Volume 8 issue 12 December 2007
Integral and Associated Lysosomal Membrane Proteins
Bernd Schröder, Christian Wrocklage, Cuiping Pan, Ralf Jäger, Bernd Kösters, Helmut Schäfer, Hans-Peter Elsässer, Matthias Mann and Andrej Hasilik

Supplemental Methods

Isolation and characterization of lysosomal membranes Placental tissue was homogenized in 250 mM sucrose, 10 mM triethanolamine, 10 mM acetic acid, 5 mM EDTA, pH 7.2 (homogenization buffer) using an Ultra-Turrax (Jahnke & Kunkel, Staufen, Germany). A postnuclear supernatant was prepared by centrifugation at 2,600 x g_max for 10 min in a GSA rotor (Sorvall, Bad Homburg, Germany). The postnuclear supernatant was layered over cushions of a Percoll solution prepared in homogenization buffer and adjusted to a density of 1.075 g/ml. After centrifugation at 47,800 x g_max for 15 min in a SS-34 rotor (Sorvall) an organelle concentrate was recovered in the sediment. This was further fractionated on a Percoll® density gradient. Step gradients were prepared in Ultracrimp® tubes (Sorvall) with the following composition (from bottom to top): 3 ml of 65% (w/v) sucrose in homogenization buffer, 8 ml Percoll® in homogenization buffer adjusted to ρ = 1.095 g/ml and 14 ml Percoll® in homogenization buffer adjusted to ρ = 1.075 g/ml. Finally, 10 ml organelle concentrate were layered on top and the gradients were centriuged at 41,700 x g_max for 30 min in a TV860 vertical rotor (Sorvall). From bottom to top 17 fractions à 2 ml were collected. Fractions 2 – 5 with the highest specific activity of lysosomal marker enzymes were combined ("dense pool") and used for further purification.

"Dense pool" (30 ml) was diluted 9-fold with isotonic Hepes buffer (250 mM sucrose, 10 mM Hepes-NaOH, 1 mM EDTA, pH 7.2) and organelles were sedimented by centrifugation at 15,800 x g_max for 20 min in a SS-34 rotor (Sorvall). The sediment was resupended in isotonic Hepes buffer. In order to facilitate removal of mitochondria and lysosomal inclusion bodies lysosomes were selectively disrupted by incubation in the presence of 20 mM methionine methyl ester in isotonic Hepes buffer and a final volume of 18 ml for 30 min at room temperature. Subsequently, the following proteinase inhibitors were added (final concentrations): 0.5 mM iodoacetamide, 10 µM leupeptin, 1 mM phenylmethanesulfonyl fluoride and 1 µM Z-Phe-Phe-diazomethylketone and pepstatin A. The sample was split and layered over two 28 ml sucrose gradients prepared from solutions of 32.5% (w/v) and 55% (w/v) sucrose in isotonic Hepes buffer. The gradients were centrifuged at 112,700 x g_max for 12 h in a SW-28 rotor (Beckman, München, Germany) and sixteen fractions were collected. The two fractions with the highest activity of β-glucocerebrosidase were combined ("sucrose gradient pool") and further fractionated in a self forming iodixanol gradient. Per gradient, 4.4 ml of "sucrose gradient pool" were mixed with 5.4 ml iodixanol working solution (50% (w/v) iodixanol, 42 mM sucrose, 10 mM Hepes-NaOH, 1 mM EDTA, pH 7.2) and 11.7 ml 10 mM Hepes-NaOH, 1 mM EDTA, pH 7.2, and transferred into an Ultracrimp® tube. A cushion of 5 ml 40% (w/v) iodixanol solution was layered below and 6 ml of 12.5% (w/v) iodixanol solution and 2.5 ml isotonic Hepes buffer above the sample. The gradients were centrifuged at 236,500 x g_max for 2 h in the TV860 vertical rotor. From top to bottom sixteen fractions 2 ml each and a 3 ml residue were collected. Lysosomal membranes were recovered from fractions 1 – 4 that were combined and are referred to as "iodixanol upper pool".

For proteomic analysis, membranes and a total sedimentable subfraction from "iodixanol upper pool" and "dense pool", respectively, were obtained after hypotonic dilution in 10 mM Tris-HCl, pH 7.4, and sedimentation at 250,000 x g_max for 2 h in a 60Ti rotor (Beckman). Samples were washed once in the same buffer and recentrifuged under the same conditions.

