- Email: [email protected]
Lysine Can Be Replaced by Histidine but Not by Arginine as the ER Retrieval Motif for Type I Membrane Proteins
Biochemical and Biophysical Research Communications 291, 751–757 (2002) doi:10.1006/bbrc.2002.6515, available online at http://www.idealibrary.com on
Lysine Can Be Replaced by Histidine but Not by Arginine as the ER Retrieval Motif for Type I Membrane Proteins Birgit Hardt and Ernst Bause 1 Institut fu¨r Physiologische Chemie, Universita¨t Bonn, Nussallee 11, 53115 Bonn, Germany
Received January 30, 2002
The OST48 subunit of the oligosaccharyltransferase complex is a type I membrane protein containing three lysines in its cytosolic domain. The two lysines in positions 3 and 5 from the C-terminus are able to direct protein localisation within the endoplasmic reticulum (ER) by COPI-mediated retrieval. Substitution of these lysines by arginine resulted in cell-surface expression of OST48, whereas ER residency was maintained when either Lys-5 or Lys-3 but not both was replaced with arginine. Localisation of OST48 was not affected by substitution of the two lysines by histidine, indicating that a His-Xaa-His sequence, in contrast to Arg-Xaa-Arg, contains ER-specific targeting information. These differences show that simple charge interactions are not sufficient for ER retention and that other structural factors also play a role. The His-Xaa-His sequence could represent a new and independent signal for directing ER localisation differing from both the arginine motif in type II proteins and the lysine motif in type I proteins. Our data do not exclude, however, that the histidine sequence simply mimicks the lysine motif as a sorting signal, being recognised by and interacting with the same receptor subunit(s) in COP-I-coated vesicles. Conclusions arising from this assumption involving the conformation of lysine at the putative COP-I binding site and the failure of Arg-Xaa-Arg to mediate ER localisation for type I proteins are discussed. © 2002 Elsevier Science (USA) Key Words: OST48; type I membrane protein; ER localisation; lysine motif; histidine motif; arginine motif.
Soluble and membrane proteins are retained in the endoplasmic reticulum (ER) by means of defined sequence motifs in their polypeptide chain, causing accumulation and seggregation from proteins destined for other subcellular compartments or the secretory pathway [1, 2]. One of the best characterised sorting signals mediating ER-residency is the KDEL sequence at the 1 To whom correspondence and reprint requests should be addressed. Fax: 0228-73 2416. E-mail: [email protected].
C-terminus of soluble proteins [3, 4]. Localisation of type I membrane proteins to the ER, on the other hand, is conferred by a double-lysine motif close to the cytosolically exposed C-terminus, whereas a double-arginine sequence close to the cytosolic N-terminus appears to be functional in ER retention of type II membrane proteins [5, 6]. The various sequence motifs direct ER localisation through a retrieval mechanism involving interaction with subunits of a multimeric coat-protein complex (COP-I), thereby initiating vesicle formation which leads to retrograde transport of the protein cargo from preGolgi compartments back to the ER [1, 2, 6 – 8]. In addition to retrieval, localisation of proteins to the ER may also occur by direct retention mechanisms operating in the organelle itself which are, however, as yet poorly characterised [9, 10]. Oligosaccharyltransferase (OST) is a hetero-oligomeric protein complex of the ER membrane catalysing the en bloc transfer of dolichyl-PP-activated oligosaccharides onto specific asparagine residues of the nascent polypeptide chain [11]. OST48, one of the subunits of the enzyme complex and probably responsible for catalysis, is a typical type I membrane protein consisting of a large luminal domain, followed by a 20 amino acid hydrophobic sequence, which functions as membrane anchor, and a 9 amino acid long peptide tail directed toward the cytosol [12, 13]. The short cytosolic peptide is highly polar and contains three lysine residues in positions 3, 5 and 7 relative to the C-terminus. Recent studies with hybrid proteins containing distinct domains of OST48 and Golgilocated Man 9-mannosidase have shown that the lysines at positions 5 and 3 of the cytosolic domain of OST48 function as a structural motif conferring ER residency by a retrograde transport pathway [10]. In this paper we describe impaired localisation of OST48 to the ER by substitution of both Lys-5 and Lys-3 with arginine but not when replaced by two histidines. MATERIALS AND METHODS Materials. Synthetic oligonucleotides, Taq DNA polymerase, Dulbecco’s Modified Eagle Medium (DMEM), fetal bovine serum, pSV.SPORT1 (Invitrogen); restriction endonucleases (MBI Fermentas);
751
0006-291X/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
DEAE dextran (Amersham Bioscience); COS 1 cells (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH); goat anti-rabbit IgG alkaline phosphatase conjugate (Sigma); fluorescein (DTAF)-conjugated AffiniPure goat anti-rabbit IgG (Dianova); nitrocellulose membranes (Schleicher & Schuell); ABI PRISM Big dye terminator cycle sequencing kit (Applied Biosystems). All other compounds were of analytical grade purity. Vector construction. The full-length cDNA encoding for pig liver OST48 was subcloned into the mammalian expression vector pSV.SPORT1 taking advantage of common restriction sites [13]. Amino acid exchanges in the C-terminal domain of OST48 were introduced by in vitro oligonucleotide-directed mutagenesis using the OST48-specific cDNA as the template [14]. For PCR amplification the following OST48-derived oligonucleotides were used as sense (s) and anti-sense (as) primers:
FIG. 1. Schematic representation of the OST48 domain structure with the cytosolic portion shown in full. Amino acid exchanges are shown shaded.
