Granulocyte colony-stimulating factor (G-CSF) enhances cocaine effects in the nucleus accumbens via a dopamine release-based mechanism

Abstract

Cocaine use disorder is associated with alterations in immune function including altered expression of multiple peripheral cytokines in humans-several of which correlate with drug use. Individuals suffering from cocaine use disorder show altered immune system responses to drug-associated cues, highlighting the interaction between the brain and immune system as a critical factor in the development and expression of cocaine use disorder. We have previously demonstrated in animal models that cocaine use upregulates the expression of granulocyte colony-stimulating factor (G-CSF)-a pleiotropic cytokine-in the serum and the nucleus accumbens (NAc). G-CSF signaling has been causally linked to behavioral responses to cocaine across multiple behavioral domains. The goal of this study was to define whether increases in G-CSF alter the pharmacodynamic effects of cocaine on the dopamine system and whether this occurs via direct mechanisms within local NAc microcircuits. We find that systemic G-CSF injection increases cocaine effects on dopamine terminals. The enhanced dopamine levels in the presence of cocaine occur through a release-based mechanism, rather than through effects on the dopamine transporter-as uptake rates were unchanged following G-CSF treatment. Critically, this effect could be recapitulated by acute bath application of G-CSF to dopamine terminals, an effect that was occluded by prior G-CSF treatment, suggesting a similar mechanistic basis for direct and systemic exposures. This work highlights the critical interaction between the immune system and psychostimulant effects that can alter drug responses and may play a role in vulnerability to cocaine use disorder.

Keywords: Cocaine; Granulocyte colony-stimulating factor; Nucleus accumbens.

Figure 1.. Granulocyte colony stimulating factor enhances cocaine-induced dopamine release in the nucleus accumbens core.

A. Male C57BL6/J mice were injected with G-CSF (50μg/kg, IP) twice – 24 hours before recording and again one hour before recording. B. Fast scan cyclic voltammetry was done in the NAc core. Dopamine release was evoked from dopamine terminals via an electrical stimulation (n = 6 per group). Dopamine is recorded at 10 Hz via a carbon fiber microelectrode, which allows for sub second dopamine monitoring (inset, right) and the dissociation of dopamine release, from uptake-related measures. C. Cocaine increases extrasynaptic dopamine levels within the NAc via multiple mechanisms. Its canonical effects occur via inhibiting the dopamine transporter (DAT) and inhibiting reuptake; however, it also acts to promote exocytotic release. These studies were designed to specifically parse release-based mechanisms from clearance-based and DAT-mediated mechanisms. D. Pre-treatment with G-CSF before recording sessions resulted in enhanced effects of cocaine on dopamine terminals in the NAc. Left, Current versus time plots showing measured dopamine levels during the application of 10μM cocaine to the slice. Right, color plots showing dopamine (green) at its oxidation potential as well as a cyclic voltammogram (inset) showing its characteristic redox signature of dopamine. E. Group data showing peak dopamine release over time after the bath application of 10μM cocaine in saline or G-CSF treated mice. F. The maximum evoked release in the presence of cocaine in each subject. Extrasynaptic dopamine levels in the presence of cocaine were enhanced following systemic G-CSF treatment. G. Michaelis-Menten modeling was used to assess the apparent km of cocaine for the dopamine transporter (a measure of relative affinity of cocaine for DAT). This was plotted over time following cocaine bath application. F. The apparent Km from each animal was plotted following stabilization. G-CSF did not affect cocaine effects at the dopamine transporter, suggesting that its effects are specific to release. Data represented as mean +/− S.E.M., * p < 0.5, ** p < 0.01.

Figure 2.. G-CSF enhancement of cocaine effects on the dopamine system is via a release-based mechanism.

A. Example of an evoked dopamine signal over time. Significant work has defined what biological factors dictate specific components of the curve, with the rising phase being mediated by active dopamine release and the falling phase via DAT-mediated clearance. Changes in either factor can contribute to changes in the total timing and amount of dopamine in the extrasynaptic space. B. The magnitude of dopamine response, which is a function of both release and clearance mechanisms, can be assessed via area under the curve measurements (AUC) which represents the total effect of cocaine on the dopamine signal. C. Release and clearance-based measurements were parsed by assessing the total dopamine release (determined as the peak signal), and tau (the time in seconds it takes for the signal to go from peak to 2/3 of peak height). These analyses and comparisons were done for all groups. D. Representative color plots showing evoked dopamine responses at baseline (top) and following bath application of 10μM cocaine (bottom) following saline or G-CSF injection. E. Current versus time plots showing all groups at baseline and following bath application of cocaine on the same scale. Left, saline-injected animals before and after cocaine. Right, G-CSF injected animals. F. AUC analysis of the cocaine curves for saline treated or G-CSF treated animals showing that G-CSF treatment enhances cocaine effects. G. A series of correlations were run in order to parse the factors contributing to these dopamine signals. Correlation of Tau in the presence of cocaine and release in the presence of cocaine, showing that they are weakly negatively correlated. This means that more time to return to baseline is associated with less release. Data represented as mean +/− S.E.M., * p < 0.5, **** p < 0.0001.

Figure 3.. G-CSF enhances cocaine effects through direct actions in the nucleus accumbens.

A. In these control animals, fast scan cyclic voltammetry was run. B. G-CSF or vehicle was applied via bath application to the NAc slices for one hour prior to cocaine application (n = 5 per group). C. G-CSF bath application had no effect on cocaine effects at DAT as measured by apparent Km. D. Group averages also showed no difference in apparent Km measures. E. Cocaine effect on dopamine release as presented as a percent of baseline. F. G-CSF bath application increased cocaine effects on dopamine release. Data represented as mean +/− S.E.M. *, p < 0.05.

Figure 4.. No effect of G-CSF bath application on cocaine in G-CSF pre-treated animals.

A. Male C57BL6/J mice were injected with G-CSF (50μg/kg, IP) twice – 24 hours before recording and again one hour before recording. B. Fast scan cyclic voltammetry was run, and G-CSF or vehicle was applied via bath application to the NAc slices for 1 hour prior to cocaine application (n = 3 per group). C. G-CSF bath application had no effect on cocaine effects at DAT as measured by apparent Km. D. Group averages also showed no difference in apparent Km measures. E. Cocaine effect on dopamine release as presented as a percent of baseline dopamine release. F. G-CSF bath application also had no effect on cocaine effects on release following G-CSF pre-treatment. Data represented as mean +/− S.E.M.