Membrane and membrane-associated proteins in Triton X-114 extracts of Mycobacterium bovis BCG identified using a combination of gel-based and gel-free fractionation strategies

Tuberculosis is an ancient disease that remains a significant global health problem. Because many membrane and membrane-associated proteins of this pathogen represent potential targets for drugs, diagnostic probes or vaccine components, we have analysed Mycobacterium bovis, bacillus Calmette-Guérin (BCG) substrain Moreau, using Triton X-114 for extraction of lipophilic proteins, followed by identification with LC coupled MS/MS. We identified 351 different proteins in total, and 103 (29%) were predicted as integral membrane proteins with at least one predicted transmembrane region and another 84 (23.9%) proteins had a positive grand average of hydropathicity (GRAVY) value, indicating increased probability for membrane association. Altogether 43 predicted lipoproteins (Lpps) were identified which is close to 50% of the total number of Lpps in the genome. Fifty-four proteins, including twenty-four predicted integral membrane proteins and seven predicted Lpps are described for the first time. The proportion of hydrophobic membrane and membrane-associated proteins shows that Triton X-114 is a highly efficient method for extraction of membrane proteins from bacteria, without the need for preisolation of membranes. ATP synthase, NAD(P) transhydrogenase, ubiquinone oxidoreductase and ubiquinol-cytochrome C reductase appear to represent major enzyme complexes in the membrane of Mycobacterium tuberculosis complex organisms.     

Proteome analysis of the plasma membrane of Mycobacterium tuberculosis

The plasma membrane of Mycobacterium tuberculosis is likely to contain proteins that could serve as novel drug targets, diagnostic probes or even components of a vaccine against tuberculosis. With this in mind, we have undertaken proteome analysis of the membrane of M. tuberculosis H37Rv. Isolated membrane vesicles were extracted with either a detergent (Triton X114) or an alkaline buffer (carbonate) following two of the protocols recommended for membrane protein enrichment. Proteins were resolved by 2D-GE using immobilized pH gradient (IPG) strips, and identified by peptide mass mapping utilizing the M. tuberculosis genome database. The two extraction procedures yielded patterns with minimal overlap. Only two proteins, both HSPs, showed a common presence. MALDI-MS analysis of 61 spots led to the identification of 32 proteins, 17 of which were new to the M. tuberculosis proteome database. We classified 19 of the identified proteins as 'membrane-associated'; 14 of these were further classified as 'membrane-bound', three of which were lipoproteins. The remaining proteins included four heat-shock proteins and several enzymes involved in energy or lipid metabolism. Extraction with Triton X114 was found to be more effective than carbonate for detecting 'putative' M. tuberculosis membrane proteins. The protocol was also found to be suitable for comparing BCG and M. tuberculosis membranes, identifying ESAT-6 as being expressed selectively in M. tuberculosis. While this study demonstrates for the first time some of the membrane proteins of M. tuberculosis, it also underscores the problems associated with proteomic analysis of a complex membrane such as that of a mycobacterium.     

The inhibitory effect of Mycoplasma fermentans on tumour necrosis factor (TNF)-alpha-induced apoptosis resides in the membrane lipoproteins

Mycoplasma have been shown to be involved in the alteration of several eukaryotic cell functions, such as cytokine production, gene expression and more. We have previously reported that infection of human myelomonocytic U937 cell line with live Mycoplasma fermentans (M. fermentans) inhibited tumour necrosis factor (TNF-alpha)-induced apoptosis. Mycoplasmal membrane lipoproteins are considered to be the most potent initiators of inflammatory reactions in mycoplasmal infections. The aim of this study was to clarify whether the inhibitory effect on TNFalpha-induced apoptosis is exerted by M. fermentans lipoproteins (LPMf). A significant reduction in TNFalpha-induced apoptosis was demonstrated by stimulation of U937 cells with M. fermentans total proteins, LPMf or MALP-2 (M. fermentans synthetic lipopeptide), but not with M. fermentans hydrophilic protein preparation (AqMf). To investigate the mechanism of M. fermentans antiapoptotic effect, the reduction of mitochondrial transmembrane potential (delta psi m) was measured. M. fermentans total proteins LPMf and MALP-2, but not AqMf, inhibited the reduction of delta psi m. In addition, M. fermentans total proteins LPMf and MALP-2, but not AqMf, downregulated the formation of active caspase-8. NF-kappaB was transactivated in cells treated with M. fermentans lipoproteins, and was essential for host cell survival, but not for the inhibition of TNFalpha-induced apoptosis by LPMf. Our results suggest that the inhibitory effect exerted by M. fermentans on TNFalpha-induced apoptosis in U937 cells is due to the membrane lipoproteins of these bacteria.     

