Antigenic stimulation of lymphocytes. II. Immunogenicity of dinitrophenyl-polyethylene oxide antigens.

Synthetic antigens consisting of dinitrophenyl groups attached to linear or branched-chain polyethylene oxide stimulate anti-DNP antibody production in rabbits and a proliferative response in vitro in immune rabbit lymphocytes. A requirement for immunogenicity is divalence. Linear DNP₂PEO is most effective at a molecular weight of 10⁴, but a clear response is obtained even at 10³. The optimal valence of DNP_nPEO (n = 4–80, MW, 4 × 10⁵) is 20. Destruction of T cells with antithymocyte serum does not impair the in vitro response of the remaining B lymphocytes to DNP₂PEO, indicating that these antigens are T independent.     

The influence of epitope density on the immunological properties of hapten-protein conjugates. I. Characteristics of the immune response to hapten-coupled albumen with varying epitope density

The relationship between the epitope density of hapten-protein conjugates (DNP· BSA), and their immunogenicity in mice has been investigated. As others have found, lightly substituted protein (DNP₅BSA) elicited primary and secondary antibody responses which were mainly IgG. In contrast, DNP₅₀BSA induced a primary IgM response with relatively little IgG, and little or no immunological memory. The transition in immunogenic behaviour from “low” to “high” occurred with a hapten: protein molar ratio around 30. DNP₅₀BSA does not contain any serologically detectable native BSA determinants or neodeterminants resulting from dinitrophenylation. Although this antigen elicits a mainly IgM response as do thymus-independent antigens, antibody production to both DNP₅BSA and DNP₅₀BSA is highly thymus dependent. The possible reasons for the thymus dependence of immune responsiveness to highepitope-density hapten-protein conjugates are discussed.     

Neonatally induced tolerance to soluble protein antigens. I. Analysis of B-cell and T helper cell function in mice tolerant to bovine serum albumin or fowl gamma-globulin

High dose tolerance to either bovine serum albumin (BSA) or fowl γ-globulin (FGG) was induced in CBA mice by neonatal injection. Tolerance to BSA lasted about 9 weeks, and that to FGG, about 18 weeks. Splenic B-cell function was analyzed using quantitative in vivo assays and in vitro limiting dilution analysis. Tolerogen-specific IgM- and non-IgM-producing B cells are depleted at least threefold in the spleens of tolerant mice. Tolerogen-specific T-helper-cell function was examined by immunization with haptenated antigens. Analysis of the recovery from tolerance indicates that the return to normal function in the tolerogen-specific B-cell and T helper fractions coincides with the return to normal responsiveness by the whole animal.     

The kinetics of aggregation phenomena. I. Minimal models for patch formation of lymphocyte membranes.

Numerous biological processes involve the assembly of one or more monomers into aggregates or networks of interconnected units. In this paper we present the initial aspects of a mathematical theory for network formation on lymphocyte membranes. We assume the fluid mosaic membrane model is valid, that a lymphocyte possesses a homogeneous set of mobile but membrane bound receptors and that these receptors can form bimolecular complexes with antigen. We show that these complexes tend to aggregate and derive expressions for their size distribution as a function of time, antigen valence and concentration, and antigenreceptor affinity.At early times, the mass of the system (receptors plus antigen) is in very small aggregates. However under appropriate conditions, a critical time is reached at which they coalesce in such a way that the mass shifts, becoming concentrated predominantly in large aggregates. We assume that this coalescence (“patch” formation) is a necessary condition for lymphocyte triggering and briefly pursue the consequences.It is shown that the time required for patch formation is a sensitive function of affinity (K), antigen valence and antigen concentration (C), and that if KC is either too high or too low patch formation will not be possible. Moreover within the range of binding constants which can lead to patching, there will be an optimum value which leads to the fastest rate of triggering, and this optimum shifts to higher affinity as the concentration of free antigen surrounding the cell decreases. For optimum KC values we estimate times typically of the order of (10–100) seconds for patch formation. The theory also suggests that if antigen valence is too low, triggering will not be possible within times of interest, without introducing other factors. It thus leads naturally to a requirement for auxiliary cells which would tend to present low valence antigens in such a way that the B lymphocytes see an effective, increased valence. The theory, although primitive, thus meets some minimal requirements in that it distinguishes binding reactions from triggering reactions, makes predictions consistent with observations on affinity maturation and the nonresponsiveness to high doses and low doses of antigen, and suggests the need for helper cells (or their products) in order for low valence antigens to be effective in lymphocyte triggering.     

