TY - JOUR
AU - Tzen, Jason T.C.
AB - Abstract An in vitro system was established to examine the targeting of proteins to maturing seed oil bodies. Oleosin, the most abundant structural protein, and caleosin, a newly identified minor constituent in seed oil bodies, were translated in a reticulocyte lysate system and simultaneously incubated with artificial oil emulsions composed of triacylglycerol and phospholipid. The results suggest that oil body proteins could spontaneously target to artificial oil emulsions in a co-translational mode. Incorporation of oleosin to artificial oil emulsions extensively protected a fragment of approximately 8 kDa from proteinase K digestion. In a competition experiment, in vitro translated caleosin and oleosin preferentially target to artificial oil emulsions instead of microsomal membranes. In oil emulsions with neutral phospholipids, relatively low protein targeting efficiency was observed. The targeting efficiency was substantially elevated when negatively charged phospholipids were supplemented to oil emulsions to mimic the native phospholipid composition of oil bodies. Mutated caleosin lacking various structural domains or subdomains was examined for its in vitro targeting efficiency. The results indicate that the subdomain comprising the proline knot motif is crucial for caleosin targeting to oil bodies. A model of direct targeting of oil-body proteins to maturing oil bodies is proposed. (Received July 10, 2001; Accepted August 29, 2001). Introduction Plant seeds store triacylglycerols (TAGs) as energy sources for germination and post-germinative growth of seedlings. The storage TAGs are confined to discrete spherical organelles called oil bodies, lipid bodies, oleosomes or spherosomes (Yatsu and Jacks 1972, Murphy 1993, Huang 1996, Napier et al. 1996). Oil bodies remain as individual small organelles even after a long period of storage in plant seeds (Slack et al. 1980). This stability is a consequence of the steric hindrance and electronegative repulsion provided by structural proteins, i.e. oleosins, on the surface of oil bodies (Tzen et al. 1992). Oleosins seem to cover the entire surface of an oil body, such that the compressed oil bodies in the cells of a mature seed never coalesce or aggregate (Tzen and Huang 1992). The TAG matrix of an oil body is surrounded by a monolayer of phospholipids (PLs) embedded with abundant oleosins and some minor proteins of higher molecular mass (Tzen et al. 1997). An oleosin molecule is proposed to comprise three distinct structural domains: an N-terminal amphipathic domain, a central hydrophobic anchoring domain, and a C-terminal amphipathic α-helical domain (Vance and Huang 1987). Sequence comparison among diverse species reveals that the central anchoring domain of oleosin is highly conserved, particularly in a relatively hydrophilic motif known as the proline knot (Tzen et al. 1992). Targeting of oleosins to seed oil bodies has been investigated in either in vivo systems via transgenic plants (Lee et al. 1991, van Rooijen and Moloney 1995, Abell et al. 1997, Sarmiento et al. 1997) or in vitro systems using microsomal membranes for integration of translated proteins (Hills et al. 1993, Loer and Herman 1993, Thoytes et al. 1995). The proline knot motif seems to play a crucial role for oleosin targeting to oil bodies (Abell et al. 1997). Based on in vitro studies, oleosins are proposed to target to the ER under the assistance of the signal-recognition particle prior to their accumulation on oil bodies (Hills et al. 1993, Loer and Herman 1993). Three minor proteins, temporarily termed Sops 1–3, have been identified exclusively in sesame oil bodies via immunofluorescence labeling (Chen et al. 1998). However, the biological functions of these three minor proteins remain unknown. Recently, a cDNA clone encoding sesame Sop1, named caleosin for its calcium-binding capacity, has been sequenced (Chen et al. 1999). Similar to oleosin structure, caleosin comprises three distinct structural domains: an N-terminal hydrophilic domain (including a calcium-binding motif), a central hydrophobic anchoring domain, and a C-terminal hydrophilic domain. In addition, a comparable proline knot motif is located in the central hydrophobic domain of caleosin. Whether this structural analogue in caleosin is responsible for its targeting to oil bodies remains to be elucidated. In this study, we established a novel in vitro system to examine protein targeting to seed oil bodies using artificial oil emulsions constituted with TAG and PL. Our results suggest that oil-body proteins spontaneously target to artificial oil emulsions in a co-translational mode. Effects of negatively charged PLs on in vitro targeting of caleosin and oleosin to artificial oil emulsions were evaluated. The structural domain required for caleosin targeting to artificial oil emulsions was also investigated by deletion mutagenesis. Based on the experimental results, a direct targeting model of oil-body proteins to maturing oil bodies is proposed. Results Migration shift of caleosin induced by calcium binding in SDS-PAGE It has been demonstrated that caleosin migrates faster in SDS-PAGE when associated with Ca2+ (Chen et al. 1999). Caleosin purified from sesame oil bodies, over-expressed in Escherichia coli, or in vitro translated in the Promega TnT® reticulocyte system, was treated with EGTA to remove endogenous cations and then incubated with Ca2+. Both EGTA-treated and Ca2+-associated caleosin molecules extracted from the above three samples were separately resolved in SDS-PAGE (Fig. 1). Migration shift induced by calcium binding was observed not only