Identification of mycoplasma membrane proteins by systematic TnphoA mutagenesis of a recombinant library

Wall-less prokaryotes in the genus Mycoplasma include over 90 species of infectious agents whose pathogenicity for humans and other animals is currently being assessed. Molecular characterization of surface proteins is critical in this regard but is hampered by the lack of genetic systems in these organisms. We used Tn phoA transposition to systematically mutagenize, in Escherichia coli , a genomic plasmid library constructed from Mycoplasma fermentans , a potential human pathogen. The strategy circumvented problems of expressing mycoplasma genes containing UGA (Trp) codons and relied on the construction of the vector pG7ZCW, designed to reduce Tn phoA transposition into vector sequences. Functional phoA gene fusions directly identified genes encoding 19 putative membrane-associated proteins of M. fermentans . Sequences of fusion constructs defined three types of export sequence: (1) non-cleavable, membrane-spanning sequences, (2) signal peptides with signal peptidase (SPase) I-like cleavage sites, and (3) signal peptides with SPase II-like lipoprotein-cleavage sites which, like most other mycoplasmal lipoprotein signals analysed to date, differed from those in several Gram-negative and Gram-positive eubacteria in their lack of a Leu residue at the −3 position. Antibodies to synthetic peptides that were deduced from two fusions to predicted lipoproteins, identified corresponding amphiphilic membrane proteins of 57 kDa and 78 kDa expressed in the mycoplasma. The P57 sequence contained a proline-rich N-terminal region analogous to an adhesin of Mycoplasma gallisepticum . The P78 protein was identical to a serologically defined phase-variant surface lipoprotein. Tn phoA mutagenesis provides an efficient means of systematically characterizing functionally diverse lipoproteins and other exported proteins in mycoplasmas.     

Lipoproteins of Mycobacterium tuberculosis: an abundant and functionally diverse class of cell envelope components

Mycobacterium tuberculosis remains the predominant bacterial scourge of mankind. Understanding of its biology and pathogenicity has been greatly advanced by the determination of whole genome sequences for this organism. Bacterial lipoproteins are a functionally diverse class of membrane-anchored proteins. The signal peptides of these proteins direct their export and post-translational lipid modification. These signal peptides are amenable to bioinformatic analysis, allowing the lipoproteins encoded in whole genomes to be catalogued. This review applies bioinformatic methods to the identification and functional characterisation of the lipoproteins encoded in the M. tuberculosis genomes. Ninety nine putative lipoproteins were identified and so this family of proteins represents ca. 2.5% of the M. tuberculosis predicted proteome. Thus, lipoproteins represent an important class of cell envelope proteins that may contribute to the virulence of this major pathogen.     

Fractionation of Tetrahymena ciliary membranes with triton X-114 and the identification of a ciliary membrane ATPase

Cilia were isolated from Tetrahymena thermophila, extracted with Triton X-114, and the detergent-soluble membrane + matrix proteins separated into Triton X-114 aqueous and detergent phases. The aqueous phase polypeptides include a high molecular mass polypeptide previously identified as a membrane dynein, detergent-soluble alpha and beta tubulins, and numerous polypeptides distinct from those found in axonemes. Integral membrane proteins partition into the detergent phase and include two major polypeptides of 58 and 50 kD, a 49-kD polypeptide, and 5 polypeptides in relatively minor amounts. The major detergent phase polypeptides are PAS-positive and are phosphorylated in vivo. A membrane-associated ATPase, distinct from the dynein-like protein, partitions into the Triton X-114 detergent phase and contains nearly 20% of the total ciliary ATPase activity. The ATPase requires Mg++ or Ca++ and is not inhibited by ouabain or vanadate. This procedure provides a gentle and rapid technique to separate integral membrane proteins from those that may be peripherally associated with the matrix or membrane.     

