Soluble precursor of membrane-associated protein kinase in uterine smooth muscle cells

Incubation of smooth muscle strips from rat uterus with isoproterenol resulted in redistribution of protein kinase activity between the cytosol and a 20,000 to 50,000g membrane fraction. Similarities in the elution properties of the cytosolic and membrane-associated forms of the enzyme on DEAE-cellulose ion exchange chromatography further suggested the two forms were the same. The nature of membrane binding of the soluble enzyme was investigated using smooth muscle microsomal and cytosol fractions. Membranes readily bound the soluble enzyme when the two subcellular compartments were reconstituted and incubated at 30 °C for 10 min. The extent of binding was proportional to the ratio of membranes to cytosol and was characterized by the inhibition of soluble enzyme activity toward exogenous substrates in a Triton X-100 reversible manner. In marked contrast to the binding of soluble protein kinase to heart particulate fractions, binding of the cytosol enzyme to smooth muscle cell membranes was unaffected by ionic strength or cAMP. The latter property indicated holoenzyme was bound in a manner similar to the free catalytic subunit of cAMP-dependent protein kinase and suggested the enzyme was bound by association between the membrane and the catalytic subunit. Binding of cytosol protein kinase to the membranes rendered the enzyme insensitive to trypsin digestion and the capacity of the smooth muscle cell membranes to bind the soluble enzyme exceeded that of other rat tissue fractions. Resistance to salt extraction and proteolysis, as well as its detergent dependence, suggested the soluble enzyme became an integral or intrinsic membrane protein following association with the membrane. The ability of membranes to incorporate [γ-³²P]ATP into phosphoprotein was lost on detergent extraction of protein kinase and restored in an apparently specific manner when extracted and washed membranes were reconstituted with soluble enzyme. The intrinsic nature of membrane protein kinase and the apparent specificity with which the soluble enzyme was hound by membranes further indicated that, in myometrium. hormone-induced translocation of protein kinase is an important mechanism by which enzyme activity is increased in the vicinity of its in situ substrates.     

Characterization and distribution of protein kinase C in ovarian tissue

Although protein kinase C, an enzyme dependent on calcium, phospholipid and diacylglycerol, has been found in high levels in ovarian tissues, its biologic function is yet unknown. In initial studies on the role of this enzyme in regulating ovarian functions, we compared protein kinase C activity in subcellular fractions of porcine corpora lutea and medium follicles. Highest protein kinase C-specific activities were found in the cytosol, followed by microsomes and mitochondria for both follicles and luteal tissues. Solubilization of all membrane-containing fractions by 0.2% Triton X-100 was required for full expression (a 4-fold average increase) of protein kinase activity. Extraction of membrane fractions with 0.5 M NaCl or sonication in a hypotonic medium revealed that 90% of the total mitochondrial protein kinase C activity and 50% of the microsomal activity was tightly membrane-bound. Characterization of both cytosolic and Triton X-100 extracted membrane preparations of luteal tissue by diethylaminoethyl (DEAE)-cellulose chromatography revealed a single peak of protein kinase C activity eluting at 80 mM NaCl. Cytosolic fractions of corpora lutea contained 3 times more protein kinase C-specific activity than did cytosolic fractions of follicles. In contrast, mitochondria from medium follicles contained 30% more specific protein kinase C activity than did luteal mitochondria. These higher cytosolic levels of protein kinase C-specific activity in corpora lutea suggest that the enzyme may play an important role in the process of luteinization or in the regulation of luteal function.     

Characterization of rat mast cell granule proteins.

Mast cell granules free of membranes were isolated by differential centrifugation of water-lysed cells. The granules were extracted sequentially with 0.5, 1.0, and 2.0 m KCl. The 0.5 m fraction contained 95% of the N-acetyl-β-glucosaminidase activity; this enzyme probably accounts for no more than 1% of the total granule protein. The 1.0 m fraction contained more than 80% of the granule chymotrypsin-like activity; the chymotrypsin-like enzyme was calculated to represent at least 15% of total granule protein. Heparin was found largely in the 1.0 m extract and in the residue after 2.0 m extraction. The heparin in both fractions had a molecular weight by gel exclusion chromatography considerably in excess of commercial porcine heparin. Acrylamide-gel electrophoresis of granules dissolved in 1% sodium dodecyl sulfate and reduced with dithiothreitol demonstrated four major protein bands. The 1.0 m fraction contained the most prominent, rapidly migrating band. The more slowly migrating, higher molecular weight bands appeared in greater proportion in the 2.0 m and residue fractions. Autodigestion of the 1.0 m extract permitted purification of the mast cell chymotrypsin-like enzyme to specific activities as high as that of crystallized bovine pancreatic α-chymotrypsin. The mast cell chymotrypsin-like enzyme purified in this way migrated on dodecyl sulfate-gel electrophoresis as a single major band with an estimated molecular weight of 29,000.     

