sufficient conditions (Fig. 5B). It is instead very likely, that the strong impact on photosynthetic electron transfer under low iron is caused by strong depletion of PSI and the cytochrome b6f complex [9, 15]. Interestingly, the rate of oxygen consumption of iron-sufficient cells Table 1. Rates of oxygen production or oxygen consumption in iron-sufficient or -deficient cells (mmol O2/mg Chl h) measured in the presence of p-benzoquinone (PBQ)
Respiration (O2 consumption)
Photosynthesis (O2 evolution)
Fe1
Fe1
Fe2
18.6 6 3.9 52.2 6 12.9
14.3 6 3 40.9 6 9.1
Dark 18 6 7.1 500 mE 1800 mE
Fe2 31.4 6 0.9
Mean value was calculated from three independent measurements from three independent biological samples.
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
To assess changes of the respiratory proteome under progressing iron deficiency, we utilized the SILAC approach as described for comparative quantitation of thylakoid proteins (see Fig. 1). For quantitative assessment of mitochondrial proteins, SDS-PAGE fractionated thylakoid membranes as well as whole cell extracts were used (Fig. 7A, closed and open symbols, respectively). It has been shown previously that respiratory mitochondrial proteins of the inner-membrane can be isolated together with thylakoids from Chlamydomonas [27, 48]. Using our quantitative approach, subunits of five respiratory complexes could be analyzed. Three subunits of complex III were quantified. Out of the three, two subunits, namely subunit 6 (C_1370004) and core I subunit, were also quantified from whole cells. All three proteins were at least twofold induced under iron deficiency. The same holds true for cytochrome c oxidase subunit 5b, which stands for complex IV. Alpha and beta subunits of mitochondrial ATP synthase were also found to be induced by about threefold. Here, the beta subunit was additionally quantified from whole cell extracts, supporting the thylakoid data. ATP synthase-associated protein ASAI increased to a lesser extent. Five subunits of complex I were quantified from thylakoids. The 75 kDa subunit (NDUFS1 subunit) of the complex I was also quantified from whole cell extracts. Comparative quantification of this subunit revealed that it decreased by about 20% under low iron in respect to iron-sufficient conditions. The 75 kDa subunit is part of the electron input module of complex I that transfers electrons from NADPH via FMH onto a chain of iron–sulfur clusters [49]. It binds three iron– sulfur clusters, two [4Fe–4S] and one [2Fe–2S] cluster. To generate more conclusive quantitative data for this complex I core subunit, mass spectrometric quantifications were also performed with an LTQ-Orbitrap hybrid mass spectrometer using whole cell extracts. For quantitative analysis of y-type ion pairs, sister peptide ions were fragmented either in the linear IT or the C-trap of the LTQ-Orbitrap mass spectrometer (Table 2, Fig. 7B). Comparative quantitative data from linear IT and C-trap fragmentation resulted in mean values and SDs of Fe- 12C6/Fe 1 13C6 ratios of 0.84 6 0.19 and 0.87 6 0.16, respectively. In comparison, comparative quantitation from 3-D ion-trap fragmentation resulted in a mean value and a corresponding SD of www.proteomics-journal.com
10
B. Naumann et al.
Proteomics 2007, 7, 0000–0000
Figure 7. Comparative quantitation of mitochondrial inner membrane proteins from iron-sufficient and -deficient conditions. (A) Quantitation of selected mitochondrial proteins. (B) Quantitation of the peptide GTESIDVSDGLGANIR of NADH ubiquinone oxidoreductase with the LTQ Orbitrap in two different settings: regular method used also with the LCQ (left panel) and HCD (right panel) (see Section 2). The dotted line in (A) visualizes a ratio of one, indicating no changes in protein expression. The solid line visualizes a ratio of two, which was regarded as the significance level for an increase in protein expression under iron deficient conditions.
