Directory UMM :Data Elmu:jurnal:B:Biological Psichatry:Vol48.Issue7.2000:
Proton Magnetic Resonance Spectroscopy of Human
Basal Ganglia: Response to Cocaine Administration
James D. Christensen, Marc J. Kaufman, Blaise deB. Frederick, Stephanie L. Rose,
Constance M. Moore, Scott E. Lukas, Jack H. Mendelson, Bruce M. Cohen, and
Perry F. Renshaw
Background: Proton magnetic resonance spectroscopy
was used to determine the effects of intravenous cocaine
or placebo administration on human basal ganglia water
and metabolite resonances.
Methods: Long echo time, proton magnetic resonance
spectra of water and intracellular metabolites were continuously acquired from an 8-cm3 voxel centered on the
left caudate and putamen nuclei before, during, and after
the intravenous administration of cocaine or a placebo in
a double-blind manner.
Results: Cocaine, at both 0.2 and 0.4 mg/kg, did not alter
the peak area for water. Cocaine at 0.2 mg/kg induced
small and reversible increases in choline-containing compounds and N-acetylaspartate peak areas. Cocaine at 0.4
mg/kg induced larger and more sustained increases in
choline-containing compounds and N-acetylaspartate
peak areas. No changes in either water or metabolite
resonances were noted following placebo administration.
Conclusions: These increases in choline-containing compounds and N-acetylaspartate peak areas may reflect
increases in metabolite T2 relaxation times secondary to
osmotic stress and/or increased phospholipid signaling
within the basal ganglia following cocaine administration.
This is the first report of acute, drug-induced changes in
the intensity of human brain proton magnetic resonance
spectroscopy resonance areas. Biol Psychiatry 2000;48:
685– 692 © 2000 Society of Biological Psychiatry
Key Words: Cocaine, osmotic stress, basal ganglia, nuclear magnetic resonance, phospholipid turnover, substance abuse
From the Brain Imaging Center (JDC, MJK, BdBF, SLR, CMM, BMC, PFR) and
Alcohol & Drug Abuse Research Center (MJK, SEL, JHM), McLean Hospital,
and the Consolidated Department of Psychiatry, Harvard Medical School
(JDC, MJK, BdBF, SLR, CMM, SEL, JHM, BMC, PFR), Belmont,
Massachusetts.
Address reprint requests to Perry F. Renshaw, M.D., Ph.D., McLean Hospital, Brain
Imaging Center, 115 Mill Street, Belmont MA 02478-9106.
Received November 23, 1999; revised March 20, 2000; accepted April 7, 2000.
© 2000 Society of Biological Psychiatry
Introduction
H
uman brain imaging studies have documented cocaine-induced alterations in cerebral metabolism and
perfusion (Gollub et al 1998; Johnson et al 1998; Kaufman
et al 1998a, 1998b; London et al 1990; Mathew et al 1996;
Pearlson et al 1993; Wallace et al 1996); however, there
remains a paucity of data on the biochemical responses of
the human brain to cocaine administration. Proton magnetic resonance spectroscopy (1H MRS) offers a means to
characterize some of cocaine’s effects by allowing noninvasive detection of brain water and proton-containing
metabolites present in millimolar concentrations.
Metabolites that are detected in the human brain in vivo
include cytosolic choline-containing compounds (Cho),
creatine and phosphocreatine (Cre), and N-acetylaspartate
(NAA). The 1H MRS choline resonance arises primarily
from phosphocholine (PC) and glycerophosphocholine
(GPC; Bluml et al 1999). Changes in the Cho peak may
reflect receptor-mediated phospholipid turnover (Exton
1994) or alterations in regional glucose utilization (Duc et
al 1997). The Cre peak, which contains contributions from
both creatine and phosphocreatine (an important energy
reservoir molecule), remains relatively stable across physiologic states (Petroff et al 1988, 1989). N-Acetylaspartate
has been suggested to serve as a marker of neuronal
density (Urenjak et al 1993). Its levels are reduced by the
injection of kainic acid, an excitotoxin, in the rat brain
(Guimaraes et al 1995) and in neurodegenerative processes (Tsai and Coyle 1995; Tzika et al 1993). Recent
work suggests that NAA also participates as a neuronal
osmolyte (Taylor et al 1995). These compounds can thus
be used as markers of cerebral biochemical function.
The purpose of this study was to determine whether
cocaine-induced biochemical changes in the basal ganglia
are detectable as changes in the 1H MRS signals of
proton-containing metabolites. Cocaine receptors associated with the dopamine transporter are thought to contribute to cocaine’s behavioral effects (Giros et al 1996).
Consequently, the basal ganglia was selected as a region of
interest, since it is enriched with dopamine terminals and
0006-3223/00/$20.00
PII S0006-3223(00)00897-0
686
J.D. Christensen et al
BIOL PSYCHIATRY
2000;48:685– 692
transporter sites and, in animals, has been shown to
undergo cocaine-induced biochemical changes. We hypothesized that cocaine would induce an increase in the
Cho peak area, by elevating striatal dopamine and,
thereby, increasing phospholipid turnover through secondmessenger signaling pathways.
Methods and Materials
All study subjects were men aged 28 6 5 years (mean 6 SD) and
were interviewed by a psychiatrist (PFR) to rule out the presence
of an Axis I disorder other than a past history of substance abuse.
No study subject met criteria for substance dependence. Ongoing
use of alcohol and tobacco averaged 6 6 6 drinks/week and 14 6
28 cigarettes/week; 19 of the 30 subjects did not smoke. All
subjects reported occasional cocaine use, 13 6 14 lifetime
exposures, with intranasal administration being the most common route of exposure. We have recently reported that illicit
cocaine use patterns are not altered in this population following
participation in an imaging study during which cocaine is
administered by slow intravenous infusion (Kaufman et al 2000).
No subject was taking any psychoactive medications at the time
of study.
All subjects underwent a complete physical examination and
were judged to be in good health. All structural brain MRI
examinations were normal. Immediately before the study all
subjects tested negative for recent cocaine, amphetamine, opiate,
phencyclidine, benzodiazepine, and barbiturate exposure by a
urine test (Triage Test, Biosite Diagnostics, San Diego) and for
alcohol exposure by a breathalyzer test (Alco Sensor III, Intoximeters, St. Louis). Subjects were implanted with an 18-gauge
angiocath in the antecubital vein for intravenous drug administration. Subjects were fit with noninvasive, cardiovascular monitoring equipment (In Vivo Research, Orlando) for continuous
vital sign measurements. This included a four-lead electrocardiogram, infrared fingertip pulse oxymetry oxygen detection system, and automated blood pressure cuff. Vital signs were
recorded at 5-min intervals throughout the study.
Magnetic resonance spectroscopy was performed using a
1.5-T clinical magnetic resonance scanner with a proton quadrature head coil (General Electric Medical Systems, Milwaukee).
The point-resolved spectroscopy localization technique (Moonen
et al 1989) was used to acquire proton (1H) signals from an
8-cm3 cubic voxel centered on the head of the left caudate
nucleus and putamen. Acquisition parameters were as follows:
interpulse repetition time (TR) 5 2 sec, echo time (TE) 5 135
msec, number of points 5 1024, spectral width 5 2 kHz, and
phase cycles 5 2. Water suppression was achieved using a series
of chemical shift–selective radiofrequency pulses and gradient
dephasing. Signal pairs consisting of an unsuppressed water
signal (16 averages) and a water-suppressed metabolite signal
(128 averages) were obtained in a sequential manner over a
5-min period. The TR and TE values were the same for both the
suppressed and unsuppressed spectra.
Three baseline signal pairs were obtained before drug challenges. Subjects (n 5 30) then were administered a placebo
intravenously (saline; n 5 10) or cocaine at low (0.2 mg/kg; n 5
10) or high (0.4 mg/kg; n 5 10) doses as a slow push over 1 min
in a double-blind manner. Seven additional signal pairs were
acquired immediately following placebo/cocaine administration,
yielding 10 pairs of water and water-suppressed spectra for each
subject, with the exception of six subjects from whom only
water-suppressed spectra were obtained.
Magnetic resonance spectroscopy data was processed in a
fully automated manner requiring no subjective operator input
(Webb et al 1994). The Cho, Cre, NAA, and water peaks were fit
to Lorentzian line shapes using an interative Marquardt algorithm for peak areas. A singular value decomposition algorithm
was used to determine peak line widths (van den Boogaart et al
1994). For this analysis, fitting of the raw, unfiltered, and
unapodized free induction decay was done in the time domain
with no Fourier reconstruction, eliminating the artificial line
width distortion that would arise from apodization.
