Association of stomatin with lipid-protein complexes in the plasma membrane and the endocytic compartment

Membrane protein - microvilli - lipid raft - GPI-anchored protein - epithelial cell The 31 kDa integral membrane protein stomatin (protein 7.2b) has a monotopic structure and a cytofacial orientation. We have shown previously that stomatin is located in plasma membrane protruding structures and forms high-order homo-oligomers in the human epithelial cell line UAC, suggesting that this protein has a structural function in the cortical morphogenesis of the cells. It is also present in a pool of juxtanuclear vesicles. In this study, we show that stomatin colocalizes with the GPI-anchored proteins placental alkaline phosphatase (PLAP) and membrane folate receptor alpha (MFRalpha) endogenously expressed in UAC cells. This observation enabled us to demonstrate two different aspects of stomatin. First, using anti-PLAP antibody internalization, we show that the peri-centrosomal vesicles containing stomatin correspond to a subset of endosomes, which can also be labeled with the late endosomal/lysosomal marker LAMP-2. Secondly, we found that stomatin is partially present in detergent-insoluble membrane domains and co-patches with PLAP on the plasma membrane, after cross-linking of PLAP by antibodies. These data indicate that stomatin and GPI-anchored proteins are linked through lipid rafts and undergo the same sorting events. We propose that stomatin, through its affinity for lipid rafts, functions in concentrating GPI-anchored proteins in membrane microvillar structures. Consistent with this hypothesis, we found that stomatin is expressed exclusively in microvilli of the apical membrane in polarized Madin-Darby canine kidney (MDCK) cells.

Isolation, molecular characterization, and tissue-specific expression of ECP-51 and ECP-54 (TIP49), two homologous, interacting erythroid cytosolic proteins

We isolated two proteins, ECP-51 and ECP-54, from human erythrocyte cytosol by affinity chromatography using a peptide of the integral membrane protein stomatin as bait. Partial amino acid sequence information obtained by microsequencing allowed us to clone the respective cDNAs. Analysis of the nucleotide sequences revealed that ECP-51 and ECP-54 are homologous (44.2% amino acid identity) and contain ATP-binding sites. ECP-54 was identified as TIP49/RUVBL1/NMP238, which is a component of a large nuclear protein complex, possibly the RNA polymerase II holoenzyme; ECP-51 is a novel protein. Using the two-hybrid system, we showed that these proteins interact with each other. The interaction of ECP-51 and ECP-54 with the stomatin peptide and the localization to the nucleus and cytoplasm suggest an additional function for these proteins as chaperone components.

Cysteine 29 is the major palmitoylation site on stomatin

The 31 kDa membrane protein stomatin was metabolically labeled with tritiated palmitic acid in the human amniotic cell line UAC and immunoprecipitated. We show that the incorporated palmitate is sensitive to hydroxylamine, indicating the binding to cysteine residues. Stomatin contains three cysteines. By expressing a myc-tagged stomatin and substituting the three cysteines by serine, individually or in combination, we demonstrate that Cys-29 is the predominant site of palmitoylation and that Cys-86 accounts for the remaining palmitate labeling. Disruption of Cys-52 alone does not show any detectable reduction of palmitic acid incorporation. Given the organization of stomatin into homo-oligomers, the presence of multiple palmitate chains is likely to increase greatly the affinity of these oligomers for the membrane and perhaps particular lipid domains within it.

Molecular cloning of hSLP-1, a novel human brain-specific member of the band 7/MEC-2 family similar to Caenorhabditis elegans UNC-24

We have isolated and characterized cDNA clones encoding a stomatin-like protein (hSLP-1) from a human cerebral cortex cDNA library. The deduced amino acid sequence (394 residues) revealed that hSLP-1 is a bipartite protein, containing a major stomatin-like part, starting at the N-terminus, and a non-specific lipid transfer protein (nsLTP)-domain at the C-terminal end, similar to the Caenorhabditis elegans protein UNC-24. Therefore, we conclude that hSLP-1 is the human homologue of UNC-24. In addition, the identification of an alternatively spliced variant demonstrated that two exon/intron boundaries are conserved in the hSLP-1 and unc-24 genes. Northern blot and RNA dot blot analyses showed that the 2. 2-kb transcript is mainly expressed in the brain, with the highest levels in the frontal lobe, cerebral cortex, caudate nucleus, amygdala, temporal lobe, putamen, substantia nigra, and hippocampus. This high-level expression of hSLP-1 in the basal ganglia may also reflect the evolutionary link to UNC-24.

