Address correspondence and reprint requests to Bita Moghaddam Department of Psychiatry, Yale University School of Medicine, VA Medical Center 116 A/2, West Haven, CT 06516, USA. E-mail: email@example.com
The prefrontal cortex (PFC) is thought to provide an excitatory influence on the output of mesoaccumbens dopamine neurons. The evidence for this influence primarily arises from findings in the rat that chemical or high-intensity and high-frequency (60–200 Hz) electrical stimulations of PFC increase burst activity of midbrain dopamine neurons, and augment terminal release of dopamine in the nucleus accumbens. However, PFC neurons in animals that are engaged in PFC-dependent cognitive tasks increase their firing frequency from a baseline of 1–3 Hz to 7–10 Hz, suggesting that the commonly used high-frequency stimulation parameters of the PFC may not be relevant to the behavioral states that are associated with PFC activation. We investigated the influence of PFC activation at lower physiologically relevant frequencies on the release of dopamine in the nucleus accumbens. Using rapid (5-min) microdialysis measures of extracellular dopamine in the nucleus accumbens, we found that although PFC stimulation at 60 Hz produces the expected increases in accumbal dopamine release, the same amplitude of PFC stimulation at 10 Hz significantly decreased these levels. These results indicate that activation of PFC, at frequencies that are associated with increased cognitive demand on this region, inhibits the mesoaccumbens dopamine system.
The descending projections of the prefrontal cortex (PFC) are believed to exert an excitatory control on midbrain dopamine neurons and dopamine release in the ventral striatum, including the nucleus accumbens (NAc). The evidence for this interaction primarily arises from neurochemical and electrophysiological studies demonstrating that stimulation of the PFC increases dopamine release in the NAc and increases burst firing of midbrain dopamine neurons (Murase et al. 1993; Taber and Fibiger 1995; Karreman and Moghaddam 1996; Tong et al. 1996; You et al. 1998).
We recently reported that electrical stimulation of the basolateral amygdala in freely moving rats does not increase the release of dopamine in the NAc unless the efferent activity of the PFC is blocked (Jackson and Moghaddam 2001). This observation suggests that, at least during conditions that activate the amygdala efferents, the PFC provides an inhibitory control over NAc dopamine release. However, such an inhibitory process would contradict the aforementioned observations that chemical stimulation with glutamate or bicuculline (Murase et al. 1993; Karreman and Moghaddam 1996), or electrical stimulation of PFC (Taber and Fibiger 1995; You et al. 1998) increases dopamine release in the NAc. A critical examination of these latter studies would suggest that the applied stimulation parameters were not similar to physiological modes of PFC activation, i.e. levels of activation during behaviors that are dependent on the functional integrity of the PFC. For example, the chemical stimulation studies may have produced supra-normal levels of activation because, at least in our study which was performed in unanesthetized animals (Karreman and Moghaddam 1996), the concentration of bicuculline that produced an increase in NAc dopamine also produced profound behavioral activation. Similarly, the electrical stimulation studies utilized very high frequencies of 60–200 Hz. This range of frequencies was chosen not because of a physiological correlate with the firing pattern of rat cortical neurons, but because rats have demonstrated a tendency to self-stimulate at these frequencies (Phillips and Fibiger 1979). However, basal firing rates of PFC neurons in unanesthetized rats are reported to be around 2 Hz, and maximum firing rates during tasks that require PFC activation, such as spatial working memory tasks or sustained visual attention, seldom exceed 10 Hz (Jung et al. 1998; Gill et al. 2000; Givens et al. 2000). Thus, based on our recent observation that stimulation of excitatory afferents to PFC inhibits the ability of amygdala stimulation to increase NAc dopamine release (Jackson and Moghaddam 2001), we hypothesized that electrical stimulation of the PFC at frequencies that mimic the physiological activation of this region in cognitive behavioral contexts should decrease NAc dopamine release. This hypothesis was tested by performing rapid (5-min) microdialysis collection of dopamine in the NAc while comparing the effect of 10 and 60 Hz stimulation of PFC on dopamine release.
Materials and methods
All animal procedures were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Yale University Animal Care and Use Committee. A total of eight male Sprague–Dawley rats (280–350 g) were used for these experiments. Under halothane anesthesia, a concentric microdialysis probe (Adams and Moghaddam 1998) was stereotaxically implanted into the NAc (AP + 1.4, L 1.1, V 8.3) and a bipolar stainless steel stimulating electrode was implanted into the ipsilateral PFC (AP + 3.2, L 0.8, V 5.3). All coordinates are relative to Bregma and are defined according to the atlas of Paxinos and Watson (1982). Probe and electrode were secured into place with dental acrylic and anchored by skull screws. After surgery, animals were placed in a clear polycarbonate cage (44 × 22 × 42 cm) with bedding, and the microdialysis probes were connected to a liquid swivel-balance arm assembly. This cage was placed in a quiet room with a 12-h light/dark cycle (lights on at 7:00 am, the same as the regular animal housing quarters). Animals had free access to food and water and were allowed to recover for at least 24 h before the start of microdialysis experiments, which were performed in the same cage/room environment in freely moving rats.
