Monoclonal antibodies as probes of conformational changes in protein-engineered cytochrome c

Determination of the nature of the antigen-antibody complex has always been the ultimate goal of three-dimensional epitope mapping studies. Various strategies for epitope mapping have been employed which include comparative binding studies with peptide fragments of antigens, binding studies with evolutionarily related proteins, chemical modifications of epitopes, and protection of epitopes from chemical modification or proteolysis by antibody shielding. In this study we report the use of protein engineering to modify residues in horse cytochrome c that are in or near the epitopes of four monoclonal antibodies specific for this protein. The results demonstrate not only that site-specific changes in the antigen binding site dramatically affect antibody binding, but, more importantly, that some of the site-specific changes cause local and long-range perturbations in structure that are detected by monoclonal antibody binding at other surfaces of the antigen. These findings emphasize the role of native conformation in the stabilization of the interaction between protein antigens and high affinity monoclonal antibodies. Furthermore, the results demonstrate that monoclonal antibodies are more sensitive probes of changes in conformation brought about by protein engineering than low resolution spectroscopic methods such as circular dichroism, where similar spectra are observed for all the analogues. These findings suggest a role for monoclonal antibodies in detecting conformational changes invoked by nonconservative amino acid substitutions or substitutions of evolutionarily conserved residues in protein-engineered or recombinant proteins.     

Monoclonal antibodies recognizing the non-tandem repeat regions of the human mucin MUC4 in pancreatic cancer

The MUC4 mucin is a high molecular weight, membrane-bound, and highly glycosylated protein. It is a multi-domain protein that is putatively cleaved into a large mucin-like subunit (MUC4α) and a C-terminal growth-factor like subunit (MUC4β). MUC4 plays critical roles in physiological and pathological conditions and is aberrantly overexpressed in several cancers, including those of the pancreas, cervix, breast and lung. It is also a potential biomarker for the diagnosis, prognosis and progression of several malignancies. Further, MUC4 plays diverse functional roles in cancer initiation and progression as evident from its involvement in oncogenic transformation, proliferation, inhibition of apoptosis, motility and invasion, and resistance to chemotherapy in human cancer cells. We have previously generated a monoclonal antibody 8G7, which is directed against the TR region of MUC4, and has been extensively used to study the expression of MUC4 in several malignancies. Here, we describe the generation of anti-MUC4 antibodies directed against the non-TR regions of MUC4. Recombinant glutathione-S-transferase (GST)-fused MUC4α fragments, both upstream (MUC4α-N-Ter) and downstream (MUC4α-C-Ter) of the TR domain, were used as immunogens to immunize BALB/c mice. Following cell fusion, hybridomas were screened using the aforementioned recombinant proteins ad lysates from human pancreatic cell lines. Three anti MUC4α-N-Ter and one anti-MUC4α-C-Ter antibodies were characterized by several inmmunoassays including enzyme-linked immunosorbent assay (ELISA), immunoblotting, immunofluorescene, flow cytometry and immunoprecipitation using MUC4 expressing human pancreatic cancer cell lines. The antibodies also reacted with the MUC4 in human pancreatic tumor sections in immunohistochemical analysis. The new domain-specific anti-MUC4 antibodies will serve as important reagents to study the structure-function relationship of MUC4 domains and for the development of MUC4-based diagnostics and therapeutics.     

Monoclonal antibodies detect and stabilize conformational states of smooth muscle myosin

Antibodies with epitopes near the heavy meromyosin/light meromyosin junction distinguish the folded from the extended conformational states of smooth muscle myosin. Antibody 10S.1 has 100-fold higher avidity for folded than for extended myosin, while antibody S2.2 binds preferentially to the extended state. The properties of these antibodies provide direct evidence that the conformation of the rod is different in the folded than the extended monomeric state, and suggest that this perturbation may extend into the subfragment 2 region of the rod. Two antihead antibodies with epitopes on the heavy chain map at or near the head/rod junction. Magnesium greatly enhances the binding of these antibodies to myosin, showing that the conformation of the heavy chain in the neck region changes upon divalent cation binding to the regulatory light chain. Myosin assembly is also altered by antibody binding. Antibodies that bind to the central region of the rod block disassembly of filaments upon MgATP addition. Antibodies with epitopes near the COOH terminus of the rod, in contrast, promote filament depolymerization, suggesting that this region of the tail is important for assembly. The monoclonal antibodies described here are therefore useful both for detecting and altering conformational states of smooth muscle myosin.     

