A nuclear-encoded protein of prokaryotic origin is essential for the stability of photosystem II in Arabidopsis thaliana
1998; Springer Nature; Volume: 17; Issue: 18 Linguagem: Inglês
10.1093/emboj/17.18.5286
ISSN1460-2075
AutoresJörg Meurer, Henning Plücken, Klaus V. Kowallik, Peter Westhoff,
Tópico(s)Mitochondrial Function and Pathology
ResumoArticle15 September 1998free access A nuclear-encoded protein of prokaryotic origin is essential for the stability of photosystem II in Arabidopsis thaliana Jörg Meurer Jörg Meurer Heinrich-Heine-Universität, Institut für Entwicklungs und Molekularbiologie der Pflanzen, Universitätsstraße 1, 40225 Düsseldorf, Germany Heinrich-Heine-Universität, Friedrich-Schiller-Universität Jena, Institut für Allgemeine Botanik, Lehrstuhl für Pflanzenphysiologie, Dornburger Straße 159, 07743 Jena, Germany Search for more papers by this author Henning Plücken Henning Plücken Heinrich-Heine-Universität, Institut für Entwicklungs und Molekularbiologie der Pflanzen, Universitätsstraße 1, 40225 Düsseldorf, Germany Search for more papers by this author Klaus V. Kowallik Klaus V. Kowallik Heinrich-Heine-Universität, Institut für Botanik, Universitätsstraße 1, 40225 Düsseldorf, Germany Search for more papers by this author Peter Westhoff Corresponding Author Peter Westhoff Heinrich-Heine-Universität, Institut für Entwicklungs und Molekularbiologie der Pflanzen, Universitätsstraße 1, 40225 Düsseldorf, Germany Search for more papers by this author Jörg Meurer Jörg Meurer Heinrich-Heine-Universität, Institut für Entwicklungs und Molekularbiologie der Pflanzen, Universitätsstraße 1, 40225 Düsseldorf, Germany Heinrich-Heine-Universität, Friedrich-Schiller-Universität Jena, Institut für Allgemeine Botanik, Lehrstuhl für Pflanzenphysiologie, Dornburger Straße 159, 07743 Jena, Germany Search for more papers by this author Henning Plücken Henning Plücken Heinrich-Heine-Universität, Institut für Entwicklungs und Molekularbiologie der Pflanzen, Universitätsstraße 1, 40225 Düsseldorf, Germany Search for more papers by this author Klaus V. Kowallik Klaus V. Kowallik Heinrich-Heine-Universität, Institut für Botanik, Universitätsstraße 1, 40225 Düsseldorf, Germany Search for more papers by this author Peter Westhoff Corresponding Author Peter Westhoff Heinrich-Heine-Universität, Institut für Entwicklungs und Molekularbiologie der Pflanzen, Universitätsstraße 1, 40225 Düsseldorf, Germany Search for more papers by this author Author Information Jörg Meurer1,2, Henning Plücken1, Klaus V. Kowallik3 and Peter Westhoff 1 1Heinrich-Heine-Universität, Institut für Entwicklungs und Molekularbiologie der Pflanzen, Universitätsstraße 1, 40225 Düsseldorf, Germany 2Heinrich-Heine-Universität, Friedrich-Schiller-Universität Jena, Institut für Allgemeine Botanik, Lehrstuhl für Pflanzenphysiologie, Dornburger Straße 159, 07743 Jena, Germany 3Heinrich-Heine-Universität, Institut für Botanik, Universitätsstraße 1, 40225 Düsseldorf, Germany *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:5286-5297https://doi.org/10.1093/emboj/17.18.5286 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info To understand the regulatory mechanisms underlying the biogenesis of photosystem II (PSII) we have characterized the nuclear mutant hcf136 of Arabidopsis thaliana and isolated the affected gene. The mutant is devoid of any photosystem II activity, and none of the nuclear- and plastome-encoded subunits of this photosystem accumulate to significant levels. Protein labelling studies in the presence of cycloheximide showed that the plastome-encoded PSII subunits are synthesized but are not stable. The HCF136 gene was isolated by virtue of its T-DNA tag, and its identity was confirmed by complementation of homozygous hcf136 seedlings. Immunoblot analysis of fractionated chloroplasts showed that the HCF136 protein is a lumenal protein, found only in stromal thylakoid lamellae. The HCF136 protein is produced already in dark-grown seedlings and its levels do not increase dramatically during light-induced greening. This accumulation profile confirms the mutational data by showing that the HCF136 protein must be present when PSII complexes are made. HCF136 homologues are found in the cyanobacterium Synechocystis species