Structural studies on transmembrane proteins. 1. Model study using bacteriorhodopsin mutants containing single cysteine residues

In developing new approaches to structural studies of polytopic transmembrane proteins, we have prepared bacteriorhodopsin mutants containing single cysteine residues at selected sites in different topological domains. Four such mutants were prepared: Gly-72----Cys and Ser-169----Cys in the presumed looped-out regions on the opposite sides of the membrane bilayer and Thr-90----Cys and Leu-92----Cys in the membrane-embedded helix C. The four mutants folded and regenerated the characteristic chromophore in detergent/phospholipid micelles and pumped protons like the wild-type bacteriorhodopsin. After reconstitution in asolectin vesicles, the sulfhydryl groups in the mutants Gly-72----Cys and Ser-169----Cys reacted with iodo[2-3H]acetic acid, while the sulfhydryl groups in the membrane-embedded mutants, Thr-90----Cys and Leu-92----Cys, did not. The sulfhydryl groups in all four mutants could be derivatized in the denatured state by reaction with iodoacetic acid or 6-acryloyl-2-(dimethylamino)naphthalene. Of these derivatives, the two from the mutants Gly-72----Cys and Ser-169----Cys folded like the wild-type bacterioopsin, whereas of the two from the helix C mutants, Thr-90----Cys and Leu-92----Cys, only the latter folded normally. However, the folding of Leu-92----Cys was also impaired when treated with the bulky 5-(iodoacetamido)fluorescein. The reactivity and the folding behavior of the cysteine mutants can thus report on the topographic domain as well as on the orientation of the helices within the membrane.     

The chromophore retinal hinders passive proton/hydroxide ion translocation through bacteriorhodopsin

Experiments have been performed to examine any influence of the chromophore retinal in bacteriorhodopsin (BR) on the passive proton/hydroxide ion flux through this integral membrane protein. BR was reconstituted into dimyristoylphosphatidylcholine (DMPC)-phosphatidylserine or DMPC-dimyristoylphosphatidylglycerol unilamellar vesicles with molar lipid to protein ratios ranging from 30 to 150. The entrapped fluorescence dye pyranine served as a reliable indicator of the internal proton concentration. Transmembrane pH-gradients were quickly established across the vesicular membrane and the kinetics of the induced fluorescence changes were compared for vesicles with incorporated native BR, BR bleached to the chromophore-free protein bacterioopsin, and BR regenerated from bacterioopsin with all-trans-retinal, respectively. For aggregated protein molecules, the H+/OH- diffusion across bacterioopsin was always considerably faster than that through the protein containing covalently bound retinal. The decay rate of the imposed pH-gradient was 4.4-9.1 and 2.0-5.1 times slower for native and regenerated BR, respectively, as compared to bacterioopsin. Stepwise regeneration of bacterioopsin with all-trans-retinal revealed a linear dependence of the predominant delta pH-decay time on the degree of regeneration. Essentially the same observations were made with monomeric protein molecules in vesicular lipid membranes. The results demonstrate that the chromophore retinal itself blocks the H+/OH- conducting pathway across the transmembrane protein BR or indirectly controls this path by inducing conformational changes in the protein upon binding.     

Evidence for post-translational membrane insertion of the integral membrane protein bacterioopsin expressed in the heterologous halophilic archaeon Haloferax volcanii

The gene coding for the integral membrane protein bacterioopsin (Bop), that is composed of seven transmembrane helices, was expressed in the halophilic archaeon Haloferax volcanii as a fusion protein with the halobacterial enzyme dihydrofolate reductase and with the cellulose binding domain of Clostridium thermocellum cellulosome. In each case, bacterioopsin was present both in the membrane and in the cytoplasmic fractions. Pulse-chase labeling experiments showed that the fusion protein in the cytoplasmic fraction is the precursor of the membrane-bound species. Bacterioopsin mutants that lack the seventh helix (BopDelta7) were found to accumulate only in the cytoplasmic fraction, whereas bacterioopsin mutants that lack either helices four and five (BopDelta4-5), or helices one and two (BopDelta1-2), were found in the cytoplasmic as well as in the membrane fractions. The seventh helix, when expressed alone, could target in trans the insertion of a separately expressed bacterioopsin mutant protein that has only the first six helices. These results support a model in which bacterioopsin is produced in H. volcanii as a soluble protein and in which its insertion into the membrane occurs post-translationally. According to this model, membrane insertion is directed by the seventh helix.     