Monoclonal antibodies to the angiotensin-converting enzyme from the human lung

The hybridoma producing monoclonal antibody (IgG1) to human angiotensin-converting enzyme (ACE) has been prepared by fusion of murine myeloma P3O1 with spleen cells of BALB/c mice immunised with a purified human lung ACE preparation. A high specificity of monoclonal antibody (MAb) binding to immobilized ACE has been demonstrated by ELISA; that of soluble ACE--by immunoadsorption test. The latter technique permits the use of impure ACE preparations for the screening procedure. This MAb did not effect ACE activity. This antibody is believed useful not only for immunoassay and immunopurification of ACE, but also as a tool for investigation of enzyme distribution in tissue as well as for studying the structure and mechanism of ACE action.     

Monoclonal antibodies as probes of the alpha-bungarotoxin and cholinergic binding regions of the acetylcholine receptor

We have probed the acetylcholine receptor (AcChR) molecule with six anti-AcChR monoclonal antibodies (mAbs) whose binding to the AcChR is inhibited or blocked by alpha-bungarotoxin (alpha BgTx). mAbs bound with a maximum stoichiometry of either one mAb (387D, 247G) or two mAbs (383C, 572C, 370C, 249E) per AcChR monomer, and the extent to which they inhibited alpha BgTx binding directly correlated with their stoichiometry of binding. The effect of mAbs on the alpha BgTx and cholinergic ligand binding properties of the AcChR molecule defined three major categories of mAbs: those that block alpha BgTx and carbamylcholine (agonist) binding, but do not block d-tubocurarine (antagonist) binding (383C, 572C, 370C and 249E); mAb 387D, which blocks agonist binding and partially blocks alpha BgTx and d-tubocurarine binding; and mAb 247G, which does not affect agonist binding, blocks at most 50% of the alpha BgTx binding sites, and decreases the affinity of the high affinity component of d-tubocurarine binding (Mihovilovic, M., and Richman, D. P. (1984) J. Biol. Chem. 259, 15051-15059). Except for mAb 247G, these mAbs strongly competed with each other for binding to the AcChR. In contrast, mAb 247G blocks about 50% of the binding of all the other mAbs. The results demonstrate the ability of mAbs to stabilize different conformational states of the AcChR and to probe cholinergic epitopes of functional importance. They also indicate the nonequivalence of the two alpha-toxin binding regions of the AcChR molecule and suggest that it is possible to identify epitopes within the alpha BgTx binding region that when bound produce differential effects on the binding of the agonist (carbamylcholine) and the antagonist (d-tubocurarine).     

Sulfhydryls of tubulin. A probe to detect conformational changes of tubulin.

