Molecular Psychiatry (2004) 9,237–251
& 2004 Nature Publishing Group All rights reserved 1359-4184/04 $25.00
www.nature.com/mp
IMMEDIATE COMMUNICATION
St John’s wort and imipramine-induced gene expression
profiles identify cellular functions relevant to
antidepressant action and novel pharmacogenetic
candidates for the phenotype of antidepressant treatment
response
M-L Wong1, F O’Kirwan1, J P Hannestad1, KJL Irizarry1, D Elashoff2 and J Licinio1,3
1
Department of Psychiatry, Center for Pharmacogenomics, Neuropsychiatric Institute, David Geffen School of Medicine at
UCLA, Los Angeles, CA, USA; 2Department of Biostatistics, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA;
3
Division of Endocrinology, Diabetes and Hypertension, Department of Medicine, David Geffen School of Medicine at UCLA,
Los Angeles, CA, USA
Both the prototypic tricyclic antidepressant imipramine (IMI) and the herbal product St John’s
wort (SJW) can be effective in the treatment of major depressive disorder. We studied
hypothalamic gene expression in rats treated with SJW or IMI to test the hypothesis that
chronic antidepressant treatment by various classes of drugs results in shared patterns of
gene expression that may underlie their therapeutic effects. Individual hypothalami were
hybridized to individual Affymetrix chips; we studied three arrays per group treatment. We
constructed 95% confidence intervals for expression fold change for genes present in at least
one treatment condition and we considered genes to be differentially expressed if they had a
confidence interval excluding 1 (or 1) and had absolute difference in expression value of 10
or greater. SJW treatment differentially regulated 66 genes and expression sequence tags
(ESTs) and IMI treatment differentially regulated 74 genes and ESTs. We found six common
transcripts in response to both treatments. The likelihood of this occurring by chance is
1.14 1023. These transcripts are relevant to two molecular machines, namely the ribosomes
and microtubules, and one cellular organelle, the mitochondria. Both treatments also affected
different genes that are part of the same cell function processes, such as glycolytic pathways
and synaptic function. We identified single-nucleotide polymorphisms in the human orthologs
of genes regulated both treatments, as those genes may be novel candidates for
pharmacogenetic studies. Our data support the hypothesis that chronic antidepressant
treatment by drugs of various classes may result in a common, final pathway of changes in
gene expression in a discrete brain region.
Molecular Psychiatry (2004) 9, 237–251. doi:10.1038/sj.mp.4001470
Published online 13 January 2004
Keywords: antidepressants; major depression; imipramine; hypericum; treatment; microarray
The search for new molecules and biological targets
for antidepressant drug discovery remains a challenge
for contemporary psychiatry. Such work is needed to
understand the fundamental molecular mechanisms
underlying major depressive disorder (MDD) and to
identify new directions for treatment. The investigation of the mechanisms for the reported therapeutic
activity of traditional herbal products, such as St
John’s wort (SJW) (see Figure 1), could uncover new
mechanisms and novel treatments for MDD.
Correspondence: Professor M-L Wong MD, Department of Psychiatry and Biobehavioral Sciences, David Geffen School of
Medicine at UCLA, 3357A Gonda Center; 695 Charles Young
Drive So., Los Angeles, CA 90095-1761, USA.
E-mail: [email protected]
Received 13 October 2003; revised 12 November 2003; accepted
17 November 2003
SJW (hypericum) extracts are among the most
widely used antidepressants in Germany with a
market share of more than 25% in 1997.1 Despite
the controversy on the effectiveness of SJW as an
antidepressant, the use of SJW has been widespread
and several preparations containing SJW are commercially available in Europe and in the US. Even though
two recent multicenter trials in the USA have
failed,2,3 randomized clinical studies have shown that
SJW extracts are significantly superior to placebo and
similarly effective as standard antidepressants in the
treatment of mild to moderately severe depression.4–6
Significant drug interactions have been reported with
cyclosporine and protease inhibitors,7,8 but a commonly cited advantage of SJW continues to be its
favorable side effect profile. Although reports of
serious and potentially fatal adverse reactions are
Antidepressant-induced gene expression
M-L Wong et al
238
Figure 1 Warty St John’s wort (United States Department
of Agriculture, Agricultural Research Service, Image Number K9982-4).
known, and probably underreported,9 the majority of
significant side effects described in clinical trials are
considered mild to moderate or transient.
Preclinical studies suggest that SJW is effective in
three major biochemical systems relevant for antidepressant activity: inhibition of the synaptic reuptake of serotonin, noradrenaline, and dopamine;10–12
it also produces monoamine reinhibition and longterm changes in receptors.10 However, there is controversy as to the exact mechanism of action for the
antidepressant effects of SJW. It is probable that both
hypericin and hyperforin have antidepressant properties. Hypericin inhibits monoamine oxidases (MAO-A
and MAO-B) and the antidepressant property of
hyperforin has been attributed to its capacity to
increase the extracellular levels of the monoamines
and glutamate in the synaptic cleft, probably as a
consequence of uptake inhibition.11,13–15 SJW extracts
contain six major natural product groups that may
Molecular Psychiatry
contribute to their pharmacological effects [acylphloroglucinols, biflavones, naphthodianthrones (hypericin),
flavonol glycosides (hyperforin), proanthocyanidins,
phenylpropanes, proanthcyanidins, and phenylpropanes];16 therefore, compounds other than hypericin
or hyperforin could also contribute to the antidepressant effects of hypericum. Repeated oral administration of SJW for 3 consecutive days has been reported
to ameliorate behavior observations in some animal
models of depression,14 and only recently the effects
of long-term administration of SJW have started to be
explored.15
Our hypothesis has been that there may be common
final molecular pathways that underlie the response
to various types of antidepressant treatment. The
identification of transcripts that are regulated in the
hypothalamus by prolonged administration of either
imipramine (IMI) or fluoxetine treatment has supported this hypothesis.17 Several investigators studying mood-stabilizing drugs have used this strategy of a
presumed final common pathway of action.18–20 The
hypothalamus was used in this study because key
biologic manifestations of MDD include alterations in
the hypothalamic functions that regulate food intake,
libido, circadian rhythms, and the synthesis and
release of hypothalamic neurohormones.21 Patients
with MDD, with melancholic features, typically have
decrease appetite, decreased sexual interest, early
morning awakening, diurnal variation in mood, and
endocrine abnormalities, such as hypogonadism,
hyposomatotropism, and hypercortisolism.22 Additionally, several antidepressants, including extracts of
SJW, when administered in a long-term basis downregulate corticotropin-releasing hormone gene (CRH)
expression in the paraventricular nucleus of the
hypothalamus (PVN).15,23–25
Here, we describe common transcripts in the
hypothalamus related to the long-term administration
of IMI and SJW. We identified human orthologs for
genes that are regulated both by SJW and by IMI, as
they may be potential novel candidates for pharmacogenetic studies. The ability to map rat genes to their
cognate human genes provides a needed bridge
between basic science and clinical research.