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.proteomics-journal.com
Plant Proteomics
Proteomics 2007, 7, 0000–0000
11
Table 2. Comparative quantification of peptides stemming from complex I NADH ubiquinon oxidoreductase (NDUFS1 subunit) from iron-sufficient and -deficient SDS-PAGE fractionated whole cells extracts
SILAC_LTQ-linear IT
SILAC_C trap
SILAC_LCQ-3-D IT
Peptide
y-Ion ratio Fe2 C12/Fe1 C13
Peptide
y-Ion ratio Fe2 C12/Fe1 C13
Peptide
y-Ion ratio Fe2 C12/Fe1 C13
GTESI
0.731 0.834 1.060 0.701 0.782 0.669 0.709 0.910 0.939 0.828 0.741 0.741 0.899 0.711 0.709 0.576 1.282 0.534 1.176 0.925 1.093 0.979 0.842 0.191
GTESI
0.695 0.823 0.632 0.812 0.717 0.860 0.856 0.874 0.858 1.241 0.917 1.169 0.811 0.915
GTESI
0.611 0.367 0.716 0.218 0.192 0.751 0.926 0.768 1.685 1.633 1.074
FASEVA
Mean value SD
FASEVA
0.870 0.164
EGWN
0.813 0.501
Quantification was performed in the linear IT and C-trap of an LTQ-Orbitrap and a 3-D ion-trap of a LCQ DecaXP plus mass spectrometer. For fragmentation of sister peptides by MS/MS in linear and 3-D IT, precursor ions were selected in a way that the m/z ratio would lie exactly in-between the m/z ratio of the two sister peptides (for example, in the case of m/z ratios of 644.25 und 647.25 Da for doubly charged sister peptides, the selected m/z value for the precursor ion would be 645.75 Da). For fragmentation in the C-trap, the m/z value of isotope-labeled ion was taken as precursor ion.
0.813 6 0.5. All these mean values are in close agreement. The fact that the SDs for the linear IT and C-trap data were rather small provides additional evidence that the 75 kDa complex I subunit is indeed only slightly compromised under iron-deprivation. Comparative quantitative data for the 19 kDa and the B13 like complex I subunits showed the same tendency as seen for the 75 kDa subunit, while data for the 18 and 16 kDa subunits revealed that they were even slightly increased under iron deficiency (Fig. 7A). It is supposed that complex I contains eight or nine iron–sulfur clusters as redox prosthetic groups [49]. In this respect, it is remarkable that the complex remains rather stable under progressing iron deficiency. Several proteins involved in carbon metabolism were quantified. Here, malate dehydrogenase was found to increase about threefold under iron deficiency. Interestingly, aconitase did also increase slightly in abundance 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
under low iron. This is of note since this polypeptide is another abundant mitochondrial [4Fe–4S] cluster protein. Serine hydroxymethyltransferase, an alanine aminotransferase and a putative glycine cleavage protein decreased as response to iron deficiency. Alanine-glyoxylate aminotransferase, another enzyme that is involved in amino acid metabolism, is even more compromised and is about tenfold diminished. A mitochondrial carbonic anhydrase, in contrast, remained rather unchanged. Interestingly, mitochondrial HSP70 did also not change in abundance, indicating that the mitochondria are not under severe stress under iron deficiency. In respect to innermembrane mitochondrial transporter proteins, analyzed by the comparative proteomic approach, only the oxoglutarate/malate carrier protein increased about twofold under low iron. Two proteins with unknown function were quantitated. The gene product of gene model C_330095, which possesses an LrgB-like domain, seems to be twofold www.proteomics-journal.com
12
B. Naumann et al.
Proteomics 2007, 7, 0000–0000
LHCII antenna size increases under iron-deprivation, which additionally renders the cells more susceptible towards photoinhibition. 3.6 The increase in functional PSII antenna size under low iron is reversibly withdrawn when cells are replenished with iron
Figure 8. Iron replenishment decreases the functional PSII antenna size and restores PSI-LHCI. Chlorophyll a fluorescence induction kinetic (A) and 77 K fluorescence emission spectrum (B) of whole cells grown under standard conditions or grown for 5 days under iron deficiency and subsequently replenished with iron. Inset in (B) shows a Western blot analysis using antibodies against Lhca3 and PsaD, the former one as an indicator for iron deficiency.
