BCG-activated Killer Cells Research Articles

In vitro generation of bacillus Calmette-Guérin-activated killer cells.

Tumor regression induced in cancer patients by local instillation of bacillus Calmette-Guérin (BCG) into the bladder is considered to be mediated by cellular immune and inflammatory reactions. In an attempt to elucidate which of these effects are relevant to tumoricidal activity, an in vitro system was employed in which the immunostimulatory effects of BCG could be studied. This report describes the induction of BCG-activated killer (BAK) cells, which effectively lyse bladder tumor cells. Human peripheral blood mononuclear cells (PBMC) were stimulated with viable and sonicated BCG (v-BCG and s-BCG, respectively) to generate BAK cells. Cytotoxicity of BAK cells was comparable with the cytotoxicity exerted by lymphokine-activated killer (LAK) cells generated by interferon (IFN)-gamma but did not reach the level of interleukin-2 (IL-2)-generated LAK cells. Induction of BAK cells was possible only with v-BCG and not with s-BCG. By depletion and enrichment of defined cell populations, the cytotoxic potential of BAK cells could be attributed to a population of CD8(+) and CD56(+) double-positive lymphocytes. Macrophages and CD4(+) cells were required for the induction of killing activity but had no such activity by themselves. Furthermore, the presence of IFN-gamma and IL-2 in the supernatants harvested during the generation of BAK cells was demonstrated. Monoclonal antibodies neutralizing these cytokines abolished BCG-mediated cytotoxicity. From these results, it is concluded that the known beneficial effect of local instillation of BCG on maintenance of the relapse-free state in superficial bladder cancer may be due to local generation of BAK cells.

BACILLUS-CALMETTE-GUERIN (BCG) AND 3D TUMORS: AN IN VITRO MODEL FOR THE STUDY OF ADHESION AND INVASION

Purpose To study adhesion, penetration and internalization of BCG and effector-cells to and into three-dimensional in vitro cell aggregates from benign and malignant urothelial origin mimicking small in vitro tumors. Materials and Methods Multicellular spheroids (MCS) were generated by “liquid-overlay” technique. Adhesion and penetration of viable FITC-labelled BCG into MCS from urothelial cancer cell lines and normal urothelial cells was studied by electron microscopy (TEM) and fluorescence microscopy. Spheroid growth during BCG-co-incubation was determined by light microscopy. Peripheral blood mononuclear cells (PBMC) were stimulated with BCG to generate BCG-activated-killer (BAK) cells. The infiltration of these effectors and of lymphokine-activated killer (LAK) cells into MCS was examined at different intervals by means of immunohistochemistry. The resulting cytotoxicity was judged in a 3 H-l-methionine release assay. Results BCG adhered to MCS from tumor cells but not to benign cell MCS. Intracellular internalization of the bacteria was detectable in superficial tumor cell-layers (1-5) whereas BCG was not found in deeper layers. Proliferation of malignant MCS was reduced in the presence of BCG. Benign MCS showed contact inhibition growth arrest, which was not altered by BCG. BAK and LAK effector cells both infiltrated tumor cell MCS as opposed to unstimulated PBMC. In contrast to LAK cells, BAK cells did not infiltrate into benign cell MCS and were not cytotoxic towards them. Conclusion With regard to the clinical situation the selective adhesion and internalization of BCG to malignant cells might explain why BCG has been rarely found in follow-up biopsies in tumor free patients. More interestingly, the selective adhesion of BCG to and infiltration of BAK effector cells into malignant cell spheroids suggests a selective mode of action of BCG.

Bacille Calmette-Guérin in superficial transitional cell carcinoma.

