Identification, purification, and subcellular localization of prostate-specific membrane antigen PSM' protein in the LNCaP prostatic carcinoma cell line

An alternatively spliced variant of prostate-specific membrane antigen (PSMA) designated PSM' was originally described following identification of its mRNA in normal prostate. We have purified the PSM' protein from LNCaP cells using two immunoaffinity columns in tandem. The first column contained a monoclonal antibody (7E11) that was reactive with the NH2 terminus of PSMA, which specifically depleted the LNCaP lysate of full-length PSMA. The nonbinding fraction was then passed over a second column composed of a monoclonal antibody (PEQ226.5), the epitope of which was located within the 134-437 domain of PSMA and shared with PSM'. The protein eluted from the second immunoaffinity column produced a Mr 95,000 band on SDS-PAGE, which was slightly lower than the full-length PSMA at Mr 100,000. The band was NH2-terminally sequenced through 15 residues, and the assigned sequence coincided with the predicted sequence for PSM' protein minus the first two NH2 terminus amino acids. The PSM' protein, therefore, began with residue 60 of PSMA (alanine). LNCaP cells were fractionated, and PSM' was localized to the cytoplasm.     

Molecular dissection of effector cell protease receptor-1 recognition of factor Xa. Assignment of critical residues involved in antibody reactivity and ligand binding

Receptor-mediated assembly of blood proteases on vascular cells maintains the hemostatic balance and initiates intracellular signal transduction. Effector cell protease receptor-1 (EPR-1) is an approximately 62-kDa vascular cell membrane receptor for the clotting protease factor Xa, participating in thrombin formation and lymphocyte activation. Here, recombinant EPR-1 fragments were engineered in the frame of intercellular adhesion molecule-1, transfected in mammalian cells, and analyzed for antibody recognition and ligand binding. Chimeric transfectants containing the EPR-1 sequence Met1-Arg60 bound the immunosuppressive anti-EPR-1 monoclonal antibody (mAb) 2E1. In contrast, transfected cells expressing the EPR-1 sequence Pro120-Ala154 were recognized by the functionally inhibitory anti-EPR-1 mAbs 9D4 and B6, bound 125I-factor Xa in a reaction quantitatively indistinguishable from that of wild-type EPR-1 transfectants, and promoted factor Xa concentration-dependent prothrombin activation in the absence of exogenous factor V/Va. Chimeric transfectants expressing the COOH terminus end of the EPR-1 extracellular domain (Ala157-Glu221) did not bind anti-EPR-1 mAbs and did not associate with factor Xa. Mutagenesis of Asn131 or Lys133 in the EPR-1 ligand recognition domain abolished factor Xa binding by 80 +/- 5.5 and 96 +/- 4%, respectively, while mutation of Lys126, Gly128, Asn129, and Asn134 was without effect. A synthetic peptide duplicating the EPR-1 sequence S123PGKPGNQNSKNEPP137 dose dependently inhibited factor V/Va-independent thrombin generation of resting endothelium (IC50 approximately 1 microM), while the adjacent EPR-1 sequence P136PKK-RERERSSHCYP150 was ineffective. These findings demonstrate that EPR-1 contains two spatially distinct functional domains implicated in lymphocyte activation (Met1-Arg60) or factor Xa binding and prothrombin activation (Pro120-Ala154). These interacting sequences may provide a novel potential target for inhibition of factor Xa-dependent vascular cell responses.     

Mapping, genomic organization and promoter analysis of the human prostate-specific membrane antigen gene

Prostate-specific membrane antigen (PSMA) is a 100 kDa type II transmembrane protein with folate hydrolase and NAALAdase activity. PSMA is highly expressed in prostate cancer and the vasculature of most solid tumors, and is currently the target of a number of diagnostic and therapeutic strategies. PSMA is also expressed in the brain, and is involved in conversion of the major neurotransmitter NAAG (N-acetyl-aspartyl glutamate) to NAA and free glutamate, the levels of which are disrupted in several neurological disorders including multiple sclerosis, amyotrophic lateral sclerosis, Alzheimer's disease and schizophrenia. To facilitate analysis of the role of PSMA in carcinoma we have determined the structural organization of the gene. The gene consists of 19 exons spanning approximately 60 kb of genomic DNA. A 1244 nt portion of the 5' region of the PSMA gene was able to drive the firefly luciferase reporter gene in prostate but not breast-derived cell lines. We have mapped the gene encoding PSMA to 11p11-p12, however a gene homologous, but not identical, to PSMA exists on chromosome 11q14. Analysis of sequence differences between non-coding regions of the two genes suggests duplication and divergence occurred 22 million years ago.     