induced. The gene product of gene model C_80056, harboring a peptidoglycan-binding domain (LysM), did not change in response to iron deficiency. In addition to mitochondrial proteins, a protein that is implicated in the glyoxylate cycle was quantitated (Fig. 7A). Malate synthase decreased twofold under low iron. The enzyme catalyses an aldol condensation reaction that produces malate from acetate and glyoxylate. Regulation of malate synthase from Chlamydomonas has been investigated [50]. Under photoheterotrophic conditions and presence of acetate the malate synthase is largely expressed but strongly down-regulated when light is switched off [50]. It is proposed that down-regulation of the malate synthase might reflect a control mediated by the energy status of the cell. In line with this, the down-regulation of malate synthase under iron deficiency might reflect a switch from photoheterotrophic growth to heterotrophic growth. The fact that under iron deficiency photosynthetic electron transfer is strongly compromised, whereas respiration it not, points also into this direction. In this respect it is remarkable that the functional 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
To assess whether the functional increase in PSII antenna size under progressing iron deficiency is reversible, chlorophyll a fluorescence induction kinetics were measured from cells that were replenished with iron after 1 and 2 days. Figure 8A shows the Fv transients in iron-replenished cells (open circles, after 1 day and gray circles, after 2 days of iron replenishment) in comparison with transients measured from iron-sufficient (black symbols) and iron-deficient cells (gray triangles). The latter two transient were taken from Fig. 6. The halftime of the Fv increase measured with cells that were replenished with iron for 1 day is nearly identical with that of iron-deficient cells. However, after 2 days of ironreplenishment speed of Fv induction is significantly slowed down and more comparable to the rate seen with iron-sufficient cells. Thus, indicating that the functional PSII antenna size decreased. In parallel we assessed status of PSI-LHCI by Western blot analysis (see inset in Fig. 8B) and low temperature fluorescence spectroscopy (Fig. 8B). The immunoblot analyses using anti-PsaD and anti-Lhca3 antibodies show that Lhca3 is absent from iron-deficient cells, while PsaD is strongly diminished in correspondence to published data [9]. Concomitant, the low temperature fluorescence spectrum with a maximum emission at 705 nm, reveals that PSI and LHCI are functionally uncoupled under low iron as described [9]. In the course of iron-replenishment, the immunoblot data demonstrate that PsaD levels are already largely restored after 1 day whereas Lhca3 levels further increase after 2 days of iron addition. Comparison of low temperature fluorescence spectra from the iron-replenished cells with the spectrum derived from iron-deficient conditions reveals that after 1 day the amplitude of fluorescence decreased though that the maximal fluorescence emission is still at 705 nm and that after 2 days, the maximum fluorescence emission peaks at 714 nm, although with a rather small amplitude. Under ironsufficient condition, low temperature fluorescence spectra of PSI-LHCI omit an emission maximum at about 715 nm [51]. Therefore, indicating that after 2 days of iron-replenishment, PSI-LHCI is on its way to become fully restored. In summary, these data suggest that after iron-replenishment, the decrease in PSII-LHCII antenna size goes hand in hand with restoration of a functional PSI-LHCI complex.
4
Discussion
Comparative proteomics as well as functional measurements were employed to investigate the impact of iron deficiency on bioenergetic pathways in C. reinhardtii. Our results revealed www.proteomics-journal.com
Proteomics 2007, 7, 0000–0000
a number of dynamic responses including remodeling of light-harvesting antenna. This included the induction of stress-related polypeptides, of a putative iron/zinc binding protein and other proteins of unknown function. Concomitant with that was a change in the metabolic status of the cell, shifting from photoheterotrophic to heterotrophic growth as a consequence of iron-deprivation. Our data suggest the operation of an intricate cellular network that senses iron contents and adjusts gene expression to modify iron homeostasis and mediate acclimation of the bioenergetic machinery. Even though photosynthetic electron transfer is largely compromised under iron deficiency, our quantitative proteomic and spectroscopic data revealed that the functional antenna size of PSII significantly increased under iron deficiency. The enlargement of the functional PSII antenna could be responsible for a pronounced photoinhibition of PSII. The photoinhibition can be explained by the fact that electron transfer beyond PSII is compromised under low iron so that the absorbed light energy exceeds the capacity of photosynthetic energy consumption. In response to that, over-reduction of electron transport carriers occurs and gives rise to the accumulation of excitation energy in the light-harvesting antennae. This