Topological analysis of the hepatitis B virus core particle by cysteine-cysteine cross-linking.

The nucleocapsid, or core particle, of hepatitis B virus is formed by 180 subunits of the core protein, which contains Cys at positions 48, 61, 107 and 183, the latter constituting the C terminus. Upon adventitious oxidation, some or all of these cysteine residues participate in the formation of disulphide bridges, leading to polymerization of the subunits within the particle. To utilize the cysteine residues as topological probes, we reduced the number of possible intersubunit crosslinks by replacing these residues individually, or in all combinations, by serine. A corresponding set of variants was constructed within the context of an assembly-competent core protein variant that lacks the highly basic C-terminal region. Analysis, by polyacrylamide gel electrophoresis under non-reducing conditions, of the oxidative crosslinking products formed by the wild-type and mutant proteins expressed in Escherichia coli, revealed a clear distinction between the three N-proximal, and the C-terminal Cys: N-proximal Cys formed intermolecular disulphide bonds only with other N-proximal cysteine residues, leading to dimerization. Cys48 and Cys61, in contrast to Cys107, could be crosslinked to the homologous cysteine residues in a second subunit, and are therefore located at the dimer interface. Cys 183 predominantly formed disulphide bonds with Cys183 in subunits other than those crosslinked by the N-proximal cysteine residues. Hence, the polymers generated by oxidation of the wild-type protein are S-S-linked dimeric N-terminal domains interconnected via Cys183/Cys183 disulphide bonds. The intermolecular crosslinks between the N-proximal cysteine residues were apparently the same in the C-terminally truncated and in the full-length proteins, corroborating the model in which the N-terminal domain and the C terminus of the HBV core protein form two distinct and structurally independent entities. The strong tendency of the N-terminal domain for dimeric interactions suggests that core protein dimers are the major intermediates in hepatitis B virus nucleocapsid assembly.     

Identification of a ligand-binding site in the Na+/bile acid cotransporting protein from rabbit ileum.

Reabsorption of bile acids occurs in the terminal ileum by a Na(+)-dependent transport system composed of several subunits of the ileal bile acid transporter (IBAT) and the ileal lipid-binding protein. To identify the bile acid-binding site of the transporter protein IBAT, ileal brush border membrane vesicles from rabbit ileum were photoaffinity labeled with a radioactive 7-azi-derivative of cholyltaurine followed by enrichment of IBAT protein by preparative SDS gel electrophoresis. Enzymatic fragmentation with chymotrypsin yielded IBAT peptide fragments in the molecular range of 20.4-4 kDa. With epitope-specific antibodies generated against the C terminus a peptide of molecular mass of 6.6-7 kDa was identified as the smallest peptide fragment carrying both the C terminus and the covalently attached radiolabeled bile acid derivative. This clearly indicates that the ileal Na(+)/bile acid cotransporting protein IBAT contains a bile acid-binding site within the C-terminal 56-67 amino acids. Based on the seven-transmembrane domain model for IBAT, the bile acid-binding site is localized to a region containing the seventh transmembrane domain and the cytoplasmic C terminus. Alternatively, assuming the nine-transmembrane domain model, this bile acid-binding site is localized to the ninth transmembrane domain and the C terminus.     