An improved strategy for selective and efficient enrichment of integral plasma membrane proteins of mycobacteria

Mycobacterial plasma membrane proteins play essential roles in many cellular processes, yet their comprehensive proteomic profiling remains challenging. This is mainly due to obstacles related to their extraction and solubilization. To tackle this problem, we have developed a novel procedure to selectively enrich mycobacterial plasma membrane proteins based on alkaline sodium carbonate washing of crude membranes followed by Triton X-114 phase partitioning. The present study assesses the efficiency of this method by proteome analysis of plasma membrane proteins from Mycobacterium bovis BCG. Extracted proteins were separated in parallel by 1-D SDS-PAGE and 2-DE and then analyzed by LC-MS/MS and MALDI-MS/MS. Our study revealed 125 proteins, of which 54 contained 1-14 predicted transmembrane domains (TMD) including nine novel proteins. The 1-D SDS-PAGE-based proteome analysis identified 81 proteins, of which 49 (60.5%) harbored TMD. This approach also revealed many hydrophobic membrane-associated/periplasmic proteins lacking TMD, but only few soluble proteins. The identified proteins were characterized with regard to biological functions and physicochemical properties providing further evidence for the high efficiency of the prefractionation method described herein.     

Phase separation of integral membrane proteins in Triton X-114 solution

A solution of the nonionic detergent Triton X-114 is homogeneous at 0 degrees C but separates in an aqueous phase and a detergent phase above 20 degrees C. The extent of this detergent phase separation increases with the temperature and is sensitive to the presence of other surfactants. The partition of proteins during phase separation in solutions of Triton X-114 is investigated. Hydrophilic proteins are found exclusively in the aqueous phase, and integral membrane proteins with an amphiphilic nature are recovered in the detergent phase. Triton X-114 is used to solubilize membranes and whole cells, and the soluble material is submitted to phase separation. Integral membrane proteins can thus be separated from hydrophilic proteins and identified as such in crude membrane or cellular detergent extracts.     

Putative lipoproteins of Streptococcus agalactiae identified by bioinformatic genome analysis

Streptococcus agalactiae is a significant pathogen causing invasive disease in neonates and thus an understanding of the molecular basis of the pathogenicity of this organism is of importance. N-terminal lipidation is a major mechanism by which bacteria can tether proteins to membranes. Lipidation is directed by the presence of a cysteine-containing 'lipobox' within specific signal peptides and this feature has greatly facilitated the bioinformatic identification of putative lipoproteins. We have designed previously a taxon-specific pattern (G+LPP) for the identification of Gram-positive bacterial lipoproteins, based on the signal peptides of experimentally verified lipoproteins (Sutcliffe I.C. and Harrington D.J. Microbiology 148: 2065-2077). Patterns searches with this pattern and other bioinformatic methods have been used to identify putative lipoproteins in the recently published genomes of S. agalactiae strains 2603/V and NEM316. A core of 39 common putative lipoproteins was identified, along with 5 putative lipoproteins unique to strain 2603/V and 2 putative lipoproteins unique to strain NEM316. Thus putative lipoproteins represent ca. 2% of the S. agalactiae proteome. As in other Gram-positive bacteria, the largest functional category of S. agalactiae lipoproteins is that predicted to comprise of substrate binding proteins of ABC transport systems. Other roles include lipoproteins that appear to participate in adhesion (including the previously characterised Lmb protein), protein export and folding, enzymes and several species-specific proteins of unknown function. These data suggest lipoproteins may have significant roles that influence the virulence of this important pathogen.     

The metabolism of neuropeptides. Phase separation of synaptic membrane preparations with Triton X-114 reveals the presence of aminopeptidase N.

