A Proteomic Characterization of the Plasma Membrane of Human Epidermis by High-Throughput Mass Spectrometry
2004; Elsevier BV; Volume: 123; Issue: 4 Linguagem: Inglês
10.1111/j.0022-202x.2004.23421.x
ISSN1523-1747
AutoresJosip Blonder, Atsushi Terunuma, Thomas P. Conrads, King C. Chan, Carole Yee, David A. Lucas, Carl F. Schaefer, Li‐Rong Yu, Haleem J. Issaq, Timothy D. Veenstra, Jonathan Vogel,
Tópico(s)Lipid Membrane Structure and Behavior
ResumoMembrane proteins are responsible for many critical cellular functions and identifying cell surface proteins on different keratinocyte populations by proteomic approaches would improve our understanding of their biological function. The ability to characterize membrane proteins, however, has lagged behind that of soluble proteins both in terms of throughput and protein coverage. In this study, a membrane proteomic investigation of keratinocytes using a two-dimensional liquid chromatography (LC) tandem-mass spectrometry (MS/MS) approach that relies on a buffered methanol-based solubilization, and tryptic digestion of purified plasma membrane is described. A highly enriched plasma membrane fraction was prepared from newborn foreskins using sucrose gradient centrifugation, followed by a single-tube solubilization and tryptic digestion of membrane proteins. This digestate was fractionated by strong cation-exchange chromatography and analyzed using microcapillary reversed-phase LC-MS/MS. In a set of 1306 identified proteins, 866 had a gene ontology (GO) annotation for cellular component, and 496 of these annotated proteins (57.3%) were assigned as known integral membrane proteins or membrane-associated proteins. Included in the identification of a large number of aqueous insoluble integral membrane proteins were many known intercellular adhesion proteins and gap junction proteins. Furthermore, 121 proteins from cholesterol-rich plasma membrane domains (caveolar and lipid rafts) were identified. Membrane proteins are responsible for many critical cellular functions and identifying cell surface proteins on different keratinocyte populations by proteomic approaches would improve our understanding of their biological function. The ability to characterize membrane proteins, however, has lagged behind that of soluble proteins both in terms of throughput and protein coverage. In this study, a membrane proteomic investigation of keratinocytes using a two-dimensional liquid chromatography (LC) tandem-mass spectrometry (MS/MS) approach that relies on a buffered methanol-based solubilization, and tryptic digestion of purified plasma membrane is described. A highly enriched plasma membrane fraction was prepared from newborn foreskins using sucrose gradient centrifugation, followed by a single-tube solubilization and tryptic digestion of membrane proteins. This digestate was fractionated by strong cation-exchange chromatography and analyzed using microcapillary reversed-phase LC-MS/MS. In a set of 1306 identified proteins, 866 had a gene ontology (GO) annotation for cellular component, and 496 of these annotated proteins (57.3%) were assigned as known integral membrane proteins or membrane-associated proteins. Included in the identification of a large number of aqueous insoluble integral membrane proteins were many known intercellular adhesion proteins and gap junction proteins. Furthermore, 121 proteins from cholesterol-rich plasma membrane domains (caveolar and lipid rafts) were identified. gene ontology liquid chromatography mass spectrometry tandem mass spectrometry two-dimensional polyacrylamide gel electrophoresisSCXstrong cation exchange The membrane proteome of a keratinocyte is comprised of an ensemble of membrane proteins whose presence is dependent on a given set of in situ conditions and is influenced by both the anatomical location and cellular differentiation status. Comprehensive profiles of membrane proteins in keratinocyte populations will facilitate our understanding of their critical roles in biological processes such as cell-to-cell adhesion, cell signaling, and ion transport. Additionally, profiling cell surface proteins will increase our understanding of the biology of different keratinocyte populations and facilitate target identification for developing biomedical therapeutics (Luche et al., 2003Luche S. Santoni V. Rabilloud T. Evaluation of nonionic and zwitterionic detergents as membrane protein solubilizers in two-dimensional electrophoresis.Proteomics. 