Repeated-measures analysis of variance (ANOVA) was used
to detect between-group differences in normalized water and
metabolite peak areas. Because we found a moderate degree of
between-subject variance in the time course of cocaine’s effects,
we also performed an area under the curve (AUC) analysis. The
AUC was calculated as the sum of the normalized metabolite
resonance intensities for each subject. Before summation the
normalized baseline value (1.00) was subtracted from each
postdrug measurement. The AUC was defined as
S [(Postdrug Time 1 2 1.00), (Postdrug Time 2 2 1.00) . . . ,
(Postdrug Time 7 2 1.00)]
Missing data values (one postdrug water-suppressed metabolite
spectrum in three subjects and a single unsuppressed water signal
in one subject) were estimated by interpolation (Winer 1971).
Results
Baseline cardiovascular parameters were normal in all
subjects and statistically equivalent across experimental
groups, with heart rate (HR) 5 60 6 8 beats per minute
(mean 6 SD), systolic pressure (Sys) 5 118 6 11 mm Hg,
diastolic pressure 5 70 6 8 mm Hg, and mean arterial
pressure 5 87 6 8 mm Hg. Cocaine induced transient,
dose-dependent increases in HR and Sys, with peak values
occurring about 15–20 min after drug administration
(Figure 1).
Cocaine administration was not associated with statistically significant changes in water peak areas. The variance of the water resonance area was, in each case, less
than 2%, documenting the consistency and reproducibility
of the technique.
A stack plot of representative metabolite spectra is
presented in Figure 2. Significant dose by time effects
were found for both Cho (p , .02) and NAA (p , .001).
Cocaine (0.2 mg/kg) induced a transient increase in the
Cho peak area, which reached a maximum of about 115%
of basal level 10 min after drug administration and then
declined to basal levels (Figure 3). In contrast, the higher
Proton MRS of Basal Ganglia after Cocaine
BIOL PSYCHIATRY
2000;48:685– 692
687
Figure 2. A stack plot showing sequential magnetic resonance
spectroscopy spectra from a single subject during the 15-min
baseline period before cocaine injection and over the 35-min
period following injection. Each spectrum was obtained without
changing any experimental parameters, and reflects the same
number of averages. There is no individual scaling between
peaks. The scale, though consistent, is arbitrary. Cho, cytosolic
choline-containing compounds; Cre, creatine and phosphocreatine; NAA, N-acetylaspartate.
Figure 1. Effects of cocaine/placebo administered at time 0 on
cardiovascular parameters. Shown are means 6 SEs of 10
subjects per dose level. Cocaine induced transient dose-dependent increases in heart rate (F 5 21.8, p , .0001; top) and
systolic blood pressure (F 5 13.0, p 5 .0001; middle). A
significant time effect was detected for diastolic pressure (F 5
4.3, p , .0001; bottom). Data are presented as mean 6 SEM and
values have been normalized to 100% for baseline measures.
cocaine dose (0.4 mg/kg) resulted in a progressive increase
in the Cho peak, which reached a maximum of about
135% of basal level 20 min following drug administration
and persisted to the end of the experiment (Figure 3, top).
For the NAA peak, 0.2 mg/kg cocaine also induced a
minimal metabolite increase, whereas 0.4 mg/kg cocaine
induced a larger increase that peaked at about 120% of
basal levels 20 min after drug administration and remained
elevated for the rest of the study (Figure 3, middle). A
slight but nonsignificant increase in the Cre peak area was
observed with high-dose cocaine (Figure 3, bottom).
Figure 4 displays a scattergram of areas under the curve
for the Cho and NAA resonance lines over time following
cocaine administration. Note that for the Cho resonance
(Figure 4, top) four of 10 subjects had an AUC . 0,
compared with seven of 10 subjects receiving 0.2 mg/kg
cocaine and nine of 10 subjects receiving 0.4 mg/kg. By
x2, the number of subjects with an increased AUC was
larger in the cohort receiving cocaine (x2 5 4.8, p , .03).
For the NAA resonance (Figure 4, bottom), four of 10
subjects had an AUC . 0, compared with five of 10
subjects receiving 0.2 mg/kg cocaine and eight of 10
subjects receiving 0.4 mg/kg. By x2, there was a weak
trend for the number of subjects with an increased AUC to
be larger in the cohort receiving cocaine (x2 5 1.7, p ,
.20). In both instances, the number of subjects with an
increased resonance intensity increased with dose. For the
Cre resonance, no trends between resonance area and
cocaine data were noted (data not shown).
Metabolite line widths were unchanged by cocaine
administration; repeated-measures ANOVA analysis detected no dose or time effects of cocaine on the Cho (5.1 6
0.8 Hz), Cre (6.1 6 1.8 Hz), and NAA (5.1 6 1.3 Hz) line
widths (mean 6 SD). This suggests that any confound in
the data resulting from subject motion is minimal.
Subject motion could also influence the observed MRS
signal amplitudes, if the volume of interest shifted over
time in such a manner as to include different amounts of
the lateral ventricles. Movement would also likely influence the shim and, therefore, the degree of water suppres-
688
J.D. Christensen et al
BIOL PSYCHIATRY
2000;48:685– 692
Figure 4. Scattergram for the area under the curve (AUC) for the
cytosolic choline-containing compounds and N-acetylaspartate
(NAA) resonance intensities following cocaine administration.
The numbers of subjects with AUC values . 0 for each dose of
drug or placebo are indicated at the top of each panel.
Figure 3. Proton metabolite peak area changes induced by
intravenous cocaine/placebo administration at time 0. Cocaine
induced dose-dependent (F 5 3.3, p 5 .05) and time-dependent
(F 5 2.5, p , .01) increases in the cytosolic choline-containing
compounds (Cho) peak area (top). Cocaine induced dosedependent increases in the N-acetylaspartate (NAA) peak area
(F 5 5.1, p , .02; middle). Cocaine administration was not
associated with statistically significant alterations in the creatine
and phosphocreatine (Cre) resonance (bottom). Data are presented as means 6 SDs.
sion; however, these effects of subject motion would not
account for the differential changes in signal intensity that
were observed in the NAA, Cho, and Cre resonance areas.
Although plasma cocaine levels were not measured in
this study, plasma analysis with gas chromatography in
separate subject groups receiving cocaine via an identical
administration protocol revealed that 0.2 (n 5 6) and 0.4
(n 5 6) mg/kg cocaine produced peak plasma levels 10
min after administration averaging 110 and 230 ng/mL,
respectively (Mendelson et al 1999).
Discussion
Our results demonstrate dose- and time-dependent effects
of cocaine on basal ganglia metabolite peak areas. These
effects temporally paralleled neuroendocrine and cardiovascular changes and plasma cocaine concentrations in
separate subject groups receiving cocaine via an identical
administration protocol as previously described (Mendelson et al 1999). In those studies, cocaine pharmacokinetics
Proton MRS of Basal Ganglia after Cocaine
following intravenous administration of 0.2 and 0.4 mg/kg
to men were determined; Tmax (and T1/2) were 6.7 min (45
min) and 8.0 min (48 min), respectively (Mendelson et al
1999). A comparable time course for cocaine-induced
striatal dopamine increases has been reproduced in both
rodents and nonhuman primates. Microdialysis studies in
rats (Bradberry et al 1993) and in primates (Iyer et al
1995) suggest that striatal dopamine levels peak 10 –20
minutes following the intravenous administration of cocaine and remain elevated for at least 100 min following
injection.
Cocaine’s effects on the Cho and NAA peak areas may
reflect changes in transverse relaxation times (T2*),
caused by osmotic stress and increased intracellular water
content. Cocaine inhibits the Na1/K1-adenosinetriphosphatase (Na1/K1-ATPase; Lien et al 1994), probably
through its action as an indirect agonist of dopamine D1
receptors (Bertorello et al 1990). Since volume regulation
is dependent upon Na1/K1-ATPase activity (Olson et al
1986), inhibition of this enzyme could lead to ion imbalance, osmotic stress, and cell swelling, as both sodium and
water move into the intracellular space (Hertz et al 1992).
Metabolite relaxation times can be greatly affected by
such changes in the cellular environment. In this regard,
osmotic stress secondary to Na1/K1-ATPase inhibition
and secondary energy failure has been associated with
enhanced metabolite mobility and increased Cho and
NAA relaxation times in some reports (Cady 1996; Cady
et al 1994), but not all (Fujimori et al 1998). Such an effect
on NAA relaxation has also been observed in brain edema
(Kamada et al 1994). Importantly, NAA has recently been
shown to serve as a neuronal osmolyte (Taylor et al 1995),
and GPC, which contributes substantially to the 1H MRS
Cho peak (see below), has been shown to function as a
renal osmolyte (Wolff and Balaban 1990).
Alternatively, cocaine may alter metabolite concentrations. For example, cocaine-stimulated phosphatidylcholine (PtdCho; brain concentration ; 21 mmol/L) hydrolysis might result in increased levels of PC (0.57 mmol/L)
and/or GPC (0.94 mmol/L; Bluml et al 1999; Klein et al
1993). With these concentration estimates, a 35% increase
in the 1H MRS Cho resonance would require the hydrolysis of 0.53 mmol/L, or approximately 2.5% of cellular
PtdCho. Turnover rates for PtdCho are difficult to measure
in vivo and may vary with regional cerebral activity;
however, Purdon and Rapoport (1998) have noted that the
turnover rates for fatty acids in phospholipids may be 100
times faster than previously thought. The relatively delayed time frame for the cocaine-induced Cho increase is
consistent with the general temporal sequence of phospholipid turnover, in which phosphatidylinositol turnover is
often the initial response, followed by PtdCho turnover
(Exton 1994).