Are bacterial proteins part of the matrix of kidney stones?

An extracellular protein, produced from Pseudomonas fluorescens strain D with molecular mass of 41.5 kDa was partially purified. Its first 12 amino acid sequence shows strong similarity to a sequence reported to belong to a protein isolated from a urate-calcium oxalate stone (Binnette & Binnette, Scan Microsc1994; 2: 233-239). A possible involvement of bacterial proteins in stone matrix is discussed.

Molecular characterization and tissue-specific expression of a murine putative G-protein-coupled receptor

We isolated by 5'- and 3'-RACE (rapid amplification of cDNA ends) clones from a murine brain cDNA library which encode a putative G-protein-coupled receptor. The composite nucleotide sequence revealed a coding region of 1197 nt; the deduced amino acid sequence of 399 amino acids showed 91.5% identity (95.7% similarity) when compared with the human homolog. An intron-like sequence, possibly involved in the regulation of expression, was found within the 5'-untranslated region. Northern blot analysis showed that the major 1.7-kb transcript is widely expressed, notably in brain and testis. In situ hybridization studies of tissue sections revealed high expression in neurons of the brain, epithelial cells of the lung, kidney and intestine, and in alveolar macrophages.

Oligomeric nature of the integral membrane protein stomatin

The 31-kDa integral membrane protein stomatin (protein 7.2b) is not only an important component of the red cell membrane but can also be found in abundance in different tissues and cell lines. The protein is thought to be anchored to the membrane by a hydrophobic domain while both N and C termini are exposed to the cytoplasm. We have previously shown in the human cell line UAC that stomatin concentrates preferentially in plasma membrane folds and protrusions. There is also evidence that stomatin is linked to the cortical actin cytoskeleton, suggesting a role in cortical morphogenesis of the cell. In this study, we demonstrate that the fundamental structure of stomatin is oligomeric. Whereas interaction of stomatin with itself was suggested by cross-linking experiments, we show by density gradient centrifugation analysis that soluble homo-oligomeric complexes of this protein are present in Triton X-100 extracts of UAC cells. We also show the existence of these oligomers by co-immunoprecipitation of the endogenous stomatin and a recombinantly expressed myc-tagged stomatin, using an anti-myc antibody. The data indicate that these complexes comprise between 9 and 12 monomers of stomatin. Two C-terminally truncated forms of stomatin do not incorporate into these oligomers, suggesting an involvement of the C terminus in the homo-oligomeric interaction.

Isolation, molecular characterization, and tissue-specific expression of a novel putative G protein-coupled receptor

We isolated a 40 kDa integral membrane protein (p40) from human erythrocyte ghosts by affinity chromatography, using a C-terminal peptide of stomatin, and obtained partial sequences which enabled us to isolate two full-length cDNAs from human bone marrow and fetal brain cDNA libraries. The cDNA sequences were identical and encoded a novel putative G protein-coupled receptor (399 amino acids). Northern and RNA dot blot analyses demonstrated that the major 4.8 kb-transcript is predominantly expressed in brain. In situ hybridization studies of tissue sections revealed high expression in neurons of the brain and spinal cord, in thymocytes, megakaryocytes, and macrophages.