Proteomic analysis using SDS-PAGE and LC-MS/MS and data processing Protein from purified lysosomal membranes and the reference sample was suspended in LDS sample sample buffer (Invitrogen, Carlsbad, CA), adjusted to 50 mM dithiothreitol and incubated for 10 min at 70 °C. Samples were applied to a NuPAGE® Novex 4–12% bis-Tris SDS-PAGE gel (Invitrogen) and protein was visualized with colloidal Coomassie brilliant blue staining kit from Invitrogen (Fig. S1). Stained segments were cut into into 1 mm³ cubes, which were washed five times with 25 mM NH₄HCO₃ in 50% (v/v) ethanol then dehydrated in 100% ethanol. Disulfides were reduced with 10 mM dithiothreitol in 50 mM NH₄HCO₃, (1 h at 56 °C) and thiols carbamidomethylated with 55 mM iodoacetamide in 50 mM NH₄HCO₃ (45 min at room temperature in the dark). Gel cubes were washed with 50 mM NH₄HCO₃, dehydrated with 100% ethanol and after repeating these steps dried in a Speed Vac concentrator for 5 min. After rehydration with 12.5 ng/µl trypsin in 50 mM NH₄HCO₃ on ice for 15 min, the excess of the solution was removed and the gel pieces covered with 50 mM NH₄HCO₃. After overnight incubation at 37 °C, the liquid was transferred to a fresh tube and residual peptides extracted twice with 3% trifluoroacetic acid in 30% (v/v) acetonitrile and twice with 100% acetonitrile. The combined extracts were processed on reverse-phase C18 StageTip columns and the eluted peptides subjected to LC-MS/MS-analysis.

Peptide separations were performed using a 15 cm in-housed fused silica emitter (75 µm inner diameter) packed with reversed-phase ReproSil-Pur C18-AQ 3 µm resin (Dr. Maisch GmbH, Ammerbuch-Entringen, Germany) in an Agilent 1100 nanoflow system. Peptide mixtures were injected at a flow rate of 500 nl/min. The elution was performed with a linear 5% to 40% (v/v) acetonitrile gradient in 0.5% (v/v) acetic acid within 140 min at a flow rate of 250 nl/min. The eluate was sprayed immediately into a LTQ-Orbitrap mass spectrometer (Thermo Electron, Bremen, Germany) equipped with a nanoelectrospray ion source (Proxeon Biosystems, Odense, Denmark). The mass spectrometer was operated in data dependent mode to automatically switch between MS and MS/MS acquisition. Survey full scan MS spectra (from m/z 300-1800) were acquired in the Orbitrap with resolution R = 60,000 at m/z 400 (after accumulation to a target value of 1,000,000 charges in the linear ion trap). The 10 most intensive ions were sequentially isolated and fragmented in the linear ion trap using collisionally induced dissociation at a target value of 10,000. Former target ions selected for MS/MS were dynamically excluded for 120 s.

Lock mass option was enabled in both MS and MS/MS mode to achieve accurate mass measurements by real time internal calibration using the poly—dimethyl-cyclo-siloxane (PCM ) ions generated in the electrospray process from ambient air (protonated (Si(CH₃)₂O)₆; m/z = 445.120025). For single SIM scan injections of the lock mass into the C-trap the lock mass “ion gain” was set at 10% of the target value of the full spectrum. Calibration for the MS/MS mode was performed at m/z = 429.088735 (PCM with neutral methane loss). The following settings were used: electrospray voltage: 2.3 kV; no sheath and auxiliary gas flow; ion transfer tube temperature: 125 °C; collision gas pressure: 1.3 mTorr; normalized collision energy using wide-band activation mode: 35% for MS/MS; ion selection threshold: 500 counts for MS/MS. An acquisition q = 0.25 and activation time of 30 ms was applied for MS/MS acquisitions. Peptide identification was performed in online nanoLC-MS/MS mode.

The acquired data was searched against the International Protein Index (human) database (version 3.14) using the MASCOT search engine (44).

Carbamidomethylation was set as fixed and oxidation of methionine as well as N-acetylation as variable modifications. Peptide mass tolerance was 10 ppm and MS/MS mass tolerance 0.5 Da. Up to two missed cleavages were allowed. Only spectra with a probability-based MASCOT peptide above the score corresponding to P < 0.05 were considered statistically significant. For identification of a protein at least two peptides fulfilling the above criterion were required, giving a 99.75% overall confidence of the identification. Sequence coverage of proteins by the identified peptides was calculated with Protein Coverage Summarizer software (NIH National Center for Research Resources, http://ncrr.pnl.gov). The number of transmembrane segments of the identified proteins was predicted with TMHMM (http://www.expasy.org).

