Relationship between the serotonergic activity and reinforcing effects of a series of amphetamine analogs

Article abstract of DOI:10.1124/jpet.104.080101

It has been reported that among drugs with mixed actions on central nervous system monoamine systems, increased serotonergic activity is associated with decreased potency as a reinforcer. The present experiment was designed to examine this relationship for amphetamine analogs that varied in serotonin releasing potency and to evaluate whether serotonergic actions can affect reinforcing efficacy. Compounds PAL 313 and 314 are para- and meta-methylamphetamine, respectively. PAL 303 and 353 are para- and meta-fluoroamphetamine, respectively. All compounds had similar potencies as in vitro releasers of dopamine (DA) and norepinephrine (NE) but differed in potency for 5-hydroxytryptamine (serotonin) (5-HT) release [EC₅₀ (nanomolar) PAL 313 = 53.4; PAL 314 = 218; PAL 303 = 939; PAL 353 = 1937]. When made available to rhesus monkeys (Macaca mulatta) (n = 4) for self-administration under a fixed-ratio 25 schedule, all were positive reinforcers with biphasic dose-response functions (0.003-1.0 mg/kg) and were equipotent. PAL 313 was self-administered at a lower rate than the other compounds, which were indistinguishable. Under a progressive-ratio schedule (n = 5), all drugs were positive reinforcers. Dose-response functions increased to a maximum or were biphasic (0.01-1.0 mg/kg), and drugs were equipotent. At maximum, PAL 313 maintained less responding than ther PAL drugs, which maintained similar maxima. Thus, all compounds were positive reinforcers under both schedules, consistent with their potent DA actions. Responding was lower when 5-HT potency was higher and comparable with DA and NE potency. The results suggest that the mechanism for this effect involves a decrease in reinforcing potency and efficacy among monoamine releasing agents when 5-HT releasing potency is increased relative to DA.

Full text of DOI:10.1124/jpet.104.080101

0022-3565/05/3132-848–854
THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS
U.S. Government work not protected by U.S. copyright
JPET 313:848–854, 2005
Vol. 313, No. 2
80101/1198072
Printed in U.S.A.
Relationship between the Serotonergic Activity and Reinforcing
Effects of a Series of Amphetamine Analogs
S. Wee, K. G. Anderson, M. H. Baumann, R. B. Rothman, B. E. Blough, and
W. L. Woolverton
Department of Psychiatry and Pharmacology and Toxicology, University of Mississippi Medical Center, Jackson, Mississippi
(S.W., K.G.A., W.L.W.); Department of Psychology, West Virginia University, Morgantown, West Virginia (K.G.A.); Clinical
Psychopharmacology Section, Intramural Research Program, National Institute on Drug Abuse, Baltimore, Maryland (M.H.B.,
R.B.R.); and Center for Organic and Medicinal Chemistry, Science and Engineering Group, RTI International, Research Triangle
Park, North Carolina (B.E.B.)
Received November 4, 2004; accepted January 25, 2005
ABSTRACT
It has been reported that among drugs with mixed actions on biphasic dose-response functions (0.003–1.0 mg/kg) and were
central nervous system monoamine systems, increased sero- equipotent. PAL 313 was self-administered at a lower rate than
tonergic activity is associated with decreased potency as a the other compounds, which were indistinguishable. Under a
reinforcer. The present experiment was designed to examine progressive-ratio schedule (n ϭ 5), all drugs were positive
this relationship for amphetamine analogs that varied in sero- reinforcers. Dose-response functions increased to a maximum
tonin releasing potency and to evaluate whether serotonergic or were biphasic (0.01–1.0 mg/kg), and drugs were equipotent.
actions can affect reinforcing efficacy. Compounds PAL 313 At maximum, PAL 313 maintained less responding than other
and 314 are para- and meta-methylamphetamine, respectively. PAL drugs, which maintained similar maxima. Thus, all com-
PAL 303 and 353 are para- and meta-fluoroamphetamine, re- pounds were positive reinforcers under both schedules, con-
spectively. All compounds had similar potencies as in vitro sistent with their potent DA actions. Responding was lower
releasers of dopamine (DA) and norepinephrine (NE) but dif- when 5-HT potency was higher and comparable with DA and
fered in potency for 5-hydroxytryptamine (serotonin) (5-HT) re- NE potency. The results suggest that the mechanism for this
lease [EC50 (nanomolar) PAL 313 ϭ 53.4; PAL 314 ϭ 218; PAL effect involves a decrease in reinforcing potency and efficacy
303 ϭ 939; PAL 353 ϭ 1937]. When made available to rhesus among monoamine releasing agents when 5-HT releasing po-
monkeys (Macaca mulatta) (n ϭ 4) for self-administration under tency is increased relative to DA.
a fixed-ratio 25 schedule, all were positive reinforcers with
Among drugs that enhance monoaminergic neurotransmis- cocaine-like drugs in self-administration is positively corre-
sion in the central nervous system (CNS), especially psycho-
stimulants, there is considerable evidence that dopamine
(DA) plays an integral part in the reinforcing effects that
contribute to their self-administration and abuse (for re-
views, see Wise, 1978; Woolverton and Johnson, 1992; How-
ell and Wilcox, 2002). Observations that are consistent with
this hypothesis include the finding that the potency of
lated with their affinity in binding at the dopamine trans-
porters (DATs) in vitro (Ritz et al., 1987; Bergman et al.,
1989; Wilcox et al., 2000). Simple potency as a reinforcer,
however, does not seem to predict potential for abuse partic-
ularly well. Rather, abuse has been more strongly associated
with efficacy or strength as a reinforcer (Brady and Griffiths,
1976). Cocaine, for example, has a relatively weak affinity at
the DAT, but it is one of the most efficacious reinforcers in
This study was supported by National Institute on Drug Abuse Grants laboratory animals and is highly abused by humans. It seems
DA-10352 (to W.L.W.) and DA-12970 (to B.E.B.). W.L.W. is the recipient of
that factors other than potency for increasing DA neuro-
transmission must contribute to reinforcing efficacy and
abuse of compounds that bind the DAT.
