Danuta J. Herzyk

Immunomodulatory Effects of Anti-CD4 Antibody in Host Resistance against Infections and Tumors in Human CD4 Transgenic Mice

ABSTRACT Anti-CD4 antibodies, which cause CD4+ T-cell depletion, have been shown to increase susceptibility to infections in mice. Thus, development of anti-CD4 antibodies for clinical use raises potential concerns about suppression of host defense mechanisms against pathogens and tumors. The anti-human CD4 antibody keliximab, which binds only human and chimpanzee CD4, has been evaluated in host defense models using murine CD4 knockout-human CD4 transgenic (HuCD4/Tg) mice. In these mice, depletion of CD4+ T cells by keliximab was associated with inhibition of anti-Pneumocystis carinii and anti-Candida albicans antibody responses and rendered HuCD4/Tg mice susceptible to P. carinii, a CD4-dependent pathogen, but did not compromise host defense against C. albicans infection. Treatment of HuCD4/Tg mice with corticosteroids impaired host immune responses and decreased survival for both infections. Resistance to experimental B16 melanoma metastases was not affected by treatment with keliximab, in contrast to an increase in tumor colonization caused by anti-T cell Thy1.2 and anti-asialo GM-1 antibodies. These data suggest an immunomodulatory rather than an overt immunosuppressive activity of keliximab. This was further demonstrated by the differential effect of keliximab on type 1 and type 2 cytokine expression in splenocytes stimulated ex vivo. Keliximab caused an initial up-regulation of interleukin-2 (IL-2) and gamma interferon, followed by transient down-regulation of IL-4 and IL-10. Taken together, the effects of keliximab in HuCD4/Tg mice suggest that in addition to depleting circulating CD4+ T lymphocytes, keliximab has the capability of modulating the function of the remaining cells without causing general immunosuppression. Therefore, keliximab therapy may be beneficial in controlling certain autoimmune diseases.

High-dose recombinant interleukin-18 induces an effective Th1 immune response to murine MOPC-315 plasmacytoma.

Interleukin (IL)-18 has profound antitumor activity when administered at high doses as a single agent for prolonged periods in BALB/c mice bearing late, well-established MOPC-315 tumors. Management with a qD x 27 schedule resulted in regression of tumors in all animals receiving 5 mg/kg/d. A protracted daily management regimen appears to be necessary to induce regression in this advanced tumor model. Biologic markers were assessed and appear to be potentially useful in evaluating the immunologic and antitumor activity of IL-18. The biomarkers of IL-18s immunologic activity include, but are not limited to, IL-1&agr;, IL-2, IL-8, IL-10, IL-12, IL-13, interferon-&ggr;, tumor necrosis factor-&agr;, and granulocyte-macrophage colony-stimulating factor. The profile of these circulating cytokines and their expression levels at baseline, and after IL-18 delivery, can be measured in the serum, as well as from splenocytes of mice or human peripheral blood mononuclear cells derived from either normal subjects or patients with cancer. We compared IL-18 and IL-12 alone or in combination for their ability to induce cytokine production and natural killer cytolytic activity. Our data support the notion that IL-18 induces a predominantly Th1 response, and that the mechanism of IL-18 activity differs from that of IL-12. The biologic activity of IL-18 management revealed by increases in serum levels of cytokines and enhancement of natural killer cytolytic activity will be useful as clinical trials initiate in 2002. Expression of interferon-&ggr; and granulocyte-macrophage colony-stimulating factor serum levels correlates directly over a broad dose escalation with the level of IL-18. Therefore, this provides a convenient pharmacodynamic reference to the biologic response to IL-18 that may serve to guide the conduct of clinical trials.