2003; 3: 249-253https://doi.org/10.1002/pmic.200390037Crossref PubMed Scopus (202) Google Scholar;Wu and Yates, 2003Wu C.C. Yates J.R. The application of mass spectrometry to membrane proteomics.Nat Biotechnol. 2003; 21: 262-267https://doi.org/10.1038/nbt0303-262Crossref PubMed Scopus (486) Google Scholar). Comprehensive profiles for this class of proteins, however, are not yet available, in part, because enumeration of existing databases of keratinocyte proteins have relied on traditional two-dimensional polyacrylamide gel electrophoretic (2D-PAGE) approaches that have proved inadequate for analyzing integral membrane proteins (Celis et al., 1995Celis J.E. Rasmussen H.H. Gromov P. et al.The human keratinocyte two-dimensional gel protein database (update 1995): Mapping components of signal transduction pathways.Electrophoresis. 1995; 16: 2177-2240Crossref PubMed Scopus (109) Google Scholar). Consequently, more efficient approaches that utilize innovative sample preparation and fractionation methods with mass spectrometry (MS) identification are required to improve the representation of membrane proteins in such global proteomic analyses. Global proteome analysis of soluble proteins has progressed to the point to where thousands of cellular proteins can be identified (Koller et al., 2002Koller A. Washburn M.P. Lange B.M. et al.Proteomic survey of metabolic pathways in rice.Proc Natl Acad Sci USA. 2002; 99: 11969-11974https://doi.org/10.1073/pnas.172183199Crossref PubMed Scopus (333) Google Scholar). In a typical shotgun proteome analysis, cells are lysed and the soluble proteins are separated from insoluble proteins via centrifugation (Washburn et al., 2001Washburn M.P. Wolters D. Yates III., J.R. Large-scale analysis of the yeast proteome by multidimensional protein identification technology.Nat Biotechnol. 2001; 19: 242-247https://doi.org/10.1038/85686Crossref PubMed Scopus (3890) Google Scholar). After proteolytic digestion, typically using trypsin, the resulting peptides are separated via multidimensional liquid chromatography (LC) and identified by tandem MS (MS/MS) that provides sequence information from which the peptide, and hence the protein, can be identified. Unfortunately, because of their lack of solubility in aqueous buffers, no approach that parallels the simplicity of these methods has been developed for membrane proteins (Luche et al., 2003Luche S. Santoni V. Rabilloud T. Evaluation of nonionic and zwitterionic detergents as membrane protein solubilizers in two-dimensional electrophoresis.Proteomics. 2003; 3: 249-253https://doi.org/10.1002/pmic.200390037Crossref PubMed Scopus (202) Google Scholar;Wu and Yates, 2003Wu C.C. Yates J.R. The application of mass spectrometry to membrane proteomics.Nat Biotechnol. 2003; 21: 262-267https://doi.org/10.1038/nbt0303-262Crossref PubMed Scopus (486) Google Scholar). Further, sufficient experimental evidence exists to suggest that conventional 2D-PAGE-based proteomic techniques lack the capability for separation and subsequent MS identification of membrane proteins as currently achievable for soluble proteins (Santoni et al., 2000Santoni V. Molloy M. Rabilloud T. Membrane proteins and proteomic: Un amour impossible?.Electrophoresis. 2000; 21: 1054-1070https://doi.org/10.1002/(SICI)1522-2683(20000401)21:6 3.3.CO;2-#Crossref PubMed Scopus (0) Google Scholar;Wu and Yates, 2003Wu C.C. Yates J.R. The application of mass spectrometry to membrane proteomics.Nat Biotechnol. 2003; 21: 262-267https://doi.org/10.1038/nbt0303-262Crossref PubMed Scopus (486) Google Scholar). In an effort to redress these issues, several solution (i.e., non-gel)-based methods for global analysis of membrane proteins have been developed with the aim of circumventing the shortcomings of 2D-PAGE-based methods. These methods typically use detergents or strong organic acids to first solubilize the membrane proteins prior to enzymatic and/or chemical proteolysis (Han et al., 2001Han D.K. Eng J. Zhou H. Aebersold R. Quantitative profiling of differentiation-induced microsomal proteins using isotope-coded affinity tags and mass spectrometry.Nat Biotechnol. 