- Email: [email protected]
Contrasting effects of stress on medial and sulcal prefrontal cortex self-stimulation
Brain
Research Bulletin, Vol. 27, pp. 225-229. o Pergmon
0361-9230191 $3.00 + .oO
Press plc, 1991. Printed in the U.S.A.
RAPID COMMUNICATION
Contrasting Effects of Stress on Medial and Sulcal Prefrontal Cortex Self-Stimulation IAIN S. MCGREGOR Department
of Psychology,
University of Sydney, NSW 2006,
Australia
Received 26 April 199 1 I. S. Contrasting effects of stress on medial and s&al prefrontal cortex self-stimulation. BRAIN RES BULL 27(2) 225-229, 1991.-Male Wistar rats were subjected to either 25 controllable or uncontrollable footshocks and then tested for changes in fixed-interval 5-second (n-5) self-stimulation of the medial prefrontal cortex (MPC), sulcal prefrontal cortex (SEC) or nucleus accumbens (NAS). Controllable footshock caused a moderate facilitation of MPC self-stimulation (30% above baseline rates) but inhibited SEC self-stimulation (32% below baseline rates). Uncontrollable footshock had no effect on MPC self-stimulation but inhibited SEC self-stimulation (52% below baseline rates). An inhibition of SF’C self-stimulation was also evident 24 hours following controllable or uncontrollable footshock. NAS self-stimulation was unaffected by footshock. Changes in locomotor activity were not consistently related to changes in self-stimulation following footshock. These results are discussed in terms of the different effects of mild stress on the release of reward-relevant neurotransmitters in the MEC, SEC and NAS. The possible role of stress-induced hypoalgesia in determining the stress-induced facilitation of MPC self-stimulation is also discussed.
MCGREGOR,
Stress Dopamine
Sulcal prefrontal cortex Medial prefrontal cortex Activity Hypoalgesia Fixed-interval
Nucleus accumbens
Self-stimulation
Rat
creases DA utilization in the MPC, but not in the SPC or NAS (4,5). Only with more severe stressors is increased DA utilization in all three regions seen (4,5). Both the NAS and SPC support self-stimulation. In both regions self-stimulation may be at least partly mediated by DA. SPC self-stimulation is decreased following lesions of ascending DA fibers from the ventral tegmental area (3) and is also inhibited by systemic injection of apomorphine (19). NAS self-stimulation is inhibited following intra-NAS injection of DA antagonists (24). If DA modulates both NAS and SPC self-stimulation then it might be predicted that stress will affect both NAS and SPC self-stimulation. However, since the effects of stress on DA in the NAS and SPC may be different to those seen in the MPC, we might predict contrasting effects of stress on self-stimulation of these three regions.
AN important observation in support of the “learned helplessness” model of depression is that uncontrollable stress impairs performance in a number of behavioral paradigms (17). It is somewhat surprising then that recent research from this laboratory has demonstrated a facilitation of medial prefrontal cortex (MPC) self-stimulation in rats subjected to such stress. Exposure to controllable or uncontrollable footshock, brief restraint or administration of the anxiogenic drug FG-7142 causes an immediate increase in MPC self-stimulation rates (12,13). A facilitation of MPC self-stimulation may be evident for up to 72 hours following a single exposure to stress (12). Stress enhances the utilization of the neurotransmitter dopamine (DA) in the MPC (1, 4, 5, 25, 26), and DA is thought to modulate the reinforcing efficacy of MPC self-stimulation (20). The stress-induced facilitation of MPC self-stimulation may thus reflect the increased release of DA in the MPC enhancing the reinforcing quality of MPC stimulation. Long-term effects of stress on MPC self-stimulation may reflect alterations in the sensitivity or number of MPC DA receptors (12,14). Stress also increases DA utilization in other forebrain regions, including the sulcal prefrontal cortex (SPC) and nucleus accumbens (NAS) (1, 4, 5, 25, 26). However, the effects of stress on DA in the SPC and NAS are different to those seen in the MPC. Firstly, the increases in DA utilization in the NAS and SPC are smaller than those seen in the MPC (4,5). Secondly, DA utilization in the SPC and NAS is not affected by milder stress, such as conditioned fear. For example, exposing a rat to an environment in which it was previously shocked in-
METHOD
Subjects, Surgery and Histology The subjects were 28 male Wistar rats weighing between 270 and 350 grams at the time of surgery. The rats were individually housed and kept in a temperature-controlled vivarium at 22 f 2°C on a 14: 10 hour light-dark cycle. Ad lib food and water were available in the home cages. The rats were stereotaxically implanted with monopolar stainless steel electrodes under ketamine-xylazine anaesthesia. The electrodes were aimed at the MPC (n = 8), SPC (n = 14) or NAS
225
(n =h). The coordinates (21) used were (mm): MPC =A: 4.2. L: 0.7. V: 3.5, SPC=A: 3.2, L: 3.5. V: 4.5 and NAS=A: 2.2, L: 1.5. V: 6.8. A stainless steel wire wrapped around two skull screws located 2 mm posterior to bregma served as the indifferent electrode. The electrode assembly was held in place with a cap made of dental acrylic cement. At the completion of the experiment the rats were sacrificed with an overdose of barbiturate. Immediately following death, an anodal current was passed through the electrode for 10 s to produce a small lesion at the tip of the electrode. Brains were rapidly removed from the cranial vault. frozen to - 12°C. and sliced on a microtome. The brain slices were then placed on glass slides and stained with Toluidine Blue. Electrode locations were determined microscopically with reference to the atlas of Paxinos and Watson (3-l). Procedure The self-stimulation methods and apparatus were as described previously (10, 12, 13). Testing occurred in four identical operant sound-attenuated chambers (37 x 25 x 32 cm) fitted with single levers. Stimulation consisted of 0.5 second trains of 100 Hz current-regulated cathodal rectangular pulses with a pulse width of 0.2 ms. The rats were trained for several weeks to acquire