The property of solutions of Triton X-114 to separate into detergent-rich and detergent-poor phases at 30 degrees C has been exploited to investigate the identities of the aminopeptidases in synaptic membrane preparations from pig striatum. When titrated with an antiserum to aminopeptidase N (EC 3.4.11.2), synaptic membranes solubilized with Triton X-100 revealed that this enzyme apparently comprises no more than 5% of the activity releasing tyrosine from [Leu]enkephalin. When assayed in the presence of puromycin, this proportion increased to 20%. Three integral membrane proteins were fractionated by phase separation in Triton X-114. Aminopeptidase activity, endopeptidase-24.11 and peptidyl dipeptidase A partitioned predominantly into the detergent-rich phase when kidney microvillar membranes were so treated. However, only 5.5% of synaptic membrane aminopeptidase activity partitioned into this phase, although the other peptidases behaved predictably. About half of the aminopeptidase activity in the detergent-rich phase could now be titrated with the antiserum, showing that aminopeptidase N is an integral membrane protein of this preparation. Three aminopeptidase inhibitors were investigated for their ability to discriminate between the different activities revealed by these experiments. Although amastatin was the most potent (IC50 = 5 X 10(-7) M) it failed to discriminate between pure kidney aminopeptidase N, the total activity of solubilized synaptic membranes and that in the Triton X-114-rich phase. Bestatin was slightly more potent for total activity (IC50 = 6.3 X 10(-6) M) than for the other two forms (IC50 = 1.6 X 10(-5) M). Puromycin was a weak inhibitor, but was more selective. The activity of solubilized membranes was more sensitive (IC50 = 1.6 X 10(-5) M) than that of the pure enzyme or the Triton X-114-rich phase (IC50 = 4 X 10(-4) M). We suggest that the puromycin-sensitive aminopeptidase activity that predominates in crude synaptic membrane preparations may be a cytosolic contaminant or peripheral membrane protein rather than an integral membrane component. Aminopeptidase N may contribute to the extracellular metabolism of enkephalin and other susceptible neuropeptides in the brain.     

Partitioning the proteome: phase separation for targeted analysis of membrane proteins in human post-mortem brain

Neuroproteomics is a powerful platform for targeted and hypothesis driven research, providing comprehensive insights into cellular and sub-cellular disease states, Gene × Environmental effects, and cellular response to medication effects in human, animal, and cell culture models. Analysis of sub-proteomes is becoming increasingly important in clinical proteomics, enriching for otherwise undetectable proteins that are possible markers for disease. Membrane proteins are one such sub-proteome class that merit in-depth targeted analysis, particularly in psychiatric disorders. As membrane proteins are notoriously difficult to analyse using traditional proteomics methods, we evaluate a paradigm to enrich for and study membrane proteins from human post-mortem brain tissue. This is the first study to extensively characterise the integral trans-membrane spanning proteins present in human brain. Using Triton X-114 phase separation and LC-MS/MS analysis, we enriched for and identified 494 membrane proteins, with 194 trans-membrane helices present, ranging from 1 to 21 helices per protein. Isolated proteins included glutamate receptors, G proteins, voltage gated and calcium channels, synaptic proteins, and myelin proteins, all of which warrant quantitative proteomic investigation in psychiatric and neurological disorders. Overall, our sub-proteome analysis reduced sample complexity and enriched for integral membrane proteins by 2.3 fold, thus allowing for more manageable, reproducible, and targeted proteomics in case vs. control biomarker studies. This study provides a valuable reference for future neuroproteomic investigations of membrane proteins, and validates the use Triton X-114 detergent phase extraction on human post mortem brain.     

Identification of Coxiella burnetii surface‐exposed and cell envelope associated proteins using a combined bioinformatics plus proteomics strategy

The Gram‐negative pathogen Coxiella burnetii is an intracellular bacterium that replicates within the phagolysosomal vacuoles of eukaryotic cells. This pathogen can infect a wide range of hosts, and is the causative agent of Q fever in humans. Surface‐exposed and cell envelope associated proteins are thought to be important for both pathogenesis and protective immunity. Herein, we propose a complementary strategy consisting of (i) in silico prediction and (ii) inventory of the proteomic composition using three enrichment approaches coupled with protein identification. The efficiency of classical Triton X‐114 phase partitioning was compared with two novel procedures; isolation of alkaline proteins by liquid‐phase IEF, and cell surface enzymatic shaving using biofunctional magnetic beads. Of the 2026 protein sequences analyzed using seven distinct bioinformatic algorithms, 157 were predicted to be outer membrane proteins (OMP) and/or lipoproteins (LP). Using the three enrichment protocols, we identified 196 nonredundant proteins, including 39 predicted OMP and/or LP, 32 unknown or poorly characterized proteins, and 17 effectors of the Type IV secretion system. We additionally identified eight proteins with moonlighting activities, and several proteins apparently peripherally associated with integral or anchored OMP and/or LP.     

