Portal de Periódicos da CAPES

Thylakoid targeting of Tat passenger proteins shows no DeltapH dependence in vivo

2003; Springer Nature; Volume: 22; Issue: 4 Linguagem: Inglês

10.1093/emboj/cdg081

1460-2075

Giovanni Finazzi, Claudia Chasen, Françis-André Wollman, Catherine de Vitry,

RNA regulation and disease

Article17 February 2003free access Thylakoid targeting of Tat passenger proteins shows no ΔpH dependence in vivo Giovanni Finazzi Giovanni Finazzi Istituto di Biofisica del CNR, Milan, Italy Search for more papers by this author Claudia Chasen Claudia Chasen Physiologie Membranaire et Moléculaire du Chloroplaste CNRS UPR1261, Institut de Biologie Physico-Chimique, 13 Rue Pierre et Marie Curie, 75005 Paris, France Search for more papers by this author Francis-André Wollman Corresponding Author Francis-André Wollman Physiologie Membranaire et Moléculaire du Chloroplaste CNRS UPR1261, Institut de Biologie Physico-Chimique, 13 Rue Pierre et Marie Curie, 75005 Paris, France Search for more papers by this author Catherine de Vitry Catherine de Vitry Physiologie Membranaire et Moléculaire du Chloroplaste CNRS UPR1261, Institut de Biologie Physico-Chimique, 13 Rue Pierre et Marie Curie, 75005 Paris, France Search for more papers by this author Giovanni Finazzi Giovanni Finazzi Istituto di Biofisica del CNR, Milan, Italy Search for more papers by this author Claudia Chasen Claudia Chasen Physiologie Membranaire et Moléculaire du Chloroplaste CNRS UPR1261, Institut de Biologie Physico-Chimique, 13 Rue Pierre et Marie Curie, 75005 Paris, France Search for more papers by this author Francis-André Wollman Corresponding Author Francis-André Wollman Physiologie Membranaire et Moléculaire du Chloroplaste CNRS UPR1261, Institut de Biologie Physico-Chimique, 13 Rue Pierre et Marie Curie, 75005 Paris, France Search for more papers by this author Catherine de Vitry Catherine de Vitry Physiologie Membranaire et Moléculaire du Chloroplaste CNRS UPR1261, Institut de Biologie Physico-Chimique, 13 Rue Pierre et Marie Curie, 75005 Paris, France Search for more papers by this author Author Information Giovanni Finazzi2, Claudia Chasen1, Francis-André Wollman 1 and Catherine de Vitry1 1Physiologie Membranaire et Moléculaire du Chloroplaste CNRS UPR1261, Institut de Biologie Physico-Chimique, 13 Rue Pierre et Marie Curie, 75005 Paris, France 2Istituto di Biofisica del CNR, Milan, Italy *Corresponding author. E-mail: [email protected] The EMBO Journal (2003)22:807-815https://doi.org/10.1093/emboj/cdg081 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The Tat pathway is a major route for protein export in prokaryotes and for protein targeting to thylakoids in chloroplasts. Based on in vitro studies, protein translocation through this pathway is thought to be strictly dependent on a transmembrane ΔpH. In this paper, we assess the ΔpH sensitivity of the Tat pathway in vivo. Using Chlamydomonas reinhardtii, we observed changes in the efficiency of thylakoid targeting in vivo by mutating the Tat signal of the Rieske protein. We then employed two endogenous pH probes located on the lumen side of the thylakoid membranes to estimate spectroscopically the ΔpH in vivo. Using experimental conditions in which the trans-thylakoid ΔpH was almost zero, we found no evidence for a ΔpH dependence of the Tat pathway in vivo. We confirmed this observation in higher plants using attached barley leaves. We conclude that the Tat pathway does not require a ΔpH under physiological conditions, but becomes ΔpH sensitive when probed in vitro/in organello because of the loss of some critical intracellular factors. Introduction Protein translocation across lipid bilayers plays a fundamental role in cellular metabolism as it allows the import of newly synthesized secretory proteins in the endoplasmic reticulum (ER), the export of proteins in the bacterial periplasmic space and the import of nuclear-encoded proteins in mitochondria and chloroplasts. In the case of organelles, nuclear-encoded respiratory or photosynthetic proteins have to cross up to three membranes in order to reach their final destination (reviewed in Wollman et al., 1999; Pfanner and Geisler, 2001). In plastids, which are separated from the cytosol by an outer and an inner