Abnormal cellular energy and phospholipid
metabolism in the left dorsolateral prefrontal
cortex of medication-free individuals with
bipolar disorder: an in vivo1H MRS study
Bipolar disorder (BD) is a prevalent and chronic
major mental illness that is associated with high
rates of disability and suicide (1). Despite strong
Frey BN, Stanley JA, Nery FG, Monkul ES, Nicoletti MA, Chen H-H,
Hatch JP, Caetano SC, Ortiz O, Kapczinski F, Soares JC. Abnormal
cellular energy and phospholipid metabolism in the left dorsolateral
prefrontal cortex of medication-free individuals with bipolar disorder: an
in vivo1H MRS study.
Bipolar Disord 2007: 9 (Suppl. 1): 119–127. ª Blackwell Munksgaard,
Objectives: While the pathophysiology of bipolar disorder (BD)
remains to be elucidated, postmortem and neuroimaging studies have
suggested that abnormalities in the dorsolateral prefrontal cortex
(DLPFC) are implicated. We compared the levels of specific brain
chemicals of interest measured with proton magnetic resonance
spectroscopy (1H MRS) in medication-free BD subjects and age- and
gender-matched healthy controls. We hypothesized that BD subjects
would present abnormal cellular metabolism within the DLPFC, as
reflected by lower N-acetyl-aspartate (NAA) and creatine +
phosphocreatine (Cr + PCr).
Methods: Thirty-two medication-free BD subjects (33.8 ± 10.2 years)
and 32 matched controls (33.8 ± 9.0 years) underwent a short echo-time
(TE ¼ 30 ms)1H MRS. An 8-cm3single voxel was placed in the left
DLPFC, and individual concentrations of NAA, Cr + PCr, choline-
containing compounds (GPC + PC), myo-inositol, and glutamate were
obtained, using the water signal as an internal reference.
Results: BD subjects had lower Cr + PCr [F(1,62)¼ 5.85; p ¼ 0.018;
one-way analysis of variance (ANOVA)] and lower GPC + PC
[F(1,62)¼ 5.79; p ¼ 0.019; one-way ANOVA] levels in the left DLPFC.
No significant differences were observed for other brain metabolites.
Conclusions: These findings provide further evidence that the
pathophysiology of BD involves impairment in the DLPFC. Our findings
can be interpreted as evidence for reduced cellular energy and
phospholipid metabolism, consistent with the hypothesis of
mitochondrial dysfunction in BD.
Benı ´cio N Freya,b,c, Jeffrey A
Stanleyd, Fabiano G Nerya,e, E Serap
Monkula,f, Mark A Nicolettia,g, Hua-
Hsuan Chenh, John P Hatcha, Sheila
C Caetanoa,d, Oswaldo Ortizg, Fla ´vio
Kapczinskib,cand Jair C Soaresa,g,h
aMOOD-CNS Program, Division of Mood and
Anxiety Disorders, Department of Psychiatry, The
University of Texas Health Science Center at San
Antonio, TX, USA,bDepartment of Biochemistry,
ICBS, Federal University of Rio Grande do Sul,
cBipolar Disorders Program, Hospital de Clı ´nicas
de Porto Alegre, Porto Alegre, RS, Brazil,
dDepartment of Psychiatry and Behavioral
Neurosciences, Wayne State University School of
Medicine, Detroit, MI, USA,eDepartment of
Psychiatry, Institute of Psychiatry, University of Sa ˜o
Paulo School of Medicine, Sa ˜o Paulo, Brazil,
fDepartment of Psychiatry, Dokuz Eylul University
School of Medicine, Izmir, Turkey,gPsychiatry
Service, South Texas Veterans Health Care System,
Audie L Murphy Division,hDepartment of
Radiology, The University of Texas Health Science
Center at San Antonio, San Antonio, TX, USA
Key words: bipolar disorder – brain imaging –
dorsolateral prefrontal cortex – magnetic
Received 3 July 2006; revised and accepted for
publication 10 November 2006
Corresponding author: Benı ´cio N Frey, MD, MSc,
MOOD-CNS Program, Department of Psychiatry,
The University of Texas Health Science Center at
San Antonio, 3939 Medical Drive, Suite 100, San
Antonio, TX 78229, USA. Fax: +1 210 562 5485;
The authors of this paper do not have any commercial associations
that might pose a conflict of interest in connection with this manu-
Bipolar Disorders 2007: 9 (Suppl. 1): 119–127
Copyright ª Blackwell Munksgaard 2007
evidence that BD is associated with neurobiolog-
ical changes, the molecular mechanisms underlying
its pathophysiology remain largely undetermined
(2). Postmortem studies demonstrate decreased
neuronal and glial density in the dorsolateral
prefrontal cortex (DLPFC) of BD subjects com-
pared to healthy controls (3, 4). Since glial cells
play a central role in providing energetic support
for neurons (5), these findings suggest that ener-
getic metabolism might be impaired in the DLPFC
of BD subjects. The DLPFC regulates executive
functions and organizes behavioral responses and
strategies in learning new tasks (6). The role of
DLPFC in the pathophysiology of BD is also
supported by recent reports that BD patients
perform more poorly on neuropsychological tests
that assess prefrontal functioning than healthy
subjects (7, 8). Further, anatomical and functional
neuroimaging studies demonstrate decreased pre-
frontal cortical volume and abnormal prefrontal
activation in BD (9).