Role of self carriers in the immune response and tolerance. III. B cell tolerance induced by hapten-modified-self involves both active T cell-mediated suppression and direct blockade.

The induction of B cell unresponsiveness with hapten-modified syngeneic murine lymphoid cells (hapten-modified self, HMS) can be achieved in vivo and in vitro. Tolerance in vivo in mice required a latent period of 3 to 4 days. Moreover, B cell unresponsiveness could not be induced by HMS in athymic nude mice, although their nu/+ littermates were rendered hyporesponsive by HMS. Pretreatment of normal mice with cyclophosphamide (cyclo) prevented their susceptibility to tolerance induction by haptenated lymphoid cells. Nude mice became sensitive to HMS-induced suppression if they were first reconstituted with spleen cells from normal (but not cyclo-treated) donors.Interestingly, labeling of H-2 antigens was not necessary for tolerance induction by HMS since haptenated teratoma cells (lacking H-2) were tolerogenic in normal recipients.In contrast, suppression of the in vitro response to haptenated flagellin occurred equally well with nude, nu/+ and anti-Ly 2 + C-treated spleen cells. These data suggest that cyclo-sensitive modified self-reactive (T) cells may regulate the immune response and mediate tolerance to HMS in vivo. However, the in vitro “blockade” of B cell reactivity may be directly mediated on hapten-specific PFC precursors.     

Desensitization of delayed-type hypersensitivity by antigen and specific antibody.

The role of antibody in the desensitization of delayed-type hypersnsitivity (DTH) to dinitrophenylated bovine gammaglobulin (DNP-BGG) was studied in rats. Rats sensitized by a subcutaneous injection of DNP₃₂-BGG in Freund's complete adjuvant (FCA) were desensitized 14 days later with various doses of DNP₃₂-BGG injected intravenously. It was found that only certain doses (100–500 μg) of DNP-BGG effectively desensitized, antigen doses outside this optimum range being ineffective in suppressing DTH. In adoptive cell transfer experiments, it was shown that sensitized peritoneal cells incubated with optimum doses of the antigen in the presence of specific antiserum in vitro failed to transfer the delayed response to normal recipients, whereas the treatment of the sensitized cells with the antigen or with the antiserum separately did not impair the ability of these cells to transfer DTH. The effect of desensitization is specific and is not permanent. The DTH reappears 3–4 wk after desensitizing injection.     

Cellular aspects of tolerance. III. The responsiveness of T cells from tolerant donors after exposure to a cross-reacting antigen

Mice, rendered tolerant to rabbit gamma globulin (RGG), were immunized with RGG or with dinitrophenylated RGG (DNP₄₀-RGG), incorporated in adjuvant. The resulting response was evaluated in terms of the half-life of trace labeled RGG (¹³¹I-RGG). An antibody response against the tolerance inducing macromolecule could be elicited with DNP₄₀-RGG, but not with RGG. Reconstitution experiments revealed that thymus derived (T) cells from tolerant donors could cooperate with bone marrow cells from normal donors in the response elicited by DNP₄₀-RGG, but could not effectively cooperate with bone marrow derived (B) cells from tolerant donors. Such B cells could cooperate with T cells from normal donors. The relative difference between native and chemically modified proteins played an important role in this tolerance circumvention, since analogous experiments with human instead of rabbit gamma globulin did not result in an effective response to determinants of the tolerance-inducing proteins. It was suggested that the number of effectively immunogenic determinants on DNP₄₀-RGG was low in B and in T cells of animals tolerant to RGG and that the probability of effective cooperation was consequently extremely low. If one of the two cell types came from a normal animal and thus could respond to a large number of determinants, the probability of effective cooperation increased so as to reveal the responsiveness of the “tolerant” cell population. There was no indication that the responsiveness of the tolerant T cell population was directed against tolerance-inducing determinants.     

Immobilization of formate dehydrogenase from Candida boidinii through cross-linked enzyme aggregates

We employed a cross-linked enzyme aggregate (CLEA) method to immobilize formate dehydrogenase (FDH) from Candida boidinii. The optimal conditions for the preparation of CLEAs were determined by examining effects of various parameters: the nature and amount of cross-linking reagent, additive concentration, cross-linking time, and pH during CLEA preparation. The recovered activities of CLEAs were significantly dependent on the concentration of glutaraldehyde; however, the recovered activity was not severely influenced by the content of dextran polyaldehyde as a mild cross-linker. Bovine serum albumin (BSA) was also used as a proteic feeder and enhanced the activity recovery by 130%. The highest recovered activity of CLEA was 18% for formate oxidation reaction and 25% for CO₂ reduction reaction. The residual activity of CLEA prepared with dextran polyaldehyde (Dex-CLEA) was over 95% after 10 cycles of reuse. The thermal stability of Dex-CLEA was increased by a factor of 3.6 more than that of the free enzyme. CLEAs of FDH could be utilized efficiently for both NADH regeneration and CO₂ reduction.     