finally results in the generation of ROS and photo-oxidative damage [52–54]. Hereby, PSII is the primary target for photoinhibition resulting in the specific degradation of the PsbA (D1) reaction center protein [55–57]. Although PSII is functionally disabled via this process, the photoinactivation of PSII can also be regarded as an efficient protective mechanism under light stress, which circumvents photo-oxidative damage and controls linear photosynthetic electron transfer chain via PSII. It has been suggested that photoinactivated PSII centers may become strong quenchers of excitation energy under severe light stress, thereby protecting neighboring PSII centers from damage by acting as effective energy sinks [58–60]. In this line we suggest that under iron-deprivation, where PSI and the cytochrome b6/f are largely down-regulated, the enlargement of the PSII antenna causes photoinhibition at even low light intensities to minimize electron input via PSII and avoid over-reduction of the electron transfer chain to protect the system from photo-oxidative damage. Restoration of damaged PSII centers occurs via a PSII repair cycle [61, 62]. The rate of recovery from light stress at a low light intensity of 21 mE?m22 ?s21 was the same for iron-sufficient and -deficient cells, indicating that excess light is not deleterious to low iron cells, as shown before [9]. However, the levels of the effective photochemical quantum yield of PSII (Fig. 5D) displayed a significant difference after recovery, indicating that the restoration of active PSII centers is already counterbalanced by photoinhibition at 21 mE?m22 ?s21 under low iron. It is of note that an increased activity of the PSII repair cycle under irondeprivation is signified by an increase in the FtsH proteases (Fig. 1B), which are competent in degradation of damaged D1 protein [63]. 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Plant Proteomics
13
The comparative proteomics data revealed that stressrelated chloroplast polypeptides, like 2-cys peroxiredoxin, a PPI (C_90033), three metalloproteases and a stress-inducible light-harvesting protein, LhcSR3, were induced under irondeprivation. It is very likely that these enzymes represent another important defense line to protect the cells from photo-oxidative damage. In addition, remodeling of LHCI has been shown to prevent photo-oxidative stress under low iron in C. reinhardtii [9]. Interestingly, proteomic profiling of isolated thlyakoids from Beta vulgaris leaves revealed a decrease of photosynthetic electron transfer complexes, including proteins of LHCII and LHCI, as a response to iron deficiency [64]. These differences between Chlamydomonas cells grown in the presence of acetate and light and B. vulgaris grown in hydroponics in response to iron-deprivation might depend on the distinct metabolic status of the two organisms due to photoheterotrophic and -autotrophic growth, respectively. Interestingly, the comparative proteomics data as well as respiratory oxygen consumption measurements revealed that the respiration machinery remains rather stable under low iron. Complex III, although containing five molecules of iron, even increased threefold as a response to iron deficiency. In the same line aconitase increased slightly under low iron. In general, the accumulation of cytochromes and abundant iron–sulfur proteins is decreased in iron-deficient cells relative to iron-replete cells from bacteria to mammals (see above). Regulation of iron utilization in nutritional iron homeostasis is exemplified by the Fur repressor, a global regulator of iron homeostasis in Escherichia coli, which determines the abundance of iron-rich respiratory components in response to iron availability [65, 66]. Recent work in the bacteria E. coli and in Pseudomonas aeruginosa pointed to additional regulatory mechanisms in iron-homeostasis [67], namely, the function of small regulatory RNAs in posttranscriptional repression of iron-using proteins. Interestingly, a corresponding mechanism that takes advantage of a protein rather than a small RNA was described in the budding yeast Saccharomyces cerevisiae under iron starvation [68]. In both cases, the repression of iron-utilizing proteins may allow allocation of iron to cellular functions more fundamental for survival during iron-deprivation. Our data suggest as well the operation of an intricate cellular network that senses iron contents and regulates gene expression to adjust iron homeostasis and mediates acclimation of the bioenergetic machinery. In Chlamydomonas, degradation of iron–sulfur proteins and cytochromes is well regulated and targets primarily PSI and the cytochrome b6/f complex under photoheterotrophic conditions, while iron-rich respiratory complexes are spared. We suggest that a hierarchy in the degradation of iron-containing protein complexes exists when iron becomes limiting, pointing to a differential and regulated system of iron allocation within and between organelles. Under low iron and photoheterotrophic conditions, a transition from photoheterotrophic to primarily heterotrophic metabolism occurs, which establishes www.proteomics-journal.com