2001; 19: 946-951https://doi.org/10.1038/nbt1001-946Crossref PubMed Scopus (817) Google Scholar;Washburn et al., 2001Washburn M.P. Wolters D. Yates III., J.R. Large-scale analysis of the yeast proteome by multidimensional protein identification technology.Nat Biotechnol. 2001; 19: 242-247https://doi.org/10.1038/85686Crossref PubMed Scopus (3890) Google Scholar). A recently described method employs a non-specific proteolytic cleavage of the extracellular, hydrophilic domains of integral membrane proteins that avoids the necessity to solubilize the entire integral membrane protein (Wu et al., 2003Wu C.C. MacCoss M.J. Howell K.E. Yates J.R. A method for the comprehensive proteomic analysis of membrane proteins.Nat Biotechnol. 2003; 21: 532-538https://doi.org/10.1038/nbt819Crossref PubMed Scopus (576) Google Scholar). This method does not disrupt the phospholipid bilayer, thus prevents digestion, and hence, subsequent MS identification, of the membrane-spanning domains of integral membrane proteins. Although many of these recently advanced solution-based approaches provide for wide membrane proteome coverage, they lack the simplicity and/or experimentally controlled proteolytic specificity of current methodologies available for proteomic analysis of soluble proteins. In this study, a single-tube extraction, solubilization, and digestion technique that permits complete solubilization and direct analysis of conventional tryptic fragments of membrane proteins was used to analyze the keratinocyte plasma membrane proteome isolated from human epidermal sheets. The ability to effectively analyze tissues is of critical importance because the cellular and membrane proteomes of cultured cells may not accurately resemble those in situ. In addition to identifying membrane proteins, it is critically important that any given technique be suitable for conducting quantitative proteome analyses of membrane proteins to compare the expression level of proteins in different keratinocyte populations. To enhance our understanding of epidermal biology, we aimed to obtain an extensive proteomic database of epidermal membrane proteins. Currently, proteomic databases of membrane proteins are lacking. Keratinocyte plasma membranes of newborn foreskins were isolated via sucrose-density gradient centrifugation and a buffered methanol solution was used to solubilize the membrane proteins, which were subsequently digested with trypsin in the same-buffered methanol solution. This digestate was fractionated using strong cation exchange liquid chromatography (SCXLC) prior to microcapillary reversed-phase (μRP) LC-MS/MS analysis. The results presented show that this simple technique not only provides the ability to identify large numbers of membrane proteins, but also enables identification of the actual membrane-spanning domains within these proteins. The focus of this study was to develop and utilize a high-throughput strategy to conduct a proteomic analysis of keratinocyte membrane proteins. The overall approach has been applied to enriched plasma membranes from epidermal cells that were solubilized and enzymatically digested in a buffered methanol solution. The key aspect of this approach is that complete and efficient tryptic digestion of solubilized membrane proteins can be carried out without requiring detergents or non-specific enzymatic proteolysis, enabling subsequent multidimensional fractionation strategies such as those widely employed in many proteomic investigations of global cell lysates. This strategy was developed and applied to a plasma membrane fraction purified from human epidermal sheets, isolated from newborn foreskins. The plasma membrane enrichment was determined by immunoblotting for α6 integrin, a known component of the keratinocyte plasma membrane (Sonnenberg et al., 1991Sonnenberg A. Calafat J. Janssen H. et al.Integrin-alpha-6-beta-4 complex is located in hemidesmosomes, suggesting a major role