PCC6803 (slr2034) and the cyanelle genome of Cyanophora paradoxa (ORF333), but are lacking in the plastomes of chlorophytes and metaphytes as well as from those of rhodo- and chromophytes. We conclude that HCF136 encodes a stability and/or assembly factor of PSII which dates back to the cyanobacterial-like endosymbiont that led to the plastids of the present photosynthetic eukaryotes. Introduction Photosystem II (PSII) is one of the four major multi-subunit protein complexes of the thylakoid membrane of oxygenic photosynthetic organisms. The reaction centre core of this photosystem consists of the P680-binding subunits D1 (PsbA) and D2 (PsbD), cytochrome b559 (PsbE–PsbF) and the PsbI protein, and performs the primary charge separation. The minimal oxygen-evolving PSII complex in addition contains CP47 (PsbB) and CP43 (PsbC), two proteins of the inner chlorophyll a antenna, as well as the extrinsic, lumenal 34 kDa protein (PsbO). There are ∼15–20 additional proteins which have been identified as constituent subunits of PSII, but their functions have only been partly elucidated. PSII, like the other thylakoid membrane complexes, is a genetic mosaic and consists of nuclear- and plastome-encoded subunits (reviewed in Hankamer et al., 1997). The functioning of PSII as a light-triggered water-plastoquinone oxidoreductase is reasonably well understood due to numerous biochemical and biophysical studies (reviewed in Hankamer et al., 1997). A two-dimensional structure at 8 Å resolution is also available (Rhee et al., 1997) that may provide a starting point towards the elucidation of its three-dimensional structure. Compared with these detailed functional and structural informations, relatively little is known about the biogenesis of this membrane complex. Greening studies with etiolated seedlings of angiosperms demonstrated that the biosynthesis of a functional PSII depends on light (Westhoff et al., 1990). However, truncated PSII complexes which lack the chlorophyll a-binding subunits D1, D2, CP47 and CP43 can accumulate in dark-grown seedlings (Westhoff et al., 1990). This suggests that the assembly of PSII proceeds stepwise and involves the formation of intermediate subcomplexes (Van Wijk et al., 1996). The stoichiometry of PSII with respect to the other thylakoid membrane complexes is not constant but varies depending upon light conditions and species (reviewed in Melis, 1991). The fact that PSII biogenesis can be regulated separately from the other thylakoid membrane complexes is evident from its differential accumulation in mesophyll and bundle-sheath cells of NADP-malic enzyme type C4 plants (Schuster et al., 1985; Oswald et al., 1990). A selective regulation of the PSII activity is also observed when cells of nitrogen-fixing filamentous cyanobacteria differentiate into heterocysts (Wolk et al., 1997). The molecular and genetic basis of this capacity for differential PSII biogenesis in both eukaryotic and prokaryotic photosynthetic organisms is unclear. In general, posttranscriptional controlling mechanisms are thought to be very important for chloroplast biogenesis (reviewed in Rochaix, 1992; Mayfield et al., 1995) and can therefore be expected to play a dominant role in PSII biogenesis as well. Several cis- and trans-regulatory components have been identified which are involved in the processing and stability of plastid mRNAs (Rochaix, 1992; Mayfield et al., 1995). The translation of plastid mRNAs requires trans-regulatory factors which at least in part interact with the 5′ untranslated regions of the target mRNAs (Danon and Mayfield, 1991; Zerges et al., 1997). The activity of these trans-regulatory factors may be controlled by the redox state of the photosynthetic apparatus as has been shown for the psbA mRNA (Danon and Mayfield, 1994). There is a particular interest in understanding the final step in the biogenesis of thylakoid membrane complexes; that is, how the plastid- and nuclear-encoded subunits assemble to functional complexes. The correct folding of both types of plastidial proteins is mediated by molecular chaperones of the GroEL–GroES (Viitanen et al., 1995) and DnaK–DnaJ–GrpE families (Marshall et al., 1990; Schlicher and Soll, 1997). All available evidence indicates that the plastidial