The most C-terminal tri-glycine segment within the polyglycine stretch of the pea Toc75 transit peptide plays a critical role for targeting the protein to the chloroplast outer envelope membrane Amy J. Baldwin and Kentaro Inoue Department of Plant Sciences, College of Agricultural & Environmental Sciences, University of California, CA, USA Most proteins found in plastids are encoded in the nuclear genome, translated on cytosolic ribosomes with cleavable N-terminal transit peptides, and imported into the organelles post-translationally. The translocon at the outer envelope membrane of chloroplasts 75 (Toc75) is postulated to function as a general protein translocation channel [1–4], and was also shown to be involved in targeting of a signal-anchored outer envel- ope membrane protein [5]. Toc75 appears to be enco- ded by a single functional gene in Arabidopsis thaliana [6] and its disruption by a T-DNA insertion caused an embryo-lethal phenotype [7], indicating the essential role of Toc75 in the viability of plants. Unlike other proteins destined for the outer mem- branes of chloroplasts or mitochondria, which do not require cleavable targeting sequences [8–10], Toc75 is synthesized with an N-terminal transit peptide that consists of two domains (Fig. 1) [11]. The first part behaves as a typical stromal targeting sequence [12], and is removed by a stromal processing peptidase Keywords chloroplast protein translocation channel; polyglycine; protein targeting; transit peptide; tripeptide segment Correspondence K. Inoue, Department of Plant Sciences, College of Agricultural & Environmental Sciences, University of California, One Shields Avenue, Davis, CA 95616, USA Fax: +1 530 752 9659 Tel: +1 530 752 7931 E-mail: kinoue@ucdavis.edu (Received 18 January 2006, accepted 13 February 2006) doi:10.1111/j.1742-4658.2006.05175.x The protein translocation channel at the outer envelope membrane of chloroplasts (Toc75) is synthesized as a larger precursor with an N-terminal transit peptide. Within the transit peptide of the pea Toc75, a major por- tion of the 10 amino acid long stretch that contains nine glycine residues was shown to be necessary for directing the protein to the chloroplast outer membrane in vitro [Inoue K & Keegstra K (2003) Plant J 34, 661–669]. In order to get insights into the mechanism by which the polyglycine stretch mediates correct targeting, we divided it into three tri-glycine segments and examined the importance of each domain in targeting specificity in vitro. Replacement of the most C-terminal segment with alanine residues resulted in mistargeting the protein to the stroma, while exchange of either of the other two tri-glycine regions had no effect on correct targeting. Further- more, simultaneous replacement of the N-terminal and middle tri-glycine segments with alanine repeats did not cause mistargeting of the protein as much as those of the N- and C-terminal, or the middle and C-terminal seg- ments. These results indicate that the most C-terminal tri-glycine segment is important for correct targeting. Exchanging this portion with a repeat of leucine or glutamic acid also caused missorting of Toc75 to the stroma. By contrast, its replacement with repeats of asparagine, aspartic acid, serine, and proline did not largely affect correct targeting. These data suggest that relatively compact and nonhydrophobic side chains in this particular region play a crucial role in correct sorting of Toc75. Abbreviations mtHsp70, mitochondrial heat shock protein 70; Plsp1, plastidic type I signal peptidase 1; psToc75, Toc75 from Pisum sativum; SPP, stromal processing peptidase; Toc, translocon at the outer envelope membrane of chloroplasts. FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS 1547 (SPP) [13]. The second domain of the pea Toc75 tran- sit peptide consists of 96 amino acids [11]. It is neces- sary to target the protein to the outer envelope membrane [13], and is cleaved off by a plastidic type I signal peptidase (Plsp1) [14,15]. There are two conserved regions in the second part of the Toc75 transit peptide; one is a stretch of 20–27 amino acids that is located near the N-terminus and is rich in hydrophobic