14
B. Naumann et al.
a new metabolic status, which in turn lowers the need for iron in the cell, since photosynthesis is down-regulated. Additionally, such remodeling enables efficient restoration of the photosynthetic machinery when iron is replenished as seen by the correlation between PSI-LHCI re-establishment and diminishment of the PSII antenna. It is tempting to speculate that pigment molecules which are set free after degradation of PSI-LHCI might be stored in LHCII and reused for restoration of PSI-LHCI after iron-replenishment. Although the amount of some Lhcb polypeptides increased (Lhcbm1, Lhcbm3, and CP26), whereas others remained stable under iron-deprivation, the twofold diminishment of active PSII centers under low iron might also contribute to the enlargement of the functional PSII antenna size under these conditions. The enlargement of the functional PSII antenna otherwise causes photoinhibition under low iron even at low light intensities which is in line with a protection of the system from photo-oxidative damage. Our results indicate that a hierarchy for iron allocations within organelles of a single cell exists that is closely linked with the metabolic status of the cell. This points to a regulatory network that interconnects iron-availability with the energy-status of the organism. Such a regulated iron-distribution is also evident at the whole organism level, where under deficiency skeletal muscle respiration is more drastically compromised compared to liver [69, 70]. Therefore, from the whole organism level to a single cell, the regulation of iron utilization appears to be a complex multilayered process with regulatory networks functioning to control the operation of essential iron-requiring metabolic pathways.
Lada Nedbal and Zuzana Benedikty are thanked for preliminary measurements of fluorescence emission from iron-sufficient and -deficient Chlamydomonas cultures. M. H. acknowledges support from the Research Foundation of the University of Pennsylvania and the Deutsche Forschungsgemeinschaft (grant Hi 739/2-1). We are indebted to Marita Hermann for excellent technical assistance and Mia Terashima for critical reading of the manuscript.
Proteomics 2007, 7, 0000–0000 [4] Onishi, T., Asakura, T., Yonetani, T., Chance, B., Electron paramagnetic resonance studies at temperatures below 77 degrees K on iron-sulfur proteins of yeast and bovine heart submitochondrial particles. J. Biol. Chem. 1971, 246, 5960– 5964. [5] Pascal, N., Douce, R., Effect of iron deficiency on the respiration of sycamore (Acer pseudoplatanus L.) cells. Plant Physiol. 1993, 103, 1329–1338. [6] Cartier, L. J., Ohira, Y., Chen, M., Cuddihee, R. W., Holloszy, J. O., Perturbation of mitochondrial composition in muscle by iron deficiency. Implications regarding regulation of mitochondrial assembly. J. Biol. Chem. 1986, 261, 13827–13832. [7] Merchant, S., in: Hoober, J. K., Wise, R. R., (Eds.)The Structure and Function of Plastids, Springer, Berlin/Heidelberg 2007, pp. 199–218. [8] Marschner, H., Mineral Nutrition of Higher Plants, Academic Press, London 1995. [9] Moseley, J. L., Allinger, T., Herzog, S., Hoerth, P. et al., Adaptation to Fe-deficiency requires remodeling of the photosynthetic apparatus, EMBO J. 2002, 21, 6709–6720. [10] Henriques, R., Jasik, J., Klein, M., Martinoia, E. et al., Knockout of Arabidopsis metal transporter gene IRT1 results in iron deficiency accompanied by cell differentiation defects, Plant Mol. Biol. 2002, 50, 587–597. [11] Varotto, C., Maiwald, D., Pesaresi, P., Jahns, P. et al., The metal ion transporter IRT1 is necessary for iron homeostasis and efficient photosynthesis in Arabidopsis thaliana, Plant J. 2002, 31, 589–599. [12] Vert, G., Grotz, N., Dedaldechamp, F., Gaymard, F. et al., IRT1, an Arabidopsis transporter essential for iron uptake from the soil and for plant growth, Plant Cell 2002, 14, 1223– 1233. [13] Martin, J. H., Fitzwater, S. E., Iron deficiency limits phytoplankton growth in the north-east Pacific subarctic. Nature 1988, 331, 341–343. [14] Behrenfeld, M. J., Kolber, Z. S., Widespread iron limitation of phytoplankton in the South Pacific ocean. Science 1999, 283, 840–843. [15] Naumann, B., Stauber, E. J., Busch, A., Sommer, F., Hippler, M., N-terminal processing of Lhca3 Is a key step in remodeling of the photosystem I-light-harvesting complex under iron deficiency in Chlamydomonas reinhardtii. J. Biol. Chem. 2005, 280, 20431–20441. [16] Doan, J. M., Schoefs, B., Ruban, A. V., Etienne, A. L., Changes in the LHCI aggregation state during iron repletion in the unicellular red alga Rhodella violacea. FEBS Lett. 2003, 533, 59–62.