Previous research has suggested several pharmacological
determinants of reinforcing efficacy. With regard to monoam-
National Institute on Drug Abuse Grant K05-DA15343. This work was previ-
ously presented at the 2004 meeting of the College on Problems of Drug
Dependence, 12–17 June 2004, San Juan, Puerto Rico.
Article, publication date, and citation information can be found at
http://jpet.aspetjournals.org.
doi:10.1124/jpet.104.080101.
ABBREVIATIONS: CNS, central nervous system; DA, dopamine; NE, norepinephrine; DAT, dopamine transporter; 5-HT, 5-hydroxytryptamine
(serotonin); SERT, serotonin transporter; HD-60, (Ϯ)-2␤-propanoyl-3␤-(2-isopropenyl)-tropane; FR, fixed ratio; PR, progressive ratio; GBR12935,
1-[2-(diphenylmethoxy)ethyl]-4-(3-phenylpropyl)-piperazine; LH, limited hold; TO, timeout; ANOVA, analysis of variance; MPPϩ, 1-methyl-4-
phenylpyridinium.
848

Self-Administration of Amphetamine Analogs
849
inergic actions, enhanced serotonergic activity may be nega- Monkey Rju2 became ill and had to be removed from the study before
all drugs had been tested. All monkeys were provided with sufficient
food to maintain stable body weight (120–180 g/day; Teklad 25%
Monkey Diet; Harlan/Teklad, Madison, WI) and had unlimited ac-
cess to water. Fresh fruit and a vitamin supplement were provided
daily and three times a week, respectively. Lighting was cycled to
maintain 16 h of light and 8 h of dark, with lights on at 6:00 AM.
The monkeys were individually housed in the experimental cubi-
cles (1.0 m3; Plaslabs, Lansing, MI). Each monkey was fitted with a
stainless steel harness attached by a tether to the rear wall of the
cubicle. The front door of the cubicle was made of transparent plastic
and the remaining walls were opaque. Two response levers (PRL-
001; BRS/LVE, Beltsville, MD) were mounted on the inside of the
door. Four jeweled stimulus lights, two red and two white, were
mounted above each lever. Drug injections were delivered by a peri-
staltic infusion pump (Cole-Parmer Instrument Co., Chicago, IL). A
Macintosh computer with custom interface and software controlled
all events in an experimental session.
Procedure. Monkeys were implanted with a silastic catheter
(0.26 cm o.d. ϫ 0.076 cm i.d.; Cole-Parmer Instrument Co.) into the
jugular (internal or external) or femoral vein under isoflurane anes-
thesia. Brachial veins were implanted with a microrenethane cath-
eter (0.08 in. o.d. ϫ 0.04 in. i.d.; Braintree Scientific, Braintree, MA)
heated and drawn to approximately half size. The proximal end of
the catheter was inserted into the vein and terminated in the vena
cava near the right atrium. The distal end was threaded subcutane-
ously to exit the back of the monkey, threaded through the spring
arm, out the rear of the cubicle, and connected to the peristaltic
pump. In the event of catheter failure, surgery was repeated using
another vein, after the veterinarian confirmed the health of the
monkey.
tively related to the self-administration of psychostimulants.
Compounds that selectively increase 5-hydroxytryptamine
(serotonin) (5-HT) neurotransmission have been found not
to maintain self-administration (Tessel and Woods, 1975;
Vanover et al., 1992; Howell and Byrd, 1995). Among am-
phetamine-like drugs that are not as selective for 5-HT ac-
tivity, Ritz and Kuhar (1989) reported a negative correlation
between potency as a reinforcer and binding affinity at the
5-HT transporter (SERT). The depletion of 5-HT by medial
forebrain bundle lesion with 5,7-dihydroxy-tryptamine seemed
to increase the reinforcing efficacy of cocaine in rats (Loh and
Roberts, 1990), whereas manipulations that increase CNS 5-HT
function can decrease cocaine self-administration (Smith et al.,
1986; Carroll et al., 1990; Howell and Byrd, 1995). Studies
using coadministered phentermine and fenfluramine as a pro-
totypical DA/5-HT releasing agents demonstrated that com-
pounds that increase both extracellular DA and 5-HT in rat
nucleus accumbens are not self-administered by rodents (Glatz
et al., 2002), do not induce conditioned place preference in rats
(Rea et al., 1998), and have low potential for abuse in human
subjects (Brauer et al., 1996). Reinforcing efficacy among a
series of cocaine analogs was shown to be negatively related to
the SERT potency relative to DAT (Roberts et al., 1999). In that
study, cocaine analogs with the highest relative SERT affinity
did not function as reinforcers.
Together, these data suggest that among drugs that in-
crease monoaminergic neurotransmission in the CNS, 5-HT
activity may reduce reinforcing efficacy and thereby reduce
self-administration. In a recent study with rhesus monkeys
(Macaca mulatta), Lile et al. (2003) reported robust self-
administration of a cocaine analog, HD-60, with at least
80-fold selectivity for the SERT relative to the DAT. This
same compound was not a positive reinforcer in rats in the
study of Roberts et al. (1999) (called WF-60). Thus, it is
possible that there are important species differences in the
influence of 5-HT activity on self-administration. In addition,
although Ritz and Kuhar (1989) proposed a negative rela-
tionship between 5-HT release and reinforcing potency, a
prospective study testing this hypothesis, or the relationship
of 5-HT release to reinforcing efficacy, has not been reported.
Experimental sessions began at noon each day and were con-
ducted 7 days per week. Thirty minutes before each session started,
catheters were filled with drugs for the sessions without infusing the
drugs into monkeys. At the start of a session, the white lights were
illuminated above both levers and pressing the right lever resulted
in the delivery of a drug injection for 10 s. During the injection, the
white lights were extinguished and the red lights were illuminated.
Pressing the left lever was counted but had no other programmed
consequence. After the session, catheters were filled with 0.9% saline
containing heparin (40 units/ml).
In baseline sessions, cocaine or saline was available for an injec-
tion. The baseline dose of cocaine or saline was initially available
under a double-alternation schedule, i.e., two consecutive daily co-
caine sessions were followed by two consecutive daily saline sessions.