side chains. Another conserved region is located about 10 residues apart from the hydrophobic region towards the C-terminus of the transit peptide. It contains 17–22 glycine residues over a stretch of 24–29 amino acids [16]. The polyglycine stretch of the pea Toc75 transit peptide at residues 91–110 (Fig. 1) can further be divided by four consecutive serine resi- dues into two regions that contain nine and six glycine residues, respectively [11,16]. By deletion and substitu- tion mutagenesis followed by in vitro import assay, the first part of the glycine-rich stretch, but not the conserved hydrophobic domain or the second polygly- cine stretch, was found to be necessary for correctly targeting the pea Toc75 protein to the outer envelope membrane [16]. Two scenarios have been postulated for the potential mechanism by which the polyglycine stretch mediates targeting of Toc75 to the chloroplast outer envelope [16]. In the first scenario, this region interacts with one or more proteins either at the intermembrane space or inner membrane of the chloroplast envelope, which keeps Toc75 from traversing the inner membrane. In the second scenario, the glycine-rich region prevents the Toc75 precursor from associating with one or more proteins that facilitate the translocation of the prepro- tein across the inner envelope membrane. A glycine repeat, when attached to a preprotein, was shown to prevent the protein from binding to a mitochondrial heat shock protein 70 (mtHsp70) [17], which exists in the matrix and assists translocation of preproteins across the mitochondrial membranes [18]. Thus, in the second scenario, a mtHsp70-like protein may exist in the intermembrane space of the chloroplast envelope and play a key role. Nevertheless, a detailed mechan- ism by which the polyglycine stretch mediates envelope targeting of Toc75 remains unknown. In this report, we extended the analysis of the poly- glycine stretch of the Toc75 transit peptide in order to better understand the targeting mechanism of the chloroplast outer envelope membrane protein. We divi- ded this region into three tri-glycine segments and examined significance of each portion by in vitro import assay. Interestingly, only the most C-terminal tri-glycine was found to be important for correctly tar- geting Toc75 to the outer envelope membrane. Results The most C-terminal tri-glycine segment is important for correct targeting of Toc75 to the chloroplast outer envelope membrane in vitro Previously, we examined the importance of certain regions within the pea Toc75 transit peptide for correct targeting by import assay using chloroplasts isolated from pea seedlings [16]. After import of radiolabeled precursor proteins, chloroplasts were treated with tryp- sin, a protease that can penetrate the outer but not the inner envelope membrane [19,20]. The chloroplasts containing imported proteins were further divided by centrifugation into supernatant and pellet fractions, and distribution of the imported proteins was ana- lyzed. In the present study, we employed this assay system to investigate the importance of certain residues within the polyglycine stretch for targeting specificity. First, we aimed to test whether or not the entire polyglycine stretch within the residues 91–100 of the pea Toc75 transit peptide is required for correct target- ing of the protein. We divided this region into three tri-glycine segments at 91–93, 95–97, and 98–100, replaced each of them with a stretch of three alanine residues (Table 1, AGG, GAG, and GGA, respect- ively), and subjected these mutated proteins to in vitro import assay. Generally, two forms of Toc75, the intermediate and mature forms, were recovered after the import reaction (Figs 1–3). This is similar to previ- ous results [11–13,15,16], which may be due to the rela- tively low activity of the Plsp1 homolog that is responsible for full maturation of Toc75 in the pea Fig. 1. The biogenesis of pea Toc75. The precursor, intermediate, and mature forms of Toc75 are indicated as prToc75, iToc75, and mToc75 with the numbers of the N-terminal amino acid residues, respectively. The stromal targeting sequence and the polyglycine stretch