5
References
[1] Finney, L. A., O’Halloran, T. V., Transition metal speciation in the cell: Insights from the chemistry of metal ion receptors. Science 2003, 300, 931–936. [2] Napier, I., Ponka, P., Richardson, D. R., Iron trafficking in the mitochondrion: Novel pathways revealed by disease. Blood 2005, 105, 1867–1874. [3] Ackrell, B. A., Maguire, J. J., Dallman, P. R., Kearney, E. B., Effect of iron deficiency on succinate- and NADH-ubiquinone oxidoreductases in skeletal muscle mitochondria. J. Biol. Chem. 1984, 259, 10053–10059.
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[17] Desquilbet, T. E., Duval, J. C., Robert, B., Houmard, J., Thomas, J. C., In the unicellular red alga Rhodella violacea iron deficiency induces an accumulation of uncoupled LHC. Plant Cell Physiol 2003, 44, 1141–1151. [18] Guikema, J. A., Sherman, L. A., Influence of iron deprivation on the membrane composition of Anacystis nidulans. Plant Physiol. 1984, 74, 90–95. [19] Laudenbach, D. E., Straus, N. A., Characterization of a cyanobacterial iron stress-induced gene similar to psbC. J. Bacteriol. 1988, 170, 5018–5026. [20] Burnap, R. L., Troyan, T., Sherman, L. A., The highly abundant chlorophyll-protein complex of iron-deficient Syne-
www.proteomics-journal.com
Proteomics 2007, 7, 0000–0000 chococcus sp. PCC 7942 (CP430 ) is encoded by the isiA gene. Plant Physiol. 1993, 103, 893–902. [21] Boekema, E. J., Hifney, A., Yakushevska, A. E., Piotrowski, M. et al., A giant chlorophyll-protein complex induced by iron deficiency in cyanobacteria. Nature 2001, 412, 745–748. [22] Bibby, T. S., Nield, J., Barber, J., Iron deficiency induces the formation of an antenna ring around trimeric photosystem I in cyanobacteria. Nature 2001, 412, 743–745. [23] Varsano, T., Wolf, S. G., Pick, U., A chlorophyll a/b-binding protein homolog that is induced by iron deficiency is associated with enlarged photosystem I units in the eucaryotic alga Dunaliella salina. J. Biol. Chem. 2006, 281, 10305– 10315. [24] Strzepek, R. F., Harrison, P. J., Photosynthetic architecture differs in coastal and oceanic diatoms. Nature 2004, 431, 689–692. [25] Hippler, M., Klein, J., Fink, T., Allinger, T., Hoerth, P., Towards functional proteomics of membrane protein complexes: Analysis of thylakoid membranes from Chlamydomonas reinhardtii. Plant J. 2001, 28, 595–606.
Plant Proteomics
15
[37] Depege, N., Bellafiore, S., Rochaix, J. D., Role of chloroplast protein kinase Stt7 in LHCII phosphorylation and state transition in Chlamydomonas. Science 2003, 299, 1572–1575. [38] Atteia, A., van Lis, R., Gelius-Dietrich, G., Adrait, A. et al., Pyruvate formate-lyase and a novel route of eukaryotic ATP synthesis in Chlamydomonas mitochondria. J. Biol. Chem., 2006, 281, 9909–9918. [39] Friso, G., Giacomelli, L., Ytterberg, A. J., Peltier, J. B. et al., In-depth analysis of the thylakoid membrane proteome of Arabidopsis thaliana chloroplasts: New proteins, new functions, and a plastid proteome database. Plant Cell, 2004, 16, 478–499. [40] Peltier, J. B., Ytterberg, A. J., Sun, Q., van Wijk, K. J., New functions of the thylakoid membrane proteome of Arabidopsis thaliana revealed by a simple, fast, and versatile fractionation strategy. J. Biol. Chem. 2004, 279, 49367– 49383. [41] Han, S., Tang, R., Anderson, L. K., Woerner, T. E., Pei, Z. M., A cell surface receptor mediates extracellular Ca(21) sensing in guard cells. Nature 2003, 425, 196–200.
[26] Kirchhoff, H., Schottler, M. A., Maurer, J., Weis, E., Plastocyanin redox kinetics in spinach chloroplasts: Evidence for disequilibrium in the high potential chain. Biochim. Biophys. Acta 2004, 1659, 63–72.