envelope membrane, translocon complexes within the outer (TOC) and inner (TIC) membranes import precursor proteins into the stroma (reviewed in Reumann and Keegstra, 1999). Stromal proteins may then be targeted to thylakoid membranes through three distinct pathways that have bacterial homologues: the Sec, SRP and Tat pathways. A fourth pathway, considered as 'spontaneous', may be organelle specific (Michl et al., 1994). This diversity in thylakoid targeting pathways, each using a specific set of substrate proteins, was first demonstrated in vitro by competition experiments (e.g. Cline et al., 1993) and then confirmed in vivo with the isolation of plant mutants that were selectively impaired in the translocation of the SRP, Sec or Tat substrates (Voelker and Barkan, 1995; Settles et al., 1997; Amin et al., 1999; Motohashi et al., 2001). While the Sec and SRP pathways translocate polypeptides in their unfolded state, and therefore require the activity of soluble chaperones (reviewed in Mori and Cline, 2001; Robinson and Bolhuis, 2001), the Tat pathway has the unique ability to transport proteins in their folded state (Clark and Theg, 1997; Hynds et al., 1998; Santini et al., 2001; Thomas et al., 2001). Targeting of the Tat substrate proteins is specified by their N-terminus presequence, which is characterized by the presence of a twin arginine (RR) motif situated upstream of a hydrophobic stretch (Chaddock et al., 1995). The energetics of protein translocation also differ among the three pathways of bacterial origin. The Sec and SRP pathways require hydrolysis of nucleotide triphosphates, ATP in the case of Sec and GTP in the case of SRP (reviewed in Kouranov and Schnell, 1996), although a proton motive force may also play a role (Ernst et al., 1994; Yuan and Cline, 1994; Mant et al., 1995). In contrast, protein translocation via the Tat pathway only requires a ΔpH (Mould and Robinson, 1991; Cline et al., 1992). Unlike substrate specificity, which has been assessed both in vitro and in vivo, the energetics of protein translocation by the various pathways have been assessed only in vitro using intact thylakoids or in organello. In this paper we address the question of the ΔpH dependence of the Tat pathway in vivo using the unicellular green alga Chlamydomonas reinhardtii, which is best suited for molecular genetics and biochemical approaches, and offers the same experimental opportunity as Chlorella sorokiniana (Finazzi and Rappaport, 1998; Rappaport et al., 1999) for the study and control of the trans-thylakoid ΔpH in vivo. In particular, the well developed genetics of C.reinhardtii have produced a number of mutants lacking the chloroplast ATPsynthase (Lemaire et al., 1986), which should have distinct ΔpH properties when compared with the wild type. Here, we focused on three major Tat passenger proteins: the 16 and 23 kDa subunits of the oxygen-evolving complex of photosystem II (PSII) (Mould and Robinson, 1991; Cline et al., 1992, 1993), and the Rieske subunit of the cytochrome b6f (cyt b6f) complex (Madueño et al., 1994; Molik et al., 2001). We demonstrate that targeting of any of these three proteins to the thylakoid membranes is ΔpH insensitive in vivo, a conclusion that challenges the conclusions on the energetic requirements of the Tat pathway previously proposed on the basis of in vitro studies. Results The Tat signal is important for thylakoid targeting of the Rieske protein in intact cells of C.reinhardtii Figure 1A shows the barley and Chlamydomonas targeting sequences of the three Tat passenger proteins used in the present study. In contrast to its bacterial counterpart, which displays a regular (RR) Tat motif, the chloroplast Rieske protein exhibits an unusual KR signal. Nevertheless it is a bona fide Tat passenger protein, as demonstrated in vitro through competition experiments using vascular plant chloroplasts (Madueño et al., 1994; Molik et al., 2001). In order to assess the importance of this specific KR signature in the biogenesis and chloroplast targeting of the Rieske protein, we converted it to a canonical Tat motif or to a non-Tat motif (petC-RR and petC-KK, respectively, in Figure 1A). Figure 1.Consequences of Tat mutagenesis on chloroplast protein import in C.reinhardtii. (A) Tat