stable lever pressing for brain stimulation on a fixed-interval 5-second (FI-5) schedule of reinforcement. Use of this schedule minimizes the nonspecific effects of stimulation upon performance while retaining sensitivity to alterations in reinforcement efficacy caused by drugs or stress (10, 12, 13). Once the rats were stabilized on FI-5, rate-intensity functions were computed for each rat and individual currents adjusted to ensure that response rates were approximately half-maximal. For six consecutive days prior to being stressed, the rats were given baseline self-stimulation and locomotor activity tests. The test sessions consisted of a 20-minute period where the rats pressed for self-stimulation at half-maximal current under FI-5. The total number of responses in 20 minutes was measured. This was immediately followed by a 20-minute locomotor activity test. The rats were placed in Coulboum Test Cages (30 X 30 X 25 cm) which had clear acrylic front and back panels and metal sides and top. The floor of each cage consisted of a grid of sixteen metal bars connected through a high impedance amplifier so that contact or the breaking of contact between four evenly spaced bars and the other twelve produced a count that was recorded by computer. The total number of counts in 20 minutes was recorded. The rats within each group (MPC, SPC, or NAS) were paired according to their self-stimulation rates on the final baseline day. Pair members were allocated randomly to an uncontrollable or controllable shock condition. On the following day, footshock was administered prior to self-stimulation. Footshock was delivered in a two shuttle boxes (70 x 11 X 14 cm) in a separate room to where self-stimulation occurred. A variable resistor shock generator (Hayes) and scrambler were used to deliver shock of 0.6 mA intensity to the grid floors of the shuttle boxes. Rats in the controllable shock condition received 25 trials where shock could be terminated through performance of a FR-1 shuttle response. If no escape response was made within 60 seconds then the trial was automatically terminated. The intertrial interval was either 15, 30, 45 or 60 seconds with a randomized order of intervals assigned prior to testing. Rats in the uncontrollable shock condition were yoked to their controllably shocked partner and received shock whenever their partner did, but independently of any emitted response.
FIG. 1. Coronal sections ranging from 2.2 to 4.7 mm anterior to bregma [modified from the atlas of Paxinos and Watson (21)]. Electrodes placements are shown in the MPC (black dots), SPC (open circles) and NAS (black squares).
Immediately following footshock the rats were given a standard 20-minute self-stimulation session followed by a 20-minute activity test. Rats were also tested for self-stimulation and activity 24 hours following shock since previous results have indicated that effects of footshock on self-stimulation can be detected at this time (2, 12, 13, 27, 28). The effects of shock on selfstimulation and activity for each group were evaluated relative to scores on the last of the six baseline days using the r-test for paired samples. RESULTS
Histology The locations of the electrodes in the 28 rats completing the experiment are shown in Fig. 1. MPC electrodes were all within the prelimbic or infralimbic cortex [terminology according to Krettek and Price (9)J. SPC electrodes were either in the dorsal or ventral agranular insular cortex. NAS electrodes were mostly within the anterior NAS. Shock Avoidance The three rats in the NAS group receiving controllable shock showed faster escape latencies than the rats in the MPC or SPC groups (ps
221
STRESS AND SELF-STIMULATION
also shown little relation between
N
ulation following
+?
c MPC
3 UC
N
'P
LS
* P * :
c
i UC
FIG. 2. Mean change in self-stimulation rates (upper) and activity (lower) at 0 and 24 hours following shock in the MPC, NAS and SPC groups given controllable (C) or uncontrollable (UC) footshock (*p
236.5 2 28.98 bar presses/20 min. The lower rates for SPC relative to MPC self-stimulation agree with previous reports (10,23). The effects of controllable and uncontrollable footshock on self-stimulation are shown in Fig. 2. There was a clear difference seen across the three groups. Controllable shock facilitated MPC self-stimulation, inhibited SPC self-stimulation and had no effect on NAS self-stimulation. There was also a significant inhibition of SPC self-stimulation 24 hours after controllable shock. Uncontrollable shock inhibited SPC self-stimulation both immediately and 24 hours following shock. There was no significant effect of uncontrollable shock on MPC or NAS self-stimulation. Activity Activity was decreased relative to baseline in nearly every group tested (Fig. 2). The effect was variable across rats and consequently only some of the group effects were statistically significant. Activity changes were not systematically related to alterations in self-stimulation rats since some groups showing large declines in activity (e.g., the MPC and NAS groups given uncontrollable shock) showed no change or an increase in selfstimulation, while other groups showing lesser decreases in activity (e.g., the two SPC groups tested at 24 hours) showed a profound inhibition of self-stimulation. Previous studies have
activity changes and self-stimstress (2, 13, 27). DISCUSSION