in caleosin from seed oil bodies and over-expressed E. coli (stained by Coomassie Blue) but also in that from in vitro translation (visualized by fluorography). The calcium-induced migration shift and relative migration positions of caleosin in the SDS-PAGE gels revealed that intact caleosin and one minor incomplete (or degraded) polypeptide were successfully in vitro transcribed/translated. Spontaneous and co-translational targeting of in vitro translated caleosin and oleosin to artificial oil emulsions An in vitro system was established to examine protein targeting to maturing seed oil bodies by incubating in vitro translated proteins with artificial oil emulsions composed of TAG and PL. For the two known oil body proteins, caleosin and oleosin, relatively high amounts of in vitro translated proteins were located in the oil fraction when artificial oil emulsions were supplied during protein translation, but found in supernatant and pellet fractions when artificial oil emulsions were incorporated after protein translation (Fig. 2). The recovered pellet in the fractionation was almost invisible and probably composed of aggregation of in vitro translated proteins. To examine the specificity of caleosin and oleosin targeting to artificial oil emulsions, a soluble seed storage protein, 2S albumin precursor, and a membrane-bound protein, HrpW, were in vitro translated accordingly for targeting analysis. For the soluble 2S albumin precursor, relatively high amounts of in vitro translated proteins were consistently found in supernatant regardless the incorporation of artificial oil emulsions. For the hydrophobic membrane protein HrpW, relatively high amounts of in vitro translated proteins were consistently found in the pellet regardless of the incorporation of artificial oil emulsions. The results indicate that oil-body proteins spontaneously target to artificial oil emulsions in a co-translational mode without apparent assistance of other factors, and that the targeting does not result from non-specific hydrophobic interaction between translated proteins and artificial oil emulsions. In the presence or absence of Ca2+, similar efficiency was observed for caleosin targeting to artificial oil emulsions. It is evident that calcium binding is not essential for caleosin targeting to artificial oil emulsions. This judgment is also supported by the targeting analysis using mutated caleosin lacking N-terminal domain derived from deletion mutagenesis (see the following results). Enhancement on resistance of oleosin to proteolytic digestion after targeting to artificial oil emulsions To examine the topology of oleosin in the above in vitro targeting system, oleosin was translated before or after inclusion of artificial oil emulsions and subjected to proteinase K digestion. In vitro translated oleosin was completely degraded within 20 min when it was translated before inclusion of artificial oil emulsions (Fig. 3A). The degradation could be significantly retarded when oleosin was translated in the presence of artificial oil emulsions prior to proteinase K digestion (Fig. 3B). After exhaustive digestion, a polypeptide of approximately 8 kDa, equivalent to the mass of oleosin central hydrophobic domain, was observed. Similar protein profile, particularly a protected polypeptide of approximately 8 kDa, could be observed when seed-purified oil bodies were subjected to proteolytic digestion (Tzen et al. 1992). The results suggest that oleosin co-translationally targeting to artificial oil emulsions apparently elevated its resistance to proteinase K digestion and its central hydrophobic domain is probably protected from the proteolytic digestion. Translated caleosin and oleosin preferentially targeting to artificial oil emulsions instead of microsomal membranes Compared with the post-translational results in Fig. 2, the relative content of co-translated caleosin or oleosin increased in the pellet fraction when incubated with microsomal membranes (Fig. 4, middle panel), in accord with the previous reports (Hills et al. 1993, Loer and Herman 1993). To evaluate the preferential targeting of oil-body proteins to the ER and budding oil bodies in maturing seeds, microsomal membranes and artificial oil emulsions were allowed to compete for translated caleosin or oleosin in the in vitro targeting system. In our experimental conditions, both caleosin and oleosin preferred targeting to artificial oil emulsions instead of microsomal membranes (Fig. 4, right panel), even though the PL content (accessible surface area for protein targeting) of microsomal membranes was approximately 250 times more than that of artificial oil emulsions in the co-existent competition. The results suggest that oil-body proteins are capable of targeting to maturing oil bodies directly without ER transportation. To examine if the preferential targeting to artificial oil emulsions is specific for oil-body proteins, the soluble 2S albumin precursor and hydrophobic membrane protein HrpW were in vitro translated accordingly for targeting analysis. A relatively high amount of the soluble 2S albumin precursor was detected in the microsomal fraction regardless the presence of artificial oil emulsions, presumably resulting from its ER targeting via the signal-recognition particle (SRP) by recognizing the N-terminal signal sequence. The majority of the hydrophobic HrpW was consistently found in the pellet fraction regardless of the inclusion of microsomal membranes and/or artificial oil emulsions. Evidently, unique oil-body proteins, but not other hydrophilic or hydrophobic proteins, specifically and preferentially target to artificial