Antibodies induced by liposomal protein exhibit dual binding to protein and lipid epitopes

Natural polyreactive antibodies can accommodate chemically unrelated epitopes, such as lipids and proteins, in a single antigen binding site. Because liposomes containing lipid A as an adjuvant can induce antibodies directed against specific lipids, we immunized mice with liposomes containing lipid A together with a protein or peptide antigen to determine whether monoclonal antibodies generated after immunization would be specifically directed both to the liposomal lipid (either cholesterol or galactosylceramide) and also to the accompanying liposomal protein or peptide. Monoclonal antibodies were obtained that bound, by ELISA, to cholesterol and to recombinant gp140 envelope protein from HIV-1, or to galactosylceramide and to an HIV-1 envelope peptide. Surface plasmon resonance studies with the former antibody showed that the liposomal cholesterol and liposomal gp140 each contributed to the overall binding energy of the antibody to liposomes containing cholesterol and protein.     

Surface engineering: optimization of antigen presentation in self-assembled monolayers.

The formation of self-assembled monolayers (SAMs) on gold surfaces containing an antigenic peptide (NANP)6 and HS(CH2)11OH, and the specific binding of a monoclonal antibody to these layers were investigated by surface plasmon resonance (SPR). Peptides were synthesized by solid-state phase synthesis and were linked either to cysteine or to an alkyl-thiol to allow covalent attachment to gold. The content of the peptide in the SAMs was systematically varied, and the binding properties of the monoclonal antibody were compared with those measured by microcalorimetry in solution. At a critical peptide concentration in the SAM an optimal antibody binding and complete surface coverage was attained. At lower peptide concentrations, the amount of adsorbed antibody decreased; at higher peptide concentrations, the binding constant decreased. These effects can be explained if the accessibility of the antigenic epitopes depends on the peptide density. Addition of free antigen induced the desorption of bound antibodies and allowed accurate measurements of the dissociation rate constant. Binding constants obtained from steady-state measurements and from measurements of the kinetic rate constants were compared.     

Solid-phase peptide quantitation assay using labeled monoclonal antibody and glutaraldehyde fixation

A solid-phase radioimmunoassay utilizing iodinated peptide-specific monoclonal antibody as a detection system instead of labeled peptide has been developed. Regional specific monoclonal antibodies to either gastrin-releasing peptide or gastrin were used as models to validate the general application of our modified assay. Conditions for radioactive labeling of the monoclonal antibody were determined to minimize oxidant damage, which compromises the sensitivity of other reported peptide quantitation assays. Pretreatment of 96-well polyvinyl chloride test plates with a 5% glutaraldehyde solution resulted in consistent retention of sufficient target peptide on the solid-phase matrix to allow precise quantitation. This quantitative method is completed within 1 h of peptide solid phasing. Pretreatment of assay plates with glutaraldehyde increased binding of target peptide and maximized antibody binding by optimizing antigen presentation. The hypothesis that glutaraldehyde affects both peptide binding to the plate and orientation of the peptide was confirmed by analysis of several peptide analogs. These studies indicate that peptide binding was mediated through a free amino group leaving the carboxy-terminal portion of the target peptide accessible for antibody binding. It was observed that the length of the peptide also affects the amount of monoclonal antibody that will bind. Under the optimal conditions, results from quantitation of gastrin-releasing peptide in relevant samples agree well with those from previously reported techniques. Thus, we report here a modified microplate assay which may be generally applied for the rapid and sensitive quantitation of peptide hormones.     

Kinetic and conformational properties of a novel T-cell antigen receptor transmembrane peptide in model membranes

Core peptide (CP; GLRILLLKV) is a 9-amino acid peptide derived from the transmembrane sequence of the T-cell antigen receptor (TCR) alpha-subunit. CP inhibits T-cell activation both in vitro and in vivo by disruption of the TCR at the membrane level. To elucidate CP interactions with lipids, surface plasmon resonance (SPR) and circular dichroism (CD) were used to examine CP binding and secondary structure in the presence of either the anionic dimyristoyl-L-alpha-phosphatidyl-DL-glycerol (DMPG), or the zwitterionic dimyristoyl-L-alpha-phoshatidyl choline (DMPC).Using lipid monolayers and bilayers, SPR experiments demonstrated that irreversible peptide-lipid binding required the hydrophobic interior provided by a membrane bilayer. The importance of electrostatic interactions between CP and phospholipids was highlighted on lipid monolayers as CP bound reversibly to anionic DMPG monolayers, with no detectable binding observed on neutral DMPC monolayers.CD revealed a dose-dependent conformational change of CP from a dominantly random coil structure to that of beta-structure as the concentration of lipid increased relative to CP. This occurred only in the presence of the anionic DMPG at a lipid : peptide molar ratio of 1.6:1 as no conformational change was observed when the zwitterionic DMPC was tested up to a lipid : peptide ratio of 8.4 : 1.     