Membrane subproteomic analysis of Mycobacterium bovis bacillus Calmette-Guérin

Mycobacterium bovis bacillus Calmette-Guérin (BCG) vaccine has been known for a long time to prevent tuberculosis (TB) worldwide since 1921. Nonetheless, we know little about BCG membrane proteome. In the present study, we utilized alkaline incubation and Triton X-114-based methods to enrich BCG membrane proteins and subsequently digested them using proteolytic enzyme. The recovered peptides were further separated by 2-D LC and identified by ESI-MS/MS. As a result, total 474 proteins were identified, including 78 integral membrane proteins (IMPs). Notably, 18 BCG IMPs were described for the first time in mycobacterium. Further analysis of the 78 IMPs indicated that the theoretical molecular mass distribution of them ranged from 8.06 to 167.86 kDa and pI scores ranged from 4.40 to 11.60. Functional classification revealed that a large proportion of the identified IMPs (67.9%, 53 out of 78) were involved in cell wall and cell processes functional group. In conclusion, here we reported a comprehensive profile of the BCG membrane subproteome. The present investigation may allow the identification of some valuable vaccine and drug target candidates and thus provide basement for future designing of preventive, diagnostic, and therapeutic strategies against TB.     

Triton X-114 cloud point extraction to subfractionate blood plasma proteins for two-dimensional gel electrophoresis

A simple and reproducible procedure for enrichment of a plasma protein subfraction suitable for two-dimensional polyacrylamide gel electrophoresis (2DE) was developed, using a Triton X-114-based cloud point extraction (CPE). Appropriate conditions for such a CPE procedure were found by SDS-PAGE to be a plasma protein concentration of about 10 mg/ml in 3% (w/v) Triton X-114. 2DE of proteins obtained by CPE of 400 μl of human plasma revealed about 200 spots constituting a spot pattern very different from the pattern of total plasma. The CPE procedure only had a limited contribution to the technical variation. Identification of about 60 spots, representing only 22 proteins, revealed that several proteins in the obtained subfraction were present in more isoforms or modifications. Among these were apolipoproteins (A-1, D, E, L1, and M), haptoglobin-related protein, phosphatidylcholine-sterol acyltransferase, serum amyloid A, and serum paraoxonase/arylesterase 1, which are proteins of a hydrophobic nature, as in plasma they relate to lipoprotein particles. Thus, Triton X-114-based CPE is a simple plasma prefractionation tool, attractive for detailed 2DE studies of hydrophobic plasma proteins and their isoforms or modifications.     

Identification of two Mycobacterium smegmatis lipoproteins exported by a SecA2-dependent pathway

The SecA2 protein is part of a specialized protein export system of mycobacteria. We set out to identify proteins exported to the bacterial cell envelope by the mycobacterial SecA2 system. By comparing the protein profiles of cell wall and membrane fractions from wild-type and DeltasecA2 mutant Mycobacterium smegmatis, we identified the Msmeg1712 and Msmeg1704 proteins as SecA2-dependent cell envelope proteins. These are the first endogenous M. smegmatis proteins identified as dependent on SecA2 for export. Both proteins are homologous to periplasmic sugar-binding proteins of other bacteria, and both contain functional amino-terminal signal sequences with lipobox motifs. These two proteins appeared to be genuine lipoproteins as shown by Triton X-114 fractionation and sensitivity to globomycin, an inhibitor of lipoprotein signal peptidase. The role of SecA2 in the export of these proteins was specific; not all mycobacterial lipoproteins required SecA2 for efficient localization or processing. Finally, Msmeg1704 was recognized by the SecA2 pathway of Mycobacterium tuberculosis, as indicated by the appearance of an export intermediate when the protein was expressed in a DeltasecA2 mutant of M. tuberculosis. Taken together, these results indicate that a select subset of envelope proteins containing amino-terminal signal sequences can be substrates of the mycobacterial SecA2 pathway and that some determinants for SecA2-dependent export are conserved between M. smegmatis and M. tuberculosis.     

