- Email: [email protected]
Effect of cocaine injections on the neuroendocrine response to the serotonin agonist MK-212
258
BIOL PSYCHIATRY 1992;32:258-269
Effect of Cocaine Injections on the Neuroendocrine Response to the Serotonin Agonist MK-212 Louis D. Van de Kar, Peter A. Rittenhouse, Patricia O'Connor, Thomas Palionis, Mark S. Brownfield, Stephanie J. Lent, Molly Cames, and Cynthia L. Bethea
This study was undertaken to examine whether several of the hormones that can be released by activation of serotonin receptors will be affected by long-term cocaine administration. Male rats received cocaine injections (15 mglkg, IP) twice daily for 7 days. Forty.two hr after the last cocaine injection, the rats were challenged with increasing doses (0, 1, S, 10 mg/kg, IP) of the S-HT1/5-HT2 agonist MK-212 (6-chloro-2-[l-piperazinyl]-pyrazine). The following observations were made: (1) cocaine reduced the rate of body weight gain; (2) cocaine inhibited the stimulatory effect of MK-212 on plasma vasopressin, ozytocin, and prolactin concentrations and on plasma renin activity and concentration; (3) cocaine did not inhibit the stimulatory effect of MIC.212 on plasma ACTH or corticosterone concentrations. The data indicate that a wide.spectrum S-HT (serotonin) agonist such as MK-212 con reveal differential neuroendocrine responses. This effect could be related to cocaine-induced changes in the different 5-HT receptor subtypes that regulate the secretion of these hormones.
Introduction Cocaine is a psychostimulant that has a considerable influence on monoaminergic neurotransmission in the brain. In addition to dopamine and norepinephrine uptake sites, cocaine binds with high affinity to serotonin (5-HT) synaptic reuptake sites (Reith et al 1983; Ritz et al 1990). This prevents the reuptake of these neurotransmitters from the synaptic cleft and leads to increased concentration of 5-HT (as well as dopamine and norepinephrine) in the synapse. Acute injection of cocaine increases ACTH secretion but reduces prolactin and lenin secretion (Borowsky and Kuhn 1991a; Lakoski et al 1991). The effect of cocaine on adrenocorticotropin (ACTH) secretion is partly mediated by 5-HT2 receptors (Borowsky and Kuhn 1991b; Levy et al 1991). Therefore, it is important
From the Departmentof Pharmacology,Stritch Schoolof Medicine, Loyola Universityof Chicago, Maywood, IL (LDV, PAR, PO); the Departmentof ComparativeBiosciences,School of VeterinaryMedicine, Universityof Wisconsin, Madison, Wi (TP, MSB); the Middleton Memorial Veterans Hospital, Geriatric Section, Madison, WI (STL, MC); and the Oregon Regional Primate ResearchCenter, Beaverton,OR (CLB). Address reprint requests to Louis D. Van de Kar, Ph.D., Department of Pharmacology,Stritch School of Medicine, Loyola University of Chicago, 2160 S. First Avenue, Maywood, IL 60153. Received May 16, 1991; revised October 28, 1991.
© 1992 Society of BiologicalPsychiatry
0006-3223/92/$05.00
Cocaine and Neuroendocfine Response to MK-212
BIOLPSYCHIATRY 1992;32:258-269
259
to know if chronic exposure to cocaine can influence the serotonergic regulation of neuroendocrine function. Knowledge of the basic mechanisms that cause cocaine's deleterious effects may contribute to the development of therapy and prevention of its abuse. Cocaine abusers often experience post-abuse anhedonia and depression, leading to renewed cocaine selfadministration (Nunes et al 1989; Trinkoffet al 1990). Decreased function of serotonergic neurons is considered to be, at least in part, a cause of endogenous depression (Meltzer and Lowy 1987). Little agreement exists about the long-term effects of cocaine on serotonergic neurons. Two studies found that repeated injection of cocaine reduced the 5-HT content in the brain (Roy et al 1978; Taylor and Ho 1977). However, in another study (Kleven et al 1988), no change in the 5-HT content in several brain regions was found after injection of a lower dose of cocaine for 10 days. Thus, additional studies are necessary to establish the long-term effects of cocaine on serotonergic function. The present study utilized the hormonal responses to a 5-HTI/5-HT2 agonist as indices of the function of serotonin receptors. Serotonergic neurons stimulate, either directly or indirectly, the release of many hormones including ACTH, corticosterone, prolactin, lenin, oxytocin, and vasopressin (for review see Van de Kar 1991). The 5-HT neurons that regulate neuroendocrine function are part of an extensive 5-HT pathway that also sends collaterals to limbic areas (Moliver 1987; Imai et al 1986). Therefore, changes in 5-HT neurons that influence behavior can be reflected in altered neuroendocrine responses to challenges with 5-HT agonists (Van de Kar 1989). Distinct 5-HT receptor subtypes mediate the serotonergic stimulation of these hormones. Activation of 5-HTtA receptors increases ACTH and corticosterone but none of the other hormones (Brownfield et al 1988; Koenig et al 1987; Lesch et al 1989, 1990; Lorens and Van de Kar 1987; Saydoff et al 1991; Van de Kar et al 1989a,b). Activation of 5-HT~c receptors also increases ACTH and corticosterone secretion (King et al 1989). On the other hand, activation of 5-HT2 receptors increases oxytocin, vasopressin, and renin secretion (Alper and Snider 1987; Alper 1990; Brownfield et al 1988; Lorens and Van de Kar 1987; Saydoff et al 1991; Van de Kar et al 1990b). Prolactin secretion is most likely mediated by activation of 5-HTle recepto~ b, rats (Van de gar et al 1989a,b). The present study focused on the influence of cocaine on the 5-HT receptors that stimulate these hormones. The 5-HTI/5-HT2 agonist and uptake inhibitor (6-chloro2-[I-piperazinyl]-pyrazine) (MK-212) (Clineschmidt 1979; Ciineschmidt et al 1978) has been previously tested in humans and was shown to increase serum prolactin and cortisol levels (Lowy and Meltzer, 1988). Furthermore, the effect of MK-212 on plasma prolactin and cortisol was blunted in patients with obsessive-compulsive disorder (Bastani et al 1990). Therefore, MK-212 was used as a challenge after injection of cocaine to rats for 7 days.
Methods Animals Male Sprague-Dawley rats (200-250 g) were purchased from Sasco-King Animal Laboratories (Oregon, ~'!). The rats were housed, two per cage, in a temperature, humidity, and illumination (12:12 hr light/dark cycle; lights on at 7 AM) controlled room. Water and food (Wayne Lab Blox, Lal~ Mills Inc., Chicago, IL) were available ad libitum. The
260
BIOLPSYCHIATRY
L.D. Van de Kar et al
1992;32:258-269
rats were killed by decapitation in an area outside the rat room. All the experimental protocols were approved by Loyola University Animal Care and Use Committee.
Drugs All drugs were dissolved in saline and injected in a volume of 1 ml/kg, IP. All cocaine injections (15 mg/kg, IP) were performed twice a day (8:30 AM and 5 PM), a dosage regimen that was previously reported to reduce brain serotonin content (Roy et al 1978). All subsequent challenges with a 5-HT agonist occurred 42 hr after the last cocaine injection. This interval was chosen to ensure complete elimination of cocaine from the circulation, and thus avoid its interaction with the 5-HT agonist. Cocaine HCI was donated by the National Institute for Drug Abuse (NIDA, RockviUe, MD) and MK-212 (6-chloro2-[l-piperazinyl]-pyrazine) was donated by Merck Sharp and Dohme (West Point, PA). MK-212 was dissolved in saline and injected in doses of 1, 5, and 10 mg/kg (ip). The rats were killed by decapitation 30 min after the injection of MK-212. All drug doses are expressed as the salt. Biochemical Determinations
Preparation of Plasma Trunk blood was collected in centrifuge tubes containing a 0.5 ml solution of 0.3 M ethylenediamine tetraacetic acid (EDTA; pH 7.4). The blood was centrifuged at 1000 x g for 20 rain at 4°C and the plasma was saved. Plasma from each rat was divided into aliquots of (1) 1.0 ml for the determination of vasopressin and oxytocin, (2) 1.0 ml for the determination of plasma renin activity (PRA), (3) 50 p,! for the determination of plasma renin concentration (PRC), (4) 30 I~l for the determination of corticosterone levels, (5) 200 p.l for the determination of plasma ACTH, and (6) 0.2 ml for the determination of prolactin levels. All plasma aliquots were stored at - 40°C until the respective hormones were determined.
Plasma VasopressinRadioimmunoassay This assay has been described in detail elsewhere (Brownfield et al 1988). The assay is a double antibody assay conducted under disequilibrium conditions. Rabbit anti-lysine vasopressin, generated in Dr. Brownfield's laboratory, and the tracer arginine vasopressin (AVP) were used in this assay. The measurement of plasma samples requires that they be extracted, a procedure detailed elsewhere (Brownfield et al 1988); the sensitivity limit is 0. I pg/tube. The antibody has less than 0.01% cross-reactivity with oxytocin and vasotocin. The intraassay variability is 4.5% and the interassay variability is 8% (Brownfield et al 1988).