are indicated as black and gray boxes, respectively. The amino acid sequence of the core polyglycine stretch at residues 91–110 is shown on top and the region where mutations were introduced is underlined. SPP, stromal processing peptidase; Plsp1, plastidic type I signal peptidase 1. The scale bar in the bottom is equivalent to 100 amino acid residues. Transit peptide of Toc75 A. J. Baldwin and K. Inoue 1548 FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS chloroplasts used for this assay [15]. Furthermore, the intermediate and mature forms of Toc75 in a given supernatant or pellet fraction showed a similar pattern of protease sensitivity (e.g., compare lanes 4 and 6, or 27 and 29 of Fig. 2A) as was shown before [11,13,16]. Thus, we did not discriminate these two forms during analyses in the present study. Proteins derived from AGG and GAG precursors were exclusively recovered in the membrane and were susceptible to trypsin (Fig. 2A, lanes 15–18, 21–24; Fig. 2B,C), indicating that they were targeted to the chloroplast outer envelope membrane in a way statisti- cally indistinguishable from the wildtype precursor (Fig. 2A, lanes 3–6; P > 0.05; Student’s t-test). By contrast, substitution of an alanine repeat for the most C-terminal tri-glycine segment (mutant GGA) resulted in proteins being targeted almost equally to the soluble and membrane fractions (Fig. 2A, lanes 27, 28; Fig. 2B). Most of the proteins recovered in the former fraction and about half of those in the latter fraction were resistant to trypsin (Fig. 2A, lanes 27–30; Fig. 2C). These data indicate that the Toc75 transit peptide with GGA mutation targeted the protein to multiple locations: the stroma, the internal mem- branes where trypsin cannot reach (i.e., either the inner envelope membrane or thylakoids), and the outer envelope membrane where trypsin can digest proteins. This targeting pattern was statistically indistinguish- able from that of another Toc75 mutant in which most of the glycine residues were replaced with alanine (Table 1, polyAla; Fig. 2A, lanes 7–12; P > 0.05; Student’s t-test). These data suggest that the most C-terminal tri-glycine segment within the polyglycine stretch is necessary for correct targeting of Toc75. Next, we wished to test whether the most C-ter- minal tri-glycine segment is sufficient for correct tar- geting of Toc75. To this end, we kept this segment intact and replaced the other six glycine residues in the polyglycine stretch with alanine residues (Table 1, AAG). We also prepared two additional mutated pro- teins as controls in which alanine residues were sub- stituted for all but either the first or the second tri-glycine segments (Table 1, GAA and AGA, respect- ively). When tested by in vitro import assay, proteins derived from GAA and AGA precursors were mistar- geted both to the stroma and to the membrane frac- tions almost evenly (Fig. 2A, lanes 33, 34, 39, 40; Fig. 2B). Proteins imported into the membrane derived from AGA precursor appeared to be slightly more susceptible to trypsin than those from GAA precursor: susceptibility of proteins in the pellet from AGA mutant was 67%, whereas that of proteins derived from GAA was about 50% (Fig. 2C). Over- all, however, we were not able to detect significant differences between the three mutants, GGA, GAA, and AGA, in their targeting patterns (distributions to the supernatant and pellets shown in Fig. 2B and sen- sitivity to trypsin presented in Fig. 2C; P > 0.05; Student’s t-test). AAG transit peptide also mistargeted Toc75 to the stroma. However, about 10 times more proteins were found in the membrane than those in the supernatant (Fig. 2A, lanes 45 and 46; Fig. 2B), and this pattern is distinct from that of the other two mutated Toc75 precursors, GAA and AGA (P<0.05; Student’s t -test). We considered the possibility that the difference between AAG and the other two mutants might be due to kinetics of the import; i.e., AAG precursor might be imported into the stroma more slowly than other mutants. In order to test this possibility, we monitored the distribution of imported proteins derived from wildtype, AGA, and AAG precursors into the supernatant and pellet