Mitogen-like monoclonal anti-actin antibodies

Monoclonal antibodies (IgM kappa) have been produced to actin isolated electrophoretically from L cell extracts. These monoclonal anti-actin antibodies bind to intact L cells and modulate DNA synthesis and cell proliferation, much like affinity-purified polyclonal rabbit antibody to the same Mr 42,000 actin. In addition, monoclonal antibodies specific for actin from Entamoeba histolytica also bound to and modulated the growth of L cells. A monoclonal antibody directed against a neuroblastoma surface antigen did not produce stimulation of L cells, and the binding activity of anti-actin monoclonal antibody to L cells was removed by absorption with actin covalently coupled to Sepharose. These observations demonstrate the specificity of interaction between the anti-actin monoclonal antibodies and the surface of intact L cells. We conclude that a surface actin-like molecule on the L cell, when bound by specific monoclonal antibody, initiates a stimulatory signal which results in enhanced cellular metabolism.     

The use of monoclonal antibodies to distinguish several chemically modified forms of human alpha 1-proteinase inhibitor.

The purpose of our investigation was to obtain monoclonal antibodies that could distinguish three forms of alpha 1-proteinase inhibitor (alpha 1-PI): native alpha 1-PI, N-chlorosuccinimide-oxidized alpha 1-PI (Ox-alpha 1-PI) and proteolytically modified alpha 1-PI (alpha 1-PI). Three specific monoclonal antibodies were characterized as to their binding properties. By using the Bio-Dot assay, it was found that all three forms of alpha 1-PI were capable of binding to antibody 6D4-6-18, that only Ox-alpha 1-PI, but not native alpha 1-PI or alpha 1-PI, could bind to antibody 6C7-5, and that alpha 1-PI and a complex between alpha 1-PI and trypsin uniquely were not able to bind to antibody 5C12-8-7. Thus it was concluded that it is possible to use monoclonal antibodies with different epitopic specificities to distinguish two chemically modified forms of alpha 1-PI from the native protein.     

Surface plasmon resonance immunosensor for the detection of Salmonella typhimurium

An immunosensor based on surface plasmon resonance (SPR) using protein G was developed for the detection of Salmonella typhimurium. A protein G layer was fabricated by binding chemically to self-assembly monolayer (SAM) of 11-mercaptoundecanoic acid (MUA) on gold (Au) surface. The formation of protein G layer on Au surface modified with 11-MUA and the binding of antibody and antigen in series were confirmed by SPR spectroscopy. The effect of detergent such as Tween-20 on binding efficiency of antibody and antigen was investigated by SPR. The binding efficiency of antigen to the antibody immobilized on Au surface was improved up to about 85% and 100% by using protein G and Tween-20, respectively. The surface morphology analyses of 11-MUA monolayer on Au substrate, protein G layer on 11-MUA monolayer and antibody layer immobilized on protein G layer were performed by atomic force microscope (AFM). Consequently, an immunosensor based on SPR for the detection of S. typhimurium using protein G was developed with a detection range of 10(2) to 10(9)CFU/ml. The current fabrication technique of a SPR immunosensor for the detection of S. typhimurium could be applied to construct other immnosensors or protein chips.     

Analysis of a common-antigen lipopolysaccharide from Pseudomonas aeruginosa.

Lipopolysaccharide isolated from Pseudomonas aeruginosa PAO1 (O5 serotype) was separated into two antigenically distinct fractions. A minor fraction, containing shorter polysaccharide chains, reacted with a monoclonal antibody to a P. aeruginosa common antigen but did not react with antibodies specific to O5-serotype lipopolysaccharide. In contrast, fractions containing long polysaccharide chains reacted only with the O5-specific monoclonal antibodies. The shorter, common-antigen fraction lacked phosphate and contained stoichiometric amounts of sulfate, and the fatty acid composition of this fraction was similar to that of the O-antigen-specific fraction. The lipid A derived from the serotype-specific lipopolysaccharide cross-reacted with monoclonal antibodies against lipid A from Escherichia coli, while the lipid A derived from the common antigen did not react. We propose that many serotypes of P. aeruginosa produce two chemically and antigenically distinct lipopolysaccharide molecules, one of which is a common antigen with a short polysaccharide and a unique core-lipid A structure.     