BIOL PSYCHIATRY
2000;48:685– 692
689
We did not observe a statistically significant increase in
the intensity of the Cre resonance; however, across a
number of studies (for review, see Kreis 1997), T2*
relaxation times for the Cre resonance are substantially
shorter than those of Cho and NAA at 1.5 T. Thus,
increases in the intensity of the Cre resonance due to an
increase in intracellular water would be smaller than those
predicted for the Cho or NAA resonances. It may also be
the case that the 10% increase observed in the Cre
resonance following 0.4 mg/kg cocaine would have
reached statistical significance had a larger number of
subjects been studied.
To date, acute, drug-induced changes in the intensities
of 1H MRS resonance lines have not been reported in the
peer-reviewed literature; however, in general agreement
with the present results, Li and colleagues (1998) have
presented a preliminary report documenting a statistically
significant, 15% increase in NAA/Cre and a nonsignificant, 10% increase in Cho/Cre 12 min after the administration of a comparable dose of cocaine (40 mg/70 kg) to
a cohort of cocaine-dependent subjects. These data were
obtained from a 5-cm3 volume centered on the left
thalamus.
Since the present data were collected at a single TE, it
is not possible to conclude whether changes in Cho and
NAA peak areas resulted from changes in concentration,
changes in relaxation times, or both effects. Kreis (1997)
has reported that the mean T2* for NAA across multiple
brain regions is 376 msec. An increase in T2* to approximately 500 msec would be needed to increase the NAA
signal intensity by 10%. Of note, Cady et al (Cady 1996;
Cady et al 1994) have reported increases of this magnitude
in brain metabolite T2* values during secondary energy
failure; however, the fact that the intensities of two
intracellular resonance lines increased, as well as the
relatively large size of these increases, tend to argue in
favor of a change in relaxation times. A 35% increase in
the Cho resonance and an 20% increase in the NAA
resonance would correspond to increases in these metabolite pools of at least 500 mmol/L and 1 mmol/L, respectively, which are not likely. To the extent that the
increased Cho and NAA resonance intensities can be
directly attributed to dopamine release within the basal
ganglia, 1H MRS may provide a unique window into
central nervous system dopaminergic neurotransmission.
Increases in the intensity of intracellular metabolites
due to osmotic stress would be consistent with the intriguing possibility that a change in basal ganglia volume might
be detected following acute cocaine administration. We
are currently investigating this issue. The chronic administration of dopamine antagonists to individuals with
schizophrenia has been noted to increase the volume of
both the caudate and the putamen (for review, see Gur et
al 1998); however, changes in basal ganglia volume have
690
J.D. Christensen et al
BIOL PSYCHIATRY
2000;48:685– 692
not been associated with the long-term administration of
dopaminergic agonists and efforts to detect basal ganglia
pathology in human cocaine users have produced mixed
results (e.g., Kish et al 1999; Little et al 1998; Staley et al
1997). An alternative experimental approach would be to
use diffusion-weighted imaging (Le Bihan et al 1986) to
detect increased intracellular water, which should lead to
increased water diffusion.
Other factors may also contribute to the increased Cho
and NAA resonance intensities. For example, cocaine
administration to nonhuman primates has been associated
with a 13–22% decrease in basal ganglia cerebral metabolic rate (Lyons et al 1996). Limited available data
suggest that local cerebral metabolic rates, as determined
by positron emission tomography, are inversely related to
the intensity of the 1H MRS choline resonance. Another
factor that may play a role is cocaine-induced vasoconstriction. The intravenous administration of cocaine has
been reported to cause decreases in cerebral blood flow
(Gollub et al 1998; Wallace et al 1996) and volume
(Kaufman et al 1998a). Alterations in cerebral blood flow
and blood volume can lead to vascular susceptibility
changes (i.e., BOLD effects) that may increase (if the local
concentration of deoxyhemoglobin is decreased) or decrease (if the concentration of deoxyhemoglobin is increased) the signal intensity of both water and nearby
tissue metabolites. In our study the signal intensity of
water, which traverses the intravascular space, was not
increased while the intensity of intracellular metabolites,
NAA and Cho, was elevated. This observation tends to
argue against a BOLD effect. Of concern, our data would
be consistent with a decrease in blood flow to an extent
that causes basal ganglia ischemia (Cady 1996; Cady et al
1994). Therefore, further work to clarify the mechanism
by which intracellular metabolite resonance intensities are
increased may be warranted.
With regard to the effects of chronic cocaine on basal
ganglia 1H MRS metabolites, Li and coworkers (1999)
have reported that there are no differences in 1H MRS
metabolite ratios between cocaine users and comparison
subjects. In contrast, a 17% reduction in NAA/Cr was
noted in the left thalamus. Chang and colleagues (1997)
have reported increased parietal white matter Cre and
myo-inositol concentrations in heavy cocaine users. In a
follow-up study (Chang et al 1999), decreased NAA and
increased myo-inositol were noted in frontal lobe gray and
white matter in a cohort of abstinent subjects with a
history of crack cocaine dependence. In our study the use
of a 135-msec TE precluded the accurate assessment of the
myo-inositol resonance. Thus, it is uncertain how the
acute, cocaine-induced increases in 1H MRS resonance
intensities are related to the neurochemical consequences
of chronic heavy drug use.
In summary, intravenous cocaine administration transiently induced dose-dependent increases of basal ganglia
Cho and NAA peak areas. This study’s results are suggestive of cocaine-induced osmotic stress, but are also consistent with an increase in phospholipid turnover within
the basal ganglia. Further studies characterizing proton
metabolite changes following cocaine administration are
warranted to clarify this issue and to determine whether
the observed changes are relevant to the physiologic and
behavioral effects of cocaine.
This research was supported by National Institute on Drug Abuse Grants
Nos. DA09448, DA04059, DA00064, DA00115, DA00329, and
DA00343.
We gratefully acknowledge the excellent technical assistance provided
by Ms. Kim Appelmans.
JDC’s current affiliation: Department of Psychiatry, University of
Louisville, Louisville, Kentucky.
References
Bertorello AM, Hopfield JF, Aperia A, Greengard P (1990):
Inhibition by dopamine of (Na1/K1)ATPase activity in
neostriatal neurons through D1 and D2 dopamine receptor
synergism. Nature 347:386 –388.
Bluml S, Seymour KJ, Ross BD (1999): Developmental changes
in choline- and ethanolamine-containing compounds measured with proton-decoupled 31P MRS in in vivo human
brain. Magn Reson Med 42: 643– 654.
Bradberry CW, Nobiletti JB, Elsworth JD, Murphy B, Jatlow P,
Roth RH (1993): Cocaine and cocaethylene: Microdialysis
comparison of brain drug levels and effects on dopamine and
serotonin. J Neurochem 60:1429 –1435.
Cady EB (1996): Metabolite concentrations and relaxation in
perinatal cerebral hypoxic-ischemic injury. Neurochem Res
21:1043–1052.
Cady EB, Lorek A, Penrice J, Wylezinska M, Cooper CE, Brown
GC, et al (1994): Brain-metabolite transverse relaxation times
in magnetic resonance spectroscopy increase as adenosine
triphosphate depletes during secondary energy failure following acute hypoxia-ischaemia in the newborn piglet. Neurosci
Lett 182:201–204.
Chang L, Ernst T, Strickland TL, Mehringer M (1999): Gender
effects on persistent cerebral metabolite changes in the frontal
lobes of abstinent cocaine users. Am J Psychiatry 156:716 –
722.
Chang L, Mehringer CM, Ernst T, Rosemarie M, Myers H,
Forney D, Satz P (1997): Neurochemical alterations in
asymptomatic abstinent cocaine users: A proton magnetic
resonance spectroscopy study. Biol Psychiatry 42:1105–1114.
Duc CO, Weber AH, Trabesinger AH, Meier D, Boesiger P
(1997): Recycling the cholines. In: International Society of
Magnetic Resonance in Medicine VI. Berkeley, CA: ISMRM,
1210.
Exton JH (1994): Phosphatidylcholine breakdown and signal
transduction. Biochem Biophys Acta 1212:26 – 42.
Fujimori H, Michaelis T, Wick M, Frahm J (1998): Proton T2
Proton MRS of Basal Ganglia after Cocaine
relaxation of cerebral metabolites during transient global
ischemia in rat brain. Magn Reson Med 39:647– 650.
Giros B, Jaber M, Jones SR, Wightman RM, Caron MO (1996):
Hyperlocomotion and indifference to cocaine and amphetamine in mice lacking the dopamine transporter. Nature
379:606 – 612.