[42] Schmidt, M., Gessner, G., Luff, M., Heiland, I. et al., Proteomic analysis of the eyespot of Chlamydomonas reinhardtii provides novel insights into its components and tactic movements. Plant Cell, 2006, 18, 1908–1930.
[27] Allmer, J., Naumann, B., Markert, C., Zhang, M., Hippler, M., Mass spectrometric genomic data mining: Novel insights into bioenergetic pathways in Chlamydomonas reinhardtii. Proteomics 2006, 6, 6207–6220.
[43] Turkina, M. V., Villarejo, A., Vener, A. V., The transit peptide of CP29 thylakoid protein in Chlamydomonas reinhardtii is not removed but undergoes acetylation and phosphorylation. FEBS Lett. 2004, 564, 104–108.
[28] Zhang, Z., Shrager, J., Jain, M., Chang, C. W. et al., Insights into the survival of Chlamydomonas reinhardtii during sulfur starvation based on microarray analysis of gene expression. Eukaryot. Cell 2004, 3, 1331–1348.
[44] Stauber, E. J., Fink, A., Markert, C., Kruse, O. et al., Proteomics of Chlamydomonas reinhardtii light-harvesting proteins. Eukaryot. Cell, 2003, 2, 978–994.
[29] Moseley, J. L., Chang, C. W., Grossman, A. R., Genomebased approaches to understanding phosphorus deprivation responses and PSR1 control in Chlamydomonas reinhardtii. Eukaryot. Cell 2006, 5, 26–44. [30] Im, C. S., Zhang, Z., Shrager, J., Chang, C. W., Grossman, A. R., Analysis of light and CO(2) regulation in Chlamydomonas reinhardtii using genome-wide approaches. Photosynth. Res. 2003, 75, 111–125. [31] Hofmann, B., Hecht, H. J., Flohe, L., Peroxiredoxins. Biol. Chem. 2002, 383, 347–364. [32] Dietz, K. J., Plant peroxiredoxins. Annu. Rev. Plant Biol. 2003, 54, 93–107. [33] Konig, J., Lotte, K., Plessow, R., Brockhinke, A. et al., Reaction mechanism of plant 2-Cys peroxiredoxin. Role of the C terminus and the quaternary structure. J. Biol. Chem. 2003, 278, 24409–24420. [34] Horling, F., Lamkemeyer, P., Konig, J., Finkemeier, I. et al., Divergent light-, ascorbate-, and oxidative stress-dependent regulation of expression of the peroxiredoxin gene family in Arabidopsis. Plant Physiol. 2003, 131, 317–325. [35] Laxa, M., Konig, J., Dietz, K. J., Kandlbinder, A., Role of the cysteine residues in Arabidopsis thaliana cyclophilin CYP203 in peptidyl-prolyl cis-trans isomerase and redox-related functions. Biochem. J. 2007, 401, 287–297. [36] Chen, G., Bi, Y. R., Li, N., EGY1 encodes a membrane-associated and ATP-independent metalloprotease that is required for chloroplast development. Plant J. 2005, 41, 364– 375.
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
[45] Six, C., Worden, A. Z., Rodriguez, F., Moreau, H., Partensky, F., New insights into the nature and phylogeny of prasinophyte antenna proteins: Ostreococcus tauri, a case study. Mol. Biol. Evol. 2005, 22, 2217–2230. [46] Elrad, D., Grossman, A. R., A genome’s-eye view of the lightharvesting polypeptides of Chlamydomonas reinhardtii. Curr. Genet. 2004, 45, 61–75. [47] Richard, C., Ouellet, H., Guertin, M., Characterization of the LI818 polypeptide from the green unicellular alga Chlamydomonas reinhardtii. Plant Mol. Biol. 2000, 42, 303–316. [48] Atteia, A., de Vitry, C., Pierre, Y., Popot, J. L., Identification of mitochondrial proteins in membrane preparations from Chlamydomonas reinhardtii. J. Biol. Chem. 1992, 267, 226– 234. [49] Brandt, U., Energy converting NADH:quinone oxidoreductase (complex I). Annu. Rev. Biochem. 2006, 75, 69–92. [50] Nogales, J., Guijo, M. I., Quesada, A., Merchan, F., Functional analysis and regulation of the malate synthase from Chlamydomonas reinhardtii. Planta 2004, 219, 325–331. [51] Wollman, F.-A., Bennoun, P., A new chlorophyll-protein complex related to Photosystem I in Biochim. Biophys. Acta.1982, 680, 352–360. [52] Mehler, A. H., Studies on reaction of illuminated chloroplasts. I. Mechanism of the reduction of oxygen and other Hill reagents. Arch. Biochem. Biophys. 1951, 33, 65–77. [53] Asada, K., Mechanisms for Scavening Reactive Molecules Generated in Chloroplasts Under Light Stress, BIOS Scientific Publishers, Oxford, UK 1994, pp. 129–142.
