J W Uhr

T Cell-Derived Lymphokines That Induce IgM and IgG Secretion in Activated Murine B Cells

Observations by Dutton etal.{1975)andSchimpI&Wecker(1972)demonstrated that cell-free supernatants obtained from activated T cells could replace T cells in the induction of antibody secretion. It is now clear that T cell-derived lymphokines affect many aspects of the growth and differentiation of B cells. In recent years, the lymphokines responsible for these activities have been functionally defmed in a number of experimental systems and the results have revealed a complexity previously not appreciated. For example, T cells can secrete several different B cell growth factors (BCGF) {Yoshizaki et al. 1983, Okada et al. 1983, Swain et al. 1983, Nakanishi et al. 1983), and a variety of nonspecific and isotype-specific differentiation factors (Dutton 1975, Harwell et al. 1976, Parker etal. 1979, Takatsu et al. 1980, Swain etal. 1981, Pure et al. 1981, Kishimoto & Ishizaka 1973, Isakson et al. 1982, Hirashima et al. 1981, Kawanishi et al. 1983). Purification of these lymphokines and their receptors on B cells will be critical in elucidating the mechanisms underlying clonai expansion, differentiation, and development of memory in lymphocytes ofthe B cell lineage.

Neoplastic B Cells as Targets for Antibody‐Ricin A Chain Immunotoxins

The targeting of toxic agents to tumor cells in vivo has been a goal of immunological research since the studies of Ehrlich (Himmelweit 1960). Successful application of this technique requires tumor-specific antibody whose activity remains unaltered following its covalent conjugation to a toxic agent. Moreover, the toxic portion of the conjugate should remain inactive until bound to the tumor cell via its antibody portion. In the studies described in this review, we have used an antibody against the cell surface immunoglobulin idiotype (Id) of a B cell tumor as our model system. B cell tumors are virtually always monoclonal in origin; hence each tumor cell that bears surface immunoglobulin expresses a particular Id (Fu et al. 1975. Schroer et al. 1974, Salsano et al. 1974). Since the Id is present on only a very small number of normal B cells (approximately l/lO*), the Id is operationally a tumor-specific antigen. Other advantages of using B cell tumors as model systems for antibody targeting studies include the availability of antibodies against other determinants on the cell surface immunoglobulin molecule (isotype, allotype, etc.) and a large body of information concerning the role of particular organs (e.g., spleen), cell types (T cells, macrophages) and lymphokines on the replication and differentiation of both normal and neoplastic B cells. In addition, a large number of humans with B cell tumors do not respond well to conventional chemo/radiotherapy (Lennert & Mohri 1978) and there is, therefore, a need to improve treatment. Recent studies utihzing antibody-directed targeting of toxic peptides to the surface of neoplastic cells are summarized in Table I and are reviewed in this volume. In general, the results of these studies have been promising. However,

Structural studies on the murine ia alloantigens-III. Tryptic peptide comparisons of allelic products of the I-E/C sub-region

Splenocytes from B10 congenic mice of the k, p, d, and r haplotypes were radiolabeled with 3H- or 14C-arginine, lysine, leucine and tyrosine. The E/C sub-region products were precipitated from NP40 lysates with specific alloantisera and the la α and β sub-units of each of the above four haplotypes were isolated by gel electrophoresis. Molecules to be compared, one 3H-labeled and the other 14C-labeled, were mixed, digested with trypsin, and the acid soluble peptides analyzed by ion-exchange chromatography. The results indicate that the α sub-units of p, d and r are approximately 90% homologous to k. In contrast, β polypeptides of the above three haplotypes have only about 50% coincidence of peptides with the k allelic product. This degree of structural variation suggests that the β sub-unit of the E/C product is encoded by the major histocompatibility complex.

Antigen-specific memory and virgin B cells differ in their requirements for conjugation to T cells

The physical interaction between carrier-specific T hybridoma cells and long-term primed, TNP-specific memory B cells (TNP-MABC) exposed to TNP-OVA was compared to that of unprimed, TNP-specific virgin B cells (TNP-ABC). The direct conjugation of the T and B cells was visualized at the light microscopic level and the number of T/B conjugates was quantified directly. The results demonstrate that the TNP-MABC, as compared to TNP-ABC, formed T/B conjugates after a shorter exposure time to antigen and at a 10-fold lower concentration of antigen. Conjugate formation was inhibited almost completely by treating the TNP-MABC with concentrations of chloroquine that only partially inhibited the ability of the TNP-ABC to form conjugates. Exposure of the T hybridoma cells or the TNP-ABC to monoclonal antibodies directed against cell surface antigens prior to conjugation indicated that L3T4, Thy-1.2, and LFA-1 antigens on the T cells and LFA-1 and I-A antigens on the TNP-ABC are involved in conjugate formation. However, in contrast to the TNP-ABC where treatment of the B cells with anti-LFA-1 blocked T/B conjugate formation, pretreatment of the TNP-MABC with anti-LFA-1 had no effect.

Preparation and application of antibodies coupled to the A chain of ricin.

