A. Schematic representation of the SILAC approach. Proteins of HepG2 cells were labelled with either ¹²C₆ L-Lysine and ¹²C₆ ¹⁴N₄ L-Arginine (“light” sample) or ¹³C₆ L-Lysine and ¹³C₆ ¹⁵N₄ L-Arginine (“heavy” sample). Cells were treated for 17 h with cotransin (30 μM) or DMSO and pooled cell lysates or supernatants containing total secretory or integral membrane proteins were separated using a SDS gradient gel (4–12%). The proteins bands were cut out, proteins were digested using trypsin and finally analysed using LC-MS/MS. The forward experiment is shown. For the reverse experiment isotopic coding was inverted. B. SILAC results. Shown are dotplots for secretory (left panel) and integral membrane proteins (right panel) which were identified and quantified by LC-MS/MS analysis (see also the for a list of the individual proteins). The ratios of protein expression following DMSO or cotransin treatment of the forward (heavy/light sample = H/L) and reverse experiments (light/heavy sample = L/H) are plotted against each other. The grey areas indicate proteins which were considered to be inhibited by cotransin.

After transfection and treatment with cotransin (17 h, 30 μM) (+) or with DMSO (-) secretory and integral membrane proteins were isolated from HepG2 cells. The proteins were identified by SDS-PAGE/immunoblotting using specific primary antibodies and horseradish peroxidise-conjugated anti-mouse or anti-rabbit IgG as secondary antibodies. The cytosolic GAPDH protein does not contain a signal sequence and served as a control for a non-sensitive protein. As examples for secretory proteins (all SPs) cotransin-sensitive Apo B-100 and cotransin-resistant PAI-1 were used. For membrane protein possessing SPs, cotransin-sensitive CDH2 and cotransin-resistant CNX were analyzed. As examples for membrane proteins containing SASs, cotransin-sensitive Erlin2 and cotransin-resistant CLDN1 are shown. The immunoblots are representative of three independent experiments. The bar graphs shown at the right side of each immunoblot represent mean intensities of the respective protein bands of these three independent experiments ±SD (densitometric analysis using ImageJ).

The blue, red and green fitted curves represent 143 non-ensitive integral membrane proteins, 21 sensitive integral membrane proteins and 50 sensitive secretory proteins respectively. A. Hydrophobicity analysis. The frequency of signal sequence hydrophobicity (sorted in classes of hydrophobicity ranging from 0–100 in steps of 4 in a scale of 0–25) was plotted against total hydrophobicity of the signal sequences (ranging from 0–100). B. Signal sequence length analysis. The frequency of signal sequence length (sorted in classes ranging from 0–50 amino acid residues in steps of 4 in a scale from 0–100) was plotted against total length of the signal sequences.

A. Sequence alignment. The sequences of the 12 sensitive SASs, identified in this study, are shown in grey (TM1 of the proteins). Cotransin sensitivity decreases from top to bottom. The sequence of the SAS of the TNF-α, which was previously shown to be cotransin-sensitive [] is shown separately. Nc indicates a SAS with an N-terminus oriented towards the cytoplasm, Ne indicates a sequence with an N-terminus facing the extracellular site (ER: luminal side); n/s indicates an as yet non-specified N tail orientation. The consensus motif consists of two groups of small amino acids in the central part (black), which are separated by two or three bulky amino acid residues and flanked by large charged, large polar or aromatic amino acids residues (white with black frame). The black box below the alignment assigns the possible positions of the amino acid residues in the motif: (#) and (*) indicate the positions of the small amino acid residues forming the first and second cavities respectively; (~) and ($) indicate the amino acid residues separating and flanking the small residues respectively. The abbreviations are: IMP2C, integral membrane protein 2 C; IMP2B, integral membrane protein 2 B; HLA II, HLA class II histocompatibility antigen gamma chain; TMEM230, transmembrane protein 230; MHC I, MHC class I antigen; ECE-1, endothelin-converting enzyme 1; Acyl-CoA, Acyl-CoA desaturase; TM Prot2, transmembrane protein 2; Erlin1, Erlin-1; LIMP2, lysosome membrane protein 2; Erlin2, Erlin-2; AGPR1, asialoglycoprotein receptor 1; TNF-α, tumor necrosis factor-α. B. Exemplary α-helical (left) and solvent-reachable surface projection (right) of the SAS of IMP2B. Green colour represents the small amino acid residues forming the two cavities, red colour the separating residues and blue colour the flanking residues. The two cavities are indicated by arrows. C. Introduction of the identified conformational consensus motif into the SAS of the cotransin-resistant AQP2 protein. Upper panel: Alignment of the SAS (highlighted in grey) of construct WT.AQP2 and mutant CM.AQP2. In the case of CM.AQP2, the motif consists of the flanking amino acid residues (white with black frame), the small residues forming the cavities in the surface of the helix (black) and the separating more bulky residues lying between the cavities. Lower panel: α-helical structure of WT.AQP2 and mutant CM.AQP2 visualized using the program PyMol (left side). The structural level is shown by the solvent reachable surface of the α-helices (right side). Green colour represents the small amino acid residues forming the two cavities, red colour the separating residues and blue colour the flanking residues. The two cavities, which are a result of the mutations, are indicated by black arrows.

A. Biosynthesis of constructs WT.AQP2 and CM.AQP2 following cotransin treatment. Columns represent the GFP fluorescence signals (arbitrary units) of the constructs after 19 h of treatment with 1.5% DMSO solvent alone (-) (dark grey columns) or 10 μM cotransin/1.5% DMSO (+) (light grey columns). Shown are mean values of three independent experiments ± SD. Fluorescence was quantified by flow cytometry measurements. B. Concentration-response curve for the cotransin-mediated biosynthesis inhibition of CM.AQP2. Data points represent the mean values of the GFP fluorescence signals of the construct after 19 h of treatment with increasing concentrations of cotransin (1–50 μM in 1.5% DMSO). Shown are mean values of three independent experiments ± SD. Fluorescence was quantified by flow cytometry measurements as above and data were normalized to the DMSO control (1.5% DMSO). The calculated IC₅₀ value is indicated. C. Upper Panel. Colocalization of the GFP fluorescence signals of the truncated constructs WT.AQP2.NT and CM.AQP2.NT (left side, green) with the CFP fluorescence signals of the cotransfected ER marker ECFP-ER (middle, red). The fluorescence signals were recorded using confocal LSM and computer-overlayed (right side; colocalization is indicated by yellow color). The xy-scans show representative cells and are representative of three independent experiments. The cartoon on the right side shows a schematic depiction of the constructs. The 4 point mutations of construct CM.AQP2.NT are indicated by (****). Lower panel. Subcellular localization of unfused soluble GFP (control protein which does not contain a signal sequence). For clarity, the GFP fluorescence signals (left side, green) and those of the plasma membrane dye trypan blue (middle, red) were recorded in this case and computer overlayed (right panel). The xy-scans show representative cells and are representative of three independent experiments.