Comparison of methods used to separate the inner and outer membranes of cell envelopes of Campylobacter spp

The outer membrane of Campylobacter coli, C. jejuni and C. fetus cell envelopes appeared as three fractions after sucrose gradient centrifugation. Each outer membrane fraction was contaminated with succinate dehydrogenase activity from the cytoplasmic membrane fraction. Similarly the inner membrane fraction was contaminated with 2-ketodeoxyoctonate and outer membrane proteins including the porin(s). The separation of these two membranes was not facilitated by variations in lysozyme treatment, cell age, presence or absence of flagella, or longer lipopolysaccharide chain length. Sodium lauroyl sarcosinate extraction resulted in an outer membrane fraction which contained some inner membrane contamination and produced multiple bands upon sucrose gradient centrifugation. Triton X-100 extraction removed the inner membrane from the outer membrane and Triton X-100/EDTA treatment extracted lipopolysaccharide-rich regions of the outer membrane which contained almost exclusively the Campylobacter porin(s). These data indicated that the inner and outer membranes of the Campylobacter cell envelope were very difficult to separate, possibly because of extensive fusions between these two membranes.     

Hemicellulose fractions and associated protein of lupin hypocotyl cell walls

The alkali extraction of polysaccharide fractions from depectinated primary cell walls of lupin hypocotyls was studied using sequential extractions at 0° and 18–22°. Aqueous 10% KOH at 0° removed hemicellulose-A (95%) heteroglycan-B (80%) and linear 1–4 linked hemicellulose-B (60%). Arabinose accounts for 88% of the monosaccharides of the linear 1–4 linked hemicellulose-B extracted between 2 and 5 h at 18–22°. Extraction of the 0° and 18–22° alkali-soluble fractions by denaturants, was also examined. 6M guanidine thiocyanate removed about 60% of the 0° 10% KOH soluble polysaccharide but little of the 18–22° soluble material. Although rapidly extracted by 10% KOH at 0° the hemicellulose-A is not extracted by this reagent. Analyses of cell walls and extracted fractions showed that there is little change in amino acid composition and little extraction of wall protein upon removal of about 60% of total wall hemicellulose with 10% KOH at 0°. It is therefore not bound to the wall through galactosylserine links. 10% KOH at 18–22° caused a marked change in composition and extracted most of the wall protein. An alkali resistant fraction high in hydroxyproline and low in serine was not extracted by 24% KOH at 18–22° in 24 hr.     

Influence of calcium on the subcellular distribution of protein kinase C in human neutrophils. Extraction conditions determine partitioning of histone-phosphorylating activity and immunoreactivity between cytosol and particulate fractions

Activation of the neutrophil respiratory burst is thought to involve a translocation and activation of protein kinase C. We report that the presence of Ca2+ during the disruption of unstimulated human neutrophils and cytoplasts resulted in an increase in protein kinase C activity (histone phosphorylation) and immunoreactive protein kinase C species in the particulate (membrane) fraction and a reduction in such activities in the cytosol. This Ca2+-induced translocation of activity was concentration-dependent and occurred at physiologically relevant concentrations of Ca2+ (30-500 nM). The Ca2+-induced membrane association of protein kinase C could be reversed by removal of Ca2+. These findings indicate that the Ca2+ concentration of the extraction buffer can determine the subcellular distribution of protein kinase C in disrupted cells and suggest that the observed location of this enzyme activity in cell fractions may not necessarily reflect the localization in intact cells. These results also raise the possibility that the distribution of protein kinase C between cytosol and membrane is a dynamic equilibrium controlled by levels of free Ca2+. Thus, Ca2+ might regulate distribution as well as activation of protein kinase C.     