The present experiments were therefore designed to evaluate When responding was stable (see below) for at least two consecutive
double-alternation sequences of cocaine and saline (i.e., eight ses-
sions), test sessions were inserted to the daily sequence between two
saline or two cocaine sessions. To prevent monkeys from learning
this session sequence, a randomly determined saline or cocaine base-
line session was inserted after every other test session. Thus, the
daily sequence of sessions was C, S, T, S, C, T, R, C, S, T, S, C, T, R,
where C, S, R, and T represent a cocaine baseline, a saline, a
randomly determined cocaine/saline, and a test session, respectively.
Drugs were tested in a different order across monkeys. All doses of
one compound were tested before moving on to the next compound.
For the first monkey tested with a given drug, doses were available
in an ascending order. For the other monkeys, doses were tested in a
random order. After a test session, a monkey was returned to base-
line conditions until responding for cocaine and saline again met
stability criteria or until a new stable baseline was established. All
doses were tested at least twice in each monkey, once with a saline
session the day before and once with a cocaine session the day before.
Fixed-ratio schedule. Drugs were made available under an FR
the relationship between 5-HT releasing potency, relative to
other monoamines, and reinforcing potency and efficacy in
rhesus monkeys. A series of amphetamine analogs with sim-
ilar in vitro potencies in releasing DA and NE, but varying
potencies in releasing 5-HT, were selected for testing.
Materials and Methods
All animal use procedures were approved by the University of
Mississippi Medical Center’s Animal Care and Use Committee and
were in accordance with National Institutes of Health guidelines.
Self-Administration Studies
Animals and Apparatus. The subjects were seven male rhesus
monkeys [88-14, AP01, AP78, and M341 for the fixed-ratio (FR)
study; AP01, M341, M1389, L500, and RJu2 for the progressive-ratio
(PR) study], weighing between 9.6 and 11.8 kg at the beginning of the
study. All the monkeys except 88-14 had histories of self-adminis- schedule that has been described previously (Wee and Woolverton,
tration of cocaine and/or other stimulants. Monkey 88-14 had previ- 2004). The response requirement was 25 lever presses per injection,
ously been trained in a drug discrimination paradigm to discriminate and each session lasted for 2 h. The baseline dose of cocaine was the
pentobarbital from saline and was naive to drug self-administration. dose that maintained the maximum responding in a dose-response

850
Wee et al.
function in all monkeys, i.e., 0.03 mg/kg/injection. Responding was
considered stable when the injections/session for cocaine varied no
more than Ϯ15% of the previous session and saline injections were
less than 15 injections/session, and there were no trends in the data.
During test sessions, one of various doses of the amphetamine ana-
logs (0.003–1.0 mg/kg) was made available for self-administration
under conditions identical to baseline conditions. When the data in
the two test sessions of a given dose were inconsistent (i.e., a dose
was a reinforcer in one test and not in the other or a reinforcer in
both tests but intake differed by more than 30 injections), the dose
was made available for at least four consecutive sessions and until
responding was stable.
Progressive-ratio schedule. Drugs also were made available to a
second group of monkeys in which responding was maintained under
a PR schedule of reinforcement comparable with that described by
Wilcox et al. (2000). The PR schedule consisted of 20 trials, with one
injection available per trial. The response requirement started at 100
responses per injection and doubled after every fourth trial. A subject
had 30 min to complete a trial [limited hold (LH) 30 min]. A trial
ended with a 10-s drug injection or the expiration of the LH. There
was a 30 min timeout (TO) after each trial. If the response require-
ment was not completed for two consecutive trials (i.e., the LH
expired) or the animal self-administered all 20 injections, the session
ended.
The baseline dose of cocaine was the lowest dose that maintained
the maximum injections in individual monkeys, i.e., 0.1 or 0.3 mg/
kg/injection. Responding was considered stable in baseline sessions
when injections/sessions varied by no more than two for both cocaine
and saline for at least two consecutive double-alternation sequences.
During test sessions, one of various doses of amphetamine and the
amphetamine analogs (0.003–1.0 mg/kg) was available for monkeys
under conditions identical to baseline sessions. When the two test
sessions of a dose showed high variability (two determinations Ն
mean Ϯ three injections) at the dose with the maximum mean
injections, the dose was redetermined twice, once after saline session
and once after cocaine baseline session. For drugs that exhibited
biphasic dose-response functions, the TO after injection was in-
creased to 60 min, and the two highest doses were tested again. This
adjustment was made out of concern that, under the 30-min TO, drug
accumulation over the session may have acted to suppress respond-
ing and thereby decreased maximum responding.
Data Analysis. The mean number of injections per session was
calculated individually from the two test sessions as a function of
dose. When a dose was tested in consecutive sessions under the FR
schedule, the last two test sessions were used in data analysis. When
a dose was retested under the PR schedule, the two retest sessions
were used in data analysis. The range of injections served as a
measure of variability in individual subjects. A dose of a drug was
considered to function as a reinforcer if the mean number of injec-
tions was above levels seen with saline, and the ranges did not
overlap.
the maximum injections in monkeys in which a drug served as a
reinforcer. The doses of the maximum injections in individual mon-
keys were averaged for a given drug and compared using a repeated
measures one-way ANOVA.
For the data under a PR schedule, the ED50 value was calculated
for each animal in which a drug served as a reinforcer using the
ascending limb of a dose-response function and nonlinear regression
analysis (GraphPad Prism 3.0). Mean ED50 values were calculated
for each drug by averaging the log values of ED50 values in all
monkeys in which the drug functioned as a reinforcer and taking the
antilog of that value. Repeated measures one-way ANOVA, using
multiple imputation to fill empty ED50 values in two monkeys for
statistical purpose, was used to compare ED50 values (GraphPad
Prism 3.0). The Student-Newman-Keuls was used as a post hoc test.