motif of the OEE23, OEE16 and Rieske proteins (bold). The hydrophobic N-terminal α-helix is underlined and the cleavage site of the stroma peptidase is indicated by an asterisk. (B) Autoradiograph of membrane-associated protein synthesized during a 10 min in vivo pulse [14C]acetate labelling in the presence of 150 μg/ml chloramphenicol (inhibitor of chloroplast-encoded protein synthesis). KR, WT-KR wild-type strain with Rieske KR motif; RR, petC-RR Rieske mutant with RR motif; KK, petC-KK Rieske mutant with KK motif; BF25, mutant lacking membrane-associated OEE23 and OEE16; ΔpetC, petC-Δ1 Rieske deletion mutant used as the recipient strain for mutagenesis. The dots indicate predicted protein positions. Tat substrates are underlined. (C) Immunodetection of protein accumulation in whole cells (cells) and thylakoid membranes (membranes). (D) Light intensity dependence of the cyt f reduction rate. A decrease in the rate of electron transfer at saturating light intensity suggests the occurrence of multiple turnovers in b6f complexes and therefore a reduction in the accumulation of functional complexes. Download figure Download PowerPoint The efficiency of translocation of the Rieske protein by the Tat pathway was probed in pulse-labelling experiments in which we labelled whole cells of C.reinhardtii for 10 min with [14C]acetate, and then purified their thylakoid membranes and analysed the membrane protein content by SDS–PAGE and autoradiography. The labelled bands of the autoradiogram corresponded to those proteins that were translated and successfully targeted to the thylakoid membranes during the time of the pulse (Figure 1B). The three Tat passenger proteins OEE23, OEE16 and Rieske are readily identified on the autoradiogram shown in Figure 1B by their absence from the membranes of the BF25 and petC-Δ1 mutants: OEE23 and OEE16 are absent in the BF25 mutant (de Vitry et al., 1989) and the Rieske protein is absent in the petC-Δ1 mutant (de Vitry et al., 1999). We observed that the rate of labelling of the Rieske protein was similar in the RR and KR lanes, which means that replacement of the KR targeting signal by an RR motif did not modify the efficiency of membrane targeting of the Rieske protein. In marked contrast, the labelled band corresponding to the Rieske protein remained below detection in the KK lane, which indicated that a KK substitution caused a drastic fall in the efficiency of membrane targeting of the Rieske protein in vivo. We used specific antibodies to assess the steady-state level of accumulation of the Rieske protein in the thylakoid membranes of the three strains (Figure 1C). As expected from the fall in thylakoid targeting efficiency, we observed a lower accumulation of the mature Rieske protein in the KK strain than in the KR (wild type) and RR strains. This was observed in whole-cell extracts (ratio 0.39 ± 0.11, mean value of five experiments) as well as in thylakoid membrane fractions (ratio 0.31 ± 0.06, mean value of four experiments) with no significant accumulation of untranslocated Rieske protein in the chloroplast stroma (not shown). Translocation and accumulation of the other Tat substrates (OEE23 and OEE16) were similar in the three strains. The reduced accumulation of the Rieske protein had consequences for cyt b6f activity. Consistent with PSI being in excess relative to the b6f complex in this mutant strain, the rate of cyt f reduction measured after a single turnover flash decreased as the light intensity increased. This is due to the increased occurrence of multiple turnovers in the cyt b6f in order to re-reduce the increased fraction of oxidized plastocyanin generated by PSI at higher light intensity. The rate of cyt f reduction started to decrease above a light intensity equal to 30% of the saturating value (Figure 1D), suggesting that the 1:1 PSI–b6f stoichiometry was attained when one-third of PSI was excited. This is in good agreement with the finding that ∼30% of the Rieske protein is accumulated in the mutant (Figure 1C). Measurement of trans-thylakoid ΔpH in intact cells of C.reinhardtii Having ascertained that the targeting specificity of the Tat pathway in C.reinhardtii chloroplasts is similar to what has previously been described in higher