Colocalization of stomatin (band 7.2b) and actin microfilaments in UAC epithelial cells

Cytolocalization of stomatin, an integral membrane protein also called erythrocyte band 7.2b, was investigated in a human epithelial cell line in which the expression of this protein is up-regulated after treatment with interleukin-6 and dexamethasone. A monoclonal antibody against stomatin was used to perform immunofluorescence and immunoelectron microscopy. The data show that stomatin concentrates preferentially in small plasma membrane protrusions. It is also found in abundance in a juxtanuclear structure possibly derived from the Golgi apparatus. Fluorescent double staining using the anti-stomatin antibody and the actin binding drug phalloidin shows a significant degree of colocalization of stomatin and cortical actin microfilaments. This association remains after actin filament disruption disruption by cytochalasin D treatment indicating a strong connection between stomatin and the membrane-associated cytoskeleton.

Cloning and analysis of a cDNA encoding the BALB/c murine erythrocyte band 7 integral membrane protein

cDNA clones encoding the BALB/c murine erythrocyte band 7 integral membrane protein (also termed protein 7.2b, or 'stomatin') were isolated by the screening of a corresponding bone-marrow lambda gt11 cDNA library with a human cDNA probe, and by 5'-RACE PCR cloning. Comparison of the murine, human and Caenorhabditis elegans protein 7.2b amino acid (aa) sequences revealed overall identities of 88% (human) and 61% (C. elegans), with the N-terminal domains showing only little similarity. The 7.2b protein sequences of the two mouse strains, BALB/c and C57BL/6J (B6), showed six rather conservative aa substitutions, three of them in the hydrophobic domain. The BALB/c murine mRNA, about 3.5 kb in size, is widely expressed in various tissues, most notably in spleen, lung and testis.

Multiple cytokines inhibit interleukin-6-dependent murine hybridoma/plasmacytoma proliferation

A panel of cytokines was tested for inhibitors of interleukin-6 (IL-6)-dependent cell proliferation. Murine type I and II interferons (mIFNs) strongly inhibited proliferation of IL-6-dependent B9 and 7TD1 cells in a dose-dependent manner. Human tumor necrosis factor-alpha (hTNF-alpha) and human transforming growth factor-beta (hTGF-beta) potently inhibited B9 and to a lesser extent 7TD1 cells, while hIL-11, human oncostatin M (hOSM), and human leukemia inhibitory factor (hLIF) had no inhibitory effects on IL-6-dependent growth. Conversely, IL-11 and OSM but not LIF stimulated B9 and 7TD1 cell growth. However, compared with IL-6, up to 1000-fold higher IL-11 and OSM concentrations were required to induce maximal cell proliferation. Increasing concentrations of IL-6 (up to 100 ng/ml) could not overcome the antiproliferative effects of mIFNs, hTNF-alpha and hTGF-beta. Supernatants from mIFN-gamma and lipopolysaccharide (LPS)-treated mouse macrophages (ANA-1 cell line) were tested in B9 cell assays to identify cytokines among stimulatory and inhibitory biological activities that can inhibit IL-6-dependent proliferation. Undiluted or relatively concentrated supernatants from ANA-1 macrophages treated with mIFN-gamma and/or LPS did not contain detectable IL-6 bioactivity. However, diluted samples contained considerable amounts of detectable IL-6 bioactivity (nanogram levels). Testing the same samples for IL-6 immunoreactivity using enzyme-linked immunoabsorbent assay revealed comparable levels of mIL-6. We conclude that IFNs, TNF-alpha, and TGF-beta and possibly other factors are potent, dominant inhibitors of IL-6-dependent plasmacytoma/hybridoma growth in vitro.

The organization of the gene (EPB72) encoding the human erythrocyte band 7 integral membrane protein (protein 7.2b)

The human gene EPB72 coding for the band 7 integral membrane protein, a major protein of the erythrocyte membrane, was isolated from a genomic DNA library and characterized. Spanning approximately 30 kb, the human EPB72 gene comprises seven exons ranging from 73 to 2331 bp; intron sizes range from 970 to approximately 11,200 bp. The first exon contains the 5'-untranslated region (61 nucleotides) and the coding sequence for the N-terminal domain; the second exon encodes the hydrophobic domain, including flanking cysteine and lysine residues. Exon 7 contains the C-terminal portion and a 2-kb 3'-untranslated region. The potential promoter region contains several consensus sequences for ubiquitous transcription factors (Sp1, AP1, AP2, CP1/2, NF kappa B, CREB, Ets-1, and CACC/GT-BF) and two imperfect sequences for erythroid factors (EKLF and GATA-1), in accordance with the ubiquitous distribution of the EPB72 mRNA in different cell types. No TATA box was apparent. An inverted Alu repeat element, flanked by nonamer direct repeats, was identified within the region -913/-620, relative to the cap site. Six additional Alu repeat elements, including one monomer and one trimer, were identified within the introns and the 3'-untranslated region. Two polyadenylation signals in the 3'-noncoding region of exon 7 enable the production of two mRNA species.