Single-Organism Model of Host Defense against Infection: A Novel Immunotoxicologic Approach to Evaluate Immunomodulatory Drugs

The immunotoxicologic effects of drugs on host defense have been studied widely using various animal models of infection. Here we describe a new approach to testing host defense by using a single organism (Candida albicans) in CBA/J mice. The model is configured to test 3 effector systems via different routes of inoculation to stimulate different effector arms of the immune response. Nonspecific immunity was evaluated by C. albicans colony-forming unit (CFU) count from the spleen at 2 hr (uptake) and ≥22 hr (clearance) following intravenous inoculation. Cell-mediated immunity was assessed by CFU count from an intramuscular injection site 6 days postinoculation. Humoral immunity was assessed by anti- Candida antibody titer, following multiple subcutaneous immunizations with C. albicans. Finally, overall immunity was evaluated following intravenous injection using survival as the endpoint. Histopathological, immunohistochemical, and electron microscopic evaluation of selected tissues revealed the involvement of the expected cell types in the different effector systems. Several immunomodulatory drugs—dexamethasone, cyclosporine, liposomal muramyltripeptide phosphatidylethanolamine, and SK&F 105685—were evaluated in the C. albicans model. Dexamethasone impaired host defense against C. albicans by suppressing all endpoints measured. Similarly, cyclosporine showed broad immunosuppressive activity, with the exception of yeast uptake from the spleen. In contrast, muramyl tripeptide-phosphatidylethanolamine enhanced all but cell-mediated immunity to C. albicans. SK&F 105685 displayed both stimulatory and inhibitory effects on immune responses to the infection. Our studies demonstrate that a single organism-based approach can be a useful method for evaluating the immunological hazards of drugs on host resistance to infection.

Preclinical Safety of Recombinant Human Interleukin-18

Recombinant human interleukin-18 (rHuIL-18) is currently in clinical trials for treatment of cancer. This report presents results of preclinical toxicity studies with rHuIL-18 in cynomolgus monkeys and recombinant murine IL-18 (rMuIL-18) in mice. The rHuIL-18 was administered intravenously in 1 or 2 different 5-day cycles at doses 0.3 to 75 mg/kg/day in monkeys. Decreases in red cell mass, neutrophil, and platelet counts, increases in monocyte and large unstained cell counts, and lymphoid hyperplasia in spleen and lymph nodes were mild, reversible, and likely related to the pharmacologic activity of IL-18. The only toxic effect was protein cast nephropathy, secondary to coprecipitation of administered IL-18 and Tamm-Horsfall protein in the distal nephron, that only occurred at 75 mg/kg/day. Other adverse effects of rHuIL-18 were related to strong immunogenicity in monkeys and were manifest only during a second dosing cycle. The rMuIL-18, at similar dosing levels and cycles in mice, resulted in reduced red cell mass, increased white blood cell counts, spleen and lymph node hyperplasia, and mild, reversible changes in intestine, liver, and lungs. Protein cast nephropathy occurred in mice at doses ≥30 mg/kg/day. In conclusion, preclinical safety studies showed that rIL-18 was well tolerated at pharmacologically active doses in both monkeys and mice.

Flow cytometry in the preclinical development of biopharmaceuticals.

Novel biomarkers are often required in the preclinical development of biopharmaceuticals in order to characterize pharmacologic and toxicologic effects and to establish pharmacodynamic and pharmacokinetic relationships. Flow cytometry is uniquely suited for measurement of these biomarkers. Large numbers of single cells in a heterogeneous population can be rapidly identified and characterized with high accuracy and reproducibility. Cells are not damaged by the detection system and can be subsequently sorted for further morphologic or functional analysis. The availability of clinical instruments and a wide range of fluorescent probes have made this technology applicable for use in toxicologic clinical pathology. Flow cytometry has played an integral role in the development of a monoclonal antibody to human CD4 (keliximab, IDEC-CE9.1, SB 210396). Lymphocyte subset analysis and assays for expression, coating, and modulation of human CD4 were used for sequential assessment of the pharmacologic activity of keliximab in transgenic mice expressing human CD4.

Effect of TNFα production inhibitors BRL 61063 and pentoxifylline on the response of rats to poly I:C