Material and methods
Hypericum extracts
The SJW extract was prepared by Botanicals Internationals (Long Beach, CA, USA) and generously
donated to us. Hypericum perforatum native plant
was extracted in ethanol/water at a ratios of 3–6 : 1,
dried, and stored protected from light and moisture at
41C. All SJW extract used in this study came from a
single lot and was free of pesticide, heavy metals, and
microbiological contaminants. The SJW extract was
analyzed at the initiation and every 6 months thereafter. High-performance liquid chromatography
(HPLC) analysis of SJW was performed by the
Pharmanex Research Center (Redwood City, CA,
USA), and the extract we used had the following
Antidepressant-induced gene expression
M-L Wong et al
composition: total hypericins 0.33%; hypericin and
pseudohypericin 0.013%; hyperforin 3.03%. The
concentration and composition of other flavonoids
in the SJW extract was not determined.
Animals
Studies were carried out in accordance with animal
protocols approved by University of California, Los
Angeles. Virus- and antibody-free male Sprague–
Dawley rats (200–250 g, Harlan, Indianapolis, IN,
USA) were housed 2/cage at 241C with lights on from
0600 to 1800. Animals were housed 3/cage in a stressfree environment for at least 5 days before the
initiation of experimental procedures. This acclimatization period was used to minimize the effects of
environmental and social stress in our experiments.
Rats were randomly separated into four groups: saline
(a control for IMI), carboxymethylcellulose (CMC, a
control for SJW), IMI, or SJW groups. Animals treated
with IMI (Sigma, St Louis, MO, USA) received 0.5 ml
intraperitoneal (i.p.) injections daily for 8 weeks with
5.0 mg/kg drug dissolved in 0.9% saline. Animals
treated with SJW extract were gavaged daily, for 8
weeks with 120 mg/kg, in 1 ml of 0.3% CMC (Sigma)
in distilled water. CMC has been used to dissolve SJW
as described previously.14,26 Control animals consisted of rats injected with vehicle (0.9% saline) i.p.
for the IMI-treated group or gavaged with vehicle
(0.3% CMC) for the SJW-treated group. Long-term
administration has been utilized because of the welldocumented findings that antidepressants are generally effective after 3–6 weeks of administration.24
Animals received their last treatment 24 h before
termination of experiments. To prevent the confounding effects of stress on gene expression, animals were
removed from their home cages by an animal handler
not involved in the decapitation process and were
decapitated within 45 s of individual removal from
home cages. Immediately after sacrificing the animals,
brains were removed. To avoid possible confounding
variations caused by circadian rhythms, all animals
were sacrificed at 10:00 to 12:00 noon. All hypothalamic tissues were dissected by the same investigator (ML W), frozen, and stored at 801C.
Microarray studies
The following groups of animals were studied: (1)
saline; (2) CMC; (3) IMI; and (4) SJW. Total RNA was
extracted from each hypotamus using TRIzol (Qiagen,
Valencia, CA, USA) reagent, cleaned with RNeasy
(Qiagen), and its quality was verified by gel. Total
RNA from individual hypothalamus was used for
each array; three animals per group were studied. We
used three to five array experiments per group. Dchip
software27 was used to do a stringent quality control
of our arrays; array data that did not pass our quality
control were not used in our analyses. Array experiments were repeated until three arrays per group met
our quality control requirements, with three independent replications of our microarray experiments as
recommended by Lockhart and Barlow.28
Protocols recommended by Affymetrix Inc. (Santa
Clara, CA, USA) were followed. Briefly, 15 mg of each
RNA sample pool were submitted to reverse transcription with SuperScript Choice System (Gibco
BRL, Rockville, MD, USA) and a T7-(dT)24 primer,
followed by in vitro transcription and biotin-labeling
with the Enzo BioArray High Yield RNA Transcript
Labeling Kit (Enzo Diagnostics, Farmingdale, NY,
USA). Labeled cRNA was fragmented and 15 mg of
the fragmented cRNA of each sample was used for
hybridization in a 2-(N-Morpholino) ethanesulfonic
acid-based buffer containing bovine serum albumin,
herring sperm DNA, control oligonucleotide B2, and
Eukaryotic hybridization controls BioB, BioC, BioD,
and Cre (Affymetrix). We used GeneChip Rat Genome
U34A arrays (Affymetrix), which contain sequences
for 8799 genes and expressed sequences tags (EST).
The samples were hybridized at 451C for 16 h under
constant rotation. After the hybridization, microarrays were washed, stained with streptavidin phycoerythrin, and scanned in a GMS 418 Array Scanner
(Affymetrix).
239
Data analysis: statistical methods
Microarray image files were loaded into the Dchip
software package. This package implements the Li &
Wong algorithm and computes two summary measures for each gene.27 The first measure is the modelbased expression index (MBEI) that summarizes the
approximate expression level over the 20 individual
probes. The second measure is the present/absent
call. This call is a decision rule utilizing a number of
details of the probe pairs to determine if there was or
was not any mRNA corresponding to that particular
gene in the sample. The data set had 8799 genes. Our
analysis plan consisted of three filtering steps to
identify a list of differentially expressed genes. Genes
were filtered in both of the treatment vs. control
separately. The first step of the analysis was to remove
genes that were absent in all arrays in all four of the
conditions (two treatments two controls). This step
removed 3051 genes from the analysis. Step two
involved the construction of 95% confidence intervals for the expression fold-change for each of the
remaining 5748 genes. The expression fold-change
value is a ratio between expression values in treatment and control conditions, thus fold-change of 1 or
1 is not different. Only those genes that had
confidence intervals excluding 1 (or 1) were
included in the final gene lists. Step three excluded
genes where the absolute difference between conditions was less than 10 (the median expression across
the arrays was 55). This last step helps remove genes
with very low levels of expression in which the foldchange can be large due to very small expression
differences.
The Dchip software provides automated quality
control by determining the genes or probes that are
outliers. These outliers are treated as missing data in
the subsequent data analysis steps. Dchip suggests
that chips with greater than 5% outliers are of poor
Molecular Psychiatry
Antidepressant-induced gene expression
M-L Wong et al
240
quality and should be repeated. In our study we
found three chips to have greater than 5% outliers
and these experimental conditions were re-run on
new chips.
Microarray-based experiments involve exceedingly
more measurements than samples even in experiments containing the largest practical sample size. In
a recent review, Slonim discussed the difficulties of
extracting meaningful information out of array data.