The 20 cysteine residues of tubulin are heterogeneously distributed throughout its three-dimensional structure. In the present work, we have used the reactivity of these cysteine residues with 5, 5'-dithiobis(2-nitrobenzoic acid) (DTNB) as a probe to detect the global conformational changes of tubulin under different experimental conditions. The 20 sulfhydryl groups can be classified into two categories: fast and slow reacting. Colchicine binding causes a dramatic decrease in the reactivity of the cysteine residues and causes complete protection of 1.4 cysteine residues. Similarly, other colchicine analogs that bind reversibly initially decrease the rate of reaction; but unlike colchicine they do not cause complete protection of any sulfhydryl groups. Interestingly, in all cases we find that all the slow reacting sulfhydryl groups are affected to the same extent, that is, have a single rate constant. Glycerol has a major inhibitory effect on all these slow reacting sulfhydryls, suggesting that the reaction of slow reacting cysteines takes place from an open state at equilibrium with the native. Ageing of tubulin at 37 degrees C leads to loss of self-assembly and colchicine binding activity. Using DTNB kinetics, we have shown that ageing leads to complete protection of some of the sulfhydryl groups and increased reaction rate for other slow reacting sulfhydryl groups. Ageing at 37 degrees C also causes aggregation of tubulin as indicated by HPLC analysis. The protection of some sulfhydryl groups may be a consequence of aggregation, whereas the increased rate of reaction of other slow reacting sulfhydryls may be a result of changes in global dynamics. CD spectra and acrylamide quenching support such a notion. Binding of 8-anilino-1-naphthalenesulfonate (ANS) and bis-ANS by tubulin cause complete protection of some cysteine residues as indicated by the DTNB reaction, but has little effect on the other slow reacting cysteines, suggesting local effects.     

Topography and conformational changes of fructose-1,6-bisphosphate aldolase.

Structure of aldolase, its interaction with nucleotides, the path of enzyme reaction and the scheme of range of conformational changes of this enzyme are presented. Retrospectives and perspectives of aldolase topography investigations are included.     

A simple approach to detect active-site-directed enzyme-enzyme interactions. The aldolase/glycerol-phosphate-dehydrogenase enzyme system

A novel approach has been elaborated to identify the mechanism of intermediate transfer in interacting enzyme systems. The aldolase/glycerol-3-phosphate-dehydrogenase enzyme system was investigated since complex formation between these two enzymes had been demonstrated. The kinetics of dihydroxyacetone phosphate conversion catalyzed by the dehydrogenase in the absence and presence of aldolase was analyzed. It was found that the second-order rate constant (kcat/Km) of the enzymatic reaction decreases due to the formation of a heterologous complex. The decrease could be attributed to an increase of the Km value since kcat did not change in the presence of aldolase. In contrast, an apparent increase in the second-order rate constant of dihydroxyacetone phosphate conversion by the dehydrogenase was observed if the triose phosphate was produced by aldolase from fructose 1,6-bisphosphate (consecutive reaction). Moreover, no effect of dihydroxyacetone phosphate on the dissociation constant of the heterologous enzyme complex could be detected by physico-chemical methods. The results suggest that the endogenous dihydroxyacetone phosphate produced by aldolase complexed with dehydrogenase is more accessible for the dehydrogenase than the exogenous one, the binding of which is impeded due to steric hindrance by bound aldolase.     

Monoclonal antibodies to type X collagen. Biosynthetic studies using an antibody to the amino-terminal domain

Monoclonal antibodies to chick type X collagen have been used to study the structure, biosynthesis, and location of type X in cartilage. The antibodies were produced by injecting purified type X collagen into female SJL/J mice and then fusing their spleen cells with Sp2/0 myeloma cells. Hybridoma culture supernatants were screened for antibodies to type X collagen by enzyme-linked immunosorbent assay and Western blots. Positive supernatants did not cross-react with other collagen types (I, II, IX, XI) or with fibronectin. Three monoclonal antibodies were chosen for further characterization. Two of them (1A6 and 6F6) recognize a pepsin-sensitive domain of type X collagen. Rotary shadowing showed that 1A6 and 6F6 both recognize the same end of type X, probably the aminoterminal non-triple helical domain. Amino acid sequencing of the intact protein and of the epitope-containing peptide confirmed that the antibody recognition sites for 1A6 and 6F6 are within the amino-terminal domain. Monoclonal antibody 2B3 reacts with the pepsinized (45 kDa) and weakly with the nonpepsinized (59 kDa) forms of type X collagen. The monoclonal antibodies were used for immunolocalization of type X in hypertrophic chondrocytes and reacted only with tissue samples from areas undergoing endochondral ossification, e.g. growth plate and fracture callus. Antibody 6F6, when coupled to Sepharose, selectively binds to type X collagen from cell and organ cultures. In a pulse-chase experiment, no processing of the 59-kDa form of type X could be detected. Two components with molecular masses of approximately 70 and 85 kDa, arising from a disulfide-bonded aggregate, were synthesized by both the permanent and calcifying cartilage organ cultures but did not react with the antibody, suggesting that these proteins are not related to type X. In summary, the pulse-chase results and the immune precipitation with monoclonal antibody 6F6 did not detect biosynthetic precursors larger than 59 kDa or proteolytically processed forms of type X.     