fractions between 3 and 30 min of the reaction. If the above possibility were correct, we should see changes in distribution of proteins, especially those from AGA, during the time course. As shown in Fig. 3, the ratios of proteins recovered in the supernatant to those in the membrane fraction appeared to be consistent among different reaction times within a single precursor (e.g., for AGA, compare lanes 19 and 20, 23 and 24, 27 and 28, and 31 and 32, respectively). Furthermore, trypsin-sen- sitivity of imported proteins was also consistent over the time course (e.g., for AGA, compare lanes 19–34). These data may indicate that (a) AGA precursor was imported into the stroma so efficiently that we could not detect translocation intermediates trapped in the Table 1. Part of the pea Toc75 transit peptide and its derivatives used in this study. Name Sequence (residues 91–100) WT GGGAGGGGGG polyAla *SA*AAAAA* AGG AAA******* GAG ****AAA*** GGA *******AAA GAA ****AAAAAA AGA AAA****AAA AAG AAA*AAA*** GGD *******DDD GGE *******EEE GGL *******LLL GGN *******NNN GGS *******SSS GGP *******PPP A. J. Baldwin and K. Inoue Transit peptide of Toc75 FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS 1549 outer membrane; (b) AAG precursor was sorted to multiple pathways in a way distinct from that of AGA precursor, or (c) the effect of AAG mutation on cor- rect targeting was significantly less than that of AGA mutation. Taken together with the previous results, we conclude that the most C-terminal tri-glycine segment is more important than the preceding two tri-glycine segments, but not sufficient for correct targeting of Toc75 completely to the chloroplast outer envelope membrane. A specific single glycine residue in the tri-glycine stretch does not determine the targeting specificity Next, we tested the importance of individual glycine residues at positions 98–100 in wildtype precursor for correct targeting. The replacement of Gly98 with alan- ine has already been shown not to affect proper target- ing [16]. Similarly, individual substitutions of glycine residues at 99 and 100 to alanine residues in wildtype AB C Fig. 2. Effects of alanine substitutions for the three tri-glycine segments in the Toc75 transit peptide on targeting specificity. (A) After the import of radiolabeled translation products (tl), chloroplasts were analyzed directly (imp), or subsequently treated without (–) or with (+) tryp- sin, lysed hypotonically, fractionated into the supernatant (S) and pellet (P) fractions by centrifugation, and analyzed by SDS ⁄ PAGE followed by fluorography. Precursor (pr), intermediate (i) and mature (m) forms of Toc75 are indicated. (B) Distribution of imported proteins in the supernatant and pellet fractions. Values indicate total amount of the intermediate and mature proteins recovered in each fraction quantified using IMAGEJ version 1.34 (National Institutes of Health, USA, http://rsb.info.nih.gov/ij/) and shown as a percentage of the total amount of pre- cursor subjected to the import reaction. The mean values and standard deviations represented by error bars were calculated based on at least three independent experiments. (C) Resistance of imported proteins to trypsin. Values indicate the ratio of trypsin-resistant proteins (both the intermediate and mature proteins) to total proteins recovered in the supernatant or pellet fractions. The mean values and error bars were calculated based on at least three independent experiments. Transit peptide of Toc75 A. J. Baldwin and K. Inoue 1550 FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS precursor did not affect targeting specificity (data not shown). In order to test if any two glycine residues are sufficient for correct targeting of Toc75, we substituted alanine for each glycine residue at positions 98–100 in AAG transit peptide. Like the polyAla mutant, pro- teins derived from all three mutants were mistargeted mainly to the stroma, in a way distinct from the pat- tern observed with AAG precursor (data not shown). These data suggest that the targeting specificity does not depend on any single residue, but requires all three glycines at positions 98–100. The tri-glycine