Tumor metastases and cell-mediated immunity in a model system in DBA/2 mice. VIII. Expression and shedding of Fc gamma receptors on metastatic tumor cell variants

The expression of receptors for the Fc portion of IgG immunoglobin molecules was studied on tumor cell lines with high and low metastatic capacity. Two tumor cell lines from DBA/2 mice that had high metastatic activity, ESb and MDAY-D2, contained a high percentage of Fc receptor positive cells, as detected in a rosette assay with IgG antibody-coated erythrocytes (EA). In contrast, the low metastatic parental line Eb, from which ESb was derived, contained only a low percentage of EA-rosette-forming cells. ESb ascites tumor cells adapted to tissue culture in the presence of 2-mercaptoethanol (2ME) had a high expression of Fc receptors, whereas a cell line adapted to tissue culture in the absence of 2ME had a low expression of Fc receptors. "Soluble" Fc receptors were detectable by their ability to bind to EA and to cause blocking of rosette formation. They were found to be present in fluids from tumor-bearing animals, such as serum and cell-free ascites. Even animals with an ascites tumor of the low-metastatic line Eb contained "soluble" Fc receptors. The results are discussed with regard to their possible significance for tumor metastasis.

Characterization of antigens on murine tumor cells reacting with alloantisera against foreign H-2 specificities: analysis by absorption with purified murine leukemia virus and normal lymphoid cells of different H-2 haplotypes

A spontaneous T cell lymphoma of DBA/2 (H-2-d) mice, SL2, was found to react with anti H-2 typing sera raised against certain foreign haplotypes as well as with anti H-2d sera. The cytotoxic anti-SL2 activity of the anti-foreign H-2 sera was detected in a newly developed microradioassay, not however, in a conventional 51Cr release test. Upon culture in vitro the reactivity of the tumor cells with the anti H-2 sera decreased. The anomalous cytotoxic anti-tumor activity of the anti-foreign H-2 sera appeared to be distinct from anti-murine leukemia virus activity, since it was not removed by absorption with either Friend of AKR leukemia virus. Partial absorption was observed with normal lymphoid cells carrying the respective foreign H-2 antigens, but not with cells of unrelated H-2 haplotypes. In each serum tested, the anti-tumor activity could also be absorbed with syngeneic H-2d lymphoid cells. These results show that the anomalous anti-tumor reactivity of certain anti H-2 typing sera, in particular of sera raised in recipients differing in H-2 from the tumor host strains, is not due to the presence of foreign (derepressed) H-2 molecules on the tumor cells. The differences observed between the tumor cells and normal cells seem to be due to unexpected antibodies in the sera reacting with public H-2 specificities which are better exposed on the tumor cells than on normal cells.

Human (HLA-A and HLA-B) and murine (H-2K and H-2D) histocompatibility antigens are cell surface receptors for Semliki Forest virus

The proteins coded for by the HLA-A and HLA-B loci in man and the H-2K and H-2D loci in mice were identified as cell surface receptors for Semliki Forest virus. This conclusion is based on the following observations: (i) Water-soluble octamers of viral coat proteins inhibit the complement-dependent cytotoxicity of antibodies directed against H-2K and H-2D antigens in mouse cells. (ii) Isolated detergent-soluble HLA-A and HLA-B antigens reconstituted in lipid vesicles inhibit the binding of viral proteins to human cells (as do the water-soluble antigens to a lesser extent). (iii) Reconstituted HLA-A and HLA-B vesicles interact in solution with Semliki Forest virus (or with vesicles containing viral spike proteins), as demonstrated by coprecipitation with antisera. (iv) Complexes between viral spoke proteins and HLA-A and HLA-B antigens or H-2K and H-2D antigens can be isolated from the cell surface by utilizing affinity chromatography or immunoprecipitation.

Studies on H-2 specificities on mouse tumour cells by a new microradioassay

Seven mouse tumour cell lines of different aetiology and origin were tested for the expression of surface alloantigens using twenty-four well defined H-2 alloantisera and anti-Thy 1.2. We used a new radioassay that involves antibody-complement treatment of the tumour target cells followed by postlabeling the surviving tumour cells with 14C-thymidine. With a relatively high frequency the anti-H-2 sera were reacting differently with the tumour cells than with respective syngeneic lymphoid cells. Thirty-six anomalous reactions out of 129 investigated were detected. Absorbtion experiments performed with H-2 antigen positive or negative lymphoid cells revealed a striking similarity between these extra-specificities and H-2 specificities of foreign haplotypes. The results are discussed in relation to the biological importance of the extra-specificities, both with regard to origin (derepression) and function (transplantation antigen properties).