oil emulsions instead of microsomal membranes. Substantial retardation in targeting efficiency by eliminating negatively charged PLs from artificial oil emulsions To inspect the effect of PL composition of oil bodies on protein targeting, in vitro translated caleosin and oleosin were separately incubated with artificial oil emulsions comprising various combinations of PLs according to their native proportions. In both caleosin and oleosin, relatively low targeting efficiency was detected for artificial oil emulsions constituted with TAG and neutral PL (PC or PE) (Fig. 5). When 20% negatively charged PS and/or PI were supplemented to PC in artificial oil emulsions, the targeting efficiency could be substantially elevated to a level similar to that using artificial oil emulsions comprising native PL extracted from sesame oil bodies. The results suggest that negatively charged PLs are effective in protein targeting to artificial oil emulsions. In vitro caleosin targeting to artificial oil emulsions crucially affiliated with the proline knot subdomain Caleosin comprises three structural domains including an N-terminal hydrophilic domain, a central hydrophobic anchoring domain, and a C-terminal hydrophilic domain; the central hydrophobic domain can be divided into two subdomains including a helical subdomain and a proline knot subdomain (Fig. 6A). To identify the crucial structural portion for caleosin targeting to artificial oil emulsions, constructs for mutated caleosin polypeptides lacking various structural domains and/or subdomains were in vitro transcribed/translated in the presence of artificial oil emulsions (Fig. 6B). Similar targeting efficiency was observed for the mutated caleosin lacking the N-terminal domain, compared with that for unaltered caleosin (Fig. 6C). Two polypeptides were found in caleosin lacking the helical subdomain (Cal-H). Migration shift induced by calcium binding was observed in both polypeptides (data not shown), and thus they probably result from protein degradation or two different structural conformations of the mutated protein. Apparent reduction in targeting efficiency was detected when the helical subdomain was deleted. However, this reduction in caleosin targeting could be mostly rescued when the N-terminal domain was concurrently removed with the helical subdomain (Cal-NH). Severe impediment in targeting efficiency was observed when the proline knot subdomain was deleted from caleosin (Cal-P). Calesoin targeting was nearly abolished to a background level (compared with the post-translation in Fig. 2) when the entire central hydrophobic domain was truncated (Cal-HP). The results indicate that the targeting of caleosin to artificial oil emulsions is crucially affiliated with its central hydrophobic domain, particularly the proline knot subdomain. Discussion An in vitro system using microsomal membranes as targets for integration of translated proteins has been employed to investigate oleosin targeting to seed oil bodies (Hills et al. 1993, Loer and Herman 1993, Thoytes et al. 1995). However, detection of oil-body protein targeting in this in vitro system raises the concern whether the detected polypeptides in the microsomal fraction are translated proteins truly integrated into the membranes or merely co-precipitated contaminant proteins aggregated due to their hydrophobic property. To investigate specific targeting of oil-body proteins, we established a novel in vitro system by constituting artificial oil emulsions mimicking maturing seed oil bodies for integration of translated proteins. Under our experimental conditions, valid targeting to artificial oil emulsions was observed only for co-translated oil-body proteins, but not for post-translated polypeptides (Fig. 2). Impediment on post-translational targeting to artificial oil emulsions presumably resulted from hydrophobic aggregation of the oil-body proteins following translation. Our results suggest that oil-body proteins are capable of spontaneously targeting to oil bodies in a co-translational mode without apparent assistance of other factors, such as SRP complex, ER membrane proteins, or molecular chaperones. In fact, neither caleosin nor oleosin possesses cleavable signal sequence required for the SRP-dependent pathway of targeting to the ER (Chen et al. 1997, Chen et al. 1999). The success of our in vitro targeting system also suggests that oil-body proteins are capable of integrating into maturing oil bodies whose surface areas presumably are not widely covered by proteins. In contrast, in vitro translated proteins including oil-body specific caleosin and oleosin could hardly target to mature oil bodies whose surface areas are entirely covered by proteins (data not shown). This observation is in accord with the result using in vitro translated soybean oleosin reported by Loer and Herman (1993). Meanwhile, in vitro translated proteins were unable to integrate into oil emulsion constituted with pure TAG (data not shown). A similar phenomenon was observed using in vitro translated rapeseed oleosins (Hills et al. 1993). Indeed, the unstable pure TAG emulsions coalesced to form an oil layer on top of the aqueous reaction solution not long after emulsification, and thus segregated from the translated proteins. It has been proposed that oleosins are transported via the ER prior to their accumulation on oil bodies (Loer and Herman 1993, Hills et al. 1993). However, our results (Fig. 2, 4) suggest another possible (probably the major) route of direct targeting of oil-body proteins to maturing (or budding) oil bodies (Fig. 7). The model of direct targeting proposes that oil-body proteins