High-Pressure Extraction of Membrane-Associated Protein Kinase C from Rat Brain

Extraction of rat brain membrane-associated protein kinase C with high specific activity was obtained by applying benzyl alcohol (a membrane fluidizer), EDTA, and high hydrostatic pressures. Approximately 50% of total brain-associated activity was extracted from membranes. The pressure-extracted activity had an eightfold enrichment in the lipid/protein ratio when compared with the cytosolic fraction. This may explain the inability of exogenous diacylglycerol to stimulate endogenous phosphorylation in pressure-extracted activity. The enzyme is extracted at greater than 1,300 atm, a result indicating it most likely has a portion inserted into the hydrophobic portion of the membrane bilayer. Perturbation of the native membrane induces a change in the membrane-associated protein kinase C-lipid interaction that permits extraction under conditions used for the cytosolic species. This is the first report of conversion of the endogenous membrane species to a cytosolic one and may be important in determining the role of protein kinase C in neuronal regulation.     

A screening procedure for the solubilization of chloroplast membrane proteins from the marine green macroalga Ulva lactuca using RP-HPLC-MALDI-MS

A protocol for purification and analysis of chloroplast membrane proteins in the green macroalga Ulva lactuca has been developed, including reversed phase high performance liquid chromatography (RP-HPLC) and matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS). Five different solvents were evaluated for extraction of membrane proteins by three methods. The highest protein yield was achieved when proteins were extracted directly from the chloroplasts using the solvent hexafluoroisopropanol. A range of proteins of increasing hydrophobicity was separated by HPLC. Analysis of both HPLC fractions and non-separated samples by MALDI-TOF-MS revealed proteins with molecular weights spanning between 1 and 376 kDa.     

Characterization and topology of the membrane domain Nqo10 subunit of the proton-translocating NADH-quinone oxidoreductase of Paracoccus denitrificans

The proton-translocating NADH-quinone oxidoreductase (NDH-1) of Paracoccus denitrificans is composed of 14 different subunits (Nqo1-Nqo14). Of these, seven subunits (Nqo7, Nqo8, and Nqo10-14) which are equivalent to the mitochondrial DNA-encoded subunits of complex I constitute the membrane segment of the enzyme complex; the remaining subunits make up the peripheral part of the enzyme. We report here on the biochemical characterization and heterologus expression of the Nqo10 subunit. The Nqo10 subunit could not be extracted from the Paracoccus membranes by NaI or alkaline treatment, which is consistent with the presumed membrane localization. By using the maltose-binding protein (MBP) fusion system, the Nqo10 subunit was overexpressed in Escherichia coli. The MBP-fused Nqo10 was expressed in membrane fractions of the host cell and was extractable by Triton X-100. The extracted fusion protein was then isolated by one-step affinity purification through an amylose column. By using immunochemical methods in conjunction with cysteine-scanning mutagenesis and chemical modification techniques, the topology of the Nqo10 subunit expressed in E. coli membranes was determined. The data indicate that the Nqo10 subunit consists of five transmembrane segments with the N- and C-terminal regions facing the periplasmic and cytoplasmic sides of the membrane, respectively. In addition, the data also suggest that the proposed topology of the MBP-fused Nqo10 subunit expressed in E. coli membranes is consistent with that of the Nqo10 subunit in the native Paracoccus membranes. From the experimentally determined topology together with computer prediction programs, a topological model for the Nqo10 subunit is proposed.     

Incorporation of tritiated thymidine into DNA of rat liver: a critique of various evaluation methods]

We have measured the incorporation of 3H-(methyl)-thymidine by cell cultures of rat foetal liver and in vivo by the livers of young rats stimulated by casein, in order to compare three methods for the extraction of DNA. The DNA was extracted by three different techniques: perchloric acid precipitation, trichloroacetic acid precipitation and phenol extraction, and its specific activity was determined. The radioactive labelling was also determined for the lipid, ribonucleic acid and protein fractions for the two first methods, in both of which 70 p. cent of the incorporated tritium is found in the DNA fraction and about 10 p. cent in each of the other fractions. The determination of the specific radioactivity of DNA gives similar results for the three extraction methods. However, since larger yields were obtained by both acid precipitation techniques than by phenol extraction, we believe them to be more suitable for studies on cell cultures.     