plant chloroplasts and bacteria, we undertook a characterization of its energetic requirements in vivo. To do this, we took advantage of the fact that a Δμ∼H+ is established across thylakoid membranes in dark-adapted green algae (Bennoun, 1982). Its composition and amplitude can be conveniently studied by measuring the response of two endogenous lumen-exposed pH probes that can be easily monitored spectroscopically in dark-adapted algae. Using this approach, the properties of Δμ∼H+ have been extensively characterized in C.sorokiniana (Finazzi and Rappaport, 1998; Rappaport et al., 1999), leading to the conclusion that ΔpH was the predominant component of Δμ∼H+ owing to ATP hydrolysis by the membrane-bound CF0–CF1 protein complex. It is known that a Δμ∼H+ is also established in dark-adapted cells of C.reinhardtii (Finazzi et al., 1997), but its detailed characterization had not yet been carried out. This characterization is presented in Figures 2 and 3. First, we measured the pH dependence of PSII- and cyt b6f-related activities in wild-type cells (Figure 2). Upon illumination by a short flash, charge separation occurs in PSII photochemical centres between a donor (P680) and an acceptor (Qa) molecule. This yields a state P680+ Qa−, which may or may not recombine to its original state: the faster that P680+ is reduced by a secondary donor (Tyr YD1161−), the less charge recombination takes place within the reaction centre and the longer lived is the reduced form of the primary acceptor Qa−. Since chlorophyll fluorescence increases when Qa is in a reduced state (Duysens and Sweers, 1963), the efficiency of charge separation after single turnover flashes can easily be monitored by fluorescence measurements. Figure 2A shows such an experiment where the fluorescence changes were recorded during a series of single turnover saturating flashes in the presence of the artificial tertiary donor hydroxylamine (HA), which destroys the manganese cluster, and 3-(3′,4′-dichlorophenyl)-1,1-dimethylurea (DCMU), an inhibitor of Qa− re-oxidation by the secondary acceptor Qb (Bennoun, 1970). These conditions slow down the re-reduction step of P680+, thus enhancing its pH dependence (Lavergne and Rappaport, 1998). The series of flash-induced fluorescence increases are presented at three different luminal pH values (set with permeant buffers as described in Materials and methods). We observed that the efficiency of charge separation decreased with the acidification of the luminal pH. This can be quantified by calculation of the fraction R of variable fluorescence increase on the first flash with respect to the maximum variable fluorescence increase reached at the end of the series of flashes. This fraction is much lower at pH 5 than at pH 7, as previously reported (Rappaport et al., 1999). Figure 2.Changes in PSII and cyt b6f activity as a function of pH in wild-type cells of C.reinhardtii. (A) Changes in the PSII-modulated chlorophyll fluorescence yield induced by a series of saturating flashes in the presence of DCMU and HA as a function of pH. The time between flashes was 200 ms, and fluorescence was sampled 50 and 100 ms after each actinic flash. (B) The kinetics of cyt f redox changes as a function of pH. Algae were excited with non-saturating flashes (20% of maximum intensity) in order to avoid multiple turnovers of the b6f complex and the generation of a substantial light-induced ΔpH. pH equilibration was performed as described in Materials and methods. Download figure Download PowerPoint Figure 3.Absolute pH dependence of PSII and cyt b6f activity in whole cells of C.reinhardtii. The pH dependence of the PSII-modulated chlorophyll fluorescence yield (A and B) and the rate of cyt f reduction (C and D) were calculated from traces as in Figure 2. The fluorescence yield is indicated by the parameter R, which expresses the increase in variable fluorescence induced by the first flash of a series (F1 − F0) normalized to the maximum variable fluorescence (Fm − F0) measured in the presence of DCMU and HA. The size of the dark- generated ΔpH is calculated for wild-type (WT) (A and C) and ATPsynthase mutant FUD50 (B and D) cells as explained in the text. Broken line, control; solid line, nigericin 10 μM. Download figure