The present experiment shows that footshock stress has different effects on self-stimulation of the MPC, SPC and NAS. The near-opposite effects of stress on MPC and SPC self-stimulation are particularly striking and add to previous reports that MPC and SPC self-stimulation have different characteristics. For example, using the conditioned taste and place preference paradigms, Robertson and Laferriere (23) showed that SPC selfstimulation has prepotent associability with gustatory cues, and MPC self-stimulation with contextual cues. Further, SPC but not MPC self-stimulation is facilitated by food deprivation (8,lO) and is associated with increases in food intake and body weight (10). These different characteristics of MPC and SPC self-stimulation agree with other reports of functional dissociations between the two regions [see (7) for review]. In general, the SPC seems to be primarily involved with feeding-related homeostasis (7,l l), while the MPC is involved in cognitive functions (including spatial mapping) and behavioral and autonomic response to stress (7,22). The effects of stress on MPC self-stimulation reported here partly agree with previous results (12,13). However, there are some differences. Firstly, in the present study, uncontrollable footshock had no effect on MPC self-stimulation while a large and significant facilitation was seen in an earlier study (13). However, this earlier study involved 60 rather than 25 uncontrollable shocks. Further, the shocks were of a longer duration due to a IX-2 escape contingency in the controllable shock group (13). This suggests that only a relatively high density of uncontrollable shock will facilitate MPC self-stimulation. A second difference is the absence of a facilitation of MPC self-stimulation 24 hours following stress. In two previous studies a facilitation of MPC self-stimulation was seen at 24 hours (12,13). This discrepancy may also relate to the comparatively low density of shock used in the present study. In particular, the density of shock used here is considerably less than that required to produce “learned helplessness” (15,17), which is characterized by behavioral effects present 24 hours following stress (17). One possible explanation of the stress-induced facilitation of MPC self-stimulation (other than simply citing increased DA release) invokes stress-induced hypoalgesia. MPC stimulation appears to have an aversive component, since rats will learn a response to escape from noncontingent trains of MPC stimulation (18). However, if a rat were hypoalgesic, then this might mask any aversive properties of MPC stimulation leading to a net increase in positive reinforcement and increased self-stimulation. High density uncontrollable footshock produces a longlasting context-independent hypoalgesia (6, 15, 16). Lower density shock (such as was used in the present study) produces only a brief hypoalgesia that is context-dependent and dissipates rapidly (15,16). This would predict that the rats given uncontrollable stress in our previous study (13) (where a high density of shock was used) were hypoalgesic when tested for self-stimulation. This may explain the facilitation of self-stimulation seen in these rats. In contrast, rats given uncontrollable shock in the present study would probably not be hypoalgesic when tested for selfstimulation due to the low density of shock given. This may explain the lack of facilitation of self-stimulation in these rats. Clearly this hypothesis is speculative, but it perhaps deserves some further investigation. The main problem would be in explaining the facilitation of MPC self-stimulation produced by controllable shock, since it is though to produce less hypoalge-
sia than uncontrollable shock [e.g., (6)]. The inhibition of SPC self-stimulation seen following stress in the present study was largely unexpected, since it was thought that the increased DA utilization produced by stress in the SPC might facilitate SPC self-stimulation. It is unlikely that the inhibition observed reflects “learned helplessness” since (as noted above) the density of shock used is insufficient to produce such an effect. Further, rats in the SPC group given controllable shock [who would not by definition be “helpless” (17)] showed a similar inhibition of self-stimulation to the rats in the uncontrollable shock group. The lower baseline rates of serf-stimulation in the SPC group might be thought to imply that SPC self-stimulation would be more easily disrupted by stress. However. the best responding rats in the SPC group (who had baseline rates of self-stimulation that were greater or similar to rats in the MPC and NAS groups) showed just as profound an inhibition of self-stimulation than the lower responding rats. It seems then that stress may have a powerful inhibitor effect on the release of some reward-relevant neurotransmitter in the SPC. At present the identity of this transmitter is unknown. The lack of effect of stress on NAS self-stimulation seen here contrasts with previous reports of an inhibitory effect of stress on NAS self-stimulation (2,27). However, several factors could explain this discrepancy. Firstly, previous studies used mice and involved a much higher density of shock than was used in the present study (227). Further, although the effect of shock on NAS self-stimulation in mice is usually inhibitory. mice of the
BALBic strain display a facilitation of NAS self-stimulation following shock (28). Since only more intense stressors increase DA utilizatton in the NAS of rats (5) it could be that a facilitation of NAS selfstimulation would be seen in rats exposed to greater levels of stress than were used in the present study. Certainly. the tendency toward a facilitation in the NAS group given uncontrollable shock (Fig. 2) suggests that this may be the case. Alternatively. it could be argued that DA is not an important determinant of NAS self-stimulation and this explains the lack of significant effect on NAS self-stimulation in the present study. It is not possible. at present, to determine whether this is the case. In conclusion, the exact mechanisms underlying the contrasting effects of stress on MPC, SPC and NAS self-stimulation are uncertain. Future studies looking at the effects of different densities of shock on self-stimulation of the MPC. SPC and NAS and investigating the pharmacological reversal of stressor effects may prove instructive in unravelling the precise mechanisms involved. The important conclusion from the present study is that stress has con~asting effects on MPC compared to SPC selfstimulation. This further suggests that these two subregions of the rodent prefrontal cortex have different functional attributes. XKNOWLEDGEMENTS This research was supported by a University of Sydney Postgraduate Schol~hip to Iain S. McGregor. Dr. Dale Atrens is thanked for providing facilities used in this research. Dr. H. Q. Lin is thanked for his excellent technical assistance.