in epidermal-cell basement-membrane adhesion.J Cell Biol. 1991; 113: 907-917https://doi.org/10.1083/jcb.113.4.907Crossref PubMed Scopus (490) Google Scholar) and fractions 9, 10, and 11 were pooled for analysis (Figure 1). Proteins were solubilized by sonication in a 60% methanol-buffered solution and directly digested with trypsin. Previous studies conducted in our laboratory have shown that although the activity of trypsin is reduced in the presence of 60% methanol, it remains sufficient to completely digest proteome samples for downstream MS/MS peptide identification (Blonder et al, in press). The tryptic peptides were resolved into 80 fractions using SCXLC. Each of these SCXLC fractions were analyzed by μLC-MS/MS resulting in the identification of 2875 unique peptides corresponding to 1306 proteins. The complete list of the proteins identified in this analysis, along with the peptides from which they were identified, is provided in supplementary Table S1. Download .doc (.57 MB) Help with doc files Table S1Comprehensive list of 1305 identified proteins. A gene ontology (GO) annotation of molecular function, biological process, or cellular component was present for 1068 of the 1306 proteins, and 866 proteins had a GO annotation for cellular component or cellular location with 496 (57.3%) classified as either known integral membrane or membrane-associated proteins, including both plasma membrane and intracellular membrane proteins (Figure 2a). If a comparable percentage of the 440 (1306–866) identified proteins not yet assigned a GO annotation for cellular location are also membrane proteins, an additional 252 (440 × 57.3%) membrane proteins would be identified for a total of 748 membrane proteins. Within the population of integral membrane proteins, 58 (11.7%) are annotated in the database as receptor proteins (supplementary Table S2). This classification of receptor proteins was limited to proteins that contained the word "receptor" in their annotated description and therefore the list of identified receptor proteins is likely much greater. Although such classifications are highly dependent on both the accuracy and completeness of the database annotation, these results suggest that the methanol-based extraction, solubilization, and digestion employed in this study represents an efficient sample preparation strategy for global membrane proteomic investigations of tissues, utilizing multidimensional fractionation and MS/MS. Considering that approximately 20–30% of open reading frames (ORF) in the genome are predicted to encode for membrane proteins (Wallin and von Heijne, 1998Wallin E. von Heijne G. Genome-wide analysis of integral membrane proteins from eubacterial, archaean, and eukaryotic organisms.Protein Sci. 1998; 7: 1029-1038Crossref PubMed Scopus (1189) Google Scholar), it is evident that a significant enrichment of membrane proteins (57%) is achievable with this approach. Although considerable membrane proteins enrichment was achieved, a notable population of intracellular proteins was also identified. Whereas there are many possible explanations, related mostly to sample preparation, a significant amount of cytosol is captured within the plasma membrane vesicles during cell lysis and released later via the solubilization process. Download .doc (.06 MB) Help with doc files Table S2Receptor proteins identified in analysis of keratinocyte plasma membranes. In this investigation, 223 (of the 1306) proteins were identified as hypothetical proteins (supplementary Table S1), representing proteins that have not been well characterized and are lacking GO annotations. We sought to better understand the molecular nature of these hypothetical proteins and conducted an analysis to map putative helical transmembrane-spanning domains within these proteins. Of the 223 hypothetical proteins identified, 104 (46%) are predicted to contain at least one transmembrane domain, as shown in Figure 2b. Of these, 26 (25%) were predicted to contain six or more transmembrane domains with a single protein predicted to contain 23 transmembrane domains. One such protein (accession number Q9BTV4) is predicted to contain four transmembrane helices