Characterization of antibody covalently coupled to liposomes

We have explored the covalent coupling of fatty acids to immunoglobulin G(IgG). N-hydroxysuccinimide ester of palmitic acid (NHSP) was used to couple palmitic acid to either a mouse monoclonal antibody to the major histocompatibility antigen, H-2^k, or goat antibody to the major glycoprotein of the Molony Leukemia Virus, gp-70. The reaction was characterized in terms of the time course, input ratio of NHSP to IgG, stoichiometry of the coupling, distribution of palmitic acid in the IgG subunits, and the antigen binding capacity of the coupled antibody. Incorporation of the fatty acid modified IgG into liposomal membranes using a detergent-dialysis method was studied as a function of extent of fatty acid coupling. Finally, the binding of IgG-coated liposomes with cells expressing proper antigens was characterized. The major conclusions were: (1) the optimal molar ratio of NHSP to IgG in the reaction was between 10 and 20, which yields about 4–5 palmitoyl chains per IgG molecule; (2) at this level of coupling, the antigen binding capacity of the IgG antibody decreased about 3–4-fold; (3) incorporation of the coupled antibody into unilamellar liposomes (about 1000 Å diameter) can be achieved with a deoxycholate-dialysis method with an optimal lipid-to-protein ratio of 10:1 (w/w); (4) there were about 48 IgG molecules incorporated per liposome under these conditions; (5) the apparent dissociation constant of the liposome-bound antibody under the optimal condition was about 6–7-fold higher than that of the native antibody; (6) binding of antibody to the target cells was accompanied by binding of liposomal lipids; both bindings could be blocked by pretreatment of cell with unmodified antibody.     

Characterization of the binding epitope of the monoclonal antibody DMAb-1 to ganglioside GM2.

Several derivatives of ganglioside GM2 were synthesized for mapping of the binding epitope of a monoclonal antibody raised against this ganglioside. The GM2 ganglioside was modified in both the hydrophobic and the hydrophobilic part of the molecule. The synthesized derivatives were characterized with fast atom bombardment mass spectrometry (FAB-MS). Affinity of the monoclonal antibody for the GM2 derivatives was determined by enzyme-linked immunosorbent assay (ELISA) on microtitre plates or by TLC immunostaining. Modifying the GM2 sialic acid by deacetylation or blocking of the carboxyl moiety abolished the binding to the monoclonal antibody while the cleaving of the glycol group on the sialic acid tail led to a 70% reduced binding affinity. Removal of the fatty acid (lyso-GM2) eliminated the binding to the antibody. GM2 derivatives with fatty acid moieties of 8 carbon atoms or less showed almost no reactivity. GM2 with saturated fatty acids 16:0, 18:0 and 20:0 had binding affinity similar to natural GM2, while the 24:0 fatty acid had only half the binding affinity. The results demonstrate the importance of ganglioside fatty acid composition with regard to ligand binding between the monoclonal antibody and its specific ganglioside antigen. Thus, caution must be shown in the application of immunaffinity methods with monoclonal antibodies for the quantitative determination of glycosphingolipids from different tissues.     

Binding of tissue-type plasminogen activator with human endothelial cell monolayers. Characterization of the high affinity interaction with plasminogen activator inhibitor-1

The formation and release of covalent complexes between tissue-type plasminogen activator (t-PA) and plasminogen activator inhibitor-1 (PAI-1) limits the application of equilibrium radioligand binding analysis to characterize the interaction between t-PA and human umbilical vein endothelial cell (HUVEC) monolayers. To avoid this difficulty, we used a recombinant mutant of t-PA, S478A rt-PA, in which alanine has been substituted for the active-site serine. Although the mutant is incapable of covalently reacting with PAI-1, 125I-labeled S478A rt-PA binding to HUVEC monolayers is specific and reversible and is characterized by a high affinity (Kd of 1.5 nM) and a large number of sites (1.5 x 10(6)/cell). This binding was shown to occur through noncovalent interaction with PAI-1 in the HUVEC monolayer by the fact that a monoclonal anti-PAI-1 antibody (MA-7D4) completely blocked S478A rt-PA binding. Two solution-phase assays with recombinant PAI-1 (rPAI-1) confirmed this noncovalent interaction: complexes between 125I-S478A rt-PA and rPAI-1 could be isolated by immunoprecipitation with anti-PAI-1 antibodies, and S478A rt-PA competed with rt-PA for inactivation by rPAI-1. In contrast diisopropylphosphate rt-PA (in which the active site serine is chemically modified) showed minimal binding to HUVEC monolayers, as a result of impaired interaction with PAI-1, in the two assays. Thus, both wild-type rt-PA and S478A rt-PA interact with the HUVEC monolayer through PAI-1. With rt-PA this results in the formation of covalent rt-PA.PAI-1 complexes that are released from the monolayer into the supernatant. With S478A rt-PA this results in the formation of noncovalent complexes that remain associated with the HUVEC monolayer, thereby identifying a large pool of reactive PAI-1 molecules in the monolayer.     