RESULTS AND DISCUSSION
Exchanges are shown in grey. The introduction of base mutations was confirmed by DNA sequencing. Cell culture and transfection. COS 1 cells were grown at 37°C under 5% CO2 as a monolayer culture in DMEM supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 g/ml streptomycin. After having reached ⬃60% confluency COS 1 cells were transfected with vector constructs encoding for wild-type OST48 and the OST48-mutants using the DEAE-dextran/chloroquine method [15]. Immunofluorescence microscopy. Transfected COS 1 cells were grown for 36 h on sterile coverslips. The cells were either fixed with 8% formaldehyde, or fixed and then permeabilised with 0.2% Triton X-100 in 200 mM Hepes, pH 7.2, for 5 min, followed by incubation in a 1:30 dilution of an affinity-purified anti-OST48-antibody in PBS containing 0.1% ovalbumin [13]. After 12 h at 4°C, antigen/antibody complexes were labeled by treatment at 37°C for 1 h with a 1:100 dilution of a DTAF-conjugated goat anti-rabbit IgG antibody and the cells were processed as previously described [10, 16]. General methods. SDS–PAGE and immunoblotting were performed as described previously [17–19]. PCR amplification, vector construction and other molecular biological techniques were performed as detailed in [14, 15, 20, 21]. Nucleotide sequencing was done according to Sanger by using the Big Dye terminator cycle sequencing kit [22].
Replacement of Both Lysines in the Cytosolic Lys-Xaa-Lys Triplet by Arginine Causes Cell Surface Expression of OST48 To characterise the functional specificity of the LysXaa-Lys motif for ER targeting, a range of OST48 mutants were synthesized in which the lysine residues in the cytosolic domain were replaced by leucine, arginine and histidine (Fig. 1). The amino acids were exchanged by in vitro directed mutagenesis using OST48 full-length cDNA subcloned into pSV.SPORT1. COS 1 cells were then transfected with the corresponding cDNAs, followed by characterisation of the overexpressed proteins by immunoblotting and determination of their subcellular localisation by indirect immunofluorescence microscopy using a monospecific polyclonal antiOST48-antibody for detection [13]. As shown in Fig. 2, COS 1 cells transfected with the vector encoding for wild-type OST48, efficiently overexpressed a polypeptide whose molecular mass (⬃48 kDa) corresponded to that calculated from the ORF in the OST48-specific cDNA (lane 2). No immunoreactive
752
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Immunoblot analysis of wild-type OST48 and OST48 mutants. COS 1 cells transfected with the various vector constructs were solubilised and aliquots of the detergent extracts containing equal amounts of cell protein, were subjected to SDS–PAGE, followed by immunoblotting using a polyclonal anti-OST48 antibody for staining [13]. Lane 1, pSV.SPORT1-transfected cells (control); lane 2, wild-type OST48; lane 3, OST48-R3; lane 4, OST48-R5/R3; lane 5, OST48-H5/H3; lane 6, OST48-H5/L3.
material was detectable on Western blots prepared from detergent extracts of COS 1 cells transfected with pSV.SPORT1, except for a minor band at a lower molecular mass (lane 1). This indicates that the level of endogeneous OST48 is rather low. Immunofluorescence analysis of transfected COS 1 cells permeabilised with Triton X-100 gave a strong labeling of intracellular structures typical for the ER (Fig. 3A), whereas nonpermeabilised cells remained unstained (Fig. 3B). Thus OST48 is expressed as an ER-resident protein. A similar ER-specific fluorescence pattern was obtained for the OST48-L7 mutant protein containing leucine in place of Lys-7 (not shown). By contrast, nonpermeabilised cells overexpressing either OST48-L3 (Fig. 3D) or OST48-L5 exerted an intense fluorescence staining at the cell surface. Thus these two protein mutants, unlike wild-type OST48 and OST48-L7, are transported to the plasma membrane, consistent with the Lys-5/Lys-3 sequence directing ER-localisation to OST48 [10]. In addition to cell-surface labeling, ERand Golgi-specific structures were also found to be stained when the cells were permeabilised (Fig. 3C). A reasonable explanation for this observation is that the capacity of protein transport may be limited, becoming overloaded due to high over-expression. A similar intracellular staining pattern was also seen for other OST48 mutants even though they were over-expressed as cell surface proteins (see below). Lysine and arginine contain a side chain which is protonated at intracellular pH. While the distance between positive charge and ␣-C atom is comparable, the -amino and ␦-guanido groups differ in size and structure. To examine whether arginine can mimick the function of Lys-5 and Lys-3 in ER-targeting, the two lysine residues were replaced stepwise by arginine (Fig. 1). As expected, the molecular masses of OST48-R3 and OST48-R5, as well as of the OST48R5/R3 double mutant over-expressed in COS 1 cells, were identical with that of wild-type OST48 (shown for OST48-R3 and OST48-R5/R3 in Fig. 2, lanes 3 and 4). Fluorescence analysis of COS 1 cells transfected with the OST48-R3- and OST48-R5-specific cDNA revealed ER typical staining, whereas nonpermeabilised cells remained unstained (shown for OST48-R3 in Figs. 3E
and 3F). By contrast, an intense cell surface fluorescence was seen on non-permeabilised cells overexpressing the OST48-R5/R3 double arginine mutant (Fig. 3H). These observations show that the ER-localisation information residing in the Lys-Xaa-Lys motif of OST48 is not affected by replacing either Lys-5 or Lys-3 but impaired after substitution of both lysines by arginines. Histidine and Lysine Cannot Be Distinguished in as Far as ER Localisation Is Concerned The positive charges in the lysine motif are assumed to play a key role in recognition and binding to the COP-I complex. Since the pK S (⬃6.5) of the histidine imidazol ring would suggest only partial protonation at intracellular pH, we replaced Lys-5 and/or Lys-3 in the lysine signal by histidine in order to see whether ER localisation of the OST48 polypeptide was affected. The immunoblot analysis shows for the OST48-H5/H3 protein (Fig. 2, lane 5) that the His-Xaa-His double mutant is overexpressed in COS 1 cells as a ⬃48 kDa protein. Similar results were obtained for the monosubstituted OST48-H5 and OST48-H3 mutants (not shown). Permeabilised COS 1 cells transfected with the OST48-H5, OST48-H3- and OST48-H5/H3-specific cDNA exhibited a fluorescence staining pattern specific for ER elements. By contrast, no staining was detectable at the cell surface of nonpermeabilised cells. This indicates that, independently of whether either Lys-5, Lys-3 (Figs. 4A and 4B) or both lysines (Figs. 4C and 4D) were replaced by histidine, these protein mutants are ER resident. Thus not only one but two histidine residues are able to replace lysine in conferring ER residency on OST48, in apparent contrast to the ArgXaa-Arg sequence (Scheme 1). Based on the observation that one lysine residue in the Lys-Xaa-Lys motif could be substituted by arginine without loss ER-localisation information, histidine in either position 5 or 3 of the His-Xaa-His sequence was exchanged against arginine (Fig. 1; Scheme 1). Immunofluorescence studies with permeabilised cells showed that OST48-R5/H3 was still overexpressed as an ER-resident protein, whereas the inverted OST48H5/R3 mutant containing arginine in the 3-position was detected at the cell surface (Fig. 4F). This suggests that the side chain of the amino acid in the 3-position is more important for recognition by COP-I components, apparently tolerating less structural variations than in the 5-position of the sequence motif. As observed for the lysine motif, substitution of either His-5 or His-3 by leucine (Fig. 1) resulted in loss of ER residency, consistent with and supporting the specific and independent function of the His-Xaa-His sequence as sorting signal for type I proteins (Fig. 2, lane 6; Fig. 4H).