stretch can be replaced with repeats of several other nonglycine residues In order to gain further details about the importance of the tri-glycine segment at residues 98–100 of the Toc75 transit peptide, we replaced this portion with repeats of various amino acids, and examined the effects on protein targeting. Particularly, we aimed to test the following three hypotheses: (a) A protein with properties similar to a known molecular chaperon, namely mtHsp70, plays a key role in recognition of the Toc75 transit peptide. (b) Non-glycine residues that frequently occur within and near the glycine-rich regions of the Toc75 transit peptides can substitute for glycine residues. (c) Another helix breaker, proline [21], can substitute for glycine. The first hypothesis is based on the report that a polyglycine stretch keeps a preprotein from binding mtHsp70 in the mitochondrial matrix [17]. Another amino acid repeat that was shown to disrupt the interac- tion of a preprotein with mtHsp70 was that of glutamic acid [17]. Hydrophobic residues, such as leucine and alanine, were found to have a high affinity to a bacterial mtHsp70 homolog, DnaK [22]. We sought to test the first hypothesis based on these observations. We replaced the tri-glycine stretch with a repeat of glutamic acid or that of leucine (Table 1, GGE and GGL, respectively), and examined the effects of these substitu- tions on targeting. Proteins derived from GGL precur- sor were targeted both to the stroma and to the membrane (Fig. 4A, lanes 21–24; Fig. 4B,C) in a way similar to GGA mutant (Fig. 4A, lanes 9–12; Fig. 4B,C). Import of GGE mutant was less efficient than that of other proteins, as evidenced by the accumu- lation of the precursor form of Toc75 that was sensitive to thermolysin after the import (Fig. 4A, lanes 14 and 16; data not shown). The intermediate form derived from GGE mutant recovered both in the supernatant and pellet fractions was resistant to trypsin (Fig. 4A, lanes 15–18; Fig. 4B,C), indicating its localization to the stroma, and thylakoid or inner membranes. These data indicate that a repeat of glutamic acid cannot replace the tri-glycine segment in the envelope targeting sequence. Thus, mtHsp70-like protein may not be involved in the polyglycine-mediated envelope targeting. In order to test the second hypothesis, we analyzed the 24–29 amino acid long regions within and nearby the conserved polyglycine stretches of Toc75 transit peptides from six plant species. As shown in Table 2, three residues, asparagine, aspartic acid, and serine, occur relatively frequently in these regions. Together they account for 17% of residues within this stretch. We replaced the critical tri-glycine within the pea Toc75 transit peptide with repeats of these three Fig. 3. Time course of the distribution of imported proteins. Radiolabeled Toc75 precursors were incubated with intact chloroplasts at room temperature for the time indicated, then chloroplasts were reisolated and analyzed as described in the legend to Fig. 2A. A. J. Baldwin and K. Inoue Transit peptide of Toc75 FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS 1551 AB C Fig. 4. Effects of replacements of the most C-terminal tri-glycine segment within the polyglycine stretch of the Toc75 transit peptide with repeats of various amino acids on targeting specificity. (A) Radiolabeled Toc75 precursors were incubated with isolated chloroplasts and ana- lyzed as described in the legend to Fig. 2A. (B) Distribution of imported proteins in the supernatant and pellet fractions quantified as de- scribed in the legend to Fig. 2B. The mean values and standard deviations represented by error bars were calculated based on at least three independent experiments. (C) Resistance of imported proteins to trypsin quantified as described in the legend to Fig. 2C. The mean values and error bars were calculated based on at least three independent experiments. Table 2. Amino acid compositions in the polyglycine stretch of Toc75 transit peptides. Sequences are those reported previously [16]. Plant species Amino acid # Gly Asn Asp Ser Phe Trp Ala Thr His Tyr Pea 91–118 17 1 1 5 1 1 1 0 0 0 Arabidopsis 101–126 19 2 3 0 2 0 0 0 0 