In Vitro Monoamine Release
In vitro release assays were conducted as described by Rothman et
al. (2003) using [3H]MPPϩ as the radioligand for both the DA and NE
release assays. Rat caudate (for DA release) or whole brain minus
cerebellum and caudate (for NE and 5-HT release) was homogenized
in ice-cold 10% sucrose containing 1 ␮M reserpine. Nomifensine (100
nM) and GBR12935 (100 nM) were added to the sucrose solution for
[3H]5-HT release experiments to block any potential [3H]5-HT re-
uptake into NE and DA nerve terminals. For the DA release assay,
100 nM desipramine and 100 nM citalopram were added to block
[3H]MPPϩ uptake into NE and 5-HT nerves. For the NE release
assay, 50 nM GBR12935 and 100 nM citalopram were added to block
[3H]MPPϩ uptake into DA and 5-HT nerves. After 12 strokes with a
Potter-Elvehjem homogenizer, homogenates were centrifuged at
1000g for 10 min at 0 to 4°C, and the supernatants were retained on
ice (synaptosomal preparation).
Synaptosomal preparations were incubated to steady state with 5
nM [3H]MPPϩ (60 min) or 5 nM [3H]5-HT (60 min) in Krebs-phos-
phate buffer (without bovine serum albumin, pH 7.4) that contained
154.4 mM NaCl, 2.9 mM KCl, 1.1 mM CaCl2, 0.83 mM MgCl2, 5 mM
glucose, 1 mg/ml ascorbic acid, and 50 ␮M pargyline plus 1 ␮M
reserpine in a polypropylene beaker with stirring at 25°C with the
appropriate blockers. After incubation to steady state, 850 ␮l of
synaptosomes preloaded with 3H ligand was added to 12 ϫ 75-mm
polystyrene test tubes that contained 150 ␮l of test drug in uptake
buffer plus 1 mg/ml bovine serum albumin. After 5 min ([3H]5-HT) or
30 min (NE and DA assays), the release reaction was terminated by
dilution with 4 ml of wash buffer (10 mM Tris-HCl, pH 7.4, contain-
ing 0.9% NaCl at 25°C) followed by rapid vacuum filtration over
Whatman GF/B filters using a Brandel harvester. The filters were
rinsed twice with 4 ml of wash buffer using the Brandel harvester,
and the retained tritium was counted by a Taurus liquid scintillation
counter at 40% efficiency after an overnight extraction in 3 ml of
Cytoscint (MP Biomedicals, Irvine, CA).
Data Analysis and Statistics. As described previously (Roth-
man et al., 1993), EC50 values for transporter assays were deter-
mined using the nonlinear least-squares curve fitting program
MLAB-PC (Civilized Software, Bethesda, MD). A correlation be-
tween in vitro measures and potency as a reinforcer under both FR
and PR schedules or between in vitro measures and the maximum
injections under both FR and PR schedules was evaluated using
Pearson correlation coefficient (GraphPad Prism 3.0 or Primer of
Biostatistics). When the data violated the assumptions for paramet-
ric analysis (i.e., equal variance), Spearman correlation coefficient
was used.
For both schedules, the group mean dose-response functions for
each drug were collapsed across the monkeys as to the dose of the
maximum injections, regardless of the absolute values of the doses.
This was done because each drug maintained qualitatively compa-
rable dose-response functions across monkeys but with the maxi-
mum responding at a different dose, in particular under the FR
schedule. Statistical analysis was done with these group means
normalized as to dose. That is, the mean maximum number of
injections was calculated for a drug by averaging the individual
maximum mean injections for the drug, regardless of dose, across
monkeys (Woolverton and Wang, 2004). Repeated measures one-way
analysis of variance (ANOVA) with the Student-Newman-Keuls as a
post hoc test was then used to assess statistically significant differ-
Drugs
Cocaine hydrochloride and amphetamine sulfate were provided by
ences among drugs in the maximum number of injections (GraphPad the National Institute on Drug Abuse (Rockville, MD). Amphetamine
Prism 3.0; GraphPad Software Inc., San Diego, CA).
Under the FR schedule, the potency of the compounds in self-
analogs PAL 314 (m-methylamphetamine), PAL 313 (p-methylam-
phetamine), PAL 353 (m-fluoroamphetamine), and PAL 303 (p-flu-
administration was compared based on the doses that maintained oroamphetamine) (Fig. 1) were synthesized using methodology de-

Self-Administration of Amphetamine Analogs
851
Fig. 1. Chemical structures of amphetamine and the sub-
stituted amphetamines tested.
scribed by Monte et al. (1997). The appropriate aldehydes were responding above saline levels at least at one dose in all
converted to their nitro-olefins using ammonium acetate and nitro-
ethane. These crude nitro-olefins were subsequently reduced with
lithium aluminum hydride to obtain the desired compounds. Each
compound was purified from the crude material as either a hydro-
chloride or fumarate salt. Purity was confirmed by combustion anal-
ysis and 1H NMR. For the self-administration study, drugs were
dissolved in 0.9% saline. Doses were expressed as the salt forms of
the drugs.
monkeys, and dose-response functions were biphasic. Mon-
key 88-14 was reliably more sensitive to all drugs and con-
sistently responded at higher rates than the other monkeys.
Based upon the dose that maintained maximum responding,
the PAL compounds did not differ significantly in potency
[Table 1; F(3,9) ϭ 2.72; p Ͼ 0.1]. The mean maximum re-
sponding maintained by PAL 313 was significantly lower
than that for PAL 314, PAL 303, and PAL 353 [Fig. 2;
F(3,12) ϭ 6.13; p Ͻ 0.05].
Results
Self-Administration under a Progressive-Ratio Sched-
ule. The baseline dose of cocaine (0.1 or 0.3 mg/kg/injection)
maintained a mean of 15.2 injections/session (S.E.M. 1.1),
whereas saline maintained a mean of 1.6 injections/session
(S.E.M. 0.2; Fig. 3). d-Amphetamine and PAL 303, 314, and 353
functioned as positive reinforcers in all monkeys. PAL 313 func-
tioned as a positive reinforcer in four of five monkeys (L500,
M1389, AP78, and RJu2). Monkey M341 initially showed high
variability at two of four doses of PAL 313. When retested, this
monkey did not self-administer PAL 313 at any dose. Dose-
response functions increased with dose over low-to-moderate
doses and were asymptotic or decreased again at higher doses.
PAL 313, PAL 314, PAL 303, and PAL 353 did not differ sig-
nificantly in potency, and d-amphetamine was approximately
6- to 12-fold more potent [Table 1; F(4,16) ϭ 17.7; p Ͻ 0.0001].