753
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. ER retention of OST48 is impaired by replacing Lys-5 and Lys-3 with arginine. COS 1 cells transfected with the cDNA encoding for either wild-type OST48 (A, B), OST-L3 (C, D), OST48-R3 (E, F), or OST48-R5/R3 (G, H) were either fixed with formaldehyde (B, D, F, H), or fixed and then permeabilised with Triton X-100 (A, C, E, G). Immunofluorescence staining was carried out using a polyclonal rabbit anti-OST48 antibody, followed by labeling of the antigen–antibody complexes with a goat anti-rabbit-IgG antibody tagged with fluorescein (DTAF) [10, 13].
CONCLUSIONS It is well established that two lysine residues located in positions 3 and 4/5 relative to the C-terminus represent a minimum sequence requirement for ER retention of type I membrane proteins. On the other hand, ER residency of type II proteins is conferred by double arginines in the N-terminal domain. Both ERtargeting motifs are thought to operate via a Golgi to
ER retrograde mechanism involving COP-I-coated vesicles [1, 2, 5, 6]. The results described in this paper show that ER-localisation information encoded for by lysine in positions 5 and 3 of the cytosolic peptide domain of OST48, is destroyed after both have been replaced by arginine. This indicates that simple charge interactions are not sufficient for signal function and that additional factors residing in the lysine side chain and/or in the amino acid sequence adjacent to the
754
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. The His-Xaa-His motif confers ER-localisation on OST48. Transfected COS 1 cells overexpressing either OST48-H3 (A, B), OST48-H5/H3 (C, D), OST48-H5/R3 (E, F) or OST48-H5/L3 (G, H) were either fixed with formaldehyde (B, D, F, H) or fixed and permeabilised (A, C, E, G) and processed as described in the legend to Fig. 3.
double-lysine motif are involved in directing ER localisation. Since the distance from the peptide backbone of the positive charge in arginine and lysine is comparable, it is reasonable to assume that loss of ERresidency as observed for the OST48-R5/R3 double mutant may be caused by steric factors introduced by the ␦-guanido group of arginine, which is larger in size than the -amino group of lysine. This may prevent recognition of the Arg-Xaa-Arg sequence by lysinespecific subunits of the COP-I complex, consistent with the current view that retrograde transport of type I
and type II membrane proteins occurs via distinct pathways involving different populations of COP-Icoated vesicles [1, 2, 7]. Whereas the OST48-R5/R3 double mutant is expressed in COS 1 cells as a cell-surface protein, arginine can replace either Lys-5 or Lys-3 without affecting ER localisation. This implies that the sorting signal is functional so long as one position is still occupied by a lysine residue. A reasonable explanation for this observation is that the binding site for the double-lysine motif in the COP-I complex displays some flexibility,
755
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
after being bound to coatomer subunits. This would be consistent with the observation that the Arg-Xaa-Arg motif lacks ER-targeting information. Comparing the side chain structure of arginine with that of lysine or histidine, the failure to be recognised by lysine- and/or histine-specific COP-I binding proteins in spite of being positively charged could result from, and thus be explained by, structural constraints imposed by one of the two NH 2 functions in the ␦-guanido group, with one of them occupying the position equivalent to the C 5methylene function in the imidazol ring being tolerated.