REFERENCES 1. Abercrombie, E. D.; Keefe, K. A.; DiFrischia. D. S.; Zigmond. M. J. Differential effect of stress on in vivo dopamine r&ease in striatum, nucleus accumbens, and medial prefrontal cortex. .I. Neurochem. 52:1655-1658; 1989. 2. Bowers, W. J.; Zacharko, R. M.; Anisman, H. Evaluation of stressor effects on intracranial self-stimulation from the nucleus accumbens and the substantia nigra in a current intensity paradigm. Behav. Brain Res. 2385-93; 1987. of the sulcal pre3. Clavier, R. M.; Gerfen, C. R. Self-stimulation frontal cortex in the rat: direct evidence for ascending dopaminergic mediation. Neurosci. Lett. 12:183-l 87; 1979. 4. Dunn, A. J. Stress-related activation of cerebral dopaminergic systems. Ann. NY Acad. Sci. 537:188-205; 1988. 5. Herman, J. P.; Guillonneau, D.; Dantzer, R.; Scatton, B.; Semerdijian-Rouquier, L.; Le Meal, M., Differential effects of inescapable footshocks and of stimuli previously paired with inescapable footshocks on dopamine turnover in cortical and limbic areas of the rat. Life. Sci. 302207-2214: 1982. 6. Jackson, R, L.: Maier, S. F.; Coon, D. J. Long-term analgesic effects of inescapable shock and learned helplessness. Science 206: 91-92; 1979. 7. Kolb. B. Functions of the frontal cortex of the rat: a comparative review. Brain Res. Rev. 8:65-98; 1984. 8. Koolhaus, J. M.; Mora, F.; Phillips, A. G. Effects of food and water dep~vation on self-stimulation of the medial and sulcal prefrontal cortex and caudate m&men in the rat. Physiol. Behav. 18:329331; 1977. 9. Krettek, J. E.; Price, J. L. The cortical projections of the mediodorsal nucleus and adjacent thalamic nuclei in the rat. J. Comp. Neurol. 171:157-192; 1977. IO. McGregor, I. S.; Atrens, D. M. Prefrontal cortex self-stimulation and energy balance. Behav. Neurosci.; in press; 1991. 11. McGregor, I. S.; Menendez, .I. A.; Atrens, D. M. Metabolic effects obtained from excitatory amino acid stimulation of the sulcal prefrontal cortex. Brain Res. 529:1-6; 1990. 12. McGregor, 1. S.; Atrens, D. M. Stressor-like effects of FG-7142 on medial prefrontal cortex self-stimulation. Brain Res. 516: 170-
174; 1990. 13. McGregor, I. S.; Balleine, B. W.; Atrens, D. M. Footshock stress facilitates self-stimulation of the medial prefrontal cortex but not the lateral hypothalamus in the rat. Brain Res. 490:397-403; 1989. 14. MacLennan, A. J.; Pelleymounter, M. A.; Atmadja, S.; Jakubovic, A.; Maier, S. F.; Fibiger, H. C., D, dopamine receptors in the rat prefrontal cortex: characterization and alteration by stress. Brain Res. 477:3OC-307; 1990. 15. Maier, S. F. Diazepam modulation of stress-induced analgesia depends on the type of analgesia. Behav. Neurosci. 104:339-347; 1990. 16. Maier, S. F. Determinants of the nature of environmentally induced hypoalgesia. Behav. Neurosci. 104:131-143; 1989. 17. Maier, S. F.; Seligman, M. E. P. Learned helplessness: theory and evidence. J. Exp. Psychol. (Gen). 105:3-46; 1976. 18. Misrendino, M. J. D.; Coons, E. E. Electrical stimulation of the medial prefrontal cortex supports both ‘pure reward’ and ‘rewardescape’ behaviour in rats. Brain Res. 483:22&232; 1989. 19. Mora, F.; Phillips, A. G.; Koolhaus, J. M.; Rolls, E. T. Prefrontal cortex and neostriamm self-stimulation in the rat: Differential effects of produced by apomorphine. Brain Res. Bull 1:421-424; 1976. pathways and circuits 20. Mora, F.; Ferrer, J. M. R. Neurotransmitters, as the neural substrates of self-stimulation of the prefrontal cortex: facts and speculations. Behav. Res. 22:127-l% 160. 21. Paxinos. G.; Watson, C. The rat brain in stereotaxic coordinates, 2nd ed. Sydney: Academic Press; 1986. 22. Ravard, S.; Camoy, P.; Herve, D.; Tassin, J.-P., Thiebot, M.-H.; Soubrie, P. Involvement of prefrontal dopamine neurones in behavioural blockade induced by controllable vs uncontrollable negative events in rats. Behav. Brain Res. 37:9-19;1990. 23. Robertson, A.; Laferriere, A. Disruption of the connections between the mediodorsal and suical prefrontal cmtices alters the associability of rewarding medial cortical simulation to place and taste stimuli in rats. Behav. Neurosci. 103:770-778; 1989. 24. Robertson, A.; Mogenson, G. J. Evidence for a role of dopamine in self-stimulation of the nucleus accumbens. Can. J. Psychol. 32: 67-76; 1978.
STRESS AND SELF-STIMULATION
25. Roth, R. H.; Tam, S-Y.; Ida, Y.; J.-X.; Deutch, A. Y. Stress and the mesocorticolimbic dopamine systems. Ann. NY Acad. Sci. 537: 138-147; 1988. 26. Thieny, A. M.; Tassin, J. P.; Blanc, G.; Glowinski, J. Selective activation of the mesocortical DA systems by stress. Nature 263: 242-243; 1976. 27. Zacharko, R. M.; Bowers, W. J.; Kokkinidis, L.; Anisman, H. Re-
229
gion-specific reductions in intracranial self-stimulation after uncontrollable stress: possible effects on reward processes. Behav. Brain Res. 9:129-141; 1983. 28. Zacharko, R. M.; Lalonde, G. T.; Kasian, M.; Anisman, H. Strainspecific effects of inescapable shock on intracranial self-stimulation from the nucleus accumbens. Brain Res. 426:164-168; 1987.