Norfloxacin Zn(II)-based complexes: acid base ionization constant determination, DNA and albumin binding properties and the biological effect against Trypanosoma cruzi

Zn(II) complexes with norfloxacin (NOR) in the absence or in the presence of 1,10-phenanthroline (phen) were obtained and characterized. In both complexes, the ligand NOR was coordinated through a keto and a carboxyl oxygen. Tetrahedral and octahedral geometries were proposed for [ZnCl₂(NOR)]·H₂O (1) and [ZnCl₂(NOR)(phen)]·2H₂O (2), respectively. Since the biological activity of the chemicals depends on the pH value, pH titrations of the Zn(II) complexes were performed. UV spectroscopic studies of the interaction of the complexes with calf-thymus DNA (CT DNA) have suggested that they can bind to CT DNA with moderate affinity in an intercalative mode. The interactions between the Zn(II) complexes and bovine serum albumin (BSA) were investigated by steady-state and time-resolved fluorescence spectroscopy at pH 7.4. The experimental data showed static quenching of BSA fluorescence, indicating that both complexes bind to BSA. A modified Stern–Volmer plot for the quenching by complex 2 demonstrated preferential binding near one of the two tryptophan residues of BSA. The binding constants obtained (K _b ) showed that BSA had a two orders of magnitude higher affinity for complex 2 than for 1. The results also showed that the affinity of both complexes for BSA was much higher than for DNA. This preferential interaction with protein sites could be important to their biological mechanisms of action. The analysis in vitro of the Zn(II) complexes and corresponding ligand were assayed against Trypanosoma cruzi, the causative agent of Chagas disease and the data showed that complex 2 was the most active against bloodstream trypomastigotes.     

Analysis of binding curves in multivalent antigen-heterogeneous antibody systems

The binding of antibodies to N-valent antigens can be utilized to gain information on the antibody affinity distribution, under the assumption, that the formation of each bond occurs as an independent event. The analysis of the most widely used plots of binding data (double reciprocal, Scatchard, Sips and the so-called “avidity” plot) leads to expressions which correlate asymptotical features of the binding curves to the antibody site concentration to the antigen valence and to the affinity moments <K^?1>, <K> and <K²>. Only in the “avidity” plot is the shape of the curve independent of the valence of the antigen, depending merely on the antibody affinity distribution.     

Characterization of the cell-mediated cytotoxic responses of isogeneic ginbuna crucian carp induced by oral immunisation with hapten-modified cellular antigens

A better understanding of the immune responses in fish elicited by oral immunisation is of importance for the development of new and effective oral vaccines for cultured fish. In the present study, we characterized specific cell-mediated cytotoxic responses in isogeneic ginbuna crucian carp (Carassius auratus langsdorfii) following oral immunisation with cellular antigens. Trinitrophenyl- (TNP) or dinitrophenyl- (DNP) modified syngeneic and allogeneic cells were used for studying the fine specificity and genetic restriction of orally-induced cytotoxic cells. Hapten-specific cytotoxic responses were detected in peripheral blood leukocytes (PBLs) of fish orally immunised with haptenated syngeneic cells. PBLs from orally immunised fish had cytolytic activity for haptenated syngeneic cells, but they showed little reactivity against both haptenated and unmodified allogeneic targets. Similarly, oral immunisation of fish with hapten-modified allogeneic cells did not induce hapten-specific cytotoxic cells which can lyse haptenated syngeneic targets. Although ginbuna crucian carp possess spontaneous cytotoxic cells that are capable of killing mammalian tumour cells, cold target inhibition studies suggested that such spontaneous cytotoxic cells were not involved in the killing of haptenated syngeneic targets. Oral immunisation of fish with haptenated syngeneic cells also induced hapten-specific cytotoxic memory responses. Oral administration of haptenated fixed cells also effectively induced hapten-specific cytotoxic cells in the treated fish. These findings suggest that oral immunisation with antigens can elicit antigen-specific cytotoxic cells that are capable of recognizing antigens in an MHC-restricted manner. In addition, our results provide indirect evidence that fish possess a mechanism for taking up exogenous non-replicating antigens from the alimentary tract and generating antigen-specific cytotoxic cells.     