Unfortunately, there is no ‘one-size-fits-all’ solution
for the analysis and interpretation of wide expression
array data.29 In fact, data analysis techniques for
microarray are still in the early phases of development.30 Bonferroni method, one proposed solution to
the multiple comparisons problem, requires the use of
statistical test with known distribution and offers a
conservative method for significance assessment.31
The implementation of Bonferroni method to array
analyses is generally not appropriate and it yields
overly conservative critical values; therefore, alternative solutions to the multiple comparisons problem
have been used.32 Furthermore, methods for P-value
correction are not generally appropriate for exploratory data analysis as false negatives are of potentially
greater concern than false positives. Our analyses are
comparable to those performed in many studies that
measured differential expression by the ratio of
expression levels between two samples; in which
genes with ratios above a fixed cutoff were said to be
differentially expressed.33–38 The advantage of our
method is that the Li & Wong algorithm estimates the
standard error of the expression measurement, which
allows us to make statistical statements about the
fold-change rather than simply using a predetermined
cutoff.27
Final gene lists were functionally classified using
the batch query in the NetAffxt Analysis Center on
the affymetrix website (http://www.affymetrix.com/
analysis/index.affx). Probability calculations were
carried out to understand the likelihood of finding
genes and ESTs common to the list of transcripts that
were differentially regulated by IMI and SJW. In these
calculations, we excluded genes that were absent in
all conditions; therefore, we calculated the probability of finding six common genes in lists of 66 and 74
randomly selected group of genes from a total of 5748
genes.
Hierarchical clustering was performed using CLUSTER and TREEVIEW software39 on the sets of genes
that had significant treatment fold changes. Genes
and arrays were clustered using Average Linkage
Clustering. Expression values in the cluster diagrams
were standardized by subtracting the mean expression and dividing by the standard deviation.
Human ortholog and SNP identification
Rat genes were mapped to their human orthologs
using a combination of web based databases and
sequence alignment tools. We mapped orthologs
across genomes using one or more locus specific
identifiers as a foreign key in the LocusLink database
Molecular Psychiatry
(http://www.ncbi.nlm.nih.gov/LocusLink/). Many rat
genes have obvious human orthologs, but in some
instances, some rat genes could not be mapped to a
human ortholog using Locus Link. We used at least
two independent computational sequence comparison methods for genes that we could not map using
LocusLink. Single-nucleotide polymorphisms (SNPs)
were identified by querying the dbSNP database,
(http://www.ncbi.nlm.nih.gov/SNP/), via a web interface available on Locus Link.
Results
We used an expression profile approach to compare
transcriptional changes related to prolonged administration of IMI and SJW. Transcripts that were
significantly altered by IMI and SJW treatments are
presented in Figures 2 and 3, respectively; these
figures show the cluster analyses of our results. Table 1
lists all the transcripts significantly changed by either
treatment grouped by functional classification. Our
findings are compatible with a predominance of
downregulation after chronic IMI, but not after SJW
treatment.
Microarray analyses
Our analysis filtered out 66 genes as differentially
expressed in the SJW versus CMC (control) comparison and 74 in the IMI versus saline comparison. Six
genes were found to be common in the two lists of
genes with a significant fold change. We calculated
the probability that this was due to chance alone. The
probability of finding one gene in common in two
randomly selected lists of genes (n ¼ 66 and 74) is
0.00015, hence the probability of finding six genes in
common is 1.14 1023. Four of those six transcripts
were similarly downregulated, one transcript was
similarly upregulated and another one had different
regulation. Figure 4 shows the cluster analysis of
these common genes. The ribosomal proteins S29
(X59051) and S4 (X14210); rat long interspersed
repetitive DNA sequence LINE3 (M13100); and
EST202275 which is similar to the sequence of the
complete mitochondrial genome of the R. novergicus
were downregulated in both IMI and SJW treatments;
microtube-associated protein 1A (MAP1A, M83196)
was upregulated; and rat carnitine palmitoyltransferase Ib (AF063302) was downregulated by SJW and
upregulated by IMI treatment.
The functional classification of transcript revealed
that differentially regulated genes and ESTs for SJW
and IMI could be separated into 12 main functional
groups, namely genes related to the following process:
calcium, cell structure, energy production/expenditure, fatty acid metabolism, hormonal, ion concentration,
neurotransmission,
protein
synthesis/
degradation, repetitive DNA sequences, signal transduction, synaptic transmission, and transcription
regulation. Genes that could not be classified in those
categories were included in miscellaneous. Several
genes in the same pathway can be modulated by one
Antidepressant-induced gene expression
M-L Wong et al
of the treatment drugs (Table 1). Also, different genes
in the same pathway can be modulated by different
antidepressant treatments. A very illustrative example
of this phenomenon is our findings involving ele-
ments of the synaptic transmission: three genes are
modulated by IMI and other three different genes are
modulated by SJW (Table 1).
241
Human ortholog identification
A number of rat genes, including ribosomal protein
S29, ribosomal protein S4, and MAP1A, was readily
mapped to their associated human genes through
Locus Link using the OMIM (Online Mendelian
Inheritance in man) identifier to query the Homologene ortholog database.
Homologene results for rat ribosomal protein S29
(GenBank identifier X59051), ribosomal protein S4
(identifier X14210), microtubule associated protein,
including the Homo sapiens MAP1A, were also easily
identified. The rat gene carnitine palmitoyltransferase
(CPT) Ib is the ortholog of the human carnitine
palmitoyltransferase 1B gene.
The rat long interspersed repetitive DNA sequence
LINE3 (M13100) proved difficult to map to the
corresponding human ortholog using gene identifiers
alone, so we performed a BLAST search against the
Non-Redundant (NR) Genbank protein database.
None of the top 40 blast hits mapped to a human
gene; among the best three scoring human hits was
the human LINE-1, which belongs to the reverse
transcriptase family. We are unable to unambiguously
classify the human LINE-1 homolog as the ortholog of
the rat LINE-3 gene.
Finally, we wanted to identify the human ortholog
of the rat expressed sequence fragment, EST202275,
which is annotated as having similarity to mitochondrial genome. The results of a BLAST query against
the NR database at NCBI showed that none of the high
scoring homologous sequences mapped to the human
genome. We utilized the entire finished human
genome sequence, including the mitochondrial genome, in another round of sequence comparison
against the rat gene fragment using the indexed
sequence comparison method called BLAT (available
Figure 2 Cluster analysis of transcripts that were significantly altered by imipramine when compared to saline. In
each treatment group, genes colored in red are upregulated
and genes colored in green are downregulated when
compared to the average gene expression across all conditions. The CMC and St John’s wort groups have been
included just as illustration here, as most of the genes in
this cluster analysis were not significantly changed by St
John’s wort treatment (See Figure 4). Note the predominance of downregulation after imipramine treatment. The
dendogram of treatment clustering is displayed above the
image and describes the degree of relatedness between
treatment groups; the length of branches denotes the degree
of similarity, short branches indicate high similarity. Active
treatment groups are more similar between themselves
when compared to control groups. Expression values were
standardized by subtracting the mean expression across
groups and dividing by the standard deviation. CMC,
carboxmethylcellulose; IMIP, imipramine; SAL, saline;
SJW, St John’s wort.
Molecular Psychiatry
Antidepressant-induced gene expression
M-L Wong et al
242
at UCSC, http://genome.ucsc.edu/cgi-bin/hgBlat).
The top three BLAT hits indicated strong homology
to the rat sequence within a region of the mitochondrial genome as well as intervals within chromosomes 11. We focused on the human mitochondrial
interval since the rat sequence was also associated
with mitochondrial sequence. A gene within this
region (AY029066) translates into a protein 24 amino
acids in length, MAPRGFSCLLLLTSEIDLPVKRRA.