Proton magnetic resonance spectroscopy (1H
MRS) allows the in vivo quantification of certain
neurochemical compounds, which are involved in
cellular energy and phospholipids metabolism
(10–13). Using a short echo-time1H MRS, several
brain metabolites, including N-acetyl-aspartate
(NAA), creatine + phosphocreatine (Cr + PCr),
myo-inositol (mI) and glutamate (Glu) can be
detected (11, 13). While the exact function of NAA
is still unclear, it has long been recognized as a
marker of neuronal integrity (14). Phosphocreatine
constitutes an intracellular energy buffering system
that transports the energy generated in the mito-
chondria to the cytosol to maintain a constant
concentration of ATP (15). The GPC + PC peak
(PC) and glycerophosphocholine (GPC), which are
products of membrane synthesis and breakdown,
respectively. Acetylcholine and free choline com-
prise <5% of the GPC + PC peak signal at
3.2 ppm. Myo-inositol is a precursor of phospha-
tidylinositol, which is an important membrane
component and a member of the phosphoinositide
second-messenger system (13). Glutamate is the
most abundant amino acid in the central nervous
system and is associated with the modulation of
long-term potentiation and memory consolidation
(16). However, an excess of this neurotransmitter
may lead to excitotoxic cell death (17).
Previous1H MRS studies have yielded inconsist-
ent results concerning the metabolic profile of the
DLPFC in adult individuals with BD. One study
showed decreased NAA/Cr + PCr levels in the
DLPFC of euthymic BD patients (18); however,
(GPC + PC),
1H MRS detects mainly phosphocholine
three other recent studies, which mostly involved
(19–21). One study showed increased levels of
glutamate + glutamine (Glx) in a small sample of
acute manic patients (22), whereas other studies
found no differences in NAA, Cr + PCr or
GPC + PClevelsbetweenBDsubjectsandcontrols
(19, 20). Studies conducted in pediatric BD subjects
revealed decreased NAA/Cr + PCr (23) and de-
creased NAA levels (24), suggesting that neuronal
of bipolar illness. It is possible that medication
treatment may normalize NAA levels during the
course of illness, or that this abnormality is present
primarily ina subset ofmoreseverely ill individuals.
Previous studies demonstrated that lithium treat-
ment might increase brain NAA levels (25, 26) and
decrease brain mI (27), as assessed using1H MRS.
The objective of this study was to investigate the
neurochemistry of the DLPFC in a sizeable sample
of medication-free BD subjects and matched
healthy comparison subjects using the
technique. Previous studies assessing the DLPFC
chemistry with1H MRS used small samples (n ¼
8–20), and only one study excluded medicated
subjects (18). Considering that mood stabilizers
and antipsychotics alter the brain metabolites (25,
28), the present study was designed to exclude this
important confounder. We hypothesized a priori
that BD subjects would present abnormal cellular
metabolism within the DLPFC, as reflected by
lower NAA and lower Cr + PCr concentrations.
BD and healthy control subjects were recruited
required to meet DSM-IV diagnostic criteria for
bipolar disorder Type I or Type II as determined
by psychiatrists using the Structured Clinical
Interview for DSM-IV (SCID) (29). Exclusion
criteria for all subjects were: age <18 years,
current serious medical conditions, history of head
trauma, organic mental disorders, and neurological
disorders. Additional exclusion criteria for BD
subjects were: use of any psychotropic medication
during the 2 weeks immediately prior to the study
(6 weeks if fluoxetine or depot drugs), and alcohol/
6 months preceding study entry. We did not
influence patients? medication status; all BD sub-
jects were unmedicated before the study entry.
Additional exclusion criteria for healthy subjects
were: any DSM-IV Axis-I disorder, as assessed by
Frey et al.
a psychiatrist using the SCID non-patient version,
any history of alcohol/substance abuse or depend-
ence, and history of any psychiatric or neurolog-
ical disorders in any of their first-degree relatives.
This study was approved by the Institutional
Review Board (The University of Texas Health
Science Center at San Antonio), and all subjects
provided their signed informed consent before
entering in the study.