BRL 61063 is a novel xanthine phosphodiesterase (PDE) type IV inhibitor with selective inhibitory activity for tumor necrosis factor (TNF) alpha production. This compound inhibits TNF alpha production by activated human blood monocytes in vitro and in animal models of endotoxemia and influenza infection. Inhibition of TNF alpha may be beneficial in many diseases; however, little is known about potential adverse effects of such inhibition on host defense. In an ex vivo study, we examined the effect of BRL 61,063 on the microbicidal and tumoricidal activity of pulmonary lavage cells during a local inflammatory response in rats challenged with Poly I:C. Pentoxifylline, a PDE inhibitor which also blocks TNF alpha production, was used for comparison. Treatment with BRL 61063 or pentoxifylline did not block the inflammatory response to Poly I:C or the activation of bronchoalveolar lavage (BAL) cells but reduced the level of tumoricidal activity attained. At the dosages used, pentoxifylline was more inhibitory than BRL 61063. Drug treatment did not prevent further stimulation of tumoricidal activity by LPS in vitro. LPS-stimulated cells from BRL 61063-treated rats reached a level of activation similar to the control group while the LPS-stimulated activity of BAL cells from pentoxifylline treated rats remained lower than control. Although pentoxifylline was more inhibitory for tumoricidal activity than BRL 61063, the latter was a more potent inhibitor of TNF alpha release as measured in vivo in LPS-challenged rats. This finding indicates that TNF alpha is not the main mediator involved in the activation of pulmonary macrophage tumoricidal function. Treatment with either BRL 61063 or pentoxifylline had little or no effect on the Poly I:C-induced candidacidal activity of BAL cells indicating that these compounds are unlikely to compromise non-specific host defense against infection.

Atiprimod (SKandF 106615), a novel macrophage targeting agent, enhances alveolar macrophage candidacidal activity and is not immunosuppresive in candida-infected mice

Azaspiranes are novel macrophage-targeting agents with activity in preclinical animal models of autoimmune disease and transplantation. The purpose of this work was to determine the effects of atiprimod (SK&F 106615), an azaspirane being developed for the treatment of rheumatoid arthritis, on rat pulmonary alveolar macrophage (AM) function and immunocompetance in Candida-infected mice. AM from rats treated with 20 mg/kg/day of atiprimod for 15 days demonstrated enhanced killing of Candida albicans ex vivo. Concentration-dependent increases in candidacidal activity were also observed as early as one hour after exposure in vitro in AM from untreated normal rats. Treatment of AM with atiprimod in vitro did not increase particulate-stimulated superoxide production or phagocytosis of Candida but decreased their ability to concentrate acridine orange, indicating an increase in lysosomal pH. Increased candidacidal activity was inhibited by superoxide dismutase and catalase, suggesting a role for reactive oxygen intermediates (ROI). Atiprimod also increased free radical-mediated killing of Candida in the presence of H2O2, iron and iodide in a cell-free system. These findings indicated that treatment with atiprimod increased the candidacidal activity of rat AM in a free radical-dependent manner. The data also suggested that atiprimod did not increase ROI production by AM, but rather increased the efficiency of radical-mediated killing. This increase may be caused by cyclization of atiprimod, facilitating electron transfer and peroxidation of lipid membranes. In vivo studies in Candida-infected CBA mice showed that atiprimod (10 mg/kg/day), did not compromise immune function in the infected mice and could be differentiated from prototypical immunosuppressive compounds used for treatment of autoimmune diseases.

Relationships between antibodies against human soluble complement receptor 1 (hsCR1) from various species

The relationships between antibodies against human soluble complement receptor 1 (hsCR1) were studied in rodents, dogs, nonhuman primates, and humans. An antibody response occurred in all species except humans. The anti-hsCR1 antibodies from the various species were characterized to determine if they recognize similar epitopes on the hsCR1 molecule. Dog and monkey sera, positive for hsCR1 binding, were used as blocking antibodies against mouse anti-hsCR1 monoclonal antibodies as well as mouse and rat anti-hsCR1-positive sera. Human sera (blood group antisera: anti-Knops, anti-McCoy, anti-Knops/McCoy, anti-Swain-Langley) and serum from one burn patient (who became seropositive despite ever receiving treatment with hsCR1) were also used to test blocking of mouse, rat, dog, and monkey anti-hsCR1. Characterization of anti-hsCR1 antibodies from different species demonstrated that hsCR1 causes divergent antibody responses among animals. While mouse, rat, and dog antibodies cross inhibit binding by approximately 50%, monkey antibodies recognize primarily different epitopes of the hsCR1 molecule. Moreover, human antibodies binding hsCR1 are completely different from the animal antibodies, including monkey. This study indicates that although hsCR1 is immunogenic in animals, there is a difference in response between species, particularly between nonprimates and primates, and finally, that this antibody response is not predictive for humans.