Research Bulletin, Vol. 27, pp. 225-229. o Pergmon
0361-9230191 $3.00 + .oO
Press plc, 1991. Printed in the U.S.A.
RAPID COMMUNICATION
Contrasting Effects of Stress on Medial and Sulcal Prefrontal Cortex Self-Stimulation IAIN S. MCGREGOR Department
of Psychology,
University of Sydney, NSW 2006,
Australia
Received 26 April 199 1 I. S. Contrasting effects of stress on medial and s&al prefrontal cortex self-stimulation. BRAIN RES BULL 27(2) 225-229, 1991.-Male Wistar rats were subjected to either 25 controllable or uncontrollable footshocks and then tested for changes in fixed-interval 5-second (n-5) self-stimulation of the medial prefrontal cortex (MPC), sulcal prefrontal cortex (SEC) or nucleus accumbens (NAS). Controllable footshock caused a moderate facilitation of MPC self-stimulation (30% above baseline rates) but inhibited SEC self-stimulation (32% below baseline rates). Uncontrollable footshock had no effect on MPC self-stimulation but inhibited SEC self-stimulation (52% below baseline rates). An inhibition of SF’C self-stimulation was also evident 24 hours following controllable or uncontrollable footshock. NAS self-stimulation was unaffected by footshock. Changes in locomotor activity were not consistently related to changes in self-stimulation following footshock. These results are discussed in terms of the different effects of mild stress on the release of reward-relevant neurotransmitters in the MEC, SEC and NAS. The possible role of stress-induced hypoalgesia in determining the stress-induced facilitation of MPC self-stimulation is also discussed.
MCGREGOR,
Stress Dopamine
Sulcal prefrontal cortex Medial prefrontal cortex Activity Hypoalgesia Fixed-interval
Nucleus accumbens
Self-stimulation
Rat
creases DA utilization in the MPC, but not in the SPC or NAS (4,5). Only with more severe stressors is increased DA utilization in all three regions seen (4,5). Both the NAS and SPC support self-stimulation. In both regions self-stimulation may be at least partly mediated by DA. SPC self-stimulation is decreased following lesions of ascending DA fibers from the ventral tegmental area (3) and is also inhibited by systemic injection of apomorphine (19). NAS self-stimulation is inhibited following intra-NAS injection of DA antagonists (24). If DA modulates both NAS and SPC self-stimulation then it might be predicted that stress will affect both NAS and SPC self-stimulation. However, since the effects of stress on DA in the NAS and SPC may be different to those seen in the MPC, we might predict contrasting effects of stress on self-stimulation of these three regions.
AN important observation in support of the “learned helplessness” model of depression is that uncontrollable stress impairs performance in a number of behavioral paradigms (17). It is somewhat surprising then that recent research from this laboratory has demonstrated a facilitation of medial prefrontal cortex (MPC) self-stimulation in rats subjected to such stress. Exposure to controllable or uncontrollable footshock, brief restraint or administration of the anxiogenic drug FG-7142 causes an immediate increase in MPC self-stimulation rates (12,13). A facilitation of MPC self-stimulation may be evident for up to 72 hours following a single exposure to stress (12). Stress enhances the utilization of the neurotransmitter dopamine (DA) in the MPC (1, 4, 5, 25, 26), and DA is thought to modulate the reinforcing efficacy of MPC self-stimulation (20). The stress-induced facilitation of MPC self-stimulation may thus reflect the increased release of DA in the MPC enhancing the reinforcing quality of MPC stimulation. Long-term effects of stress on MPC self-stimulation may reflect alterations in the sensitivity or number of MPC DA receptors (12,14). Stress also increases DA utilization in other forebrain regions, including the sulcal prefrontal cortex (SPC) and nucleus accumbens (NAS) (1, 4, 5, 25, 26). However, the effects of stress on DA in the SPC and NAS are different to those seen in the MPC. Firstly, the increases in DA utilization in the NAS and SPC are smaller than those seen in the MPC (4,5). Secondly, DA utilization in the SPC and NAS is not affected by milder stress, such as conditioned fear. For example, exposing a rat to an environment in which it was previously shocked in-
METHOD
Subjects, Surgery and Histology The subjects were 28 male Wistar rats weighing between 270 and 350 grams at the time of surgery. The rats were individually housed and kept in a temperature-controlled vivarium at 22 f 2°C on a 14: 10 hour light-dark cycle. Ad lib food and water were available in the home cages. The rats were stereotaxically implanted with monopolar stainless steel electrodes under ketamine-xylazine anaesthesia. The electrodes were aimed at the MPC (n = 8), SPC (n = 14) or NAS
225
(n =h). The coordinates (21) used were (mm): MPC =A: 4.2. L: 0.7. V: 3.5, SPC=A: 3.2, L: 3.5. V: 4.5 and NAS=A: 2.2, L: 1.5. V: 6.8. A stainless steel wire wrapped around two skull screws located 2 mm posterior to bregma served as the indifferent electrode. The electrode assembly was held in place with a cap made of dental acrylic cement. At the completion of the experiment the rats were sacrificed with an overdose of barbiturate. Immediately following death, an anodal current was passed through the electrode for 10 s to produce a small lesion at the tip of the electrode. Brains were rapidly removed from the cranial vault. frozen to - 12°C. and sliced on a microtome. The brain slices were then placed on glass slides and stained with Toluidine Blue. Electrode locations were determined microscopically with reference to the atlas of Paxinos and Watson (3-l). Procedure The self-stimulation methods and apparatus were as described previously (10, 12, 13). Testing occurred in four identical operant sound-attenuated chambers (37 x 25 x 32 cm) fitted with single levers. Stimulation consisted of 0.5 second trains of 100 Hz current-regulated cathodal rectangular pulses with a pulse width of 0.2 ms. The rats were trained for several weeks to acquire stable lever pressing for brain stimulation on a fixed-interval 5-second (FI-5) schedule of reinforcement. Use of this schedule minimizes the nonspecific effects of stimulation upon performance while retaining sensitivity to alterations in reinforcement efficacy caused by drugs or stress (10, 12, 13). Once the rats were stabilized on FI-5, rate-intensity functions were computed for each rat and individual currents adjusted to ensure that response rates were approximately half-maximal. For six consecutive days prior to being stressed, the rats were given baseline self-stimulation and locomotor activity tests. The test sessions consisted of a 20-minute period where the rats pressed for self-stimulation at half-maximal current under FI-5. The total number of responses in 20 minutes was measured. This was immediately followed by a 20-minute locomotor activity test. The rats were placed in Coulboum Test Cages (30 X 30 X 25 cm) which had clear acrylic front and back panels and metal sides and top. The floor of each cage consisted of a grid of sixteen metal bars connected through a high impedance amplifier so that contact or the breaking of contact between four evenly spaced bars and the other twelve produced a count that was recorded by computer. The total number of counts in 20 minutes was recorded. The rats within each group (MPC, SPC, or NAS) were paired according to their self-stimulation rates on the final baseline day. Pair members were allocated randomly to an uncontrollable or controllable shock condition. On the following day, footshock was administered prior to self-stimulation. Footshock was delivered in a two shuttle boxes (70 x 11 X 14 cm) in a separate room to where self-stimulation occurred. A variable resistor shock generator (Hayes) and scrambler were used to deliver shock of 0.6 mA intensity to the grid floors of the shuttle boxes. Rats in the controllable shock condition received 25 trials where shock could be terminated through performance of a FR-1 shuttle response. If no escape response was made within 60 seconds then the trial was automatically terminated. The intertrial interval was either 15, 30, 45 or 60 seconds with a randomized order of intervals assigned prior to testing. Rats in the uncontrollable shock condition were yoked to their controllably shocked partner and received shock whenever their partner did, but independently of any emitted response.
FIG. 1. Coronal sections ranging from 2.2 to 4.7 mm anterior to bregma [modified from the atlas of Paxinos and Watson (21)]. Electrodes placements are shown in the MPC (black dots), SPC (open circles) and NAS (black squares).
Immediately following footshock the rats were given a standard 20-minute self-stimulation session followed by a 20-minute activity test. Rats were also tested for self-stimulation and activity 24 hours following shock since previous results have indicated that effects of footshock on self-stimulation can be detected at this time (2, 12, 13, 27, 28). The effects of shock on selfstimulation and activity for each group were evaluated relative to scores on the last of the six baseline days using the r-test for paired samples. RESULTS
Histology The locations of the electrodes in the 28 rats completing the experiment are shown in Fig. 1. MPC electrodes were all within the prelimbic or infralimbic cortex [terminology according to Krettek and Price (9)J. SPC electrodes were either in the dorsal or ventral agranular insular cortex. NAS electrodes were mostly within the anterior NAS. Shock Avoidance The three rats in the NAS group receiving controllable shock showed faster escape latencies than the rats in the MPC or SPC groups (ps
221
STRESS AND SELF-STIMULATION
also shown little relation between
N
ulation following
+?
c MPC
3 UC
N
'P
LS
* P * :
c
i UC
FIG. 2. Mean change in self-stimulation rates (upper) and activity (lower) at 0 and 24 hours following shock in the MPC, NAS and SPC groups given controllable (C) or uncontrollable (UC) footshock (*p
236.5 2 28.98 bar presses/20 min. The lower rates for SPC relative to MPC self-stimulation agree with previous reports (10,23). The effects of controllable and uncontrollable footshock on self-stimulation are shown in Fig. 2. There was a clear difference seen across the three groups. Controllable shock facilitated MPC self-stimulation, inhibited SPC self-stimulation and had no effect on NAS self-stimulation. There was also a significant inhibition of SPC self-stimulation 24 hours after controllable shock. Uncontrollable shock inhibited SPC self-stimulation both immediately and 24 hours following shock. There was no significant effect of uncontrollable shock on MPC or NAS self-stimulation. Activity Activity was decreased relative to baseline in nearly every group tested (Fig. 2). The effect was variable across rats and consequently only some of the group effects were statistically significant. Activity changes were not systematically related to alterations in self-stimulation rats since some groups showing large declines in activity (e.g., the MPC and NAS groups given uncontrollable shock) showed no change or an increase in selfstimulation, while other groups showing lesser decreases in activity (e.g., the two SPC groups tested at 24 hours) showed a profound inhibition of self-stimulation. Previous studies have
activity changes and self-stimstress (2, 13, 27). DISCUSSION