All subjects received a physical examination, and a
medical and psychiatric history was taken. The
diagnoses were established by trained psychiatrists
using the SCID (29). The severity of manic and
depressive symptoms was assessed using the Young
Mania Rating Scale (YMRS) (30), and the
Hamilton Depression Rating Scale (HAMD) (31),
respectively. All subjects underwent a laboratory
testing of liver, thyroid, kidney, electrolytes, blood
count, cortisol, and b-hCG (females), as well as a
screening for cannabis, cocaine, stimulants, opi-
oids, benzodiazepines, hallucinogens, and alcohol.
All subjects with abnormal laboratory tests, pos-
itive substance use, or pregnancy (females) were
excluded from the study.
1H MRS procedure
1H MRS was carried out in a 1.5-T Philips
Gyroscan Intera scanner (Philips Medical Systems,
Bothell, WA, USA). Axial, sagittal and coronal
T1-weighted localizer images were first obtained to
verify patient positioning and determine voxel
placement. Then, a 2 · 2 · 2 cm (8 cm3) voxel
was placed in the left DLPFC (Fig. 1), using the
superior frontal sulcus, the lateral fissure, and the
genu of corpus callosum as anatomical landmarks
point-resolved spectroscopy sequence (PRESS)
with TE ¼ 30 ms, TR ¼ 3.0 s, bandwidth: 2 kHz,
4,096 complex data points. Water unsuppressed
spectra were also acquired for absolute quantifica-
tion of metabolites in units of mmol/kg wet weight
(33). The quantification of the spectral metabolites
NAA, glutamate, glutamine,
GPC + PC, taurine, alanine, aspartate, gamma-
amino-butyric acid, glucose, and N-acetyl-aspar-
tyl-glutamate, as well as lipid and macromolecule
resonances (34), was done using the Linear Com-
bination Model (LC Model) software (35), an
operator-independent fitting routine (Fig. 2). Only
the results of the more reliable metabolites (NAA,
PCr + Cr, GPC + PC, myo-inositol and glutam-
ate) were used in the analysis.
1H MRS data were acquired using a
mI,PCr + Cr,
We performed multivariate analysis of covariance
with subject age as the covariate. Following a
Fig. 1. Dorsolateral
(A) ¼ axial view; (B) ¼ sagittal view; (C) ¼ coronal view.
Abnormal energy and metabolism in bipolar disorder
statistically significant multivariate test we per-
formed post hoc univariate analysis of covariance
with age as the covariate to compare the groups
(BD versus controls) on each metabolite. No
further adjustment to reported p-values was made
for analysis of multiple metabolites. On an explor-
atory basis, we also used analysis of covariance to
examine the associations between the metabolite
concentrations and clinical features including
mood state, Axis-I comorbidities, BD subtype
(BD Type I and BD Type II), and presence or
absence of lifetime psychiatric hospitalization. We
used partial correlations, adjusting for age to
examine the associations between the metabolite
concentrations and age at onset of illness, length of
illness, HAMD and YMRS scores. All analyses
were carried out using SPSS version 14.0.1 soft-
ware (SPSS Inc., Chicago, IL, USA). The two-
tailed significance criterion was set at p < 0.05
without adjustment for multiple comparisons.
Demographic and clinical characteristics of the
sample are displayed in Table 1. The BD sample
consisted of 32 individuals [mean (SD) age ¼
33.8 ± 10.2 years; 21females,
depressed, 7 hypomanic, 1 mixed and 7 euthymic;
20 BD Type I, 12 BD Type II]. Seventeen BD
subjects (53%) had a history of comorbid anxiety
disorders, eight (25%) had a history of past
11 males; 17
alcohol/substance abuse or dependence, and seven
(22%) had no Axis-I comorbidity. The healthy
comparison sample comprised 32 age- and gender-
matched healthy volunteers [mean (SD) age ¼
33.8 ± 9.0 years; 22 females, 10 males]. BD and
control subjects did not differ significantly in terms
of age (t ¼ 0.0; df ¼ 62; p ¼ 1.0; t-test), gender
(v2¼ 3.15; df ¼ 2; p ¼ 0.2). BD patients had
lower educational status than controls (Mann–
Whitney U-test, Z ¼ 3.8, p < 0.001). Mean levels
of all metabolites of BD and healthy subjects are
displayed in Table 2. The multivariate analysis
revealed that the BD and control subjects differed
significantly on the profile of neuro-metabolites
(Wilks? lambda ¼ 0.73, F ¼ 4.3, df ¼ 5,57, p ¼
0.002). Post hoc univariate testing without correc-
tion for multiple metabolites showed that BD
subjects had significantly lower Cr + PCr levels in
the left DLPFC than control subjects [F ¼ 5.85;
df ¼ 1,62; p ¼ 0.018 (Fig. 3)]. BD subjects also
had significantly lower GPC + PC levels than
controls [F ¼ 5.79; df ¼ 1,62; p ¼ 0.019 (Fig. 4)].