Gollub RL, Breiter HC, Kantor H, Kennedy D, Gastfriend D,
Mathew RT, et al (1998): Cocaine decreases cortical cerebral
blood flow but does not obscure regional activation in
functional magnetic resonance imaging in human subjects.
J Cereb Blood Flow Metab 18:724 –734.
Guimaraes AR, Schwartz P, Prakash MR, Carr CA, Berger UV,
Jenkins BG, et al (1995): Quantitative in vivo 1H nuclear
magnetic resonance spectroscopic imaging of neuronal loss in
rat brain. Neuroscience 69:1095–1101.
Gur RE, Maany BA, Mozley PD, Swanson C, Bilker W, Gur RC
(1998): Subcortical MRI volumes in neuroleptic-naive and
treated patients with schizophrenia. Am J Psychiatry 155:
1711–1717.
Hertz L, Code WE, Sykova E (1992): Ions, water and energy in
brain cells: A synopsis of interrelations. Can J Physiol
Pharmacol 70:S100 –S106.
Iyer RN, Nobiletti JB, Jatlow PI, Bradberry CW (1995): Cocaine
and cocaethylene: effects on extracellular dopamine in the
primate. Psychopharmacology 120:150 –155.
Johnson B, Lamki L, Fang B, Barron B, Wagner L, Wells L, et
al (1998): Demonstration of dose dependent global and
regional cocaine-induced reductions in brain blood flow using
a novel approach to quantitative single photon emission
computerized tomography. Neuropsychopharmacology 18:
377–384.
Kamada K, Houkin K, Hida K, Matsuzawa H, Iwasaki Y, Abe H,
Nakada T (1994): Localized proton spectroscopy of focal
brain pathology in hurnans: Significant effects of edema on
spin-spin relaxation time. Magn Reson Med 31:537–540.
Kaufman MJ, Levin JM, Kukes TJ, Villafuerte RA, Hennen J,
Lukas SE, et al (2000): Illicit cocaine use patterns in intravenous-naive cocaine users following investigational intravenous cocaine administration. Drug Alcohol Depend 58:35– 42.
Kaufman MJ, Levin JM, Maas LC, Rose SL, Kukes TJ, Mendelson JH, et al (1998a): Cocaine decreases relative cerebral
blood volume in humans: A dynamic susceptibility contrast
magnetic resonance imaging study. Psychopharmacology
138:76 – 81.
Kaufman MJ, Levin JM, Ross MH, Lange N, Rose SL, Kukes
TJ, et al (1998b): Cocaine induced cerebral vasoconstriction
detected in humans with magnetic resonance angiography.
JAMA 279:376 –380.
Kish SJ, Kalasinsky KS, Furukawa Y, Guttman M, Ang L, Li L,
et al (1999): Brain choline acetyltransferase activity in
chronic, human users of cocaine, methamphetamine, and
heroin. Mol Psychiatry 4:26 –32.
Klein J, Gonzalez R, Koppen A, Loffelholz K (1993): Free
choline and choline metabolites in rat brain and body fluid:
Sensitive determination and implications for choline supply
to the brain. Neurochem Int 31:293–300.
Kreis R (1997): Quantitative localized 1H MR spectroscopy for
clinical use. Prog Nucl Magn Reson Spectrosc 11:155–195.
Le Bihan D, Breton E, Lallemand D, Grenier P, Cabanis E,
BIOL PSYCHIATRY
2000;48:685– 692
691
Laval-Jeantet M (1986): MR imaging of intravoxel incoherent motions: Applications to diffusion and perfusion in
neurologic disorders. Radiology 161:401– 407.
Li S-J, Wang Y, Pankiewicz J, Stein EA (1999): Neurochemical
adaptation to cocaine abuse: Reduction of N-acetyl aspartate
in thalamus of human cocaine users. Biol Psychiatry 45:
1481–1487.
Li S-J, Wang Y, Risinger R, Rainey C, Harsch H, Pankiewicz J,
et al (1998): Acute cocaine administration induces NAA
increase detected by 1H MRS. In: International Society of
Magnetic Resonance in Medicine VI. Berkeley, CA: ISMRM,
1714.
Lien R, Mishra OP, Graham E, Delivoria-Papadopoulos M,
Anday EK (1994): Alteration of brain cell membrane function
following cocaine exposure in the fetal guinea pig. Brain Res
637:249 –254.
Little KY, McLaughlin DP, Zhang L, McFinton PR, Dalack GW,
Cook EH Jr, et al (1998): Brain dopamine transporter messenger RNA and binding sites in cocaine users: A postmortem
study. Arch Gen Psychiatry 550:793–799.
London ED, Cascella NG, Wong DF, Phillips RL, Dannals RF,
Links JM, et al (1990): Cocaine-induced reduction of glucose
utilization in human brain. A study using positron emission
tomography and [fluorine 18]-fluorodeoxyglucose. Arch Gen
Psychiatry 47:567–574.
Lyons D, Friedman DP, Nader MA, Porrino LJ (1996): Cocaine
alters cerebral metabolism within the ventral striatum and
limbic cortex of monkeys. J Neurosci 16:1230 –1238.
Mathew RJ, Wilson WH, Lowe JV, Humphries D (1996): Acute
changes in cranial blood flow after cocaine hydrochloride.
Biol Psychiatry 40:609 – 616.
Mendelson JH, Mello NK, Sholar MB, Siegel AJ, Kaufman MJ,
Levin JM, et al (1999): Cocaine pharmacokinetics in men and
in women during the follicular and luteal phases of the
menstrual cycle. Neuropsychopharmacology 21:294 –303.
Moonen CT, von Kienlin M, van Zijl PC, Cohen J, Gillen J, Daly
P, Wolf G (1989): Comparison of single-shot localization
methods (STEAM and PRESS) for in vivo proton NMR
spectroscopy. NMR Biomed 2:201–208.
Olson J, Sankar R, Holtzman D, James A, Fleischhacker D
(1986): Energy-dependent volume regulation in primary cultured cerebral astrocytes. J Cell Physiol 128:209 –215.
Pearlson GD, Jeffery PJ, Harris GJ, Ross CA, Fischman MW,
Camargo EE (1993): Correlation of acute cocaine-induced
changes in local cerebral blood flow with subjective effects.
Am J Psychiatry 150:495– 497.
Petroff OAC, Ogino T, Alger JR (1988): High-resolution proton
magnetic resonance spectroscopy of rabbit brain: Regional
metabolite levels and postmortem change. J Neurochem
51:163–171.
Petroff OAC, Spencer DD, Alger JR, Prichard JW (1989):
High-field proton magnetic resonance spectroscopy of human
cerebrum during surgery for epilepsy. Neurology 39:1197–
1202.
Purdon AD, Rapoport SI (1998): Energy requirements for two
aspects of phospholipid metabolism in mammalian brain.
Biochem J 335:313–318.
Staley JK, Talbot JZ, Ciliax BJ, Miller GW, Levey AI, Kung
692
BIOL PSYCHIATRY
2000;48:685– 692
MP, et al (1997): Radioligand binding and immunoautoradiographic evidence for a lack of toxicity to dopaminergic nerve
terminals in human cocaine overdose victims. Brain Res
274:219 –229.
Taylor DL, Davies SEC, Obrenovitch TP, Doheny MH, Patsalos
PN, Clark JB, Symon L (1995): Investigation into the role of
N-acetylaspartate in cerebral osmoregulation. J Neurochem
65:275–281.
Tsai G, Coyle JT (1995): N-Acetylaspartate in neuropsychiatric
disorders. Prog Neurobiol 46:531–540.
Tzika AA, Ball WSL, Vigneron DB, Dunn RS, Kirks DR (1993):
Clinical proton MR spectroscopy of neurodegenerative disease in childhood. AJNR Am J Neuroradiol 14:1267–1281.
Urenjak J, Williams SR, Gadian DG, Noble M (1993): Proton
nuclear magnetic resonance spectroscopy unambiguously
identifies different neural cell types. J Neurosci 13:981–989.
J.D. Christensen et al
van den Boogaart A, Ala-Korpela M, Jokisaari J, Griffiths JR
(1994): Time and frequency domain analysis of NMR data
compared: an application to 1D 1H spectra of lipoproteins.
Magn Reson Med 31:347–358.
Wallace EA, Wisniewski G, Zubal G, van Dyck CH, Pfau SE,
Smith EO, et al (1996): Acute cocaine effects on absolute
cerebral blood flow. Psychopharmacology 128:17–20.
Webb PG, Sailasuta N, Kohler SJ, Raidy T, Moats RA, Hurd RE
(1994): Automated single-voxel proton MRS: Technical development and multisite verification. Magn Reson Med 31:
365–373.
Winer BJ (1971): Statistical Principles in Experimental Design.
New York: McGraw-Hill.