At maximum, d-amphetamine maintained more injections/ses-
sion than the PAL compounds, whereas PAL 313 maintained
fewer injections/session than the other compounds [Fig. 3;
F(3,12) ϭ 585; p Ͻ 0.001]. The maximum number of injections
of PAL 314, PAL 303, and PAL 353 were indistinguishable.
When drugs with biphasic dose-response functions, amphet-
amine, PAL 303, and PAL 353 were reexamined with a TO of 60
min, the shapes of the dose-response functions were unchanged;
i.e., they were asymptotic or biphasic. Furthermore, the maxi-
mum number of injections did not exceed those seen with the
shorter TO for any drug.
Self-Administration under a Fixed-Ratio Schedule.
In test sessions, the baseline dose of cocaine (0.03 mg/kg/
injection) maintained a mean of 54 injections/session (S.E.M.
10.3), whereas saline maintained a mean of four injections/
session (S.E.M. 1.5; Fig. 2). All PAL compounds maintained
Fig. 2. Self-administration of PAL compounds under an FR25 schedule of
reinforcement. Drugs were available for self-administration for 2 h/day.
Each data point represents the mean injections/session of each dose for
four rhesus monkeys, and vertical error bars represent the S.E.M. values.
The point above Sal or Coc represents self-administration of saline or the
baseline dose of cocaine in test sessions, respectively. Data were normal-
ized as to dose to adjust for individual differences in sensitivity. Max,
dose that maintained maximum injections in each animal; MaxϪ1, half-
log dose lower than Max; Maxϩ1, half-log dose higher than Max. ء, p Ͻ
0.05 compared with PAL 314, PAL 303, or PAL 353. Inset, dose-response
functions of PAL compounds. The drugs were tested in a different dose
range in one monkey because of different sensitivity to the drugs. Thus,
data points are the mean of three or four monkeys. When the data point
was from less than three monkeys, the number of subjects is indicated in
parentheses.
In Vitro Monoamine Release Assay. d-Amphetamine
and the PAL compounds released DA, NE, and 5-HT in a
concentration-dependent manner, and concentration-effect
functions were parallel (data not shown). All the compounds
showed comparable potencies releasing DA with the EC50
values between 8.0 and 51.5 nM (Table 2). Similarly, the
EC50 values for the PAL compounds did not differ for NE
release with the EC50 values between 7.2 and 28.0 nM (Table

852
Wee et al.
TABLE 1
Potencies and maximum responding of PAL compounds in self-administration
Data are mean Ϯ S.E.M., and numbers in parentheses are numbers of animals tested. For the FR schedule, the mean dose that maintained responding at the peak of the
biphasic curve was calculated for potency. For the PR schedule, an ED50 dose of a dose-response function was obtained in individual monkey and averaged across the monkeys
for the mean. Potency is expressed as micromoles per kilogram per injection. For the PR potency, log mean ED50 (milligrams per kilogram per injection) Ϯ S.E.M. was added
in parentheses to indicate variation. The maximum injection indicates the maximum injections per session.
Schedule
PAL 313
PAL 314
PAL 303
PAL 353
d-Amphetamine
FR
Potency
Maximum injection
0.5 Ϯ 0.22 (4)
16.8 Ϯ 3.0 (4)
0.23 Ϯ 0.09 (4)
43.8 Ϯ 12.2 (4)
0.13 Ϯ 0.03 (4)
56.1 Ϯ 13.9 (4)
0.09 Ϯ 0.02 (4)
56.1 Ϯ 17.7 (4)
0.05 Ϯ 0.03a (4)
37.3 Ϯ 9.2a (4)
PR
Potency
Maximum injection
0.48 (–0.9 Ϯ 0.18) (4) 0.46 (–0.9 Ϯ 0.13) (5) 0.26 (–1.3 Ϯ 0.14) (5) 0.26 (–1.2 Ϯ 0.18) (5) 0.04*** (–1.8 Ϯ 0.11) (4)
8.9 Ϯ 1.7 (5) 11.6 Ϯ 1.2 (5) 12.6 Ϯ 1.3 (5) 11.9 Ϯ 1.1 (5) 13.8 Ϯ 0.6 (5)
*** p Ͻ 0.0001 compared with each of the PAL compounds.
a Data from Woolverton et al. (2001).
Relationship between Self-Administration and in
Vitro Effects. For the FR schedule, there was no significant
correlation between potency as a reinforcer (Table 1) and in
vitro potency (Table 2) releasing DA (r ϭ 0.55, df ϭ 3, p ϭ
0.34) or 5-HT (r ϭ Ϫ0.84, df ϭ 3, p ϭ 0.08). However, potency
as a reinforcer was positively correlated with the ratio of the
in vitro potencies releasing DA/5-HT (r ϭ 0.98, df ϭ 3, p ϭ
0.004). There was no correlation with the ratio of the in vitro
potencies releasing DA/NE (r ϭ 0.78, df ϭ 3, p ϭ 0.12).
Maximum responding under the FR schedule was not corre-
lated with any measure of in vitro potency. For the PR
schedule, there was also a positive correlation between the
ratio of the in vitro potencies releasing DA/5-HT and rein-
forcing potency (r ϭ 0.97, df ϭ 4, p ϭ 0.02). However, potency
as a reinforcer was not correlated with any other measure of
in vitro potency. Efficacy as a reinforcer under the PR sched-
ule was not correlated with any measure of in vitro potency.
Fig. 3. Self-administration of d-amphetamine and PAL compounds under
a progressive-ratio schedule of reinforcement. Each data point represents
the mean injections/session for four (d-amphetamine) or five monkeys (all
PAL compounds), and vertical error bars represent the S.E.M. values.
The point above Sal or Coc represents self-administration of saline in test
sessions or cocaine during baseline sessions, respectively. Data were
normalized as to dose to adjust for individual differences in sensitivity.