Mass spectrometric analysis of GAP-43/neuromodulin reveals the presence of a variety of fatty acylated species

GAP-43 (neuromodulin) is a protein kinase C substrate that is abundant in developing and regenerating neurons. Thioester-linked palmitoylation at two cysteines near the GAP-43 N terminus has been implicated in directing membrane binding. Here, we use mass spectrometry to examine the stoichiometry of palmitoylation and the molecular identity of the fatty acid(s) attached to GAP-43 in vivo. GAP-43 expressed in either PC12 or COS-1 cells was acetylated at the N-terminal methionine. Approximately 35% of the N-terminal GAP-43 peptides were also modified by palmitate and/or stearate on Cys residues. Interestingly, a variety of acylated species was detected, in which one of the Cys residues was acylated by either palmitate or stearate, or both Cys residues were acylated by palmitates or stearates or a combination of palmitate and stearate. Depalmitoylation of membrane-bound GAP-43 did not release the protein from the membrane, implying that additional forces function to maintain membrane binding. Indeed, mutation of four basic residues within the N-terminal domain of GAP-43 dramatically reduced membrane localization of GAP-43 without affecting palmitoylation. These data reveal the heterogeneous nature of S-acylation in vivo and illustrate the power of mass spectrometry for identification of key regulatory protein modifications.     

Wild-type and mutant bacterioopsins D85N, D96N, and R82Q: high-level expression in Escherichia coli

The integral membrane protein bacterioopsin, found in the extremely halophilic archaebacterium Halobacterium halobium, was expressed in Escherichia coli as a fusion protein containing 13 heterologous amino acids at the amino terminus. The expressed protein was localized primarily to the E. coli cytoplasmic membrane (greater than 80%) and had an in vivo half-life of 26 min. The amount of bacterioopsin in E. coli crude lysates was quantitated immunologically from Western blots and was expressed at 10-20-fold higher levels than seen previously (i.e., 17 mg/L; 5.6% of the total protein). Three distinct forms of the protein were detected immunologically: two of the forms were generated by the removal of either one or four amino acid residues at the amino terminus; the third form remained unaltered.     

Topography of the membrane domain of the liver Na+-dependent bile acid transporter

The ileal apical and liver basolateral bile acid transporters catalyze the Na+-dependent uptake of these amphipathic molecules in the intestine and liver. They contain nine predicted helical hydrophobic sequences (H1-H9) between the exoplasmic N-glycosylated N terminus and the cytoplasmic C terminus. Previous in vitro translation and in vivo alanine insertion scanning studies gave evidence for either nine or seven transmembrane segments, with H3 and H8 noninserted in the latter model. N-terminal GFP constructs containing either successive predicted segments or only the last two domains of the liver transporter following a membrane anchor signal were expressed in HEK-293 cells, and a C-terminal glycosylation flag allowed detection of membrane insertion. Western blot analysis with anti-GFP antibody after alkali and PNGase treatment showed that H1, H2, H3 behaved as competent transmembrane (TM) sequences. Results from longer constructs were difficult to interpret. H9, however, but not H8 was membrane-inserted. To analyze the intact transporter, a C-terminal YFP fusion protein was expressed as a functionally active protein in the plasma membrane of HEK-293 cells as seen by confocal microscopy. After limited tryptic digestion to ensure the accessibility of only exoplasmic lysine or arginine residues, molecular weight (MW) analysis of the five cleavage products on SDS-PAGE predicted the presence of seven transmembrane segments, H1, H2, H3, H4, H5, H6, and H9, with H7 and H8 exoplasmic. This new method provided evidence for seven membrane segments giving a new model of the membrane domain of this protein and probably the homologous ileal transporter, with H7/H8 as the transport region.     

In vivo membrane assembly of the E.coli polytopic protein, melibiose permease, occurs via a Sec-independent process which requires the protonmotive force.