0 Soybean 84–112 21 2 2 0 1 1 0 1 1 0 Lotus 64–91 22 1 1 0 1 1 0 2 0 0 Potato 101–124 19 1 1 2 0 0 0 0 0 1 Tomato 101–124 19 1 1 3 0 0 0 0 0 0 Total 117 8 9 10 5 3 1 3 1 1 % 73.6 5.0 5.7 6.3 3.1 1.9 0.6 1.9 0.6 0.6 Transit peptide of Toc75 A. J. Baldwin and K. Inoue 1552 FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS residues. All three mutants, GGN, GGD, and GGS showed targeting patterns similar to AAG mutant: proteins were targeted mainly to the membrane frac- tion and were susceptible to trypsin, indicating that they were in the outer envelope (Fig. 4A, lanes 25–42; Fig. 4B,C). Small portions (1–2%) of proteins were also recovered in the soluble fraction (Fig. 4B). Inter- estingly, they showed different susceptibility to trypsin (Fig. 4C): those derived from GGD and GGS were resistant, indicating their location in the stroma, whereas GGN-derived proteins were degraded, imply- ing their location in the intermembrane space. Another interesting observation was that overall import effi- ciency of GGN mutant was 18%, which was signifi- cantly higher than those of other precursors (3–12%; Fig. 4B). Finally, we prepared a construct in which the glycine repeat was replaced with a repeat of proline (Table 1, GGP). The distribution of proteins derived from GGP precursor to the supernatant and the membrane frac- tions was similar to those of GGN, GGD, GGS, and AAG (Figs 2B and 4B). Interestingly, similar to the case of GGN mutant, a small but significant amount of proteins from GGP found in the supernatant were susceptible to trypsin (Fig. 4B,C), indicating that they did not traverse the inner envelope membrane. Taken together, tri-proline and tri-asparagine can substitute for tri-glycine in envelope targeting to an extent somewhat better than repeats of aspartic acid and serine residues. Discussion In this report, we aimed to get insights into the mech- anism by which the polyglycine stretch within the Toc75 transit peptide mediates targeting of the protein to the chloroplast outer envelope membrane. Through mutagenesis and protein import assay in vitro, we were able to show that the most C-terminal tri-glycine within the polyglycine stretch is important for correctly targeting the protein to the outer membrane. Interest- ingly, replacements of the tri-glycine with repeats of asparagine, aspartic acid, serine, and proline caused a lesser degree of mistargeting compared to those with alanine, leucine, and glutamic acid repeats. We have not been able to identify a possible structure conserved and specific between repeats of the former four amino acid residues and glycine. Nevertheless, they are relat- ively small and not hydrophobic compared to the three amino acids that could not replace glycine. Thus, it may be possible that the compact and hydrophilic properties of this region are important for envelope targeting. How would this region keep the protein from cros- sing the inner envelope membrane? Neither of the two scenarios postulated before [16] can be excluded at this point. In the first scenario, there may be a proteina- ceous component either in the chloroplast envelope intermembrane space or in the inner membrane that binds to the flexible, small, and hydrophilic pocket that corresponds to 98–100 of the pea Toc75 transit peptide, and holds the protein at the envelope mem- brane. In the second scenario, this pocket could pre- vent Toc75 from interacting with a proteinaceous component that directs the protein to the stroma. Potential candidates for this component include a sub- unit of Toc complex such as Toc12 [23], a Hsp70 homolog in the intermembrane space of the chloro- plast envelope [23–26] that may have a different fea- ture than mtHsp70, and components of the translocon at the inner envelope membrane of chloroplasts such as Tic22 [27,28]. Constructs generated in this study should be useful to address these hypotheses. Experimental procedures Preparation of plasmids containing cDNAs for mutated pea Toc75 Substitutions of amino acid residues in the pea Toc75 pre- cursor were made using a QuikChangeÒ Site-Directed Mut- agenesis Kit (Stratagene, Cedar Creek, TX, USA). Sets of primers with variations of a