Monoclonal antibodies recognizing epitopes carried on both glycolipids and glycoproteins of the human milk fat globule membrane.

The molecules of the human milk fat globule membrane (MFGM) which bind four murine monoclonal antibodies (LICR LON M3, M8, M18 and M24) raised against the human MFGM have been identified. By using 'Western' blotting [Burnette (1981) Anal. Biochem. 112, 195-203] it was shown that each antibody reacted with a different set of proteins. M3 and M24 were similar in their pattern of reaction with the membrane proteins, but were quite distinct from M8 and M18, which also differed from each other. Glycopeptides prepared from the MFGM by exhaustive Pronase digestion were able to inhibit partially the binding of M3 and M24, and prevent totally the binding of M8 and M18, to the MFGM in an enzyme-linked immunoabsorbent assay. Oligosaccharides obtained by the deproteination of human milk also completely inhibited the binding of M3, M18 and M24 to the MFGM. However, the binding of M8 was not inhibited by these saccharides, and therefore M8 may not be recognizing a simple carbohydrate determinant. By using an enzyme-linked assay, M8 and M18 were shown not to bind to MFGM glycolipid, whereas M3 and M24 did, and this was confirmed by overlaying thin layer chromatograms of MFGM lipids with these antibodies. Both M3 and M24 showed a similar complex pattern of reaction, binding to more than one glycolipid moiety. By these means all four antibodies have been shown to react with antigens which involve carbohydrate side chains carried on different proteins, and two were also shown to react with such determinants on glycolipids.     

Monoclonal antibodies: use to detect developmentally regulated antigens on D. discoideum amebae

We have used monoclonal antibodies to detect developmentally regulated cell surface antigens on D. discoideum amebae. A study of an antigen detected using an antibody produced by a hybridoma line implicates a previously undescribed component in the process of cell aggregation. This antigen (consisting of a doublet of 69,000 and 73,000 molecular weight) is first detected during the early hours of cell starvation and is present until cells begin slug formation. The developmental appearance of the antigen is not controlled by cAMP pulses and is distinct from that of Contact A sites. Fab fragments directed against the antigen are potent inhibitors of aggregation but do not inhibit the differentiation of cells to aggregation competence.     

Monoclonal antibodies to alpha-chain regions of human fibrinogen that participate in polymer formation

Monoclonal antibodies have been generated against a cross-link-containing derivative of alpha polymer (alpha XLCNBr), isolated following CNBr digestion of fibrin [Sobel, J. H., Ehrlich, P. H., Birken, S., Saffran, A. J., & Canfield, R. E. (1983) Biochemistry (preceding paper in this issue)]. One cloned cell line (F-102) was chosen for characterization based on its apparent specificity for the A alpha-chain region A alpha 518-584 (CNBr X). A second line (F-103) was selected because of its anti-A alpha 241-476 (CNBr VIII) properties. These two regions of the A alpha chain have previously been implicated as major contributors to the cross-linking process that leads to alpha-polymer formation. Radioimmunoassays have been developed, employing the immunoglobulins produced by clones F-102 and F-103. These assays have been applied, in conjunction with high-performance liquid chromatography purified tryptic and chymotryptic derivatives of CNBr VIII and CNBr X, to localize the respective determinants involved in antibody binding. In each case, virtually full immunoreactivity was exhibited by both the CNBr fragment and a single tryptic or chymotryptic peptide originating from it. These findings indicate that sequence-specific, rather than conformational, determinants were operative in the generation of antibodies F-102 and F-103. The epitope recognized by F-102 was localized to the region of A alpha 540-554, while the F-103 binding site resided within A alpha 259-276. When these radioimmunoassays were applied to study the relative immunoreactivity exhibited by a variety of fibrinogen derivatives, the results obtained support earlier suggestions that the COOH-terminal portion of the A alpha chain contains regions of random conformation.     