Genetic linkage analysis of the Ak1, Col5a1, Epb7.2, Fpgs, Grp78, Pbx3, and Notch1 genes in the region of mouse chromosome 2 homologous to human chromosome 9q

The genes for adenylate kinase-1 (AK1), folyl polyglutamate synthetase (FPGS), the collagen pro alpha 1(V) chain (COL5A1), erythrocyte protein band 7.2b (EPB72), and a proto-oncogene homeobox (PBX3) all map to the distal portion of human chromosome 9q (HSA9q) but have not previously been mapped by linking analysis in the mouse. In this study, we have used two interspecific backcrosses to map the mouse homologues of each of these genes to mouse chromosome 2 (MMU2). The Ak1, Col5a1, Epb7.2, Fpgs, and Pbx3 genes were mapped with respect to the genes for Grp78, Rxra, Notch1 (the mouse homologue of TAN1), Spna2, Abl, and Hc (the mouse homologue of C5), all of which have previously been mapped by linkage analysis on MMU2 and have human homologues that map to HSA9q. Two of the reference loci for MMU2, D2Mit1 and Acra, were also mapped in the same cross to facilitate comparisons with existing maps. The consensus gene order deduced by combining data from both crosses is D2Mit1-(Dbh,Notch1)-(Col5a1,Rxra)-Spna2-Ab l-(Ak1,Fpgs)- (Grp78,Pbx3)-(Epb7.2,Hc,Gsn)-Acra. These loci therefore form part of the conserved synteny between HSA9q and MMU2.

Identification of the phosphorylation site on human erythrocyte band 7 integral membrane protein: implications for a monotopic protein structure

Band 7 membrane protein was metabolically labelled with [32P]phosphate in the presence of cAMP, isolated and digested, the labelled peptides were purified and sequenced. Ser-9 was identified as the only phosphorylation site. This proves that the N-terminal region is located at the cytoplasmic side of the membrane and implies a monotopic rather than the predicted bitopic structure.

The gene coding for erythrocyte protein band 7.2b (EPB72) is located in band q34.1 of human chromosome 9

Human erythrocyte integral membrane protein band 7 (also termed protein 7.2b or stomatin) is involved in the Na+/K+ permeability of red cells. A cDNA clone coding for this protein was used as a probe to determine the chromosomal localization of the gene (EPB72). Southern blot analysis of somatic cell hybrid DNA panels localized the human EPB72 gene to chromosome band 9q34.1.

[Coping with illness and body image following different surgical techniques in breast cancer]

Coping and body image were studied in 80 patients with breast cancer. We compared three groups, who had undergone different operative strategies: temporary expander prothesis (n = 19), excisional biopsy (n = 20) and modified radical mastectomy (n = 40). Instruments were: the Herschbach coping questionnaire, the Strauss and Appelt body image questionnaire and a semi-structured interview. Patients after excisional biopsy showed significantly less illness-related stress than the other two groups, both in the number of all items named (p less than 0.02) and in the weighted total (p less than 0.05). Patients after modified mastectomy had chosen a different, "unfavourable" pattern of coping mechanisms than patients after expander prothesis or excisional biopsy: more resignation (p less than 0.05), less fighting and talking (p less than 0.05, respectively). In the body image questionnaire, no significant difference was found. However, in the semi-structured interview, patients after mastectomy mentioned more frequently that they avoided looking at the scar (p less than 0.05). These results underline the importance of preoperative counselling.