Plasma Oxytocin Radioimmunoassay This assay is identical to the vasopressin assay described above and was described in detail elsewhere (Saydoff et al 1991); it also is a double-antibody assay conducted under disequilibrium conditions. Rabbit anti-oxytocin obtained from Hoechst-Boehring (San
Cocaine and Neuroendocrine Response to MK-212
BIOLPSYCHIATRY 1992;32:258-269
261
Diego, CA) was generated in rabbits. Synthetic oxytocin was obtained from Bachem Fine Chemicals (Torrance, CA) with purity of the peptide established by thin layer chromatography, amino acid analysis, and high-pressure liquid chromatography. The measurement of plasma samples requires that they are extracted, and this procedure is detailed elsewhere (Brownfield et al 1988). The antibody has less than 0.01% cross-reactivity with AVP and vasotocin. The intraassay and interassay variabilities are 5%, and 9%, respectively.
Plasma Renin Activity (PRA ) PRA was measured by radioimmunoassay for generated angiotensin (ANG I) as previously described (Van de Kar et al 1989a). In this assay, renin generates ANG 1 from the endogenous renin substrate. Therefore, PRA is a composite of the plasma concentration of renin and renin substrate. The radioimmunoassay for AN(3 I is described below.
Plasma Renin Concentration (PRC) In this assay, a saturating concentration of renin substrate is added to the plasma to allow generation of ANG 1 at maximal velocity. Thus, this assay reflects the total concentration of the enzyme renin in plasma, independent of the concentration of renin substrate. In the PRC assay, renin substrate is obtained from plasma of rats that were nephrectomized and received a dexamethasone injection (0.2 rag/rat) 24 hr before sacrifice. The details of this assay were described by us ebewhere (Richardson Morton et al 1989). The radioimmunoassay of ANG I was perfozmed with antiserum at a dilution of 1:16,000 and a total binding of 35% as previously described (Richardson Morton et al 1989). The sensitivity limit of the RIA was 10 pg ANG I per tube and the intraassay variability was 4.4%. The interassay variability was 12.6%.
Piasn:J Corticosterone Radioimmunoassay Corticosterone radioimmunoassay was performed on unextracted plasma samples (2 and 5 I~i) in which binding proteins have been denatured by boiling, as previously described (Richardson Morton et ai 1989), using procedures and antiserum from Radioassay Systems Laboratories (Carson, CA). The sensitivity limit is 0.02 ng/tube and the intraassay and interassay variabilities were 4.5% and 11.9%, respectively.
Plasma ACTH Radioimmunoassay ACTH radioimmunoassay on unextracted plasma samples (20-50 Ixl) was performed as previously described (Carnes et al 1986). Briefly, the ACTH antiserum was obtained from lgG Corp (Nashville TN). ACTH standards (1-39) were obtained from Calbiochem and ~2SI-ACTH from INCSTAR (Stillwater, MN). The sequence recognition of the antiserum is 5-18. In addition, this antiserum does not significantly recognize a-MSH, [3-MSH, [3-endorphin, [3-1ipotropin, ACTH ! 1-24, or ACTH 1-16-amide. The minimum detectable concentration is 0.25 pg/tube and the intraassay and interassay variations were 4.2% and 14.6%, respectively.
262
BIOL PSYCHIATRY 1992;32:258-269
L.D. Van de Kar et al
Plasma Prolactin Radioimmunoassay Prolactin radioimmunoassay was performed with reagents provided by the National Institute of Arthritis, Diabetes, Digestive and Kidney Disorders (NIADDK). Anti-rat prolactin serum S-8 was used at a dilution of 1:5000, as described previously (Van de Kar and Bethea 1982). Briefly, NIADDK rat prolactin (preparation rPRL-I-5) was used for iodinated tracer, and NIADDK rat prolactin (preparation rPRL-RP-3) was used as the reference preparation. The intraassay variability was 6.8% and all the samples from one experiment were determined together in one assay.
Statistics The data are represented as the group means and the standard errors of the mean (SEM). Statistical analysis of the data was performed by two-way analysis of variance (ANOVA), and individual group means were compared by Student-Newman-Keuls' test (Steel and Torrie 1960) using a computer program (NWA STATPAK, Portland OR).
Results Injection of cocaine (15 mg/kg, IP) for 7 days produced a significant reduction in the rate of body weight gain. The body weights on the last day were 291 +- 4.1 g for the vehicle group and 278 - 3.8 g for the cocaine group. The statistical analysis revealed a significant increase in body weigh (Fis,a66t = 56.43, p < 0.00001), a significant effect of cocaine (FI~,.~661 = 9.11, p < 0.0026), and a significant cocaine x weight gain interaction (F[5.a661 -- 2.31, p < 0.044). A posthoc Newman Keuls analysis revealed that the significant difference from the saline-treatod rats were observed in cocaine-treated rats that were weighed 42 hr after the last cocaine injection. The basal levels of the hormones measured were not altered in the cocaine-treated rats. Injection of MK-212 (l, 5, or l0 mg/kg, IP) to the vehicle-treated rats dose dependently increased the plasma concentration of all hormones except vasopressin and renin which only increased after injection of the highest dose. The effect of MK-212 on plasma prolactin concentration was significantly inhibited in the cocaine-injected rats (Figure 1). The two-way ANOVA showed a significant cocaine x MK-212 interaction (Fi3..~41 = 7.138, p < 0.01). Similarly, the effect of the highest dose of MK-212 on oxytocin was inhibited in cocaine pretreated rats (Figure 2). The two-way ANOVA did not reveal a significant cocaine x MK-212 interaction (FD,471 ---1.36, NS). However, a subsequent comparison of the group means using Newman Keuls' test revealed that plasma oxytocin concentration in cocaine pretreated rats that received the highest dose of MK-212 (10 mg/kg) were significantly lower than those of vehiclepretreated rats that received the same dose of MK-212 (Figure 2). In the case of vasopressin and renin, only the highest dose of MK-212 significantly increased their plasma concentrations (Figures 2 and 3). The effect of MK-212 on plasma vasopressin was abolished in cocaine-treatod rats (Figure 2). The two-way ANOVA indicated a cocaine × MK-212 interaction (Ft3,s~I = 2.579, p < 0.0516) that was not significant, but a post-hoc Newman Keuls' test revealed that the MK-212 (10 mg/kg)induced increase in vasopressin was significantly blocked in cocaine-treated rats. The effect of MK-2 i 2 on plasma renin activity and concentration was inhibited by pretreatment with cocaine (Figure 3). The two-way ANOVA revealed a significant cocaine x MK-
Cocaine and NeuroendocrineResponse to MK-212
BIOL PSYCHIATRY
263
1992;32:258-269
# o SALINE
270
Figure 1. MK-212 (0, I, 5, 10 mg/kg, IP) induced increase in plasma prolactin concentration in rats treated with cocaine (15 mg/kg, twice daily) for 7 days. Data represent mean --. SEM of eight rats per group. **Significant difference from the corresponding saline (0 dose of MK-212) group, p < 0.01 (2-way ANOVA and Newman Keuls' test). #Significant difference from the corresponding group injected with the same dose of MK-212, p < 0.05 (2-way ANOVA and Newman Keuls' test).
240
l=o l=
r-U Z
~.I 0= I~ .,¢ r~
COCAINE
21o
**t I
18o 150 120 90
_/
60
Ill
0
o
.
.
.
.
.
_...,,,~ .lip
1
. . . . . . . .
I
lO
DOSE OF MK-2t2 (mg/kg IP)
# 120
•
i
COCAINE
"
0i
tlo
r,o
30
o
A
i
i
. . . .
|
. . . .
1
iI
lo
# 4O
l
u~ r~ rA
g
.
,< ,< @
-
-
-
. . . .
O~
1
~
"
1o
DOSE OF M K - 2 1 2 ( m g / k g I~')
Figure 2. MK-212 (0, I, 5, I0 mg/kg, IP) induced increase in plasma oxytocin (top) and vasopressin (bottom) in rats treated with cocaine (15 mg/kg, twice daily) for 7 days. Data represent mean --- SEM of eight rats per group. Significant difference from the corresponding saline (0 dose of MK-212) group, *p < 0.05, **p < 0.01 (2-way ANOVA and Newman Keuls' test). ~Significant difference from the corresponding group injected with the same dose of MK-212, p < 0.05 (2-way ANOVA and Newman Keuls' test).
264
BIOL PSYCHIATRY 1992;32:258-269
L.D. Van de Kar et al
#
40
o SALINE
35
*
4, COCAINE 25
z~
2o "~ 10
D,, 0
i!
0
I
10
# ==
i.
.o
6O
Figure 3. MK-212 (0, !, 5, 10 mg/kg, IP) induced increase in plasma lenin activity (top) and concentration (bottom) in rats treated with cocaine (15 mg/kg, twice daily) for 7 days. Data represent mean - SEM of eight rats per group. *Significant difference from the corresponding saline (0 dose of MK-212) group, p < 0.05 (2way ANOVA and Newman Keuls' test). #Sigaificant differencefrom the correspondinggroup injected with the same dose of MK-212, p < 0,05 (2-way ANOVA and Newman Keuls' test).
m
=.
