Effects of antibodies to choriogonadotropin in malignant growth. I. Rat 3230 AC mammary adenocarcinoma

Human chorionic gonadotropin as a model for a fetal antigen

Human chorionic gonadotropin (hCG) has an important biological function and has clinical use as a hormone, yet its applications as an antigen are also numerous. Pregnancy tests have brought immunological reactions into the high street pharmacy and the home. As a secreted antigen hCG provides for choriocarcinoma of gestational or germ cell origin our closest approximation to the ideal tumour marker. It is unique in providing the basis of a biochemical screening test for a specific cancer in a high risk population. It has a valuable role in diagnosis and prognosis and in monitoring this disease and detecting tumours is unknown. It has been investigated as a target for radioimmunolocalization studies with 131I-labelled anti-hCG antibodies, and found useful in some patients. Attempts have been made to improve the technique using liposomes containing a second antibody. As a target for antibody therapy in fertility control and oncology, by active or passive immunization methods, much depends on whether hCG functions as a cell membrane antigen.

Metastatic phenotype correlates with high expression of membrane-associated complete beta-human chorionic gonadotropin in vivo

BACKGROUND

Investigations using living human cancer cells and the nude mouse model were conducted to evaluate the expression of human chorionic gonadotropin (hCG) in various cancers grown in vitro and in vivo. The aim was to determine whether membrane-associated hCG in any of its forms is a characteristic metastatic marker, and at what levels or ratios.

METHODS

Human cancer cell lines known to produce tumors that metastasize spontaneously when grown in nude mice (n = 4) were compared with those that do not produce such tumors (n = 4) using analytical (quantitative) flow cytometry. Monoclonal antibodies directed to epitopes of intact hCG (hCG-holo) and its subunits, including beta-human chorionic gonadotropin with its carboxy-terminal peptide (hCG beta-CTP), allowed for the determination of hCG beta-CTP/hCG-holo ratios.

RESULTS

No significant difference in hCG beta-CTP/hCG-holo ratios was found between the cultured human cancer cells that do not metastasize spontaneously (ratio = 2.39) and those that do (ratio = 2.13), and no difference was seen in their growth rate in nude mice. However, the cells isolated from tumors that do not metastasize spontaneously showed a decrease in their ratios to values less than 1. They reverted to their original values after reestablishment in culture and subsequent passages. In contrast, the ratios shown by cells isolated from tumors that metastasize spontaneously increased to 3 to 6 times their original values in culture, then reverted to their original values after reestablishment in culture and subsequent passages.

CONCLUSIONS

To our knowledge, these data demonstrate the following for the first time: 1) There is a direct in vivo correlation between human cancer cells that metastasize spontaneously in nude mice and the expression of membrane-associated complete hCG beta (hCG beta-CTP); and the correlation identifies this molecule as a characteristic metastatic phenotype marker. 2) The marked ratio variations under different conditions indicate that the metastatic phenotype is an unstable event. 3) Growth and local invasion in vivo correlates with the expression of hCG-holo.

Immune recognition of a molecule naturally presented as a monomeric or an oligomeric structure: the model of the human chorionic gonadotropin alpha subunit

The immune recognition of a molecule naturally presented as a monomeric or an oligomeric structure is analyzed using the human chorionic gonadotropin alpha subunit (hCG-alpha) as a model. Indeed, hCG-alpha circulates as either a free subunit or combined to the beta subunit (hCG-beta) to form the dimeric hCG hormone. A T cell study was performed in BALB/c (H-2d) mice which were found to be high responders to hCG-alpha. Mice were immunized with the free hCG-alpha or the dimeric hCG alpha/beta, and their lymph node cells were challenged in vitro with either alpha subunits from different species, hCG or peptides spanning the entire primary structure of hCG-alpha. Proliferation and IL-2 assays demonstrated that hCG-alpha-primed lymph node cells responded equally well to hCG-alpha and hCG alpha/beta, suggesting that both the free and combined hCG-alpha subunits are processed in a similar way. Among the various synthetic peptides used, only those mimicking the hCG-alpha(59-92) C-terminus portion were able to stimulate hCG-alpha-primed lymph node cells, demonstrating that this region contains immunodominant T cell recognition site(s). The hCG-alpha(23-43) and (32-59) peptides, although incapable of stimulating T cells primed with hCG-alpha, elicited a T cell response when used as immunogens. These regions encompassed cryptic epitopes which were not generated during hCG-alpha processing in H-2d mice. The T cell epitopes of hCG-alpha above described as immunodominant or cryptic on the free alpha subunit, had similar characteristics when the alpha/beta dimer was used as the immunogen. In contrast, T cells primed with peptides mimicking immunodominant sites recognized differently the hCG-alpha and the hCG alpha/beta antigens. Moreover, the analysis of the B cell response to all the immunogenic hCG-alpha peptides indicated that they bear B and T cell epitopes as well. Antibodies elicited against the hCG-alpha(59-92) or (32-59) peptide were capable of recognizing the alpha subunit in its free form but not in the alpha/beta hCG dimer. Such study deserves attention for the comprehensive mechanisms of the immune response to hCG as well as for the design of anti-hCG vaccines.

