Physiological concentrations of insulin support the in vitro growth of LB T-cell lymphoma. We could not detect similar insulin dependence in other tumor cell lines. This study reports that insulin also enhances the growth of LB cells in vivo. Mice treated with Streptozotocin (SZ) developed partial resistance to LB lymphoma growth and they survived longer (p < 0.0025) than non-diabetic mice after LB-cell inoculation. A few diabetic mice developed complete tumor resistance, manifested by total regression of the lymphoma. SZ-treated diabetic mice reconstituted with external insulin died as fast as non-diabetic mice when both were inoculated with the same number of LB cells. The SZ-treated diabetic mice did not develop resistance to the growth of BCLI B-cell leukemia, which demonstrated only a marginal proliferative response to insulin in vitro. Mice fed a low-energy diet exhibited low insulin levels and also developed resistance to lymphoma growth (50% survival 21 days vs. 15 days; p < 0.0005), supporting the concept that insulin enhances LB T-cell tumor development in mice.

A flow cytometry method has been employed to quantitatively compare the expression of A, B and H antigens on various red blood cells (RBC). The H substance was directly labelled by fluorescein-conjugated anti-H lectin and the A and B antigens by indirect staining first with monoclonal anti-A or anti-B antibodies followed by fluorescently, fluorescein (FITC) or phycoerythrin (PE), labelled anti-mouse immunoglobulin (Ig) antibodies. More than a ten-fold difference in cellular fluorescence intensity was found within each sample. Both the percentage and the mean fluorescence of the positive subpopulation for each antigen were determined. Each RBC population was characterized with respect to the expression of A, B or H antigen by a compound mean value that was the calculated product of these two parameters. The results demonstrated a reciprocal relationship between the compound means of A or B and H. The ratio of A/H or B/H was found to be most informative. Homozygotes for A or B had ratios of greater than 200 and greater than 30, respectively, while heterozygotes (AO or BO) had ratios of less than 5. This method could also distinguish between A1 and A2; RBC carrying the A1 phenotype (as determined by agglutination with anti-A1 lectin) showed a higher A/H ratio than those carrying A2. In contrast to the reciprocity in the expression of A (or B) and H found in RBC obtained from different individuals, a direct correlation was found in the expression of these antigens by individual cells within a given population.(ABSTRACT TRUNCATED AT 250 WORDS)