Concentric microdialysis probes were constructed as described elsewhere (Adams and Moghaddam 1998). The probes had an outer diameter of 330 µm and exposed membrane of 2.0 mm. The perfusion solution contained 145 mm NaCl, 2.7 mm KCl, 1.0 mm MgCl2 and 1.2 mm CaCl2. Probes were perfused at a flow rate of 0.5 µL/min during the 24-h surgical recovery period and 2.5 µL/min during the experiment. Dialysis samples were collected every 5 min and the amount of dopamine in the samples was measured by HPLC with electrochemical detection as described elsewhere (Adams and Moghaddam 1998).
Three stable baseline samples (< 10% variation) were collected prior to PFC stimulation. Each rat was stimulated at 10 Hz or 60 Hz (100 µA, 0.5-ms monophasic pulses, stimulus duration = 0.5 s, interstimulus interval = 5 s) for a total duration of 20 min. Stimulation current was provided by a constant current source (Grass) driven by a pulse stimulator (Grass). The 60-Hz stimulation was performed in six of the same animals that had received 10-Hz stimulation three hours earlier. To ensure that the time of stimulation did not contribute to the direction of response, two of the animals received (only) the 10-Hz stimulation in the afternoon about the same time (approx 1–2 pm) that the other animals received the 60-Hz stimulation. Responses in these animals were similar to the other six animals at 10 Hz; therefore, this data was combined for all eight animals.
The average of the first three samples was considered as the baseline and defined to be 100%. All values given are expressed as percentages of baseline ± SEM. Within-group analysis of the main effect of stimulation was performed by one-way analysis of variance (anova) for repeated measures with time as the within-subjects factor and stimulation frequency (10 or 60 Hz) as the between-groups factor. The level of significance for all procedures was p < 0.05.
At the end of each experiment, animals were anesthetized with chloral hydrate. Passing a 50-mA current through the electrode for 30 s marked the location of the stimulating electrode. Animals were perfused intracardially with saline followed by 10% buffered formalin. Fixed brains were cut at 250-µm intervals and sections stained with cresyl violet. Probe and electrode placements were verified for all the data presented in this study, and are shown in Fig. 1.
The baseline levels of dopamine were 1.31 ± 0.12 fmol/µL before the 10-Hz stimulation and 1.14 ± 0.15 fmol/µL before the 60-Hz stimulation. Stimulation of the PFC at the low frequency of 10 Hz (100 µA, 0.5-ms pulse, 0.5-s stimulus duration, 5-s interstimulus interval) produced a stimulus-locked decrease in NAc dopamine levels to 86% ± 2% of baseline levels (Fig. 2). This decrease was significant across time (F = 2.2, n = 8, p = 0.02), persisted throughout the stimulation period, and returned to baseline immediately upon stimulation termination. In contrast, stimulation at 60 Hz (100 µA, 0.5-ms pulse, 0.5-s stimulus duration, 5-s interstimulus interval) increased dopamine levels to a maximum of 131% ± 13% of baseline levels. This increase was significant across time (F = 2.7, n = 6, p = 0.007). A significant stimulation frequency–time interaction was found between the two stimulation frequencies (F = 14.6, p = 0.002).
Animals were in a rest/sleep state before stimulation. There was no overt behavioral reaction to either frequency of stimulation. Most of the animals (five out of eight) exhibited mild behavioral activation during the stimulation, as characterized by slow movement about the cage and brief periods of grooming or rearing at either stimulation frequency.
Electrical stimulation of PFC at 10 Hz, which closely corresponds to the maximum in vivo firing rate of PFC neurons in animals engaged in cognitive tasks, decreased dopamine release in the NAc. This decrease was locked to the electrical stimulation, returning to baseline levels immediately upon termination of stimulation. This observation agrees with our previous findings suggesting that the PFC provides an inhibitory control over NAc dopamine release under tonic conditions (Takahata and Moghaddam 2000), and during basolateral amygdala stimulation (Jackson and Moghaddam 2001).
In contrast to the inhibitory effect of 10-Hz stimulation, PFC stimulation at 60 Hz increased NAc dopamine. This latter finding is in agreement with earlier reports. For example Taber and Fibiger (1995), using identical stimulus parameters as those used in the present study, observed a similar magnitude of NAc dopamine increase (131% compared with 138%). You et al. (1998) also reported an increase in NAc dopamine during PFC stimulation at > 25 Hz (500 µA) to approximately 130% of baseline. Although these investigators did not report a significant effect in response to 6- or 12-Hz stimulation, a close examination of their data suggests a small decrease in dopamine release that may have been masked by the slow collection interval used in the study.
