Ultrasonic vocalizations: evidence for an affective opponent process during cocaine self-administration

Abstract

Rationale: Preclinical models of cocaine addiction in the rodent have shown that cocaine induces both positive and negative affective states. These observations have led to the notion that the initial positive/euphoric state induced by cocaine administration may be followed by an opposing, negative process. In the rodent, one method for inferring positive and negative affective states involves measuring their ultrasonic vocalizations (USVs). Previous USV recordings from our laboratory suggested that the transition between positive and negative affect might involve decaying or sub-satiety levels of self-administered cocaine.

Objectives: In order to explicitly test the role of cocaine levels on these affective states, the present study examined USVs when calculated body levels of cocaine were clamped (i.e., held at a constant level via experimenter-controlled infusions) at, below, or above subjects' self-determined drug satiety thresholds.

Results: USVs indicated that (1) positive affect was predominantly observed during the drug loading period, but declined quickly to near zero during maintenance and exhibited little relation to calculated drug level, and (2) in contrast, negative affect was observed at sub-satiety cocaine levels, but was relatively absent when body levels of cocaine were clamped at or above subjects' satiety thresholds.

Conclusions: The results reinforce the opponent-process hypothesis of addiction and suggest that an understanding of the mechanisms underlying negative affect might serve to inform behavioral and pharmacological therapies.

Figure 1. Animals acquire and stabilize SA behaviors across two weeks of SA training

A) Lever responding across training significantly increased before stabilizing B) Peak drug level (mg/kg) across training increased over sessions. The average drug levels during the sub, circa, and supra-satiety conditions are shown using the solid, dotted, and dashed lines, respectively. C) Bodyweight (g) decreased across SA. D) Total drug consumption (mg/kg) increased across training. All data points are shown as the Mean ± SEM. Session is represented on the x-axes while individual dependent measures are represented on the y-axes.

Figure 2. Drug levels above satiety attenuate lever responding while drug levels below satiety increase responding

A) Lever responses across hours in the sub-, circa- and supra-satiety drug clamps with the lever. Example calculated cocaine from one animal during B) self-administration C) the sub-satiety clamp with a lever D) the circa-satiety clamp with a lever and E) the supra-satiety clamp with a lever. The vertical dashed lines at 90 minutes represent the start of computer-controlled micro-infusions for each of the clamp conditions (C–E) and the tick marks along the bottom of each graph represent lever presses performed by the animal across the session.

Figure 3. Fifty-kHz ultrasonic vocalizations increase following cocaine administration (load-up) before subsequently decaying

The average number of ultrasonic vocalizations (USVs; mean ± SEM) is shown as a function of the number of infusions earned. Twenty-two kHz USVs were highest prior to cocaine SA and decayed following the first infusion of cocaine. Fifty-kHz USVs rapidly increased following the first infusion but decayed as SA continued. During the pre-drug period (i.e., ‘0’ infusions earned) individual subjects emitted exclusively 22- or 50-kHz USVs during 85% of observations. The remaining observations were characterized by baseline rates of USV emission or a mix of 22- and 50-kHz USVs with one call category always predominating (Mean=15-fold difference). Nine of eleven subjects exhibited increases in 50-kHz USVs following the first infusion of cocaine while the remaining two subjects exhibited primarily decreases in 22-kHz USVs over the same period. +: p < 0.10, *: p <0.05 when compared to the 22-kHz range. α: p< 0.05 when compared within frequency to baseline.

Figure 4. Sub-satiety drug levels produce increases in 22-kHz ultrasonic vocalizations

The average number of ultrasonic vocalizations (USVs; mean ± SEM) emitted during the last four hours of each condition. Twenty-two kHz USVs were greatest when calculated drug levels were clamped below satiety (sub-satiety drug clamp). In comparison to 50-kHz USVs, all eleven subjects showed robust increases in short-22kHz USVs during the sub-satiety clamp. Twenty-two kHz calls were also observed during the maintenance period of SA (week 2) and were greatest when calculated drug levels were lowest (inset). Inset: 22-kHz USVs during the last four hours of the week 2 recording are shown as a function of drug level quartiles.Q₁: 0–25^th percentile; Q₂: 25–50^th percentile; Q₃: 50–75^th percentile; Q₄: 75–100^th percentile. †:P < 0.05 when compared to the lowest quartile (Q₁). *: p <0.05 when compared within-frequency to the baseline condition. +: p<0.05 when compared within-frequency to the circa-satiety clamp. All other statistical comparisons are found in the text.