Expression of IgG memory response in vitro to thymus-dependent and thymus-independent antigens

A model is described in which expression of IgG secondary antihapten responses of large magnitude can be initiated in vitro without resorting to in vivo boosting prior to culture. The number of IgG plaque-forming cells (PFC) is frequently as much as 100-fold greater than that of IgM PFC. Spleen cells from mice primed with trinitrophenylated keyhole limpet hemocyanin (TNP-KLH) several months earlier are stimulated in vitro to produce an anti-TNP plaque-forming cell response 7–10 days later. The in vitro IgG response can be elicited with either a thymus-dependent antigen (TNP-KLH) or thymus-independent antigens (TNP-T₄ bacteriophage or DNP-dextran). The kinetics of the responses to these two forms of antigen differ in that the thymus-independent response peaks two days earlier. The IgG response to both forms of antigen requires the presence of 2-mercaptoethanol (2-ME) even though macrophages are not depleted prior to culture. In the absence of the reducing agent both thymus-dependent and thymus-independent IgG responses were diminished ≥90%. The magnitude of the response to thymus-independent antigens emphasizes the ability of these materials to elicit IgG expression in memory B cells provided optimal conditions for memory development and in vitro expression exist.     

Strong Influence of the Processing of the Antigen on Negative Effects on T Cell Activation by Regions Outside the Determinant Area of Bovine αs1-Casein

α_s1-Casein can elicit a proliferative response in responding T cell clone 3D20 cells (specific for I-A^b plus fragment 136–151), even when using fixed splenic antigen-presenting cells (APC) not carrying antigen processing ability. The order of potency of each tested antigen for fixed APC was the determinant peptide (136–151) > the long peptide (136–195) > the intact protein (199 residues), indicating that regions outside the determinant area negatively affected the stimulatory potency of the antigens. On the other hand, the order for normal splenic APC was the short peptide > the intact protein > the long peptide. This shows that negative effects by regions outside the determinant area were strongly influenced by the antigen processing.     

Recognition of DNP-amino acid determinants by primed guinea pig T cells

Guinea pigs were immunized with DNP-mycobacteria prepared by reaction of the bacteria with dinitrofluorobenzene. T cells from these animals responded as well against DNPazo-proteins and proteins haptenated by reaction with DNP-sulfonic acid as against proteins haptenated by the same method as the immunogen. These cells responded also to free DNP-O-tyrosine, to the DNPazo derivatives of N-acetyltyrosine and of aminocaproic acid, and to α-N-DNP-l-asparagine. The experiments suggest that at least part of the DNP-responsive lymphocyte population found in animals primed with DNP-mycobacteria recognized a determinant no larger than the hapten with a single amino acid side chain, and that there was no requirement for the side chain to be aromatic (such as tyrosine or histidine) in order for recognition to occur. Determinants of this size were capable, in free form, of stimulating primed cells to mount an in vitro proliferative response. DNP-amino acids were not effective, however, as primary immunogens.     

Internalization of alpha 2 macroglobulin in receptosomes. Studies with monovalent electron microscopic markers

Previous studies using multivalent, peroxidase-labeled antibody for localization of α₂ M have demonstrated that the binding of α₂-macroglobulin (α₂ M) to randomly distributed receptors on the surfaces of fibroblasts initiates the accumulation of α₂ M-receptor complexes in clathrin-coated pits. The α₂ M-receptor complexes are then internalized into a specialized vesicle termed the receptosome. In the present study we have used three different monovalent ligands to localize α₂ M and show that the endocytosis of α₂ M-receptor complexes by the receptosomal pathway is not initiated as a result of antibody-induced cross-linking of the α₂ M-receptor complexes. To perform these studies the following monovalent markers of α₂ M were prepared for electron microscopic visualization: (1) a monovalent hybrid antibody directed against α₂ M and ferritin; (2) a monovalent hybrid antibody directed against α₂ M-peroxidase; and (3) a direct 1:1 conjugate of α₂ M-peroxidase. We find that all three of the markers are internalized by the ligand pathway previously described using multivalent labels. The steps involved are clustering of α₂ M receptor complexes in coated regions of the plasma membrane followed by endocytosis of α₂ M into receptosomes. Our results are contrasted with previous studies on lymphocytes in which antibody induced cross-linking of membrane antigens was necessary for triggering their pinocytosis. The methods described in this paper are applicable for visualizing at the electron microscopic level the internalization of other ligands and hormones.     