The present experiment shows that footshock stress has different effects on self-stimulation of the MPC, SPC and NAS. The near-opposite effects of stress on MPC and SPC self-stimulation are particularly striking and add to previous reports that MPC and SPC self-stimulation have different characteristics. For example, using the conditioned taste and place preference paradigms, Robertson and Laferriere (23) showed that SPC selfstimulation has prepotent associability with gustatory cues, and MPC self-stimulation with contextual cues. Further, SPC but not MPC self-stimulation is facilitated by food deprivation (8,lO) and is associated with increases in food intake and body weight (10). These different characteristics of MPC and SPC self-stimulation agree with other reports of functional dissociations between the two regions [see (7) for review]. In general, the SPC seems to be primarily involved with feeding-related homeostasis (7,l l), while the MPC is involved in cognitive functions (including spatial mapping) and behavioral and autonomic response to stress (7,22). The effects of stress on MPC self-stimulation reported here partly agree with previous results (12,13). However, there are some differences. Firstly, in the present study, uncontrollable footshock had no effect on MPC self-stimulation while a large and significant facilitation was seen in an earlier study (13). However, this earlier study involved 60 rather than 25 uncontrollable shocks. Further, the shocks were of a longer duration due to a IX-2 escape contingency in the controllable shock group (13). This suggests that only a relatively high density of uncontrollable shock will facilitate MPC self-stimulation. A second difference is the absence of a facilitation of MPC self-stimulation 24 hours following stress. In two previous studies a facilitation of MPC self-stimulation was seen at 24 hours (12,13). This discrepancy may also relate to the comparatively low density of shock used in the present study. In particular, the density of shock used here is considerably less than that required to produce “learned helplessness” (15,17), which is characterized by behavioral effects present 24 hours following stress (17). One possible explanation of the stress-induced facilitation of MPC self-stimulation (other than simply citing increased DA release) invokes stress-induced hypoalgesia. MPC stimulation appears to have an aversive component, since rats will learn a response to escape from noncontingent trains of MPC stimulation (18). However, if a rat were hypoalgesic, then this might mask any aversive properties of MPC stimulation leading to a net increase in positive reinforcement and increased self-stimulation. High density uncontrollable footshock produces a longlasting context-independent hypoalgesia (6, 15, 16). Lower density shock (such as was used in the present study) produces only a brief hypoalgesia that is context-dependent and dissipates rapidly (15,16). This would predict that the rats given uncontrollable stress in our previous study (13) (where a high density of shock was used) were hypoalgesic when tested for self-stimulation. This may explain the facilitation of self-stimulation seen in these rats. In contrast, rats given uncontrollable shock in the present study would probably not be hypoalgesic when tested for selfstimulation due to the low density of shock given. This may explain the lack of facilitation of self-stimulation in these rats. Clearly this hypothesis is speculative, but it perhaps deserves some further investigation. The main problem would be in explaining the facilitation of MPC self-stimulation produced by controllable shock, since it is though to produce less hypoalge-
sia than uncontrollable shock [e.g., (6)]. The inhibition of SPC self-stimulation seen following stress in the present study was largely unexpected, since it was thought that the increased DA utilization produced by stress in the SPC might facilitate SPC self-stimulation. It is unlikely that the inhibition observed reflects “learned helplessness” since (as noted above) the density of shock used is insufficient to produce such an effect. Further, rats in the SPC group given controllable shock [who would not by definition be “helpless” (17)] showed a similar inhibition of self-stimulation to the rats in the uncontrollable shock group. The lower baseline rates of serf-stimulation in the SPC group might be thought to imply that SPC self-stimulation would be more easily disrupted by stress. However. the best responding rats in the SPC group (who had baseline rates of self-stimulation that were greater or similar to rats in the MPC and NAS groups) showed just as profound an inhibition of self-stimulation than the lower responding rats. It seems then that stress may have a powerful inhibitor effect on the release of some reward-relevant neurotransmitter in the SPC. At present the identity of this transmitter is unknown. The lack of effect of stress on NAS self-stimulation seen here contrasts with previous reports of an inhibitory effect of stress on NAS self-stimulation (2,27). However, several factors could explain this discrepancy. Firstly, previous studies used mice and involved a much higher density of shock than was used in the present study (227). Further, although the effect of shock on NAS self-stimulation in mice is usually inhibitory. mice of the
BALBic strain display a facilitation of NAS self-stimulation following shock (28). Since only more intense stressors increase DA utilizatton in the NAS of rats (5) it could be that a facilitation of NAS selfstimulation would be seen in rats exposed to greater levels of stress than were used in the present study. Certainly. the tendency toward a facilitation in the NAS group given uncontrollable shock (Fig. 2) suggests that this may be the case. Alternatively. it could be argued that DA is not an important determinant of NAS self-stimulation and this explains the lack of significant effect on NAS self-stimulation in the present study. It is not possible. at present, to determine whether this is the case. In conclusion, the exact mechanisms underlying the contrasting effects of stress on MPC, SPC and NAS self-stimulation are uncertain. Future studies looking at the effects of different densities of shock on self-stimulation of the MPC. SPC and NAS and investigating the pharmacological reversal of stressor effects may prove instructive in unravelling the precise mechanisms involved. The important conclusion from the present study is that stress has con~asting effects on MPC compared to SPC selfstimulation. This further suggests that these two subregions of the rodent prefrontal cortex have different functional attributes. XKNOWLEDGEMENTS This research was supported by a University of Sydney Postgraduate Schol~hip to Iain S. McGregor. Dr. Dale Atrens is thanked for providing facilities used in this research. Dr. H. Q. Lin is thanked for his excellent technical assistance.
