Search

The three-dimensional arrangement of ammo acids in a protein dictates its biological properties One of the urgent challenges of modern biology is to decipher the relationship between the linear sequence of ammo acids and the conformation adopted by proteins Success is predicting protein conformation from the primary structure can be anticipated to permit rational design of therapeutic formulations to treat intractable human diseases and of probes for research in fundamental biological problems [ABSTRACT FROM AUTHOR]

Radioimmunotherapeutic and radioimmunodiagnostic procedures have been the subjects of nearly 200 clinical trials since 1978. Applications range from diagnosis of tumor deposits to systemic radiation therapy. Originally, murine IgG was the most common antibody form used clinically. Later the antibody fragments, Fab and F(ab')2, were extensively investigated as the alternative radioisotope carriers displaying more favorable pharmacokinetics. Future clinical applications will include products of molecular engineering, such as single-chain antigen-binding proteins, "humanized" (also known as "chimeric") antibodies, human antibodies, and bifunctional antibodies. Selection of antibodies for clinical trials is an extensive and detailed process. The factors to be considered include characterization of the antibody with regard to its interactions with the antigen(s) of interest, nonspecific binding to non-target tissues, and in vivo pharmacokinetics. In addition, the biological consequences of chemical modifications made during production of the radiolabeled material must be assessed. Ideally, the choice of the radioisotope and radiolabeling methods is individually tailored for each antibody to yield products that display minimally impaired biological properties for any given application. Only seldom can such a detailed approach be afforded. Thus, prior experiences with antibodies and radioisotopes described in the literature are useful in designing the experimental strategies. For each of the commonly used radioisotopes, there are several well-tested methods. A brief overview of available methodologies and specific procedures for radiolabeling with radiometals is given in the following sections. [ABSTRACT FROM AUTHOR]

Polyclonal antibodies often serve as the starting point for interesting studies of new antibody functions and their links with immunoregulation and autoimmune disease. For example, the catalytic activity of naturally formed antibodies was first discovered using human autoantibody mixtures (1,2). Thyroglobulin (Tg), a large water-soluble glycoprotein (mass approx 660 kDa) is a classical target for autoimmune responses. Thyroglobulin is stored in the thyroid colloid and can constitute up to 70% of the total protein in this gland. The protein is iodinated within the thyroid by peroxidase-catalyzed reactions and proteolytic processing of the iodinated Tg leads to formation of the thyroid hormones T3 and T4. Autoantibodies to Tg are found in >80% of patients with Hashimoto's thyroiditis, a disease characterized by thyroid damage and hyperthyroidism. Here, we describe a method to measure the catalytic breakdown of Tg by autoantibody fractions. Since Tg contains several repeat domains and antigenic epitopes (3,4), the hydrolytic specificity of the antibodies was determined using a panel of commercially available peptide-methylcoumarinamide (peptide-MCA) substrates. Cleavage of the amide bond linking an amino acid and the coumarin moiety in these substrates serves as a convenient surrogate for peptide bond hydrolysis. The peptide-MCA substrates have previously been used to assay the activity of endopeptidases and exopeptidases (5). [ABSTRACT FROM AUTHOR]

High-specific-activity radiolabeled polypeptide probes permit sensitive detection of antigen-antibody interactions. In the case of small polypeptides labeled with 125I, various iodinated forms can be separated by reversed-phase HPLC and resolved peaks characterized by protein chemistry methods. We describe here methods for preparation and purification of vasoactive intestinal (VIP) labeled at Tyr10 with 125I, and the use of (Tyr10_125I)VIP to characterize antibody binding and enzymatic activities. [ABSTRACT FROM AUTHOR]

A subpopulation of antibodies with high affinity for the substrate ground state is also capable of chemical catalysis (1). However, high turnover by catalysts is dependent on efficient binding to the transition state, not the ground state. Direct screening for catalysis may permit isolation of more efficient antibody catalysts, since this approach obviates the requirement for strong binding to substrate ground states (or presumed transition state analogs). To this end, we have developed methods to purify recombinant antibody fragments rapidly, followed by measurement of their catalytic activity using peptide-methylcoumarinamide conjugates or radiolabeled peptides as substrates. [ABSTRACT FROM AUTHOR]