Publisher Summary This chapter describes the selective cytotoxicity of a variety of normal and neoplastic cells by toxic peptides that have been covalently coupled to specific antibodies. The synthesis and applications of such antibody–toxin conjugates is also discussed in the chapter. Ricin is an example of a toxin that has been used in this manner. Ricin is obtained from castor beans and is composed of two 3 x 10 4 dalton polypeptide chains bridged by a disulfide bond. One chain (the B chain) has binding specificity for galactose and is responsible for the binding of the toxin to the surface of ricin-sensitive cells. Once bound, the other peptide chain (A chain) enters the cytoplasm and catalytically and irreversibly inhibits protein synthesis. The covalent coupling of antibody to the A chain of ricin can be performed by a disulfide exchange reaction mediated by the heterobifunctional, thiol-containing cross-linker N-succinimidyl-3-(2-pyridyldithio) propionate (SPDP). The hybrid product mimics the structure of the native toxin in that the A chain is covalently linked through a disulfide bridge to a binding moiety. Covalent coupling is performed in three steps: (1) purification of ricin A chain; (2) introduction of 2-pyridyl disulfides to the primary amino groups on the antibody; and (3) ricin A chain substitution via disulfide exchange.

In vivo reconstitution of ricin-like activity from its A and B chain subunits.

The capacity of highly purified preparations of ricin A and B chains to reconstitute ricin toxicity both in vitro and in vivo was studied. When the nontoxic A and B chain subunits were mixed and electrophoresed on sodium dodecyl sulphate-polyacrylamide gels (SDS-PAGE), reconstituted ricin was observed. The mixtures killed cells of the human Burkitts lymphoma cell line Daudi in vitro and killed mice after i.v. injection. It was also shown that when mice were injected with one ricin subunit followed by administration of the complementary polypeptide up to 8 hr later, they died with lesions similar to those of ricin-induced death. This observation suggests that A and B chains recombine either in the serum or on the surface of cells. The rate of clearance of A and B chains from the blood of rats indicated that sufficient concentrations of either chain were present in the circulation 8 hr after injection to account for the observed toxicity. The above studies therefore suggest that the subunits of ricin have a very high affinity for each other and are capable of reconstituting biologically active ricin in vitro and in injected mice.

Primary structural analysis of the Tla region gene products, QA-2 and TL

Abstract The Tla region on the murine 17th chromosome encodes at least three gene products (Qa-1, Qa-2 and TL) which are structurally similar to the major transplantation antigens H-2K, -D and -L. In this communication we compare the tryptic peptides of two Tla region-encoded gene products, Qa-2 and TL. The results indicate that TL and Qa-2 may be only modestly related in amino acid sequence and support the previous suggestion that distinct subfamilies of class I gene products with unique biochemical and functional properties may exist.

MOLECULAR MODIFICATIONS IN VSV-INFECTED CELLS

ABSTRACT Three major observations have emerged from our studies of molecular changes in P815 cells infected with vesicular stomatitis virus (VSV). 1) Glycosylation of H-2 and/or viral glycoprotein is a prerequisite for lysis of infected cells by VSV-sensitized, H-2 identical killer T cells. Treatment of P815 cells before and during VSV infection with the antibiotic Tunicamycin, which specifically inhibited addition of sugars to polypeptides of glycoproteins, inhibited both glycosylation of proteins and lysis of infected P815 cells by syngeneic, virus-immune killer cells. 2) VSV infection caused a 50% decrease in the amount of H-2 on the surfaces of P815 cells 3 hr. after VSV infection. 3) Surface H-2 and viral proteins do not form a detergent-stable association. P815 cells were surface-labeled, then VSV-infected and lysed. The lysate was treated with rabbit anti-VSV (or control) serum plus S. aureus . Immunoprecipitation of VSV antigens failed to deplete H-2 antigens.

Purification and characterization of antigen-binding virgin and memory B cells

Publisher Summary This chapter focuses on the methods for purifying resting antigen-specific B cells so that their activation by antigens and T cells can be assessed. Many methods for studying the physiology and activation of B lymphocytes are available. The chapter describes a method for purifying a population of TNP-specific B cells (TNP-ABC) from spleens of normal or primed mice. These cells are suitable for the analysis of early activation events of B cells in response to antigen. However, to interpret such data, an accurate measurement of the purity of the enriched population is required. The antigen-binding capacity of these cells as measured by rosetting or binding fluorescent antigen gives a far higher estimate of purity than functional assays, such as proliferation, plaque formation, and an enumeration of clones in the splenic focus assay.

The Use of Immunotoxins to Kill Neoplastic B Cells

Paul Ehrlich first discussed the potential use of antibodies as carriers of pharmacologic agents (1). During the last 10 years there have been many attempts to apply this concept to the elimination of neoplastic and other target cells using antibodies coupled to toxic agents. A cell-binding antibody conjugated to a plant or bacterial toxin has been termed an “immunotoxin”. One such toxin, ricin, like most toxic proteins produced by bacteria and plants, has a toxic Polypeptide (A chain) attached to a cell binding Polypeptide (B chain) (2). The B chain is a lectin that binds to galactosecontaining glycoproteins or glycolipids on the cell surface. By mechanisms that are not yet well understood, the A chain of the cell bound ricin gains access to the cell cytoplasm presumably by receptor-mediated endocytosis and penetration of the membrane of the endocytic vesicle (3). There is evidence that the B chain can also facilitate the translocation of the A chain through the membrane of the endocytic vesicle, possibly by forming a pore (4–7). In the cytoplasm, the A chain of ricin inhibits protein synthesis by enzymatically inactivating the EF-2 binding portion of the 60S ribosomal subunit. It is thought that one molecule of A chain in the cytoplasm of a susceptible cell can kill it (3).