Prostate-specific membrane antigen: a novel folate hydrolase in human prostatic carcinoma cells

A novel monoclonal antibody has been developed that reacts strongly with human prostatic cancer, especially tumors of high grade. This antibody (7E11C-5) is currently in Phase 3 trials as an imaging agent for metastatic disease. We have cloned the gene that encodes the antigen that is recognized by the 7E11C-5 monoclonal antibody and have designated this unique protein prostate-specific membrane (PSM) antigen. PSM antigen is a putative class II transmembranous glycoprotein exhibiting a molecular size of Mr 94,000. Functionally, class II membrane proteins serve as transport or binding proteins or have hydrolytic activity. Preliminary studies have demonstrated binding of pteroylmonoglutamate (folate) to membrane fractions that also cross-reacted with the PSM monoclonal antibody. We observed substantial carboxypeptidase activity as folate hydrolase associated with PSM antigen. The purpose of our study was to demonstrate that human prostatic carcinoma cells expressing PSM antigen exhibit folate hydrolase activity using methotrexate triglutamate (MTXGlu3) and pteroylpentaglutamate (PteGlu5) as substrates. Isolated membrane fractions from four human prostate cancer cell lines (LNCaP, PC-3, TSU-Prl, and Duke-145) were examined for folate hydrolase activity using capillary electrophoresis. After timed incubations at various pH ranges and in the presence and absence of thiol reagents, separation of pteroyl(glutamate)n derivatives was achieved with an electrolyte of sodium borate and SDS, while absorbance was monitored at 300 nm. The results demonstrate clearly that LNCaP cells, which highly express PSM, hydrolyze gamma-glutamyl linkages of MTXGlu3. The membrane-bound enzyme is an exopeptidase, because it progressively liberates glutamates from MTXGlu3 and PteGlu5 with accumulation of MTX and PteGlu1, respectively. The semipurified enzyme has a broad activity from pH 2.5 to 9.5 and exhibits activity maxima at pH 5 and 8. Enzymatic activity is maintained in the presence of reduced glutathione, homocysteine, and p-hydroxymercuribenzoate (0.05-0.5 mm) but was inhibited weakly by DTT (>/=0.2 mm). By contrast to LNCaP cell membranes, membranes isolated from other human prostate adenocarcinoma cells (PC-3, Duke-145, and TSU-Pr1) did not exhibit comparable hydrolase activity, nor did they react with 7E11-C5 monoclonal antibody. After transfection of PC-3 cells with a full-length 2.65-kb PSM cDNA subcloned into a pREP7 eukaryotic expression vector, non-PSM antigen-expressing PC-3 cells developed immunoreactivity to 7E11-C5 monoclonal antibody and demonstrated folate hydrolase activities and optimum pH activity profiles identical to those of LNCaP cells. The membrane-bound enzymes from both LNCaP- and PC-3-transfected cells also have a capacity to hydrolyze an alpha-linked glutamyl moiety from N-acetyl-alpha-aspartylglutamate. We have identified that PSM antigen is a pteroyl poly-gamma-glutamyl carboxypeptidase (folate hydrolase) and is expressed strongly in human prostate cancer. Cancer cells that express this enzyme are resistant to methotrexate therapy. Those developing future therapeutic strategies in the treatment of prostate cancer that utilize folate antagonists need to consider this mechanism of resistance.     