By traversing the genomic annotation back to the
transcript level, we were able to identify the locus as
belonging to Humanin and the mitochondrial ribosomal 16S RNA.
SNP identification
SNP information on the six genes that are regulated by
SJW and IMI treatments is presented in Table 2.
Discussion
This study focused on surveying a number of
genes and EST to elucidate the molecular mechanisms of the action of antidepressants beyond
the receptor level and to identify a presumed
final common pathway of antidepressant action.
Figure 5 synthesizes our approach to identify cellular
functions and genes that are regulated by antidepressants of different classes. The identification of such
Figure 4 Cluster analysis of six genes that are significantly
regulated by St John’s wort and imipramine treatments.
Note that four of the six transcripts were downregulated by
St John’s wort and imipramine treatments. In each treatment
group, genes colored in red are upregulated and genes
colored in green are downregulated when compared to the
average gene expression across all conditions. Expression
values were standardized by subtracting the mean expression across groups and dividing by the standard deviation.
CMC, carboxmethylcellulose; IMIP, imipramine; SAL, saline; SJW, St John’s wort.
Figure 3 Cluster analysis of transcripts that were significantly altered by St John’s wort. In each treatment group,
genes colored in red are upregulated and genes colored in
green are downregulated when compared to the average
gene expression across all conditions. The saline and
imipramine groups have been included just as illustration
here, as most of the genes in this cluster analysis were not
significantly changed by imipramine treatment (See Figure
4). The dendogram of treatment clustering is displayed
above the image and describes the degree of relatedness
between treatment groups; the length of branches denotes
the degree of similarity, short branches indicate high
similarity. Active treatment groups are more similar
between themselves when compared to control groups.
Expression values were standardized by subtracting the
mean expression across groups and dividing by the
standard deviation. CMC, carboxmethylcellulose; IMIP,
imipramine; SAL, saline; SJW, St John’s wort.
Molecular Psychiatry
Antidepressant-induced gene expression
M-L Wong et al
243
Molecular Psychiatry
Antidepressant-induced gene expression
M-L Wong et al
244
Molecular Psychiatry
Antidepressant-induced gene expression
M-L Wong et al
245
genes is useful for neuroscience research,
pharmacogenetic association studies, and drug
development.
Several groups, including ours, have used stressfree animals to study the effects of chronic antidepressant treatment.15,17,23,24 This is useful in the
understanding of chronic antidepressant effects, as
nondepressed humans treated with antidepressants
have been reported to have physiological responses to
such treatments.40 We expected that SJW would
regulate many transcripts in a similar way to a
prototypic antidepressant drug. Contrary to expecta-
tion, our study identified only six genes that are
regulated by IMI and SJW in the hypothalamus, but
we found that treatment with both IMI and SJW elicit
transcripts that are important in the same cellular
functions of building (ribosomal protein S4 and S29),
scaffolding/intracellular
transport
(microtubuleassociated protein 1A), and fueling the cell (carnitine
palmitoyltransferase Ib and EST202275). Although
these findings require further confirmation, we discuss their relevance, as these transcripts are
fair representatives of their respective functional
groups.
Molecular Psychiatry
Antidepressant-induced gene expression
M-L Wong et al
246
Table 2
SNP information on six genes that are regulated by St John’s wort and imipramine treatments
Gene/protein
SNP link
(1) Ribosomal protein S29
GenBank GI: 13904868
Number detected SNPs: 12a
Number Unclassified SNPs: 11a
Number intronic SNPs: N/Ab
Number exonic SNPs: 1b
Number coding SNPs: 1b
Number amino-acid changes: 1 (nonconservative)b
http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?locusId ¼ 6235
(2) Ribosomal protein S4
Genbank GI: 17981705
Number detected SNPs: 71a
Number unclassified SNPs: 66a
Number intronic SNPs: 4b
Number exonic SNPs: 4b
Number coding SNPs: 3b
Number amino-acid changes: 3b
http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?locusId ¼ 6191
Some inconsistencies because of two diff annotations:
(1) Contig annotation
(2) GenBank mapping
(3) Microtube-associated protein 1a
http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?locusId ¼ 4130
Genbank: GI: 21536457
Number detected SNPs: 176a
Number Unclassified SNPs: 176 (genbank mapping)a
Number intronic SNPs: 1b
Number exonic SNPs: 112b
Number coding SNPs: 104b
Number amino-acid changes:78b
(4) LINE1 rev transcr. homolog
GenBank: 126295
Number Detected SNPs: 1a
Number Unclassified SNPs: N/Aa
Number intronic SNPs: N/Ab
Number exonic SNPs: N/Ab
Number coding SNPs: N/Ab
Number amino-acid changes: N/Ab
http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?type ¼ rs&rs ¼ 792629
(5) Carnitine palmitoyltransferase 1B
GenBank GI: 23238252
Number detected SNPs:52a
Number unclassified SNPs: 46a
Number intronic SNPs:18b
Number exonic SNPs: 36b
Number coding SNPs: 8b
Number amino-acid changes: 8b
http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?locusId ¼ 1375
(6) Humanin 1
http://www.giib.or.jp/mtsnp/index_e.html
http://www.giib.or.jp/mtsnp/search_mtSNP_e.html
Mitochondrial SNP Project
GenBank GI: 14017399
Number detected SNPs: 81a
Number Unclassified SNPs: N/Aa
Number intronic SNPs: N/Ab
Number exonic SNPs: N/Ab
Number coding SNPs: N/Ab
Number amino-acid changes: N/Ab
N/A, not available.
a
GenBank Mapping Annotation Statistics.
b
Contig Assembly Annotation Statistics.
Molecular Psychiatry
These SNPs were collected from 16S rRNA
Antidepressant-induced gene expression
M-L Wong et al
Antidepressant effect in the function of cell
building—down regulation of ribosomal proteins
mRNA: ribosomal protein S4 and S29
Several lines of evidence indicate that ribosomal
RNAs are responsible for protein synthesis; therefore,
ribosomal proteins could have been recruited during
evolution to stabilize the configuration of ribosomal
RNA and improve protein synthesis.41 Thus, ribosomal proteins may also have extraribosomal functions,
which are still largely unknown. Protein production
is important for several cellular functions, and it is
essential for cell proliferation and neurogenesis.
Evidence that treatment with antidepressants can
alter neuronal cell proliferation in vivo and in vitro
has already been documented. In vitro experiments
suggest that fluoxetine is capable of inhibiting and
stimulating cell proliferation in non-neuronal
cells.42,43 Prolonged treatment with fluoxetine increased neurogenesis in adult rat hippocampus, but
in vitro findings showed a dual effect of antidepressant in cerebellar cell cultures.44 The role possible of
altered functioning of ribosomal protein in human
disease is largely unexplored.