spontaneously target to maturing oil bodies in a co-translational mode without packaging and transportation via the ER while TAG, PC, PS, and PI are first synthesized in the smooth ER, directed toward maturing oil bodies, and assembled into mature lipid-storage organelles. Regardless the lack of sequence homology, both caleosin and oleosin, the two known oil-body proteins, possess a similar proline knot motif (Frandsen et al. 2001). The proline knot motif in oleosin has been proposed to play an essential role for its targeting to oil bodies (Abell et al. 1997). In our in vitro targeting analysis, the crucial structural portion required for caleosin targeting to artificial oil emulsion is the central hydrophobic domain, particularly the subdomain including the proline knot motif. The helical subdomain preceding the proline knot subdomain of caleosin increases the hydrophobicity of the protein, and thus probably enhances its stability in seed oil bodies via hydrophobic interaction. Apparent reduction in targeting efficiency of caleosin lacking the helical subdomain (Fig. 6C) may not imply direct involvement of this subdomain in protein targeting to artificial oil emulsions since the targeting efficiency could be mostly recovered when the hydrophilic N-terminal domain was also truncated. Instead, the apparent reduction may result from post-targeting structural instability of attenuated hydrophobic interaction between artificial oil emulsions and the central anchoring domain of the mutated caleosin lacking the helical subdomain. Negatively charged PLs have been proposed to be involved in targeting of certain proteins to various membrane structures using in vitro systems, such as model phospholipid membranes (Kennedy et al. 1997) and vesicle bilayers (Leventis and Silvius 1998). Negatively charged PS and PI are present in a consistent amount (30–40%) in the PLs of oil bodies from diverse seeds (Tzen et al. 1993). Taking advantage of the in vitro targeting system established in this study, we found that inclusion of negatively charged PS and PI in artificial oil emulsions substantially enhanced the targeting efficiency of caleosin and oleosin to those artificial oil emulsions (Fig. 5). It is likely that the negatively charged PS and PI, presumably resided in the surface area of seed oil bodies, are involved in protein targeting to these lipid-storage organelles. Nevertheless, it remains to be studied what the structural requirements in oil-body proteins are that interact with the surface negatively charged PLs of oil bodies for their efficient targeting. Materials and Methods Purification of sesame oil bodies Mature sesame (Sesamum indicum L.) seeds were gifts from the Crop Improvement Department, Tainan District Agricultural Improvement Station. The seeds were soaked in water for 10 min prior to purification of oil bodies. Extracted oil bodies were subjected to further purification including two-layer flotation by centrifugation, detergent washing, ionic elution, treatment of chaotropic agent, and integrity testing with hexane (Tzen et al. 1997). Over-expression of sesame caleosin in Escherichia coli A full-length cDNA clone of sesame caleosin (Accession no. AF109921) was constructed in the non-fusion expression vector, pET29a(+) (Novagen), using an NdeI site at the initial methionine position and a XhoI site in the polylinker of the vector. The recombinant plasmid, termed pECal, was used to transform E. coli strain NovaBlue (DE3). Over-expression was induced by 0.1 mM IPTG in a bacteriophage T7 RNA polymerase/promoter system. Three h after induction, the E. coli cells were harvested, lysed by sonication in 10 mM sodium phosphate buffer, pH 7.5, and then subjected to further analyses. In vitro transcription/translation of sesame caleosin Caleosin was in vitro translated using a TnT® T7 coupled reticulocyte lysate system (Promega). Coupled transcription/translation was executed for 90 min at 30°C according to the supplier’s protocol using 0.5 µg pECal construct as a DNA template in a reaction mixture of 25 µl. The translated caleosin was radio-labeled with Amersham International Redivue l-[35S]methionine. Effect of calcium binding on caleosin migration in SDS-PAGE Oil bodies (50 mg) in 500 µl Tris-buffered-saline containing 10 mM Tris-HCl buffer, pH 7.5 were mixed with an equal volume of 200 mM ethylene-glycol-bis(β-aminoethyl ether)-tetraacetic acid (EGTA) for 5 min at room temperature. The oil bodies were subjected to examination of caleosin migration in SDS-PAGE (Laemmli 1970) directly or after incubation with 100 mM CaCl2. Similar treatments were also applied to caleosin over-expressed in E. coli or in vitro translated in the Promega reticulocyte lysate system. After electrophoresis, proteins extracted from oil bodies or E. coli cells were stained with Coomassie Blue R-250 while the in vitro translated caleosin in the SDS-PAGE gel was fixed with methanol : water : acetic acid (50 : 40 : 10, v/v/v), dried on a sheet of Whatman 3MM filter paper, and visualized by fluorography using a Kodak X-ray film exposed for 48 h. Purification of TAG and PL To separately extract neutral lipids (mainly TAG) and PLs, sesame oil bodies were initially mixed with an equal volume of diethyl ether (Peng and Tzen 1998). After centrifugation, the upper ether fraction was harvested and evaporated under nitrogen to collect TAG. The lower water fraction was further extracted with chloroform/methanol (2 : 1, v/v). After centrifugation, extracted PLs in the lower chloroform fraction were collected. The native TAG and PLs extracted from sesame oil bodies were subjected to the following preparation of