Immobilized metal affinity chromatography (IMAC), introduced in 1975 (1), relies on the formation of coordinate bonds between metal ions immobilized on a suitable support and electron donor groups in proteins. A polyhistidine tag (five or six His residues) placed at either the C- or N-terminus of a recombinant protein can form a stable chelate with immobilized transition metals. This allows fractionation of the target protein to 90-95% purity levels in a single chromatographic step (2-4). Metals like Cu2+ and Ni2+ bind surface-accessible His residues selectively. The high affinity of polyhistidine tags for commercially available metal resins (Kd 10−13M or more) allows the use of stringent conditions to remove loosely bound proteins, while retaining the recombinant protein bound to the immobilized metal. [ABSTRACT FROM AUTHOR]

Asthma is associated with increased catalytic autoantibodies to vaso-active intestinal peptide (VIP), a bronchodilator and anti-inflammatory neuropeptide found in nerve endings supplying airway smooth muscle and epithelial structures (1). Asthmatic airways obtained at autopsy are deficient in VIP (2), raising the possibility that efficient catalytic hydrolysis of VIP by autoantibodies synthesized locally in the airways or derived by transudation from systemic circulation may be a pathogenetic factor in asthma. With the aim of investigating the possible cause-effect relationship between autoantibodies to VIP and asthma, we have cloned catalytic antibody light (L) chains from the immune repertoire of an exercise-induced asthma patient. We chose to work with the L chains for the following reasons. First, L chains retain the ability to recognize VIP and appear to turn over more rapidly than intact antibodies. Faster catalysis observed with L chains may arise from diminished affinity for the substrate ground state and a more flexible active site than the intact antibody-combining site. Second, Fv species from combinatorial libraries of heavy (H)- and L-chain-variable regions are not likely to reflect the composition of antibodies found in vivo, in that the probability of pairing of the natural H- and L-chain partners in these libraries is low (3). Third, free L chains are found in circulation in vivo (4). Glutathione levels are increased in the asthmatic airway (5). A reducing environment may lead to increased levels of L chains by cleavage of disulfide links. We anticipate that availability of recombinant human L chains that catalyze the hydrolysis of VIP will permit direct examination of their disease-causing potential, for example, by examination of airway abnormalities in transgenic mice overexpressing the L chains. [ABSTRACT FROM AUTHOR]

This chapter describes efficient systems for construction, expression, and screening of comprehensive libraries of murine or human antibody Fab fragments displayed monovalently on the surface of filamentous (fd) phage. Both systems use phagemid vectors that co-express combinations of randomly assembled pairs of light (L) and heavy (H) coding regions under transcriptional control of the lacZ promoter. The combinatorial Fab gene cassette is inserted in-frame with a truncated version of the phage surface protein pII1 (ΔpIII), which results in the positioning of the assembled Fab molecule at one tip of the phage particle. [ABSTRACT FROM AUTHOR]

The display of antibody fragments on the surface of filamentous bacteriophages (1-7) constitutes a powerful system for the selection of molecules with desired specificities. In phage display, the antibody fragment is coupled to the minor coat protein (protein3) of bacteriophage M13 phage and is, in this way, both anchored in the phage capsid and exposed on the phage surface, linking specificity and genetic information. This permits direct isolation and sequence determination of the gene encoding the antibody fragment. The gene can subsequently be exposed to further engineering and selection in order to improve affinity and specificity. [ABSTRACT FROM AUTHOR]

The application of phage display to the screening of antibody V-gene libraries promises to replace animal immunization and hybridoma technology as a means of antibody generation (1,2). Synthetic libraries offer the potential for expanding the immune system repertoire and designing libraries for particular sets of antigens. We have developed a method for the generation of synthetic antibody gene libraries by which spiked oligonucleotides are assembled by a ligase chain reaction protocol (3). This is a synthetic adaptation of a reaction that was previously used as a diagnostic tool (4,5). [ABSTRACT FROM AUTHOR]