www.proteomics-journal.com
16
B. Naumann et al.
[54] Asada, K., Radical Production and Scavenging in the Chloroplast. In Advances in Photosynthesis. Photosynthesis and the Environment, Kluwer, Amsterdam 1996, pp. 123– 150. [55] Barber, J., Andersson, B., Too much of a good thing: Light can be bad for photosynthesis. Trends Biochem. Sci. 1992, 17, 61–66. [56] Niyogi, K. K., Photoprotection revisited: Genetic and molecular approaches. Annu. Rev. Plant Physiol. Plant Mol. Biol. 1999, 50, 333–359. [57] Niyogi, K. K., Safety valves for photosynthesis. Curr. Opin. Plant Biol. 2000, 3, 455–460. [58] Lee, H. Y., Hong, Y. N., Chow, W. S., Photoinactivation of photosystem II complexes and photoprotection by nonfunctional neighbours in Capsicum annuum L. leaves. Planta 2001, 212, 332–342. [59] Matsubara, S., Chow, W. S., Populations of photoinactivated photosystem II reaction centers characterized by chlorophyll a fluorescence lifetime in vivo. Proc. Natl. Acad. Sci. USA 2004, 101, 18234–18239. [60] Chow, W. S., Lee, H. Y., He, J., Hendrickson, L. et al., Photoinactivation of photosystem II in leaves. Photosynth. Res. 2005, 84, 35–41. [61] Baena-Gonzalez, E., Aro, E. M., Biogenesis, assembly and turnover of photosystem II units. Philos. Trans. R. Soc. B 2002, 357, 1451–1459. [62] Kanervo, E., Suorsa, M., Aro, E. M., Functional flexibility and acclimation of the thylakoid membrane. Photoch. Photobio. Sci. 2005, 4, 1072–1080.
2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Proteomics 2007, 7, 0000–0000 [63] Nixon, P. J., Barker, M., Boehm, M., de Vries, R., Komenda, J., FtsH-mediated repair of the photosystem II complex in response to light stress. J. Exp. Bot. 2005, 56, 357–363. [64] Andaluz, S., Lopez-Millan, A. F., De Las Rivas, J., Aro, E. M. et al., Proteomic profiles of thylakoid membranes and changes in response to iron deficiency. Photosynth. Res. 2006, 89, 141–155. [65] Abdul-Tehrani, H., Hudson, A. J., Chang, Y. S., Timms, A. R. et al., Ferritin mutants of Escherichia coli are iron deficient and growth impaired, and fur mutants are iron deficient. J. Bacteriol. 1999, 181, 1415–1428. [66] McHugh, J. P., Rodríguez-Quiñones, F., Abdul-Tehrani, H., Svistunenko, D. A. et al., Global iron-dependent gene regulation in Escherichia coli: A new mechanism for iron homeostasis. J. Biol. Chem. 2003, 13, 13. [67] Masse, E., Arguin, M., Ironing out the problem: New mechanisms of iron homeostasis. Trends Biochem. Sci. 2005, 30, 462–468. [68] Puig, S., Askeland, E., Thiele, D. J., Coordinated remodeling of cellular metabolism during iron deficiency through targeted mRNA degradation. Cell 2005, 120, 99–110. [69] Willis, W. T., Dallman, P. R., Impaired control of respiration in iron-deficient muscle mitochondria. Am. J. Physiol. 1989, 257, C1080–C1085. [70] Hagler, L., Askew, E. W., Neville, J. R., Mellick, P. W. et al., Influence of dietary iron deficiency on hemoglobin, myoglobin, their respective reductases, and skeletal muscle mitochondrial respiration. Am. J. Clin. Nutr. 1981, 34, 2169– 2177.
www.proteomics-journal.com