Proteomic analysis of chlorosome-depleted membranes of the green sulfur bacterium Chlorobium tepidum

Green sulfur bacteria are obligate anaerobic phototrophs, which in addition to outer and plasma membranes contain chlorosomes. The analysis of the membrane proteome of Chlorobium tepidum from chlorosome-depleted membranes is described in this study. The membranes were purified by sucrose density centrifugation and characterized by 1-DE and 2-DE coupled with MS, absorption spectroscopy, and electron microscopy. 1-DE and 2-DE were employed to analyze the membrane proteins and to characterize the capabilities of the methods. Solubilization of the membrane proteins prior to 2-DE was improved by using a series of zwitterionic detergents. Based on the resolved spots after 2-DE, the combination of amidosulfobetaine 14 with Triton X-100 is more efficient than the combination of CHAPS, N-decyl-N,N-dimethyl-3-ammonio-1-propane sulfonate, and Triton X-100. From the application of 1-DE and 2-DE, 167 and 202 unique proteins were identified, respectively, using PMF by MALDI-TOF MS. Both methods resulted in the detection of 291 different proteins of which only 88 were predicted membrane proteins, indicating the limitation of membrane protein detection after separation with electrophoresis methods. In addition, 53 of these proteins were identified as outer membrane proteins.     

Selective release of the Treponema pallidum outer membrane and associated polypeptides with Triton X-114.

The effects of the nonionic detergent Triton X-114 on the ultrastructure of Treponema pallidum subsp. pallidum are presented in this study. Treatment of Percoll-purified motile T. pallidum with a 1% concentration of Triton X-114 resulted in cell surface blebbing followed by lysis of blebs and a decrease in diameter from 0.25-0.35 micron to 0.1-0.15 micron. Examination of thin sections of untreated Percoll-purified T. pallidum showed integrity of outer and cytoplasmic membranes. In contrast, thin sections of Triton X-114-treated treponemes showed integrity of the cytoplasmic membrane but loss of the outer membrane. The cytoplasmic cylinders generated by detergent treatment retained their periplasmic flagella, as judged by electron microscopy and immunoblotting. Recently identified T. pallidum penicillin-binding proteins also remained associated with the cytoplasmic cylinders. Proteins released by Triton X-114 at 4 degrees C were divided into aqueous and hydrophobic phases after incubation at 37 degrees C. The hydrophobic phase had major polypeptide constituents of 57, 47, 38, 33-35, 23, 16, and 14 kilodaltons (kDa) which were reactive with syphilitic serum. The 47-kDa polypeptide was reactive with a monoclonal antibody which has been previously shown to identify a surface-associated T. pallidum antigen. The aqueous phase contained the 190-kDa ordered ring molecule, 4D, which has been associated with the surface of the organisms. Full release of the 47- and 190-kDa molecules was dependent on the presence of a reducing agent. These results indicate that 1% Triton X-114 selectively solubilizes the T. pallidum outer membrane and associated proteins of likely outer membrane location.     

Tandem application of cationic colloidal silica and Triton X-114 for plasma membrane protein isolation and purification: towards developing an MDCK protein database

Plasma membrane (PM) proteins are attractive therapeutic targets because of their accessibility to drugs. Although genes encoding PM proteins represent 20-30% of eukaryotic genomes, a detailed characterisation of their encoded proteins is underrepresented, due, to their low copy number and the inherent difficulties in their isolation and purification as a consequence of their high hydrophobicity. We describe here a strategy that combines two orthogonal methods to isolate and purify PM proteins from Madin Darby canine kidney (MDCK) cells. In this two-step method, we first used cationic colloidal silica (CCS) to isolate adherent (Ad) and non-adherent (nAd) PM fractions, and then subjected each fraction to Triton X-114 (TX-114) phase partitioning to further enrich for hydrophobic proteins. While CCS alone identified 255/757 (34%) membrane proteins, CCS/TX-114 in combination yielded 453/745 (61%). Strikingly, of those proteins unique to CCS/TX-114, 277/393 (70%) had membrane annotation. Further characterisation of the CCS/TX-114 data set using Uniprot and transmembrane hidden Markov model revealed that 306/745 (41%) contained one or more transmembrane domains (TMDs), including proteins with 25 and 17 TMDs. Of the remaining proteins in the data set, 69/439 (16%) are known to contain lipid modifications. Of all membrane proteins identified, 93 had PM origin, including proteins that mediate cell adhesion, modulate transmembrane ion transport, and cell-cell communication. These studies reveal that the application of CCS to first isolate Ad and nAd PM fractions, followed by their detergent-phase TX-114 partitioning, to be a powerful method to isolate low-abundance PM proteins, and a useful adjunct for in-depth cell surface proteome analyses.