Multiple determinants direct the orientation of signal-anchor proteins: the topogenic role of the hydrophobic signal domain

Abstract

The orientation of signal-anchor proteins in the endoplasmic reticulum membrane is largely determined by the charged residues flanking the apolar, membrane-spanning domain and is influenced by the folding properties of the NH2-terminal sequence. However, these features are not generally sufficient to ensure a unique topology. The topogenic role of the hydrophobic signal domain was studied in vivo by expressing mutants of the asialoglycoprotein receptor subunit H1 in COS-7 cells. By replacing the 19-residue transmembrane segment of wild-type and mutant H1 by stretches of 7-25 leucine residues, we found that the length and hydrophobicity of the apolar sequence significantly affected protein orientation. Translocation of the NH2 terminus was favored by long, hydrophobic sequences and translocation of the COOH terminus by short ones. The topogenic contributions of the transmembrane domain, the flanking charges, and a hydrophilic NH2-terminal portion were additive. In combination these determinants were sufficient to achieve unique membrane insertion in either orientation.

Figure 1

Mutant H1ΔLeu19 is expressed as two differently glycosylated forms. (A) COS-7 cells were transfected with cDNAs of H1, H1Δ, H1Leu19, H1ΔLeu19, and HC (the exoplasmic portion of H1 equipped with the cleavable signal sequence of influenza virus hemagglutinin) as indicated. The transfected cells were labeled for 30 min with [³⁵S]methionine, solubilized, and subjected to immunoprecipitation using an antiserum directed against the COOH-terminal sequence of H1. The immunoprecipitates were analyzed by gel electrophoresis and fluorography. (B) Immunoprecipitates were treated without (−) or with endoglycosidase H (E) before analysis by gel electrophoresis. The positions of marker proteins are shown with their molecular weights indicated in kD. The band indicated by an asterisk represents a partially glycosylated species (see text).

Figure 2

The unglycosylated form of mutant H1ΔLeu19 is inserted in the membrane in an inverted orientation. (A) Saponin extraction: cells were transfected with the indicated constructs, labeled, and extracted with 0.1% saponin. The saponin extract (S) and the remaining cells (C) were separately immunoprecipitated and analyzed by gel electrophoresis and fluorography. Untreated cells were solubilized and immunoprecipitated as a measure of the total material (TOT). (B) Alkaline extraction: transfected and labeled cells were homogenized and incubated at pH 11.5. The samples were either immunoprecipitated directly (TOT) or after separation into pellet (P) and supernatant fractions (S). (C) Protease protection: transfected cells were labeled, homogenized, and incubated without (−) or with trypsin (T) or with trypsin in the presence of detergent (TD). Immunoprecipitates were analyzed by SDS–gel electrophoresis and fluorography. The band indicated by an asterisk represents the partially glycosylated species. (D) Schematic representation of the membrane orientation of constructs H1Δ and H1ΔLeu19. The H1 sequence is shown in black and the Leu₁₉ domain as an empty rectangle. The glycosylation sites as indicated by open circles and N-linked glycans by closed squares.

Figure 4

Effect of different hydrophobic domains on membrane insertion of H1, H1Δ, and H1ΔQ. The constructs H1 (A), H1Δ (B), and H1ΔQ (C) with the wild-type transmembrane domain of H1 (lane 1) or with hydrophobic segments consisting of 7–25 leucine residues (lanes 2–8) were expressed in COS-7 cells, labeled, immunoprecipitated, and analyzed by gel electrophoresis and fluorography. Membrane integration assessed by saponin extraction (D) and protease sensitivity (E) is shown for the constructs with the shortest hydrophobic segments of 7 leucines (see legend to Fig. 2). The position of the marker proteins of 29 and 35 kD are indicated.

Figure 3

Amino acid sequence of the signal–anchor domain of H1 mutant constructs. The hydrophobic transmembrane segments and their flanking sequences are listed. H1-4g and H1-4gLeu# are identical to H1-4 and H1-4Leu# except for the insertion of the tripeptide sequence MTM following asparagine-13 in the NH₂terminal portion, which creates a potential glycosylation site.

Figure 6

Effect of different hydrophobic domains on membrane insertion of H1-4 and H1-4g. The constructs H1-4 (A) and H1-4g (B) with the wild-type transmembrane domain of H1 (lanes 1 and 2) or with hydrophobic segments consisting of 7–25 leucine residues (lanes 3–14) were expressed in COS-7 cells, labeled, and immunoprecipitated. Samples were treated without (−) or with endoglycosidase H (E) before analysis by gel electrophoresis and fluorography.

Figure 7

Quantitation of the topology of the constructs H14Leu# and H1-4gLeu#. The insertion experiments including those shown in Fig. 6 were quantified by densitometric scanning of the fluorographs. The fraction of once glycosylated protein, i.e., with N_exo/C_cyt orientation, is presented as percent of the total of all forms as described in the legend to Fig. 5. The values for H14Leu# represent the mean of three or more experiments with standard deviations; those for H1-4gLeu# represent single determinations performed simultaneously.

Figure 5

Topology of signal–anchor mutants. Insertion experiments, including those shown in Fig. 4, were quantified by densitometric scanning of fluorographs. The fraction of unglycosylated protein, i.e., with N_exo/C_cyt orientation, is presented as percent of the total of all forms. The values for constructs with polyleucine domains are plotted as a function of the number of leucines in this segment (Leu#). Corresponding constructs with the transmembrane segment of wild-type H1 are shown to the right (wt). The values represent the mean of three or more experiments with standard deviations.