Max, dose of the maximum injections in each animal. Max-1, half-log dose
lower than Max; Max-2, full-log dose lower than Max. ءءء, p Ͻ 0.001
compared with each of the other compounds. Inset, dose-response func-
tion of amphetamine and PAL compounds. The drugs were tested in
Discussion
As predicted based upon their nanomolar potencies as re-
leasers of DA, all of the tested compounds functioned as
positive reinforcers in all monkeys responding under an FR
schedule of reinforcement. Dose-response functions under
the FR schedule were biphasic, as is typically seen under
different dose ranges in monkeys because of different sensitivities to the these conditions (Young and Herling, 1986; Bergman et al.,
drugs. Thus, data points are the mean of four or five monkeys. When the
1989). Although not tested in the present study, d-amphet-
amine has been found to function as a reinforcer under con-
ditions similar to those used here (Woolverton et al., 2001).
Additionally, all but one of the compounds, PAL 313, served
as a positive reinforcer in all monkeys responding under a PR
schedule. PAL 313 was a reinforcer in four of the five mon-
data point was pooled from less than four monkeys, the number of
subjects is indicated in parentheses.
TABLE 2
In vitro potency as releasers of monoamine neurotransmitters
Values are mean Ϯ S.D. for three experiments.
EC50
keys tested. The compounds did not differ in potency in either
self-administration assay and had comparable potencies as
DA releasers in vitro, results that are at least consistent with
the conclusion that DA release is involved in their reinforcing
effects. Although a similar argument could be made for NE,
NE agonists have not previously been reported to function as
positive reinforcers (Woolverton, 1987; Wee and Woolverton,
2004). The reinforcing effects of these compounds in the
present experiment are consistent with full or partial am-
phetamine-like discriminative stimulus effects that have
been reported previously for PAL 303, PAL 313, and m-
methylamphetamine (Higgs and Glennon, 1990; Marona-
Lewicka et al., 1995). Together, these results suggest am-
phetamine type of abuse potential of the compounds.
DA/NE DA/5-HT
͓3H͔DA
͓3H͔NE
͓3H͔5-HT
nM
d-Amphetamine 8.0 Ϯ 0.43 7.2 Ϯ 0.44 1756 Ϯ 94a
1.1
1.5
1.8
1.8
2.0
0.004
0.01
0.05
0.15
0.83
PAL 353
PAL 303
PAL 314
PAL 313
24.2 Ϯ 1.1 16.1 Ϯ 1.7 1937 Ϯ 202
51.5 Ϯ 1.7 28.0 Ϯ 1.8
33.3 Ϯ 1.3 18.3 Ϯ 1.4
44.1 Ϯ 2.6 22.2 Ϯ 1.3
939 Ϯ 76
218 Ϯ 22
53.4 Ϯ 4.1
a The value for 5-HT release was published previously (Rothman et al., 2001).
2). As for 5-HT release, the order of the potencies was PAL
313 Ͼ PAL 314 Ͼ PAL 303 Ͼ PAL 353 Ͼ d-amphetamine.
Thus, the DA/NE potency ratio ranged between 1.1 and 2.0,
whereas the DA/5-HT potency ratio ranged between 0.004
and 0.83.
The lack of a significant negative correlation between po-
tency as a reinforcer under the FR schedule and potency for

Self-Administration of Amphetamine Analogs
853
releasing 5-HT in vitro is inconsistent with the hypothesis differ on these measures, raises the possibility that there
proposed by Ritz and Kuhar (1989). It is worth noting, nev- may be a threshold of 5-HT activity for these effects to be
ertheless, that the correlation for this relationship ap- evident. Additional research would be required to determine
proached statistical significance (p ϭ 0.08) and may have whether this threshold is for absolute 5-HT potency or 5-HT
achieved statistical significance with additional degrees of potency relative to DA potency.
freedom. Additionally, the most potent 5-HT releaser, PAL
There are at least two possible accounts of the reduced
313, was clearly less potent than the least potent 5-HT re- reinforcing efficacy of compounds with increased 5-HT/DA
leaser, d-amphetamine. As an alternative view, the correla- potency. It is possible that increased 5-HT release opposes
tion between the ratio of DA/5-HT releasing potency in vitro the activation of DA systems that are involved in the rein-
and potency as a reinforcer was significant, suggesting that forcing effect. Such an effect would not be apparent in the in
the mix of dopaminergic and serotonergic actions is a deter- vitro results, but it could be measured in vivo. Indeed, in an
minant of potency as a reinforcer. Results were similar for experiment using in vivo microdialysis in rats (Clark et al.,
the PR schedule in that potency as a 5-HT releaser was not 2004), PAL 313 showed diminished efficacy for increasing
correlated with potency as a reinforcer among the PAL com- extracellular DA relative to the other compounds, even with
pounds, although again the least potent 5-HT releaser d- comparable DA releasing potency in vitro. In research with
amphetamine was the most potent reinforcer. As for the FR monoamine uptake blockers, Howell and colleagues have
schedule, the ratio of DA/5-HT in vitro potencies predicted published data consistent with an opposing effect of 5-HT
potency as a reinforcer under the PR.
activity on DA activity. More specifically, cocaine-induced
The compounds differed in the maximum rate of self-ad- increases in extracellular DA concentrations, measured in
ministration under the FR schedule, to the extent that rate of vivo, were diminished after pretreatment with the selective
responding was lower for PAL 313 than for the other com- 5-HT reuptake blocker alaproclate in monkeys (Czoty et al.,
pounds. PAL 313 was the most potent in vitro 5-HT releaser 2002). Alaproclate also decreased the metabolic effects of
and had the highest in vitro DA/5-HT potency ratio. Although cocaine in monkeys, as measured by positron emission to-
this result raises the possibility of a relationship between mography (Howell et al., 2002). The present results are con-
DA/5-HT activity and reinforcing efficacy, this conclusion sistent with the conclusion that 5-HT actions oppose DA
cannot be drawn unambiguously. Rate of self-administration actions involved in reinforcing effects.
under a simple FR schedule is determined not only by the
The possibility also should be considered that 5-HT activity
reinforcing effect of a drug but also by other effects of the could act to decrease self-administration of a DA agonist via
drug that can affect rate. These have been termed “nonspe- a punishment mechanism. It has been demonstrated that i.v.
cific” effects and may include effects on motor, sensory, or drug injections can punish ongoing behavior maintained by
integrative function; satiety; or punishing effects. Any or all food delivery (Katz and Goldberg, 1986; Woolverton, 2003).
