Effect of insulin on ultrastructure and glycogenesis in primary cultures of adult rat hepatocytes

Insulin in the presence of high concentrations of glucose has a beneficial trophic effect on the development of primary cultures of hepatocytes. Compared to the situation observed in hormone-free control cultures, the flattening of the reaggregated hepatocytes is enhanced, and the reconstituted cell trabeculae are enlarged and tend to form a confluent monolayer after 3 days; the survival time is prolonged from 3 to 5 or 6 days. Ultrastructural modifications are also initiated by insulin; numerous glycogen particles appear after 24 h, in between the cisternae of the proliferated smooth endoplasmic reticulum. After 48 h, large amounts of glycogen are stored, and numerous polysomes are present. A small number of cells showed an increased synthesis of lipid droplets in the lumen of the smooth endoplasmic reticulum and form liposomes at the same time. After 72 h, cytolysomes filled with glycogen develop, simulating glycogenosis type II. Simultaneously, microtubules and microfilaments, closely related to numerous polysomes, appear in cytoplasmic extensions constituting undulating membranes. The biochemical data demonstrate that, in the absence of insulin, a high concentration of glucose stimulates glycogenesis and hinders glycogenolysis. This effect of glucose on polysaccharide synthesis is progressively lost. The addition of insulin to the culture induces after 48 and 72 h, a three- to fivefold increase of the glucose incorporation into glycogen, as compared to the controls. The presence of insulin is required to maintain the hepatocyte's capacity to store glycogen. Glycogen synthetase is converted into its active form under the influence of glucose. Insulin increases the rate of activation.     

Tumorigenicity of herpesvirus-transformed cells correlates with production of plasminogen activator.

Our studies first demonstrated that established hamster cell lines transformed in vitro by herpesviruses activate plasminogen more effectively than normal hamster fibroblasts. This ability is probably due to increased levels of the enzyme plasminogen activator (PA). In the studies described here, the 333-8-9 cell line, originally transformed by herpes simplex virus type 2 strain 333, was used to derive subclonal lines that maintained stable PA phenotypes over the course of long in vitro passage. We were interested in correlating tumor formation by the subclones with their fibrinolytic capacity. Cells were, therefore, single-cell subcloned twice, and resulting cultures were tested for ability to activate plasminogen in vitro. PA activity was then quantitated by [125I]fibrin lysis assay, and high- and low-activity subclones were isolated; these retained high- or low-activity phenotypes. Syngeneic newborn hamsters were inoculated with these subclones and observed for the appearance of palpable tumors. A strong correlation between enzyme activity and tumor formation was observed in four separate trials; animals receiving high-PA subclones developed tumors more rapidly than those inoculated with the parental cell line. Tumors were also excised from test animals, and the cell lines established from the tumors were tested in vitro at different passages for their ability to activate plasminogen. These tumor cells were then reinoculated into syngeneic animals to confirm the tumorigenicity of cell lines with high fibrinolytic activity. In these experiments, the positive correlation between PA production and tumorigenicity was confirmed.     

Mutant forms of cytochrome P-450 controlling both 18- and 11beta-steroid hydroxylation in the rat.

A reciprocal relationship between steroid 18- and 11beta-hydroxylase activities in the salt susceptible (S) and the salt resistant (R) strains of rats was previously shown to be controlled by a single genetic locus with two alleles and inheritance by co-dominance (Rapp, J. P., and Dahl, L. K. (1972), Endocrinology 90, 1435). The strain specific steroidogenic patterns, characterized by the relative magnitudes of 18- and 11beta-hydroxylase activities, were found to be determined by adrenal mitochondrial cytochrome P-450 particles. Carbon monoxide inhibition of 18- and 11beta-hydroxylation of deoxycorticosterone in these strains showed that the CO/O2 ratio causing 50% inhibition (i.e., Warburg's partition constant, K) was identical for 18- and 11beta-hydroxylation within a strain, but different for both 18- and 11 beta hydroxylation between strains. (K values were: S rats, 18-hydroxylation = 11.4 +/- 1.4; S rats, 11beta-hydroxylation = 11.0 +/- 1.2; R rats, 18-hydroxylation = 56.4 +/- 13.7; R rats, 11beta-hydroxylation = 46.7 +/- 11.7). This between-strain difference was unique for 18- and 11beta-hydroxylation; i.e., it was not seen with cholesterol side-chain cleavage or 21-hydroxylation. Moreover, the strain-specific K values for 18- and 11beta-hydroxylase and the strain-specific steroidogenic patterns due to the relative magnitudes of 18- and 11beta-hydroxylase activities segregated together in an F2 population. These data strongly suggest the same cytochrome P-450 is involved in both 18- and 11beta-hydroxylation and that this cytochrome is mutated between S and R rats. K values for the reaction corticosterone leads to 18-hydroxycorticosterone were different between S and R strains, indicating that the mutant cytochrome was also involved in this hydroxylation, but K values for the conversion corticosterone leads to aldosterone were not different between strains. This was interpreted to mean that each step in the sequence corticosterone leads to 18-hydroxycorticosterone leads to aldosterone was mediated by a different cytochrome, the K value for the second step being the lower and dominating the overall reaction. It was speculated that the second step could be a second hydroxylation at position 18 to yield 18,18-dihydroxycorticosterone which could be unstable and decompose into aldosterone and water.     