Differences in relative abundance of proteins in the "iodixanol upper pool" membranes and the "dense pool" reference sample were estimated from the number of spectral counts of each protein (6,11). The total numbers of MS/MS spectra with a significant MASCOT score were 54.822 and 33.050, respectively. For each identified protein a logarithmic enrichment ratio (R_SC) was calculated (11) that accounted for differences in size of the data sets and the sampling depth:

where R_SC is the logarithmic enrichment ratio in the lysosomal membrane preparation; SC_IUP and SC_DP the numbers of spectral counts assigned to a certain protein in the "iodixanol upper pool" and "dense pool" membrane data sets respectively; Total_IUP and Total_DP the overall numbers of matched MS/MS-spectra as mentioned above. The correction factor of 0.5 allowed the calculation of a R_SC for proteins not detected in the reference data set (spectral count SC_DP of 0) also. For statistical evaluation of observed enrichment ratios a Chi²-Test with Yates-correction was applied. The obtained p-values (P_ind) were adjusted for multiple statistical testing according to Bonferroni (P_adj).

Supplemental Materials

Supplemental Figure 1: SDS-PAGE separation of lysosomal membrane proteins for LC-MS/MS-analysis. Three subfractions (insoluble, top/interphase and bottom phase) were obtained from 170 µg total protein. Protein was detected by staining with colloidal Coomassie brilliant blue. The lanes were sliced as indicated and the slices processed for protein identification as described in the supplementary Materials and Methods.

Figure 1 (.jpg)

Supplemental Table 1: Enrichment of marker proteins in the purified membranes. The calculation of the enrichment was based on the specific activity of β-glucocerebrosidase at 37 °C (n = 3) and an ELISA detection of LAMP-2 (n = 3).

Subcellular fraction	β-Glucocerebrosidase  (-fold)*	LAMP-2 (-fold)**
“Dense pool”	1	1
Sedimentable fraction of "dense pool"	1.2 (± 0.1)	1.9 (± 0.2)
Purified lysosomal membranes	7.9 (± 2.1)	13.8 (± 1.9)

* Enzymatic assay (n=3)
**ELISA (n=3)

Supplemental Table 2: Amino acid sequences of nonredundant peptides of proteins that were identified with at least two hits in the dataset derived from purified lysosomal membranes. The columns from left to right contain the database accession number, the name of the protein, the peptide sequence, detected modifications, the number of missed tryptic cleavage sites and the Mascot peptide score.

Table 2 (.xls)

Supplemental Table 3: List of proteins detected in purified lysosomal membranes with two or more peptides. Proteins are colour-coded according to their enrichment or depletion in the purified lysosomal membranes. Significantly enriched proteins (R_SC > 1.25, P_adj < 0.025) are shown with a white background. Those with a R_SC > 1.25 not meeting the statistical criterion (P_adj > 0.025) are highlighted in yellow and depleted proteins (R_SC < 1.25) in orange. Proteins are categorized according to their known or presumed function.The entries in categories L1 to L7 correspond to the list shown in Table 1. The remainder, those not meeting the statistical criterion of the enrichment as well as the depleted proteins are represented in categories R01 – R37. The enriched and the depleted proteins in these categories are grouped separately and each group is sorted according to the estimated "hydrophilic moiety" coverage. This estimation is based on data shown in Table S4. The sorting can be altered after a downloading. SC: spectral counts either in the purified lysosomal membrane dataset or the reference sample, respectively. R_SC: logarithmic enrichment ratio, ×_STAT: χ² test statistics, P_ind: p-value of individual χ² test, P_adj: p-value adjusted for multiple testing according to Bonferroni.

Table 3 (.xls)

Supplemental Table 4: Experimental and estimated coverage of the "hydrophilic moiety" in lysosomal proteins. The amino acid sequences of 52 transmembrane proteins from the categories L1 – L7 were inspected to determine the number of residues comprizing the "hydrophobic moiety", which represented the sum of residues in transmembrane segments bound by neighbouring lysine or arginine residues. Its sequence coverage was calculated directly after localizing the detected peptides (Table S2) within these extended transmembrane segments. The remainder of the protein was considered to comprize the "hydrophilic moiety" and its experimental coverage was calculated as a complement of that of the "hydrophobic moiety". An estimation of the coverage of the "hydrophilic moiety" (I) was performed according to the following formula I = C*∑/(L*T/r + ∑ - L*T), where ∑ = total number of residues, L = average number of residues per lysine/arginine – enclosed transmembrane segment, T = number of predicted transmembrane segments, r = ratio of the average sequence coverage in the "hydrophilic" and "hydrophobic moieties". Values of I are showm in column R and the average values L and r, as calculated at the bottom of the table were highlighted in red.

Table 4 (.xls)