Monoclonal antibodies recognizing protease-generated neoepitopes from cartilage proteoglycan degradation. Application to studies of human link protein cleavage by stromelysin.

Monoclonal antibodies were raised that specifically recognize the NH2-terminal neoepitope sequence present in link protein cleavage products derived from stromelysin-degraded proteoglycan aggregate. Competitive enzyme-linked immunosorbent assay, using synthetic peptides as inhibitors, showed that one of these antibodies (CH-3) required, for antibody recognition, the free NH2-terminal amino acid isoleucine (residue 17 of the intact protein) in the sequence NH2-IQAENG at the stromelysin cleavage site of link protein 3. Human proteoglycan aggregate was digested with recombinant human stromelysin, bovine chymotrypsin, bovine trypsin, and porcine elastase, and their respective link protein degradation products were tested for immunoreactivity with antibody CH-3. Only stromelysin- and chymotrypsin-generated link protein 3 were recognized by antibody CH-3. Both of these enzymes generate link protein NH2 termini with the sequence 17IQAENG. . .; hence these studies indicated that monoclonal antibody CH-3 recognized this neoepitope sequence in only specific proteolytically modified link protein molecules. Since the occurrence of link protein 3 increases with aging, the incidence of CH-3 epitope in proteoglycans isolated from human knee articular cartilage of individuals of different ages was investigated. The prevalence of CH-3 epitope was found to be highest in newborn and adolescent articular cartilage samples. However, little CH-3 epitope was detected in older adult cartilage, although considerably more link protein 3 was present in these samples. These results suggest that additional proteolytic agents are responsible for the increased occurrence of link protein degradation products with aging.     

Monoclonal antibodies that distinguish between subspecies of human interferon-alpha and that detect interferon oligomers

Monoclonal antibodies to human interferons (HuIFN) of the alpha-class have been prepared by screening against 125I-labeled IFN in a rapid liquid-phase radioimmunoassay. All of the six antibodies produced react with HuIFN-alpha 2 and with some components of HuIFN-alpha N (Namalwa); three of the antibodies also bind HuIFN-alpha 1, and these either do not bind or bind very weakly the 25K component of Namalwa. Reaction of the antibodies with IFN components blotted onto nitrocellulose after separation on reducing gels suggests that two of the antibodies are against conformational determinants, whereas the epitopes recognized by the other antibodies are not destroyed by reduction or SDS treatment; these antibodies can be used to detect the presence of oligomers in IFN preparations. From the reaction of the antibodies with different alpha-IFN in immunoblots, in an antiviral assay, and in an ELISA, it was concluded that at least five different epitopes are recognized by the six antibodies, only one of which is non-neutralizing.     

Antipeptide antibodies against conserved regions of human interferon-alpha: evidence for conformational variations between IFN-alpha subtypes.

Antibodies to two conserved regions (residues 29-36 and 139-151) of human interferon-alpha were raised by immunizing rabbits with four short synthetic peptides coupled to carriers. The antibodies were tested for reactivity with recombinant interferon-alpha by ELISA. Despite the amino acid conservation of the two regions, there are significant variations in the reactivity of the antibodies with the IFN-alpha subtypes. The reactivity is enhanced significantly when the disulfide bonds of the interferon molecule are reduced. The results indicate that there are subtype-specific differences in the presentation of the epitopes in these conserved regions of human interferon-alpha.     

Monoclonal antibodies as probes of the antigenic structure of tobacco mosaic virus.