To investigate the mechanism of polytopic membrane protein insertion in Escherichia coli, we have examined the protein and energy requirements for in vivo membrane assembly of the prototypic 12 transmembrane domain sugar co-transporter, melibiose permease (MelB). MelB membrane assembly was analyzed both kinetically, by pulse labeling experiments, and functionally by measuring the activity of the inserted permease. Strikingly, the rate of MelB membrane assembly is decreased approximately 4-fold upon dissipation of the transmembrane electrochemical proton gradient, delta(mu)H+, indicative of a strong requirement for delta(mu)H+. Interestingly, selective dissipation of either the electrical (delta(psi)) or the chemical (delta(pH)) component of delta(mu)H+ demonstrates that either form of energy is required for MelB membrane assembly. In contrast, MelB membrane assembly does not require SecA, SecY or SecE, all three proteins which are strictly required for protein translocation. Neither the rate of MelB membrane assembly nor the amount of functional permease is affected by inactivation or depletion of these Sec proteins. These results strongly suggest that polytopic membrane proteins such as MelB insert into the cytoplasmic membrane by a mechanism fundamentally different from protein translocation.     

Sequence requirements for membrane assembly of polytopic membrane proteins: molecular dissection of the membrane insertion process and topogenesis of the human MDR3 P-glycoprotein.

The biogenesis of membrane proteins with a single transmembrane (TM) segment is well understood. However, understanding the biogenesis and membrane assembly of membrane proteins with multiple TM segments is still incomplete because of the complexity and diversity of polytopic membrane proteins. In an attempt to investigate further the biogenesis of polytopic membrane proteins, I used the human MDR3 P-glycoprotein (Pgp) as a model polytopic membrane protein and expressed it in a coupled cell-free translation/translocation system. I showed that the topogenesis of the C-terminal half MDR3 Pgp molecule is different from that of the N-terminal half. This observation is similar to that of the human MDR1 Pgp. The membrane insertion properties of the TM1 and TM2 in the N-terminal half molecule are different. The proper membrane anchorage of both TM1 and TM2 of the MDR3 Pgp is affected by their C-terminal amino acid sequences, whereas only the membrane insertion of the TM1 is dependent on the N-terminal amino acid sequences. The efficient membrane insertion of TM3 and TM5 of MDR3 Pgp, on the other hand, requires the presence of the putative TM4 and TM6, respectively. The TM8 in the C-terminal half does not contain an efficient stop-transfer activity. These observations suggest that the membrane insertion of putative TM segments in the human MDR3 Pgp does not simply follow the prevailing sequential event of the membrane insertion by signal-anchor and stop-transfer sequences. These results, together with my previous findings, suggest that different isoforms of Pgp can be used in comparison as a model system to understand the molecular mechanism of topogenesis of polytopic membrane proteins.     

Monoclonal antibodies against human erythrocyte band 3 protein. Localization of proteolytic cleavage sites and stilbenedisulfonate-binding lysine residues

Monoclonal antibodies against the membrane domain of human red blood cell band 3 protein have been prepared and used in topographical studies of the arrangement of the polypeptide in the membrane. One of the antibodies binds to a site near the N terminus of the membrane domain; another binds to a site near the C terminus. The latter has been used to localize a site of intracellular trypsin digestion. The cleavage site, in human band 3, corresponds to Lys-761 in mouse band 3; the site is 168 residues from the C terminus of the protein. This is the first intracellular site in the membrane domain (other than the N terminus) that has been localized in the primary structure. The antibody that binds to the N-terminal portion of the membrane domain has been used to identify a new S-cyanylation cleavage site about 7,000 daltons from the C terminus. Proteolysis/cross-linking experiments with the stilbenedisulfonate derivative H2DIDS (4,4'-diisothiocyanostilbene-2,2'-disulfonate) reveal that one end of the H2DIDS reacts covalently with a lysine residue that is between about 70 and 168 residues from the C terminus of band 3. In addition to placing restrictions on the location of the H2DIDS-binding lysine, these studies provide direct evidence that the C-terminal 28,000-dalton papain fragment crosses the membrane at least three times. With previous data on the remainder of the membrane domain, there is now direct evidence that the band 3 polypeptide crosses the membrane at least eight times.     