Antidepressant effect in the function of cell
scaffolding/transport—upregulation of microtubule
mRNA: microtubule-associated protein 1A(MAP1A)
Microtubules have a crucial organizing role in all
eucaryotic cells. A variety of accessory proteins bind
to microtubules and serve various functions. MAPs
can stabilize microtubules against disassembly or
cross-link microtubules in the cytosol. MAP assemblies can be separated in two types, type I MAP
(MAP1A and 1B) and type II (MAP2, 4 and Tau). Type
I MAPs are large filamentous molecules found in
axons and dendrites of neurons. Chronic but not acute
treatment with desipramine inhibited microtubule
assembly in vivo and the degree of phosphorylation of
serine residues of MAP2 was significantly increased
after chronic administration of desipramine without
changes in the total concentration of MAP2.45 Given
that MAP2 and Tau are major cytoskeletal proteins in
neurons, which are involved in the pathogenesis of
Alzheimer’s and other neuropsychiatric diseases,46,47
and that it is not uncommon that antidepressants are
included in the treatment regimen of those conditions, a better understanding of the effects of
antidepressant treatment on microtubule proteins
would be relevant.
Antidepressant effect in the function of cell fueling:
downregulation of EST202275 (similar to
mitochondrial genome mRNA) and modulation of rat
carnitine palmitoyltransferase Ib 1, 2, and 3 genes.
Antidepressants can affect brain energy metabolism
and substrate oxidation pattern. It has been long
suggested that antidepressants may act by making
more energy available to overcome the depressed
stated.48 Initial observations supported that treatment
with desmethylimipramine for 7 days resulted in
increase in glucose utilization in several brain areas,
but that treatment for 28 days was inhibitory.49 Data
on oxidative energy metabolism suggest that IMI had
a maximum stimulation of respiration in the brain
mitochondria that was sustained after 1- and 2-week
treatment.50 It should however be noted that IMI at
high concentrations can inhibit ATPase activity;51
moreover, uncoupling effects on oxidative phosphorylation in rat liver mitochondria have been described
with melipramine.52 It is possible that the human
ortholog of the gene represented in rat EST202275 is
the mitochondrial 16S rRNA gene. This gene seem to
also encode humanin, a unique oncopeptide that
inhibits neuronal death by an antagonistic interaction
with members of the apoptotic signaling machinery.53
This finding would support the hypothesis that
antidepressants might be neuroprotective agents,
possibly exerting their neuroprotective effects
through diverse cellular and molecular mechanisms.
Additional mitochondrial process can also be
affected by antidepressants, such as monoamine
oxidase activity and the oxidation of fatty acids.
CNS and peripheral mitochondrial monoamine oxidase (MAO) are influenced by antidepressants.54,55
Long-term effects of several antidepressant drugs,
including IMI, can inhibit brain mitochondrial MAO
activities of MAO-A and MAO-B forms.56 Oxidation
of fatty acids in the mitochondria is initiated by a
sequence of events, of which CPT I and CPTII are
important elements. Mitochondrial b-oxidation of
long-chain fatty acids produces a large amount of
ATP; it is the main source of energy for skeletal
muscle during exercise and for cardiac muscle, and
during prolonged fasting. In our experiments, modulation of carnitine palmitoyltransferase Ib 1,2, and 3
genes by SJW and IMI treatments were of opposite
effect. It has been known that antidepressants disturb
lipid turnover in different cell types, such as
lymphocytes, monocytes, and human histiocytic
lymphoma cell line U-937.57,58
There is evidence that mitochondrial DNA polymorphisms may increase the susceptibility to Alzheimer disease, indicating that mitochondrial function
might be relevant to a psychiatric phenotype.59–61
247
Repeated DNA sequences: long interspersed elements
(LINE)
Rat LINE3 (or L1Rn) was included among the few
genes that were differentially regulated by chronic
antidepressant treatment with both IMI and SJW. This
gene belongs to a family of repetitive DNA elements,
which is ubiquitous in mammals, namely the L1
family. Rat LINE3 has six putative open reading
frames; the functions of these putative proteins are
unknown. About 50 000 copies of L1 elements occur
in the human genome, it accounts for about 5% of the
total human DNA. L1 elements transpose through
reverse transcription of an RNA intermediate.62 Given
that over 40% of the mammalian genome can be
composed of repetitive elements of four major classes,
could it be predictable that repetitive DNA sequences
would be implicated in paradigms involving longMolecular Psychiatry
Antidepressant-induced gene expression
M-L Wong et al
248
pressin, growth hormone-releasing factor, proenkephalin, pituitary glycoprotein hormone alpha
subunit, and promelanin-concentrating hormone),
neuroimmune factors (transthyretin, thymosin
beta-10, prothymosin alpha), glutamate (N-methyl
D-aspartate 2A receptor), metabolic pathway (glycolytic) genes, protein synthesis and degradation (proteasome) and synaptic function (SC2 synaptic
glycoprotein, synaptojanin 1, synaptosomal associated protein (SNAP-25A), syntaxin 1a and 2, and
vesicle-associated membrane protein 1) (refer to Table
1). Interestingly, differential expression of vesicleassociated membrane protein 2 (VAMPs/synaptobrevin-2) has been found after antidepressant and
electroconvulsive treatment in rat frontal cortex.63
Additionally, signaling and calcium-related components appear to be modulated by long-term antidepressant treatment. These systems have been recently
implicated in the pathophysiology of major depression and in the mechanism of action of antidepressant drugs.21,64
Figure 5 Summary of the study’s strategy, findings, and
potential applications.
term treatments? The function of most families of
repeated DNA is unknown: only 1–2% of those
correspond to families of known function (ribosomal
RNA, histones).
Intriguingly, the theme of repeated DNA sequences
is present in a significant proportion of differentially
regulated transcripts. For instance, several copies of
ribosomal proteins sequences can be found in the
genome. Several copies of the mitochondrial DNA
genome are located in the mitochondrial matrix; each
cell contains hundreds of mitochondrias and thousands of mtDNA copies. Type I MAP has several
repeats on the amino-acid sequence KKEX, which is
implicated as a binding site for tubulin.
Additional relevant systems identified by the
functional characterization of differentially regulated
genes
As we focused on hypothalamic tissue, it should not
be surprising that our results have not shown many
significant changes in classical neurotransmitter
systems. Nevertheless, our results support the concept that the cellular and molecular effects of
antidepressants might involve neurohormones (vasoMolecular Psychiatry
Transcriptional regulation
The changes in gene expression that are observed in
response to both acute and chronic antidepressant
treatment provide insight into the underlying regulatory networks that modulate neural plasticity necessary to respond to chronic stress and subsequent
depression induced by such chronic stress. The
results of chronic antidepressant treatment ought to
be conceptualized in the context of the transcriptional
regulators, which ultimately mediate the observed
changes in gene expression. Specific effects of
antidepressants or electroconvulsive treatment have
already been reported for immediate early genes and
transcription factors such as c-fos, zif268, NGF1-A,
phosphorylation of CRE-binding protein, DfosB, and
Narp.65–72
In our analysis, we identified four genes that were
downregulated in response to both IMI and SJW. Such
genes could be potential markers for predicting
response to antidepressant treatment; however, the
fact that genes are coordinately downregulated suggests there are specific combinations of transcriptional regulators that control the expression of these
four genes. Even though the transcription factors
themselves may not exhibit significant differences in
expression levels, changes in their ability to regulate
gene expression can still occur. We have identified
four transcription factors that undergo changes in
expression specifically in response to IMI. Of these
four transcription factors, AI23025 is a known
inhibitor of DNA binding via dominant-negative
effects that prevent other transcription factors from
building a suitable scaffold for the recruitment of
polymerase.