artificial oil emulsions. Preparation of artificial oil emulsions For preparation of artificial oil emulsions, a 1-ml suspension containing 20 mg TAG and 200 µg PL was prepared in a 1.5-ml eppendorf tube. Native PLs purified from sesame oil bodies or various combinations of PLs (PC, PE, PS, and PI purchased from Sigma) dissolved in chloroform were placed at the bottom of the eppendorf tube, and the chloroform was allowed to evaporate under nitrogen. After evaporation, purified TAG was incorporated and then sonicated with a 3-mm-diameter probe in a Sonics & Materials VCX-400 ultrasonic processor with the amplitude at 30% for 15 s. The sample was cooled on ice for 5 min. The sonication was repeated two more times to generate artificial oil emulsions. The artificial oil emulsions were diluted one thousand times before they were supplied to in vitro translated proteins for targeting analysis. Construction of plasmids for in vitro translation of oleosin, 2S albumin precursor, and HrpW A recombinant plasmid termed pEOle, was constructed with a full-length cDNA clone of rice oleosin (Accession no. U43930) in the non-fusion expression vector, pET28a(+) (Novagen), using an NcoI site at the initial methionine position and a XhoI site in the polylinker of the vector. A recombinant plasmid termed pEAlb, was constructed with a full-length cDNA clone of sesame 2S albumin precursor (Accession no. AF091841) in the fusion expression vector, pET30c(+) (Novagen), using an EcoRI site and a XhoI site in the polylinker of the vector. Sesame 2S albumin precursor comprises an N-terminal cleavable signal sequence of 21 residues responsible for targeting to the ER via the SRP-dependent pathway, and its processed mature polypeptide is a soluble storage protein accumulated in seed protein bodies (Tai et al. 1999). A recombinant plasmid termed pEHrp, was constructed with a full-length cDNA clone of HrpW (Accession no. U25812) in the non-fusion expression vector, pET29a(+) (Novagen), using an NdeI site at the initial methionine position and a XhoI site in the polylinker of the vector. HrpW, composed of six predicted transmembrane helices, is one of the membrane-bound subunits consisting of a protein apparatus for secretion of elicitors responsible for hypersensitive response and pathogenicity (Huang et al. 1995). Analysis of targeting efficiency of translated proteins to artificial oil emulsions In the presence of 1 µl diluted artificial oil emulsions, caleosin, oleosin, 2S albumin precursor, or HrpW was in vitro translated as described previously using pECal, pEOle, pEAlb, or pEHrp as the reaction template. After translation, the reaction mixture (25 µl) was added with 5 µl undiluted artificial oil emulsions (to assist formation of oil layer in later centrifugation) and 175 µl washing solution containing 0.5 M sucrose and 0.1 M Na2CO3 at pH 11.5. The sample was kept on ice for 30 min with occasional mixing, and then fractionated into oil layer, supernatant, and pellet by centrifugation at 14,000 rpm for 30 min at 4°C using an eppendorf centrifuge. The oil layer and the pellet were separately resuspended in 195 µl (equivalent to the volume of the supernatant) washing solution. Relative contents of in vitro translated proteins in the three fractions were resolved in SDS-PAGE and detected by fluorography as mentioned previously. Proteolytic digestion of in vitro translated oleosin Oleosin was in vitro translated in a reaction mixture of 25 µl before or after inclusion of artificial oil emulsions, and then subjected to proteolytic digestion using 200 ng proteinase K (Sigma). The digestion was executed in a buffer containing 10 mM Tris-HCl, pH 8.0 and 10 mM CaCl2 at room temperature for 0, 20, 40, 60, 90, or 120 min. Proteolytic digestion was stopped at various periods of time by adding phenylmethylsulfonyl fluoride to a final concentration of 0.1 mM. The degradation of translated oleosin was monitored in SDS-PAGE by fluorography as described previously. Competition between microsomal membranes and artificial oil emulsions for in vitro translated caleosin, oleosin, 2S albumin precursor, and HrpW In a competition experiment, in vitro targeting efficiency was analyzed as mentioned previously except using canine pancreatic microsomal membranes (Promega) and/or artificial oil emulsions as target sources for integration of translated proteins. In the 25 µl reaction mixture, 1.2 µl microsomal membranes (according to the supplier’s protocol) and/or 1 µl diluted artificial oil emulsions were supplied to translated caleosin, oleosin, 2S albumin precursor, or HrpW. The PL content (>50 µg) of microsomal membranes was approximately 250 times more than that (0.2 µg) of artificial oil emulsions in the co-existent competition. Construction of plasmids for mutated caleosin Constructs of mutated caleosin lacking various structural domains and/or subdomains were obtained by PCR-based mutagenesis. In each construct, two 5′-phosphorylated divergent primers were designed to anneal to the desired mutation site in the pECal plasmid. For those constructs encoding truncated caleosin polypeptides lacking the N-terminal domain, an initial methionine codon was introduced via one of the paired primers. The paired primers are 5′-catatgtatatctccttcttaa-3′ and 5′-gtgatagcttcccttataatggct-3′ for pECal-N (lacking N-terminal domain), 5′-attgaaaccaatttggcgaagt-3′ and 5′-agttatcctactctcccgggtt-3′ for pECal-H (lacking helical subdomain), 5′-catatgtatatctccttcttaa-3′ and 5′-ggtttcaatgtgatagcttccct-3′ for pECal-NH (lacking N-terminal domain and helical subdomain), 5′-cagcgccacattaatgacgata-3′ and 5′-cacaaggccaaacatggaagcga-3′ for pECal-P (lacking