Download PowerPoint Figure 2B shows the kinetics of cyt f re-reduction, measured after a single turnover flash at the same set of luminal pH values. Lowering the pH induced a marked decrease in the rate of cyt f re-reduction, as expected in the case of a reaction (the oxidation of the electron carrier plastoquinol) that couples electron transfer to proton release in the lumen (reviewed in Bendall, 1982). Thus, our experimental system gave us access to two independent reactions that are sensitive to the luminal pH. Their pH sensitivity can be exploited to establish two independent calibration curves that serve to monitor the luminal pH in vivo (Figure 3, squares). In order to estimate the amplitude of the trans-thylakoid ΔpH existing in dark-adapted algae, we first measured the PSII- and cyt b6f-related reactions in the absence of pH equilibration (no permeant buffer or ionophores added). When analysed with respect to each calibration curve, the two sets of data pointed to the same luminal pH (∼5.5) in the dark-adapted wild type (Figure 3A and C, broken lines). The same procedure was repeated in the presence of 1 μM nigericin (solid lines), which is an H+/K+ ion exchanger that specifically dissipates the trans-thylakoid pH gradient without suppressing the electrical component of Δμ∼H+ (Shavit and San Pietro, 1967). Therefore, nigericin equilibrates the luminal pH with that of the stroma, which has a large buffering capacity. In the absence of permeant buffers, no equilibration between the external medium and the cells was observed upon addition of the ionophore, i.e. no effect of the external pH was observed (data not shown), in agreement with previous reports for C.sorokiniana (Finazzi and Rappaport, 1998). The two calibration curves yield a pH of ∼7 in the presence of nigericin, which reflects the stromal pH in these experimental conditions; this value is in agreement with previous estimations for dark-adapted chloroplasts (Heldt et al., 1973) Thus, a ΔpH of ∼1.5 units builds up in darkness across the thylakoid membranes of dark-grown cells of wild-type C.reinhardtii. The generation of Δμ∼H+ in C.sorokiniana in darkness has been ascribed to ATP hydrolysis by the membrane-bound ATPsynthase (Rappaport et al., 1999). Therefore, we repeated the same set of PSII- and cyt b6f-sensitive pH measurements using the C.reinhardtii FUD50 mutant that bears a deletion of most of the gene encoding the β subunit and thus has no chloroplast ATPsynthase/ATPase activity (Lemaire et al., 1986). In the presence of nigericin, the mutant had the same characteristics as the wild type with a pH value of 7. However, the pH value remained the same in the absence of protonophore (Figure 3B and D). This experiment indicates that the FUD50 mutant is unable to develop a significant trans-thylakoid ΔpH, thus confirming that the proton source for the ΔpH formed in darkness is the chloroplast ATPsynthase. Role of trans-thylakoid ΔpH in targeting and accumulation of active PSII and cyt b6f complexes in C.reinhardtii The above study provided the opportunity to investigate the influence of the trans-thylakoid ΔpH on the functional assembly of the photosynthetic electron transfer chain. If it was strictly required for thylakoid targeting of those PSI, PSII and cyt b6f subunits that use the Tat pathway, we would expect the ATPsynthase mutant cells grown in darkness to display large pleiotropic changes in electron transfer reactions and in thylakoid protein accumulation, as in maize mutants defective in the Tat pathway (Voelker and Barkan, 1995). Thus, we compared the total number of active reaction centres, the ratio of PSII to PSI primary and the maximal rate of electron transfer through the b6f complex in the wild type and the FUD50 mutant grown in darkness (Figure 4A). We found no difference between the two, a conclusion further substantiated by the finding that similar amounts of CP47 and cyt f accumulated in their thylakoid membranes whereas, as expected, the major CF1 subunits were missing from the FUD50 mutant (Figure 4B). The three Tat proteins (the Rieske protein, OEE23 and OEE16) accumulated to the same extent in the two dark-grown strains (Figure 4C). In addition, their rate of targeting to the thylakoid membranes, measured