Cow's milk-free diet does not prevent diabetes in NOD mice.

Early dietary exposure to cow's milk proteins has been proposed as an important environmental factor in the development of IDDM both in humans and in diabetes-prone rodents. To examine the significance of cow's milk protein in IDDM, 120 NOD mice were maintained, starting from conception until sacrifice, on one of four diets: standard PMI Picolab Rodent Diet 20, a milk-free modification of the standard Picolab diet, a milk-free diet incorporating 0.036% bovine serum albumin (BSA), and a milk-free diet including 0.036% bovine IgG (BGG). The cumulative IDDM incidence at 7 months for these mice in a specific pathogen-free environment on the respective diets was 78, 93, 93, and 67% for females, and 17, 54, 17, and 0% for males. The ages of diabetes onset and insulitis scores were similar for mice on each diet. The unexpectedly lower incidence of IDDM in mice on the milk-free diet that included BGG raises the possibility this cow's milk protein might possibly have some protective effect against the development of IDDM in NOD mice. Our main finding was that the standard, milk-free, and BSA-containing diets resulted in comparable incidences of IDDM in NOD mice, demonstrating that neither cow's milk whey proteins in general nor BSA in particular are significantly important as etiologic dietary agents in IDDM in NOD mice.

In vitro and in vivo kinetics of recombinant vaccinia virus cancer-gene therapy.

BACKGROUND: Gene therapy of cancer is a promising therapeutic modality. Recombinant vaccinia viruses (RecVV), engineered to produce cytokines, may be effective in this area. This study's purpose was to investigate the kinetics of RecVV infection, measuring protein production and in vivo viral growth pattern. METHODS: RecVV were constructed by homologous recombination, encoding murine interleukin-2 (mIL-2). After tumor cell infection, mIL-2 production was measured in vitro. Tumor-bearing and naive hosts were inoculated with RecVV and wild type vaccinia. Livers, spleens, and (where applicable) tumors were sequentially harvested, and tissue viral levels were measured. RESULTS: Infected tumor cells made high levels of mIL-2 after infection with RecVV encoding for this cytokine. Naive mice were able to clear recombinant but not wild type VV from their livers and spleens by days 3 and 5, respectively. Tumor-bearing animals had persistent RecVV titers in the tumor tissue at day 8. CONCLUSIONS: RecVV can infect tumor cells, causing the production of a large amount of foreign protein but are attenuated relative to wild type virus in the murine host.