0
0
1
10
DOSE OF t d K - 2 t 2 (mg/kg, ip)
212 interaction for both plasma renin activity (Fp.s,,I = 2.90, p < 0.043) and plasma lenin concentration (Ft,.s21 -- 2.96, p < 0.041). In contrast with the above-mentioned hormones, there was no difference in the doseresponse curve effects of MK-212 on plasma ACTH and corticosterone between cocaine and vehicle-pretreated rats (Figure 4). The two-way ANOVA exhibited a cocaine × MK212 interaction that was not significant for either ACTH (Fp.s41 = 1.73, p < 0.16) or plasma corticosterone (Fp.s~j = 0.678, p < 0.56). Furthermore, a post-hoe Newman Keuls' test revealed no difference between the cocaine and saline-treated groups for any of the MK-212 doses.
Discussion The results of this study suggest that cocaine injections for 7 days can differentially alter the subsequent hormonal responses to a direct 5-HT agonist. Though the effect of MK212 on several hormones (vasopressin, oxytocin, prolactin, and renin) was blunted after cocaine injections, the hypothalamic-pituitary-adrenal (HPA) axis was spared. This raises the question, What is the difference between the HPA axis and the other neuroendocrine systems? The most obvious difference is in the nature of the 5-HT receptor subtypes that control the secretion of these hormones. Activation of 5-HT~A receptors only increases ACTH
Cocaine and Neuroendocrine Response to MK-212
BIOLPSYCHIATRY
265
1992;32:258-269
go0 70o
*
COCAINE
~'~ t
600 ~[~ 500 ,~ 400 ~:~ 300 200 I~
~
Ioo 0
. . . . . . . .
0
I
illll
1
30 ~
. . . .
,
10
*
zs
0
I
Figure 4. MK-212 (0, I, 5, 10 mg/kg, IP) induced increase in plasma ACTH (top) and corticostemne (bottom) in rats treated with cocaine (15 mg/kg, twice daily) for 7 days. Data represent mean .+SEM of eight rats per group. *Significant difference from the corresponding saline (0 dose of MK212) group, p < 0.05 (2-way ANOVA and Newman Keuls' test).
10
DOSE OF MK-212 (mg/kg, ip)
and corticosterone secretion (Koenig et al 1987; Lorens and Van de Kar 1987). Activation of 5-HTIA receptors does not seem to increase the secretion of vasopressin, renin, or prolactin (Brownfield et al 1988; Lorens and Van de Kar 1987; Sayoff et al 1991; Van der Kar et al 1989a; Van de Kar et al 1990b). MK-212 is a 5-HTI and 5-HT2 agonist (Conn and Sanders-Bush 1985; Cunningham et al 1986). Its affinity for the various 5-HT receptor subtypes is not well described, though it is considered a low affinity and high efficacy agonist (Ciineschmidt 1979; Clineschmidt et al 1978) with a moderate affinity for 5-HT~A, 5-HT~e, and 5-HT~c receptors and a lower affinity for 5-HT2 receptors (Hoyer 1988). MK-212 was previously shown to increase lenin and vasopressin secretion by activation of 5-HT2 or 5-HT~c receptors (Brownfield et al 1988; Lorens and Van de Kar 1987). The effect of the high dose of MK-212 (10 mg/kg, IP) on both renin and vasopressin secretion was completely blocked by low doses (0.1-0.3 mg/kg) of the 5-HT,c/5-HT2 antagonists LY53857 and ritanserin (Brownfield et al 1988; Lorens and Van de Kar 1987). These observations suggest that even at this high dose, MK-212 activates mainly 5-HT~c/5-HT2 receptors to stimulate renin and vasopressin secretion. Hence, the cocaine-induced inhibition of the effect of MK-212 on renin and vasopressin secretion suggests a decline in the function of 5-HTic or 5-HT2 receptors. The influence of MK-212 on oxytocin secretion is at least partly mediated by 5-HT2 or 5-HT~c receptors as it was inhibited by ritanserin (Saydoff et al 1991). With respect
266
BIOLPSYCHIATRY
L.D. Van de Kar et al
1992:32:258-269
to prolactin secretion, blocking 5-HT~c or 5-HT2 receptors with LY53857 only partly inhibited the effect of MK-212 on prolactin secretion, suggesting that not only 5-HT1c/ 5-HT2 but also other 5-HT receptors mediate the effect of MK-212 on prolactin (Van de Kar et al 1989b). The MK-212-induced stimulation of ACTH and corticosterone secretion is subject to debate, but most data suggest that it is mediated by activation of either 5-HTIA or 5-HT~c receptors (King et al 1989; Koenig et al 1987). Thus, the data suggest that the effect of MK-212 on most hormones is mediated, at least in part, by either 5HT~c or 5-HT2 receptors. In addition, the effect of MK-212 on the HPA axis could be partly mediated by activation of 5-HT~,,, receptors. Along this line of reasoning, the repeated injection of cocaine could have a more profound effect on 5-HTtc or 5-HT2 receptors than on 5-HT~A receptors. We previously found that repeated injection of cocaine reduces the function of scrotonergic nerve terminals that stimulate ACTH and corticosterone secretion, because they inhibited the effect of the 5-HT releaser p-chloroamphetamine on plasma ACTH and corticosterone concentrations (Van de Kar et al 1990a, 1992). In contrast, pretreatment with cocaine potentiated the corticosterone response to the direct 5-HT agonist RU 24969 whereas the corticosterone response to another 5-HT agonist, m-CPP, was unaltered (V an de Kar et al 1992). Hence, the current data with MK-212 arc consistent with previous data using other direct 5-HT agonists. Taken together, the data suggest that cocaine does not alter all the 5-HT receptor subtypes that stimulate ACTH secretion. Another interpretation of the data could be that the effect of cocaine on serotonergic neurons is more profoundly focused on the nerve terminals and that not all serotonergic nerve terminals are affected in the same manner. Cocaine binds with high affinity to the 5-HT (and also dopamine and norepinephrine) uptake sites on nerve terminals (Reith et al 1983: Ritz et al 1990). The binding of cocaine to the 5-HT uptake site prevents its reuptake from the synaptic cleft which leads to increased concentration of 5-HT in the synapse. Acute IV injection of cocaine suppresses the firing rate of serotonergic neurons in the dorsal raphe nucleus via activation of somato-dendritic 5-HT~^ receptors (Pitts and Marwah 1986; Lakoski and Cunningham 1988; Cunningham and Lakoski 1988). Thus, repeated administration of cocaine could alter the function of neurons originating in the dorsal raphe nucleus, and their nerve terminals in the forebrain. MK-212 is not only a 5-HT agonist but also a 5-HT uptake inhibitor (Clineschmidt et al 1978). Thus, part of its endocrine effects could be mediated by the increase of synaptic 5-HT concentration as a consequence of 5-HT uptake inhibition. There is evidence that the serotonergic neurons that regulate neuroendocrine function are not homogenous. The serotonergic neurons that stimulate prolactin and lenin secretion arc located in the midbrain dorsal raphe nucleus (Barofsky et al 1983; Van de Kar and Bethea 1982; Van de Kar et al 1982). In contrast, the serotonergic neurons that regulate ACTH/corticosterone secretion are neither in the dorsal raphe, nor in the median raphe nucleus (Van de Kar et al 1982). Furthermore, the serotonergic fibers that stimulate vasopressin secretion might be ascending fibers whose cell bodies arc located caudal to the hypothalamus. This information comes from knife cuts that interrupted ascending fibers to the hypothalamus and blocked the effect of a 5-HT releasing drug on plasma vasopressin and prolactin concentrations and renin activity, but not on plasma corticosterone concentration (Karteszi et al 1982; Brownfield et al 1988; Van de Kar et al 1985). Thus, the different serotonergic fibers that stimulate the secretion of the respective hormones could underlie their divergent response to MK-212 after pretreatment with cocaine. In conclusion, the data suggest that repeated administration of cocaine has a profound
Cocaine and Neuroendocrine Response to MK-212
BIOLPSYCHIATRY 1992;32:258-269
267
inhibitory effect on serotonergic neuronal function in rats. The most likely receptor subtypes affected are 5-HT2 or 5-HT~c receptors, and the most likely neurons affected are the ones originating from the dorsal raphe nucleus. The data also suggest a need to test whether decreased serotonergic function also occurs in human cocaine abusers. MK212 seems to be a useful diagnostic tool in humans as it increases plasma prolactin and cortisol, an effect that was blunted in patients with obsessive-compulsive disorder (Bastani et al 1990). Therefore, this 5-HT agonist/uptake inhibitor could be useful to detect serotonergic dysfunction in cocaine abusers. The authors thank Ms. KayokoKunimotoand Mr. Joseph Yrachetafor theirexcellenttechnicalassistance. We also thankthe followingfor the generousgiftsof drugs:the NationalInstitutefor DrugAbuse(NIDA, Rockville, MD) for cocaine and Merck Sh~r~ and Dohm¢ (West Point PA) for ~K-212, Supported in part by USPHS Grants DA04865 (LDV), MH45812 (LDV and MSB), and AmericanHeart Associationof MetropolitanChicago (LDV) and the Veterans Administration(MC).