SCHEME 1
allowing for compensation of the structural constraints imposed by one, but not by two, of the bulky ␦-guanido groups. The observation that both the OST48-R5 and the OST48-R3 hybrid mutants are expressed as ER resident proteins, contradicts data previously described by Shin et al. [23] who suggested that Lys-3 in the double-lysine motif but not Lys-5 should be functionally invariant. This apparent discrepancy points up that the binding specificity may depend on the nature of the amino acids adjacent to the sequence motif. Our data show clearly that, in contrast to the ArgXaa-Arg sequence and in contrast to studies reported elsewhere in the literature [23, 24], two histidines may replace the lysines in ER targeting, at least functionally. This is surprising because the histidine imidazol group (pK S ⬃6.5) (i) is approximately 10% protonated at intracellular pH; (ii) like arginine has a positive charge which is delocalised; and (iii) the planar imidazol ring is rigid compared to the lysine or arginine side chain. Thus it appears reasonable to postulate that the His-Xaa-His sequence may represent a differential ERlocalisation motif interacting with COP-I subunits, differing from those which specifically bind the lysine motif in type I and the arginine signal in type II proteins. Assuming that the histidine and lysine motifs interact with the same receptor proteins in COP-Icoated vesicles, several observations may be made in terms of signal binding site specificty: (i) the side chain of histidine must be protonated since charge interactions, although apparently not sufficient, are likely to play a key role in signal recognition; (ii) since the distance from the polypeptide backbone to the N 1 nitrogen in imidazol is comparable to that for the -amino group in lysine, binding to COP-I structures may stabilize a putative positive charge in this positon; (iii) the methylene group at C 5 of the planar imidazol ring appears not to interfere with or prevent signal recognition, suggesting that this region is not in direct contact with binding protein(s) in the COP-I complex. Taking into account the structural properties of the histidine side chain, including its rigidity, planarity and charge distribution, it is possible that the imidazol ring mimicks the side chain conformation for lysine
ACKNOWLEDGMENTS This work was supported by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 284). The authors are indebted to Dr. R. A. Klein (Universita¨ t Bonn) for critical reading of the manuscript.
REFERENCES 1. Teasdale, R. D., and Jackson M. R. (1996) Signal-mediated sorting of membrane proteins between the endoplasmic reticulum and the Golgi apparatus. Annu. Rev. Cell Dev. Biol. 12, 27–54. 2. Wieland, F., and Harter, C. (1999) Mechanisms of vesicle formation: Insights from the COP system. Curr. Opin. Cell Biol. 11, 440 – 446. 3. Lewis, M. J., and Pelham, H. R. (1992) Ligand-induced redistribution of a human KDEL receptor from the Golgi complex to the endoplasmic reticulum. Cell 68, 353–364. 4. Yamamoto, K., Fujii, R., Toyofuka, Y., Saito, T., Koseki, H., Hsu, V. W., and Aoe, T. (2001) The KDEL receptor mediates a retrieval mechanism that contributes to quality control at the endoplasmic reticulum. EMBO J. 20, 3082–3091. 5. Cosson, P., and Letourneur, F. (1994) Coatomer interaction with di-lysine endoplasmic reticulum retention motifs. Science 263, 1629 –1631. 6. Schutze, M.-P., Peterson, P. A., and Jackson, M. R. (1994) An N-terminal double-arginine motif maintains type II membrane proteins in the endoplasmic reticulum. EMBO J. 13, 1696 –1705. 7. Orci, L., Stamnes, M., Ravazzola, M., Amherdt, M., Perrelet, A., So¨ llner, T. H., and Rothman, J. E. (1997) Bidirectional transport by distinct populations of COPI-coated vesicles. Cell 90, 335– 349. 8. Cosson, P., and Letourneur, F. (1997) Coatomer (COPI)-coated vesicles: role in intracellular transport and protein sorting. Curr. Opin. Cell Biol. 9, 484 – 487. 9. Fu, J., and Kreibich, G. (2000) Retention of subunits of the oligosaccharyltransferase complex in the endoplasmic reticulum. J. Biol. Chem. 275, 3984 –3990. 10. Hardt, B., Aparicio, R., and Bause, E. (2001) Analysis of structural signals conferring localisation of pig OST48 to the endoplasmic reticulum. Biol. Chem. 382, 1039 –1047. 11. Kornfeld, R., and Kornfeld, S. (1985) Assembly of asparaginelinked oligosaccharides. Annu. Rev. Biochem. 54, 631– 664. 12. Silberstein, S., Kelleher, D. J., and Gilmore, R. (1992) The 48kDa subunit of the mammalian oligosaccharyltransferase complex is homologous to the essential yeast protein WBP1. J. Biol. Chem. 267, 23658 –23663. 13. Hardt, B., Aparicio, R., and Bause, E. (2000)The oligosaccharyl-
756
Vol. 291, No. 4, 2002
14.
15.
16.
17. 18.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
transferase complex from pig liver: cDNA cloning, expression and functional characterisation. Glycoconj. J. 17, 767–779. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D. Seidman, J. G., Smith, J. A., and Struhl, K. (1987) Current Protocols in Molecular Biology, Wiley, New York. Sambrook, J., Fritsch, E. F., and Maniatis T. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Bieberich, E., and Bause, E. (1995) Man 9-mannosidase from human kidney is expressed in COS cells as a Golgi resident type II transmembrane N-glycoprotein. Eur. J. Biochem. 233, 644 – 649. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head bacteriophage T4. Nature 227, 680 – 685. Blake, M. S., Johnson, K. H., Russel-Jones, G. J., and Gotschlich, E. C. (1984) A rapid, sensitive method for detection of alkaline phosphatase-conjugated anti-antibody on Western blots. Anal. Biochem. 136, 175–179.
19. Knecht, D. A., and Dimond, R. L. (1984) Visualization of antigenic proteins on Western blots. Anal. Biochem. 136, 180 –184. 20. Sakai, K., Gelfand, D. H., Stoffels, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis, K. B., and Ehrlich, H. A. (1988) Primerdirected enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487– 491. 21. White, B. A. (1993) PCR-Protocols—Current Methods and Applications, Humana Press, Ottawa, Canada. 22. Sanger, F., Nicklen, S., and Coulson A. R. (1977) DNA sequencing with chain termination inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467. 23. Shin, J., Dunbrack, Jr., R. L., Lee, S., and Strominger, J. L. (1991) Signals for retention of transmembrane proteins in the endoplasmic reticulum studied with CD4 truncation mutants. Proc. Natl. Acad. Sci. USA 88, 1918 –1922. 24. Jackson, M. R., Nilsson, T., and Peterson, P. A. (1990). Identification of a consensus motif for retention of transmembrane proteins in the endoplasmic reticulum. EMBO J. 9, 3153–3162.