Effect of thymic factor on human lymphoid cells of umbilical cord blood and of children with T cell deficiency

We report on the in vitro effect of a thymic factor (TF) extracted from pig thymuses, on human lymphoid cells from umbilical cord blood and from peripheral blood of 8 T cell-deficient patients. E rosette formation was not affected by TF when tested on cells from peripheral blood of normal adults. With cells from umbilical cord blood of 13 healthy, full-therm newborn babies, the difference between the percent (mean) of ERFCs before (16.31 +/- 11.13) and after (28.85 +/- 17.10) incubation with TF was statistically significant (p less than 0.05). In most samples, TF transformed about 10-20% of the cells. In the T cell-deficient group the increase in ERFCs of the peripheral blood lymphocytes, though consistent, was variable in degree from case to case. Our data indicate that precursor cells in some individuals with T cell deficiency are very sensitive to TF. Patients with highly responsive precursors appear to be the best candidates for a therapeutic approach with TF when thymus transplant is not possible.

Fc receptors and Ia antigens

Antibody against Ia antigens inhibits the ability of B lymphocytes to bind aggregated immunoglobulin and to form EA rosettes. The explanation suggested for this phenomenon has been that Ia antigens are identical to or closely associated with Fc receptors. But a variety of observations preclude acceptance of this explanation since inhibition is also demonstrable with antibody against a wide variety of B cell surface components, including H-2K, H-2D, beta2 microglobulin, immunoglobulin, Ly 4.2. Furthermore, Fc receptor function can be separated from Ia antigens by capping or by isolation of membrane components. It seems likely that the mechanism of inhibition of Fc receptors by antibody against Ia antigens is part of a broader spectrum of effects induced when antibody binds to cell membrane antigens.

Interactions of Fc receptors with antibodies against Ia antigens and other cell surface components

Two Fc receptor-dependent tests were investigated to study the question of a relationship between Fc receptors and known cell surface antigens, in particular I region-associated (Ia) antigens: (a) a rosette assay with antibody-coated erythrocytes (EA) as indicator cells and normal mouse lymphoid cells as source of rosette-forming cells, and (b) a cytotoxicity test with antibody-coated erythrocytes as target cells and normal mouse spleen cells as a source of cytotoxic cells (K cells). EA rosettes were specifically inhibited by antibodies reacting with Ia antigens. Various other antisera reacting with antigens on B lymphocytes, like anti-Ly 4.2 (raised in H-2 identical mice), rabbit antimouse B-cell serum, or rabbit antimouse immunoglobulin, also specifically inhibited the rosettes. No inhibition occurred in the presence of allogeneic or xenogeneic antisera reacting with T lymphocytes. K-cell cytotoxicity was specifically inhibited by each of the antisera (reacting with either B cells or T cells). F(ab')2 fragments of anti-Ia antibodies could still specifically inhibit EA rosettes but they could not inhibit K-cell cytotoxicity. Similar results were obtained with F(ab')2 fragments of anti-immunoglobulin antibodies. These results indicate that the mechanism of inhibition of Fc receptors in the two tests was different. In neither of the tests could we find any evidence for a unique association between the Fc receptors and Ia antigens. The Fc receptors on K cells did not seem to be associated at all with Ia antigens.

Immune Responses Against Native and Chemically Modified Albumins in Mice

Albumin antigens were chemically modified at their charged residues by various means (i.e., methylation, acetylation, succinylation) and the effect of the modification reactions on humoral and cellular immune responses was compared. The modifications resulted in profound changes of the molecules' physicochemical properties (charge, conformation, state of aggregation) and in a destruction of the majority of serologically detectable native albumin determinants. On the level of the humoral antibody response the effects of the modifications were i) pronounced reduction of the response against native determinants and ii) induction of an antibody response against new antigenic determinants (NAD) on the modified proteins.