References Alper RH (1990): Hemodynamic and renin response to ( - )-DOI, a selective 5-t4T~ receptor agonist, in conscious rats. Eur J Pharmacol 175:323-332. Alper RH, Snider JM (1987): Activation of semtonin2 (5-HT2) receptors by quipazine increases arterial pressure and renin secretion in conscious rats. J Pharmacol Exp Ther 243:829-833. Barofsky A-L, Taylor J, Massari VJ (1983): Dorsal raphe-hypothalamic projections provide the stimulatory semtonergic input to suckling-induced pmlactin release. Endocrinology !! 3:1894-1903. Bastani B, Nash JF, Meltzer HY (1990): Prolactin and cortisol responses to MK-212, a serotonin agonist, in obsessive-compulsive disorder. Arch Gen Psychiatry 47:833-839. Borowsky B, Kuhn CM (1991a): Chronic cocaine administration sensitizes behavioral but not neuroendocrine responses. Brain Res 543:301-306. Borowsky B, Kuhn CM (! 991b): Monoamine mediation of cocaine-induced hypothalamo-pituitaryadrenal activation. J Pharmacol Exp Ther 256:204-210. Brownfield MS, Greathouse J, Lorens SA, Armstrong J., Urban JH, Van de Kar LD (1988): Neuropharmacological characterization of serotonergic stimulation of vasopressin secretion in conscious rats. Neuroendocrinology 47:277-283. Cames M, Bmwnfield MS, Kalin N, Lent S, Barksdale CM (1986): Episodic secretion of ACTH in rats. Peptides 7:219-224. Clineschmidt B V (1979): MK-212: A serotonin-like agonist in the CNS. Gen Pharmacol 10:287290. Clineschmidt BV, Totaro JA, Pflueger AB, McGuffin JC (1978): Inhibition of the serotonergic uptake system by MK-212 (6-chloro-2-[ l-piperazinyl]-pyrazine). PharmacolRes Carom 10(3):219228. Conn PJ, Sanders-Bush E (1985): Serotonin-stimulated phosphoinositide turnover. Mediation by the S2-binding site in rat cerebral cortex but not in subcortical regions. J Pharmacol Exp Ther 234:195-203. Cunningham KA, Callahan PM, Appel JB (1986): Discriminative stimulus properties of the semtonin agonist MK-212. Psychopharmacology 90:193-197. Cunningham KA, Lakoski JM (1988): Electrophysiological effects of cocaine and procaine on dorsal raphe serotonin neurons. Eur J Pharmacol 148:457-462. Hayer D (1988): Functional correlates of serotonin 5-HTI recognition sites. J Receptor Res 8:5981.
268
BIOLPSYCHIATRY
L.D. Van de Kar et al
1992;32:258-269
lmai H, Steindler DA, Kitai ST (1986): The organization of divergent axonal projections from the midbrain raphe nuclei in the rat. J Comp Neurol 243:363-380. Karteszi M, Van de Kar LD, Makara G, Stark E, Ganong WF (1982): Evidence that the mediobasal hypothalamus is involved in serotonergic stimulation of renin secretion. Neuroendocrinology 34:323-326. King BH, Brazell C, Dourish CT, Middlemiss DN (1989): MK-212 increases rat plasma ACTH concentration by activation of the 5-HT,c receptor subtype. Neurosci Lett 105:174-176. Kleven MS, Woolverton WL, Selden LS (1988): Lack of long-term monoamine depletions following repeated or continuous exposure to cocaine. Brain Res Bull 21:233-237. Koenig Jl, Gudelsky 13A, Meltzer HY (1987): Stimulation of corticosterone and beta-endorphin secretion in the rat by selective 5-HT receptor subtype activation. Fur J Pharraacol 137:!-8. Lakoski LM, Cunningham KA (1988): Cocaine interactions with central monoaminergic systems: Electrophysiological approaches. Trends Pharmacol Sci 9:177-180. Lakoski JM, Rittenhouse PA, Bonadonna AM, Van de Kar LD (1991): Acute, but not repeated, cocaine administration decreases renin secretion in the conscious male rat. Neurosci Lett 127:181184. Lesch K-P, Rupprecht R, Poten B, Muller U, Sohne K, Fritze J (1989): Endocrine responses to 5-hydroxytryptamine-lA receptor activation by ipsapirone in humans. Biol Psychiatry 26:203-205. Lesch K-P, Sohne K, Poten B, SchoeUnhammer t3, Rupprecht R (1990): Corticotropin and cortisol secretion after central 5-hydroxytryptamine-tA (5-HT~^) receptor activation: Effects of 5-HT receptor and beta-adrenoceptor antagonists. J Clin Endocrinol Metab 70:670--674. Levy AD, Li Q, Kerr JE, et al (1991): Cocaine-induced elevation of plasma ACTH and corticosterone is mediated by serotonergic neurons. J Pharmacol Exp Ther 259:495-500. Lorens SA, Van de Kar LD (1987): Differential effects of serotonin (5-HTIA and 5-HT.,) agonists and antagonists on renin and corticosterone secretion. Neuroendocrinology 45:3-,)5--310. Lowy MT, Meltzer HY (1988): Stimulation of serum cortisol and prolactin secretion in humans by MK-212, a centrally active serotonin agonist. Biol Psychiatry 23:818-828. Mcltzer HY, Lowy MT (1987): The serotonin hypothesis of depression. In Meltzer HY (ed), Psychopharmacology: The Third Generation of Progress. New York: Raven Press, pp 513526 Moliver ME (1987): Serotonergic neuronal systems: What their anatomic organization tells us about function. J Clin Psychopharmacol (Suppl 6) 7:3S-23S. Nunes EV, Quitkin FM, Klein DF (1989): Psychiatric diagnosis in cocaine abuse. Psychiatr Res 28:105-114, Pitts DK, Marwah J (1986): Electrophysiological effects of cocaine on central monoaminergic neurons. Eur J Pharmacol ! 3 ! :95-98. Reith MEA, Sershen H, Allen DL, Lajtha A (1983): A portion of [3HI cocaine binding in brain is associated with serotonergic neurons. Mol Pharmacol 23:6(10-.606. Richardson Morton KD, Van de Kar LD, Brownfield MS, Bethea CL (1989): Neuronal cell bodies in the hypothalamie paraventricular nucleus mediate stress-induced renin and corticosterone secretion. Neuroendocrinology 50:73-80. Ritz MC, Cone El, Kuhar MJ (1990): Cocaine inhibition of ligand binding at dopamine, norepinephrine and serotonin transporters: A structure-activity study. Life Sci 46:635--645. Roy SN, Bhattacharyya AK, Pradhan S, Pradhan SN (1978): Behavioural and neurochemical effects of repeated administration of cocaine in rats. Neuropharmacology 17:559-564. Saydoff JA, Rittenhouse PA, Van de Kar LD, Brownfield MS (1991): Enhanced serotonelgic transmission stimulates oxytocin secretion in conscious male rats. J Pharmacol Exp Ther 257:9599. Steel R13D, Torrie JH (1960): Principles and Procedures of Statistics with Special Reference to the Biological Sciences. New York: McGraw Hill.
Cocaine and Neuroendocrine Response to MK-212
BIOLPSYCHIATRY 1992;32:258-269
269
Taylor D, Ho BT (1977): Neurochemical effects of cocaine following acute and repeated injections. J Neurosci Re$ 3:95-101. Trinkoff AM, Ritter C, Anthony JC (1990): The prevalence and self-reported consequences of cocaine use. An exploratory and descriptive analysis. Drug Alcohol Depend 26:217-225. Van de Kar LD (1989): Neuroendocrine aspects of the serotonergic hypothesis of depression. Neuro$ci Biobehav Rev 13:237-246. 'Van de Kar LD (1991): Neuroendocrine pharmacology of serotonergic ($-HT) neurons. Annu Rev Pharmacol Toxicol 31:289-320. Van de Kar LD, Bethea CL (1982): Pharmacological evidence that serotonergic stimulation of prolactin secretion is mediated via the dorsal raphe nucleus. Neuroendocrinology 35:225-230. Van de Kar LD, Wilkinson CW, Skrobik Y, Brownfield MS, Ganong WF (1982): Evidence that serotonergic neurons in the dorsal raphe nucleus exert a stimulatory effect on the secretion of renin but not of corticosterone. Brain Res 235:233-243. Van de Kar LD, Kmeszi M, Bethea CL, Ganong WF 0985): Serotonergic stimulation of prolactin and oorticosterone secretion is mediated by different pathways from the mediobasal hypothal. amus. Neuroendocrinology 41:380-384. Van de Kar LD, Cames M, Maslowski RJ, et al (1989a): Neuroendocrine evidence for denervation supersensitivity of serotonin receptors. Effects of the 5-HT agonist RU 24969 on corticotropin, corticosterone, prolactin and renin secretion. J Pharmacol Exp Ther 251:428-434. Van de Kar LD, Lorens SA, Urban JH, Bethea CL (19891)): Effect of selective serotonin (5-HT) agonists and 5-HT2 antagonist on prolactin secretion. Neuropharmacology 28:299-305. Van de Kar LD, Rittenhouse PA, Iyer LV, et al (1990a): Chronic cocaine alters the neuroendocrine responses to serotonin (5-HT) agonists (Abst~ac0 FASEB J 4:A594--A594. Van de Kar LD, Urban JH, Brownfield MS (1990b): Serotonergic regulation of renin and vasopressin secretion. In Paoletti R, Vanhoutte P M, Brunello N, Maggi F M (eds), Serotonin: From Cell Biology to Pharmacology and Therapeutics. Dordrecht, The Netherlands: Kluwer Academic Publishers, pp 123-129. Van de Kar LD, Bonadorma AM, Rittenhouse PA (1992): Prior exposure to chronic cocaine inhibits the serotonergic stimulation of ACTH and corticosterone secretion. Neuropharmacology 31:169175.