Structure of the soluble domain of a membrane-anchored thioredoxin-like protein from Bradyrhizobium japonicum reveals unusual properties

TlpA is an unusual thioredoxin-like protein present in the nitrogen-fixing soil bacterium Bradyrhizobium japonicum. A hydrophobic N-terminal transmembrane domain anchors it to the cytoplasmic membrane, whereby the hydrophilic thioredoxin domain becomes exposed to the periplasmic space. There, TlpA catalyses an essential reaction, probably a reduction, in the biogenesis of cytochrome aa(3). The soluble thioredoxin domain (TlpA(sol)), devoid of the membrane anchor, was purified and crystallized. Oxidized TlpA(sol) crystallized as a non-covalent dimer in the space group P2(1)2(1)2(1). The X-ray structure analysis was carried out by isomorphous replacement using a xenon derivative. This resulted in a high-resolution (1.6 A) three-dimensional structure that displayed all of the features of a classical thioredoxin fold. A number of peculiar structural details were uncovered: (i) Only one of the two active-site-cysteine sulphurs (Cys72, the one closer to the N terminus) is exposed on the surface, making it the likely nucleophile for the reduction of target proteins. (ii) TlpA(sol) possesses a unique structural disulphide bond, formed between Cys10 and Cys155, which connects an unprecedented N-terminal alpha helix with a beta sheet near the C terminus. (iii) An insertion of about 25 amino acid residues, not found in the thioredoxin prototype of Escherichia coli, contributes only marginally to the thioredoxin fold, but forms an extra, surface-exposed alpha helix. This region plus another surface-exposed stretch (-Ile-Gly-Arg-Ala-), which is absent even in the closest TlpA relatives, might be considered as specificity determinants for the recognition of target proteins in the periplasm. The TlpA(sol) structure paves the way towards unraveling important structure-function relationships by rational mutagenesis.     

The elastin-binding protein of Staphylococcus aureus (EbpS) is expressed at the cell surface as an integral membrane protein and not as a cell wall-associated protein.

The elastin-binding proteins EbpS of Staphylococcus aureus strains Cowan and 8325-4 were predicted from sequence analysis to comprise 486 residues. Specific antibodies were raised against an N-terminal domain (residues 1-267) and a C-terminal domain (residues 343-486) expressed as recombinant proteins in Escherichia coli. Western blotting of lysates of wild-type 8325-4 and Newman and the corresponding ebpS mutants showed that EbpS migrated with an apparent molecular mass of 83 kDa. The protein was found exclusively in cytoplasmic membrane fractions purified from protoplasts or lysed cells, in contrast to the clumping factor ClfA, which was cell-wall-associated. EbpS was predicted to have three hydrophobic domains H1-(205-224), H2-(265-280), and H3-(315-342). A series of hybrid proteins was formed between EbpS at the N terminus and either alkaline phosphatase or beta-galactosidase at the C terminus (EbpS-PhoA, EbpS-LacZ). PhoA and LacZ were fused to EbpS between hydrophobic domains H1-H2 and H2-H3, and distal to H3. Expression of enzymatic activity in E. coli showed that EbpS is an integral membrane protein with two membrane-spanning domains H1 and H3. N-terminal residues 1-205 and C-terminal residues 343-486 were predicted to be exposed on the outer face of the cytoplasmic membrane. The ligand-binding domain of EbpS is known from previous studies to be present in the N terminus between residues 14-34 and probing whole cells with anti-EbpS1-267 antibodies indicated that this region is exposed on the surface of intact cells. This was also confirmed by the observation that wild-type S. aureus Newman cells bound labeled tropoelastin whereas the ebpS mutant bound 72% less. In contrast, the C terminus, which carries a putative LysM peptidoglycan-binding domain, is not exposed on the surface of intact cells and presumably remains buried within the peptidoglycan. Finally, expression of EbpS was correlated with the ability of cells to grow to a higher density in liquid culture, suggesting that EbpS may have a role in regulating cell growth.     