proline knot subdomain), or 5′-attgaaaccaatttggcgaagt-3′ and 5′-cacaaggccaaacatggaagcga-3′ for pECal-HP (lacking helical and proline knot subdomains). Amplification was carried out using pfu DNA polymerase (Stratagene) for 12 cycles at 94°C, 1 min; 50°C, 1 min; 72°C, 14 min. The blunt-ended linear PCR product was resolved in an agarose gel, purified by ethanol precipitation, and self-ligated to form a circular plasmid. The mutated DNA was transformed and amplified in E. coli (DH10B). The mutations were confirmed by sequencing using the SequenceTM Version 2.0 DNA Sequence kit (USB). Furthermore, the mutated constructs were transformed and over-expressed in E. coli strain NovaBlue (DE3), and the relative mobility of the over-expressed polypeptides was resolved in SDS-PAGE (data not shown) to ensure the accuracy of construct expression. All the mutated constructs were subjected to analysis of targeting efficiency of their translated products to artificial oil emulsions using the procedure described previously. Acknowledgements The authors wish to thank Professor Chih-Ning Sun for critical reading of the manuscript, Professor Hsiou-Chen Huang for providing the HrpW clone, and Dr. Tien-Joung Yiu for supplying mature sesame seeds. The work was supported by a grant from the National Science Council, Taiwan, ROC (NSC 89-2313-B-005-095 to JTC Tzen). 1 Corresponding author: E-mail, tctzen@dragon.nchu.edu.tw; Fax, +886-4-2285-3527. View largeDownload slide Fig. 1 SDS-PAGE of seed-purified, E. coli over-expressed, and in vitro translated caleosin treated with or without Ca2+. Sesame oil body proteins (10 µg), the total proteins (5 µg) of E. coli cells containing over-expressed caleosin, and in vitro translated caleosin in the Promega reticulocyte lysate system were separately treated with 100 mM EGTA and then incubated with 100 mM Ca2+. The proteins were either stained with Coomassie Blue (for oil-body and E. coli proteins) or visualized by fluorography (for in vitro translated caleosin). Labels on the left indicate the molecular masses of proteins. View largeDownload slide Fig. 1 SDS-PAGE of seed-purified, E. coli over-expressed, and in vitro translated caleosin treated with or without Ca2+. Sesame oil body proteins (10 µg), the total proteins (5 µg) of E. coli cells containing over-expressed caleosin, and in vitro translated caleosin in the Promega reticulocyte lysate system were separately treated with 100 mM EGTA and then incubated with 100 mM Ca2+. The proteins were either stained with Coomassie Blue (for oil-body and E. coli proteins) or visualized by fluorography (for in vitro translated caleosin). Labels on the left indicate the molecular masses of proteins. View largeDownload slide Fig. 2 Targeting efficiency of in vitro translated caleosin, oleosin, 2S albumin precursor, and HrpW. Artificial oil emulsions were supplied to expressed caleosin (with or without EGTA), oleosin, 2S albumin precursor, and HrpW either before or after in vitro translation. After reaction, three fractions, oil layer (O), supernatant (S), and pellet (P), were collected by centrifugation for each reaction mixture. The translated proteins in the three fractions were resolved in SDS-PAGE with loading samples adjusted to represent amounts derived from an equal quantity of the reaction mixture, and then visualized by fluorography. View largeDownload slide Fig. 2 Targeting efficiency of in vitro translated caleosin, oleosin, 2S albumin precursor, and HrpW. Artificial oil emulsions were supplied to expressed caleosin (with or without EGTA), oleosin, 2S albumin precursor, and HrpW either before or after in vitro translation. After reaction, three fractions, oil layer (O), supernatant (S), and pellet (P), were collected by centrifugation for each reaction mixture. The translated proteins in the three fractions were resolved in SDS-PAGE with loading samples adjusted to represent amounts derived from an equal quantity of the reaction mixture, and then visualized by fluorography. View largeDownload slide Fig. 3 Protection of in vitro translated oleosin from proteolytic digestion after targeting to artificial oil emulsions. Oleosin was in vitro translated either before (A) or after (B) incorporation of artificial oil emulsions, and then subjected to proteinase K digestion. The degraded fragments of oleosin at various intervals of digestion were resolved in SDS-PAGE with loading samples adjusted to represent amounts derived from an equal quantity of the reaction mixture, and then visualized by fluorography. Labels on the left indicate the molecular masses of polypeptides. View largeDownload slide Fig. 3 Protection of in vitro translated oleosin from proteolytic digestion after targeting to artificial oil emulsions. Oleosin was in vitro translated either before (A) or after (B) incorporation of artificial oil emulsions, and then subjected to proteinase K digestion. The degraded fragments of oleosin at various intervals of digestion were resolved in SDS-PAGE with loading samples adjusted to represent amounts derived from an equal quantity of the reaction mixture, and then visualized by fluorography. Labels on the left indicate the molecular masses of polypeptides. View largeDownload slide Fig. 4 Competition between microsomal membranes and artificial oil emulsions for translated caleosin, oleosin, 2S albumin precursor, and HrpW. Microsomal membranes and artificial oil emulsions were separately or simultaneously supplied to translated caleosin, oleosin, 2S albumin precursor or HrpW. The relative contents of translated proteins were detected in the three fractions, oil layer (O), supernatant (S), and pellet (P) using