homologues of these chaperones act similarly to their eubacterial and mitochondrial counterparts. However, it is not yet clear whether the actual assembly step requires additional complex-specific factors. A complex-specific factor has recently been described for the cyanobacterial photosystem I, but its mechanism of action has yet to be investigated in detail (Bartsevich and Pakrasi, 1997). We are interested in understanding the biogenesis of the thylakoid membrane and particularly that of PSII at the molecular level. To identify genes that are involved in these processes we are pursuing a genetic approach using Arabidopsis thaliana as the experimental system (Meurer et al., 1996b). Arabidopsis, besides Chlamydomonas (Rochaix, 1995) and maize (Miles, 1982), offers a number of attractive features which makes it the system of choice for studying the biogenesis of the photosynthetic apparatus in photosynthetic eukaryotes. With the high chlorophyll fluorescence (hcf) phenotype (Miles, 1982) as a selection criterion, a systematic search for mutants defective in photosynthetic light reactions and electron transport was initiated. The mutations were generated by ethyl methanesulfonate treatment (Meurer et al., 1996b) or by T-DNA insertion mutagenesis (Feldmann, 1991). To date, ∼60 different recessive nuclear mutants have been isolated (Meurer et al., 1996b; J.Meurer and K.Meierhoff, personal communication). The mutants were characterized by chlorophyll fluorescence induction, P700 absorption kinetics, immunoblotting, and RNA gel blot analysis. This allowed us to classify the mutants with respect to the thylakoid membrane complex being affected by the mutation. Here we report on mutant hcf136 which is specifically deficient in PSII activity but not primarily affected in the functions of the other thylakoid membrane complexes. RNA gel-blot and immunoblotting analyses revealed that all known plastid- and nuclear-encoded mRNAs for PSII subunits are present in the mutant, but none of the corresponding subunits. Protein labelling studies revealed that the plastome-encoded PSII subunits are synthesized but do not accumulate. This indicates that the HCF136 gene encodes a factor that is essential for the stability or assembly of PSII. The HCF136 gene could be cloned because of its T-DNA tag and was found to encode a protein that has homologues in the cyanobacterial genome and the cyanelle genome of Cyanophora paradoxa. This is the first report on the molecular identification of a nuclear-encoded gene that is essential for the biogenesis of PSII in higher plants and that originated from the prokaryotic ancestor of extant plastids. Results Selection and phenotype of the hcf136 mutant The hcf136 mutant of A.thaliana ecotype Wassilewskya was generated by T-DNA insertion mutagenesis and isolated from the mutant collection of K.Feldmann (Feldmann, 1991). Genetic analysis showed that the mutation is recessive. The kanamycin resistance marker carried by the T-DNA and the mutant phenotype co-segregated, thus indicating that the mutation was due to the insertion of the T-DNA (data not shown). The hcf136 mutant was selected by its high chlorophyll fluorescence which could be detected by eye under UV light in the dark (Miles, 1982). When germinated on soil, homozygous mutant seedlings were found to be lethal. They developed pale green cotyledons but no primary leaves. Cultivation on a sucrose-supplemented Gelrite medium rescued the mutant seedlings leading to a wild-type-like habitus. However, the mutant seedlings became somewhat paler during culture in the light, indicating that even a relatively low photon flux density of ∼20–50 μmol/m2/s causes photooxidative damage and results in the progressive loss of chlorophyll. Mutant plants did not produce any flowers. Hcf136 is deficient in PSII activity The functional state of PSII can easily be monitored by non-invasive chlorophyll fluorescence measurements (Krause and Weis, 1991). Analysis of hcf136 mutant seedlings revealed no variable fluorescence with this technique (Figure 1A), indicating that PSII was completely inactive and that the linear electron flow was interrupted. Measurements of the absorbance kinetics of the P700 reaction pigment at 820 nm (Figure 