Cancer and cell wall deficient bacteria

Since the 1920's, a small number of researchers have been quite regularly isolating highly pleomorphic bacteria from the blood and tumors of humans and animals with cancer. These bacteria show characteristics of cell wall deficient bacteria, and can apparently be observed regularly in the blood of cancer patients by darkfield microscopy. Cancer appears to be induced by injecting these bacteria into experimental animals, and some forms of cancer can be prevented by prevaccination with killed bacteria, isolated from experimental animals affected with the specific cancer. The bacteria regularly produce a protein ressembling chorionic gonadotropin hormone, a substance which appears to protect trophoblastic and cancer cells from immune recognition. There is some evidence that a plasmid may be responsible for this property, and even that these bacteria may in some manner be intimately associated with retroviruses. Further studies are proposed to clarify the role these bacteria play in cancer.

Expression of human choriogonadotropin-like material correlates with metastatic phenotype of R3230 AC rat adenocarcinoma

Human choriogonadotropin (hCG)-like material has been found in variable amounts on the surface of cells of human and animal tumors. Intravenous injection of R3230 AC rat adenocarcinoma cells, one of the models investigated, results in multiple lung foci seeding. We analyzed the phenotypic diversity of this tumor by cloning and culturing two distinct cell subpopulations from a cell culture of this tumor, hereafter called OR or original cell culture. One was obtained after repeated exposure of the OR to increasing concentrations of concanavalin A and wheat germ agglutinin. A single clone was isolated and was named lectin-resistant (LR) cell line. The LR cells did not metastasize but maintained stable tumorigenicity and morphology over at least 10 passages. A second cell line was obtained by repeated passage and injection of cells from a single metastatic node. After repeating the process five times, a single clone of cells was selected from the final variant and was called lung metastatic (LM) cell line. The LM cultured cells maintained stable tumorigenicity, morphology, and metastatic properties for no more than 10 passages. OR, LR, and LM cells were assessed by their doubling time (DT), chromosome counts, and hCG immunocytochemistry. The results demonstrated that the LM cell line had a higher chromosome count than the LR and the OR cell lines, and its DT was the shortest. Immunocytochemistry of the transplanted OR neoplasm showed scattered expression of the hCG-like material. By the same techniques a complete lack of reactivity of the LR cells was found. However, almost all cells of the LM line were strongly positive for hCG-like material. After a few passages, the great majority of the LM cells also became unreactive. Our data demonstrate: (i) the existence of marked heterogeneity of the expression of hCG-like material in the primary tumor cell population; (ii) that the expression of hCG-like material correlates with the metastasizing capacity of the cells; and (iii) that there is a phenotypic instability for the expression of hCG-like material by tumor cells when maintained in vitro.

Humoral response of normal and athymic (nude) mice to human choriogonadotropin immunogens

The production of antibodies against human choriogonadotropin (hCG) was studied in normal and athymic (nu/nu) mice of two strains (C57/BL and Balb/c), injected with native (whole) hCG or an immunogen consisting of a synthetic hCG beta COOH-terminal peptide, residues 109-145, conjugated to diphtheria toxoid and mixed with a synthetic muramyl dipeptide analog (nor-MDP) as adjuvant. Both the short-term effect of native hCG dissolved in saline and injected IM (primary response), and the long-term effect of the native hCG and of the hCG immunogen dissolved in saline, emulsified in squalene-Arlacel A, and injected SC as a depot injection (secondary or memory response), were considered. The results obtained indicate that native hCG may be classified as a T-cell independent antigen in the sense that it can elicit low levels of IgM antibodies on a short term basis in athymic mice that have either no or very low T-cell levels. In long-term studies using hCG and the hCG immunogen no antibodies could be detected in athymic mice 14 days after a booster inoculation given 28 days after primary immunization, a regimen that produced high levels of antibodies in normal mice. Because of their inability to sustain humoral responses to native hCG as well as to other hCG immunogens, athymic mice seem well suited for in vivo studies of some of the biological effects of hCG.

Antibodies to bacterial vaccines demonstrating specificity for human choriogonadotropin (hCG) and immunochemical detection of hCG-like factor in subcellular bacterial fractions

Investigations were done to determine whether vaccines prepared with chemically killed Staphylococcus haemolyticus RU1 and Streptococcus bovis AV46 (bacteria that have been demonstrated to express human choriogonadotropin [hCG]-like material on their surface) elicited antibodies in rabbits with specificity for hCG determinants. In addition, the anatomical locus of the hCG-like factor was determined by separation of bacterial subcellular fractions. The results demonstrated that these bacterial vaccines elicited antibodies immunologically similar to those antibodies produced in response to the whole human trophoblastic hormone, a similarity extending even to cross-reactivity with human luteinizing hormone. The bacterial hCG-like material appeared to be localized in the membranes of the cell wall, and most was present in the soluble membranous and cytoplasmic constituents. Its expression in bacteria was a strain characteristic and not a species characteristic.