There was a non-significant trend for higher mI
levels in the left DLPFC of BD subjects compared
to controls (F ¼ 3.46; df ¼ 1,61; p ¼ 0.06). There
were no significant differences in NAA (p ¼ 0.93)
or Glu (p ¼ 0.21) levels between patients and
controls. In addition, there was no significant
association between any of the brain metabolites
and mood state, Axis-I comorbidities, BD subtype,
df ¼ 1;p ¼ 0.79), or handedness
Cr + PCr
Fig. 2. Typical proton magnetic resonance spectroscopy spectrum. NAA ¼ N-acetyl-aspartate; Cr + PCr ¼ creatine plus phos-
phocreatine; GPC + PC ¼ choline-containing compounds; mI ¼ myo-inositol; Glu ¼ glutamate.
Frey et al.
or lifetime hospitalizations (all p > 0.05). Finally,
mI levels were positively correlated with length of
illness (r ¼ 0.42; p ¼ 0.023). There was no signif-
icant correlation between age at onset, HAMD or
YMRS scores and any brain metabolite (all
p > 0.05).
We found lower levels of Cr + PCr in the DLPFC
of medication-free BD subjects compared to
healthy volunteers. This finding is in line with
Table 1. Demographic characteristics of bipolar subjects and healthy controls
Bipolar subjects (n ¼ 32)Healthy controls (n ¼ 32)p-value
Age (years), mean (SD)
Age range (years)
Education history (SCID-I) (%)
Grade 7–12 (without graduating)
Graduated high school or equivalent
Graduated 2-year college
Graduated 4-year college
Part graduate/professional school
Completed graduate/professional school
BD subtype (Type I/Type II)
Mood state (%)
Axis-I comorbidity (%)
Alcohol/substance abuse or dependence
Length of illness (months), mean (SD)
Age at onset (years), mean (SD)
HAMD score, mean (SD)
YMRS score, mean (SD)
Lifetime hospitalization, yes (%)
BD ¼ bipolar disorder; HAMD ¼ Hamilton Depression Rating Scale; SCID-I ¼ Structured Clinical Interview for DSM-IV – Axis I;
YMRS ¼ Young Mania Rating Scale; NA ¼ not applicable.
Table 2. Concentrations of brain metabolites in bipolar disorder subjects
and healthy controls
(n ¼ 32)
(n ¼ 32)p-valuea
Cr + PCr
GPC + PC
Values are presented as mean (SD), mmol ⁄kg wet weight.
ap-values displayed are not adjusted for multiplicity due to the
analysis of multiple metabolites. The unadjusted post hoc
univariate p-values are shown following a significant multivariate
test. Following Bonferroni correction, none of the p-values would
remain statistically significant at the p < 0.05 level.
NAA ¼ N-acetyl-aspartate; Cr + PCr ¼ creatine plus phospho-
creatine; GPC + PC ¼ choline-containing compounds; mI ¼
myo-inositol; Glu ¼ glutamate.
Fig. 3. Cr + PCr levels in bipolar subjects and healthy con-
trols.Cr + PCr ¼ creatine
bipolar subjects; HC ¼ healthy controls. *p ¼ 0.018.
Abnormal energy and metabolism in bipolar disorder
spectroscopy (31P MRS) studies that showed
decreased PCr in the frontal lobe in bipolar
depression (36, 37). Using
et al. (38) reported that depressed BD patients had
lower Cr + PCr in the left frontal lobe than
euthymic BD patients, but no differences were
found between BD patients and controls. Deicken
et al. (39) recently showed that subjects with
familial BD have lower Cr + PCr concentration
in the right and left hippocampus than controls,
suggesting that abnormal Cr + PCr levels may not
be restricted to the prefrontal cortex in BD.
Under physiologic conditions, PCr is synthesized
from Cr and the ATP generated in the mitochon-
dria in a reaction catalyzed by the creatine kinase
(CK) enzyme. This reaction is illustrated in the
1H MRS, Hamakawa
Cr þ ATP
PCr þ ADP þ Hþ
PCr is then transported to the cytosol acting as
an energetic buffer to regenerate the ATP con-
sumed by the cell. Cr + PCr is considered a
marker of the cellular energy status (40), and
decreased Cr + PCr concentrations reflect de-
creased energetic metabolism, possibly due to
mitochondrial dysfunction. Therefore, our results
are consistent with the hypothesis that BD sub-
jects have abnormal energetic metabolism in the
DLPFC. Whereas PCr is a major source of high-
energy phosphates required for cellular home-
ostasis (41), Cr stabilizes mitochondrial CK in its
octameric form, preventing the opening of the
mitochondrial transition pore, an early event of
apoptosis (42). Our finding is consistent with the
hypothesis of mitochondrial dysfunction in BD
(43) and raises the question of whether decreased
Cr + PCr is associated with the neuropatholog-
ical changes observed in BD (44).