Activation and membrane binding of carboxypeptidase E

Carboxypeptidase E (CPE) is a carboxypeptidase B-like enzyme that is thought to be involved in the processing of peptide hormones and neurotransmitters. Soluble and membrane-associated forms of CPE have been observed in purified secretory granules from various hormone-producing tissues. In this report, the influence of membrane association on CPE activity has been examined. A substantial amount of the membrane-associated CPE activity is solubilized upon extraction of bovine pituitary membranes with either 100 mM sodium acetate buffer (pH 5.6) containing 0.5% Triton X-100 and 1 M NaCl, or by extraction with high pH buffers (pH greater than 8). These treatments also lead to a two- to threefold increase in CPE activity. CPE extracted from membranes with either NaCl/Triton X-100 or high pH buffers hydrolyzes the dansyl-Phe-Ala-Arg substrate with a lower Km than the membrane-associated CPE. The Vmax of CPE present in extracts and membrane fractions after the NaCl/Triton X-100 treatment is twofold higher than in untreated membranes. Treatment of membranes with high pH buffers does not affect the Vmax of CPE in the soluble and particulate fractions. Pretreatment of membranes with bromoacetyl-D-arginine, an active site-directed irreversible inhibitor of CPE, blocks the activation by NaCl/Triton X-100 treatment. Thus the increase in CPE activity upon extraction from membranes is probably not because of the conversion of an inactive form to an active one, but is the result of changes in the conformation of the enzyme that effect the catalytic activity.     

Icthyophthirius multifiliis has membrane-associated immobilization antigens

Sera from fish that survive infections with the ciliated protozoon, Ichthyophthirius multifiliis, immobilize the parasite in vitro. In order to identify cell surface antigens involved in the immobilization response, integral membrane proteins were extracted from tomites with Triton X-114 and used to immunize rabbits. The rabbit antisera immobilized the parasite in vitro and antigens were localized to cell and ciliary plasma membranes by indirect immunofluorescent microscopy. The membrane protein fractions from both whole cells and tomite cilia were characterized by 1- and 2-dimensional SDS-PAGE. A 43,000-dalton (D) glycoprotein with an isoelectric point of 7.0 is the predominant protein in these fractions, comprising 12% and 60% of the total protein of whole cell and ciliary membranes, respectively. Western blot analysis of ciliary proteins with immune rabbit sera indicated that the 43,000-D glycoprotein is the principal antigen.     

Separation of the primary dehydrogenase from the cytochromes of the nicotinamide adenine dinucleotide (reduced form) oxidase of Bacillus megaterium

A selective extraction procedure was developed for sequentially extracting a fraction containing the primary dehydrogenase and a fraction containing the cytochromes of the nicotinamide adenine dinucleotide (reduced form) (NADH) oxidase of Bacillus megaterium KM membranes. The primary dehydrogenase (NADH-2,6-dichlorophenolindophenol oxidoreductase) activity was extracted from sonically treated membranes with 0.4% sodium deoxycholate for 30 min at 4 C. The insoluble residue was extracted with 0.4% sodium deoxycholate in 1 m KCl for 30 min at 25 C. A combination of the two extracts and dilution in Mg(2+) gave good recovery of the original membrane NADH oxidase activity. The primary dehydrogenase fraction contained 41% of the membrane protein, no cytochromes, flavine adenine dinucleotide as the sole acid-extractable flavine, and most of the membrane ribonucleic acid (RNA). The cytochrome-containing fraction had 16% of the membrane protein, 61% of the membrane cytochrome with the same relative amounts of cytochromes a and b as the original membrane, no acid-extractable flavine, little RNA, and no oxidoreductase activity. The oxidoreductase fraction remained soluble after removal of deoxycholate whereas the cytochrome fraction became insoluble after removal of deoxycholate-KCl, but the precipitated fraction could be redissolved in 0.4% sodium deoxycholate. Treatment of both fractions with ribonuclease to destroy all of the RNA present did not affect the ability of the fractions to recombine into a functional oxidase unit. Treatment of either fraction with phospholipase A prevented restoration of a functional oxidase when the oxidoreductase and cytochrome fractions were treated in solution, but no affect on restoration of oxidase was observed when the phospholipase A treatment was carried out with the soluble oxidoreductase fraction and the insoluble cytochrome fraction.     