Additionally, the combination of transcription
factors, including a known repressor, in combination
with the modulation of other genes, including
humanin, which inhibits specific pathways, such as
apoptosis, suggests a real shift in the genetic program
Antidepressant-induced gene expression
M-L Wong et al
after antidepressant treatment. Chronic antidepressants may modify the expression levels of genes
known to antagonize various pathways and cellular
systems.
These results represent the net changes of many cell
types present in hypothalamic tissue; therefore, it is
not unexpected that most of the changes we identified
are below two-fold. It was not within the scope of this
work to replicate the relatively small changes of CRH
gene expression in specific hypothalamic areas, such
as in the PVN (reduction of 40%) that have consistently been reported after chronic antidepressant or
SJW treatment.15,23–25
Of note, two transcripts that displayed large downregulation belong to the immune/inflammation related group; they are the transthyretin (for IMI
treatment) and the interleukin-3 beta subunit (for
SJW treatment) genes.
The findings reported here do not replicate those
reported by Landgrebe et al.73 as they characterized
the effects of treatment with mirtazapine or paroxetine in total brain RNA of mice using a customized
microarray. Differences in study design could account
for these discrepancies. We studied a specific brain
region (hypothalamus), while they studied whole
brain; our study examined rats, theirs, mice. Finally,
the duration of our long-term treatment (8 weeks) was
different from the paradigm used in that study (1 or 4
weeks). In spite of these methodological differences,
which preclude direct comparison of results, it is
noteworthy that Landgrebe et al’s data provide
evidence that some antidepressants can predominantly downregulate gene expression, and that we
found similar downregulation effects after chronic
IMI treatment.
We need to consider the possibility that our
results underreport actual changes because we did
not focus on a specific neuronal nucleus; consequently, additional genes could be identified in future
studies by adopting specific dissection strategies.
Furthermore, as it is the case in most microarray
studies, the number of genes queried in our study
was disproportionally larger than the number of
samples per group. To address that key issue, we
used analyses approaches described in the literature
to define differentially expressed genes. Further
studies are needed to characterize the localization,
time course, and intensity of changes we report here.
It is unclear how many of these changes in mRNA will
reflect into protein expression and be functionally
relevant.
A novel observation emerging from this study is
that several essential general cell functions may be
affected by chronic antidepressant treatments (see
figure 5). These functions include protein synthesis
and degradation, cellular scaffolding and intracellular
transport, mitochondrial and glycolytic energy metabolism, and cellular signaling through modulation of
calcium biding proteins. Pathways that may be
modulated by chronic antidepressant treatments
include functions that may protect neurons from
cellular injury. Our experiments provide a framework
for the changes that occur during long-term administration of hypericum and IMI. It is possible that the
regulation of genes in the same functional group
could result in the same final outcome along a
signaling pathway. These findings could form the
basis of subsequent bioinformatics investigations into
the pathways underlying both the progression and
treatment of depression.
The ability to understand the results of microarray
experiments and translate them into novel tools and
targets for advancing the state of knowledge in the
study of depression and response to antidepressant
treatment can be greatly enhanced by identifying
neuronal signaling pathways. Such models could
serve as a framework for testing our hypotheses and
refining them where necessary. The lack of complete
understanding requires model building and hypothesis testing. The fact that several of the genes
identified in this study contain dense numbers of
SNPs is exciting because those SNPs may represent
potential new candidates for the pharmacogenomic
assessment of the phenotype of antidepressant treatment response. The combination of carefully planned
genetic screens and informatics mining of expression
data to extend and complete candidate pathways can
eventually lead to the development of a new generation of treatment strategies and possibly diagnostic
approaches for depression.
249
Acknowledgements
We thank Aaron Bertalmio, Xueying Gu, Che Hutson,
and Dr João R Oliveira for their technical help. We
express our gratitude to Botanicals International for
providing a large amount of SJW extract from a single
lot for our studies, and to Dr. H. Howardsun from
Pharmanox for his analysis of SJW extract. We thank
Drs. David Heber and Vay Lian W. Go for their
support.
This work was supported by NIH grants AT00151,
MH062777, RR00865, and RR017365 (M-LW), RR017611, RR016996, HL004526, DK058851, DK063240,
HG002500, and GM061394 (JL), and by awards from
NARSAD (ML-W), Dana Foundation, and Stanley
Foundation (JL), and the Neuropsychiatric Institute
(University of California, Los Angeles, CA, USA).
References
1 Muller WE, Kasper S. Clinically used antidepressant drugs.
Pharmacopsychiatry 1997; 30(Suppl 2): 71.
2 Group HDTS. Effect of Hypericum perforatum (St John’s wort) in
major depressive disorder: a randomized controlled trial. JAMA
2002; 287: 1807–1814.
3 Shelton RC, Keller MB, Gelenberg A, Dunner DL, Hirschfeld R,
Thase ME et al. Effectiveness of St John’s wort in major
depression: a randomized controlled trial. JAMA 2001; 285:
1978–1986.
4 Linde K, Ramirez G, Mulrow CD, Pauls A, Weidenhammer W, Mel
D. St John’s wort for depression—an overview and meta-analysis
of randomised clinical trials. BMJ 1996; 313: 253–258.
Molecular Psychiatry
Antidepressant-induced gene expression
M-L Wong et al
250
5 Stevinson C, Ernst E. Hypericum for depression. An update of the
clinical evidence. Eur Neuropsychopharmacol 1999; 9: 501–505.
6 Lecrubier Y, Clerc G, Didi R, Kieser M. Efficacy of St. John’s wort
extract WS 5570 in major depression: a double-blind, placebocontrolled trial. Am J Psychiatry 2002; 159: 1361–1366.
7 Piscitelli SC, Kress DR, Bertz RJ, Pau A, Davey R. The effect of
ritonavir on the pharmacokinetics of meperidine and normeperidine. Pharmacotherapy 2000; 20: 549–553.
8 Ruschitzka F, Meier PJ, Turina M, Luscher TF, Noll G. Acute heart
transplant rejection due to Saint John’s wort. Lancet 2000; 355:
548–549.
9 Barnes J, Mills SY, Abbot NC, Willoughby M, Ernst E. Different
standards for reporting ADRs to herbal remedies and conventional
OTC medicines: face-to-face interviews with 515 users of herbal
remedies. Br J Clin Pharmacol 1998; 45: 496–500.
10 Muller WE, Rolli M, Schafer C, Hafner U. Effects of hypericum
extract (LI 160) in biochemical models of antidepressant activity.
Pharmacopsychiatry 1997; 2: 102–107.