REFERENCES 1. Abercrombie, E. D.; Keefe, K. A.; DiFrischia. D. S.; Zigmond. M. J. Differential effect of stress on in vivo dopamine r&ease in striatum, nucleus accumbens, and medial prefrontal cortex. .I. Neurochem. 52:1655-1658; 1989. 2. Bowers, W. J.; Zacharko, R. M.; Anisman, H. Evaluation of stressor effects on intracranial self-stimulation from the nucleus accumbens and the substantia nigra in a current intensity paradigm. Behav. Brain Res. 2385-93; 1987. of the sulcal pre3. Clavier, R. M.; Gerfen, C. R. Self-stimulation frontal cortex in the rat: direct evidence for ascending dopaminergic mediation. Neurosci. Lett. 12:183-l 87; 1979. 4. Dunn, A. J. Stress-related activation of cerebral dopaminergic systems. Ann. NY Acad. Sci. 537:188-205; 1988. 5. Herman, J. P.; Guillonneau, D.; Dantzer, R.; Scatton, B.; Semerdijian-Rouquier, L.; Le Meal, M., Differential effects of inescapable footshocks and of stimuli previously paired with inescapable footshocks on dopamine turnover in cortical and limbic areas of the rat. Life. Sci. 302207-2214: 1982. 6. Jackson, R, L.: Maier, S. F.; Coon, D. J. Long-term analgesic effects of inescapable shock and learned helplessness. Science 206: 91-92; 1979. 7. Kolb. B. Functions of the frontal cortex of the rat: a comparative review. Brain Res. Rev. 8:65-98; 1984. 8. Koolhaus, J. M.; Mora, F.; Phillips, A. G. Effects of food and water dep~vation on self-stimulation of the medial and sulcal prefrontal cortex and caudate m&men in the rat. Physiol. Behav. 18:329331; 1977. 9. Krettek, J. E.; Price, J. L. The cortical projections of the mediodorsal nucleus and adjacent thalamic nuclei in the rat. J. Comp. Neurol. 171:157-192; 1977. IO. McGregor, I. S.; Atrens, D. M. Prefrontal cortex self-stimulation and energy balance. Behav. Neurosci.; in press; 1991. 11. McGregor, I. S.; Menendez, .I. A.; Atrens, D. M. Metabolic effects obtained from excitatory amino acid stimulation of the sulcal prefrontal cortex. Brain Res. 529:1-6; 1990. 12. McGregor, 1. S.; Atrens, D. M. Stressor-like effects of FG-7142 on medial prefrontal cortex self-stimulation. Brain Res. 516: 170-
174; 1990. 13. McGregor, I. S.; Balleine, B. W.; Atrens, D. M. Footshock stress facilitates self-stimulation of the medial prefrontal cortex but not the lateral hypothalamus in the rat. Brain Res. 490:397-403; 1989. 14. MacLennan, A. J.; Pelleymounter, M. A.; Atmadja, S.; Jakubovic, A.; Maier, S. F.; Fibiger, H. C., D, dopamine receptors in the rat prefrontal cortex: characterization and alteration by stress. Brain Res. 477:3OC-307; 1990. 15. Maier, S. F. Diazepam modulation of stress-induced analgesia depends on the type of analgesia. Behav. Neurosci. 104:339-347; 1990. 16. Maier, S. F. Determinants of the nature of environmentally induced hypoalgesia. Behav. Neurosci. 104:131-143; 1989. 17. Maier, S. F.; Seligman, M. E. P. Learned helplessness: theory and evidence. J. Exp. Psychol. (Gen). 105:3-46; 1976. 18. Misrendino, M. J. D.; Coons, E. E. Electrical stimulation of the medial prefrontal cortex supports both ‘pure reward’ and ‘rewardescape’ behaviour in rats. Brain Res. 483:22&232; 1989. 19. Mora, F.; Phillips, A. G.; Koolhaus, J. M.; Rolls, E. T. Prefrontal cortex and neostriamm self-stimulation in the rat: Differential effects of produced by apomorphine. Brain Res. Bull 1:421-424; 1976. pathways and circuits 20. Mora, F.; Ferrer, J. M. R. Neurotransmitters, as the neural substrates of self-stimulation of the prefrontal cortex: facts and speculations. Behav. Res. 22:127-l% 160. 21. Paxinos. G.; Watson, C. The rat brain in stereotaxic coordinates, 2nd ed. Sydney: Academic Press; 1986. 22. Ravard, S.; Camoy, P.; Herve, D.; Tassin, J.-P., Thiebot, M.-H.; Soubrie, P. Involvement of prefrontal dopamine neurones in behavioural blockade induced by controllable vs uncontrollable negative events in rats. Behav. Brain Res. 37:9-19;1990. 23. Robertson, A.; Laferriere, A. Disruption of the connections between the mediodorsal and suical prefrontal cmtices alters the associability of rewarding medial cortical simulation to place and taste stimuli in rats. Behav. Neurosci. 103:770-778; 1989. 24. Robertson, A.; Mogenson, G. J. Evidence for a role of dopamine in self-stimulation of the nucleus accumbens. Can. J. Psychol. 32: 67-76; 1978.
STRESS AND SELF-STIMULATION
25. Roth, R. H.; Tam, S-Y.; Ida, Y.; J.-X.; Deutch, A. Y. Stress and the mesocorticolimbic dopamine systems. Ann. NY Acad. Sci. 537: 138-147; 1988. 26. Thieny, A. M.; Tassin, J. P.; Blanc, G.; Glowinski, J. Selective activation of the mesocortical DA systems by stress. Nature 263: 242-243; 1976. 27. Zacharko, R. M.; Bowers, W. J.; Kokkinidis, L.; Anisman, H. Re-
229
gion-specific reductions in intracranial self-stimulation after uncontrollable stress: possible effects on reward processes. Behav. Brain Res. 9:129-141; 1983. 28. Zacharko, R. M.; Lalonde, G. T.; Kasian, M.; Anisman, H. Strainspecific effects of inescapable shock on intracranial self-stimulation from the nucleus accumbens. Brain Res. 426:164-168; 1987.