The fact that the Cr + PCr signal has long been
used as an internal reference in1H MRS studies
(45) resulted in a lack of investigation of possible
abnormalities of Cr + PCr in BD. In addition,
studies conducted in major depressive and schizo-
phrenic patients demonstrated that the use of
Cr + PCr as an internal reference might lead to
a misinterpretation of the data (46, 47). In order to
avoid this potential problem we analyzed only the
individual concentrations of the neurometabolites.
WealsodemonstratedlowerlevelsofGPC + PC
in the left DLPFC in BD subjects compared to
differences in GPC + PC concentrations between
BD subjects and controls in the DLPFC (18–22).
Two studies that investigated other prefrontal
regions found increased GPC + PC/Cr + PCr in
the right cingulate cortex (48), and a trend for lower
GPC + PCintheorbitofrontalgraymatter(49).In
fact, if Cr + PCr is truly lower in BD, the previous
report of increased GPC + PC/Cr + PCr in the
right cingulate cortex (48) could have been misin-
terpreted. Evidence of altered brain phospholipid
metabolism in BD was also demonstrated with31P
demonstrated that BD subjects have significantly
lower brain phosphomonoesters (PMEs) than nor-
mal controls (36, 51–54). Given that the membrane
precursors PC and phosphoethanolamine (PE) are
the main components of the PME peak (12), our
finding supports the hypothesis that BD is associ-
ated with altered membrane phospholipid metabo-
lism. Reduced GPC + PC signal indicates reduced
cellular membrane phospholipids content or vol-
ume, which is consistent with postmortem studies
showing decreased neuronal and glial cell density in
the DLPFC of BD subjects (3, 4). Interestingly,
PE and its metabolite ethanolamine inhibit mito-
chondrial electron transfer activity, suggesting that
altered phospholipid metabolism might further
affect mitochondrial functioning. Conversely (and
perhaps more likely), mitochondrial dysfunction
may fail to provide the energy required for normal
membrane metabolism (56).
BD subjects showed a non-significant trend
toward higher mI levels in the left DLPFC than
controls. This trend is in line with our previous
report that manic/mixed patients have higher left-
to-right mI in the DLPFC (20). No differences in
NAA or Glu levels between bipolars and controls
were observed in the present study. Whereas no
previous study investigated absolute Glu levels in
the DLPFC in BD, 1 study that included untreated
31P MRS studies have consistently
Fig. 4. GPC + PC levels in bipolar subjects and healthy
controls.GPC + PC ¼ choline-containing
BP ¼ bipolar subjects; HC ¼ healthy controls. *p ¼ 0.019.
Frey et al.
adult individuals found decreased NAA/Cr + PCr
in the DLPFC of BD (18), and subsequent adult
studies failed to replicate this finding (19–22).
Nonetheless, all negative studies had included
patient samples that consisted mostly of medicated
individuals, therefore allowing the possibility that
negative results had been largely driven by possible
medication effects. It is indeed possible that any
regional abnormalities on NAA levels that are
present and related to the illness may be altered by
putative neuroprotective effects of mood stabilizers
(25, 57). It is also important to note that several
pediatric studies (23, 24, 58) found reductions in
NAA levels in the DLPFC in bipolar individuals,
suggesting the possibility that such abnormality
may be more particular to certain illness subtypes,
perhaps the most severe ones. This is also impor-
tant when we consider our current sample, com-
prised of outpatients, without any currently
psychotic subject, mostly with mild to moderate
illness severity. Considering the prior adult findings
in untreated subjects in the study by Winsberg
et al. (18) and growing literature demonstrating
such changes in pediatric BD, we cannot rule out
the possibility that such change would be found in
a sub-group of more severely ill untreated subjects.
In general, the levels of brain metabolites did not
associate with the clinical presentation of BD. We
found no significant associations between any
metabolite and mood state, Axis-I comorbidities,
BD subtype, lifetime hospitalization, age at onset,
HAMD or YMRS scores. Since all subjects were
recruited from the community, it is possible that we
recruited a less severely ill sample compared to
previous studies. We did find a positive correlation
between mI levels and length of illness, suggesting
that chronicity may lead to alterations of the
phosphoinositide signaling system in BD (59).
4 weeks of lithium treatment increase NAA (25,
26) and decrease mI concentrations (27). However,
the long-term effects of chronic psychotropic use
on brain chemistry have not been extensively
3.6 months of lithium treatment decreased non-
significantly gray matter Glx (glutamate + gluta-
mine) and increased non-significantly gray matter
mI in BD patients; no effects of 1.7 months of
valproate treatment were observed (60). Therefore,
even though we assessed only medication-free
subjects in this study, the potential effects of
previous medication exposure on brain metabolites
cannot be absolutely ruled out.