sequence corresponding to nuc- leotide numbers 277–314 of the pea Toc75 (psToc75) cod- ing sequence and the plasmid pET-psToc75 as a template [12] were used to generate constructs for mutants GGA, GGE, GGL, GGN, GGD, GGS, and GGP. cDNA sequences for GAA and AGA mutants were prepared using a plasmid encoding GGA as a template and primers that anneal to nucleotide numbers 277–314 of psToc75. Plas- mids carrying sequences encoding AGG and GAG were prepared using sets of primers corresponding to nucleotide numbers 256–293 and 268–299 of psToc75, respectively, and pET-psToc75 as a template. A plasmid encoding the AGG mutant and primers corresponding to 256–293 of psToc75 were used to prepare a construct encoding AAG. Identities of all the clones were confirmed by sequencing of the entire coding sequence. Protein import assay using isolated chloroplasts Chloroplasts were isolated from 10- to 14-day old soil- grown pea as described [29]. Radiolabeled precursor pro- teins were prepared from cDNA constructs using T N TÒ Coupled Reticulocyte Lysate System (Promega, Madison, WI, USA) with [ 35 S]Met and T7 RNA polymerase. Import and trypsin treatment were performed essentially as A. J. Baldwin and K. Inoue Transit peptide of Toc75 FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS 1553 described [16]. Briefly, the radiolabeled precursor proteins (10 lL) were incubated with chloroplasts containing 12.5 lg chlorophyll, import buffer (50 mm Hepes⁄ KOH, 330 mm sorbitol, pH 8.0) and 3 mm Mg-ATP in a total vol- ume of 50 lL in the light for 30 min at room temperature. After the reaction, intact chloroplasts were re-isolated by 40% (v ⁄ v) Percoll and washed once with the import buffer. The re-isolated chloroplasts were further resuspended into 100 lL of the import buffer with or without 1.25 lg trypsin (Sigma, St. Louis, MO, USA), incubated for 30 min on ice, and to which 100 lL of the import buffer containing 1.25 lg trypsin inhibitor (Sigma) was added. Chloroplasts were re-isolated by 40% (v ⁄ v) Percoll, washed with the import buffer, and lysed with 10 mm Hepes ⁄ KOH, pH 8.0, and 10 mm MgCl 2 . Soluble and membrane fractions were obtained after centrifugation at 16 000 g at 4 °C for 30 min. Proteins from the soluble fraction were precipitated with 80% (v ⁄ v) acetone. Both fractions were resuspended in the sample buffer and analyzed by SDS ⁄ PAGE followed by fluorography. Acknowledgements We thank Dr Daniel Potter for his critical reading of the manuscript, and also members of the Inoue labor- atory for their helpful discussions. The project was supported by the National Research Initiative of the USDA Cooperative State Research, Education and Extension Service, grant number 2003-02860 to K.I. References 1 Hinnah SC, Hill K, Wagner R, Schlicher T & Soll J (1997) Reconstitution of a chloroplast protein import channel. EMBO J 16, 7351–7360. 2 Hinnah SC, Wagner R, Sveshnikova N, Harrer R & Soll J (2002) The chloroplast protein import channel Toc75: pore properties and interaction with transit pep- tides. Biophys J 83, 899–911. 3 Jarvis P & Robinson C (2004) Mechanisms of protein import and routing in chloroplasts. Curr Biol 14, R1064–R1077. 4 Soll J & Schleiff E (2004) Protein import into chloro- plasts. Nat Rev Mol Cell Biol 5, 198–208. 5 Tu SL, Chen LJ, Smith MD, Su YS, Schnell DJ & Li HM (2004) Import pathways of chloroplast interior pro- teins and the outer-membrane protein OEP14 converge at Toc75. Plant Cell 16 , 2078–2088. 6 Jackson-Constan D & Keegstra K (2001) Arabidopsis genes encoding components of the chloroplastic protein import apparatus. Plant Physiol 125, 1567– 1576. 7 Baldwin A, Wardle A, Patel R, Dudley P, Park SK, Twell D, Inoue K & Jarvis P (2005) A molecular-genetic study of the Arabidopsis Toc75 gene family. Plant Physiol 138, 715–733. 8 Schleiff E & Klosgen RB (2001) Without a little help from ‘my’ friends: direct insertion of proteins into chloroplast membranes? Biochim Biophys Acta 1541, 22–33. 9 Shore GC, McBride HM, Millar DG, Steenaart NA & Nguyen M (1995) Import and insertion of proteins into the mitochondrial outer membrane. Eur J Biochem 227, 9–18. 