Cloning and nucleotide sequence of cDNA encoding human erythrocyte band 7 integral membrane protein

cDNA clones encoding the human erythrocyte band 7 membrane protein were isolated by immunoscreening from bone marrow and HeLa cell lambda gt 11 cDNA libraries, and their nucleotide sequences were determined. HeLa- and bone marrow cell-derived sequences were identical, except for one nucleotide; the deduced sequence of 287 amino acids was confirmed by sequence identity with peptides of the erythroid protein. Structure analysis assigned band 7 protein to the type Ib transmembrane proteins.

Isolation and partial characterization of the human erythrocyte band 7 integral membrane protein

Monoclonal antibodies to the Mr 31,000 major integral membrane protein of the human erythrocyte band 7 region were used to identify the corresponding polypeptide chain and epitope-carrying fragments on immunoblots. Analysis of the erythrocyte membrane, membrane fractions, and cytosol revealed that the Mr 31,000 band 7 integral membrane protein is unique and not related to any of the other water-soluble or membrane-bound band 7 components. Cross-reacting proteins were identified in the membranes of other mammalian erythrocytes and in cell lines of epithelial and lymphoid origin. Proteolytic digestion of intact human erythrocytes or erythrocyte membranes demonstrated that the band 7 integral membrane protein has an intracellular domain larger than Mr 12,000; it does not have an extracellular one. One of the monoclonal antibodies was employed for the isolation of band 7 integral membrane protein by immunoaffinity chromatography; subsequent Edman degradation revealed a blocked N-terminus.

Psychosomatic aspects of galactorrhea

We interviewed 33 women with non-puerperal galactorrhea, 5 (15%) of whom had HPRL levels greater than 18 ng/ml and 24 controls, namely women with benign breast lesions. We used a semi-structured interview covering the duration of symptoms, preceding life events and the effect on the relationship of the couple. We also used the Beck depression inventory, the Strauss and Appelt body image questionnaire and an 8 item Visual Analogue Scale (VAS). Galactorrheic women were more depressive (P less than 0.1), had more prior life events (P less than 0.001), longer duration of symptoms (P less than 0.01), and less fear of their disease (P less than 0.05) than did controls. Both groups had similar results with the body image questionnaire. Within the study group, results were independent of prolactin (HPRL) levels or amenorrhea.

Binding of recombinant interleukin-1 beta to the third complement component and alpha 2-macroglobulin after activation of serum by immune complexes

Activation of human normal serum with tetanus/antitetanus immune complexes (TAT-IC) resulted in increased binding of 125I-labeled interleukin-1 beta (IL-1 beta) to serum factors, as opposed to untreated serum. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by autoradiography showed labeling of two large molecular mass factors of an apparent molecular weight (Mr) of 200,000 and 400,000, respectively. These complexes could be dissociated by reduction. No complexes were formed when reducing compounds were added to serum-TAT-IC-125I-IL-1 beta mixtures. Complex formation was largely prevented by alkylating compounds. Molecular sieve chromatography of TAT-IC-activated serum confirmed that 125I-IL-1 beta became bound to high Mr serum proteins. Fractions containing high molecular 125I-IL-1 serum protein complexes partially retained IL-1-like activity since they induced proliferation of an IL-1-dependent murine T helper (D10G4) cell lineage. The 125I-IL-1 beta binding factors could be immunoprecipitated from TAT-IC-activated serum 125I-IL-1 beta solutions by antisera to alpha 2-macroglobulin (alpha 2M) or to the third complement component (C3). SDS-PAGE of the immunoprecipitates showed radioactive bands corresponding to the expected Mr resulting from complex formation between 125I-IL-1 beta and these two proteins. Treatment of purified plasma alpha 2M and C3 with trypsin or activation with methylamine, which causes cleavage of the internal thiol ester and the appearance of free thiol groups in these proteins, mediated binding of 125I-IL-1 beta to alpha 2M and C3b. The results suggest that cleavage of the internal thiol ester in C3 and alpha 2M makes these plasma proteins susceptible to binding of 125I-IL-1 beta and that free thiol groups do play a role in the formation of 125I-IL-1 beta plasma protein complexes. Activated C3 and alpha 2M may function as IL-1 beta carrier proteins in biologic fluids, in addition to their other physiologic roles.