Analysis of biochemical phase shift oscillators by a harmonic balancing technique

The use of harmonic balancing techniques for theoretically investigating a large class of biochemical phase shift oscillators is outlined and the accuracy of this approximate technique for large dimension nonlinear chemical systems is considered. It is concluded that for the equations under study these techniques can be successfully employed to both find periodic solutions and to indicate those cases which can not oscillate. The technique is a general one and it is possible to state a step by step procedure for its application. It has a substantial advantage in producing results which are immediately valid for arbitrary dimension. As the accuracy of the method increases with dimension, it complements classical small dimension methods. The results obtained by harmonic balancing analysis are compared with those obtained by studying the local stability properties of the singular points of the differential equation. A general theorem is derived which identifies those special cases where the results of first order harmonic balancing are identical to those of local stability analysis, and a necessary condition for this equivalence is derived. As a concrete example, the n-dimensional Goodwin oscillator is considered where p, the Hill coefficient of the feedback metabolite, is equal to three and four. It is shown that for p = 3 or 4 and n less than or equal to 4 the approximation indicates that it is impossible to construct a set of physically permissible reaction constants such that the system possesses a periodic solution. However for n greater than or equal to 5 it is always possible to find a large domain in the reaction constant space giving stable oscillations. A means of constructing such a parameter set is given. The results obtained here are compared with previously derived results for p = 1 and p = 2.     

Immunologic approaches to the treatment of human cancer based on a guinea pig model

Summary Local immunotherapy is a form of cancer treatment where exogenous antigen is introduced into the area of the tumor. Under favorable circumstances, the perfused tumor regresses, systemic tumor-specific transplantation immunity is augmented, and distant microscopic metastases regress. Successful local immunotherapy requires an immunologically competent host, small tumor burden, and tumor located usually in the skin. A wide variety of biologic agents are capable of promoting local immunotherapy. BCG has been most widely studied. The antitumor activity of two different preparations of the Tice substrain of BCG were compared. No significant differences in antitumor activity were found. Alternative approaches to intralesional injection were sought. Intradermal injection of BCG adjacent to dermal tumors, prior to surgery, led to eradication of axillary metastases and to the development of tumor-specific transplantation immunity.Successful local BCG immunotherapy is a by-product of the host response to BCG infection. Involved are lymphocytes specifically sensitized to mycobacterial antigens, lymphocyte mediators, and macrophages which develop the capacity to kill tumor cells. Tumor-cell killing may be mediated by exocytosis of macrophage lysosomes into tumor cells. Complete and permanent tumor eradication probably requires the development of tumor-specific transplantation immunity mediated by sensitized lymphocytes. Local infection of the tumor may augment the development of this tumor-specific immunity.     

Induction of murine p30 by superinfecting herpesviruses.

The interaction of endogenous type C viruses with superinfecting herpes simplex virus type 2 (HSV-2) was investigated in two murine cell lines. Replication of HSV-2 was suboptimal in random-bred Swiss/3T3A cells and, in initial experiments, infection with a low virus-to-cell ratio resulted in carrier cultures with enhanced murine leukemia virus (MuLV) p30 expression. Immunofluorescence tests with Swiss/3T3A cells productively infected with HSV-2 also showed HSV-associated cytoplasmic antigens and enhanced MuLV p30 expression when compared with uninfected controls. Inactivation of HSV-2 with UV light did not abolish this reaction, although the number of cells expressing p30 was reduced. HSV-2 replicated more efficiently in a line of NIH Swiss cells (N c1 A c1 10). These cells are not readily inducible for type C expression by conventional methods; however, untreated and UV-inactivated HSV-2 induced both HSV-2-associated antigens and MuLV p30 in these cells. Although the Birch strain of human cytomegalovirus induced MuLV p30, neither mouse cytomegalovirus nor vesicular stomatitis virus induced MuLV p30 in either cell line.     

Distinguishing cytomegalovirus, mycoplasma, and cellular DNA polymerases.

Cytomegalovirus-induced DNA polymerase can be distinguished from infected-cell enzymes by activity in 100 mM (NH4)2SO4. Virus polymerase is stimulated to 145% of control, whereas mock-infected cell polymerase is inhibited to 12% of control without added salt. Mycoplasmas induce a DNA polymerase in cell extracts that is stimulated to 130 to 180% by 25 mM (NH4)2SO4. Mycoplasma DNA polymerase may be mistaken for a virus-induced polymerase when virus stocks are contaminated. Identification of virus, cellular, and mycoplasma DNA polymerases in total cell extracts is described using sedimentation rate and effect of inhibitors on DNA polymerase activities.