1B) supported this inference. A strong light-pulse, given after illumination with far-red light, induced a further increase in the absorbance in mutant plants, whereas a decrease was detected in the wild type. Such an increase is also observed in plants treated with the herbicide DCMU [3-(3′,4′-dichlorophenyl)-1,1-dimethylurea] which completely blocks the electron transport between PSII and photosystem I (PSI) (compare with Meurer et al., 1996b). The altered redox kinetics of P700 in mutant plants are therefore due to a lacking electron flow from PSII, and are not caused by deficiencies in PSI. Figure 1.Spectroscopic analysis of hcf136 mutant and wild-type plants by measurements of chlorophyll fluorescence induction (A), P700 redox kinetics (B) and chlorophyll fluorescence emission at 77 K (C). For experimental details see Materials and methods. Download figure Download PowerPoint Chlorophyll fluorescence measurements at 77K verified the above findings (Figure 1C). The emission band at 733–735 nm, which is characteristic for a functional PSI, was clearly detectable in the mutant, corroborating evidence that PSI is functional or at least present in mutant plants. In contrast, the two emission peaks at 688 and 695 nm, which are indicative of the PSII reaction centre, were undetectable. Instead, the two bands were replaced by a new emission peak at 685 nm which is not found in the wild type. This additional band probably originates from the increased emission of the peripheral light-harvesting complex of PSII (LHCII) which in turn is caused by the inhibition of exiton transfer from the peripheral to the inner light-harvesting complex in the mutant (Krugh and Miles, 1995). Taken together, all spectroscopic data indicate that PSII is completely inactive in hcf136 and that this defect represents the primary lesion. Chloroplast ultrastructure in hcf136 PSII and its associated light-harvesting complex are preferentially located in the stacked grana regions, while the cytochrome b6f complex and PSI are found in the margins of the grana as well as in stroma thylakoids (Anderson, 1986). Wild-type chloroplasts display the typical differentiation into stroma and grana lamellae, whereas hcf136 thylakoids reveal a strikingly different ultrastructure (Figure 2). The grana appear enlarged by 6- to 8-fold and extend almost throughout the chloroplast. Mutant grana lamellae are closely appressed to each other, indicating that the spacing of the thylakoid membranes within one granum is reduced. Figure 2.Electron micrographs of chloroplasts from wild-type and hcf136 leaves. Both wild-type and mutant plants were grown on a sucrose-supplemented agar medium, and plants of comparable growth stages were selected for the electron microscopic analysis. Bar = 1 μm. Download figure Download PowerPoint PSII polypeptides do not accumulate in mutant thylakoids The spectroscopical analyses indicated that the mutational defect of hcf136 resides in PSII. To support this contention and to investigate the PSII subunits affected by the mutation, thylakoid membranes of hcf136 were analysed by immunoblotting (Figure 3) using a collection of antisera raised against individual PSII polypeptides and representative polypeptides of other photosynthetic membrane complexes (Meurer et al., 1996b). The plastome-encoded subunits of PSII analysed are almost undetectable (CP47, CP43, D2 and D1) or drastically depleted (cytochrome b559) in mutant thylakoids. Even the nuclear-encoded subunits of the oxygen-evolving complex, the 34 and 23 kDa proteins, fail to accumulate to significant levels in hcf136. In contrast, the amount of the 24 kDa LHCII polypeptide is not reduced as estimated by silver-stained SDS gels (data not shown). Figure 3.Immunoblot analysis of thylakoid membrane proteins from hcf136 and wild-type. Three-week-old plants were used for the analysis. Size fractionated membrane proteins were transferred to nitrocellulose, and specific polypeptides were immunodetected with the indicated antisera. The lanes were loaded with 8 μg (WT and hcf136) or 2 μg (WT 1/4) protein, respectively. The nature of the high molecular weight band reacting with the antiserum to the 34 kDa protein in both the mutant and wild-type is not known. Download figure Download