Wolff SD, Balaban RS (1990): Regulation of the predominant
renal medullary organic solutes in vivo. Annu Rev Physiol
52:727–746.
Basal Ganglia: Response to Cocaine Administration
James D. Christensen, Marc J. Kaufman, Blaise deB. Frederick, Stephanie L. Rose,
Constance M. Moore, Scott E. Lukas, Jack H. Mendelson, Bruce M. Cohen, and
Perry F. Renshaw
Background: Proton magnetic resonance spectroscopy
was used to determine the effects of intravenous cocaine
or placebo administration on human basal ganglia water
and metabolite resonances.
Methods: Long echo time, proton magnetic resonance
spectra of water and intracellular metabolites were continuously acquired from an 8-cm3 voxel centered on the
left caudate and putamen nuclei before, during, and after
the intravenous administration of cocaine or a placebo in
a double-blind manner.
Results: Cocaine, at both 0.2 and 0.4 mg/kg, did not alter
the peak area for water. Cocaine at 0.2 mg/kg induced
small and reversible increases in choline-containing compounds and N-acetylaspartate peak areas. Cocaine at 0.4
mg/kg induced larger and more sustained increases in
choline-containing compounds and N-acetylaspartate
peak areas. No changes in either water or metabolite
resonances were noted following placebo administration.
Conclusions: These increases in choline-containing compounds and N-acetylaspartate peak areas may reflect
increases in metabolite T2 relaxation times secondary to
osmotic stress and/or increased phospholipid signaling
within the basal ganglia following cocaine administration.
This is the first report of acute, drug-induced changes in
the intensity of human brain proton magnetic resonance
spectroscopy resonance areas. Biol Psychiatry 2000;48:
685– 692 © 2000 Society of Biological Psychiatry
Key Words: Cocaine, osmotic stress, basal ganglia, nuclear magnetic resonance, phospholipid turnover, substance abuse
From the Brain Imaging Center (JDC, MJK, BdBF, SLR, CMM, BMC, PFR) and
Alcohol & Drug Abuse Research Center (MJK, SEL, JHM), McLean Hospital,
and the Consolidated Department of Psychiatry, Harvard Medical School
(JDC, MJK, BdBF, SLR, CMM, SEL, JHM, BMC, PFR), Belmont,
Massachusetts.
Address reprint requests to Perry F. Renshaw, M.D., Ph.D., McLean Hospital, Brain
Imaging Center, 115 Mill Street, Belmont MA 02478-9106.
Received November 23, 1999; revised March 20, 2000; accepted April 7, 2000.
© 2000 Society of Biological Psychiatry
Introduction
H
uman brain imaging studies have documented cocaine-induced alterations in cerebral metabolism and
perfusion (Gollub et al 1998; Johnson et al 1998; Kaufman
et al 1998a, 1998b; London et al 1990; Mathew et al 1996;
Pearlson et al 1993; Wallace et al 1996); however, there
remains a paucity of data on the biochemical responses of
the human brain to cocaine administration. Proton magnetic resonance spectroscopy (1H MRS) offers a means to
characterize some of cocaine’s effects by allowing noninvasive detection of brain water and proton-containing
metabolites present in millimolar concentrations.
Metabolites that are detected in the human brain in vivo
include cytosolic choline-containing compounds (Cho),
creatine and phosphocreatine (Cre), and N-acetylaspartate
(NAA). The 1H MRS choline resonance arises primarily
from phosphocholine (PC) and glycerophosphocholine
(GPC; Bluml et al 1999). Changes in the Cho peak may
reflect receptor-mediated phospholipid turnover (Exton
1994) or alterations in regional glucose utilization (Duc et
al 1997). The Cre peak, which contains contributions from
both creatine and phosphocreatine (an important energy
reservoir molecule), remains relatively stable across physiologic states (Petroff et al 1988, 1989). N-Acetylaspartate
has been suggested to serve as a marker of neuronal
density (Urenjak et al 1993). Its levels are reduced by the
injection of kainic acid, an excitotoxin, in the rat brain
(Guimaraes et al 1995) and in neurodegenerative processes (Tsai and Coyle 1995; Tzika et al 1993). Recent
work suggests that NAA also participates as a neuronal
osmolyte (Taylor et al 1995). These compounds can thus
be used as markers of cerebral biochemical function.
The purpose of this study was to determine whether
cocaine-induced biochemical changes in the basal ganglia
are detectable as changes in the 1H MRS signals of
proton-containing metabolites. Cocaine receptors associated with the dopamine transporter are thought to contribute to cocaine’s behavioral effects (Giros et al 1996).
Consequently, the basal ganglia was selected as a region of
interest, since it is enriched with dopamine terminals and
0006-3223/00/$20.00
PII S0006-3223(00)00897-0
686
J.D. Christensen et al
BIOL PSYCHIATRY
2000;48:685– 692
transporter sites and, in animals, has been shown to
undergo cocaine-induced biochemical changes. We hypothesized that cocaine would induce an increase in the
Cho peak area, by elevating striatal dopamine and,
thereby, increasing phospholipid turnover through secondmessenger signaling pathways.
Methods and Materials
All study subjects were men aged 28 6 5 years (mean 6 SD) and
were interviewed by a psychiatrist (PFR) to rule out the presence
of an Axis I disorder other than a past history of substance abuse.
No study subject met criteria for substance dependence. Ongoing
use of alcohol and tobacco averaged 6 6 6 drinks/week and 14 6
28 cigarettes/week; 19 of the 30 subjects did not smoke. All
subjects reported occasional cocaine use, 13 6 14 lifetime
exposures, with intranasal administration being the most common route of exposure. We have recently reported that illicit
cocaine use patterns are not altered in this population following
participation in an imaging study during which cocaine is
administered by slow intravenous infusion (Kaufman et al 2000).
No subject was taking any psychoactive medications at the time
of study.
All subjects underwent a complete physical examination and
were judged to be in good health. All structural brain MRI
examinations were normal. Immediately before the study all
subjects tested negative for recent cocaine, amphetamine, opiate,
phencyclidine, benzodiazepine, and barbiturate exposure by a
urine test (Triage Test, Biosite Diagnostics, San Diego) and for
alcohol exposure by a breathalyzer test (Alco Sensor III, Intoximeters, St. Louis). Subjects were implanted with an 18-gauge
angiocath in the antecubital vein for intravenous drug administration. Subjects were fit with noninvasive, cardiovascular monitoring equipment (In Vivo Research, Orlando) for continuous
vital sign measurements. This included a four-lead electrocardiogram, infrared fingertip pulse oxymetry oxygen detection system, and automated blood pressure cuff. Vital signs were
recorded at 5-min intervals throughout the study.
Magnetic resonance spectroscopy was performed using a
1.5-T clinical magnetic resonance scanner with a proton quadrature head coil (General Electric Medical Systems, Milwaukee).
The point-resolved spectroscopy localization technique (Moonen
et al 1989) was used to acquire proton (1H) signals from an
8-cm3 cubic voxel centered on the head of the left caudate
nucleus and putamen. Acquisition parameters were as follows:
interpulse repetition time (TR) 5 2 sec, echo time (TE) 5 135
msec, number of points 5 1024, spectral width 5 2 kHz, and
phase cycles 5 2. Water suppression was achieved using a series
of chemical shift–selective radiofrequency pulses and gradient
dephasing. Signal pairs consisting of an unsuppressed water
signal (16 averages) and a water-suppressed metabolite signal
(128 averages) were obtained in a sequential manner over a
5-min period. The TR and TE values were the same for both the
suppressed and unsuppressed spectra.
Three baseline signal pairs were obtained before drug challenges. Subjects (n 5 30) then were administered a placebo
intravenously (saline; n 5 10) or cocaine at low (0.2 mg/kg; n 5
10) or high (0.4 mg/kg; n 5 10) doses as a slow push over 1 min
in a double-blind manner. Seven additional signal pairs were
acquired immediately following placebo/cocaine administration,
yielding 10 pairs of water and water-suppressed spectra for each
subject, with the exception of six subjects from whom only
water-suppressed spectra were obtained.
Magnetic resonance spectroscopy data was processed in a
fully automated manner requiring no subjective operator input
(Webb et al 1994). The Cho, Cre, NAA, and water peaks were fit
to Lorentzian line shapes using an interative Marquardt algorithm for peak areas. A singular value decomposition algorithm
was used to determine peak line widths (van den Boogaart et al
1994). For this analysis, fitting of the raw, unfiltered, and
unapodized free induction decay was done in the time domain
with no Fourier reconstruction, eliminating the artificial line
width distortion that would arise from apodization.
Repeated-measures analysis of variance (ANOVA) was used
to detect between-group differences in normalized water and
metabolite peak areas. Because we found a moderate degree of
between-subject variance in the time course of cocaine’s effects,
we also performed an area under the curve (AUC) analysis. The
AUC was calculated as the sum of the normalized metabolite
resonance intensities for each subject. Before summation the
normalized baseline value (1.00) was subtracted from each
postdrug measurement. The AUC was defined as
S [(Postdrug Time 1 2 1.00), (Postdrug Time 2 2 1.00) . . . ,
(Postdrug Time 7 2 1.00)]
Missing data values (one postdrug water-suppressed metabolite
spectrum in three subjects and a single unsuppressed water signal
in one subject) were estimated by interpolation (Winer 1971).