Considering that PFC afferents do not make direct synaptic contact with mesoaccumbens dopamine neurons (Carr and Sesack 2000), the main question raised by our findings is why would different stimulation frequencies of the PFC have opposite influences on these neurons? There are several identified inhibitory networks that can account for an inhibitory influence of PFC on NAc dopamine release at ‘normal’ physiological levels of stimulation, from which the 10-Hz stimulation parameter was derived. One mechanism may involve activation of corticostriatal efferents and stimulation of GABAergic medium spiny output neurons in the NAc (Sesack and Pickel 1992). These GABA neurons, which project to the ventral tegmental area (VTA), could in turn inhibit the mesoaccumbens dopamine neurons. A second inhibitory pathway may involve activation of PFC projections to the GABAergic neurons in the VTA, which could produce local inhibition of dopamine neurons in this region (Carr and Sesack 2000).
The 60-Hz stimulation would also be expected to activate these inhibitory pathways. However, driving cortical efferents at supra-normal frequencies may result in substantially higher levels of glutamate release that overtax the glutamate uptake mechanisms and results in a large overflow of glutamate in cortical projection regions including the VTA. In support of this mechanism, Rossetti et al. (1998) examining the frequency-dependent effect of PFC stimulation on extracellular glutamate levels in the VTA, reported no increase after 4- or 16-Hz stimulation, whereas 60-Hz stimulation produced a profound and sustained increase in VTA glutamate levels to 246% of baseline. This increase, which is a measure of glutamate overflow, suggests that at high stimulation frequencies, glutamate could reach dopamine neurons that do not receive direct synaptic input from PFC. This will produce indiscriminate activation (and bursting) of all dopamine neurons, including those that project to the NAc. This indirect activation may be strong enough to counteract the normal inhibitory influence that occurs at lower frequencies of stimulation, resulting in a net increase in NAc dopamine release. Alternatively, the differences in the number of applied pulses in the 10- and 60-Hz stimulation paradigms may have produced the differential effect on release. However, this appears unlikely in that the effects of the two stimulation paradigms were in opposite directions.
One caveat to our conclusion that behaviorally relevant activation of PFC may primarily produce an inhibitory effect of NAc dopamine release is that previous studies have reported that reducing the basal activity of PFC by lesions (Jaskiw et al. 1990), intra-PFC application of lidocaine (Murase et al. 1993), tetrodotoxin (Karreman and Moghaddam 1996) or amphetamine (Louilot et al. 1989; Kolachana et al. 1995; Karreman and Moghaddam 1996) decreases dopamine turnover or release in the ventral striatum, suggesting that the PFC may produce a tonic excitatory influence on these levels. However, lesion studies are confounded by compensatory mechanisms, and studies involving amphetamine should be interpreted cautiously because amphetamine is a potent releaser of serotonin, which is known to activate PFC pyramidal neurons (Marek and Aghajanian 1999). Similarly, studies performed in anesthetized animals (e.g. Murase et al. 1993) are somewhat difficult to interpret because PFC neurons are not active under anesthesia. The decrease in NAc dopamine in response to intra-PFC application of lidocaine or tetrodotoxin (Murase et al. 1993; Karreman and Moghaddam 1996) is also compounded by the fact that as a result of the close proximity of PFC and NAc, these drugs may have infused ventrally and exerted their inhibitory effect directly at the level of NAc. This mechanism would explain the delayed onset of this decrease in both studies.
Understanding the mechanistic basis of PFC modulation of ventral striatal dopamine function is of great interest because this modulation may be fundamental to executive control of goal-directed behavior, and has been implicated in the pathophysiology of psychiatric disorders such as schizophrenia and addiction that are thought to be associated with abnormalities in both PFC and limbic dopamine function (Weinberger et al. 1986; Robbins 1996; Volkow et al. 1999; Grace 2000). The present study demonstrates that, in contrast to the previous notion that the PFC positively modulates the activity of the mesoaccumbens dopamine system under stimulation conditions that resemble the physiological range of the firing rate of PFC neurons during performance of PFC-dependent cognitive tasks, release of dopamine in the NAc is inhibited. However, our results also indicate that PFC regulation is biphasic, and under experimental conditions that may have relevance to pathophysiological states the PFC provokes an abnormal increase in limbic dopamine function.
The authors are grateful to Karyn Groth for technical assistance. This work was supported by National Institute of Mental Health awards MH01616, MH48404, MH44866 and the Veterans Association National Centers for PTSD and Schizophrenia.