Palmitic Is the Main Fatty Acid Carried by Lipids of Detergent-Resistant Membrane Fractions from Neural and Non-Neural Cells

Lipids extracted from detergent-resistant membrane fractions, thought to derive from membrane domains, were analyzed for fatty acid composition. The proportion of palmitic acid in fractions isolated from neurons (cerebellar granule cells) and from neural-like cell lines (neuroblastoma-glioma NG108-15) nearly doubled (reaching about 54% of total fatty acids) with respect to cell WCL, indicating their enrichment in palmitic acid-carrying lipids. The proportion of palmitic acid in detergent-resistant fractions obtained from caveolin-transfected NG108-15 cells was comparable with that obtained from caveolin-negative cells, ruling out a specific role of this protein in recruiting palmitoylated lipid species. The enrichment in palmitic acid was remarked also in membrane fractions isolated from non-neuronal cell lines (A431) using either detergents or detergent-free techniques. Lipid fractionation and mass spectrometry experiments show that palmitic acid–rich phosphatidylcholine species are responsible of the peculiar fatty acid composition of these fractions. All together these results suggest that the enrichment in palmitic acid–rich phosphatidylcholine species is a common feature of neural and non-neural cell lines and may play a major role in the biogenesis of membrane domains.     

The neutrophil glycoprotein Mo1 is an integral membrane protein of plasma membranes and specific granules

The glycoprotein Mo1 has previously been demonstrated to be on the cell surface and in the specific granule fraction of neutrophils and to be translocated to the cell surface during degranulation. It is not known, however, whether Mo1 is an integral membrane protein or a soluble, intragranular constituent loosely associated with the specific granule membrane. Purified neutrophils were disrupted by nitrogen cavitation and separated on Percoll density gradients into four fractions enriched for azurophilic granules, specific granules, plasma membrane, and cytosol, respectively. The glycoproteins in these fractions were labeled with 3H-borohydride reduction, extracted with Triton X-114, and immunoprecipitated with 60.3, an anti-Mo1 monoclonal antibody Mo1 was detected only in the specific granule and plasma membrane fractions and partitioned exclusively into the detergent-rich fraction consistent with Mo1 being an integral membrane protein. In addition, treatment of specific granule membranes with a high salt, high urea buffer to remove absorbed or peripheral proteins failed to dissociate Mo1. These data support the hypothesis that Mo1 is an integral membrane protein of plasma and specific granule membranes in human neutrophils.     

Protein kinases of rat liver endoplasmic reticulum. Solubilisation, partial characterisation and comparison with protein kinases of rat liver cytosol.

Protein kinase associated with rat liver microsomes was only partly extracted by treatment with 1.5 M KCl. The enzyme was solubilised by Triton X-100 or sodium deoxycholate at the same or slightly higher detergent concentrations than microsomal marker components. The enzyme activity increased 2-3 fold upon solubilisation. Three peaks with protein kinase activity (fractions MI, MII and MIII) were resolved on DEAE-cellulose chromatography. Fraction MIII but not fractions MI or MII was activated by adenosine 3':5'-monophosphate (cyclic AMP). All fractions catalysed the phosphorylation of protamine and histones but not that of casein or phosvitin. Fractions MI and MIII had a similar substrate specificity and phosphorylated histones at a relatively much higher rate than did fraction MII. The isoelectric points were 8.1 for fraction MI, 5.5 for fraction MII and 4.9 for fraction MIII. On incubation of fraction MIII with cyclic AMP it was split into two catalytically active components with pI 8.1 and 7.35. The component with pI 8.1 was predominant and corresponded to fraction MI. Five protein kinase peaks were resolved from rat liver cytosol by DEAE-cellulose chromatography. Three of them (fractions CIa, CIIb and CIII) had the same properties as each of the microsomal kinase fractions. A forth fraction (CIIa) was cyclic-AMP-dependent and had the same substrate specificity as fractions MI and MIII. Its pI was 5.1, and it was split into two components by cyclic AMP (pI 8.1 and 7.35). In binding studies fraction CIIb bound more efficiently to microsomes than fraction CIII, while fractions CIa, CIIa and the microsomal protein kinase fractions did not bind appreciably. When microsomes were treated with trypsin exposed protein kinase was inactivated and the latency of the remaining enzyme increased substantially. Most of fraction MII was inactivated by trypsin while fraction MIII was resistant. The possible orientation of protein kinase fractions MII and MIII in the microsomal membrane is discussed.     