In vitro synthesis of lactose permease to probe the mechanism of membrane insertion and folding

Insertion and folding of polytopic membrane proteins is an important unsolved biological problem. To study this issue, lactose permease, a membrane transport protein from Escherichia coli, is transcribed, translated, and inserted into inside-out membrane vesicles in vitro. The protein is in a native conformation as judged by sensitivity to protease, binding of a monoclonal antibody directed against a conformational epitope, and importantly, by functional assays. By exploiting this system it is possible to express the N-terminal six helices of the permease (N(6)) and probe changes in conformation during insertion into the membrane. Specifically, when N(6) remains attached to the ribosome it is readily extracted from the membrane with urea, whereas after release from the ribosome or translation of additional helices, those polypeptides are not urea extractable. Furthermore, the accessibility of an engineered Factor Xa site to Xa protease is reduced significantly when N(6) is released from the ribosome or more helices are translated. Finally, spontaneous disulfide formation between Cys residues at positions 126 (Helix IV) and 144 (Helix V) is observed when N(6) is released from the ribosome and inserted into the membrane. Moreover, in contrast to full-length permease, N(6) is degraded by FtsH protease in vivo, and N(6) with a single Cys residue at position 148 does not react with N-ethylmaleimide. Taken together, the findings indicate that N(6) remains in a hydrophilic environment until it is released from the ribosome or additional helices are translated and continues to fold into a quasi-native conformation after insertion into the bilayer. Furthermore, there is synergism between N(6) and the C-terminal half of permease during assembly, as opposed to assembly of the two halves as independent domains.     

Transmembrane helix-helix association: relative stabilities at low pH

We have previously studied the unfolding equilibrium of bacterioopsin in a single phase solvent, using F?rster mechanism fluorescence resonance energy transfer (FRET) as a probe, from tryptophan donors to a dansyl acceptor. We observed an apparent unfolding transition in bacterioopsin perturbed by increasing ethanol concentrations [Nannepaga et al. (2004) Biochemistry 43, 50-59]. We have further investigated this transition and find that the unfolding is pH-dependent. We have now measured the apparent pK of acid-induced unfolding of bacterioopsin in 90% ethanol. When the acceptor is on helix B (Lys 41), the apparent pK for unfolding is 4.75; on the EF connecting loop (Cys 163), 5.15; and on helix G (Cys 222), 5.75. Five-helix proteolytic fragments are less stable. The apparent unfolding pKs are 5.46 for residues 72-248 (Cys 163) and 7.36 for residues 1-166 (Lys 41). When interpreted in terms of a simple equilibrium model for unfolding, the apparent pKs give relative free energies of unfolding in the range of -0.54 to -3.5 kcal/mol. The results suggest that the C-terminal helix of bacterioopsin is less stably folded than the N-terminal helices. We analyzed the pairwise helix-helix interaction surfaces of bacteriorhodopsin and three other seven-transmembrane-helix proteins on the basis of crystal structures. The results show that the interaction surfaces are smoother and the helix axis separations are closer in the amino-terminal two-thirds of the proteins compared with the carboxyl-terminal one-third. However, the F helix is important in stabilizing the folded structure, as shown by the instability of the 1-166 fragment. Considering the high-resolution crystal structure of bacteriorhodopsin, there are no obvious helix-helix interactions involving protein side chains which would be destabilized by protonation at the estimated pH of the unfolding transitions. However, a number of helix-bridging water molecules could become protonated, thereby weakening the helix-helix interactions.     