The antigenic structure of tobacco mosaic virus has been analysed by measuring the ability of nine monoclonal antibodies to distinguish between wild-type virus and 13 mutants showing single and double amino acid substitutions in the coat protein. Although the majority of antibodies detected those substitutions that were located at the outer surface of the virion, some of them also recognized conformation alterations induced by exchanges occurring deep inside the subunit. In the case of five mutants, the antibody reactivity was reduced compared with wild-type virus, while in the case of three others, it was significantly higher. Each monoclonal antibody possessed a unique discrimination pattern with respect to the different substitutions. The simultaneous presence of two exchanges led to the complete disappearance of any binding with six of the nine antibodies and to reduced binding with three others. The superior discriminatory capacity of monoclonal antibodies compared with polyclonal antisera was demonstrated by the fact that three exchanges not detected with antisera were found to alter the antigenicity when tested with monoclonal antibodies.     

Monoclonal antibodies to bovine UDP-galactosyltransferase. Characterization, cross-reactivity, and utilization as structural probes

A series of mouse monoclonal antibodies has been developed against a soluble form of bovine UDP-galactose:N-acetylglucosamine galactosyltransferase purified to apparent chemical homogeneity by a combination of affinity and immunoadsorption chromatography. The purified enzyme consists of two molecular mass variants of 42 and 48 kDa. Individual monoclonal antibodies were selected for by their ability to recognize immobilized affinity-purified galactosyltransferase and were not reactive against bovine alpha-lactalbumin and bovine immunoglobulins. Based on competitive binding assays and Western blot analysis with either galactosyltransferase or lactose synthetase (covalently cross-linked alpha-lactalbumin galactosyltransferase), these monoclonal antibodies can be subdivided into four groups. Group A (3 clones) recognize an epitope at or near the alpha-lactalbumin binding site. In addition, this group is cross-reactive with soluble galactosyltransferase from human milk and pleural effusion. Group B (6 clones) and D (1 clone) appear to recognize two different epitopes on the 6-kDa fragment which is released when the 48-kDa galactosyltransferase polypeptide is converted to the 42-kDa form, apparently by proteolysis. Groups A and C (1 clone) recognize epitopes found on both the 48- and 42-kDa polypeptide. Interestingly, immunofluorescence studies indicate that only two monoclonal antibody groups (C and D) are able to decorate membrane-bound galactosyltransferase (Golgi-associated) in formalin-fixed, methanol-, or detergent-permeabilized cells. Thus, these groups of monoclonal antibodies appear to identify four separate structural/functional domains on soluble galactosyltransferase, two of which are not readily accessible for binding in situ.     

Enzyme immunoassay using monoclonal antibodies to study conformational changes of human serum albumin

A method for studying conformational changes induced in the human albumin molecule, either in its purified form or in serum, is described. Plates were coated with albumin or human serum at varying pHs and were reacted with peroxidase-labeled anti-albumin monoclonal antibodies of different specificities. The data showed that albumin molecules were coated in conformations induced as a result of pH changes, allowing us to demonstrate that pH modifications involved the N-terminal portion of the albumin molecule whether in its purified form or in serum. This method should be applicable to the study of conformational modifications in other proteins as well.     

Monoclonal antibodies to human brain acetylcholinesterase: properties and applications

1. Acetylcholinesterase (AChE) was purified 20,000-fold in a 43% yield from 90 g of human cerebellum by combined immunoaffinity and ligand affinity chromatography. The purified enzyme migrated as a 68,000-dalton band during polyacrylamide gel electrophoresis under denaturing and reducing conditions. 2. Balb/c mice were immunized with multiple 10-micrograms injections of this material in order to raise monoclonal antibodies to human brain AChE. Three such antibodies were obtained and characterized. 3. Each antibody cross-reacted distinctively with AChEs from other mammals. No antibody recognized human plasma butyrylcholinesterase but all reacted with AChE from human red blood cells. 4. Antibodies HR5 and HR3 performed well in two-site immunoassays for AChE. With these assays we compared autopsy samples of cortical region A9 from six controls (nonneurological cases) and five patients with Alzheimer's disease. The latter showed a highly significant 60% deficit of AChE protein. 5. The present antibodies will permit additional immunochemical studies of cholinergic systems in dementia.