10 Hofmann NR & Theg SM (2005) Chloroplast outer membrane protein targeting and insertion. Trends Plant Sci 10, 450–457. 11 Tranel PJ, Froehlich J, Goyal A & Keegstra K (1995) A component of the chloroplastic protein import appara- tus is targeted to the outer envelope membrane via a novel pathway. EMBO J 14, 2436–2446. 12 Inoue K, Demel R, de Kruijff B & Keegstra K (2001) The N-terminal portion of the preToc75 transit peptide interacts with membrane lipids and inhibits binding and import of precursor proteins into isolated chloroplasts. Eur J Biochem 268, 4036–4043. 13 Tranel PJ & Keegstra K (1996) A novel, bipartite tran- sit peptide targets OEP75 to the outer membrane of the chloroplastic envelope. Plant Cell 8, 2093–2104. 14 Inoue K, Potter D, Shipman RL, Perea JV & Theg SM (2005) Involvement of a type I signal peptidase in bio- genesis of chloroplasts – Towards identification of the enzyme for maturation of the chloroplast protein trans- location channel. In Photosynthesis: Fundamental Aspects to Global Perspectives (van der Est A & Bruce D, eds), pp. 933–935. Allen Press, Lawrence, KS. 15 Inoue K, Baldwin AJ, Shipman RL, Matsui K, Theg SM & Ohme-Takagi M (2005) Complete maturation of the plastid protein translocation channel requires a type I signal peptidase. J Cell Biol 171, 425–430. 16 Inoue K & Keegstra K (2003) A polyglycine stretch is necessary for proper targeting of the protein transloca- tion channel precursor to the outer envelope membrane of chloroplasts. Plant J 34, 661–669. 17 Okamoto K, Brinker A, Paschen SA, Moarefi I, Hayer- Hartl M, Neupert W & Brunner M (2002) The protein import motor of mitochondria: a targeted molecular ratchet driving unfolding and translocation. EMBO J 21, 3659–3671. 18 Mokranjac D & Neupert W (2005) Protein import into mitochondria. Biochem Soc Trans 33, 1019–1023. 19 Cline K, Werner-Washburne M, Andrews J & Keegstra K (1984) Thermolysin is a suitable protease for probing the surface of intact pea chloroplasts. Plant Physiol 75, 675–678. 20 Jackson DT, Froehlich JE & Keegstra K (1998) The hydrophilic domain of Tic110, an inner envelope mem- brane component of the chloroplastic protein transloca- Transit peptide of Toc75 A. J. Baldwin and K. Inoue 1554 FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS tion apparatus, faces the stromal compartment. J Biol Chem 273, 16583–16588. 21 Gunasekaran K, Nagarajaram HA, Ramakrishnan C & Balaram P (1998) Stereochemical punctuation marks in protein structures: glycine and proline containing helix stop signals. J Mol Biol 275, 917–932. 22 Flynn GC, Pohl J, Flocco MT & Rothman JE (1991) Peptide-binding specificity of the molecular chaperone BiP. Nature 353 , 726–730. 23 Becker T, Hritz J, Vogel M, Caliebe A, Bukau B, Soll J & Schleiff E (2004) Toc12, a novel subunit of the inter- membrane space preprotein translocon of chloroplasts. Mol Biol Cell 15, 5130–5144. 24 Marshall JS, DeRocher AE, Keegstra K & Vierling E (1990) Identification of heat shock protein hsp70 homo- logues in chloroplasts. Proc Natl Acad Sci USA 87, 374–378. 25 Waegemann K & Soll J (1991) Characterization of the protein import apparatus in isolated outer envelopes of chloroplasts. Plant J 1, 149–158. 26 Schnell DJ, Kessler F & Blobel G (1994) Isolation of components of the chloroplast protein import machin- ery. Science 266, 1007–1012. 27 Kouranov A, Chen X, Fuks B & Schnell DJ (1998) Tic20 and Tic22 are new components of the protein import apparatus at the chloroplast inner envelope membrane. J Cell Biol 143, 991–1002. 28 Bedard J & Jarvis P (2005) Recognition and envelope translocation of chloroplast preproteins. J Exp Bot 56, 2287–2320. 29 Bruce BD, Perry S, Froehlich J & Keegstra K (1994) In vitro import of proteins into chloroplasts. Plant Mol Biol Manual J1, 1–15. A. J. Baldwin and K. Inoue Transit peptide of Toc75 FEBS Journal 273 (2006) 1547–1555 ª 2006 The Authors Journal compilation ª 2006 FEBS 1555 . derivatives used in this study. Name Sequence (residues 91–100) WT GGGAGGGGGG polyAla *SA*AAAAA* AGG AAA******* GAG ****AAA*** GGA *******AAA GAA ****AAAAAA AGA. The most C-terminal tri-glycine segment within the polyglycine stretch of the pea Toc75 transit peptide plays a critical role for targeting the protein