757
Lysine Can Be Replaced by Histidine but Not by Arginine as the ER Retrieval Motif for Type I Membrane Proteins Birgit Hardt and Ernst Bause 1 Institut fu¨r Physiologische Chemie, Universita¨t Bonn, Nussallee 11, 53115 Bonn, Germany
Received January 30, 2002
The OST48 subunit of the oligosaccharyltransferase complex is a type I membrane protein containing three lysines in its cytosolic domain. The two lysines in positions 3 and 5 from the C-terminus are able to direct protein localisation within the endoplasmic reticulum (ER) by COPI-mediated retrieval. Substitution of these lysines by arginine resulted in cell-surface expression of OST48, whereas ER residency was maintained when either Lys-5 or Lys-3 but not both was replaced with arginine. Localisation of OST48 was not affected by substitution of the two lysines by histidine, indicating that a His-Xaa-His sequence, in contrast to Arg-Xaa-Arg, contains ER-specific targeting information. These differences show that simple charge interactions are not sufficient for ER retention and that other structural factors also play a role. The His-Xaa-His sequence could represent a new and independent signal for directing ER localisation differing from both the arginine motif in type II proteins and the lysine motif in type I proteins. Our data do not exclude, however, that the histidine sequence simply mimicks the lysine motif as a sorting signal, being recognised by and interacting with the same receptor subunit(s) in COP-I-coated vesicles. Conclusions arising from this assumption involving the conformation of lysine at the putative COP-I binding site and the failure of Arg-Xaa-Arg to mediate ER localisation for type I proteins are discussed. © 2002 Elsevier Science (USA) Key Words: OST48; type I membrane protein; ER localisation; lysine motif; histidine motif; arginine motif.
Soluble and membrane proteins are retained in the endoplasmic reticulum (ER) by means of defined sequence motifs in their polypeptide chain, causing accumulation and seggregation from proteins destined for other subcellular compartments or the secretory pathway [1, 2]. One of the best characterised sorting signals mediating ER-residency is the KDEL sequence at the 1 To whom correspondence and reprint requests should be addressed. Fax: 0228-73 2416. E-mail: [email protected].
C-terminus of soluble proteins [3, 4]. Localisation of type I membrane proteins to the ER, on the other hand, is conferred by a double-lysine motif close to the cytosolically exposed C-terminus, whereas a double-arginine sequence close to the cytosolic N-terminus appears to be functional in ER retention of type II membrane proteins [5, 6]. The various sequence motifs direct ER localisation through a retrieval mechanism involving interaction with subunits of a multimeric coat-protein complex (COP-I), thereby initiating vesicle formation which leads to retrograde transport of the protein cargo from preGolgi compartments back to the ER [1, 2, 6 – 8]. In addition to retrieval, localisation of proteins to the ER may also occur by direct retention mechanisms operating in the organelle itself which are, however, as yet poorly characterised [9, 10]. Oligosaccharyltransferase (OST) is a hetero-oligomeric protein complex of the ER membrane catalysing the en bloc transfer of dolichyl-PP-activated oligosaccharides onto specific asparagine residues of the nascent polypeptide chain [11]. OST48, one of the subunits of the enzyme complex and probably responsible for catalysis, is a typical type I membrane protein consisting of a large luminal domain, followed by a 20 amino acid hydrophobic sequence, which functions as membrane anchor, and a 9 amino acid long peptide tail directed toward the cytosol [12, 13]. The short cytosolic peptide is highly polar and contains three lysine residues in positions 3, 5 and 7 relative to the C-terminus. Recent studies with hybrid proteins containing distinct domains of OST48 and Golgilocated Man 9-mannosidase have shown that the lysines at positions 5 and 3 of the cytosolic domain of OST48 function as a structural motif conferring ER residency by a retrograde transport pathway [10]. In this paper we describe impaired localisation of OST48 to the ER by substitution of both Lys-5 and Lys-3 with arginine but not when replaced by two histidines. MATERIALS AND METHODS Materials. Synthetic oligonucleotides, Taq DNA polymerase, Dulbecco’s Modified Eagle Medium (DMEM), fetal bovine serum, pSV.SPORT1 (Invitrogen); restriction endonucleases (MBI Fermentas);
751
0006-291X/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
DEAE dextran (Amersham Bioscience); COS 1 cells (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH); goat anti-rabbit IgG alkaline phosphatase conjugate (Sigma); fluorescein (DTAF)-conjugated AffiniPure goat anti-rabbit IgG (Dianova); nitrocellulose membranes (Schleicher & Schuell); ABI PRISM Big dye terminator cycle sequencing kit (Applied Biosystems). All other compounds were of analytical grade purity. Vector construction. The full-length cDNA encoding for pig liver OST48 was subcloned into the mammalian expression vector pSV.SPORT1 taking advantage of common restriction sites [13]. Amino acid exchanges in the C-terminal domain of OST48 were introduced by in vitro oligonucleotide-directed mutagenesis using the OST48-specific cDNA as the template [14]. For PCR amplification the following OST48-derived oligonucleotides were used as sense (s) and anti-sense (as) primers:
FIG. 1. Schematic representation of the OST48 domain structure with the cytosolic portion shown in full. Amino acid exchanges are shown shaded.