11 Muller WE, Singer A, Wonnemann M, Hafner U, Rolli M, Schafer
C. Hyperforin represents the neurotransmitter reuptake inhibiting
constituent of hypericum extract. Pharmacopsychiatry 1998; 1:
16–21.
12 Neary JT, Bu Y. Hypericum LI 160 inhibits uptake of serotonin and
norepinephrine in astrocytes. Brain Res 1999; 816: 358–363.
13 Kaehler ST, Sinner C, Chatterjee SS, Philippu A. Hyperforin
enhances the extracellular concentrations of catecholamines,
serotonin and glutamate in the rat locus coeruleus. Neurosci Lett
1999; 262: 199–202.
14 Chatterjee SS, Bhattacharya SK, Wonnemann M, Singer A, Muller
WE. Hyperforin as a possible antidepressant component of
hypericum extracts. Life Sci 1998; 63: 499–510.
15 Butterweck V, Winterhoff H, Herkenham M. St John’s wort,
hypericin, and imipramine: a comparative analysis of mRNA
levels in brain areas involved in HPA axis control following shortterm and long-term administration in normal and stressed rats.
Mol Psychiatry 2001; 6: 547–564.
16 Erdelmeier CA. Hyperforin, possibly the major non-nitrogenous
secondary metabolite of Hypericum perforatum L. Pharmacopsychiatry 1998; 1: 2–6.
17 Wong ML, Khatri P, Licinio J, Esposito A, Gold PW. Identification
of hypothalamic transcripts upregulated by antidepressants.
Biochem Biophys Res Commun 1996; 229: 275–279.
18 Berridge MJ, Downes CP, Hanley MR. Neural and developmental
actions of lithium: a unifying hypothesis. Cell 1989; 59: 411–419.
19 Manji HK, Chen G. PKC, MAP kinases and the bcl-2 family of
proteins as long-term targets for mood stabilizers. Mol Psychiatry
2002; 7(Suppl 1): S46–S56.
20 Williams RS, Cheng L, Mudge AW, Harwood AJ. A common
mechanism of action for three mood-stabilizing drugs. Nature
2002; 417: 292–295.
21 Wong M-L, Licinio J. Research and treatment approaches to
depression. Nature Rev Neurosci 2001; 2: 343–351.
22 Gold PW, Wong ML, Chrousos GP, Licinio J. Stress system
abnormalities in melancholic and atypical depression: molecular,
pathophysiological, and therapeutic implications. Mol Psychiatry
1996; 1: 257–264.
23 Brady LS, Whitfield Jr HJ, Fox RJ, Gold PW, Herkenham M. Longterm antidepressant administration alters corticotropin-releasing
hormone, tyrosine hydroxylase, and mineralocorticoid receptor
gene expression in rat brain. Therapeutic implications. J Clin
Invest 1991; 87: 831–837.
24 Brady LS, Gold PW, Herkenham M, Lynn AB, Whitfield Jr HJ. The
antidepressants fluoxetine, idazoxan and phenelzine alter corticotropin-releasing hormone and tyrosine hydroxylase mRNA levels
in rat brain: therapeutic implications. Brain Res 1992; 572: 117–
125.
25 Reul JM, Stec I, Soder M, Holsboer F. Chronic treatment of rats
with the antidepressant amitriptyline attenuates the activity of the
hypothalamic–pituitary–adrenocortical system. Endocrinology
1993; 133: 312–320.
26 Bhattacharya SK, Chakrabarti A, Chatterjee SS. Activity profiles of
two hyperforin-containing hypericum extracts in behavioral
models. Pharmacopsychiatry 1998; 1: 22–29.
Molecular Psychiatry
27 Li C, Wong WH. Model-based analysis of oligonucleotide arrays:
expression index computation and outlier detection. Proc Natl
Acad Sci USA 2001; 98: 31–36.
28 Lockhart DJ, Barlow C. Expressing what’s on your mind: DNA
arrays and the brain. Nat Rev Neurosci 2001; 2: 63–68.
29 Slonim DK. From patterns to pathways: gene expression data
analysis comes of age. Nat Genet 2002; 32(Suppl): 502–508.
30 Zhang MQ. Large-scale gene expression data analysis: a new
challenge to computational biologists. Genome Res 1999; 9: 681–
688.
31 Rice JA. Mathematical Statistics and Data Analysis. Wadsworth
Publishing Company: Pacific Grove, CA, 1988.
32 Olshen AB, Jain AN. Deriving quantitative conclusions
from microarray expression data. Bioinformatics 2002; 18: 961–
970.
33 Chu S, DeRisi J, Eisen M, Mulholland J, Botstein D, Brown PO et
al. The transcriptional program of sporulation in budding yeast.
Science 1998; 282: 699–705.
34 DeRisi JL, Iyer VR, Brown PO. Exploring the metabolic and genetic
control of gene expression on a genomic scale. Science 1997; 278:
680–686.
35 Schena M, Shalon D, Heller R, Chai A, Brown PO, Davis RW.
Parallel human genome analysis: microarray-based expression
monitoring of 1000 genes. Proc Natl Acad Sci USA 1996; 93:
10614–10619.
36 Wodicka L, Dong H, Mittmann M, Ho MH, Lockhart DJ. Genomewide expression monitoring in Saccharomyces cerevisiae. Nat
Biotechnol 1997; 15: 1359–1367.
37 Fambrough D, McClure K, Kazlauskas A, Lander ES. Diverse
signaling pathways activated by growth factor receptors induce
broadly overlapping, rather than independent, sets of genes. Cell
1999; 97: 727–741.
38 Holstege FC, Jennings EG, Wyrick JJ, Lee TI, Hengartner CJ, Green
MR et al. Dissecting the regulatory circuitry of a eukaryotic
genome. Cell 1998; 95: 717–728.
39 Eisen MB, Spellman PT, Brown PO, Botstein D. Cluster analysis
and display of genome-wide expression patterns. Proc Natl Acad
Sci USA 1998; 95: 14863–14868.
40 Michelson D, Galliven E, Hill L, Demitrack M, Chrousos G, Gold P.
Chronic imipramine is associated with diminished hypothalamic–
pituitary–adrenal axis responsivity in healthy humans. J Clin
Endocrinol Metab 1997; 82: 2601–2606.
41 Wool IG. Extraribosomal functions of ribosomal proteins. Trends
Biochem Sci 1996; 21: 164–165.
42 Edgar VA, Sterin-Borda L, Cremaschi GA, Genaro AM. Role of
protein kinase C and cAMP in fluoxetine effects on human T-cell
proliferation. Eur J Pharmacol 1999; 372: 65–73.
43 Genaro AM, Edgar VA, Sterin-Borda L. Differential effects of
fluoxetine on murine B-cell proliferation depending on the
biochemical pathways triggered by distinct mitogens. Biochem
Pharmacol 2000; 60: 1279–1283.
44 Manev H, Uz T, Smalheiser NR, Manev R. Antidepressants alter
cell proliferation in the adult brain in vivo and in neural cultures
in vitro. Eur J Pharmacol 2001; 411: 67–70.