The mechanisms by which BCG exerts its antitumour activity remain unclear. Attachment of BCG to the bladder via FN has been shown to be an important step in initiating its antitumorigenic activity. The mechanism(s) by which BCG operates requires LAK cells, BCG-activated killer cells, T lymphocytes (CD4) helper cells and CD8 suppressor/cytotoxic cells) and monocytes. The optimal route of administration is intravesical. The efficacy of a BCG vaccine depends on the viability, dose and strain. Differences in efficacy and side-effects have not been shown between different strains. Low-dose regimens successfully protect from recurrences, with fewer side-effects. The initial schedule of BCG is a course of six instillations in 6 weeks; when the patient fails this course, two possibilities arise. The first is maintenance therapy; response rates improve but there is more local and systemic toxicity. The second is a further 6-week course, and this seems most useful in those with a sustained response to the initial treatment. The clinical response to BCG therapy can be monitored using cytokine measurements or p53 determinations. Toxicity remains a major problem in BCG treatment and triple antituberculosis combination therapy should be given for 3 months in those with severe systemic side-effects. The use of prophylactic isoniazid is not recommend to decrease side-effects. The clinical results of BCG have been good, with success rates of 58-100%, with a minimal follow-up of one year in prophylaxis. BCG seems superior to intravesical therapy, but at the cost of inducing more adverse effects. BCG is not indicated for low- and intermediate-risk patients, in whom chemotherapy is the first choice. BCG can also be used to eliminate tumour after an incomplete TUR, or in patients who are unfit for surgery, with a 60-70% success rate. The primary and best treatment for CIS is intravesical BCG; encouraging results have been reported, with success rate of 42-83% after a minimal follow-up of one year. Although currently BCG seems to be the choice for high-risk superficial TCC, many questions remain unanswered, especially about the mechanism(s) of action, the optimal dose and clinical schedule.

The induction of bacillus-Calmette-Guérin-activated killer cells requires the presence of monocytes and T-helper type-1 cells.

Previously we have described the induction of MHC-unrestricted killer cells against bladder tumour cells by bacillus Calmette-Guérin (BCG), termed BCG-activated killer (BAK) cells. In the present paper we deal with the accessory-cell requirement for the activation of BAK cells. We show that monocytes are required for activating BAK cells, since no cytotoxicity can be induced in the absence of monocytes. Therefore, these phagocytes may represent the first step during the activation cascade of BAK cells. Furthermore, the presence of CD4+ T cells was essential for generating BAK cells: depleting peripheral blood mononuclear cells of CD4+ cells prior to stimulation with BCG abolished the cytotoxicity against bladder tumour cells. Experiments with monoclonal antibodies (mAb) neutralizing the activity of either interleukin-2 (IL-2) or interferon gamma (IFN gamma) underlined the importance of these cytokines: both mAb blocked the induction of BAK cells. Since both cytokines are related to the so-called Th1 pattern of T cells, we consider the second step of the generation of BAK cells as follows: monocytes presenting antigens of BCG trigger Th1-like cells in a preferred manner. These Th1-like T cells secrete IL-2 and IFN gamma and, thus, activate the BAK effector cells. Since CD4+ cells are dominant in the cells infiltrating the bladder wall after intravesical instillation of BCG in vivo, we postulate an important role for the Th1 subpopulation. We further postulate that the occurrence of macrophages in this infiltrate seems to be significant in the maintenance of the relapse-free state of the patient.

On the mode of action of intravesical bacillus Calmette-Guérin: in vitro characterization of BCG-activated killer cells.

Previously we had shown that, upon activation with viable bacillus Calmette-Guérin (BCG), peripheral blood mononuclear cells (PBMNC) could be rendered cytotoxic against otherwise insensitive natural killer (NK)-resistant bladder cancer cell lines. This phenomenon had been termed the BCG-activated killer (BAK) cell phenomenon. By means of depletion and enrichment procedures of mononuclear cell subpopulations derived from BCG-activated PBMNC we further characterized the cytolytic BAK effector cells functionally in an in vitro cytotoxicity assay against the bladder carcinoma cell line BT-A and phenotypically in their pathway of activation. Neither macrophages nor CD4+ T-helper/inducer cells exerted cytotoxic BAK activity. This cytotoxicity was restricted to the CD8+CD56+ subpopulation of T-cytotoxic/NK cells. Furthermore, activation of BAK cells via interferon gamma (IFN-gamma) was evidenced by the complete inhibition of BAK cell generation with an IFN-gamma antibody.