N-terminal and central regions of the human CD44 extracellular domain participate in cell surface hyaluronan binding

CD44 molecules are cell surface receptors for hyaluronan (HA). To define regions of the extracellular domain of CD44 that are important for HA binding, we have studied the ability of HA-blocking CD44 mAbs to bind to CD44 from a variety of sources. Five CD44 mAbs (5F12, BRIC235, 3F12, BU-75, and HP2/9) of 21 studied were identified that at least partially blocked FITC-labeled HA (HA-FITC) binding to the standard form of CD44 (CD44S) in CD44-transfected Jurkat cells. Analysis of reactivity of HA-blocking CD44 mAbs defined three distinct epitopes. Lack of reactivity of mAb 5F12 with a CD44 fusion protein (CD44-Rg) containing an N-terminal truncation of 20 amino acids (aa), as well as reactivity of mAb 5F12 with an N-terminal CD44 synthetic peptide (CD44-9A), demonstrated that the N-terminal proximal region of CD44 (aa 1 to 20) was involved in mAb 5F12 binding. A mutant cell line, CEM-NKR, derived from the T-ALL cell line, CEM, did not bind mAb 5F12 nor bind HA, whereas wild-type CEM did bind mAb 5F12 and HA. Sequence analysis of wild-type CEM and CEM-NKR CD44 cDNA demonstrated a G to A point mutation at position 575 in the CD44 cDNA of CEM-NKR, resulting in an arginine to histidine mutation at aa position 154. Taken together, our studies demonstrated that there are three epitopes to which HA-blocking mAbs bind in the extracellular domain of CD44, and that the CD44 N-terminal proximal and central regions are two regions in the extracellular domain of CD44 that may interact and either mediate or regulate HA binding to cell surface CD44.     

Structure-function study of the extracellular domain of the human IFN-alpha receptor (hIFNAR1) using blocking monoclonal antibodies: the role of domains 1 and 2

We have performed a structure-function analysis of extracellular domain regions of the human IFN-alpha receptor (hIFNAR1) using mAbs generated by immunizing mice with a soluble hIFNAR1-IgG. Five mAbs described in this study recognize different epitopes as determined by a competitive binding ELISA and by alanine substitution mutant analyses of the hIFNAR1-IgG. Two mAbs, 2E1 and 4A7, are able to block IFN-stimulated gene factor 3 (ISGF3) formation and inhibit the antiviral cytopathic effect induced by several IFN-alpha (IFN-alpha 2/1, -alpha 1, -alpha 2, -alpha 5, and -alpha 8). None of these anti-IFNAR1 mAbs were able to block activity of IFN-beta. mAb 4A7 binds to a domain 1-hIFNAR1-IgG but not to a domain 2-hIFNAR1-IgG, which suggests that its binding region is located in domain 1. The binding of the most potent blocking mAb, 2E1, requires the presence of domain 1 and domain 2. The most critical residue for 2E1 binding is a lysine residue at position 249, which is in domain 2. These findings suggest that both domain 1 and domain 2 are necessary to form a functional receptor and that a region in domain 2 is important. IFN-beta recognizes regions of the hIFNAR complex that are distinct from those important for the IFN-alpha.     

Definition of the specificity of monoclonal antibodies against porcine CD45 and CD45R: report from the CD45/CD45R and CD44 subgroup of the Second International Swine CD Workshop