BIOL PSYCHIATRY 1992;32:258-269
Effect of Cocaine Injections on the Neuroendocrine Response to the Serotonin Agonist MK-212 Louis D. Van de Kar, Peter A. Rittenhouse, Patricia O'Connor, Thomas Palionis, Mark S. Brownfield, Stephanie J. Lent, Molly Cames, and Cynthia L. Bethea
This study was undertaken to examine whether several of the hormones that can be released by activation of serotonin receptors will be affected by long-term cocaine administration. Male rats received cocaine injections (15 mglkg, IP) twice daily for 7 days. Forty.two hr after the last cocaine injection, the rats were challenged with increasing doses (0, 1, S, 10 mg/kg, IP) of the S-HT1/5-HT2 agonist MK-212 (6-chloro-2-[l-piperazinyl]-pyrazine). The following observations were made: (1) cocaine reduced the rate of body weight gain; (2) cocaine inhibited the stimulatory effect of MK-212 on plasma vasopressin, ozytocin, and prolactin concentrations and on plasma renin activity and concentration; (3) cocaine did not inhibit the stimulatory effect of MIC.212 on plasma ACTH or corticosterone concentrations. The data indicate that a wide.spectrum S-HT (serotonin) agonist such as MK-212 con reveal differential neuroendocrine responses. This effect could be related to cocaine-induced changes in the different 5-HT receptor subtypes that regulate the secretion of these hormones.
Introduction Cocaine is a psychostimulant that has a considerable influence on monoaminergic neurotransmission in the brain. In addition to dopamine and norepinephrine uptake sites, cocaine binds with high affinity to serotonin (5-HT) synaptic reuptake sites (Reith et al 1983; Ritz et al 1990). This prevents the reuptake of these neurotransmitters from the synaptic cleft and leads to increased concentration of 5-HT (as well as dopamine and norepinephrine) in the synapse. Acute injection of cocaine increases ACTH secretion but reduces prolactin and lenin secretion (Borowsky and Kuhn 1991a; Lakoski et al 1991). The effect of cocaine on adrenocorticotropin (ACTH) secretion is partly mediated by 5-HT2 receptors (Borowsky and Kuhn 1991b; Levy et al 1991). Therefore, it is important
From the Departmentof Pharmacology,Stritch Schoolof Medicine, Loyola Universityof Chicago, Maywood, IL (LDV, PAR, PO); the Departmentof ComparativeBiosciences,School of VeterinaryMedicine, Universityof Wisconsin, Madison, Wi (TP, MSB); the Middleton Memorial Veterans Hospital, Geriatric Section, Madison, WI (STL, MC); and the Oregon Regional Primate ResearchCenter, Beaverton,OR (CLB). Address reprint requests to Louis D. Van de Kar, Ph.D., Department of Pharmacology,Stritch School of Medicine, Loyola University of Chicago, 2160 S. First Avenue, Maywood, IL 60153. Received May 16, 1991; revised October 28, 1991.
© 1992 Society of BiologicalPsychiatry
0006-3223/92/$05.00
Cocaine and Neuroendocfine Response to MK-212
BIOLPSYCHIATRY 1992;32:258-269
259
to know if chronic exposure to cocaine can influence the serotonergic regulation of neuroendocrine function. Knowledge of the basic mechanisms that cause cocaine's deleterious effects may contribute to the development of therapy and prevention of its abuse. Cocaine abusers often experience post-abuse anhedonia and depression, leading to renewed cocaine selfadministration (Nunes et al 1989; Trinkoffet al 1990). Decreased function of serotonergic neurons is considered to be, at least in part, a cause of endogenous depression (Meltzer and Lowy 1987). Little agreement exists about the long-term effects of cocaine on serotonergic neurons. Two studies found that repeated injection of cocaine reduced the 5-HT content in the brain (Roy et al 1978; Taylor and Ho 1977). However, in another study (Kleven et al 1988), no change in the 5-HT content in several brain regions was found after injection of a lower dose of cocaine for 10 days. Thus, additional studies are necessary to establish the long-term effects of cocaine on serotonergic function. The present study utilized the hormonal responses to a 5-HTI/5-HT2 agonist as indices of the function of serotonin receptors. Serotonergic neurons stimulate, either directly or indirectly, the release of many hormones including ACTH, corticosterone, prolactin, lenin, oxytocin, and vasopressin (for review see Van de Kar 1991). The 5-HT neurons that regulate neuroendocrine function are part of an extensive 5-HT pathway that also sends collaterals to limbic areas (Moliver 1987; Imai et al 1986). Therefore, changes in 5-HT neurons that influence behavior can be reflected in altered neuroendocrine responses to challenges with 5-HT agonists (Van de Kar 1989). Distinct 5-HT receptor subtypes mediate the serotonergic stimulation of these hormones. Activation of 5-HTtA receptors increases ACTH and corticosterone but none of the other hormones (Brownfield et al 1988; Koenig et al 1987; Lesch et al 1989, 1990; Lorens and Van de Kar 1987; Saydoff et al 1991; Van de Kar et al 1989a,b). Activation of 5-HT~c receptors also increases ACTH and corticosterone secretion (King et al 1989). On the other hand, activation of 5-HT2 receptors increases oxytocin, vasopressin, and renin secretion (Alper and Snider 1987; Alper 1990; Brownfield et al 1988; Lorens and Van de Kar 1987; Saydoff et al 1991; Van de Kar et al 1990b). Prolactin secretion is most likely mediated by activation of 5-HTle recepto~ b, rats (Van de gar et al 1989a,b). The present study focused on the influence of cocaine on the 5-HT receptors that stimulate these hormones. The 5-HTI/5-HT2 agonist and uptake inhibitor (6-chloro2-[I-piperazinyl]-pyrazine) (MK-212) (Clineschmidt 1979; Ciineschmidt et al 1978) has been previously tested in humans and was shown to increase serum prolactin and cortisol levels (Lowy and Meltzer, 1988). Furthermore, the effect of MK-212 on plasma prolactin and cortisol was blunted in patients with obsessive-compulsive disorder (Bastani et al 1990). Therefore, MK-212 was used as a challenge after injection of cocaine to rats for 7 days.
Methods Animals Male Sprague-Dawley rats (200-250 g) were purchased from Sasco-King Animal Laboratories (Oregon, ~'!). The rats were housed, two per cage, in a temperature, humidity, and illumination (12:12 hr light/dark cycle; lights on at 7 AM) controlled room. Water and food (Wayne Lab Blox, Lal~ Mills Inc., Chicago, IL) were available ad libitum. The
260
BIOLPSYCHIATRY
L.D. Van de Kar et al
1992;32:258-269
rats were killed by decapitation in an area outside the rat room. All the experimental protocols were approved by Loyola University Animal Care and Use Committee.
Drugs All drugs were dissolved in saline and injected in a volume of 1 ml/kg, IP. All cocaine injections (15 mg/kg, IP) were performed twice a day (8:30 AM and 5 PM), a dosage regimen that was previously reported to reduce brain serotonin content (Roy et al 1978). All subsequent challenges with a 5-HT agonist occurred 42 hr after the last cocaine injection. This interval was chosen to ensure complete elimination of cocaine from the circulation, and thus avoid its interaction with the 5-HT agonist. Cocaine HCI was donated by the National Institute for Drug Abuse (NIDA, RockviUe, MD) and MK-212 (6-chloro2-[l-piperazinyl]-pyrazine) was donated by Merck Sharp and Dohme (West Point, PA). MK-212 was dissolved in saline and injected in doses of 1, 5, and 10 mg/kg (ip). The rats were killed by decapitation 30 min after the injection of MK-212. All drug doses are expressed as the salt. Biochemical Determinations
Preparation of Plasma Trunk blood was collected in centrifuge tubes containing a 0.5 ml solution of 0.3 M ethylenediamine tetraacetic acid (EDTA; pH 7.4). The blood was centrifuged at 1000 x g for 20 rain at 4°C and the plasma was saved. Plasma from each rat was divided into aliquots of (1) 1.0 ml for the determination of vasopressin and oxytocin, (2) 1.0 ml for the determination of plasma renin activity (PRA), (3) 50 p,! for the determination of plasma renin concentration (PRC), (4) 30 I~l for the determination of corticosterone levels, (5) 200 p.l for the determination of plasma ACTH, and (6) 0.2 ml for the determination of prolactin levels. All plasma aliquots were stored at - 40°C until the respective hormones were determined.
Plasma VasopressinRadioimmunoassay This assay has been described in detail elsewhere (Brownfield et al 1988). The assay is a double antibody assay conducted under disequilibrium conditions. Rabbit anti-lysine vasopressin, generated in Dr. Brownfield's laboratory, and the tracer arginine vasopressin (AVP) were used in this assay. The measurement of plasma samples requires that they be extracted, a procedure detailed elsewhere (Brownfield et al 1988); the sensitivity limit is 0. I pg/tube. The antibody has less than 0.01% cross-reactivity with oxytocin and vasotocin. The intraassay variability is 4.5% and the interassay variability is 8% (Brownfield et al 1988).