Dissecting the Immunobiological Effects of Bacillus Calmette-Guérin (BCG) In Vitro: Evidence of a Distinct BCG-Activated Killer (BAK) Cell Phenomenon

Several immunobiological effects of intravesical bacillus Calmette-Guérin (BCG) during immunotherapy of superficial bladder cancer have been suggested as possible mediators of the mode of action. In an attempt to elucidate which of these effects is relevant to tumoricidal activity, an in vitro cytotoxicity assay was employed in which the direct effects of BCG and of cytokines against four transitional carcinoma cell lines were studied. Furthermore, peripheral blood mononuclear cells (PBMNC) were analyzed for their cytotoxic potential against these target cells. We found that none of the cytokines interleukin-1, interleukin-2, interferon-gamma, tumor necrosis factor alpha, lymphotoxin, or BCG alone were cytotoxic against the bladder carcinoma cell lines. However, a pronounced cytotoxicity against these targets resistant to natural killer cells could be induced in PBMNC by coincubation with viable BCG. We termed this the BCG-activated killer (BAK) cell phenomenon. In contrast to lymphokine-activated (LAK) cells, these BAK cells needed prolonged activation for 7 days and did not enhance the cytotoxicity against K562 target cells sensitive to natural killer cells. Nonviable, heat-inactivated BCG was significantly less effective, and sonificated fractions of BCG were not effective in stimulating PBMNC towards BAK cell activity. In vitro dissection of effects observed during BCG intravesical therapy may give more insight into the mode of action of BCG and may help to separate primary tumoricidal effector mechanisms from secondary concomitant phenomena. Further characterization of the BAK cell may result in an improvement of intravesical BCG immunotherapy.

Induction of bacillus-Calmette-Guérin-activated killer cells from human peripheral blood mononuclear cells against human bladder carcinoma cell lines in vitro.

Cytotoxicity against two human bladder carcinoma cell lines (BT-A and BT-B) was investigated using human peripheral blood mononuclear cells (PBMC) stimulated with viable bacillus Calmette-Guérin (BCG) or sonicated BCG (s-BCG). We applied a cytotoxicity assay based on radioactive labelling of tumour cells by incorporation of L[3H]methionine. The results were compared with the cytotoxicity exerted by lymphokine-activated killer (LAK) cells generated by interleukin-2 (IL-2) and interferon gamma (IFN gamma). BCG-stimulated PBMC showed a cytotoxic potential against BT-A and BT-B comparable to that of IFN gamma-generated LAK cells, but this did not reach the level of IL-2-generated LAK cells. We termed these cytotoxic effectors BCG-activated killer (BAK) cells. In contrast to their cytotoxicity against bladder tumour cells, BAK cells did not differ from unstimulated PBMC in the killing of K562 cells. Only viable but not sonicated BCG was able to induce cytotoxicity against BT-A and BT-B. We could demonstrate the presence of the cytokines IFN gamma, IL-2, tumour necrosis factor alpha (TNF alpha) and TNF beta in the supernatants harvested during the generation of BAK cells. Monoclonal antibodies neutralizing IFN gamma were able to inhibit BCG-mediated cytotoxicity, giving evidence of the involvement of IFN gamma in the induction of BAK cells. Furthermore, we performed experiments to investigate the cytotoxic potential of distinct cell populations. The cells effective in BCG-activated killing of bladder tumour cells could be localized within the CD8+/CD56+ lymphocyte subset. CD4+ cells and macrophages did not exhibit cytolytic activity. Our findings imply that the activation by BCG of CD8+/CD56+ killer cells might be an important antitumoral mechanism during BCG therapy against superficial urothelial bladder cancer.