Introduction of mutations in antibody combining sites provides a powerful means to assess the functional role of individual residues and segments of the combining sites in interactions with antigen. Once cloned antibody fragments are available, several methods can be utilized to introduce mutations. Commonly used methods include those of Kunkel et al. (1) involving the use of a uracil-labeled template, and Kramer and Fritz (2) in which gapped duplex DNA is used to construct oligonucleotide-directed mutants. Several polymerase chain reaction-based mutagenesis techniques have been described (3-5). However, available thermostable polymerases are error-prone, leading to undesired sequence changes. For example, Taq polymerase has an average error rate of 0.8% (6,7). In this chapter, we described an efficient method for mutagenesis of cloned antibody fragments in which mutant oligonucleotides are utilized to initiate the synthesis of a phosphorothioate nucleotide containing DNA strands using single-stranded phage or phagemid DNA as template. The original strand is removed by exonuclease digestion, the nonmutant strand is nicked with a restriction enzyme, gapped with exonuclease III, and mutant double-stranded DNA is generated using polymerase and ligase enzymes (8-10). The method has been applied to generate point mutations and deletion mutations in an anti-VIP light (L) chain in the example described here. Insertion mutations can also be performed using appropriate oligonucleotides. In addition, oligonucleotides synthesized using mixtures of nucleotide phosphoramidites in individual cycles can be used to produce libraries of random mutants. A major advantage of this approach is the high mutation efficiency. pCANTAB5his6, a phagemid vector derived from pUC119, was used in the present example. In principle, any M13 vector can be employed. An easy-to-use kit with excellent instructions is supplied by Amersham (Sculptor). In combination with appropriate strategies to select mutant clones (see Tramontano, Chapter 16, and Huang et al., Chapter 27), new constructs can be prepared rapidly and clones with the desired activity isolated. [ABSTRACT FROM AUTHOR]

In 1988 the first report on the expression of a fully functional recombinant Fab in E. coli (1) was one of the initiating events in the development of the field of antibody engineering. Shortly after came the report of the cloning and expression of antibody-variable domains as single-chain Fv or sFv (2). This provided a method for rapid cloning of stable antibody domains from hybridoma cells into E. coli in a fully functional form. More recent developments in the field, such as expression and selection of libraries of sFv or Fab on phage particles, have allowed the de novo generation of antibodies with binding properties equivalent to or, in some cases, better than those derived by traditional hybridoma methods (3). [ABSTRACT FROM AUTHOR]

The variable regions of antibody subunits are the "business ends" of these molecules responsible for interactions with antigens and activation of the biological functions of the constant regions. Rearrangement of variable-, diversity-, and joining-region genes and the hypermutability of the complementarity determining regions (CDRs) permit development of new antigen-binding specificities. With the discovery of natural catalytic activity in autoantibodies (1-3) and the confirmation of this phenomenon (4,5), it has become clear that hypervariability in antibody genes may permit a natural evolution of enzyme-like sites in antibodies. Additional evidence for this hypothesis consists of observations that monoclonal antibodies (MAb) to a natural antigen, vasoactive intestinal polypeptide (VIP), display proteolytic activity (6,7). As the first step toward establishing the physicochemical basis of the catalytic activity, we have cloned and expressed the light (L)-chain subunit and single-chain FV (scFV) fragment of one such antibody. The recombinant L chain and the scFV display interactions with VIP typical of an antibody-combining site. Both molecules are efficient catalysts (8). The distinctive feature of catalytic hydrolysis of VIP by these recombinant proteins is high-affinity substrate binding (low Km), a property not shared by conventional enzymes. The L chain also catalyzes the hydrolysis of a protease substrate unrelated in sequence to VIP, but this activity appears to be the result of low affinity recognition of basic residues (8). By primary structure analyses, molecular modeling, and kinetic examination of the catalytic activities, it appears that the residues in the antibody active site providing important contributions in the binding and hydrolysis steps may be different. Site-directed mutagenesis of the recombinant protein active sites is ongoing to validate this hypothesis. [ABSTRACT FROM AUTHOR]

Antibodies to native B-DNA are common in systemic lupus erythematosus (SLE), and their measurement serves as a diagnostic and prognostic marker of the disease (1). Autoreactive antibodies are also found in the sera of normal individuals (2). Natural autoantibodies crossreact with a variety of conserved, and seemingly unrelated, autoantigens. Most of these antibodies are of low affinity, although rare exceptions of high affinity natural anti-DNA antibodies have been reported (3). The role of cells that make natural autoantibodies and whether they are progenitors of cells that make disease-associated autoantibodies remains unclear (4). [ABSTRACT FROM AUTHOR]

