Antitumor efficacy of adenocarcinoma cells engineered to produce interleukin 12 (IL-12) or other cytokines compared with exogenous IL-12

BACKGROUND: Numerous animal model studies have examined the ability of genetically engineered tumor cells to release cytokines and to elicit an immune memory against the parental tumor. Often only a single cytokine is studied, and few comparative studies have been conducted. PURPOSE: We evaluated the antitumor efficacy of adenocarcinoma cells engineered to release interleukin (IL)-12 in a mouse model system. The efficacy of this cytokine was compared with that of other cytokines released by engineered adenocarcinoma cells and that of exogenous IL-12 administered both locally and intraperitoneally. METHODS: BALB/cAnCr mice were inoculated with syngeneic parental mammary adenocarcinoma (TSA) cells in quantities sufficient to lead to tumors in all inoculated mice. TSA cells engineered to release IL-12 (TSA-IL12) were also injected into normal and selectively immunosuppressed BALB/cAnCr mice. Tumor incidence, growth, and rejection patterns were evaluated by the measurement of neoplastic masses and by the study of the histologic and ultrastructural features of the tumor site. The effects of local or intraperitoneal administration of recombinant IL-12 (rIL-12) on tumor-bearing animals were also studied. RESULTS: Most mice rejected TSA-IL12 cells through a CD8-positive, T-lymphocyte-dependent reaction associated with macrophage infiltration, vessel damage, and necrosis. The systemic immunity of mice that had rejected TSA-IL12 cells to a subsequent challenge with parental TSA cells was less efficient than that elicited by TSA cells engineered to release IL-4 or IL-10 but equivalent to that elicited by TSA cells engineered to release IL-2, IL-7, and interferon alfa. Compared with TSA cells engineered to produce other cytokines, TSA-IL12 cells were the most efficient in curing mice with established TSA tumors; injection of 0.1 million proliferating cells contralaterally to the tumor growth area cured five of 15 mice bearing 1-day-old tumors; injection of the same dose of proliferating cells into the tumor growth area cured two of 20 tumor-bearing mice. However, two 5-day courses with a nontoxic dose (0.1 microgram) of rIL-12 given intraperitoneally cured a similar proportion of these animals (six of 20). Only two of 20 mice with 7-day-old TSA tumors were cured by vaccination with proliferating TSA-IL12 cells, whereas 24 of 30 mice with such tumors were cured by intraperitoneal administration of rIL-12. CONCLUSIONS: TSA cells engineered to release IL-12 are rejected by most mice; the ensuing immune memory for TSA parental cells, however, was less efficient than that elicited by proliferating TSA cells engineered to release other cytokines (e.g., IL-4, IL-10, and possibly interferon gamma). The immune reaction elicited by TSA-IL12 cells was the most efficient in curing mice with established TSA tumors; notably though, the same or a better cure rate was obtained with rIL-12 given intraperitoneally.     

Regulation of local host-mediated anti-tumor mechanisms by cytokines: direct and indirect effects on leukocyte recruitment and angiogenesis

The regulation of tumor growth by cytokine-induced alterations in host effector cell recruitment and activation is intimately associated with leukocyte adhesion and angiogenic modulation. In the present study, we have developed a novel tumor model to investigate this complex series of events in response to cytokine administration. Gelatin sponges containing recombinant human basic fibroblast growth factor (rhFGFb) and B16F10 melanoma cells were implanted onto the serosal surface of the left lateral hepatic lobe in syngeneic C57BL/6 mice. The tumor model was characterized by progressive tumor growth initially localized within the sponge and the subsequent development of peritoneal carcinomatosis. Microscopic examination of the sponge matrix revealed well developed tumor-associated vascular structures and areas of endothelial cell activation as evidenced by leukocyte margination. Treatment of mice 3 days after sponge implantation with a therapeutic regimen consisting of pulse recombinant human interleukin-2 (rhIL-2) combined with recombinant murine interleukin-12 (rmIL-12) resulted in a marked hepatic mononuclear infiltrate and inhibition of tumor growth. In contrast to the control group, sponges from mice treated with rhIL-2/rmIL-12 demonstrated an overall lack of cellularity and vascular structure. The regimen of rhIL-2 in combination with rmIL-12 was equally effective against gelatin sponge implants of rhFGFb/B16F10 melanoma in SCID mice treated with anti-asialo-GM1 in the absence of a mononuclear infiltration, suggesting that T, B, and/or NK cells were not the principal mediators of the anti-tumor response in this tumor model. The absence of vascularity within the sponge after treatment suggests that a potential mechanism of rhIL-2/rmIL-12 anti-tumor activity is the inhibition of neovascular growth associated with the establishment of tumor lesions. This potential mechanism could be dissociated from the known activities of these two cytokines to induce the recruitment and activation of host effector cells. Moreover, this model provides a unique opportunity to study the cellular and molecular mechanism(s) underlying both tumor angiogenesis and leukocyte recruitment to metastatic lesions.     

Recruitment of hepatic NK cells by IL-12 is dependent on IFN-gamma and VCAM-1 and is rapidly down-regulated by a mechanism involving T cells and expression of Fas

NK cells have been shown to be important antitumor or antiviral effector cells in the liver. In the present study we have examined the factors that regulate the initial recruitment and subsequent fate of hepatic NK and T cells in mice treated with IL-12 or IL-2. Daily administration of IL-12 caused a rapid initial increase in NK cells followed by a subsequent decrease that coincided with an accumulation of T cells. The recruitment of hepatic NK cells by IL-12, but not the subsequent T cell infiltrate, was abrogated in IFN-gamma(-/-) mice. In contrast, daily administration of IL-2 caused a sustained increase in liver-associated NK cells that was not diminished in IFN-gamma(-/-) mice. The IL-12-induced recruitment in both hepatic NK and T cells was abrogated by in vivo treatment with anti-VCAM-1 mAbs, while treatment with anti-ICAM-1 Abs decreased only the recruitment of T cells in the IL-12-treated mice. The rapid loss of newly recruited hepatic NK cells in IL-12-treated mice did not occur in SCID mice or in B.MRL-Fas(lpr) (Fas-) and B6Smn.C3H-Fasl(gld) (FasL-) mutant mice, suggesting that T cells can actively eliminate hepatic NK cells through a Fas-dependent mechanism. These findings also imply that during the endogenous innate immune response to infectious agents or tumors or in the host response induced by cytokine therapies, the biologic effects of NK cells may be limited by T cell-mediated effects.     

Three dimensional finite element investigations into the effects of thickness and notch radius on the fracture toughness of polycarbonate

Fracture toughness of polycarbonate (PC), a commercially important glassy amorphous polymer, is known to be sensitively dependent on a number of factors including molecular weight, ageing time, loading rate and specimen geometry. In this work, we analyze the effect of notch radius and specimen thickness on the near tip fields and the consequence of these on the mode I fracture initiation. To this end, we have performed extensive three dimensional Finite Element simulations within the framework of large deformation elasto-plasticity based on a realistic constitutive model that has been carefully calibrated for PC. Using a simple set of criteria for fracture initiation by void nucleation or ductile tearing, we are able to reproduce experimentally observed brittle to ductile transitions that occur in PC with decrease in thickness and increase in notch radius.