Helper (T) lymphocyte responses as a test for cellular immunity were investigated in cell transfer experiments. When using limiting doses of immunogen for primary immunization, it could be shown that none of the modification reactions had any negative impact on the capacity of the albumin to induce helper cells reacting with the native albumin. In the same experiments, however, a marked reduction on the level of albumin-specific antibody-forming cell precursors (B cells) was observed. These results suggest that there are different structural requirements for the induction of B and T cell responses against native protein antigens.

Compared to antibodies, helper cells induced by differently modified albumins seemed to have a higher degree of cross-reactivity. However, they also revealed a notable discriminating capacity, suggesting that both antibodies and helper cells were able to distinguish on the modified proteins NAD from native determinants. Possible interpretations of the results are discussed.

A new sensitive assay for antibody against cell surface antigens based on inhibition of cell-dependent antibody-mediated cytotoxicity. I. Specificity and sensitivity

Inhibition of cell-dependent antibody-mediated cytotoxicity has been investigated as a new assay for antibody against cell surface antigens. The cytotoxicity system consisted of effector cells (normal mouse spleen cells), target cells ((61)Cr-labeled chicken erythrocytes), and antitarget cell antibody. Addition of antibody against cell surface antigens in the effector cell population regularly inhibited the cytotoxicity measured in this system. This cytotoxicity inhibition assay (CIA) detected antibody with a variety of specificities: anti-H-2, anti-Thy 1.2, anti-immunoglobulin, and antimouse bone marrow-derived lymphocyte antigen. When the inhibition by anti-H-2 sera was analyzed using effector cells from congenic mice, the activity was found to be directed against specificities mapping in the H-2K, H-2D, and I regions of the H-2 complex, correlating well with the specificities characterized by complement-dependent assays. A comparison between the sensitivity of the CIA and complement-dependent lysis revealed that the CIA was 2-11 times more sensitive for anti-H-2 antisera and 20-780 times more sensitive for certain antisera against subpopulations of the spleen cells (i.e., T cells or B cells). The CIA proved to be precise, sensitive, and reliable. It may become a very useful antibody assay in various species including man.

Cytotoxic Immune Cells with Specificity for Defined Soluble Antigens

Antigen-coated target cells were specifically lysed (a) by spleen cells from mice that had been immunized against the antigen presented in a soluble form and (b) by non-immune mouse spleen cells in the presence of specific antibodies. In the immune cell cytotoxic system, which consists of antibody-producing cells (APC) and cytotoxic effector cells (CEC), limiting dilution assays showed that the APC were diluted out before the CEC so that the amount of antibody being produced in vitro was limiting the cytotoxicity. In the presence of optimal concentrations of antibody, CEC from immune and non-immune cells were equally effective, regardless of whether the immune cells and the antibody were of the same or of different specificity.

In vitro cytotoxicity in the immune spleen cell system started after a lag period of about 3 hr and was depressed to about 50% after the cells had been treated with trypsin. The cytotoxic activity of non-immune spleen cells in the presence of antibody started without a lag period and was not affected by trypsin treatment of the cells.

Both the immune and the non-immune cell cytotoxic system were blocked by unrelated antigen-antibody complexes. Antibody-complexed target cells were much more potent inhibitors than were soluble complexes. Small numbers of target cells heavily coated with antibody were as efficient in blocking as were larger numbers of target cells lightly coated with antibody. The inhibition by antibodycomplexed antigen-coated target cells was neutralized by the addition of soluble antigen. This “unblocking” was thought to be due to a competition between soluble and insoluble antigen for antibody and the creation of non-inhibitory soluble complexes. The relevance of these findings for anti-tumor immune systems is discussed.

Cytotoxic immune cells with specificity for defined soluble antigens. IV. Antibody as mediator of specific cytotoxicity

Spleen cells from mice immunized against ovalbumin (OA) or dinitrophenylated mouse serum albumin (DM) were found to be specifically cytotoxic in vitro towards target cells (chicken red blood cells) coated with these antigens. Inhibition of specific cytotoxicity was observed when free soluble antigen was added to the incubation mixtures. DM-immune cell cytotoxicity could be specifically and completely inhibited by DNP-lysine and was thus shown to be hapten specific. Complete and specific inhibition was also observed for OA-immune cell cytotoxicity using OA as inhibitor, but compared with the inhibition curves obtained with DNP-lysine, the OA cytotoxicity inhibition curves were shifted by a factor of about one hundred towards lower molar inhibitor concentrations. Very similar results were observed when the serum antibodies of DM- and OA-immune animals were analyzed by passive hemagglutination inhibition. With increasing time after immunization, both cytotoxicity inhibition curves and agglutination inhibition curves, shifted to lower antigen or hapten concentrations. Specific cytotoxicity against antigen-coated target cells was induced in nonimmune spleen cells (a) by serum from immune animals, and (b) by supernatants from in vitro immune cell cultures. In both instances, the factor which induced antigen-specific cytotoxic activity could be absorbed on anti-mouse Ig columns, thus demonstrating its immunoglobulin nature. The ability of target cell bound antibodies to induce cytotoxicity in nonimmune spleen cells was restricted to the 7S antibody class.