The ability to make monoclonal rodent antibodies has revolutionized immunology. To reduce the immunogenicity of these antibodies for human in vivo use, methods have been developed to create artificial recombinant antibodies (1). Chimeric recombinant antibodies containing mouse variable domains and human constant domains are characterized by essentially the same antigen-binding properties as the mouse antibodies from hybridomas. The immunogenicity of the chimeric antibodies is lower owing to replacement of mouse constant domains for human ones. Various humanized, reshaped, and other antibody constructs have been created in different laboratories (1). [ABSTRACT FROM AUTHOR]

Monoclonal antibodies (MAb) with prespecified enzyme-like catalytic activity are obtained from the immune response against transition-state analogs (TSA). The premise for generating antibody catalysts, as put forth by Jencks (1), was validated through early examples of specific hydrolytic catalysts for ester and carbonate substrates (2, 3). The initial success with TSA of acyl transfer processes as haptens led to derivation of antibodies as catalysts for a variety of reactions (4). This approach has been expanded to include haptenic molecules not directly related in structure to a deduced feature of the reaction's transition state, but that could deploy groups in the antibody that participate in a catalytic mechanism not otherwise available for the reaction in ordinary immunological recognition (5). The unifying concept in these approaches is the insight to the reaction mechanism that allows the construction of a suitable mimic of the transition-state or other high-energy structures. Individual approaches explore the scope of the antibody's enzymatic qualities and the validity of the design strategy for generating the catalyst. The success of these endeavors typically requires several screening steps for the identification of a catalyst among a set of MAb obtainable from hybridoma cloning experiments. [ABSTRACT FROM AUTHOR]

Catalysis of various chemical transformations by antibodies opens broad perspectives in the fundamental and applied branches of life science (1). Antibodies are ideally suited for therapeutic applications owing to their capability of homing to a particular target. In addition, they happily withstand stringent conditions and, therefore, are easy to isolate and manipulate. These advantages have provoked considerable interest in the idea that new catalytic antibodies could substitute for enzymes in therapy. [ABSTRACT FROM AUTHOR]

In 1974, Niels Jerne (1) proposed a theory that pictured the immune system as a network of idiotypic-anti-idiotypic interactions regulating the immune response to an antigen. The theoretical basis for the construction of idiotypic structures mimicking external antigen results from the fact that the conformation of some of the anti-idiotypic antibodies may be complementary to the three-dimensional structure of the binding site of the first antibody, and therefore, represents the internal image of the original epitope. The experimental validation of this concept has been provided by numerous studies where anti-idiotypic antibodies were not only found to mimic the structure of antigens, but were also able to induce a functional activity that mimics the physiological activity of the original antigen. As an alternative to the transition-state approach for eliciting catalytic antibodies, our approach consists of exploiting the internal image properties of anti-idiotypic antibodies to produce catalytic antibodies. In the first step, an antibody is raised that recognizes the active site of an enzyme. The binding site of this first antibody (Ab1) has structural features complementary to those of the enzyme. This monoclonal Ab1 is selected based on its ability to inhibit the enzymatic reaction. In the second step, Ab1 is used as antigen to produce anti-idiotypic antibodies. Among these anti-idiotypic antibodies (Ab2), some may represent internal images of the enzyme active site. Owing to the enormous variability in the antibody population, it is possible that among the anti-idiotypic antibodies, a few not only possess the binding function for the substrate, but also catalyze its transformation. If this is the case, the combining site of such "enzyme-like" antibodies may possess not only the general structural features common with the enzyme active site, but also the amino acid residues essential for catalytic activity. [ABSTRACT FROM AUTHOR]

Despite major progress in various fields of chemical and molecular biology, generation of a synthetic molecule that possesses the catalytic qualities of enzymes has not been achieved. In order to overcome this obstacle, numerous attempts have been carried out to synthesize antibodies that express catalytic activity. [ABSTRACT FROM AUTHOR]