RESULTS AND DISCUSSION
Exchanges are shown in grey. The introduction of base mutations was confirmed by DNA sequencing. Cell culture and transfection. COS 1 cells were grown at 37°C under 5% CO2 as a monolayer culture in DMEM supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 g/ml streptomycin. After having reached ⬃60% confluency COS 1 cells were transfected with vector constructs encoding for wild-type OST48 and the OST48-mutants using the DEAE-dextran/chloroquine method [15]. Immunofluorescence microscopy. Transfected COS 1 cells were grown for 36 h on sterile coverslips. The cells were either fixed with 8% formaldehyde, or fixed and then permeabilised with 0.2% Triton X-100 in 200 mM Hepes, pH 7.2, for 5 min, followed by incubation in a 1:30 dilution of an affinity-purified anti-OST48-antibody in PBS containing 0.1% ovalbumin [13]. After 12 h at 4°C, antigen/antibody complexes were labeled by treatment at 37°C for 1 h with a 1:100 dilution of a DTAF-conjugated goat anti-rabbit IgG antibody and the cells were processed as previously described [10, 16]. General methods. SDS–PAGE and immunoblotting were performed as described previously [17–19]. PCR amplification, vector construction and other molecular biological techniques were performed as detailed in [14, 15, 20, 21]. Nucleotide sequencing was done according to Sanger by using the Big Dye terminator cycle sequencing kit [22].
Replacement of Both Lysines in the Cytosolic Lys-Xaa-Lys Triplet by Arginine Causes Cell Surface Expression of OST48 To characterise the functional specificity of the LysXaa-Lys motif for ER targeting, a range of OST48 mutants were synthesized in which the lysine residues in the cytosolic domain were replaced by leucine, arginine and histidine (Fig. 1). The amino acids were exchanged by in vitro directed mutagenesis using OST48 full-length cDNA subcloned into pSV.SPORT1. COS 1 cells were then transfected with the corresponding cDNAs, followed by characterisation of the overexpressed proteins by immunoblotting and determination of their subcellular localisation by indirect immunofluorescence microscopy using a monospecific polyclonal antiOST48-antibody for detection [13]. As shown in Fig. 2, COS 1 cells transfected with the vector encoding for wild-type OST48, efficiently overexpressed a polypeptide whose molecular mass (⬃48 kDa) corresponded to that calculated from the ORF in the OST48-specific cDNA (lane 2). No immunoreactive
752
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Immunoblot analysis of wild-type OST48 and OST48 mutants. COS 1 cells transfected with the various vector constructs were solubilised and aliquots of the detergent extracts containing equal amounts of cell protein, were subjected to SDS–PAGE, followed by immunoblotting using a polyclonal anti-OST48 antibody for staining [13]. Lane 1, pSV.SPORT1-transfected cells (control); lane 2, wild-type OST48; lane 3, OST48-R3; lane 4, OST48-R5/R3; lane 5, OST48-H5/H3; lane 6, OST48-H5/L3.
material was detectable on Western blots prepared from detergent extracts of COS 1 cells transfected with pSV.SPORT1, except for a minor band at a lower molecular mass (lane 1). This indicates that the level of endogeneous OST48 is rather low. Immunofluorescence analysis of transfected COS 1 cells permeabilised with Triton X-100 gave a strong labeling of intracellular structures typical for the ER (Fig. 3A), whereas nonpermeabilised cells remained unstained (Fig. 3B). Thus OST48 is expressed as an ER-resident protein. A similar ER-specific fluorescence pattern was obtained for the OST48-L7 mutant protein containing leucine in place of Lys-7 (not shown). By contrast, nonpermeabilised cells overexpressing either OST48-L3 (Fig. 3D) or OST48-L5 exerted an intense fluorescence staining at the cell surface. Thus these two protein mutants, unlike wild-type OST48 and OST48-L7, are transported to the plasma membrane, consistent with the Lys-5/Lys-3 sequence directing ER-localisation to OST48 [10]. In addition to cell-surface labeling, ERand Golgi-specific structures were also found to be stained when the cells were permeabilised (Fig. 3C). A reasonable explanation for this observation is that the capacity of protein transport may be limited, becoming overloaded due to high over-expression. A similar intracellular staining pattern was also seen for other OST48 mutants even though they were over-expressed as cell surface proteins (see below). Lysine and arginine contain a side chain which is protonated at intracellular pH. While the distance between positive charge and ␣-C atom is comparable, the -amino and ␦-guanido groups differ in size and structure. To examine whether arginine can mimick the function of Lys-5 and Lys-3 in ER-targeting, the two lysine residues were replaced stepwise by arginine (Fig. 1). As expected, the molecular masses of OST48-R3 and OST48-R5, as well as of the OST48R5/R3 double mutant over-expressed in COS 1 cells, were identical with that of wild-type OST48 (shown for OST48-R3 and OST48-R5/R3 in Fig. 2, lanes 3 and 4). Fluorescence analysis of COS 1 cells transfected with the OST48-R3- and OST48-R5-specific cDNA revealed ER typical staining, whereas nonpermeabilised cells remained unstained (shown for OST48-R3 in Figs. 3E
and 3F). By contrast, an intense cell surface fluorescence was seen on non-permeabilised cells overexpressing the OST48-R5/R3 double arginine mutant (Fig. 3H). These observations show that the ER-localisation information residing in the Lys-Xaa-Lys motif of OST48 is not affected by replacing either Lys-5 or Lys-3 but impaired after substitution of both lysines by arginines. Histidine and Lysine Cannot Be Distinguished in as Far as ER Localisation Is Concerned The positive charges in the lysine motif are assumed to play a key role in recognition and binding to the COP-I complex. Since the pK S (⬃6.5) of the histidine imidazol ring would suggest only partial protonation at intracellular pH, we replaced Lys-5 and/or Lys-3 in the lysine signal by histidine in order to see whether ER localisation of the OST48 polypeptide was affected. The immunoblot analysis shows for the OST48-H5/H3 protein (Fig. 2, lane 5) that the His-Xaa-His double mutant is overexpressed in COS 1 cells as a ⬃48 kDa protein. Similar results were obtained for the monosubstituted OST48-H5 and OST48-H3 mutants (not shown). Permeabilised COS 1 cells transfected with the OST48-H5, OST48-H3- and OST48-H5/H3-specific cDNA exhibited a fluorescence staining pattern specific for ER elements. By contrast, no staining was detectable at the cell surface of nonpermeabilised cells. This indicates that, independently of whether either Lys-5, Lys-3 (Figs. 4A and 4B) or both lysines (Figs. 4C and 4D) were replaced by histidine, these protein mutants are ER resident. Thus not only one but two histidine residues are able to replace lysine in conferring ER residency on OST48, in apparent contrast to the ArgXaa-Arg sequence (Scheme 1). Based on the observation that one lysine residue in the Lys-Xaa-Lys motif could be substituted by arginine without loss ER-localisation information, histidine in either position 5 or 3 of the His-Xaa-His sequence was exchanged against arginine (Fig. 1; Scheme 1). Immunofluorescence studies with permeabilised cells showed that OST48-R5/H3 was still overexpressed as an ER-resident protein, whereas the inverted OST48H5/R3 mutant containing arginine in the 3-position was detected at the cell surface (Fig. 4F). This suggests that the side chain of the amino acid in the 3-position is more important for recognition by COP-I components, apparently tolerating less structural variations than in the 5-position of the sequence motif. As observed for the lysine motif, substitution of either His-5 or His-3 by leucine (Fig. 1) resulted in loss of ER residency, consistent with and supporting the specific and independent function of the His-Xaa-His sequence as sorting signal for type I proteins (Fig. 2, lane 6; Fig. 4H).