Plasma Oxytocin Radioimmunoassay This assay is identical to the vasopressin assay described above and was described in detail elsewhere (Saydoff et al 1991); it also is a double-antibody assay conducted under disequilibrium conditions. Rabbit anti-oxytocin obtained from Hoechst-Boehring (San
Cocaine and Neuroendocrine Response to MK-212
BIOLPSYCHIATRY 1992;32:258-269
261
Diego, CA) was generated in rabbits. Synthetic oxytocin was obtained from Bachem Fine Chemicals (Torrance, CA) with purity of the peptide established by thin layer chromatography, amino acid analysis, and high-pressure liquid chromatography. The measurement of plasma samples requires that they are extracted, and this procedure is detailed elsewhere (Brownfield et al 1988). The antibody has less than 0.01% cross-reactivity with AVP and vasotocin. The intraassay and interassay variabilities are 5%, and 9%, respectively.
Plasma Renin Activity (PRA ) PRA was measured by radioimmunoassay for generated angiotensin (ANG I) as previously described (Van de Kar et al 1989a). In this assay, renin generates ANG 1 from the endogenous renin substrate. Therefore, PRA is a composite of the plasma concentration of renin and renin substrate. The radioimmunoassay for AN(3 I is described below.
Plasma Renin Concentration (PRC) In this assay, a saturating concentration of renin substrate is added to the plasma to allow generation of ANG 1 at maximal velocity. Thus, this assay reflects the total concentration of the enzyme renin in plasma, independent of the concentration of renin substrate. In the PRC assay, renin substrate is obtained from plasma of rats that were nephrectomized and received a dexamethasone injection (0.2 rag/rat) 24 hr before sacrifice. The details of this assay were described by us ebewhere (Richardson Morton et al 1989). The radioimmunoassay of ANG I was perfozmed with antiserum at a dilution of 1:16,000 and a total binding of 35% as previously described (Richardson Morton et al 1989). The sensitivity limit of the RIA was 10 pg ANG I per tube and the intraassay variability was 4.4%. The interassay variability was 12.6%.
Piasn:J Corticosterone Radioimmunoassay Corticosterone radioimmunoassay was performed on unextracted plasma samples (2 and 5 I~i) in which binding proteins have been denatured by boiling, as previously described (Richardson Morton et ai 1989), using procedures and antiserum from Radioassay Systems Laboratories (Carson, CA). The sensitivity limit is 0.02 ng/tube and the intraassay and interassay variabilities were 4.5% and 11.9%, respectively.
Plasma ACTH Radioimmunoassay ACTH radioimmunoassay on unextracted plasma samples (20-50 Ixl) was performed as previously described (Carnes et al 1986). Briefly, the ACTH antiserum was obtained from lgG Corp (Nashville TN). ACTH standards (1-39) were obtained from Calbiochem and ~2SI-ACTH from INCSTAR (Stillwater, MN). The sequence recognition of the antiserum is 5-18. In addition, this antiserum does not significantly recognize a-MSH, [3-MSH, [3-endorphin, [3-1ipotropin, ACTH ! 1-24, or ACTH 1-16-amide. The minimum detectable concentration is 0.25 pg/tube and the intraassay and interassay variations were 4.2% and 14.6%, respectively.
262
BIOL PSYCHIATRY 1992;32:258-269
L.D. Van de Kar et al
Plasma Prolactin Radioimmunoassay Prolactin radioimmunoassay was performed with reagents provided by the National Institute of Arthritis, Diabetes, Digestive and Kidney Disorders (NIADDK). Anti-rat prolactin serum S-8 was used at a dilution of 1:5000, as described previously (Van de Kar and Bethea 1982). Briefly, NIADDK rat prolactin (preparation rPRL-I-5) was used for iodinated tracer, and NIADDK rat prolactin (preparation rPRL-RP-3) was used as the reference preparation. The intraassay variability was 6.8% and all the samples from one experiment were determined together in one assay.
Statistics The data are represented as the group means and the standard errors of the mean (SEM). Statistical analysis of the data was performed by two-way analysis of variance (ANOVA), and individual group means were compared by Student-Newman-Keuls' test (Steel and Torrie 1960) using a computer program (NWA STATPAK, Portland OR).
Results Injection of cocaine (15 mg/kg, IP) for 7 days produced a significant reduction in the rate of body weight gain. The body weights on the last day were 291 +- 4.1 g for the vehicle group and 278 - 3.8 g for the cocaine group. The statistical analysis revealed a significant increase in body weigh (Fis,a66t = 56.43, p < 0.00001), a significant effect of cocaine (FI~,.~661 = 9.11, p < 0.0026), and a significant cocaine x weight gain interaction (F[5.a661 -- 2.31, p < 0.044). A posthoc Newman Keuls analysis revealed that the significant difference from the saline-treatod rats were observed in cocaine-treated rats that were weighed 42 hr after the last cocaine injection. The basal levels of the hormones measured were not altered in the cocaine-treated rats. Injection of MK-212 (l, 5, or l0 mg/kg, IP) to the vehicle-treated rats dose dependently increased the plasma concentration of all hormones except vasopressin and renin which only increased after injection of the highest dose. The effect of MK-212 on plasma prolactin concentration was significantly inhibited in the cocaine-injected rats (Figure 1). The two-way ANOVA showed a significant cocaine x MK-212 interaction (Fi3..~41 = 7.138, p < 0.01). Similarly, the effect of the highest dose of MK-212 on oxytocin was inhibited in cocaine pretreated rats (Figure 2). The two-way ANOVA did not reveal a significant cocaine x MK-212 interaction (FD,471 ---1.36, NS). However, a subsequent comparison of the group means using Newman Keuls' test revealed that plasma oxytocin concentration in cocaine pretreated rats that received the highest dose of MK-212 (10 mg/kg) were significantly lower than those of vehiclepretreated rats that received the same dose of MK-212 (Figure 2). In the case of vasopressin and renin, only the highest dose of MK-212 significantly increased their plasma concentrations (Figures 2 and 3). The effect of MK-212 on plasma vasopressin was abolished in cocaine-treatod rats (Figure 2). The two-way ANOVA indicated a cocaine × MK-212 interaction (Ft3,s~I = 2.579, p < 0.0516) that was not significant, but a post-hoc Newman Keuls' test revealed that the MK-212 (10 mg/kg)induced increase in vasopressin was significantly blocked in cocaine-treated rats. The effect of MK-2 i 2 on plasma renin activity and concentration was inhibited by pretreatment with cocaine (Figure 3). The two-way ANOVA revealed a significant cocaine x MK-
Cocaine and NeuroendocrineResponse to MK-212
BIOL PSYCHIATRY
263
1992;32:258-269
# o SALINE
270
Figure 1. MK-212 (0, I, 5, 10 mg/kg, IP) induced increase in plasma prolactin concentration in rats treated with cocaine (15 mg/kg, twice daily) for 7 days. Data represent mean --. SEM of eight rats per group. **Significant difference from the corresponding saline (0 dose of MK-212) group, p < 0.01 (2-way ANOVA and Newman Keuls' test). #Significant difference from the corresponding group injected with the same dose of MK-212, p < 0.05 (2-way ANOVA and Newman Keuls' test).
240
l=o l=
r-U Z
~.I 0= I~ .,¢ r~
COCAINE
21o
**t I
18o 150 120 90
_/
60
Ill
0
o
.
.
.
.
.
_...,,,~ .lip
1
. . . . . . . .
I
lO
DOSE OF MK-2t2 (mg/kg IP)
# 120
•
i
COCAINE
"
0i
tlo
r,o
30
o
A
i
i
. . . .
|
. . . .
1
iI
lo
# 4O
l
u~ r~ rA
g
.
,< ,< @
-
-
-
. . . .
O~
1
~
"
1o
DOSE OF M K - 2 1 2 ( m g / k g I~')
Figure 2. MK-212 (0, I, 5, I0 mg/kg, IP) induced increase in plasma oxytocin (top) and vasopressin (bottom) in rats treated with cocaine (15 mg/kg, twice daily) for 7 days. Data represent mean --- SEM of eight rats per group. Significant difference from the corresponding saline (0 dose of MK-212) group, *p < 0.05, **p < 0.01 (2-way ANOVA and Newman Keuls' test). ~Significant difference from the corresponding group injected with the same dose of MK-212, p < 0.05 (2-way ANOVA and Newman Keuls' test).
264
BIOL PSYCHIATRY 1992;32:258-269
L.D. Van de Kar et al
#
40
o SALINE
35
*
4, COCAINE 25
z~
2o "~ 10
D,, 0
i!
0
I
10
# ==
i.
.o
6O
Figure 3. MK-212 (0, !, 5, 10 mg/kg, IP) induced increase in plasma lenin activity (top) and concentration (bottom) in rats treated with cocaine (15 mg/kg, twice daily) for 7 days. Data represent mean - SEM of eight rats per group. *Significant difference from the corresponding saline (0 dose of MK-212) group, p < 0.05 (2way ANOVA and Newman Keuls' test). #Sigaificant differencefrom the correspondinggroup injected with the same dose of MK-212, p < 0,05 (2-way ANOVA and Newman Keuls' test).
m
=.