Affinity binding of cells to cryogel adsorbents with immobilized specific ligands: effect of ligand coupling and matrix architecture

The capture of human acute myeloid leukemia KG-1 cells expressing the CD34 surface antigen and the fractionation of human blood lymphocytes were evaluated on polyvinyl alcohol (PVA)-cryogel beads and dimethyl acrylamide (DMAAm) monolithic cryogel with immobilized protein A. The affinity ligand (protein A) was chemically coupled to the reactive PVA-cryogel beads and epoxy-derivatized monolithic cryogels through different immobilization techniques and the binding efficiency of the cell surface receptors specific antibody-labeled cells to the gels/beads was determined. The binding of cells to monolithic cryogel was higher (90-95%) compared with cryogel beads (76%). B-lymphocytes, which bound to the protein A-cryogel beads, were separated from T-lymphocytes with yields for the two cell types 74 and 85%, respectively. About 91% of the bound B-cells could be recovered without significantly impairing their viability. Our results show differences in the percentage of cell-binding to the immunosorbents caused by ligand density, flow shear forces and bond strength between the cells and the affinity surface once distinct chemical coupling of protein A, size of beads, sequence of antibody binding to protein A adsorbents, morphology and geometry of surface matrices were compared.     

Conformational dynamics of individual antibodies using computational docking and AFM

Molecular recognition between a receptor and a ligand requires a certain level of flexibility in macromolecules. In this study, we aimed at analyzing the conformational variability of receptors portrayed by monoclonal antibodies that have been individually imaged using atomic force microscopy (AFM). Individual antibodies were chemically coupled to activated mica surface, and they have been imaged using AFM in ambient conditions. The resulting topographical surface of antibodies was used to assemble the three subunits constituting antibodies: two antigen‐binding fragments and one crystallizable fragment using a surface‐constrained computational docking approach. Reconstructed structures based on 10 individual topographical surfaces of antibodies are presented for which separation and relative orientation of the subunits were measured. When compared with three X‐ray structures of antibodies present in the protein data bank database, results indicate that several arrangements of the reconstructed subunits are comparable with those of known structures. Nevertheless, no reconstructed structure superimposes adequately to any particular X‐ray structure consequence of the antibody flexibility. We conclude that high‐resolution AFM imaging with appropriate computational reconstruction tools is adapted to study the conformational dynamics of large individual macromolecules deposited on mica. Copyright © 2013 John Wiley & Sons, Ltd.     

Determination of kinetic parameters of epitope-paratope interaction based on solid phase binding: An inexpensive alternate to biospecific interaction analysis

A method has been developed for biospecific interaction analysis between antigen and antibody using solid phase binding approach. Real time kinetics between monoclonal antibody and human chorionic gonadotropin have been studied. Kinetic constants of the bimolecular reaction are determined. Affinity constants measured by several independent methods have been found to be relatively consistent. Convenient and simple procedures to determine affinity constant, K_onand K_off of monoclonal antibody-human chorionic gonadotropin interaction using binding of [¹²⁵I]hCG to immobilized monoclonal antibody are presented. Values obtained compare well with those obtained using surface plasmon resonance technology, making this method a viable alternative.     

Production and characterization of four monoclonal antibodies against porcine pancreatic colipase

Four monoclonal antibodies directed against porcine colipase have been generated by hybridization of myeloma cells with spleen cells of BALB/c immunized mice. Antibodies were screened by binding to immobilized colipase in a solid-phase assay. Monoclonal antibodies were purified by affinity chromatography on colipase coupled to Sepharose. All monoclonal antibodies are of the IgG1 class with high affinity for the antigen. The dissociation constant of the complex formed in solution between porcine colipase and antibody varied from 1.1 X 10(-10) M to 1.8 X 10(-8) M. Epitope specificity was studied for each antibody and in pairs with an enzyme-linked immunosorbent assay (ELISA). Results indicate that the four monoclonal antibodies react with at least three different antigenic regions of colipase. Finally, three monoclonal antibodies were found to be potent inhibitors of colipase activity. Antiporcine monoclonal antibodies appear to be suitable probes for studying the lipid affinity site of the protein cofactor of pancreatic lipase.     