Ichthyophthirius multifiliis Has Membrane-Associated Immobilization Antigens

Sera from fish that survive infections with the ciliated protozoon, Ichthyophthirius multifiliis , immobilize the parasite in vitro. In order to identify ceil surface antigens involved in the immobilization response, integral membrane proteins were extracted from tomites with Triton X-l14 and used to immunize rabbits. The rabbit antisera immobilized the parasite in vitro and antigens were localized to cell and ciliary plasma membranes by indirect immunofluorescent microscopy. The membrane protein fractions from both whole cells and tomite cilia were characterized by 1 - and 2-dimensiona! SDS-PAGE. A 43,000-dalton (D) glycoprotein with an isoelectric point of 7.0 is the predominant protein in these fractions, comprising 12% and 60% of the total protein of whole cell and ciliary membranes, respectively. Western blot analysis of ciliary proteins with immune rabbit sera indicated that the 43,000-D glycoprotein is the principal antigen.     

Studies on protein kinase C tightly-bound to rat liver plasma membrane and its protease-activated form

1. Rat liver plasma membrane contained two types of protein kinase C which could be extracted by Ca2(+)-chelator and detergent, respectively. The activities of these two enzymes were nearly equivalent. 2. The detergent-extracted protein kinase C, tightly-bound to membrane, was separated into two subtypes by hydroxyapatite column chromatography. Based on the elution profile and the Ca2+/phospholipid requirement, the major and the minor components were identified as type III and type II protein kinase C, respectively. 3. The detergent-extracted protein kinase C was converted to an active fragment with Mr 45,000 by limited proteolysis with trypsin. Incubation under physiological level of ionic strength increased the stability of this active enzyme and protected it from further inactivation by trypsin. 4. Phosphorylation of H1 histone by the protease-activated kinase was stimulated 1.5-2-fold by phosphatidylserine. However, this enzyme phosphorylated multiple proteins in rat liver subcellular fractions in Ca2(+)- and phospholipid-independent manner. 5. These results suggest that the protein kinase C (mainly type III enzyme) tightly-bound to rat liver plasma membrane may have important role through protein phosphorylation by the native or the protease-activated kinase.     

Detection of lateral heterogeneity in the cytoplasmic membrane of<Emphasis Type="Italic">Bacillus subtilis</Emphasis>

Lateral heterogeneity in the cytoplasmic membrane of Bacillus subtilis was found by using density gradient centrifugation. Crude membranes (CM) present in the whole cell lysate were separated into three fractions of increasing density (F, CI, CII). Substantial difference exists in the amount of protein recovered from these fractions, the relative ratio being 15 : 35 : 50. The qualitative protein composition (by SDS-PAGE) of the fractions varies markedly as well. The lipid components extracted from the fractions are also distributed in different proportions, viz. 40 : 40 : 20. The spectrum of fatty acids (FA), detected in lipids of F fraction and analyzed by GC-MS exhibits the same profile as that found in CM; in contrast, fractions CI and CII undergo extensive FA reconstruction. Thermotropic behavior of fractions measured by the steady-state fluorescence anisotropy of 1,6-diphenyl-1,3,5-hexatriene indicates significant variations of microviscosity (r(s)) within the F, CI and CII fractions. The protein-to-lipid ratio plays evidently a key role in affecting the physical state of the cytoplasmic membrane. Microdomains of different density coexist in the membrane and exhibit heterogeneity in both chemical composition and "physical state"; the increased de novo synthesis of FA induced by the cold exclusively in fractions CI and CII indicates correlation with an altered physiological state of bacterial metabolism.     

Proteomics of Medicago sativa cell walls

A method for the sequential extraction and profiling by two-dimensional gel electrophoresis (2-DE) of Medicago sativa (alfalfa) stem cell wall proteins is described. Protein extraction included freezing, grinding in a sodium acetate buffer, separation by filtration of cell walls from cytosolic contents, and extensive washing. Cell wall proteins were then extracted sequentially with a solution containing 200 mM CaCl2 and 50 mM sodium acetate, followed by extraction with 3.0 M LiCl and 50 mM sodium acetate. Cell wall proteins from both the CaCl2 and LiCl fractions were profiled by 2-DE. Approximately 150 protein spots were extracted from these two gels, digested with trypsin, and analyzed using nanoscale HPLC coupled to a hybrid quadrupole time-of-flight (Q-tof) tandem mass spectrometer (LC/MS/MS). More than 100 proteins were identified and used in conjunction with the 2-DE profiles to generate proteomic reference maps for cell walls of this important legume. Identified proteins include classical cell wall proteins as well as proteins traditionally considered as non-secreted. Two unique extracellular proteins were also identified.