Swine cell binding analyses of a set of 48 monoclonal antibodies (mAbs), including eleven standards, assigned to the CD44 and CD45 subset group of the Second International Swine CD Workshop yielded 13 clusters. Although none of these corresponded to CD44, seven mAbs formed a cluster which was identified as being specific for restricted epitopes of CD45 (CD45R). In addition, a T-cell subset specific cluster comprised of four mAbs was also identified. Two mAbs (STH106 and SwNL 554.1) reacted exclusively with CD8 bright lymphocytes, the other two (2B11 and F01G9) with a subset of CD4 lymphocytes. The other 10 clusters were either specific for MHC-class I like molecules or overlapped with clusters identified by the adhesion molecule subgroup and are therefore just briefly discussed in this report. The specificity of all the mAbs in the CD45R cluster was verified by their ability to immunoprecipitate distinct proteins and to react with CHO cells expressing individual isoforms of CD45. Three CD45R mAbs (3a56, MIL5, -a2) did react with a 210 kDa isoform(s), while another three (STH267, FG2F9, 6E3/7) only recognized a 226 kDa isoform(s). The remaining one (MAC326) precipitated both a 210 and 226 kDa protein. The specificity of all the mAbs in the CD45R cluster, and of the CD45 common mAbs, was confirmed by their reactivity with CHO cells transfected with cDNAs encoding the extracellular and transmembrane portions of distinct CD45R isoforms. Those mAbs recognizing a 210 kDa protein reacted with CHO cells expressing the CD45RC isoform, while those capable of precipitating a 226 kDa, but not the 210 kDa, polypeptide recognized CHO cells expressing either the CD45RAC and the relatively rare CD45RA isoform. MAC326 was unique in its inability to react with CHO cells engineered to produce the CD45RC and CD45RAC isoforms. Thus, three mAbs (6E3/7, STH267, and FG2F9) appear to be specific for an epitope(s) encoded by the A exon, while one (MAC326) recognizes a determinant encoded by the C exon. The remaining three mAbs (3a56, -a2, MIL5) are apparently specific for an epitope(s) which results from the fusion of the C exon to the invariant leader sequence and is destroyed by inclusion of the A exon. All three CD45 common mAbs, K252.1E4, MAC323 and 74.9.3, did react with the CHO cells lines expressing either the CD45RA, CD45RC, CD45RAC or CD45RO isoforms, but not with untransfected CHO cells. When the natural expression of CD45 isoforms was examined by reacting lymphocytes with CD45R mAbs, a high level expression of isoforms containing the A exon-generated domain was detected in all B cells while the majority of CD4+ T cells had undetectable or lower expression density of this protein than B cells. In contrast, the density of expression of the CD45 isoform(s) containing the C exon-generated domain ranged from undetectable to high in CD4+ T cells whereas the amounts were approximately ten-fold lower in B cells. Overall this panel of CD45 mAbs will be very useful in analyzing the maturation and differentiation of swine lymphoid cells subsets.     

Estrogen induces adenosine deaminase gene expression in MCF-7 human breast cancer cells: role of estrogen receptor-Sp1 interactions

Adenosine deaminase (ADA) gene expression is induced by 17beta-estradiol (E2) in MCF-7 human breast cancer cells, whereas the antiestrogens 4'-hydroxytamoxifen and ICI 182,780 exhibit partial estrogen receptor (ER) agonist/antagonist and antagonist activities, respectively. Previous studies have shown that the -211 to +11 region of the ADA gene promoter contains six GC-rich sites (I-VI) that bind Sp1 protein, and these elements are required for high basal expression. In transient transfection studies with pADA211, which contains the -211 to +11 ADA gene promoter linked to a bacterial chloramphenicol acetyl transferase (CAT) reporter gene, E2 and tamoxifen (but not ICI 182,780) induced CAT activity. Ligand-induced transactivation was observed only in cells cotransfected with expression plasmids for wild-type ER or HE11, which does not contain the DNA-binding domain of the ER. Cotransfection with HE15 and HE19, which contain the DNA-binding domain and activation function-1 (AF-1) and AF-2 of the ER, respectively, did not result in E2-induced activity. Subsequent deletion analysis of the ADA gene promoter showed that Sp1 binding site IV (-79 to -73) was primarily responsible for hormone responsiveness. ER activation of ADA gene expression is another example of an E2-induced gene that is dependent on ER/Sp1 interactions with a site-specific GC-rich motif.     

Elevated expression of eIF4E in confined early breast cancer lesions: possible role of hypoxia

The translation-initiation factor eIF4E is rate-limiting for protein synthesis, and its over-expression results in oncogenic transformation of mammalian cells. eIF4E facilitates the synthesis of several powerful tumor angiogenic factors (FGF-2 and VEGF) by selectively enhancing their translation. In breast carcinomas, eIF4E is commonly over-expressed, but the pathology where this elevation is initially manifested is presently unknown. To probe whether the elevation of eIF4E marks an early stage of cancer development, we focused our research on early cancerous lesions. We have analyzed 70 invasive ductal carcinomas (IDCs), 78 ductal carcinomas in situ (DCIS), 51 benign lesions and 4 model cell lines for elevated expression of eIF4E by several different methods: Northern/Western blots, immuno-histochemistry and in situ RT-PCR. eIF4E expression was markedly increased in IDC and in islets of viable cells in the center of poorly vascularized DCIS, which are not easily identifiable by standard histological stains. We also show that expression of eIF4E is increased by hypoxia and, presumably, in hypoxic areas of these lesions. We propose that clonal expansion of cancer cells, permanently over-expressing eIF4E, gives them a critical advantage to survive hypoxia and marks the transition toward the vascular phase of cancer progression. Hence, eIF4E may be useful in stratifying DCIS lesions according to their malignant stage.     