753
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. ER retention of OST48 is impaired by replacing Lys-5 and Lys-3 with arginine. COS 1 cells transfected with the cDNA encoding for either wild-type OST48 (A, B), OST-L3 (C, D), OST48-R3 (E, F), or OST48-R5/R3 (G, H) were either fixed with formaldehyde (B, D, F, H), or fixed and then permeabilised with Triton X-100 (A, C, E, G). Immunofluorescence staining was carried out using a polyclonal rabbit anti-OST48 antibody, followed by labeling of the antigen–antibody complexes with a goat anti-rabbit-IgG antibody tagged with fluorescein (DTAF) [10, 13].
CONCLUSIONS It is well established that two lysine residues located in positions 3 and 4/5 relative to the C-terminus represent a minimum sequence requirement for ER retention of type I membrane proteins. On the other hand, ER residency of type II proteins is conferred by double arginines in the N-terminal domain. Both ERtargeting motifs are thought to operate via a Golgi to
ER retrograde mechanism involving COP-I-coated vesicles [1, 2, 5, 6]. The results described in this paper show that ER-localisation information encoded for by lysine in positions 5 and 3 of the cytosolic peptide domain of OST48, is destroyed after both have been replaced by arginine. This indicates that simple charge interactions are not sufficient for signal function and that additional factors residing in the lysine side chain and/or in the amino acid sequence adjacent to the
754
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. The His-Xaa-His motif confers ER-localisation on OST48. Transfected COS 1 cells overexpressing either OST48-H3 (A, B), OST48-H5/H3 (C, D), OST48-H5/R3 (E, F) or OST48-H5/L3 (G, H) were either fixed with formaldehyde (B, D, F, H) or fixed and permeabilised (A, C, E, G) and processed as described in the legend to Fig. 3.
double-lysine motif are involved in directing ER localisation. Since the distance from the peptide backbone of the positive charge in arginine and lysine is comparable, it is reasonable to assume that loss of ERresidency as observed for the OST48-R5/R3 double mutant may be caused by steric factors introduced by the ␦-guanido group of arginine, which is larger in size than the -amino group of lysine. This may prevent recognition of the Arg-Xaa-Arg sequence by lysinespecific subunits of the COP-I complex, consistent with the current view that retrograde transport of type I
and type II membrane proteins occurs via distinct pathways involving different populations of COP-Icoated vesicles [1, 2, 7]. Whereas the OST48-R5/R3 double mutant is expressed in COS 1 cells as a cell-surface protein, arginine can replace either Lys-5 or Lys-3 without affecting ER localisation. This implies that the sorting signal is functional so long as one position is still occupied by a lysine residue. A reasonable explanation for this observation is that the binding site for the double-lysine motif in the COP-I complex displays some flexibility,
755
Vol. 291, No. 4, 2002
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
after being bound to coatomer subunits. This would be consistent with the observation that the Arg-Xaa-Arg motif lacks ER-targeting information. Comparing the side chain structure of arginine with that of lysine or histidine, the failure to be recognised by lysine- and/or histine-specific COP-I binding proteins in spite of being positively charged could result from, and thus be explained by, structural constraints imposed by one of the two NH 2 functions in the ␦-guanido group, with one of them occupying the position equivalent to the C 5methylene function in the imidazol ring being tolerated.
SCHEME 1
allowing for compensation of the structural constraints imposed by one, but not by two, of the bulky ␦-guanido groups. The observation that both the OST48-R5 and the OST48-R3 hybrid mutants are expressed as ER resident proteins, contradicts data previously described by Shin et al. [23] who suggested that Lys-3 in the double-lysine motif but not Lys-5 should be functionally invariant. This apparent discrepancy points up that the binding specificity may depend on the nature of the amino acids adjacent to the sequence motif. Our data show clearly that, in contrast to the ArgXaa-Arg sequence and in contrast to studies reported elsewhere in the literature [23, 24], two histidines may replace the lysines in ER targeting, at least functionally. This is surprising because the histidine imidazol group (pK S ⬃6.5) (i) is approximately 10% protonated at intracellular pH; (ii) like arginine has a positive charge which is delocalised; and (iii) the planar imidazol ring is rigid compared to the lysine or arginine side chain. Thus it appears reasonable to postulate that the His-Xaa-His sequence may represent a differential ERlocalisation motif interacting with COP-I subunits, differing from those which specifically bind the lysine motif in type I and the arginine signal in type II proteins. Assuming that the histidine and lysine motifs interact with the same receptor proteins in COP-Icoated vesicles, several observations may be made in terms of signal binding site specificty: (i) the side chain of histidine must be protonated since charge interactions, although apparently not sufficient, are likely to play a key role in signal recognition; (ii) since the distance from the polypeptide backbone to the N 1 nitrogen in imidazol is comparable to that for the -amino group in lysine, binding to COP-I structures may stabilize a putative positive charge in this positon; (iii) the methylene group at C 5 of the planar imidazol ring appears not to interfere with or prevent signal recognition, suggesting that this region is not in direct contact with binding protein(s) in the COP-I complex. Taking into account the structural properties of the histidine side chain, including its rigidity, planarity and charge distribution, it is possible that the imidazol ring mimicks the side chain conformation for lysine
ACKNOWLEDGMENTS This work was supported by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 284). The authors are indebted to Dr. R. A. Klein (Universita¨ t Bonn) for critical reading of the manuscript.