45 Miyamoto S, Asakura M, Sasuga Y, Osada K, Bodaiji N, Imafuku J
et al. Effects of long-term treatment with desipramine on
microtubule proteins in rat cerebral cortex. Eur J Pharmacol
1997; 333: 279–287.
46 Spillantini MG, Murrell JR, Goedert M, Farlow MR, Klug A, Ghetti
B. Mutation in the tau gene in familial multiple system tauopathy
with presenile dementia. Proc Natl Acad Sci USA 1998; 95: 7737–
7741.
47 Hutton M, Lendon CL, Rizzu P, Baker M, Froelich S, Houlden H
et al. Association of missense and 50 -splice-site mutations in
tau with the inherited dementia FTDP-17. Nature 1998; 393: 702–
705.
48 Mata M, Fink DJ, Gainer H, Smith CB, Davidsen L, Savaki H et al.
Activity-dependent energy metabolism in rat posterior pituitary
primarily reflects sodium pump activity. J Neurochem 1980; 34:
213–215.
49 Gerber III JC, Choki J, Brunswick DJ, Reivich M, Frazer A. The
effect of antidepressant drugs on regional cerebral glucose
utilization in the rat. Brain Res 1983; 269: 319–325.
Antidepressant-induced gene expression
M-L Wong et al
50 Katyare SS, Rajan RR. Effect of long-term in vivo treatment with
imipramine on the oxidative energy metabolism in rat brain
mitochondria. Comp Biochem Physiol C Pharmacol Toxicol
Endocrinol 1995; 112: 353–357.
51 Eto K, Fukuda T, Araki Y, Inoue B, Ogata M. Effect of tricyclic
drugs on mitochondrial membrane. Acta Med Okayama 1985; 39:
289–295.
52 Burbenskaya NM, Nartsissov YR, Tsofina LM, Komissarova IA.
The uncoupling effect of some psychotropic drugs on oxidative
phosphorylation in rat liver mitochondria. Biochem Mol Biol Int
1998; 45: 261–268.
53 Maximov V, Martynenko A, Hunsmann G, Tarantul V. Mitochondrial 16S rRNA gene encodes a functional peptide, a potential
drug for Alzheimer’s disease and target for cancer therapy. Med
Hypotheses 2002; 59: 670–673.
54 Roth JA, Gillis CN. Inhibition of lung, liver and brain monoamine
oxidase by imipramine and desipramine. Biochem Pharmacol
1974; 23: 1138–1140.
55 Nag M, Nandi N. In vitro and in vivo effect of chloropromazine,
imipramine and lithium chloride on monoamine oxidase activity
in rat brain mitochondria. Biosci Rep 1987; 7: 701–704.
56 Egashira T, Takayama F, Yamanaka Y. Effects of long-term
treatment with dicyclic, tricyclic, tetracyclic, and noncyclic
antidepressant drugs on monoamine oxidase activity in mouse
brain. Gen Pharmacol 1996; 27: 773–778.
57 Kodavanti UP, Mehendale HM. Cationic amphiphilic drugs
and phospholipid storage disorder. Pharmacol Rev 1990; 42:
327–354.
58 Xia Z, Appelkvist EL, DePierre JW, Nassberger L. Tricyclic
antidepressant-induced lipidosis in human peripheral monocytes
in vitro, as well as in a monocyte-derived cell line, as monitored by
spectrofluorimetry and flow cytometry after staining with Nile red.
Biochem Pharmacol 1997; 53: 1521–1532.
59 Lin FH, Lin R, Wisniewski HM, Hwang YW, Grundke-Iqbal I,
Healy-Louie G et al. Detection of point mutations in codon 331 of
mitochondrial NADH dehydrogenase subunit 2 in Alzheimer’s
brains. Biochem Biophys Res Commun 1992; 182: 238–246.
60 Corral-Debrinski M, Horton T, Lott MT, Shoffner JM, McKee AC,
Beal MF et al. Marked changes in mitochondrial DNA deletion
levels in Alzheimer brains. Genomics 1994; 23: 471–476.
61 Davis RE, Miller S, Herrnstadt C, Ghosh SS, Fahy E, Shinobu LA et
al. Mutations in mitochondrial cytochrome c oxidase genes
segregate with late-onset Alzheimer disease. Proc Natl Acad Sci
USA 1997; 94: 4526–4531.
62 Boeke JD, Garfinkel DJ, Styles CA, Fink GR. Ty elements transpose
through an RNA intermediate. Cell 1985; 40: 491–500.
63 Yamada M, Takahashi K, Tsunoda M, Nishioka G, Kudo K, Ohata
H et al. Differential expression of VAMPs/synaptobrevin-2 after
antidepressant and electroconvulsive treatment in rat frontal
cortex. Pharmacogenomics J 2003; 2: 377–382.
64 Manji HK, Drevets WC, Charney DS. The cellular neurobiology of
depression. Nat Med 2001; 7: 541–547.
65 Dahmen N, Fehr C, Reuss S, Hiemke C. Stimulation of immediate
early gene expression by desipramine in rat brain. Biol Psychiatry
1997; 42: 317–323.
66 Torres G, Horowitz JM, Laflamme N, Rivest S. Fluoxetine induces
the transcription of genes encoding c-fos, corticotropin-releasing
factor and its type 1 receptor in rat brain. Neuroscience 1998; 87:
463–477.
67 Johansson IM, Bjartmar L, Marcusson J, Ross SB, Seckl JR,
Olsson T. Chronic amitriptyline treatment induces hippocampal
NGFI-A, glucocorticoid receptor and mineralocorticoid receptor
mRNA expression in rats. Brain Res Mol Brain Res 1998; 62:
92–95.
68 Bjartmar L, Johansson IM, Marcusson J, Ross SB, Seckl JR, Olsson
T. Selective effects on NGFI-A, MR, GR and NGFI-B hippocampal
mRNA expression after chronic treatment with different subclasses of antidepressants in the rat. Psychopharmacology (Berlin)
2000; 151: 7–12.
69 Thome J, Sakai N, Shin K, Steffen C, Zhang YJ, Impey S et al.
cAMP response element-mediated gene transcription is upregulated by chronic antidepressant treatment. J Neurosci 2000; 20:
4030–4036.
70 Chen J, Nye HE, Kelz MB, Hiroi N, Nakabeppu Y, Hope BT et al.
Regulation of delta FosB and FosB-like proteins by electroconvulsive seizure and cocaine treatments. Mol Pharmacol 1995; 48:
880–889.
71 Reti IM, Baraban JM. Sustained increase in Narp protein expression following repeated electroconvulsive seizure. Neuropsychopharmacology 2000; 23: 439–443.
72 Nibuya M, Nestler EJ, Duman RS. Chronic antidepressant administration increases the expression of cAMP response element
binding protein (CREB) in rat hippocampus. J Neurosci 1996; 16:
2365–2372.
73 Landgrebe J, Welzl G, Metz T, van Gaalen MM, Ropers H, Wurst W
et al. Molecular characterisation of antidepressant effects in the
mouse brain using gene expression profiling. J Psychiatr Res 2002;
36: 119–129.
251
Molecular Psychiatry