PowerPoint Reduced levels (∼70%) are also observed for subunits A/B and D of PSI which were investigated as representative polypeptides of this photosystem (Figure 3). These reductions are already detectable in 1-week-old seedlings (data not shown), suggesting that the hcf136 mutation affects also PSI. However, it is often observed with PSII mutants that their PSI content is also affected, indicating that the reduced PSI levels in PSII mutants are due to secondary effects of the mutation. In the case of hcf136 this view is supported by the spectroscopic analyses (see above) which revealed that the PSI reaction centres accumulating are functional. Therefore, a direct effect of the hcf136 mutation on PSI appears to be unlikely but cannot be excluded at the current stage of investigation. No differences in steady-state levels were observed for subunits of the cytochrome b6f complex (cytochrome f and subunit IV) and of the chloroplast ATP synthase (α and β subunits), demonstrating that the HCF136 gene product is not needed for the assembly of these thylakoid membrane complexes. Taken together, the immunoblot analyses indicate that hcf136 is primarily affected in PSII and that the reduction in PSI reaction centres is likely due to secondary effects of the mutation. The plastome-encoded PSII polypeptides are synthesized but not stable in hcf136 To understand why PSII polypeptides do not accumulate in hcf136 plants, the levels and patterns of the plastomeand nuclear-encoded PSII transcripts were investigated by RNA gel blot hybridization. Representative plastomeand nuclear-encoded genes for components of other thylakoid membrane complexes were included in this analysis (Meurer et al., 1996b). The experiments revealed that both the amount and the sizes of all analysed transcripts were not altered in the mutant (data not shown). The inability of hcf136 to accumulate PSII subunits is therefore not caused by a missing transcript encoding one of these structural PSII proteins, but must be due to a translational defect or, alternatively, to the instability of the PSII subunits once they have been synthesized. To distinguish between these two possibilities, the rate of synthesis of thylakoid membrane proteins was investigated in intact mutant seedlings by pulse-labelling experiments with [35S]methionine (Figure 4). To obtain an easily interpretable labelling pattern, the synthesis of the nuclear-encoded chloroplast proteins was blocked with cycloheximide, and only the plastome-encoded PSII proteins were investigated. Figure 4 shows that the protein labelling patterns of mutant and wild type are similar with respect to the numbers of polypeptides detectable and their intensity of labelling. The PSII subunits CP47, CP43, D1 and D2 can be identified due to their known electrophoretic mobilities (Kim et al., 1994). They are synthesized in the mutant, but the incorporation of [35S]methionine into these proteins is reduced by ∼50% as compared with the wild type. Therefore, the lack of the plastome-encoded PSII subunits in hcf136 cannot be explained by a translational deficiency, but must be predominantly caused by an increased instability of these PSII subunits in the mutant background. Figure 4.In vivo protein synthesis with primary leaves of 12-day-old mutant and wild-type seedlings. Incorporation of [35S]methionine in the presence of cycloheximide was performed as described in Materials and methods. Wild-type and mutant proteins with equivalent amounts of radioactivity [100 000 c.p.m. (WT and hcf136) or 25 000 c.p.m. (WT 1/4) and 12 500 c.p.m. (WT 1/8)] were electrophoresed on polyacrylamide–SDS gels, blotted onto a nitrocellulose membrane and analysed by fluorography. The identity of the indicated bands was determined as described (Meurer et al., 1996a). The molecular masses (in kDa) were estimated by co-electrophoresis with commercially available size standards. Download figure Download PowerPoint Figure 4 also reveals that the synthesis of PSA-A/B is significantly reduced in the hcf136 mutant. This can be due to enhanced proteolysis within the short incubation time of 15 min or to an effect on synthesis of these proteins in the mutant background. Cloning of HCF136 sequences Since segregation and Southern