The existence of opioid receptors as "narcotic analgesic binding components" in mammalian brain was first documented in the early 1970s (1-3). Shortly thereafter, endogenous opioid peptides (endorphins) that also bound to these receptors were found in mammalian brain tissue. Subsequent pharmacological work revealed the existence of three main subclasses of receptors designated µ, δ, and κ. This work was confirmed by the recent cloning of these three subclasses (4-8) that bear sequence homologies to each other and to the G-protein-coupled receptor super-family. Before the successes in cloning the subclasses, several investigators seeking to answer questions about the molecular basis of the different opioid-receptor classes prepared anti-idiotypic antibodies to the opioid receptors. This approach was taken because unsuccessful experiments showed that antiopioid-receptor antibodies could not be prepared by immunization with cells expressing the receptors. The general experimental approach has been to use subclass-selective ligands as immunogens to raise antihapten antibodies (Id+, AB1), which in turn are used as immunogens to prepare anti-idiotypic antibodies (AB2). Anti-idiotypic antibodies have been described in theory as containing an "internal image" of the ligand at their binding site (9, 10). This implies that portions of the binding region of the anti-idiotypic antibody can demonstrate binding properties and, in some cases, pharmacological properties similar to the original ligand. In accordance with this view, the resulting anti-idiotypic anti opioidreceptor antibodies have been shown to mimic their respective agonists both kinetically and pharmacologically (11-13). Thus, antiopioid receptor antibodies have proven to be important tools for studying the functionality and localization of receptors within their native environment (12,14). A significant advantage to using AB2s as probes for membrane-bound receptors is that they can be fluorescently labeled at sites distant from their receptor-binding site, whereas this cannot be done with the much smaller opioid ligands. Furthermore, fluorescently labeled AB2s can be used to monitor binding to individual cells within a large total population of cells, whereas more commonly used radioreceptor-binding assays can only address the average binding properties of an entire cell population (15). [ABSTRACT FROM AUTHOR]

Serum from normal individuals contains a repertoire of antibodies even in the absence of apparent immunization. The B-cell compartment of the immune system can recognize not only the antigens of the external environment, but also internal or self-antigens. Normal serum contains natural autoantibodies of the IgM and IgG classes that recognize a wide range of self-antigens, including nuclear antigens, intracellular and membrane components, and circulating plasma proteins (1,2). Some of the antigens recognized by natural autoantibodies are also the targets of autoantibodies in autoimmune diseases, e.g., thyroglobulin (Tg), neutrophil cytoplasmic antigens, glomerular basement membrane antigens, intrinsic factor, and factor VIII. Despite the presence of autoreactive B- and T-cells in the normal available immune repertoire, autoimmune disease remains a relatively rare event. Autoimmune diseases cannot be diagnosed by the mere finding of increased titers of autoantibodies in serum. The question arises whether the structure, specificity, affinity, idiotypy, and genetics of autoantibodies in healthy subjects differ from those of pathological autoimmunity (3,4). Determining the fine specificity of autoantibodies may lead to a means of distinguishing natural vs disease-associated autoantibodies. [ABSTRACT FROM AUTHOR]

Chapter 9 developed the concept of antibody clonality and described several methods that have been used to define antibody clonality in polyclonal sera. Affinity immunoblotting (1) has proven to be a powerful and high-resolution method for estimating clonality of immune responses to a variety of antigens. This chapter describes the affinity immunoblotting method in detail. [ABSTRACT FROM AUTHOR]

The seminal development shaping our current understanding of the immune system was the formulation of the concept of clonal selection by Sir Macfarlane Burnet (1). This idea revolutionized the approach to formulation of experimental design in immunology. At the heart of clonal selection was the idea that antibody specificity preexisted before antigen arrival and that the antigen triggered the system by selecting only cells preprogrammed to make antibody molecules with significant affinity for that particular antigen. The former instructional theories of antibody formation clearly were rendered obsolete by this concept. Many scholars attribute the subsequent formal proof of clonal selection to Raff et al. (2). [ABSTRACT FROM AUTHOR]

This chapter was written with the express goal of introducing several basic concepts involved when antigens and antibodies interact in vitro to the novice reader and to serve as a review of these concepts to the more advanced reader. Most of these concepts also apply to in vivo reactivity. Antigen-antibody reactions in vivo frequently involve subsequent reactions of the complexes with complement, but these reactions will be largely ignored for the purpose of this chapter. Reactivity properties of antibodies are determined by several important structural features of the antigens and antibody preparations used. Major properties of antigens that influence their reactivities with antibodies include structural features such as whether individual epitopes are repeated on the molecule, the relative proportion of sequential and conformational epitopes, and the degree of purity of the antigen preparation. For purposes of this chapter, it will be assumed that the antigens used as examples are pure, but the knowledgeable reader will understand that this is almost never the case. [ABSTRACT FROM AUTHOR]