The Immune Response Against Hapten-Autologous Protein Conjugates in the Mouse

The present experiments were carried out to determine whether antibody formation in response to a hapten coupled to a non-immunogenic carrier was thymus-dependent. Mouse serum albumin (MSA) was used as the nonimmunogenic carrier. The anti-hapten responses to hapten-MSA conjugates of different hapten density were compared to the anti-hapten responses to hapten coupled to heterologous carriers in the following types of mice: 1) normal mice, 2) thymectomized, irradiated, and bone marrow cell reconstituted (TXBM) mice, 3) “sham”-thymectomized mice, and 4) TXBM mice further reconstituted with thymocytes (TXBMT mice).

The anti-hapten response to hapten-MSA conjugates of medium and high density were all found to be suppressed in TXBM mice, whereas in TXBMT mice it was fully reconstituted. Thus, the antibody response to hapten-MSA conjugates in the mouse is thymus-dependent regardless of hapten density.

Immune responses against native and chemically modified albumins in mice. I. Analysis of non-thymus-processed (B) and thymus-processed (T) cell responses against methylated bovine serum albumin

Immune cells induced by bovine serum albumin (BSA) and its methylated derivative (MBSA) have been compared in a cooperative cell transfer system for their content of BSA-specific antibody-forming cell precursors (AFCP, B) and BSA-specific helper (T) cells. When MBSA immune cells were transferred together with hapten-primed cells into recipient mice which were stimulated by a hapten-BSA conjugate, their cooperative secondary anti-hapten response was as good as in case of transferred BSA immune cells. Their secondary anti-BSA response, however, was markedly reduced (reduction factor > 30). Hapten-MBSA conjugates had the same capacity to react with BSA-specific helper cells in the cooperative secondary anti-hapten response as hapten-BSA conjugates but had a reduced ability to react with BSA-specific AFCP cells. In spite of the pronounced reduction of the B cell response, MBSA had the same threshold dose as BSA for activating BSA-specific T cells. These data suggest that B and T cells recognize different epitopes on the BSA molecule, only those recognized by B cells being affected by the methylation procedure.

Determination of antibody class in a system of cooperating antigenic determinants

19S and 7S memory is analyzed in a system of cooperating antigenic determinants. Cooperation occurs in the induction of both 19S and 7S secondary antibodies, and for both responses carrier specificity can be entirely accounted for by presensitization of the animal to carrier determinants. The class distribution of secondary anti-hapten antibody depends on the dose of the hapten primary-carrier conjugate used for priming, and on the time interval between priming with the hapten primary-carrier conjugate and secondary injection. The conditions of priming with the secondary carrier influence the extent of the secondary response but not the class distribution of secondary antibody. The data confirm the cooperation hypothesis of antibody induction. Specifically, we interpret them to mean that in hapten-carrier cooperation, the hapten-specific memory cells are predetermined for the class of the emerging antibodies. Together with the hapten-specific memory cells, the carrier-specific helpers are responsible for the extent of the secondary response.

The requirement of more than one antigenic determinant for immunogenicity

Rabbits primarily stimulated with a BSA (bovine serum albumin)-sulfanilic acid complex will produce a good secondary response to the sulfanilic acid hapten if the carrier used in the secondary stimulus is again BSA, and not if the secondary carrier is HGG (human gamma globulin). In the latter situation, a good secondary response is obtained, however, if the rabbits are pretreated a few weeks earlier with free HGG. We conclude that the immune stimulus involves the recognition of carrier determinants unrelated to the hapten. As the receptors for recognition of unrelated determinants are probably situated on different cells, we suggest that the immune stimulus leading to antibody formation requires the interaction of two antigen-bridged cells.