The Rate of Cocaine Self-Administration as a Function of Unit Dose: the Result of a Graded or Quantal Pharmacodynamic Response?

Andrew Norman, Mantana Norman, Michael Tabet, Vladimir Tsibulsky. University of Cincinnati, Cincinnati, United States.

The change in frequency of cocaine self-administration as a function of unit dose is widely assumed to represent a graded pharmacodynamic response, indicating a dose-dependent increase in reinforcement and/or the rate-limiting effects of cocaine. However, this theory ignores the cumulative concentration of cocaine after multiple doses. An alternative pharmacokinetic (PK)/pharmacodynamic (PD) theory (1) states that the cocaine unit dose-response curve is the product of the cocaine t1/2 and the minimum concentration at which a quantal PD response occurs (Cmin/satiety threshold). Cmin is assumed to be independent of unit dose. If so, the antagonist-induced increase in Cmin should also be independent of unit dose.

Male Sprague-Dawley rats (350-500 g over the course of the study) were trained to self-administer cocaine HCl. Once a stable rate of self-administration was maintained at cocaine unit doses of either 0.75, 1.5, 3 or 6 μmol kg-1, the D1 dopamine receptor antagonist SCH23390 (15 nmol kg-1 i.v.) was administered and the session continued.

The inter-injection intervals during maintained self-administration were stable and were proportional to the unit dose. A rectangular hyperbolic function provided a good fit to the data but the resulting EC50 was always higher than the highest cocaine unit dose used. SCH23390 accelerated the rate of cocaine self-administration, which resulted in a shift to the right in the unit dose-response curve. Intervals gradually returned to towards baseline values over 3-5 hours and the time course and magnitude of the change in cocaine concentration ratio, based on the calculated cocaine level at the time of each lever press (Cmin) was the same across all unit doses.

Despite the apparent antagonist-induced rightward shift with the graded response model the maximal response, and EC50, are not measurable over a range of non-lethal unit doses. The magnitude and time-course of the antagonist-induced increase in Cmin was independent of the unit dose consistent with the PK/PD model assuming a quantal response. It is concluded that all unit doses contain the same PD information and a dose-response curve is unnecessary for assessing the effects of antagonists. Instead, the unit dose simply affects the resolution of the measurement of the antagonist effect.

(1) Norman and Tsibulsky (2006) Brain Res. 1116:143.