and although no information regarding the functional significance or cellular location of this protein is known for human cells, this protein has 97% amino acid identity to a mouse protein that has been shown to be expressed within the several different tissue types including hematopoietic stem cells and skin and has been mapped to the integral membrane (http://genome-www5.stanford.edu/cgi-bin/sourceResult). Another hypothetical protein (Q9ULK5) was found to contain four putative transmembrane domains. This protein has 89% amino acid identity to the loop tail-associated protein (LTAP) in mouse, as shown in supplementary Fig S1. Mouse LTAP has been shown to be an integral membrane protein involved in morphogenesis of the embryonic epithelium (Kibar et al., 2001Kibar Z. Vogan K.J. Groulx N. Justice M.J. Underhill D.A. Gros P. Ltap, a mammalian homolog of drosophila strabismus/van gogh, is altered in the mouse neural tube mutant loop-tail.Nat Genet. 2001; 28: 251-255https://doi.org/10.1038/90081Crossref PubMed Scopus (363) Google Scholar). Although present in a number of different tissue types, it is prominently expressed in blastocyst, trophoblast stem cells, and the neuroectoderm throughout early neurogenesis. Mouse LTAP has four mapped transmembrane domains that show good agreement with TMHMM mapping of the human homolog identified within this study. The hypothetical protein Q9ULK5 also contains the C-terminal PDZ domain (ETSV) as in the mouse homolog. Because these proteins are annotated as hypothetical, no direct evidence of their cellular location is available. The hypothetical proteins identified in this study, however, may represent novel keratinocyte membrane proteins, considering the number of predicted transmembrane segments and the high-level enrichment for membrane proteins in this study. Download .doc (.03 MB) Help with doc files Figure S1Predicted amino acid sequence of hypothetical human protein Q9ULK5 (upper row) aligned with that of mouse loop tail-associated protein (LTAP). The sites of absolute homology are in bold and the carboxy-terminal PDZ domain within both proteins is underlined. The approximate positions of the four predicted transmembrane domains for the mouse protein are shown above the aligned sequences. A recent proteomic analysis of membrane and soluble proteins from rat brain lysate by multidimensional protein identification technology (MudPIT) succeeded in the identification of 1610 proteins of which 454 (28%) were predicted to be integral membrane proteins (Wu et al., 2003Wu C.C. MacCoss M.J. Howell K.E. Yates J.R. A method for the comprehensive proteomic analysis of membrane proteins.Nat Biotechnol. 2003; 21: 532-538https://doi.org/10.1038/nbt819Crossref PubMed Scopus (576) Google Scholar). In this study, we were able to identify 496 integral membrane proteins from human epidermis, representing 57% of the proteins with a GO annotation for cellular location. Although these studies were carried out on different tissues and employed dissimilar sample preparation strategies in terms of protein isolation, solubilization, and enzymatic digestion, the results indicate that both approaches allow comparable results in terms of protein identifications and corresponding membrane protein coverage. Whereas the number of identified proteins is an important metric by which to judge the efficacy of a global proteomic analysis, a key aspect to bona fide protein characterization is the extent of protein coverage achievable for each protein. For membrane proteomic analysis, the ability to identify membrane-spanning peptides provides a reasonable measure of the efficiency of the sample preparation technique for disrupting interactions between the protein and the lipid bilayer. The ability to identify membrane-spanning peptides from a preparation employing a standard tryptic digestion depends on the ability of trypsin to gain access to accessible arginyl or lysyl residues near or within the membrane-imbedded domain. The ability of this sample preparation technique to characterize hydrophobic integral membrane proteins from complex mixtures is exemplified in Fig S2, which shows the identification of the facilitated glucose transporter member 1 (GLUT 1) protein, a widely distributed constitutive transporter protein whose expression has been linked to neoplastic transformation of malignant epithelial tissues (Baer et al., 1997Baer S.C. Casaubon L. Younes M. Expression of the human erythrocyte glucose transporter glut1 in cutaneous neoplasia.J Am Acad Dermatol. 