Fibrinogen coating density affects the conformation of immobilized fibrinogen: implications for platelet adhesion and spreading

Adhesion of platelets to immobilized fibrinogen appears to play an important role in a variety of physiologic and pathologic phenomena. We previously observed that the fibrinogen concentration used to coat polystyrene wells affected the morphology and distribution of GP IIb/IIIa receptors on the surface of platelets adherent to the fibrinogen. One possible explanation for these differences is that fibrinogen immobilized at high density adopts a different conformation than fibrinogen immobilized at low density. To address this possibility, we studied the binding of a panel of anti-fibrinogen monoclonal antibodies (mAbs) to fibrinogen immobilized at different coating densities. Three different patterns of binding were observed: 1) a linear increase in binding to wells coated with 1-10 microg/ml fibrinogen, followed by a lesser increase or plateau at higher fibrinogen concentrations (mAbs Fd4-4E1, Fd4-7B3, 1D4, 4-2); 2) minimal reactivity at all fibrinogen concentrations (mAbs GC4-1A12, 2C34); 3) a biphasic response, with a linear increase up to 10 microg/ml fibrinogen and then a significant decline in binding at higher fibrinogen concentrations (mAbs 311, 31A9, FPA 19/7, 9C3, 1C5-A5/2, 44-3). The patterns of mAb binding to fibrinogen immobilized from plasma were similar. Most mAbs that demonstrated a biphasic response bound poorly or not at all to soluble fibrinogen, while mAbs that demonstrated a linear/plateau response were able to bind soluble fibrinogen. At equal surface densities, mAbs that bound biphasically, particularly mAb 1C5-A5/2, were more reactive to urea-denatured than native fibrinogen. mAbs 1C5-A5/2 and 44-3 are specific for gamma 1-78 and 95-265, respectively, suggesting that the fibrinogen gamma-chain may be sensitive to changes in conformation induced by immobilization. In summary, these data suggest that fibrinogen immobilized at 1-10 microg/ml adopts a conformation unlike soluble fibrinogen, while fibrinogen immobilized at > 30 microg/ml adopts a more solution-like conformation. These differences in fibrinogen conformation may partially account for the ability of platelets to bind to immobilized fibrinogen without the addition of agonist, as well as the differences in spreading and GPIIb/IIIa distribution on platelets adherent to high- versus low-density immobilized fibrinogen.     

Biochemical characterization and molecular cloning of a novel endothelial-specific sialomucin

We have generated rat monoclonal antibodies (MoAbs) against cell surface antigens of the mouse endothelioma cell line bEND.3. Three antibodies (V.1A7, V.5C7, and V.7C7) were selected, all of which recognize a 75-kD antigen on bEND.3 cells and bind selectively to endothelial cells in cryostat sections of mouse tissues. A cDNA for the antigen was isolated from a bEND.3 pCDM8 expression library by using transient expression in COS-7 cells and immunoselection with the three MoAbs. This cDNA coded for a novel, type I membrane protein of 248 amino acids with an extracellular domain rich in threonine and serine residues (35%). The protein is sensitive to O-sialoglycoprotein endopeptidase, indicating that it belongs to the class of sialomucin-like proteins. Therefore, we suggest the name endomucin. Treatment of isolated endomucin by sialidase and O-glycosidase reduced the apparent molecular weight to 45 kD and abolished binding of all three antibodies, indicating that carbohydrates are directly or indirectly involved in the formation of the antibody epitopes. Immunohistological analysis of all examined mouse tissues showed that endomucin is an endothelial antigen found in venous endothelium as well as in capillaries, but not on arterial endothelium. Interestingly, high endothelial venules of peripheral and mesenteric lymph nodes as well as of Peyers's patches were negative for staining with the three MoAbs.     