Contact sensitivity in rabbits sensitized with dinitrophenylated Mycobacterium tuberculosis

The conjugation of Mycobacterium tuberculosis with DNFB results in the formation of a haptenated preparation that induces the formation of contact sensitivity when administered subcutaneously. This contact sensitivity can be measured in vivo by topical application of the free chemical and in vitro by lymphocyte transformation. The antigens suitable for the in vitro detection are those preparations obtained by the haptenation of cell membranes. Haptenation of serum proteins, of homologous and heterologous origin, does not produce antigens suitable for in vitro assay. The antigen requirements for the in vitro transformation assay of contact sensitivity are similar for adjuvant induced sensitivity as well as for free chemical induced sensitivity.     

Osmotic pressure method to measure salt induced folding/unfolding of bovine serum albumin

A new approach has been developed to monitor protein folding by utilizing osmotic pressure and a range of salt concentrations in a well characterized protein, bovine serum albumin (BSA). It is hypothesized that both the ‘effective’ osmotic molecular weight, A_e, and the solute/solvent interaction parameter, I, in the empirical relation $\text{M}{}_{\mathrm{solvent}}\text{M}{}_{\mathrm{solute}}\text{= (}\text{RTϱ}\text{A}{}_{\mathrm{e}}\text{)}\text{1}\text{gp}\text{+ I}$ [1] can be used as measures of protein folding. I is a measure of solvent perturbed by the solute and is thought to depend directly upon the solvent accessible surface area (ASA). It is reasoned that larger solvent accessible surface area of an unfolded or denatured protein should perturb more water and produce larger I-values. Thus I-values allow calculation of a unfolded protein fraction, f_u^a, due to changes in relative solvent accessible surface area. It has been observed that A_e decreases for filamentous, denatured proteins due to segmental motion of the molecule [2]. This allows calculation of unfolded protein fraction from the effective molecular weight, f_u^m. Colloid osmotic pressure of BSA was measured in a range of salt concentrations at 25°C, and pH = 7 (above the isoelectric point of BSA at pH = 5.4). Both S and I were used to monitor protein folding as the salt concentration was varied. In general, larger and variable I-values and smaller A_e were observed at salt concentrations less than 50 mmolal NaCl (I_max = 8.9), while constant I = 4.1 and A_e = 66,500 were observed above 50 mmolal NaCl. The two expressions for fractional unfolding (f_u^a and f_u^m) are in general agreement. Small differences in the parameters below 50 mmolal salt concentration are explained with well known shifts in the relative amounts of α-helix, β-sheet and random coil in denatured BSA. The relative amounts of these shifts agree with predictions in the literature attributed to continuous BSA expansion rather than an ‘all-or-none’ conversion.     

Purification of the creatine kinase isozymes of the green sunfish (Lepomis cyanellus) with Blue Sepharose CL-6B.

Three homodimeric creatine kinase isozymes (A₂, B₂, and C₂) of the green sunfish (Lepomis cyanellus) were purified by a combination of affinity chromatography, gel filtration, and preparative starch gel electrophoresis. The final preparations were isozymically pure and were used to elicit antibodies in rabbits. The use of the group-specific adsorbant Blue Sepharose CL-6B (Pharmacia) and specific elution conditions for creatine kinase facilitated purification. Fish creatine kinase isozymes are sensitive to denaturation and cannot be readily purified by procedures routinely used for mammalian creatine kinase isozymes.     

Improvement of antibody affinity by introduction of basic amino acid residues into the framework region

Antibodies are widely used not only as therapeutic agents but also as research tools and diagnostic agents, and extensive efforts have been made to generate antibodies that have higher affinity. It was recently reported that introduction of charged residues into the framework region of an antibody improved its affinity; however, the underlying molecular mechanism has not been elucidated. In this study, we used kinetic and thermodynamic analyses of the antibody–antigen interaction to investigate the molecular mechanism by which an antibody with introduced charged residues recognizes its antigen with higher affinity. The introduction of basic amino acid residues resulted in improvement of the affinity whereas the introduction of acidic residues weakened the interaction. For two mutant antigen-binding fragments (Fabs) with improved affinity (named K5- and R5-mutants), the balance between the association rate constant k_on and the dissociation rate constant k_off was distinct despite each mutant having the same number of charged residues. Moreover, thermodynamic analysis of the interactions in the transition state revealed a difference between the K5- and R5-mutants in terms of enthalpic energy change following formation of the encounter complex with the antigen. These results suggest that the affinity of the K5- and R5-mutants is improved by distinct mechanisms. Although the mutations destabilize the Fab and necessitate further studies, our strategy is expected to become a versatile and simple means to improve the affinity of antibodies to their antigens.