blot data indicated that the mutated hcf136 gene was tagged by T-DNA, the genomic regions flanking the right border of the T-DNA were isolated by inverse PCR (Ochman et al., 1993) as described in Materials and methods. Sequence analysis of the cloned PCR-amplified fragment and Southern hybridization with genomic DNA from hcf136 and wild-type plants (data not shown) confirmed that the isolated PCR fragments indeed flank the right T-DNA border. To examine the effect of the T-DNA insertion on the expression of the HCF136 gene the isolated PCR genomic fragment was used for Northern hybridization (Figure 5). With RNA from wild-type plants the probe hybridized to a single transcript of ∼1.4 kb in length. No hybridizing RNA of that size could be detected in RNA from mutant plants but some degradation products appeared at ∼0.4 kb. Reprobing the filters with psbO sequences which encode the 34 kDa polypeptide of the oxygen-evolving complex of PSII revealed no differences in the hybridization intensity of the psbO mRNA, thus demonstrating that equal RNA amounts have been analysed in both cases (Figure 5). It follows from this that the T-DNA insertion leads to the inactivation of the presumptive HCF136 gene by inhibiting the stable accumulation of its transcript. Figure 5.Expression analysis of the HCF136 gene. Eight micrograms of leaf RNA from 3-week-old mutant and wild-type plants grown under the same conditions was analysed by Northern hybridization as described in Materials and methods. As a probe, a 530 bp DraI restriction fragment of the inverse PCR amplified products was used which contained exclusively genomic sequences directly adjacent to the T-DNA. To check for equal loading the filter was stripped and rehybridized with a psbO probe (Meurer et al., 1996a). Download figure Download PowerPoint To determine the sequence of the affected gene, a λZAP-cDNA library prepared from Arabidopsis leaves was screened with the genomic sequence that had been used for Northern hybridizations. Twenty-two of the positive cDNAs obtained were partially sequenced from both ends and found to contain identical sequences. The largest clone of these cDNAs, pcAt136–28, was 1400 bp in size and was sequenced on both strands. Sequence analysis revealed an open reading frame of 403 amino acids encoding a protein of 44.1 kDa. To confirm that the HCF136 gene disruption was indeed responsible for the high chlorophyll fluorescence phenotype of hcf136, the pcAt136–28 cDNA was fused to the 35S promoter of the Cauliflower Mosaic Virus (see Materials and methods), and the chimeric gene was introduced into homozygous mutant plants via Agrobacterium tumefaciens using a root transformation protocol (Valvekens et al., 1988). In each of the 14 independently generated transformants the HCF136 cDNA was able to complement the mutation and to restore wild-type characteristics. The transformed mutants turned normal green and showed chlorophyll fluorescence induction kinetics indistinguishable from wild type (Figure 6). In particular, the ratio of the variable to the maximal fluorescence (Fv/Fm) which reflects the efficiency of the primary photochemical events in PSII was fully restored in the complemented lines (Figure 6). The complementation experiment thus ultimately demonstrates that the mutant gene disrupted by the T-DNA is the causative factor of the mutation, and that the cDNA sequences isolated encode a functional HCF136 protein. Figure 6.Chlorophyll fluorescence induction kinetics of hcf136 seedlings transformed with the HCF136 cDNA under control of the 35S promoter. For details see Materials and methods. Download figure Download PowerPoint Characteristics and evolution of the HCF136 protein The HCF136 reading frame present in the cDNA pcAt136–28 encodes a protein of 403 amino acids (Figure 7). The flanking region of its ATG start codon matches the consensus motif CA ATG GC which has been found around the translational initiation sites of a large number of plant genes (Lütcke et al., 1987) suggesting that this ATG is the translational initiation codon. However, the genomic sequence of HCF136 (http://www.kazusa.or..) revealed that there is an additional in-frame ATG codon. This putative translational initiation codon is located 23 