Some limitations of the present study should be
addressed. The single-voxel technique did not allow
concomitant investigation of other brain regions
1H MRS, it has been demonstrated that
thought to be involved in BD. Biochemical exam-
ination of other implicated brain regions will be
needed before an integrated and more complete
understanding of BD pathophysiology will be
possible. Another limitation is that differences in
water and metabolite content due to different
contributions of brain tissue and cerebrospinal fluid
may be a potential source of error. To minimize this
potential problem of partial volume effects, we
guided the voxel placement using high-resolution
T1-weighted images referred to standard anatom-
ical landmarks. Although no association between
mood state and metabolites concentration was
found, the fact that patients were on various mood
states might affect the results. Finally, the statistical
analysis of multiple metabolites in relatively small
samples presents a challenge for brain imaging
researchers to balance the trade-off between preser-
vation of statistical power and strong protection
against Type I errors. The method employed here,
unprotected univariate tests following a statistically
significant multivariate test, is a relatively safe
procedure, but it cannot absolutely hold the fam-
ily-wise Type I error rate to <5% in a strong sense.
Our study also has important methodological
strength. The recruitment of a large sample of
medication-free BD subjects overcomes two im-
portant limitations affecting previous studies. The
use of this sample avoids the known confounding
effects of some psychotropic medications and
provides sufficient statistical power to test the
primary hypotheses. Further, the use of a short
echo-time sequence with long repetition-time mini-
mizes errors caused by the different relaxation times
of water and brain metabolites. Finally, the use of
individual metabolite concentrations excludes an-
other potential confounder and improves the
sensitivity in comparison to studies where metabo-
lite concentrations were expressed as ratios.
In conclusion, the present study demonstrates
lower Cr + PCr and GPC + PC levels in the left
DLPFC of medication-free BD subjects compared
to age and gender matched healthy individuals.
These findings indicate that altered cellular energy
and phospholipid metabolism may be involved in
the pathophysiology of BD, and are consistent
with the hypothesis of mitochondrial dysfunction
(43). Prospective studies assessing the long-term
effects of mood stabilizers on brain metabolites are
warranted to further investigate the clinical rele-
vance of the present findings.
068766, RR020571, Krus Endowed Chair (UTHSCSA), Veter-
Abnormal energy and metabolism in bipolar disorder
ans Administration (Merit Review Award), UTHSCSA GCRC
(M01-RR-01346), NARSAD and CAPES Foundation (Brazil).
1. Muller-Oerlinghausen B, Berghofer A, Bauer M. Bipolar
disorder. Lancet 2002; 359: 241–247.
2. Soares JC. Contributions from brain imaging to the
elucidation of pathophysiology of bipolar disorder. Int
J Neuropsychopharmacol 2003; 6: 171–180.
3. Rajkowska G, Halaris A, Selemon LD. Reductions in
neuronal and glial density characterize the dorsolateral
prefrontal cortex in bipolar disorder. Biol Psychiatry 2001;
4. Uranova N, Orlovskaya D, Vikhreva O et al. Electron
microscopy of oligodendroglia in severe mental illness.
Brain Res Bull 2001; 55: 597–610.
5. Magistretti PJ, Pellerin L. Cellular mechanisms of brain
energy metabolism and their relevance to functional brain
imaging. Philos Trans R Soc Lond B Biol Sci 1999; 354:
6. Mega MS, Cummings JL. Frontal-subcortical circuits and
neuropsychiatric disorders. J Neuropsychiatry Clin Neu-
rosci 1994; 6: 358–370.
7. Bearden CE, Glahn DC, Monkul ES et al. Sources of
declarative memory impairment in bipolar disorder:
mnemonic processes and clinical features. J Psychiatr Res
2006; 40: 47–58.
8. Clark L, Iversen SD, Goodwin GM. A neuropsychological
investigation of prefrontal cortex involvement in acute
mania. Am J Psychiatry 2001; 158: 1605–1611.
9. Strakowski SM, Delbello MP, Adler CM. The functional
neuroanatomy of bipolar disorder: a review of neuroim-
aging findings. Mol Psychiatry 2005; 10: 105–116.
10. Stanley JA. In vivo magnetic resonance spectroscopy and
its application to neuropsychiatric disorders. Can J Psy-
chiatry 2002; 47: 315–326.
11. Stanley JA, Pettegrew JW, Keshavan MS. Magnetic
resonance spectroscopy in schizophrenia: Methodological
issues and findings – part I. Biol Psychiatry 2000; 48: 357–
12. Kato T, Inubushi T, Kato N. Magnetic resonance spectr-
oscopy in affective disorders. J Neuropsychiatry Clin
Neurosci 1998; 10: 133–147.
13. Soares JC, Krishnan KR, Keshavan MS. Nuclear magnetic
resonance spectroscopy: new insights into the pathophys-
iology of mood disorders. Depression 1996; 4: 14–30.
14. Urenjak J, Williams SR, Gadian DG, Noble M. Proton
nuclear magnetic resonance spectroscopy unambiguously
identifies different neural cell types. J Neurosci 1993; 13:
15. Hemmer W, Wallimann T. Functional aspects of creatine
kinase in brain. Dev Neurosci 1993; 15: 249–260.