the same procedure described in Fig. 2. View largeDownload slide Fig. 4 Competition between microsomal membranes and artificial oil emulsions for translated caleosin, oleosin, 2S albumin precursor, and HrpW. Microsomal membranes and artificial oil emulsions were separately or simultaneously supplied to translated caleosin, oleosin, 2S albumin precursor or HrpW. The relative contents of translated proteins were detected in the three fractions, oil layer (O), supernatant (S), and pellet (P) using the same procedure described in Fig. 2. View largeDownload slide Fig. 5 Effects of PL composition on targeting efficiency of in vitro translated caleosin and oleosin. In vitro translated caleosin and oleosin were separately incubated with artificial oil emulsions composed of TAG and various PL compositions (native PL extracted from sesame oil bodies or different combinations of commercial PLs indicated on the left margin). The relative contents of translated caleosin and oleosin were detected in the three fractions, oil layer (O), supernatant (S), and pellet (P) using the same procedure described in Fig. 2, and quantitated by a densitometer (numerical values under the photos) using the TINA quantifier program. View largeDownload slide Fig. 5 Effects of PL composition on targeting efficiency of in vitro translated caleosin and oleosin. In vitro translated caleosin and oleosin were separately incubated with artificial oil emulsions composed of TAG and various PL compositions (native PL extracted from sesame oil bodies or different combinations of commercial PLs indicated on the left margin). The relative contents of translated caleosin and oleosin were detected in the three fractions, oil layer (O), supernatant (S), and pellet (P) using the same procedure described in Fig. 2, and quantitated by a densitometer (numerical values under the photos) using the TINA quantifier program. View largeDownload slide Fig. 6 (A) A simplified secondary structural model of sesame caleosin on the surface of oil bodies. A monolayer of PLs, depicted by gray circles attached with two tails, segregates the hydrophobic TAG matrix of an oil body from hydrophilic cytosol. Numbers indicate the relative positions (counting from N-terminus) of residues at the junctions of various domains and subdomains. This model is simplified from the structural model of caleosin proposed by Chen et al. (1999). (B) A schematic diagram of mutated caleosin polypeptides encoded by various deletion constructs. Mutated caleosin polypeptides lacking various domains and/or subdomains are encoded by deletion constructs and aligned in their primary structures. The N-terminal domain, helical subdomain, proline knot subdomain, and C-terminal domain are denoted as open boxes with centered letters, N, H, P, and C, respectively. (C) Targeting efficiency of the mutated caleosin polypeptides. In vitro translated caleosin polypeptides were separately analyzed for their targeting to artificial oil emulsions. The relative contents of mutated caleosin polypeptides were detected in the three fractions, oil layer (O), supernatant (S), and pellet (P) using the same procedure described in Fig. 2. View largeDownload slide Fig. 6 (A) A simplified secondary structural model of sesame caleosin on the surface of oil bodies. A monolayer of PLs, depicted by gray circles attached with two tails, segregates the hydrophobic TAG matrix of an oil body from hydrophilic cytosol. Numbers indicate the relative positions (counting from N-terminus) of residues at the junctions of various domains and subdomains. This model is simplified from the structural model of caleosin proposed by Chen et al. (1999). (B) A schematic diagram of mutated caleosin polypeptides encoded by various deletion constructs. Mutated caleosin polypeptides lacking various domains and/or subdomains are encoded by deletion constructs and aligned in their primary structures. The N-terminal domain, helical subdomain, proline knot subdomain, and C-terminal domain are denoted as open boxes with centered letters, N, H, P, and C, respectively. (C) Targeting efficiency of the mutated caleosin polypeptides. In vitro translated caleosin polypeptides were separately analyzed for their targeting to artificial oil emulsions. The relative contents of mutated caleosin polypeptides were detected in the three fractions, oil layer (O), supernatant (S), and pellet (P) using the same procedure described in Fig. 2. View largeDownload slide Fig. 7 A model of direct targeting of oil-body proteins to maturing oil bodies. This model illustrates the assembly of three essential constituents (TAG, PL, and protein) in oil bodies. Both TAG and PL (neutral PC and negatively charged PS and PI) are synthesized in the smooth ER and transported to maturing oil bodies. Oil-body proteins target to maturing oil bodies spontaneously and co-translationally without ER transportation. Budding areas of the ER and the mature oil body in cytosol were cut open to reveal TAG inside the structure. View largeDownload slide Fig. 7 A model of direct targeting of oil-body proteins to maturing oil bodies. This model illustrates the assembly of three essential constituents (TAG, PL, and protein) in oil bodies. Both TAG and PL (neutral PC and negatively charged PS and PI) are synthesized in the smooth ER and transported to maturing oil bodies. Oil-body proteins target to maturing oil bodies spontaneously and co-translationally without ER transportation. Budding areas of the ER and the mature oil body in cytosol were cut open to reveal TAG inside the structure. Abbreviations EGTA ethylene-glycol-bis(β-aminoethyl ether)-tetraacetic acid PC phosphatidylcholine PE phosphatidylethanolamine PI phosphatidylinositol PL(s) phospholipid(s) PS phosphatidylserine SRP signal-recognition particle TAG(s) triacylglycerol(s). References Abell, B.M., Holbrook, L.A., Abenes, M., Murphy, D.J., Hills, M.J. and Moloney, M.M. ( 1997) Role of the proline knot motif in oleosin endoplasmic reticulum topology and oil body targeting. Plant Cell  9: 1481–1493. Google Scholar Chen, E.C.F., Tai, S.S.K., Peng, C.C. and Tzen, J.T.C. ( 1998) Identification of three novel unique proteins in seed oil bodies of sesame. Plant Cell Physiol.  