REFERENCES 1. Teasdale, R. D., and Jackson M. R. (1996) Signal-mediated sorting of membrane proteins between the endoplasmic reticulum and the Golgi apparatus. Annu. Rev. Cell Dev. Biol. 12, 27–54. 2. Wieland, F., and Harter, C. (1999) Mechanisms of vesicle formation: Insights from the COP system. Curr. Opin. Cell Biol. 11, 440 – 446. 3. Lewis, M. J., and Pelham, H. R. (1992) Ligand-induced redistribution of a human KDEL receptor from the Golgi complex to the endoplasmic reticulum. Cell 68, 353–364. 4. Yamamoto, K., Fujii, R., Toyofuka, Y., Saito, T., Koseki, H., Hsu, V. W., and Aoe, T. (2001) The KDEL receptor mediates a retrieval mechanism that contributes to quality control at the endoplasmic reticulum. EMBO J. 20, 3082–3091. 5. Cosson, P., and Letourneur, F. (1994) Coatomer interaction with di-lysine endoplasmic reticulum retention motifs. Science 263, 1629 –1631. 6. Schutze, M.-P., Peterson, P. A., and Jackson, M. R. (1994) An N-terminal double-arginine motif maintains type II membrane proteins in the endoplasmic reticulum. EMBO J. 13, 1696 –1705. 7. Orci, L., Stamnes, M., Ravazzola, M., Amherdt, M., Perrelet, A., So¨ llner, T. H., and Rothman, J. E. (1997) Bidirectional transport by distinct populations of COPI-coated vesicles. Cell 90, 335– 349. 8. Cosson, P., and Letourneur, F. (1997) Coatomer (COPI)-coated vesicles: role in intracellular transport and protein sorting. Curr. Opin. Cell Biol. 9, 484 – 487. 9. Fu, J., and Kreibich, G. (2000) Retention of subunits of the oligosaccharyltransferase complex in the endoplasmic reticulum. J. Biol. Chem. 275, 3984 –3990. 10. Hardt, B., Aparicio, R., and Bause, E. (2001) Analysis of structural signals conferring localisation of pig OST48 to the endoplasmic reticulum. Biol. Chem. 382, 1039 –1047. 11. Kornfeld, R., and Kornfeld, S. (1985) Assembly of asparaginelinked oligosaccharides. Annu. Rev. Biochem. 54, 631– 664. 12. Silberstein, S., Kelleher, D. J., and Gilmore, R. (1992) The 48kDa subunit of the mammalian oligosaccharyltransferase complex is homologous to the essential yeast protein WBP1. J. Biol. Chem. 267, 23658 –23663. 13. Hardt, B., Aparicio, R., and Bause, E. (2000)The oligosaccharyl-
756
Vol. 291, No. 4, 2002
14.
15.
16.
17. 18.
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
transferase complex from pig liver: cDNA cloning, expression and functional characterisation. Glycoconj. J. 17, 767–779. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D. Seidman, J. G., Smith, J. A., and Struhl, K. (1987) Current Protocols in Molecular Biology, Wiley, New York. Sambrook, J., Fritsch, E. F., and Maniatis T. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Bieberich, E., and Bause, E. (1995) Man 9-mannosidase from human kidney is expressed in COS cells as a Golgi resident type II transmembrane N-glycoprotein. Eur. J. Biochem. 233, 644 – 649. Laemmli, U. K. (1970) Cleavage of structural proteins during the assembly of the head bacteriophage T4. Nature 227, 680 – 685. Blake, M. S., Johnson, K. H., Russel-Jones, G. J., and Gotschlich, E. C. (1984) A rapid, sensitive method for detection of alkaline phosphatase-conjugated anti-antibody on Western blots. Anal. Biochem. 136, 175–179.
19. Knecht, D. A., and Dimond, R. L. (1984) Visualization of antigenic proteins on Western blots. Anal. Biochem. 136, 180 –184. 20. Sakai, K., Gelfand, D. H., Stoffels, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis, K. B., and Ehrlich, H. A. (1988) Primerdirected enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487– 491. 21. White, B. A. (1993) PCR-Protocols—Current Methods and Applications, Humana Press, Ottawa, Canada. 22. Sanger, F., Nicklen, S., and Coulson A. R. (1977) DNA sequencing with chain termination inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467. 23. Shin, J., Dunbrack, Jr., R. L., Lee, S., and Strominger, J. L. (1991) Signals for retention of transmembrane proteins in the endoplasmic reticulum studied with CD4 truncation mutants. Proc. Natl. Acad. Sci. USA 88, 1918 –1922. 24. Jackson, M. R., Nilsson, T., and Peterson, P. A. (1990). Identification of a consensus motif for retention of transmembrane proteins in the endoplasmic reticulum. EMBO J. 9, 3153–3162.
757