16. Jia Z, Lu YM, Agopyan N, Roder J. Gene targeting
reveals a role for the glutamate receptors mGluR5 and
GluR2 in learning and memory. Physiol Behav 2001; 73:
17. Coyle JT, Puttfarcken P. Oxidative stress, glutamate, and
neurodegenerative disorders. Science 1993; 262: 689–695.
18. Winsberg ME, Sachs N, Tate DL, Adalsteinsson E,
Spielman D, Ketter TA. Decreased dorsolateral prefrontal
N-acetyl aspartate in bipolar disorder. Biol Psychiatry
2000; 47: 475–481.
19. Brambilla P, Stanley JA, Nicoletti MA et al. 1H magnetic
resonance spectroscopy investigation of the dorsolateral
prefrontal cortex in bipolar disorder patients. J Affect
Disord 2005; 86: 61–67.
20. Frey BN, Folgierini M, Nicoletti M et al. A proton
magnetic resonance spectroscopy investigation of the
dorsolateral prefrontal cortex in acute mania. Hum Psy-
chopharmacol 2005; 20: 133–139.
21. Bertolino A, Frye M, Callicott JH et al. Neuronal pathol-
ogy in the hippocampal area of patients with bipolar
disorder: a study with proton magnetic resonance spectro-
scopic imaging. Biol Psychiatry 2003; 53: 906–913.
22. Michael N, Erfurth A, Ohrmann P et al. Acute mania is
accompanied by elevated glutamate/glutamine levels with-
in the left dorsolateral prefrontal cortex. Psychopharma-
cology (Berl) 2003; 168: 344–346.
23. Chang K, Adleman N, Dienes K, Barnea-Goraly N, Reiss
A, Ketter T. Decreased N-acetylaspartate in children with
familial bipolar disorder. Biol Psychiatry 2003; 53: 1059–
24. Sassi RB, Stanley JA, Axelson D et al. Reduced NAA
levels in the dorsolateral prefrontal cortex of young bipolar
patients. Am J Psychiatry 2005; 162: 2109–2115.
25. Moore GJ, Bebchuk JM, Hasanat K et al. Lithium
increases N-acetyl-aspartate in the human brain: in vivo
evidence in support of bcl-2?s neurotrophic effects? Biol
Psychiatry 2000; 48: 1–8.
26. Silverstone PH, Wu RH, O’Donnell T, Ulrich M, Asghar
SJ, Hanstock CC. Chronic treatment with lithium, but not
sodium valproate, increases cortical N-acetyl-aspartate
concentrations in euthymic bipolar patients. Int Clin
Psychopharmacol 2003; 18: 73–79.
27. Moore GJ, Bebchuk JM, Hasanat K et al. Temporal
dissociation between lithium-induced changes in frontal
lobe myo-inositol and clinical response in manic-depressive
illness. Am J Psychiatry 1999; 156: 1902–1908.
28. Bertolino A, Callicott JH, Mattay VS et al. The effect of
treatment with antipsychotic drugs on brain N-acetylas-
partate measures in patients with schizophrenia. Biol
Psychiatry 2001; 49: 39–46.
29. First MB, Spitzer RL, Gibbon M, Williams JB. Structured
Clinical Interview for DSM-IV (SCID-I). New York:
Biomedics Research Department, 1998.
30. Young RC, Biggs JT, Ziegler VE, Meyer DA. A rating
scale for mania: reliability, validity and sensitivity. Br J
Psychiatry 1978; 133: 429–435.
31. Hamilton M. A rating scale for depression. J Neurol
Neurosurg Psychiatry 1960; 23: 56–62.
32. Jackson GD, Duncan JS. MRI Anatomy: A New Angle on
the Brain. New York: Churchill Livingstone, 1996.
33. Stanley JA, Drost DJ, Williamson PC, Thompson RT. The
use of a priori knowledge to quantify short echo in vivo 1H
MR spectra. Magn Reson Med 1995; 34: 17–24.
34. Seeger U, Klose U, Mader I, Grodd W, Nagele T.
Parameterized evaluation of macromolecules and lipids in
proton MR spectroscopy of brain diseases. Magn Reson
Med 2003; 49: 19–28.
35. Provencher SW. Estimation of metabolite concentrations
from localized in vivo proton NMR spectra. Magn Reson
Med 1993; 30: 672–679.
36. Kato T, Takahashi S, Shioiri T, Murashita J, Hamakawa
H, Inubushi T. Reduction of brain phosphocreatine in
bipolar II disorder detected by phosphorus-31 magnetic
resonance spectroscopy. J Affect Disord 1994; 31: 125–133.
37. Kato T, Shioiri T, Murashita J et al. Lateralized abnor-
mality of high-energy phosphate metabolism in the frontal
lobes of patients with bipolar disorder detected by phase-
encoded 31P-MRS. Psychol Med 1995; 25: 557–566.