39: 935–941. Google Scholar Chen, J.C.F., Lin, R.H., Huang, H.C. and Tzen, J.T.C. ( 1997) Cloning, expression and isoform classification of a minor oleosin in sesame oil bodies. J. Biochem.  122: 819–824. Google Scholar Chen, J.C.F., Tsai, C.C.Y. and Tzen, J.T.C. ( 1999) Cloning and secondary structure analysis of caleosin, a unique calcium-binding protein in oil bodies of plant seeds. Plant Cell Physiol.  40: 1079–1086. Google Scholar Frandsen, G.I., Mundy, J. and Tzen, J.T.C. ( 2001) Oil bodies and their associated proteins, oleosin and caleosin. Physiol. Plant.  112: 301–307. Google Scholar Hills, M.J., Watson, M.D. and Murphy, D.J. ( 1993) Targeting of oleosins to the oil bodies of oilseed rape (Brassica napus L.). Planta  189: 24–29. Google Scholar Huang, A.H.C. ( 1996) Oleosin and oil bodies in seeds and other organs. Plant Physiol.  110: 1055–1061. Google Scholar Huang, H.C., Lin, R.H., Chang, C.J., Collmer, A. and Deng, W.L. ( 1995) The complete hrp gene cluster of Pseudomonas syringae pv. syringae 61 includes two blocks of genes required for Harpin secretion that are arranged colinearly with Yersinia ysc homologs. Mol. Plant Microbe Interact.  8: 733–746. Google Scholar Kennedy, M.T., Brockman, H. and Rusnak, F. ( 1997) Determinants of calcineurin binding to model membranes. Biochemistry  36: 13579–13585. Google Scholar Laemmli, U.K. ( 1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature  227: 680–685. Google Scholar Leventis, R. and Silvius, J.R. ( 1998) Lipid-binding characteristics of the polybasic carboxy-terminal sequence of K-ras4B. Biochemistry  37: 7640–7648. Google Scholar Lee, W.S., Tzen, J.T.C., Kridl, J.C., Radke, S.E. and Huang, A.H.C. ( 1991) Maize oleosin is correctly targeted to seed oil bodies in Brassica napus transformed with the maize oleosin gene. Proc. Natl. Acad. Sci. USA  88: 6181–6185. Google Scholar Loer, D.S. and Herman, E.M. ( 1993) Cotranslational integration of soybean (Glycine max) oil body membrane protein oleosin into microsomal membranes. Plant Physiol.  101: 993–998. Google Scholar Murphy, D.J. ( 1993) Structure, function and biogenesis of storage lipid bodies and oleosins in plants. Prog. Lipid Res.  32: 247–280. Google Scholar Napier, J.A., Stobart, A.K. and Shewry, P.R. ( 1996) The structure and biogenesis of plant oil bodies: the role of the ER membrane and the oleosin class of proteins. Plant Mol. Biol.  31: 945–956. Google Scholar Peng, C.C. and Tzen, J.T.C. ( 1998) Analysis of the three essential constituents of oil bodies in developing sesame seeds. Plant Cell Physiol.  39: 35–42. Google Scholar Sarmiento, C., Ross, J.H., Herman, E. and Murphy, D.J. ( 1997) Expression and subcellular targeting of a soybean oleosin in transgenic rapeseed. Implications for the mechanism of oil-body formation in seeds. Plant J.  11: 783–796. Google Scholar Slack, C.R., Bertaud, W.S., Shaw, B.P., Holland, R., Browse, J. and Wright, H. ( 1980) Some studies on the composition and surface properties of oil bodies from the seed cotyledons of safflower and linseed. Biochem. J.  190: 551–561. Google Scholar Tai, S.S.K., Wu, L.S.H., Chen, E.C.F. and Tzen, J.T.C. ( 1999) Molecular cloning of 11S globulin and 2S albumin, the two major seed storage proteins in sesame. J. Agric. Food Chem.  47: 4932–4938. Google Scholar Thoytes, P.J.E., Millichip, M.I., Stobart, A.K., Griffiths, W.T., Shewry, P.R. and Napier, J.A. ( 1995) Expression and in vitro targeting of a sunflower oleosin. Plant Mol. Biol.  29: 403–410. Google Scholar Tzen, J.T.C., Cao, Y.Z., Laurent, P., Ratnayake, C. and Huang, A.H.C. ( 1993) Lipids, proteins, and structure of seed oil bodies from diverse species. Plant Physiol.  101: 267–276. Google Scholar Tzen, J.T.C. and Huang, A.H.C. ( 1992) Surface structure and properties of plant seed oil bodies. J. Cell Biol.  117: 327–335. Google Scholar Tzen, J.T.C., Lie, G.C. and Huang, A.H.C. ( 1992) Characterization of the charged components and their topology on the surface of plant seed oil bodies. J. Biol. Chem.  267: 15626–15634. Google Scholar Tzen, J.T.C., Peng, C.C., Cheng, D.J., Chen, E.C.F. and Chiu, J.M.H. ( 1997) A new method for seed oil body purification and examination of oil body integrity following germination. J. Biochem.  121: 762–768. Google Scholar Vance, V.B. and Huang, A.H.C. ( 1987) The major protein from lipid bodies of maize. Characterization and structure based on cDNA cloning. J. Biol. Chem.  262: 11275–11279. Google Scholar van Rooijen, G.J.H. and Moloney, M.M. ( 1995) Plant seed oil-bodies as carriers for foreign proteins. Bio/Technology.  13: 72–77. Google Scholar Yatsu, L.Y. and Jacks, T.J. ( 1972) Spherosome membranes. Half unit-membranes. Plant Physiol.  49: 937–943. Google Scholar
TI - An in vitro System to Examine the Effective Phospholipids and Structural Domain for Protein Targeting to Seed Oil Bodies
JF - Plant and Cell Physiology
DO - 10.1093/pcp/pce160
DA - 2001-11-15
UR - https://www.deepdyve.com/lp/oxford-university-press/an-in-vitro-system-to-examine-the-effective-phospholipids-and-jZSJtQTKzM
SP - 1245
EP - 1252
VL - 42
IS - 11
DP - DeepDyve
ER -