of these might be influenced by an increase in 5-HT activity. Delivery of response contingent electric shock also can de-
The PR schedule, by increasing response requirement for crease self-administration of cocaine (Bergman and Johan-
successive injections, measures maximum behavioral output son, 1981). In preliminary experiments (W. L. Woolverton,
maintained by an injection. From this information an esti- unpublished data), the addition of histamine to an injection
mate of a drug’s efficacy as a reinforcer, or its maximum of cocaine can act to punish cocaine self-administration. Con-
reinforcing effect, can be inferred. The present PR, by arrang- sidering the clinical reports of unpleasant side effects of
ing a TO after each injection, is designed to minimize the drugs that increase 5-HT neurotransmission (Sternbach,
influence of nonspecific effects on responding. The TO allows 1991; Birmes et al., 2003; Finfgeld, 2004), it is conceivable
these effects to dissipate, at least partially, between injec- that the addition of 5-HT actions could suppress self-admin-
tions. Results from the PR schedule are consistent with a istration of a DA agonist.
conclusion that the reinforcing efficacy of each of the PAL
It is interesting to note that PAL 313 was approximately
compounds was lower than that of d-amphetamine. More- equipotent in releasing all three monoamines. In this effect,
over, the reinforcing efficacy of PAL 313 was lower than that it parallels cocaine, which is equipotent as a blocker of up-
of the other compounds. This efficacy relationship was di- take of DA, NE, and 5-HT in vitro. Although cocaine dose-
rectly related to 5-HT potency, with d-amphetamine the least response functions were not determined in the present ex-
potent and PAL 313 the most potent 5-HT releaser. The periments, cocaine has been repeatedly tested under these
implication is that increased potency as a 5-HT releaser is conditions (Wilcox et al., 2000; Woolverton and Wang, 2004)
associated with diminished reinforcing efficacy.
and is clearly a more efficacious reinforcer than PAL 313.
Overall, results from both schedules of reinforcement lend The reason(s) for this difference is unclear, although it does
support to the hypothesis that potency as a 5-HT releaser indicate the relationship between monoamine activity and
contributes negatively to potency as a reinforcer, but with the reinforcing efficacy is not simple. This conclusion is under-
refinement that the DA/5-HT potency ratio may be a better scored by the reports, noted previously, of self-administration
predictor than 5-HT potency alone. The observation that the of HD-60 by monkeys (Lile et al., 2003) but not by rats
compound with the highest 5-HT releasing potency (PAL (Roberts et al., 1999). HD-60 is a high-potency 5-HT uptake
313) was the least efficacious reinforcer, whereas the com- blocker relative to its potency as a DA uptake blocker. It has
pound with the lowest 5-HT releasing potency (d-amphet- been reported that monoamine releasers can produce a ro-
amine) was the most efficacious reinforcer, supports the con- bust increase in the extracellular monoamines in vitro,
clusion of Roberts et al. (1999) for DA uptake blockers. The whereas reuptake transporter inhibitors of similar potencies
observation that the compound (PAL 313) with the highest were less effective (Rothman and Baumann, 2002). Mono-
5-HT potency had the lowest potency and efficacy as a rein- amine transporter inhibitors depend on the basal firing rates
forcer, whereas the other PAL compounds did not apparently of neurons to increase extracellular monoamines, and at the

854
Wee et al.
Bush E, and Nichols DE (1997) Dihydrobenzofuran analogues of hallucinogens. 4.
Mescaline derivatives. J Med Chem 40:2997–3008.
Rea WP, Rothman RB, and Shippenberg TS (1998) Evaluation of the conditioned
reinforcing effects of phentermine and fenfluramine in the rat: concordance with
clinical studies. Synapse 30:107–111.
Ritz MC and Kuhar MJ (1989) Relationship between self-administration of amphet-
amine and monoamine receptors in brain: comparison with cocaine. J Pharmacol
Exp Ther 248:1010–1017.
Ritz MC, Lamb RJ, Goldberg SR, and Kuhar MJ (1987) Cocaine receptors on
dopamine transporters are related to self-administration of cocaine. Science (Wash
DC) 237:1219–1223.
same time, seem to stimulate the negative feedback inhibi-
tion, actions that may contribute to a relatively smaller in-
crease in the neurotransmitters. This difference may contrib-
ute to the greater influence of serotonergic activity on the
self-administration of PAL 313 than on the self-administra-
tion of cocaine.
Acknowledgments
Roberts DC, Phelan R, Hodges LM, Hodges MM, Bennett B, Childers S, and Davies
H (1999) Self-administration of cocaine analogs by rats. Psychopharmacology
144:389–397.
We acknowledge the expert advice of Dr. Warren May on the
statistical analysis and the technical assistance of Hamilton McGee,
Stephanie Dyson, and Joy Flenner.
Rothman RB and Baumann MH (2002) Serotonin releasing agents. Neurochemical,
therapeutic and adverse effects. Pharmacol Biochem Behav 71:825–836.
Rothman RB, Lewis B, Dersch CM, Xu H, Radesca L, de Costa BR, Rice KC, Kilburn
References
RB, Akunne HC, and Pert
A (1993) Identification of a GBR12935 homolog,
LR1111, which is over 4000-fold selective for the dopamine transporter, relative to
serotonin and norepinephrine transporters. Synapse 14:34–39.
Rothman RB, Vu N, Partilla JS, Roth BL, Hufeisen SJ, Compton-Toth BA, Birkes J,
Young R, and Glennon RA (2003) In vitro characterization of ephedrine-related
stereoisomers at biogenic amine transporters and the receptosome reveals selec-
tive actions as norepinephrine transporter substrates. J Pharmacol Exp Ther
307:138–145.
Smith FL, Yu DS, Smith DG, Leccese AP, and Lyness WH (1986) Dietary tryptophan
supplements attenuate amphetamine self-administration in the rat. Pharmacol
Biochem Behav 25:849–855.
Sternbach H (1991) The serotonin syndrome. Am J Psychiatry 148:705–713.
Tessel RE and Woods JH (1975) Fenfluramine and N-ethyl amphetamine: compar-
ison of the reinforcing and rate-decreasing actions in the rhesus monkey. Psycho-
pharmacologia 43:239–244.