1997; 37: 575-577Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar). Six peptides were identified from this protein, including the membrane-spanning peptide SFEMLILGR. In addition, peptides that are part of the membrane-spanning domains for the neutral amino acid transporter B and CTL2 proteins were also identified (Table I). The majority of identified proteins shown in Table I and supplemental Table S1 are localized within the plasma membrane, which indicates efficient solubilization and disruption of the lipid bilayer allowing direct tryptic digestion of integral membrane domains. For each protein in Table II, many of which are rarely seen in typical 2D-PAGE-based data sets, the corresponding GRAVY value is calculated; a positive value indicates that a protein is aqueous insoluble. Download .doc (.06 MB) Help with doc files Figure S2The sequence coverage of the facilitated glucose transporter protein (GLUT 1) that contains 12 mapped transmembrane domains (red/bold font). Tandem mass spectrum of the doubly charged precursor ion (1065.6 MH+, Xcorr 3.4568) detecting a fully tryptic transmembrane-spanning peptide SFEMLILGR that spans the fourth mapped transmembrane domain. Tandem mass spectrum of doubly charged precursor ion (1444.7 MH+, Xcorr 3.4194) representing a fully tryptic transmembrane peptide GTADVTHDLQEMK. The sequence of GLUT 1 showing all detected peptides (underlined) and the corresponding transmembrane domains (red/bold).Table ISelected subset of membrane proteins identified from keratinocyte plasma membranesDescriptionAccessionaAccession number, Swiss-Prot Release of 08/22/03.GRAVYbGRAVY value calculated using ProtParam tool at http://us.expasy.org/TMDcNumber of mapped transmembrane domains by the transmembrane hidden Markov model (TMHMM).PeptidedNumber of peptides identified.Protein locationGlucose transporter type 1P111660.532114126Plasma membraneSodium/potassium ATPase, α-1 braneP050230.0112411024Plasma membraneCalcium-transporting ATPase 1P20020-0.1763911012Plasma membraneCD9 antigenP219260.48105744Plasma membraneCation channel TRPM4BQ8TD43-0.02998364Plasma membraneOligosaccharyl transferaseP469770.237873136Plasma membraneChloride channel protein 7P517980.23851102Plasma membraneGap junction alpha-1 proteinP17302-0.2202145Plasma membraneGap junction beta-2 proteinP290330.2876143Plasma membraneTransmembrane protein PT27Q9HC070.3271652Plasma membraneMonocarboxylate transporter 1 (MCT 1)P539850.348001112Plasma membraneDB83 proteinP570880.4291534Plasma membraneSodium/hydrogen exchanger 1P196340.208711122Plasma membraneClaudin-1O958320.5336534Plasma membraneClaudin-3O155510.60532Plasma membraneCholine transporter-like proteinQ8WWI50.54551103Plasma membraneNeutral amino acid transporterQ157580.643807104Plasma membraneEquilibrative nucleoside transporterQ998080.684615112Plasma membraneLarge neutral amino acids transporterQ016500.739053114Plasma membraneDefender against cell deathP469660.82477932Plasma membraneCTL2 geneQ8IWA50.397592104Plasma membraneClathrin-coated vesicle proton pumpQ930500.01913485Plasma membranePotassium-transporting ATPase α chain 1P206480.07311492Plasma membraneProbable cation-transporting ATPaseQ9HD200.114037103Plasma membraneSolute carrier family 12 member 7Q9Y6660.120037122Plasma membraneLysophosphatidic acid acyltransferaseQ9NRZ70.14228742Plasma membraneTransmembrane 9 superfamily member 2Q998050.27179592Plasma membraneTransmembrane 9 superfamily member 3Q9HD450.2458494Plasma membraneTransmembrane 9 superfamily member 4Q925440.171294Plasma membraneBA207C16.3–hypothetical proteinQ9NQL60.25833494Plasma membraneHSPC121Q9P035-0.01876752Plasma membraneTranslocation protein SEC63 homologQ9UGP8-0.68115935Plasma membraneVacuolar ATP synthaseP274491.0432VacuolarVacuolar proton translocating ATPaseQ9Y4870.12920662VacuolarT cell immune