Major advances in the analysis of biomolecular structure have emerged from the development of highly specific monoclonal antibodies (MAb) by Köhler and Milstein in 1975 (1). MAb are produced from hybridoma cell lines that secrete a single species of antibody to a unique antigen. The advantages of MAb as reagents for antibody engineering are: 1.They are extremely pure immunological reagents that can be produced by immunization with a heterogenous antigenic preparation;2.They are homogenous with respect to affinity and specificity;3.The specificity of the antibody is directed to a single epitope;4.The desired MAb can be selected for activity, affinity, and subclass by applying appropriate screening methods; and5.Virtually unlimited quantities of MAb can be produced. [ABSTRACT FROM AUTHOR]

Antigen-binding sites in antibodies are formed by the variable regions of light (L) and heavy (H) chains. Delineation of antigen interactions with the individual subunits of antibodies is of interest for several reasons, including: 1.Free L chains are present within B-lymphocytes and in the circulation of patients with certain types of B-lymphocyte tumors (1,2);2.Derivation of high-affinity antibodies in vitro from randomly combined libraries of L and H chains is dependent on pairing of subunits that display appropriate interactions with each other and the antigen (3); and3.Pairing of L and H chains from different antibodies may permit the development of new functions in the hybrid. For example, pairing of a catalytic L chain with H chains with a defined antigen-binding specificity could generate a catalyst with specificity dictated by the H chain. [ABSTRACT FROM AUTHOR]

In mammals, the genes encoding the variable (V) domains of the immunoglobulin heavy (H) chain are assembled during lymphocyte development by rearrangements of variable (VH), diversity (DH), and junctional (JH) gene segments (1). Selection of these gene elements from their corresponding libraries of germ-line genes is governed by a site-specific, developmentally ordered process. In humans, the organization of the VH locus has been completely delineated. This cluster comprises 6 functional JH genes, more than 30 DH segments, and approx 100 nonallelic VH genes that have been categorized in at least 7 families. VH gene members within a given family are highly homologous with >80% sequence identity, whereas the degree of homology between members of distinct families is usually <70%. Human VH families vary in size ranging from one member in the VH6 family to over 30 in the VH3 family (2). [ABSTRACT FROM AUTHOR]

A three-dimensional structure of reasonably high resolution is the starting point of any protein engineering project, whether intended to improve the affinity of an antibody, change the specificity of an enzyme, or to test a hypothesis regarding the role of a particular residue in determining structure or stability. The atomic coordinates determined by X-ray crystallography or, increasingly, NMR, provides the means to visualize the shape of a protein as well as the distances between atoms within a protein, or between a protein and ligand or another macromolecule. The free energy changes that determine protein structure, stability, and interaction properties ultimately arise from the distances that separate the various types of atoms involved, including proteins, ligands, and solvent molecules. [ABSTRACT FROM AUTHOR]

Antibodies are multisubunit proteins, but unlike many other multisubunit proteins, it is possible to study structural and functional properties of an antibody subunit free of the others. Thus, antibodies are singular proteins in that they are highly amenable to analysis by a fundamental scientific research strategy, i.e., investigation of complex systems by study of its parts. The metabolism of even a superficially simple bacterial cell cannot be understood without a rigorous understanding of the kinetic properties of each of the enzymes involved. In much the same way, the quaternary interactions of protein subunits cannot be understood except in those relatively rare cases in which the subunit and individual domain components of a complex protein can be examined individually, yielding structural and biophysical information. [ABSTRACT FROM AUTHOR]

Antibodies possess a vast repertoire of specificity and affinity. To understand the molecular basis of antibody function, we require high-resolution X-ray crystallographic structures and good solution structures of free and antigen-bound antibodies. The number of reported antibody structures grows each year. Yet the number of structures deposited with the Brookhaven Protein Database (PDB) (1) remains relatively small, with 43 deposited entries at the time of writing, when compared to the available sequence data. It is therefore important to develop an effective method of predicting the structure of antibody-combining sites. The validity of predicted structures can then be confirmed by mutagenesis in the combining site. The models can also provide valuable structural information to "humanize" antibodies for therapy effectively, to develop immunosensors, and even for the complete de novo design of new antibodies with different functions. [ABSTRACT FROM AUTHOR]

We have been collecting nucleotide and amino acid sequences of proteins of immunological interest, and aligning them in order to understand the structure and function relations of these proteins (1). To aid in organizing and analyzing this collection, a computer program, called SEQHUNT, was written. [ABSTRACT FROM AUTHOR]