Epitope masking of rat esophageal carcinoma tumor-associated antigen by certain coexisting glycolipid and phospholipid molecules: a potential mechanism for tumor cell escape from the host immune responses

A monoclonal antibody (mAb-5G) produced against a tumorigenic rat esophageal cell line, B2T, was shown to react specifically with a unique glycolipid antigen expressed on the cell surface of tumorigenic and certain non-tumorigenic, immortalized rat esophageal cell lines [Cancer Immunol Immunother 36: 94 (1993)]. In enzyme-linked immunosorbent assay experiments, mAb-5G reacted with crude lipid extracts prepared from B2T cells cultured in vitro, but showed very little reactivity with crude lipid extracts prepared from the same cell line passaged once in vivo, unless the antigen was separated from other lipid components by column or thin-layer chromatography (TLC). When a secondary tissue-culture cell line was established from the above B2T tumor tissues and serially subcultured in vitro, the percentage of positively stained cells was increased significantly in immunofluorescence assay. It was also demonstrated that the amount of extractable antigen was increased as the cells were subcultured in vitro up to passage 15, and stabilized thereafter. These results indicate the presence of certain lipid components in crude lipid extracts from B2T cells grown in vivo that are capable of interfering with antigen-antibody binding. On TLC plates, these interfering lipids were identified as phosphatidylcholine, phosphatidylserine, sphingomyelin and gangliosides. The interfering lipids did not bind the antibody, rather they appeared to interfere with antigen accessibility. These lipid substances may modify tumor cell surface antigen(s), thus protecting the tumor cells from host immune destruction.     

Docosahexaenoic acid-containing phosphatidylcholine affects the binding of monoclonal antibodies to purified Kb reconstituted into liposomes

Class I major histocompatibility complex (MHC I) molecules are transmembrane proteins that bind and present peptides to T-cell antigen receptors. The role of membrane lipids in controlling MHC I structure and function is not understood, although membrane lipid composition influences cell surface expression of MHC I. We reconstituted liposomes with purified MHC I (Kb) and probed the effect of lipid composition on MHC I structure (monoclonal anti-MHC I antibody binding). Four phospholipids were compared; each had a phosphocholine head group, stearic acid in the sn-1 position, and either oleic, alpha-linolenic, arachidonic, or docosahexaenoic acid (DHA) in the sn-2 position. The greatest binding of monoclonal antibody AF6-88.5, which detects a conformationally sensitive epitope in the extracellular region of the MHC I alpha-chain, was achieved with DHA-containing proteoliposomes. Other epitopes (CTKb, 5041.16.1) showed some sensitivity to lipid composition. The addition of beta2-microglobulin, which associates non-covalently with the alpha-chain and prevents alpha-chain aggregation, did not equalize antibody binding to proteoliposomes of different lipid composition, suggesting that free alpha-chain aggregation was not responsible for disparate antibody binding. Thus, DHA-containing membrane lipids may facilitate conformational change in the extracellular domains of the alpha-chain, thereby modulating MHC I function through effects on that protein's structure.     

Quenching of fluorescein-conjugated lipids by antibodies. Quantitative recognition and binding of lipid-bound haptens in biomembrane models, formation of two-dimensional protein domains and molecular dynamics simulations.

Three model biomembrane systems, monolayers, micelles, and vesicles, have been used to study the influence of chemical and physical variables of hapten presentation at membrane interfaces on antibody binding. Hapten recognition and binding were monitored for the anti-fluorescein monoclonal antibody 4-4-20 generated against the hapten, fluorescein, in these membrane models as a function of fluorescein-conjugated lipid architecture. Specific recognition and binding in this system are conveniently monitored by quenching of fluorescein emission upon penetration of fluorescein into the antibody's active site. Lipid structure was shown to play a large role in affecting antibody quenching. Interestingly, the observed degrees of quenching were nearly independent of the lipid membrane model studied, but directly correlated with the chemical structure of the lipids. In all cases, the antibody recognized and quenched most efficiently a lipid based on dioctadecylamine where fluorescein is attached to the headgroup via a long, flexible hydrophilic spacer. Dipalmitoyl phosphatidylethanolamine containing a fluorescein headgroup demonstrated only partial binding/quenching. Egg phosphatidylethanolamine with a fluorescein headgroup showed no susceptibility to antibody recognition, binding, or quenching. Formation of two-dimensional protein domains upon antibody binding to the fluorescein-lipids in monolayers is also presented. Chemical and physical requirements for these antibody-hapten complexes at membrane surfaces have been discussed in terms of molecular dynamics simulations based on recent crystallographic models for this antibody-hapten complex (Herron et al., 1989. Proteins Struct. Funct. Genet. 5:271-280).     