Results
Baseline cardiovascular parameters were normal in all
subjects and statistically equivalent across experimental
groups, with heart rate (HR) 5 60 6 8 beats per minute
(mean 6 SD), systolic pressure (Sys) 5 118 6 11 mm Hg,
diastolic pressure 5 70 6 8 mm Hg, and mean arterial
pressure 5 87 6 8 mm Hg. Cocaine induced transient,
dose-dependent increases in HR and Sys, with peak values
occurring about 15–20 min after drug administration
(Figure 1).
Cocaine administration was not associated with statistically significant changes in water peak areas. The variance of the water resonance area was, in each case, less
than 2%, documenting the consistency and reproducibility
of the technique.
A stack plot of representative metabolite spectra is
presented in Figure 2. Significant dose by time effects
were found for both Cho (p , .02) and NAA (p , .001).
Cocaine (0.2 mg/kg) induced a transient increase in the
Cho peak area, which reached a maximum of about 115%
of basal level 10 min after drug administration and then
declined to basal levels (Figure 3). In contrast, the higher
Proton MRS of Basal Ganglia after Cocaine
BIOL PSYCHIATRY
2000;48:685– 692
687
Figure 2. A stack plot showing sequential magnetic resonance
spectroscopy spectra from a single subject during the 15-min
baseline period before cocaine injection and over the 35-min
period following injection. Each spectrum was obtained without
changing any experimental parameters, and reflects the same
number of averages. There is no individual scaling between
peaks. The scale, though consistent, is arbitrary. Cho, cytosolic
choline-containing compounds; Cre, creatine and phosphocreatine; NAA, N-acetylaspartate.
Figure 1. Effects of cocaine/placebo administered at time 0 on
cardiovascular parameters. Shown are means 6 SEs of 10
subjects per dose level. Cocaine induced transient dose-dependent increases in heart rate (F 5 21.8, p , .0001; top) and
systolic blood pressure (F 5 13.0, p 5 .0001; middle). A
significant time effect was detected for diastolic pressure (F 5
4.3, p , .0001; bottom). Data are presented as mean 6 SEM and
values have been normalized to 100% for baseline measures.
cocaine dose (0.4 mg/kg) resulted in a progressive increase
in the Cho peak, which reached a maximum of about
135% of basal level 20 min following drug administration
and persisted to the end of the experiment (Figure 3, top).
For the NAA peak, 0.2 mg/kg cocaine also induced a
minimal metabolite increase, whereas 0.4 mg/kg cocaine
induced a larger increase that peaked at about 120% of
basal levels 20 min after drug administration and remained
elevated for the rest of the study (Figure 3, middle). A
slight but nonsignificant increase in the Cre peak area was
observed with high-dose cocaine (Figure 3, bottom).
Figure 4 displays a scattergram of areas under the curve
for the Cho and NAA resonance lines over time following
cocaine administration. Note that for the Cho resonance
(Figure 4, top) four of 10 subjects had an AUC . 0,
compared with seven of 10 subjects receiving 0.2 mg/kg
cocaine and nine of 10 subjects receiving 0.4 mg/kg. By
x2, the number of subjects with an increased AUC was
larger in the cohort receiving cocaine (x2 5 4.8, p , .03).
For the NAA resonance (Figure 4, bottom), four of 10
subjects had an AUC . 0, compared with five of 10
subjects receiving 0.2 mg/kg cocaine and eight of 10
subjects receiving 0.4 mg/kg. By x2, there was a weak
trend for the number of subjects with an increased AUC to
be larger in the cohort receiving cocaine (x2 5 1.7, p ,
.20). In both instances, the number of subjects with an
increased resonance intensity increased with dose. For the
Cre resonance, no trends between resonance area and
cocaine data were noted (data not shown).
Metabolite line widths were unchanged by cocaine
administration; repeated-measures ANOVA analysis detected no dose or time effects of cocaine on the Cho (5.1 6
0.8 Hz), Cre (6.1 6 1.8 Hz), and NAA (5.1 6 1.3 Hz) line
widths (mean 6 SD). This suggests that any confound in
the data resulting from subject motion is minimal.
Subject motion could also influence the observed MRS
signal amplitudes, if the volume of interest shifted over
time in such a manner as to include different amounts of
the lateral ventricles. Movement would also likely influence the shim and, therefore, the degree of water suppres-
688
J.D. Christensen et al
BIOL PSYCHIATRY
2000;48:685– 692
Figure 4. Scattergram for the area under the curve (AUC) for the
cytosolic choline-containing compounds and N-acetylaspartate
(NAA) resonance intensities following cocaine administration.
The numbers of subjects with AUC values . 0 for each dose of
drug or placebo are indicated at the top of each panel.
Figure 3. Proton metabolite peak area changes induced by
intravenous cocaine/placebo administration at time 0. Cocaine
induced dose-dependent (F 5 3.3, p 5 .05) and time-dependent
(F 5 2.5, p , .01) increases in the cytosolic choline-containing
compounds (Cho) peak area (top). Cocaine induced dosedependent increases in the N-acetylaspartate (NAA) peak area
(F 5 5.1, p , .02; middle). Cocaine administration was not
associated with statistically significant alterations in the creatine
and phosphocreatine (Cre) resonance (bottom). Data are presented as means 6 SDs.
sion; however, these effects of subject motion would not
account for the differential changes in signal intensity that
were observed in the NAA, Cho, and Cre resonance areas.
Although plasma cocaine levels were not measured in
this study, plasma analysis with gas chromatography in
separate subject groups receiving cocaine via an identical
administration protocol revealed that 0.2 (n 5 6) and 0.4
(n 5 6) mg/kg cocaine produced peak plasma levels 10
min after administration averaging 110 and 230 ng/mL,
respectively (Mendelson et al 1999).
Discussion
Our results demonstrate dose- and time-dependent effects
of cocaine on basal ganglia metabolite peak areas. These
effects temporally paralleled neuroendocrine and cardiovascular changes and plasma cocaine concentrations in
separate subject groups receiving cocaine via an identical
administration protocol as previously described (Mendelson et al 1999). In those studies, cocaine pharmacokinetics
Proton MRS of Basal Ganglia after Cocaine
following intravenous administration of 0.2 and 0.4 mg/kg
to men were determined; Tmax (and T1/2) were 6.7 min (45
min) and 8.0 min (48 min), respectively (Mendelson et al
1999). A comparable time course for cocaine-induced
striatal dopamine increases has been reproduced in both
rodents and nonhuman primates. Microdialysis studies in
rats (Bradberry et al 1993) and in primates (Iyer et al
1995) suggest that striatal dopamine levels peak 10 –20
minutes following the intravenous administration of cocaine and remain elevated for at least 100 min following
injection.
Cocaine’s effects on the Cho and NAA peak areas may
reflect changes in transverse relaxation times (T2*),
caused by osmotic stress and increased intracellular water
content. Cocaine inhibits the Na1/K1-adenosinetriphosphatase (Na1/K1-ATPase; Lien et al 1994), probably
through its action as an indirect agonist of dopamine D1
receptors (Bertorello et al 1990). Since volume regulation
is dependent upon Na1/K1-ATPase activity (Olson et al
1986), inhibition of this enzyme could lead to ion imbalance, osmotic stress, and cell swelling, as both sodium and
water move into the intracellular space (Hertz et al 1992).
Metabolite relaxation times can be greatly affected by
such changes in the cellular environment. In this regard,
osmotic stress secondary to Na1/K1-ATPase inhibition
and secondary energy failure has been associated with
enhanced metabolite mobility and increased Cho and
NAA relaxation times in some reports (Cady 1996; Cady
et al 1994), but not all (Fujimori et al 1998). Such an effect
on NAA relaxation has also been observed in brain edema
(Kamada et al 1994). Importantly, NAA has recently been
shown to serve as a neuronal osmolyte (Taylor et al 1995),
and GPC, which contributes substantially to the 1H MRS
Cho peak (see below), has been shown to function as a
renal osmolyte (Wolff and Balaban 1990).
Alternatively, cocaine may alter metabolite concentrations. For example, cocaine-stimulated phosphatidylcholine (PtdCho; brain concentration ; 21 mmol/L) hydrolysis might result in increased levels of PC (0.57 mmol/L)
and/or GPC (0.94 mmol/L; Bluml et al 1999; Klein et al
1993). With these concentration estimates, a 35% increase
in the 1H MRS Cho resonance would require the hydrolysis of 0.53 mmol/L, or approximately 2.5% of cellular
PtdCho. Turnover rates for PtdCho are difficult to measure
in vivo and may vary with regional cerebral activity;
however, Purdon and Rapoport (1998) have noted that the
turnover rates for fatty acids in phospholipids may be 100
times faster than previously thought. The relatively delayed time frame for the cocaine-induced Cho increase is
consistent with the general temporal sequence of phospholipid turnover, in which phosphatidylinositol turnover is
often the initial response, followed by PtdCho turnover
(Exton 1994).