Assignment of the four disulfides in the N-terminal somatomedin B domain of native vitronectin isolated from human plasma

The primary sequence of the N-terminal somatomedin B (SMB) domain of native vitronectin contains 44 amino acids, including a framework of four disulfide bonds formed by 8 closely spaced cysteines in sequence patterns similar to those found in the cystine knot family of proteins. The SMB domain of vitronectin was isolated by digesting the protein with endoproteinase Glu-C and purifying the N-terminal 1-55 peptide by reverse-phase high performance liquid chromatography. Through a combination of techniques, including stepwise reduction and alkylation at acidic pH, peptide mapping with matrix-assisted laser desorption ionization mass spectrometry and NMR, the disulfide bonds contained in the SMB domain have been determined to be Cys(5):Cys(9), Cys(19):Cys(31), Cys(21):Cys(32), and Cys(25):Cys(39). This pattern of disulfides differs from two other connectivities that have been reported previously for recombinant forms of the SMB domain expressed in Escherichia coli. This arrangement of disulfide bonds in the SMB domain from native vitronectin forms a rigid core around the Cys(19): Cys(31) and Cys(21):Cys(32) disulfides. A small positively charged loop is created at the N terminus by the Cys(5): Cys(9) cystine. The most prominent feature of this disulfide-bonding pattern is a loop between Cys(25) and Cys(39) similar to cystine-stabilized alpha-helical structures commonly observed in cystine knots. This alpha-helix has been confirmed in the solution structure determined for this domain using NMR (Mayasundari, A., Whittemore, N. A., Serpersu, E. H., and Peterson, C. B. (2004) J. Biol. Chem. 279, 29359-29366). It confers function on the SMB domain, comprising the site for binding to plasminogen activator inhibitor type-1 and the urokinase receptor.     

Membrane insertion of uracil permease, a polytopic yeast plasma membrane protein.

Uracil permease is a multispanning protein of the Saccharomyces cerevisiae plasma membrane which is encoded by the FUR4 gene and produced in limited amounts. It has a long N-terminal hydrophilic segment, which is followed by 10 to 12 putative transmembrane segments, and a hydrophilic C terminus. The protein carries seven potential N-linked glycosylation sites, three of which are in its N-terminal segment. Overexpression of this permease and specific antibodies were used to show that uracil permease undergoes neither N-linked glycosylation nor proteolytic processing. Uracil permease N-terminal segments of increasing lengths were fused to a reporter glycoprotein, acid phosphatase. The in vitro and in vivo fates of the resulting hybrid proteins were analyzed to identify the first signal anchor sequence of the permease and demonstrate the cytosolic orientation of its N-terminal hydrophilic sequence. In vivo insertion of the hybrid protein bearing the first signal anchor sequence of uracil permease into the endoplasmic reticulum membrane was severely blocked in sec61 and sec62 translocation mutants.     

Topology and endoplasmic reticulum retention signals of the lysosomal storage disease-related membrane protein CLN6

CLN6 is a polytopic membrane protein of unknown function resident in the endoplasmic reticulum (ER). Mutant CLN6 causes the lysosomal storage disorder neuronal ceroid lipofuscinosis. Defining the topology of CLN6, and the structural domains and motifs required for interaction with cytosolic and luminal proteins may allow insights into its function. In this study we analysed the topology, ER retention and oligomerization of CLN6. We demonstrated, by differential membrane permeabilization of transfected BHK cells using specific detergents and two distinct antibodies, that CLN6 contains an N-terminal cytoplasmic domain, seven transmembrane domains, and a luminal C terminus. Mutational analyses and confocal immunofluorescence microscopy showed that changes of potential ER localization signals in the N- or C-terminal domain (a triple arginine cluster, and a dileucine motif) did not alter the subcellular localization of CLN6. The deletion of a dilysine motif impaired partially the ER localization of CLN6. Furthermore, expression analyses of fusion and deletion constructs in non-neuronal and neuronal cells suggested that two portions of CLN6 contributed to its retention within the ER. We showed that the N-terminal domain was necessary but not sufficient for ER retention of CLN6 and that deletion of transmembrane domains 6 and 7 was accompanied with the loss of ER localization and, in some instances, trafficking to the cisGolgi. From these data we concluded that CLN6 maintains its ER localization by expressing retention signals present in both the N-terminal cytosolic domain and in the carboxy-proximal transmembrane domains 6 and 7. Additionally, the ability of CLN6 to homodimerize may also prevent exit from the ER via an interaction with membrane-associated factors.