Characterization of human red cell Rh (rhesus-)specific polypeptides by limited proteolysis

Human red cells of various Rh phenotypes were surface-labelled with 125I and the Rh-specific labelled polypeptides were isolated by preparative SDS-PAGE. The polypeptides were subjected to limited proteolysis and the resulting fragments were analysed by SDS-PAGE and autoradiography. Chymotryptic peptide maps of proteins obtained from Rh(D)-positive and -negative types appeared completely identical, whereas tryptic peptide maps revealed a difference: a fragment of Mr 17,500 was associated with the Rh(D) antigen, and one of Mr 19,000 with the Rh(C/c,E/e) antigens. Treatment of Rh polypeptides with carboxypeptidase Y prior to tryptic digestion resulted in a shift of nearly all tryptic fragments, including a fragment of Mr 8,000, indicating that the surface label was incorporated into the C-terminal part of the molecule.

Characterization of the human erythrocyte complement receptor CR1 (C3b receptor) by epitope mapping

Monoclonal antibodies and an anti-idiotypic serum against human complement receptor CR1 (C3b receptor, immune adherence receptor) were used to identify CR1 and some of its proteolytic fragments by an immunoblotting technique. The anti-idiotypic serum had a specificity for the C3b-binding site, as could be shown by its cross-reactivity with complement factor H. The monoclonal antibodies GARP-4 and GARP-37 were specific for epitopes located nearby the ligand-binding site, because they blocked the immune adherence reaction. For the immunoblotting technique, it was essential to use non-reducing conditions, since reduction of CR1 destroyed the epitopes. Therefore, mainly large (disulphide-linked) fragments of CR1 were obtained. A chymotryptic fragment of Mr 56,000 identified by GARP-4, was the smallest cleavage product to be associated with the C3b-binding domain. Different proteases gave CR1 degradation products of similar Mr, indicating the presence of distinct domains, three of which had a Mr approximately 38,000. A schematic model of CR1 substructure was deduced from the epitope mapping data.

Chemical and carbon-13 nuclear magnetic resonance studies of the blood group M and N active sialoglycopeptides from human glycophorin A

The NH2-terminal sialoglycopeptides from human erythrocyte glycophorin A have been obtained by specific proteolytic cleavage and gel filtration chromatography. By cyanogen bromide cleavage, a glycosylated octapeptide was obtained from blood group M donors having an amino acid composition and 13C NMR spectrum consistent with the structure (formula: see text) was demonstrated. By Staphylococcus aureus protease cleavage, a glycosylated pentapeptide was obtained from N donors having the same structure as II, without the carboxyl-terminal sequence Val . Ala . Hse. Methanolysis/gas chromatographic analysis and 13C NMR spectroscopy of I and II and their asialo derivatives reveal that the M- and N-active sialoglyco-octapeptides both have identical oligosaccharide structures, each containing three O-linked tetrasaccharides with the structure NeuNAc alpha 2-3Gal beta 1-3(NeuNAc alpha 2-6)GalNAc alpha 1-O-Ser(Thr). The demonstration of the anomeric form of GalNAc-peptide linkages revealed by 13C NMR has previously been inaccessible by chemical analysis. Conformationally, I and II appear identical and both manifest several unusual resonance shifts suggestive of a glycopeptide secondary structure involving four specific hydrogen bonds. Calcium ion titration was also found to induce shifts in the NeuNAc 13C resonances that may be of functional significance. Serological studies reveal that both the M and N glyco-octapeptides and the N glyco-pentapeptide retain all of the M and N activity of the parent structure. Deamination and/or desialylation completely destroys this activity. These data are consistent with a model in which the M or N determinant is the NH2-terminal amino acid and a NeuNAc residue(s). From these data it is concluded that there is no chemical basis for assertions in the literature that M and N antigens differ in their oligosaccharide structure or that the N antigen is biosynthetically transformed to the M antigen by sialylation.