Purification, characterization, and expression of CFTR nucleotide-binding domains

The nucleotide binding domains (NBDs) within CFTR were initially predicted to lie in the cell cytoplasm, and to gate anion permeability through a pore that was present in membrane spanning alpha helices of the overall polypeptide. Our studies designed to characterize CFTR suggest several important features of the isolated nucleotide binding domain. NBD-1 appears to bind nucleotides with similar affinity to the full-length CFTR protein. In solution, the domain contains a high beta sheet content and self-associates into ordered polymers with molecular mass greater than 300,000 Daltons. The domain is very lipophilic, disrupts liposomes, and readily enters the planar lipid bilayer. Clinically important mutations in the domain may disrupt the nucleotide binding capabilities of the protein, either through a direct effect on the nucleotide binding site, or through effects that influence the overall folding of the domain in vitro. Finally, after expression in human epithelial cells (including epithelial cells from a CF patient), the first nucleotide binding domain targets the plasma membrane even in the absence of other constituents of full-length CFTR and mediates anion permeability in these cells.     

Analysis of MT1-MMP and MMP2 expression in human gastric cancers

The vascular endothelium is intimately involved in a wide variety of normal physiological processes, including coagulation/anticoagulation, the maintenance of vascular tone, and pathological processes, including reperfusion injury, inflammatory syndromes, and tumor cell metastasis. This review discusses the importance of increased adhesive molecule expression on the endothelial surface in promoting circulating inflammatory cell-endothelial adherence in inflammatory conditions, as well as the role of the vascular endothelium in reperfusion injury, altered microvascular permeability states, and atherogenesis.     

Anti-idiotypic monoclonal antibodies to GM1 identify ganglioside binding proteins

Ganglioside stimulated neurite outgrowth may be due to ganglioside binding to membrane proteins or to intercalation into the membrane. To test that ganglioside binding proteins could be found on neuronal surfaces, anti-idiotypic ganglioside monoclonal antibodies (AIG mAbs) were generated to mimic the biological properties of the GM1 ganglioside. The AIG mAbs were identified by their ability to bind to a known GM1 binding protein, the beta-subunit of cholera toxin. For the two AIG mAbs studied, AIG5 and AIG20, binding to beta-CT was blocked most strongly by GM1. This data also suggests that AIG5 and AIG20 mimic different but overlapping epitopes of the ganglioside GM1. Western blotting and immunoprecipitation of mammalian tissues reveals four potential ganglioside binding proteins of molecular weight 93, 66, 57, and 45 kDa. Immunocytochemistry demonstrates neuronal surface label with the AIG mAbs, which suggests that gangliosides, enriched on the neuronal surface membrane, are co-localized with putative ganglioside binding proteins. In bioassays, the AIG mAbs promote neuronal sprouting. This shows that these antibodies can be used to study the biological effects of ganglioside binding to neuronal surface proteins, and the role of gangliosides in the activation of neurite outgrowth.     

Endoglin (CD 105) expression in human lymphoid organs and placenta

Endoglin (CD 105) is a cell surface antigen widely expressed on vascular endothelium, syncytiotrophoblast, some tissue macrophages, certain culture cells (including early leukemic B-lineage) and some endothelial cell lines. Though its relation to the transforming growth factor-beta (TGF-beta) receptor system is well documented, its function and detailed pattern of expression still remain to be clarified. We examined the differential tissue distribution of endoglin in human lymphoid organs and placenta with several anti-CD 105 monoclonal antibodies (mAbs) using an indirect immunoperoxidase technique, and performed semi-quantitative measurements using an image-analyzing system for comparison. Arterial, venous and capillary endothelia in these organs were reactive with anti-CD 105 mAbs at varying intensities. Interestingly, a distinctly stronger staining pattern was observed in the high endothelial venules (HEVs) which may indicate a special role for endoglin in lymphocyte trafficking. Syncytiotrophoblast expressed endoglin strongly on their apical cell membrane. Extravillous trophoblasts at certain locations selectively expressed endoglin on their cell membranes, suggesting a special role for this surface antigen during trophoblast differentiation.