Vanover KE, Nader MA, and Woolverton WL (1992) Evaluation of the discriminative
stimulus and reinforcing effects of sertraline in rhesus monkeys. Pharmacol Bio-
chem Behav 41:789–793.
Wee S and Woolverton WL (2004) Evaluation of the reinforcing effects of atomoxetine
in monkeys: comparison to methylphenidate and desipramine. Drug Alcohol De-
pend 75:271–276.
Wilcox KM, Rowlett JK, Paul IA, Ordway GA, and Woolverton WL (2000) On the
relationship between the dopamine transporter and the reinforcing effects of local
anesthetics in rhesus monkeys: practical and theoretical concerns. Psychopharma-
cology 153:139–147.
Wise RA (1978) Catecholamine theories of reward: a critical review. Brain Res
152:215–247.
Woolverton WL (1987) Evaluation of the role of norepinephrine in the reinforcing
effects of psychomotor stimulants in rhesus monkeys. Pharmacol Biochem Behav
26:835–839.
Bergman J and Johanson CE (1981) The effects of electric shock on responding
maintained by cocaine in rhesus monkeys. Pharmacol Biochem Behav 14:423–426.
Bergman J, Madras BK, Johnson SE, and Spealman RD (1989) Effects of cocaine and
related drugs in nonhuman primates. III. Self-administration by squirrel mon-
keys. J Pharmacol Exp Ther 251:150–155.
Birmes P, Coppin D, Schmitt L, and Lauque D (2003) Serotonin syndrome: a brief
review. CMAJ 168:1439–1442.
Brady JV and Griffiths RR (1976) Behavioral procedures for evaluating the relative
abuse potential of CNS drugs in primates. Fed Proc 35:2245–2253.
Brauer LH, Johanson CE, Schuster CR, Rothman RB, and de Wit H (1996) Evalu-
ation of phentermine and fenfluramine, alone and in combination, in normal,
healthy volunteers. Neuropsychopharmacology 14:233–241.
Carroll ME, Lac ST, Asencis M, and Kragh R (1990) Fluoxetine reduces intravenous
cocaine self-administration in rats. Pharmacol Biochem Behav 35:237–244.
Clark RD, Baumann MH, Blough BE, and Rothman RB (2004) Effects of monoamine
releasing agents on extracellular dopamine and serotonin in rat nucleus accum-
bens: Relationship to locomotor activation (Abstract). College on Problems of Drug
Dependence; 2004 June 12–17; San Juan, Puerto Rico. Poster 42.
Czoty PW, Ginsburg BC, and Howell LL (2002) Serotonergic attenuation of the
reinforcing and neurochemical effects of cocaine in squirrel monkeys. J Pharmacol
Exp Ther 300:831–837.
Finfgeld DL (2004) Serotonin syndrome and the use of SSRIs. J Psychosoc Nurs Ment
Health Serv 42:16–20.
Glatz AC, Ehrlich M, Bae RS, Clarke MJ, Quinlan PA, Brown EC, Rada P, and
Hoebel BG (2002) Inhibition of cocaine self-administration by fluoxetine or d-
fenfluramine combined with phentermine. Pharmacol Biochem Behav 71:197–204.
Higgs RA and Glennon RA (1990) Stimulus properties of ring-methyl amphetamine
analogs. Pharmacol Biochem Behav 37:835–837.
Howell LL and Byrd LD (1995) Serotonergic modulation of the behavioral effects of
cocaine in the squirrel monkey. J Pharmacol Exp Ther 275:1551–1559.
Howell LL, Hoffman JM, Votaw JR, Landrum AM, Wilcox KM, and Lindsey KP
(2002) Cocaine-induced brain activation determined by positron emission tomog-
raphy neuroimaging in conscious rhesus monkeys. Psychopharmacology 159:154–
160.
Woolverton WL (2003) A novel choice method for studying drugs as punishers.
Pharmacol Biochem Behav 76:125–131.
Woolverton WL, Hecht GS, Agoston GE, Katz JL, and Newman AH (2001) Further
studies of the reinforcing effects of benztropine analogs in rhesus monkeys. Psy-
chopharmacology 154:375–382.
Woolverton WL and Johnson KM (1992) Neurobiology of cocaine abuse. Trends
Pharmacol Sci 13:193–200.
Woolverton WL and Wang Z (2004) Relationship between injection duration, trans-
porter occupancy and reinforcing strength of cocaine. Eur J Pharmacol 486:251–
257.
Howell LL and Wilcox KM (2002) Functional imaging and neurochemical correlates
of stimulant self-administration in primates. Psychopharmacology 163:352–361.
Katz JL and Goldberg SR (1986) Effects of H1-receptor antagonists on responding
punished by histamine injection or electric shock presentation in squirrel mon-
keys. Psychopharmacology 90:461–467.
Lile JA, Wang Z, Woolverton WL, France JE, Gregg TC, Davies HM, and Nader MA
(2003) The reinforcing efficacy of psychostimulants in rhesus monkeys: the role of
pharmacokinetics and pharmacodynamics. J Pharmacol Exp Ther 307:356–366.
Loh EA and Roberts DCS (1990) Break-points on a progressive ratio schedule
reinforced by intravenous cocaine increase following depletion of forebrain seroto-
nin. Psychopharmacology 101:262–266.
Young AM and Herling S (1986) Drugs as reinforcers: studies in laboratory animals,
in Behavioral Analysis of Drug Dependence (Goldberg SR and Stolerman IP eds) pp
9–67, Academic Press, New York.
Address correspondence to: Dr. William L. Woolverton, Division of Neuro-
biology and Behavior Research, Department of Psychiatry and Human Behav-
ior, University of Mississippi Medical Center, 2500 N. State St., Jackson, MS
39216-4505. E-mail: wwoolverton@psychiatry.umsmed.edu
Marona-Lewicka D, Rhee G, Sprague JE, and Nichols DE (1995) Psychostimulant-
like effects of p-fluoroamphetamine in the rat. Eur J Pharmacol 287:105–113.
Monte AP, Waldman SR, Marona-Lewicka D, Wainscott DB, Nelson DL, Sanders-