Incorporation of biological information in cancer risk assessment: Example — Vinyl chloride

Vinyl chloride (VC) is used as an example to demonstrate how biological information can be incorporated into quantitative risk assessment. The information included is the pharmacokinetics of VC in animals and humans and the data-generated hypothesis that VC primarily affects the initiation stage of the multistage carcinogenesis. The emphasis in this paper is on the improvement of risk assessment methodology rather than the risk assessment of VC per se.Sufficient data are available to construct physiologically-based pharmacokinetic models for both animals and humans. These models are used to calculate the metabolized dose corresponding to exposure scenarios in animals and in humans.On the basis of the data on liver angiosarcomas and carcinomas in rats, the cancer risk per unit of metabolized dose is comparable, irrespective of routes (oral or inhalation) of exposure. The tumor response from an intermittent/partial lifetime exposure is shown to be consistent with that from a lifetime exposure when VC is assumed to affect the first (initiation) stage of the multistage carcinogenic process. Furthermore, the risk estimates calculated on the basis of animal data are shown to be consistent with the human experience.     

Perfusion method for preparing pig brain cortex for Golgi-Cox impregnation

The perfusion procedure described in this paper produces high quality impregnation of pig visual and somatosensory cortical neurons with a Golgi-Cox solution. Starting within 30 min after death, pig heads were perfused with a fixative solution composed of a mixture (v/v) of liquid phenol, 5%; formalin, 14%; ethylene glycol, 25%; methanol, 28%; and water, 28% for two periods of 4 hr each. After perfusion, the heads were chilled for at least 18 hr. The entire brain was removed from the skull and then placed in 10% buffered formalin, where it remained for at least 10 days before taking the blocks that were to be immersed in the Golgi-Cox solution. Three weeks spent in the Golgi-Cox solution typically produced uniform neuron impregnation. The tissue blocks were then embedded in celloidin and sectioned at 120 micron. This procedure avoids the following difficulties: Golgi-Cox methods that produced excellent results with rodent or primate tissue were unsuccessful with pig tissue, placing fresh tissue in Golgi-Cox solution resulted in incomplete neuron impregnation, and immersion fixation in 10% buffered formalin without perfusion resulted in excessive staining of glia.     

Somatostatin receptor interacting protein defines a novel family of multidomain proteins present in human and rodent brain.

By using the yeast two-hybrid system we identified a novel protein from the human brain interacting with the C terminus of somatostatin receptor subtype 2. This protein termed somatostatin receptor interacting protein is characterized by a novel domain structure, consisting of six N-terminal ankyrin repeats followed by SH3 and PDZ domains, several proline-rich regions, and a C-terminal sterile alpha motif. It consists of 2185 amino acid residues encoded by a 9-kilobase pair mRNA; several splice variants have been detected in human and rat cDNA libraries. Sequence comparison suggests that the novel multidomain protein, together with cortactin-binding protein, forms a family of cytoskeletal anchoring proteins. Fractionation of rat brain membranes indicated that somatostatin receptor interacting protein is enriched in the postsynaptic density fraction. The interaction of somatostatin receptor subtype 2 with its interacting protein was verified by overlay assays and coimmunoprecipitation experiments from transfected human embryonic kidney cells. Somatostatin receptor subtype 2 and the interacting protein display a striking overlap of their expression patterns in the rat brain. Interestingly, in the hippocampus the mRNA for somatostatin receptor interacting protein was not confined to the cell bodies but was also observed in the molecular layer, suggesting a dendritic localization of this mRNA.     

Parallel dose-response curves in combination experiments

A possible experimental design for combination experiments is to compare the doseresponse curve of a single agent with the corresponding curve of the same agent using either a fixed amount of a second one or a fixed dose ratio. No interaction is then often defined by a parallel shift of these curves. We have performed a systematic study for various types of doseresponse relations both for the dose-additivity (Loewe additivity) and for the independence (Bliss independence) criteria for defining zero interaction. Parallelism between doseresponse curves of a single agent and those of the same agent in the presence of a fixed amount of another one is found for the Loewe-additivity criterion for linear doseresponse relations. For nonlinear relations, one has to differentiate between effect parallelism (parallel shift on the effect scale) and dose parallelism (parallel shift on the dose scale). In the case of Loewe additivity, zero-interaction dose parallelism is found for power, Weibull, median-effect and logistic doseresponse relations, given that special parameter relationships are fulfilled. The mechanistic model of competitive interaction exhibits dose parallelism but not effect parallelism for Loewe additivity. Bliss independence and Loewe additivity lead to identical results for exponential doseresponse curves. This is the only case for which dose parallelism was found for Bliss independence. Parallelism between single-agent doseresponse relations and Loewe additivity mixture relations is found for examples with a fixed doseratio design. However, this is again not a general property of the design adopted but holds only if special conditions are fulfilled. The comparison of combination doseresponse curves with single-agent relations has to be performed taking into account both potency and shape parameters. The results of this analysis lead to the conclusion that parallelism between zero interaction combination and single-agent doseresponse relations is found only for special cases and cannot be used as a general criterion for defining zero-interaction in combined-action assessment even if the correct potency shift is taken into account.     

Insulin Sensitivity of Heifers on Different Diets

The hyperinsulinemic euglycemic clamp technique was used to investigate the effect on insulin sensitivity of 2 different diets used in practical cattle feeding in calves. Ten 4 to 5-month-old heifer calves were allocated to 2 feeding groups, LO or HI, to obtain growth rates of 400 g/day or 900 g/day. The heifers were fed and housed individually for 5 weeks. Growth rates close to calculated rates were obtained with the diets used. Weekly blood samples were collected from the jugular vein for analysis of glucose, insulin, cortisol, total serum protein, urea, cholesterol and nonesterified fatty acids. During week 5, insulin sensitivity was estimated using the hyperinsulinemic euglycemic clamp technique. Insulin sensitivity did not differ between the groups, but the plasma glucose levels were higher during weeks 3 and 4 for the HI group compared to the LO group. It may be concluded that the amount of concentrate in the diet was too low to induce changes in either the basal plasma insulin levels or the insulin sensitivity in the HI group.