as the rate of labelling of the corresponding polypeptides present in membranes purified from dark-grown cells pulse-labelled for 10 min with [14C]acetate, was identical for the two strains (Figure 4D). Thus, thylakoids from the wild type and from FUD50 showed identical protein patterns for the three Tat passenger proteins, an observation that rules out any prominent physiological role for ΔpH in targeting these proteins to their final membrane location. Figure 4.Consequences of ΔpH removal by the ATPsynthase mutation FUD50 on protein translocation in the thylakoids of C.reinhardtii. (A) Accumulation of functional PSII and b6f complexes in dark-grown wild-type (WT) and ATPsynthase mutant FUD50 strains. PSI and PSII activities were calculated as described in Materials and methods. Cyt b6f activity was estimated as the rate of cyt f reduction, as in Figure 2. All parameters were normalized to the cellular concentration. (B) Immunodetection of accumulation of α and β subunits of CF1, cyt f and chlorophyll binding protein CP47 of PSII. (C) Immunodetection of accumulation of the Rieske protein of b6f complex and of the oxygen-evolving complex OEE23 and OEE16 subunits of PSII. (D) Autoradiography of membrane-associated protein synthesized during a 10 min pulse in vivo [14C]acetate labelling in the presence of 150 μg/ml chloramphenicol. Tat substrate proteins are underlined, and Sec substrates are in italics. The dots indicate the presence of the protein. Download figure Download PowerPoint Effect of the H+/K+ exchanger nigericin on protein targeting to thylakoids in vivo The above results demonstrated that the absence of a ΔpH in the FUD50 mutant had no influence on the efficiency of protein targeting to thylakoids and on protein functional assembly. In order to perform an independent assessment of the role of the trans-thylakoid ΔpH in protein translocation through the Tat pathway in vivo, we looked for the effect of nigericin, which specifically removes the ΔpH component of Δμ∼H+. Wild-type cells grown in the light were incubated for 1 h in the dark and pulse-labelled for 10 min with [14C]acetate in the absence or presence of nigericin. Again, we found no evidence for a ΔpH-dependent change in the yield of membrane targeting of the Tat passenger proteins whether the effect of nigericin was assessed in the dark (Figure 5) or in the light (data not shown). Similar results were obtained for nigericin concentrations ranging from 1 to 10 μM, an indication that the ionophore concentration was saturating (data not shown). However, we noted that nigericin had a significant effect on the labelling of chloroplast-encoded products (Figure 5, dots), which are identified on the right-hand side of Figure 5. This effect of nigericin probably reflects the requirement of a ΔpH to activate the elongation step of translation, as suggested for the chloroplasts of vascular plants (Mühlbauer and Eichacker, 1998). Figure 5.Consequences of ΔpH removal by the H+/K+ exchanger nigericin on protein translocation in the thylakoids of C.reinhardtii wild-type cells. Autoradiography of membrane-associated protein synthesis during a 10 min in vivo pulse [14C]acetate labelling. Light-grown cells were labelled in the dark in the presence (+) or absence (−) of 10 μM nigericin. The dots indicate predicted protein positions. Addition of 8 μg/ml cycloheximide (an inhibitor of nuclear-encoded protein synthesis) during pulse-labelling is also indicated. Download figure Download PowerPoint Role of trans-thylakoid ΔpH on the targeting and accumulation of active PSII and cyt b6f complexes in barley leaves The finding that thylakoid targeting of Tat passenger proteins showed no ΔpH sensitivity in vivo in C.reinhardtii prompted us to perform a similar in vivo experiment with vascular plants. Attached leaves of barley were gently scratched and pulse-labelled for 30 min with droplets of [35S]Cys/Met with and without nigericin. We then compared the rate of labelling of a Sec substrate (OEE33) and a Tat substrate (OEE23), immunodetected in Figure 6A, which we recovered after isolation of thylakoid membranes and extraction with dithiothreitol–Na2CO3 (Figure 6B). Again, the pattern of relative labelling of the two proteins remained similar