BIOL PSYCHIATRY
2000;48:685– 692
689
We did not observe a statistically significant increase in
the intensity of the Cre resonance; however, across a
number of studies (for review, see Kreis 1997), T2*
relaxation times for the Cre resonance are substantially
shorter than those of Cho and NAA at 1.5 T. Thus,
increases in the intensity of the Cre resonance due to an
increase in intracellular water would be smaller than those
predicted for the Cho or NAA resonances. It may also be
the case that the 10% increase observed in the Cre
resonance following 0.4 mg/kg cocaine would have
reached statistical significance had a larger number of
subjects been studied.
To date, acute, drug-induced changes in the intensities
of 1H MRS resonance lines have not been reported in the
peer-reviewed literature; however, in general agreement
with the present results, Li and colleagues (1998) have
presented a preliminary report documenting a statistically
significant, 15% increase in NAA/Cre and a nonsignificant, 10% increase in Cho/Cre 12 min after the administration of a comparable dose of cocaine (40 mg/70 kg) to
a cohort of cocaine-dependent subjects. These data were
obtained from a 5-cm3 volume centered on the left
thalamus.
Since the present data were collected at a single TE, it
is not possible to conclude whether changes in Cho and
NAA peak areas resulted from changes in concentration,
changes in relaxation times, or both effects. Kreis (1997)
has reported that the mean T2* for NAA across multiple
brain regions is 376 msec. An increase in T2* to approximately 500 msec would be needed to increase the NAA
signal intensity by 10%. Of note, Cady et al (Cady 1996;
Cady et al 1994) have reported increases of this magnitude
in brain metabolite T2* values during secondary energy
failure; however, the fact that the intensities of two
intracellular resonance lines increased, as well as the
relatively large size of these increases, tend to argue in
favor of a change in relaxation times. A 35% increase in
the Cho resonance and an 20% increase in the NAA
resonance would correspond to increases in these metabolite pools of at least 500 mmol/L and 1 mmol/L, respectively, which are not likely. To the extent that the
increased Cho and NAA resonance intensities can be
directly attributed to dopamine release within the basal
ganglia, 1H MRS may provide a unique window into
central nervous system dopaminergic neurotransmission.
Increases in the intensity of intracellular metabolites
due to osmotic stress would be consistent with the intriguing possibility that a change in basal ganglia volume might
be detected following acute cocaine administration. We
are currently investigating this issue. The chronic administration of dopamine antagonists to individuals with
schizophrenia has been noted to increase the volume of
both the caudate and the putamen (for review, see Gur et
al 1998); however, changes in basal ganglia volume have
690
J.D. Christensen et al
BIOL PSYCHIATRY
2000;48:685– 692
not been associated with the long-term administration of
dopaminergic agonists and efforts to detect basal ganglia
pathology in human cocaine users have produced mixed
results (e.g., Kish et al 1999; Little et al 1998; Staley et al
1997). An alternative experimental approach would be to
use diffusion-weighted imaging (Le Bihan et al 1986) to
detect increased intracellular water, which should lead to
increased water diffusion.
Other factors may also contribute to the increased Cho
and NAA resonance intensities. For example, cocaine
administration to nonhuman primates has been associated
with a 13–22% decrease in basal ganglia cerebral metabolic rate (Lyons et al 1996). Limited available data
suggest that local cerebral metabolic rates, as determined
by positron emission tomography, are inversely related to
the intensity of the 1H MRS choline resonance. Another
factor that may play a role is cocaine-induced vasoconstriction. The intravenous administration of cocaine has
been reported to cause decreases in cerebral blood flow
(Gollub et al 1998; Wallace et al 1996) and volume
(Kaufman et al 1998a). Alterations in cerebral blood flow
and blood volume can lead to vascular susceptibility
changes (i.e., BOLD effects) that may increase (if the local
concentration of deoxyhemoglobin is decreased) or decrease (if the concentration of deoxyhemoglobin is increased) the signal intensity of both water and nearby
tissue metabolites. In our study the signal intensity of
water, which traverses the intravascular space, was not
increased while the intensity of intracellular metabolites,
NAA and Cho, was elevated. This observation tends to
argue against a BOLD effect. Of concern, our data would
be consistent with a decrease in blood flow to an extent
that causes basal ganglia ischemia (Cady 1996; Cady et al
1994). Therefore, further work to clarify the mechanism
by which intracellular metabolite resonance intensities are
increased may be warranted.
With regard to the effects of chronic cocaine on basal
ganglia 1H MRS metabolites, Li and coworkers (1999)
have reported that there are no differences in 1H MRS
metabolite ratios between cocaine users and comparison
subjects. In contrast, a 17% reduction in NAA/Cr was
noted in the left thalamus. Chang and colleagues (1997)
have reported increased parietal white matter Cre and
myo-inositol concentrations in heavy cocaine users. In a
follow-up study (Chang et al 1999), decreased NAA and
increased myo-inositol were noted in frontal lobe gray and
white matter in a cohort of abstinent subjects with a
history of crack cocaine dependence. In our study the use
of a 135-msec TE precluded the accurate assessment of the
myo-inositol resonance. Thus, it is uncertain how the
acute, cocaine-induced increases in 1H MRS resonance
intensities are related to the neurochemical consequences
of chronic heavy drug use.
In summary, intravenous cocaine administration transiently induced dose-dependent increases of basal ganglia
Cho and NAA peak areas. This study’s results are suggestive of cocaine-induced osmotic stress, but are also consistent with an increase in phospholipid turnover within
the basal ganglia. Further studies characterizing proton
metabolite changes following cocaine administration are
warranted to clarify this issue and to determine whether
the observed changes are relevant to the physiologic and
behavioral effects of cocaine.
This research was supported by National Institute on Drug Abuse Grants
Nos. DA09448, DA04059, DA00064, DA00115, DA00329, and
DA00343.
We gratefully acknowledge the excellent technical assistance provided
by Ms. Kim Appelmans.
JDC’s current affiliation: Department of Psychiatry, University of
Louisville, Louisville, Kentucky.
References
Bertorello AM, Hopfield JF, Aperia A, Greengard P (1990):
Inhibition by dopamine of (Na1/K1)ATPase activity in
neostriatal neurons through D1 and D2 dopamine receptor
synergism. Nature 347:386 –388.
Bluml S, Seymour KJ, Ross BD (1999): Developmental changes
in choline- and ethanolamine-containing compounds measured with proton-decoupled 31P MRS in in vivo human
brain. Magn Reson Med 42: 643– 654.
Bradberry CW, Nobiletti JB, Elsworth JD, Murphy B, Jatlow P,
Roth RH (1993): Cocaine and cocaethylene: Microdialysis
comparison of brain drug levels and effects on dopamine and
serotonin. J Neurochem 60:1429 –1435.
Cady EB (1996): Metabolite concentrations and relaxation in
perinatal cerebral hypoxic-ischemic injury. Neurochem Res
21:1043–1052.
Cady EB, Lorek A, Penrice J, Wylezinska M, Cooper CE, Brown
GC, et al (1994): Brain-metabolite transverse relaxation times
in magnetic resonance spectroscopy increase as adenosine
triphosphate depletes during secondary energy failure following acute hypoxia-ischaemia in the newborn piglet. Neurosci
Lett 182:201–204.
Chang L, Ernst T, Strickland TL, Mehringer M (1999): Gender
effects on persistent cerebral metabolite changes in the frontal
lobes of abstinent cocaine users. Am J Psychiatry 156:716 –
722.
Chang L, Mehringer CM, Ernst T, Rosemarie M, Myers H,
Forney D, Satz P (1997): Neurochemical alterations in
asymptomatic abstinent cocaine users: A proton magnetic
resonance spectroscopy study. Biol Psychiatry 42:1105–1114.
Duc CO, Weber AH, Trabesinger AH, Meier D, Boesiger P
(1997): Recycling the cholines. In: International Society of
Magnetic Resonance in Medicine VI. Berkeley, CA: ISMRM,
1210.
Exton JH (1994): Phosphatidylcholine breakdown and signal
transduction. Biochem Biophys Acta 1212:26 – 42.
Fujimori H, Michaelis T, Wick M, Frahm J (1998): Proton T2
Proton MRS of Basal Ganglia after Cocaine
relaxation of cerebral metabolites during transient global
ischemia in rat brain. Magn Reson Med 39:647– 650.
Giros B, Jaber M, Jones SR, Wightman RM, Caron MO (1996):
Hyperlocomotion and indifference to cocaine and amphetamine in mice lacking the dopamine transporter. Nature
379:606 – 612.