response cDNA7 proteinQ134880.18012173VacuolarTransport protein Sec61P383780.56104EREndoplasmic reticulum calcium ATPaseP166150.096641107ERMannosyltransferase (Not56-like protein)Q926850.47557182ERa Accession number, Swiss-Prot Release of 08/22/03.b GRAVY value calculated using ProtParam tool at http://us.expasy.org/c Number of mapped transmembrane domains by the transmembrane hidden Markov model (TMHMM).d Number of peptides identified. Open table in a new tab Table IIIdentified transmembrane linker/receptor proteins within the keratinocyte adhesion apparatusDescriptionAccessionaAccession number, Swiss-Prot Release of 08/22/03.LocationPeptidebNumber of peptides identified.TMDcNumber of mapped transmembrane domains by the transmembrane hidden Markov model (TMHMM) at http://www.cbs.dtu.dk/services/TMHMM/GRAVYdGRAVY value calculated using ProtParam tool at http://us.expasy.org/Claudin-1O95832Tight junction440.53365Claudin-3O15551Tight junction240.605Junction adhesion moleculeQ9Y624Tight junction51-0.091973Cadherin-1P12830Adherens junction41-0.35136Cadherin-3P22223Adherens junction21-0.403498EGF-receptorP00533Plasma membrane81-0.316033Na,K-ATPase, β-3P54709Plasma membrane81-0.144086Desmoglein 1Q02413Desmosome111-0.275119Desmoglein 2Q14126Desmosome31-0.305640Desmoglein 3P32926Desmosome61-0.201301Desmocollin 1Q08554Desmosome61-0.441275Desmocollin 3Q14574Desmosome61-0.403013Gap junction α1 (C × 43)P17302Gap junction54-0.22021Gap junction β2 (C × 26)P29033Gap junction340.28761Gap junction β3 (C × 31)O75712Gap junction140.047778Gap junction β6 C × 30)O95452Gap junction140.054789Integrin α2P17301Adherens101-0.099915Integrin αVP06756Adherens31-0.219638Integrin α6P23229Adherens131-0.407611Integrin β1P05556Adherens51-0.406516Integrin β2P05107Adherens61-0.315604Integrin β4P16144Adherens121-0.447202a Accession number, Swiss-Prot Release of 08/22/03.b Number of peptides identified.c Number of mapped transmembrane domains by the transmembrane hidden Markov model (TMHMM) at http://www.cbs.dtu.dk/services/TMHMM/d GRAVY value calculated using ProtParam tool at http://us.expasy.org/ Open table in a new tab The number of identified proteins and the extent of their coverage are important measures by which to judge the efficacy of a global proteomic analysis. The utility of the proteomic approach can also be validated by the degree of the coverage of known plasma membrane proteins previously characterized using non-MS based methods. Cell adhesion is a complex and well-coordinated physiological process that is vital for maintaining the barrier and facilitating the overall biological function of the epidermis. The intercellular adhesion apparatus of the human epidermis is comprised of four distinct types of cell-to-cell junctions: tight junctions, adherens junctions, desmosomes, and gap junctions, whereas cell-to-matrix junctions are comprised of hemidesmosomes and focal contacts (Hentula et al., 2001Hentula M. Peltonen J. Peltonen S. Expression profiles of cell-cell and cell-matrix junction proteins in developing human epidermis.Arch Dermatol Res. 2001; 293: 259-267https://doi.org/10.1007/s004030100213Crossref PubMed Scopus (15) Google Scholar) (Figure 3). Most of the plasma membrane and membrane-associated proteins known to part of these adhesion structures have been identified in this proteome analysis. Tight junctions are the most apical component of the junctional complex in simple epithelia and contribute to apical–basolateral polarity, whereas in the epidermis, tight junctions are located in the granular and upper spinous layers (Tsuruta et al., 2002Tsuruta D. Green K.J. Getsios S. Jones J.C.R. The barrier function of skin: How to keep a tight lid on water loss.Trends Cell Biol. 2002; 12: 355-357https://doi.org/10.1016/S0962-8924(02)02316-4Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar) (Figure 3). In both types of epithelia, tight junctions are involved in controlling permeability and diffusion of solutes. Each tight junction contains three classes of integral membrane proteins: claudins, junction adhesion molecules, and occludins (Furuse et al., 2002Furuse M. Hata M. Furuse K. et al.Claudin-based tight junctions are c
Referência(s)