whether or not the leaves were treated with nigericin (Figure 6C). The ability of nigericin to dissipate the trans-thylakoid ΔpH in intact leaves was confirmed by the same experimental approach as used for C.reinhardtii (Figure 6D); unlike algae, plants fully deactivate the chloroplast ATPsynthase in darkness and therefore display no ΔpH in darkness (broken versus dotted lines). In our experimental conditions, a resting ΔpH of ∼1 unit that remained after illumination of the leaves for 5 min collapsed upon treatment with nigericin (compare solid and dotted lines). Figure 6.Consequences of ΔpH removal by the H+/K+ exchanger nigericin on protein translocation in barley leaves. (A) Immunodetection of OEE33 and OEE23 in thylakoid extracted from pulsed leaves (mb). (B) Coomassie Blue staining of proteins of the pellet (P) and supernatant (S) of chaotropic-treated membranes. (C) Autoradiography of membrane-associated protein synthesis during a 30 min pulse [35S]Met/Cys in vivo labelling in the presence or absence of 100 μM nigericin. Tat substrates are underlined and Sec substrates are in italics. (D) Thylakoid membranes uncoupled by nigericin in intact barley leaves. The fluorescence yield is indicated by the parameter R, as in Figure 3. Broken line, dark-adapted sample; solid line, illuminated sample; dotted line, illuminated sample plus nigericin 10 μM. Illumination was followed by a dark adaptation in order to allow re-oxidation of Qa. The calibration curve (squares) is the same when measured in thylakoids isolated from vascular plants (Lavergne and Rappaport, 1998). However, see its close resemblance to that obtained in vivo in C.reinhardtii (Figure 3A). Download figure Download PowerPoint Discussion A thylakoid-targeting Tat pathway in C.reinhardtii Translocation mutants of vascular plants showing severe phenotypic deficiencies have allowed the identification of three thylakoid membrane components of the Tat translocation machinery: cpTatC, Tha4 and Hcf106, which are similar to the Escherichia coli tatC, tatA and tatB gene products, respectively (reviewed in Mori and Cline, 2001; Robinson and Bolhuis, 2001). The C.reinhardtii EST database (http://www.biology.duke.edu/chlamy) contains homologues of cpTatC, Tha4 and Hcf106, which a chloroP algorithm (Emanuelsson et al., 1999) predicts to have chloroplast targeting sequences. Thus, it is reasonable to assume that C.reinhardtii chloroplasts have the same Tat machinery in charge of the same protein substrates as vascular plant chloroplasts. These substrates are identified by the presence of a twin arginine motif in their signal sequence (Teter and Klionsky, 1999; Mori and Cline, 2001; Robinson and Bolhuis, 2001). Among the major Tat substrates are two OEE subunits, OEE23 and OEE16, whose thylakoid targeting sequences are highly conserved between vascular plants and C.reinhardtii chloroplasts. The Tat signature of a third Tat substrate, the Rieske protein, underwent a substitution from RR in bacteria to KR in chloroplasts of vascular plants and algae, but has been shown experimentally by competition experiments to be Tat targeted to the thylakoids with OEE23 (Molik et al., 2001). In this work, we have tested the relationships between Tat targeting of the Rieske and its conversion to the holoform by altering the endogenous KR motif to a canonical RR motif. Neither thylakoid targeting nor biogenesis of an active Rieske was affected by the substitution. This behaviour excludes the possibility that the KR sequence would be required to slow down the passage of the Rieske protein through the Tat translocon in order to allow the binding of the Fe–S cluster before translocation (Molik et al., 2001). In contrast, conversion of the KR motif to a non-Tat KK motif had dramatic effects on thylakoid targeting, leading to a 70% decrease in the assembly of the Rieske in active cyt b6f complexes. This mutational approach of the Rieske targeting in vivo confirms the role of Tat signalling in the chloroplast of living cells of C.reinhardtii. Protein translocation by the Tat pathway in vivo Up to now, most studies of the energetics of protein translocation have been performed in vitro and in organello. These studies have shown conclusively that the Tat pathway does not requ