Frey et al.
38. Hamakawa H, Kato T, Shioiri T, Inubushi T, Kato N. Download full-text
Quantitative proton magnetic resonance spectroscopy of
the bilateral frontal lobes in patients with bipolar disorder.
Psychol Med 1999; 29: 639–644.
39. Deicken RF, Pegues MP, Anzalone S, Feiwell R, Soher B.
Lower concentration of hippocampal N-acetylaspartate in
familial bipolar I disorder. Am J Psychiatry 2003; 160:
40. Kemp GJ. Non-invasive methods for studying brain
energy metabolism: what they show and what it means.
Dev Neurosci 2000; 22: 418–428.
41. Aubert A, Costalat R. A model of the coupling between
brain electrical activity, metabolism, and hemodynamics:
application to the interpretation of functional neuroimag-
ing. Neuroimage 2002; 17: 1162–1181.
42. O’Gorman E, Beutner G, Dolder M, Koretsky AP,
Brdiczka D, Wallimann T. The role of creatine kinase in
inhibition of mitochondrial permeability transition. FEBS
Lett 1997; 414: 253–257.
43. Kato T, Kato N. Mitochondrial dysfunction in bipolar
disorder. Bipolar Disord 2000; 2: 180–190.
44. Rajkowska G. Depression: what we can learn from
postmortem studies. Neuroscientist 2003; 9: 273–284.
45. Stork C, Renshaw PF. Mitochondrial dysfunction in
bipolar disorder: Evidence from magnetic resonance
spectroscopy research. Mol Psychiatry 2005; 10: 900–919.
46. Gruber S, Frey R, Mlynarik V et al. Quantification of
metabolic differences in the frontal brain of depressive
patients and controls obtained by 1H-MRS at 3 tesla.
Invest Radiol 2003; 38: 403–408.
47. Wood SJ, Berger G, Velakoulis D et al. Proton magnetic
resonance spectroscopy in first episode psychosis and
ultra high-risk individuals. Schizophr Bull 2003; 29: 831–
48. Moore CM, Breeze JL, Gruber SA et al. Choline, myo-
inositol and mood in bipolar disorder: a proton magnetic
resonance spectroscopic imaging study of the anterior
cingulate cortex. Bipolar Disord 2000; 2: 207–216.
49. Cecil KM, DelBello MP, Morey R, Strakowski SM.
Frontal lobe differences in bipolar disorder as determined
by proton MR spectroscopy. Bipolar Disord 2002; 4: 357–
50. Yildiz A, Sachs GS, Dorer DJ, Renshaw PF.31P nuclear
magnetic resonance spectroscopy findings in bipolar ill-
ness: a meta-analysis. Psychiatry Res 2001; 106: 181–191.
51. Deicken RF, Weiner MW, Fein G. Decreased temporal
lobe phosphomonoesters in bipolar disorder. J Affect
Disord 1995; 33: 195–199.
52. Deicken RF, Fein G, Weiner MW. Abnormal frontal lobe
phosphorous metabolism in bipolar disorder. Am J Psy-
chiatry 1995; 152: 915–918.
53. Kato T, Takahashi S, Shioiri T, Inubushi T. Alterations in
brain phosphorous metabolism in bipolar disorder detect-
ed by in vivo 31P and 7Li magnetic resonance spectros-
copy. J Affect Disord 1993; 27: 53–59.
54. Kato T, Takahashi S, Shioiri T, Inubushi T. Brain
phosphorous metabolism in depressive disorders detected
J Affect Disord 1992; 26: 223–230.
55. Modica-Napolitano JS, Renshaw PF. Ethanolamine and
phosphoethanolamine inhibit mitochondrial function in
vitro: implications for mitochondrial dysfunction hypothe-
sis in depression and bipolar disorder. Biol Psychiatry
2004; 55: 273–277.
56. Purdon AD, Rapoport SI. Energy requirements for two
aspects of phospholipid metabolism in mammalian brain.
Biochem J 1998; 335: 313–318.
57. Coyle JT, Manji HK. Getting balance: drugs for bipolar
disorder share target. Nat Med 2002; 8: 557–558.
58. Olvera RL, Caetano SC, Fonseca M et al. N-acetyl
aspartate reduction in the dorsolateral prefrontal cortex
of pediatric bipolar patients. Biol Psychiatry 2006; 59
(Suppl. 1): 223S.
59. Soares JC, Dippold CS, Wells KF, Frank E, Kupfer DJ,
Mallinger AG. Increased platelet membrane phosphatidy-
linositol-4,5-bisphosphate in drug-free depressed bipolar
patients. Neurosci Lett 2001; 299: 150–152.
60. Friedman SD, Dager SR, Parow A et al. Lithium and
valproic acid treatment effects on brain chemistry in
bipolar disorder. Biol Psychiatry 2004; 56: 340–348.
Abnormal energy and metabolism in bipolar disorder