0
0
1
10
DOSE OF t d K - 2 t 2 (mg/kg, ip)
212 interaction for both plasma renin activity (Fp.s,,I = 2.90, p < 0.043) and plasma lenin concentration (Ft,.s21 -- 2.96, p < 0.041). In contrast with the above-mentioned hormones, there was no difference in the doseresponse curve effects of MK-212 on plasma ACTH and corticosterone between cocaine and vehicle-pretreated rats (Figure 4). The two-way ANOVA exhibited a cocaine × MK212 interaction that was not significant for either ACTH (Fp.s41 = 1.73, p < 0.16) or plasma corticosterone (Fp.s~j = 0.678, p < 0.56). Furthermore, a post-hoe Newman Keuls' test revealed no difference between the cocaine and saline-treated groups for any of the MK-212 doses.
Discussion The results of this study suggest that cocaine injections for 7 days can differentially alter the subsequent hormonal responses to a direct 5-HT agonist. Though the effect of MK212 on several hormones (vasopressin, oxytocin, prolactin, and renin) was blunted after cocaine injections, the hypothalamic-pituitary-adrenal (HPA) axis was spared. This raises the question, What is the difference between the HPA axis and the other neuroendocrine systems? The most obvious difference is in the nature of the 5-HT receptor subtypes that control the secretion of these hormones. Activation of 5-HT~A receptors only increases ACTH
Cocaine and Neuroendocrine Response to MK-212
BIOLPSYCHIATRY
265
1992;32:258-269
go0 70o
*
COCAINE
~'~ t
600 ~[~ 500 ,~ 400 ~:~ 300 200 I~
~
Ioo 0
. . . . . . . .
0
I
illll
1
30 ~
. . . .
,
10
*
zs
0
I
Figure 4. MK-212 (0, I, 5, 10 mg/kg, IP) induced increase in plasma ACTH (top) and corticostemne (bottom) in rats treated with cocaine (15 mg/kg, twice daily) for 7 days. Data represent mean .+SEM of eight rats per group. *Significant difference from the corresponding saline (0 dose of MK212) group, p < 0.05 (2-way ANOVA and Newman Keuls' test).
10
DOSE OF MK-212 (mg/kg, ip)
and corticosterone secretion (Koenig et al 1987; Lorens and Van de Kar 1987). Activation of 5-HTIA receptors does not seem to increase the secretion of vasopressin, renin, or prolactin (Brownfield et al 1988; Lorens and Van de Kar 1987; Sayoff et al 1991; Van der Kar et al 1989a; Van de Kar et al 1990b). MK-212 is a 5-HTI and 5-HT2 agonist (Conn and Sanders-Bush 1985; Cunningham et al 1986). Its affinity for the various 5-HT receptor subtypes is not well described, though it is considered a low affinity and high efficacy agonist (Ciineschmidt 1979; Clineschmidt et al 1978) with a moderate affinity for 5-HT~A, 5-HT~e, and 5-HT~c receptors and a lower affinity for 5-HT2 receptors (Hoyer 1988). MK-212 was previously shown to increase lenin and vasopressin secretion by activation of 5-HT2 or 5-HT~c receptors (Brownfield et al 1988; Lorens and Van de Kar 1987). The effect of the high dose of MK-212 (10 mg/kg, IP) on both renin and vasopressin secretion was completely blocked by low doses (0.1-0.3 mg/kg) of the 5-HT,c/5-HT2 antagonists LY53857 and ritanserin (Brownfield et al 1988; Lorens and Van de Kar 1987). These observations suggest that even at this high dose, MK-212 activates mainly 5-HT~c/5-HT2 receptors to stimulate renin and vasopressin secretion. Hence, the cocaine-induced inhibition of the effect of MK-212 on renin and vasopressin secretion suggests a decline in the function of 5-HTic or 5-HT2 receptors. The influence of MK-212 on oxytocin secretion is at least partly mediated by 5-HT2 or 5-HT~c receptors as it was inhibited by ritanserin (Saydoff et al 1991). With respect
266
BIOLPSYCHIATRY
L.D. Van de Kar et al
1992:32:258-269
to prolactin secretion, blocking 5-HT~c or 5-HT2 receptors with LY53857 only partly inhibited the effect of MK-212 on prolactin secretion, suggesting that not only 5-HT1c/ 5-HT2 but also other 5-HT receptors mediate the effect of MK-212 on prolactin (Van de Kar et al 1989b). The MK-212-induced stimulation of ACTH and corticosterone secretion is subject to debate, but most data suggest that it is mediated by activation of either 5-HTIA or 5-HT~c receptors (King et al 1989; Koenig et al 1987). Thus, the data suggest that the effect of MK-212 on most hormones is mediated, at least in part, by either 5HT~c or 5-HT2 receptors. In addition, the effect of MK-212 on the HPA axis could be partly mediated by activation of 5-HT~,,, receptors. Along this line of reasoning, the repeated injection of cocaine could have a more profound effect on 5-HTtc or 5-HT2 receptors than on 5-HT~A receptors. We previously found that repeated injection of cocaine reduces the function of scrotonergic nerve terminals that stimulate ACTH and corticosterone secretion, because they inhibited the effect of the 5-HT releaser p-chloroamphetamine on plasma ACTH and corticosterone concentrations (Van de Kar et al 1990a, 1992). In contrast, pretreatment with cocaine potentiated the corticosterone response to the direct 5-HT agonist RU 24969 whereas the corticosterone response to another 5-HT agonist, m-CPP, was unaltered (V an de Kar et al 1992). Hence, the current data with MK-212 arc consistent with previous data using other direct 5-HT agonists. Taken together, the data suggest that cocaine does not alter all the 5-HT receptor subtypes that stimulate ACTH secretion. Another interpretation of the data could be that the effect of cocaine on serotonergic neurons is more profoundly focused on the nerve terminals and that not all serotonergic nerve terminals are affected in the same manner. Cocaine binds with high affinity to the 5-HT (and also dopamine and norepinephrine) uptake sites on nerve terminals (Reith et al 1983: Ritz et al 1990). The binding of cocaine to the 5-HT uptake site prevents its reuptake from the synaptic cleft which leads to increased concentration of 5-HT in the synapse. Acute IV injection of cocaine suppresses the firing rate of serotonergic neurons in the dorsal raphe nucleus via activation of somato-dendritic 5-HT~^ receptors (Pitts and Marwah 1986; Lakoski and Cunningham 1988; Cunningham and Lakoski 1988). Thus, repeated administration of cocaine could alter the function of neurons originating in the dorsal raphe nucleus, and their nerve terminals in the forebrain. MK-212 is not only a 5-HT agonist but also a 5-HT uptake inhibitor (Clineschmidt et al 1978). Thus, part of its endocrine effects could be mediated by the increase of synaptic 5-HT concentration as a consequence of 5-HT uptake inhibition. There is evidence that the serotonergic neurons that regulate neuroendocrine function are not homogenous. The serotonergic neurons that stimulate prolactin and lenin secretion arc located in the midbrain dorsal raphe nucleus (Barofsky et al 1983; Van de Kar and Bethea 1982; Van de Kar et al 1982). In contrast, the serotonergic neurons that regulate ACTH/corticosterone secretion are neither in the dorsal raphe, nor in the median raphe nucleus (Van de Kar et al 1982). Furthermore, the serotonergic fibers that stimulate vasopressin secretion might be ascending fibers whose cell bodies arc located caudal to the hypothalamus. This information comes from knife cuts that interrupted ascending fibers to the hypothalamus and blocked the effect of a 5-HT releasing drug on plasma vasopressin and prolactin concentrations and renin activity, but not on plasma corticosterone concentration (Karteszi et al 1982; Brownfield et al 1988; Van de Kar et al 1985). Thus, the different serotonergic fibers that stimulate the secretion of the respective hormones could underlie their divergent response to MK-212 after pretreatment with cocaine. In conclusion, the data suggest that repeated administration of cocaine has a profound
Cocaine and Neuroendocrine Response to MK-212
BIOLPSYCHIATRY 1992;32:258-269
267
inhibitory effect on serotonergic neuronal function in rats. The most likely receptor subtypes affected are 5-HT2 or 5-HT~c receptors, and the most likely neurons affected are the ones originating from the dorsal raphe nucleus. The data also suggest a need to test whether decreased serotonergic function also occurs in human cocaine abusers. MK212 seems to be a useful diagnostic tool in humans as it increases plasma prolactin and cortisol, an effect that was blunted in patients with obsessive-compulsive disorder (Bastani et al 1990). Therefore, this 5-HT agonist/uptake inhibitor could be useful to detect serotonergic dysfunction in cocaine abusers. The authors thank Ms. KayokoKunimotoand Mr. Joseph Yrachetafor theirexcellenttechnicalassistance. We also thankthe followingfor the generousgiftsof drugs:the NationalInstitutefor DrugAbuse(NIDA, Rockville, MD) for cocaine and Merck Sh~r~ and Dohm¢ (West Point PA) for ~K-212, Supported in part by USPHS Grants DA04865 (LDV), MH45812 (LDV and MSB), and AmericanHeart Associationof MetropolitanChicago (LDV) and the Veterans Administration(MC).