Chemical modification of surfactant protein A alters high affinity binding to rat alveolar type II cells and regulation of phospholipid secretion

Alveolar type II cells express a high affinity receptor for pulmonary surfactant protein A (SP-A), and the interaction of SP-A with these cells leads to inhibition of surfactant lipid secretion. We have investigated the binding of native and modified forms of SP-A to isolated rat alveolar type II cells. Native and deglycosylated forms of SP-A readily competed with 125I-SP-A for cell surface binding. Alkylation of SP-A with excess iodoacetamide yielded forms of SP-A that did not inhibit surfactant lipid secretion and did not compete with 125I-SP-A for cell surface binding. Reductive methylation of SP-A with H2CO and NaCNBH3 yielded forms of SP-A with markedly reduced receptor binding activity that also exhibited significantly reduced capacity to inhibit lipid secretion. Modification of SP-A with cyclohexanedione reversibly altered cell surface binding and the activity of SP-A as an inhibitor of lipid secretion. Two monoclonal antibodies that block the function of SP-A as an inhibitor of lipid secretion completely prevented the high affinity binding of SP-A to type II cells. A monoclonal antibody that recognizes epitopes on SP-A but failed to block the inhibition of secretion also failed to completely attenuate high affinity binding to the receptor. Concanavalin A inhibits phospholipid secretion of type II cells by a mechanism that is reversed in the presence of excess alpha-methylmannoside. Concanavalin A did not block the high affinity binding of 125I-SP-A to the receptor. Neither the high affinity binding nor the inhibitor activity of SP-A was prevented by the presence of mannose or alpha-methylmannoside. The SP-A derived from humans with alveolar proteinosis is a potent inhibitor of surfactant lipid secretion but failed to completely displace 125I-SP-A binding from type II cells. From these data we conclude that: 1) cell surface binding activity of rat SP-A is directly related to its capacity to inhibit surfactant lipid secretion; 2) monoclonal antibodies directed against SP-A can be used to map binding domains for the receptor; 3) the lectin activity of SP-A against mannose ligands does not appear to be essential for cell surface binding; 4) concanavalin A does not compete with SP-A for receptor binding; and 5) the human SP-A derived from individuals with alveolar proteinosis exhibits different binding characteristics from rat SP-A.     

Tomoregulin internalization confers selective cytotoxicity of immunotoxins on prostate cancer cells

We have recently identified and validated the prostate cancer antigen Tomoregulin as a target for the radioimmunotherapy for prostate cancer. Here, we provide evidence that Tomoregulin is an internalizing antigen and a potential target for immunotoxins. First, the cell surface localization of Tomoregulin was confirmed by flow cytometry, and its expression levels were determined by whole-cell binding assays. Second, laser scanning confocal microscopy revealed Tomoregulin internalization into the cytoplasm on antibody binding at 37 degrees C. The internalized Tomoregulin was found to colocalize with acidic vesicles. Third, internalization kinetics assays using (125)I-labeled anti-Tomoregulin mouse monoclonal antibody 2H8 demonstrated that the amount of internalized antigen-antibody complexes increased with time and reached approximately 25% of the total surface antigen after 60 to 90 minutes. Because 2H8 is capable of binding to Tomoregulin on the cell surface and can be internalized, we finally evaluated 2H8 as a means of targeting toxic payloads to prostate cancer cells. 2H8 was coupled to the cytotoxin saporin through a secondary antibody (Mab-ZAP) in indirect immunotoxin assays. Cell killing occurred on Tomoregulin-positive cells (Clone69) at the immunotoxin concentrations not affecting the Tomoregulin-negative cells (PC-3). In contrast to 2H8, the control antibody (mouse anti-c-Myc antibody 9E10) had no effect on cells in the presence of Mab-ZAP. Thus, Tomoregulin internalization confers selective cytotoxicity of immunotoxins on prostate cancer cells, and Tomoregulin-mediated delivery of immunotoxin has potential as a prostate cancer therapy.