Gollub RL, Breiter HC, Kantor H, Kennedy D, Gastfriend D,
Mathew RT, et al (1998): Cocaine decreases cortical cerebral
blood flow but does not obscure regional activation in
functional magnetic resonance imaging in human subjects.
J Cereb Blood Flow Metab 18:724 –734.
Guimaraes AR, Schwartz P, Prakash MR, Carr CA, Berger UV,
Jenkins BG, et al (1995): Quantitative in vivo 1H nuclear
magnetic resonance spectroscopic imaging of neuronal loss in
rat brain. Neuroscience 69:1095–1101.
Gur RE, Maany BA, Mozley PD, Swanson C, Bilker W, Gur RC
(1998): Subcortical MRI volumes in neuroleptic-naive and
treated patients with schizophrenia. Am J Psychiatry 155:
1711–1717.
Hertz L, Code WE, Sykova E (1992): Ions, water and energy in
brain cells: A synopsis of interrelations. Can J Physiol
Pharmacol 70:S100 –S106.
Iyer RN, Nobiletti JB, Jatlow PI, Bradberry CW (1995): Cocaine
and cocaethylene: effects on extracellular dopamine in the
primate. Psychopharmacology 120:150 –155.
Johnson B, Lamki L, Fang B, Barron B, Wagner L, Wells L, et
al (1998): Demonstration of dose dependent global and
regional cocaine-induced reductions in brain blood flow using
a novel approach to quantitative single photon emission
computerized tomography. Neuropsychopharmacology 18:
377–384.
Kamada K, Houkin K, Hida K, Matsuzawa H, Iwasaki Y, Abe H,
Nakada T (1994): Localized proton spectroscopy of focal
brain pathology in hurnans: Significant effects of edema on
spin-spin relaxation time. Magn Reson Med 31:537–540.
Kaufman MJ, Levin JM, Kukes TJ, Villafuerte RA, Hennen J,
Lukas SE, et al (2000): Illicit cocaine use patterns in intravenous-naive cocaine users following investigational intravenous cocaine administration. Drug Alcohol Depend 58:35– 42.
Kaufman MJ, Levin JM, Maas LC, Rose SL, Kukes TJ, Mendelson JH, et al (1998a): Cocaine decreases relative cerebral
blood volume in humans: A dynamic susceptibility contrast
magnetic resonance imaging study. Psychopharmacology
138:76 – 81.
Kaufman MJ, Levin JM, Ross MH, Lange N, Rose SL, Kukes
TJ, et al (1998b): Cocaine induced cerebral vasoconstriction
detected in humans with magnetic resonance angiography.
JAMA 279:376 –380.
Kish SJ, Kalasinsky KS, Furukawa Y, Guttman M, Ang L, Li L,
et al (1999): Brain choline acetyltransferase activity in
chronic, human users of cocaine, methamphetamine, and
heroin. Mol Psychiatry 4:26 –32.
Klein J, Gonzalez R, Koppen A, Loffelholz K (1993): Free
choline and choline metabolites in rat brain and body fluid:
Sensitive determination and implications for choline supply
to the brain. Neurochem Int 31:293–300.
Kreis R (1997): Quantitative localized 1H MR spectroscopy for
clinical use. Prog Nucl Magn Reson Spectrosc 11:155–195.
Le Bihan D, Breton E, Lallemand D, Grenier P, Cabanis E,
BIOL PSYCHIATRY
2000;48:685– 692
691
Laval-Jeantet M (1986): MR imaging of intravoxel incoherent motions: Applications to diffusion and perfusion in
neurologic disorders. Radiology 161:401– 407.
Li S-J, Wang Y, Pankiewicz J, Stein EA (1999): Neurochemical
adaptation to cocaine abuse: Reduction of N-acetyl aspartate
in thalamus of human cocaine users. Biol Psychiatry 45:
1481–1487.
Li S-J, Wang Y, Risinger R, Rainey C, Harsch H, Pankiewicz J,
et al (1998): Acute cocaine administration induces NAA
increase detected by 1H MRS. In: International Society of
Magnetic Resonance in Medicine VI. Berkeley, CA: ISMRM,
1714.
Lien R, Mishra OP, Graham E, Delivoria-Papadopoulos M,
Anday EK (1994): Alteration of brain cell membrane function
following cocaine exposure in the fetal guinea pig. Brain Res
637:249 –254.
Little KY, McLaughlin DP, Zhang L, McFinton PR, Dalack GW,
Cook EH Jr, et al (1998): Brain dopamine transporter messenger RNA and binding sites in cocaine users: A postmortem
study. Arch Gen Psychiatry 550:793–799.
London ED, Cascella NG, Wong DF, Phillips RL, Dannals RF,
Links JM, et al (1990): Cocaine-induced reduction of glucose
utilization in human brain. A study using positron emission
tomography and [fluorine 18]-fluorodeoxyglucose. Arch Gen
Psychiatry 47:567–574.
Lyons D, Friedman DP, Nader MA, Porrino LJ (1996): Cocaine
alters cerebral metabolism within the ventral striatum and
limbic cortex of monkeys. J Neurosci 16:1230 –1238.
Mathew RJ, Wilson WH, Lowe JV, Humphries D (1996): Acute
changes in cranial blood flow after cocaine hydrochloride.
Biol Psychiatry 40:609 – 616.
Mendelson JH, Mello NK, Sholar MB, Siegel AJ, Kaufman MJ,
Levin JM, et al (1999): Cocaine pharmacokinetics in men and
in women during the follicular and luteal phases of the
menstrual cycle. Neuropsychopharmacology 21:294 –303.
Moonen CT, von Kienlin M, van Zijl PC, Cohen J, Gillen J, Daly
P, Wolf G (1989): Comparison of single-shot localization
methods (STEAM and PRESS) for in vivo proton NMR
spectroscopy. NMR Biomed 2:201–208.
Olson J, Sankar R, Holtzman D, James A, Fleischhacker D
(1986): Energy-dependent volume regulation in primary cultured cerebral astrocytes. J Cell Physiol 128:209 –215.
Pearlson GD, Jeffery PJ, Harris GJ, Ross CA, Fischman MW,
Camargo EE (1993): Correlation of acute cocaine-induced
changes in local cerebral blood flow with subjective effects.
Am J Psychiatry 150:495– 497.
Petroff OAC, Ogino T, Alger JR (1988): High-resolution proton
magnetic resonance spectroscopy of rabbit brain: Regional
metabolite levels and postmortem change. J Neurochem
51:163–171.
Petroff OAC, Spencer DD, Alger JR, Prichard JW (1989):
High-field proton magnetic resonance spectroscopy of human
cerebrum during surgery for epilepsy. Neurology 39:1197–
1202.
Purdon AD, Rapoport SI (1998): Energy requirements for two
aspects of phospholipid metabolism in mammalian brain.
Biochem J 335:313–318.
Staley JK, Talbot JZ, Ciliax BJ, Miller GW, Levey AI, Kung
692
BIOL PSYCHIATRY
2000;48:685– 692
MP, et al (1997): Radioligand binding and immunoautoradiographic evidence for a lack of toxicity to dopaminergic nerve
terminals in human cocaine overdose victims. Brain Res
274:219 –229.
Taylor DL, Davies SEC, Obrenovitch TP, Doheny MH, Patsalos
PN, Clark JB, Symon L (1995): Investigation into the role of
N-acetylaspartate in cerebral osmoregulation. J Neurochem
65:275–281.
Tsai G, Coyle JT (1995): N-Acetylaspartate in neuropsychiatric
disorders. Prog Neurobiol 46:531–540.
Tzika AA, Ball WSL, Vigneron DB, Dunn RS, Kirks DR (1993):
Clinical proton MR spectroscopy of neurodegenerative disease in childhood. AJNR Am J Neuroradiol 14:1267–1281.
Urenjak J, Williams SR, Gadian DG, Noble M (1993): Proton
nuclear magnetic resonance spectroscopy unambiguously
identifies different neural cell types. J Neurosci 13:981–989.
J.D. Christensen et al
van den Boogaart A, Ala-Korpela M, Jokisaari J, Griffiths JR
(1994): Time and frequency domain analysis of NMR data
compared: an application to 1D 1H spectra of lipoproteins.
Magn Reson Med 31:347–358.
Wallace EA, Wisniewski G, Zubal G, van Dyck CH, Pfau SE,
Smith EO, et al (1996): Acute cocaine effects on absolute
cerebral blood flow. Psychopharmacology 128:17–20.
Webb PG, Sailasuta N, Kohler SJ, Raidy T, Moats RA, Hurd RE
(1994): Automated single-voxel proton MRS: Technical development and multisite verification. Magn Reson Med 31:
365–373.
Winer BJ (1971): Statistical Principles in Experimental Design.
New York: McGraw-Hill.
Wolff SD, Balaban RS (1990): Regulation of the predominant
renal medullary organic solutes in vivo. Annu Rev Physiol
52:727–746.