amino acids in front of the one present in pcAt136–28, but lacks the sequence context typical for plant translational initiation sites. Because of this, and since the complementation test demonstrated that the HCF136 reading frame present in pcAt136–28 encodes a functional protein we conclude that this upstream AUG codon is not used as the translational initiation codon. Figure 7.Amino acid sequence of HCF136 and comparison with ORF333 of Cyanophora paradoxa (C.p.) and ORF342 (slr2034) of Synechocystis sp. PCC6803 (Syn). Identical amino acids are star-marked, conserved exchanges are dotted. The nucleotide sequence of the HCF136 cDNA has been submitted to DDBJ/EMBL/GenBank databases under the accession No. Y15628. Download figure Download PowerPoint Database searches failed to detect any sequence motives or domains that could shed some light on the function of the HCF136 protein. These searches did, however, identify significant similarities with two ORFs of yet unknown function (Figure 7). ORF333 is encoded in the plastome of the glaucocystophycean alga Cyanophora paradoxa and ORF342 (slr2034) by the genome of the cyanobacterium Synechocystis sp. PCC6803. The three proteins revealed 34% identical and 14% similar amino acid positions which are more or less evenly distributed. In addition to three minor insertions, the central part of the HCF136 protein contains 19 amino acids which are not found in the glaucocystophycean and cyanobacterial homologues. We conclude that the HCF136 gene was present in the genome of the cyanobacterial-like endosymbiont that gave rise to the extant plastids of all photosynthetic eukaryotes, but was transferred to the nucleus before the rhodophyte/chromophyte and chlorophyte lineages separated. In both the cyanobacterium and the glaucocystophyte, the HCF136 homologues are located directly upstream of the psbE-F-L-J operon, whose gene order and composition has been conserved in both the cyanobacteria and the plastomes of all photosynthetic eukaryotes sequenced so far (http://www.ebi.ac.u..). The psbE-F-L-J operon exclusively encodes PSII proteins, and the clustering of the HCF136 homologues with this operon may be taken as further support that HCF136 is functionally associated with PSII. HCF136 encodes a lumenal protein that is located in stromal thylakoids The N-terminal region of the HCF136 reading frame is similar to transit sequences that direct nuclear-encoded proteins into the chloroplast (von Heijne et al., 1989), suggesting that the HCF136 protein is located in this organelle. To prove this assumption and to determine precisely the intracellular location of the HCF136 protein, immunoblot analyses were carried out with fractionated intact chloroplasts. As a prerequisite, an antiserum was raised to HCF136 protein which had been produced in Escherichia coli, and the antiserum was furthermore affinity-purified. The antiserum labelled a 37 kDa protein in crude extracts prepared from Arabidopsis and spinach plants but did not detect any protein in hcf136 extracts (data not shown). This finding demonstrated the specificity of the antiserum produced and corroborated that the T-DNA insertion prevented the accumulation of HCF136 protein. The apparent molecular mass of the detected protein was ∼7 kDa less than the predicted molecular mass of the entire HCF136 reading frame (Figure 8A). Such a reduced size was to be expected if the N-terminal part functions as a plastidial transit peptide and is removed after import into the organelle. In line with this expectation, the HCF136 protein was found to be located in intact chloroplasts (Figure 8A). Subfractionation of the intact chloroplasts by sucrose density gradient centrifugation showed that the HCF136 protein is associated with the thylakoid membrane and not with the envelope (Figure 8B). Washing with 0.2 M Na2CO3 (Figure 8A) released the HCF136 protein from the chloroplast membrane fraction suggesting that it is attached to this membrane. The hydropathy analysis (data not shown) supported this inference and showed that the first 18 N-terminal amino acids of the mature HCF136 protein form a hydrophobic patch, while the remainder of the protein is entirely hydrophilic. Fractionation of the thylakoid membranes in
Referência(s)