References Alper RH (1990): Hemodynamic and renin response to ( - )-DOI, a selective 5-t4T~ receptor agonist, in conscious rats. Eur J Pharmacol 175:323-332. Alper RH, Snider JM (1987): Activation of semtonin2 (5-HT2) receptors by quipazine increases arterial pressure and renin secretion in conscious rats. J Pharmacol Exp Ther 243:829-833. Barofsky A-L, Taylor J, Massari VJ (1983): Dorsal raphe-hypothalamic projections provide the stimulatory semtonergic input to suckling-induced pmlactin release. Endocrinology !! 3:1894-1903. Bastani B, Nash JF, Meltzer HY (1990): Prolactin and cortisol responses to MK-212, a serotonin agonist, in obsessive-compulsive disorder. Arch Gen Psychiatry 47:833-839. Borowsky B, Kuhn CM (1991a): Chronic cocaine administration sensitizes behavioral but not neuroendocrine responses. Brain Res 543:301-306. Borowsky B, Kuhn CM (! 991b): Monoamine mediation of cocaine-induced hypothalamo-pituitaryadrenal activation. J Pharmacol Exp Ther 256:204-210. Brownfield MS, Greathouse J, Lorens SA, Armstrong J., Urban JH, Van de Kar LD (1988): Neuropharmacological characterization of serotonergic stimulation of vasopressin secretion in conscious rats. Neuroendocrinology 47:277-283. Cames M, Bmwnfield MS, Kalin N, Lent S, Barksdale CM (1986): Episodic secretion of ACTH in rats. Peptides 7:219-224. Clineschmidt B V (1979): MK-212: A serotonin-like agonist in the CNS. Gen Pharmacol 10:287290. Clineschmidt BV, Totaro JA, Pflueger AB, McGuffin JC (1978): Inhibition of the serotonergic uptake system by MK-212 (6-chloro-2-[ l-piperazinyl]-pyrazine). PharmacolRes Carom 10(3):219228. Conn PJ, Sanders-Bush E (1985): Serotonin-stimulated phosphoinositide turnover. Mediation by the S2-binding site in rat cerebral cortex but not in subcortical regions. J Pharmacol Exp Ther 234:195-203. Cunningham KA, Callahan PM, Appel JB (1986): Discriminative stimulus properties of the semtonin agonist MK-212. Psychopharmacology 90:193-197. Cunningham KA, Lakoski JM (1988): Electrophysiological effects of cocaine and procaine on dorsal raphe serotonin neurons. Eur J Pharmacol 148:457-462. Hayer D (1988): Functional correlates of serotonin 5-HTI recognition sites. J Receptor Res 8:5981.
268
BIOLPSYCHIATRY
L.D. Van de Kar et al
1992;32:258-269
lmai H, Steindler DA, Kitai ST (1986): The organization of divergent axonal projections from the midbrain raphe nuclei in the rat. J Comp Neurol 243:363-380. Karteszi M, Van de Kar LD, Makara G, Stark E, Ganong WF (1982): Evidence that the mediobasal hypothalamus is involved in serotonergic stimulation of renin secretion. Neuroendocrinology 34:323-326. King BH, Brazell C, Dourish CT, Middlemiss DN (1989): MK-212 increases rat plasma ACTH concentration by activation of the 5-HT,c receptor subtype. Neurosci Lett 105:174-176. Kleven MS, Woolverton WL, Selden LS (1988): Lack of long-term monoamine depletions following repeated or continuous exposure to cocaine. Brain Res Bull 21:233-237. Koenig Jl, Gudelsky 13A, Meltzer HY (1987): Stimulation of corticosterone and beta-endorphin secretion in the rat by selective 5-HT receptor subtype activation. Fur J Pharraacol 137:!-8. Lakoski LM, Cunningham KA (1988): Cocaine interactions with central monoaminergic systems: Electrophysiological approaches. Trends Pharmacol Sci 9:177-180. Lakoski JM, Rittenhouse PA, Bonadonna AM, Van de Kar LD (1991): Acute, but not repeated, cocaine administration decreases renin secretion in the conscious male rat. Neurosci Lett 127:181184. Lesch K-P, Rupprecht R, Poten B, Muller U, Sohne K, Fritze J (1989): Endocrine responses to 5-hydroxytryptamine-lA receptor activation by ipsapirone in humans. Biol Psychiatry 26:203-205. Lesch K-P, Sohne K, Poten B, SchoeUnhammer t3, Rupprecht R (1990): Corticotropin and cortisol secretion after central 5-hydroxytryptamine-tA (5-HT~^) receptor activation: Effects of 5-HT receptor and beta-adrenoceptor antagonists. J Clin Endocrinol Metab 70:670--674. Levy AD, Li Q, Kerr JE, et al (1991): Cocaine-induced elevation of plasma ACTH and corticosterone is mediated by serotonergic neurons. J Pharmacol Exp Ther 259:495-500. Lorens SA, Van de Kar LD (1987): Differential effects of serotonin (5-HTIA and 5-HT.,) agonists and antagonists on renin and corticosterone secretion. Neuroendocrinology 45:3-,)5--310. Lowy MT, Meltzer HY (1988): Stimulation of serum cortisol and prolactin secretion in humans by MK-212, a centrally active serotonin agonist. Biol Psychiatry 23:818-828. Mcltzer HY, Lowy MT (1987): The serotonin hypothesis of depression. In Meltzer HY (ed), Psychopharmacology: The Third Generation of Progress. New York: Raven Press, pp 513526 Moliver ME (1987): Serotonergic neuronal systems: What their anatomic organization tells us about function. J Clin Psychopharmacol (Suppl 6) 7:3S-23S. Nunes EV, Quitkin FM, Klein DF (1989): Psychiatric diagnosis in cocaine abuse. Psychiatr Res 28:105-114, Pitts DK, Marwah J (1986): Electrophysiological effects of cocaine on central monoaminergic neurons. Eur J Pharmacol ! 3 ! :95-98. Reith MEA, Sershen H, Allen DL, Lajtha A (1983): A portion of [3HI cocaine binding in brain is associated with serotonergic neurons. Mol Pharmacol 23:6(10-.606. Richardson Morton KD, Van de Kar LD, Brownfield MS, Bethea CL (1989): Neuronal cell bodies in the hypothalamie paraventricular nucleus mediate stress-induced renin and corticosterone secretion. Neuroendocrinology 50:73-80. Ritz MC, Cone El, Kuhar MJ (1990): Cocaine inhibition of ligand binding at dopamine, norepinephrine and serotonin transporters: A structure-activity study. Life Sci 46:635--645. Roy SN, Bhattacharyya AK, Pradhan S, Pradhan SN (1978): Behavioural and neurochemical effects of repeated administration of cocaine in rats. Neuropharmacology 17:559-564. Saydoff JA, Rittenhouse PA, Van de Kar LD, Brownfield MS (1991): Enhanced serotonelgic transmission stimulates oxytocin secretion in conscious male rats. J Pharmacol Exp Ther 257:9599. Steel R13D, Torrie JH (1960): Principles and Procedures of Statistics with Special Reference to the Biological Sciences. New York: McGraw Hill.
Cocaine and Neuroendocrine Response to MK-212
BIOLPSYCHIATRY 1992;32:258-269
269
Taylor D, Ho BT (1977): Neurochemical effects of cocaine following acute and repeated injections. J Neurosci Re$ 3:95-101. Trinkoff AM, Ritter C, Anthony JC (1990): The prevalence and self-reported consequences of cocaine use. An exploratory and descriptive analysis. Drug Alcohol Depend 26:217-225. Van de Kar LD (1989): Neuroendocrine aspects of the serotonergic hypothesis of depression. Neuro$ci Biobehav Rev 13:237-246. 'Van de Kar LD (1991): Neuroendocrine pharmacology of serotonergic ($-HT) neurons. Annu Rev Pharmacol Toxicol 31:289-320. Van de Kar LD, Bethea CL (1982): Pharmacological evidence that serotonergic stimulation of prolactin secretion is mediated via the dorsal raphe nucleus. Neuroendocrinology 35:225-230. Van de Kar LD, Wilkinson CW, Skrobik Y, Brownfield MS, Ganong WF (1982): Evidence that serotonergic neurons in the dorsal raphe nucleus exert a stimulatory effect on the secretion of renin but not of corticosterone. Brain Res 235:233-243. Van de Kar LD, Kmeszi M, Bethea CL, Ganong WF 0985): Serotonergic stimulation of prolactin and oorticosterone secretion is mediated by different pathways from the mediobasal hypothal. amus. Neuroendocrinology 41:380-384. Van de Kar LD, Cames M, Maslowski RJ, et al (1989a): Neuroendocrine evidence for denervation supersensitivity of serotonin receptors. Effects of the 5-HT agonist RU 24969 on corticotropin, corticosterone, prolactin and renin secretion. J Pharmacol Exp Ther 251:428-434. Van de Kar LD, Lorens SA, Urban JH, Bethea CL (19891)): Effect of selective serotonin (5-HT) agonists and 5-HT2 antagonist on prolactin secretion. Neuropharmacology 28:299-305. Van de Kar LD, Rittenhouse PA, Iyer LV, et al (1990a): Chronic cocaine alters the neuroendocrine responses to serotonin (5-HT) agonists (Abst~ac0 FASEB J 4:A594--A594. Van de Kar LD, Urban JH, Brownfield MS (1990b): Serotonergic regulation of renin and vasopressin secretion. In Paoletti R, Vanhoutte P M, Brunello N, Maggi F M (eds), Serotonin: From Cell Biology to Pharmacology and Therapeutics. Dordrecht, The Netherlands: Kluwer Academic Publishers, pp 123-129. Van de Kar LD, Bonadorma AM, Rittenhouse PA (1992): Prior exposure to chronic cocaine inhibits the serotonergic stimulation of ACTH and corticosterone secretion. Neuropharmacology 31:169175.