Overview
The Hypothesis of NMDA Receptor
Hypofunction for Schizophrenia
1,2
1,3
Huey-Jen Chang, M.D. , Hsien-Yuan Lane, M.D., Ph.D. ,
4*
Guochuan E. Tsai, M.D., Ph.D.
Schizophrenia, a multifactorial mental disorder with polygenic inheritance as well as
environmental influences, encompasses a characteristic group of symptoms and neurocognitive
deficits. Cognitive function, a major determinant of quality of life and overall function in
schizophrenia, contributes more to the prognosis of the disease than positive symptoms, such as
delusions or hallucinations. Although its exact etiological mechanisms remain relatively
unknown, extensive studies are ongoing to explore. Among them, one of the primary causal
factors is dysfunction of the N-methyl-D-aspartate (NMDA)-type glutamate receptors. This
article reviews the clinical limitations of current antipsychotics in treating the core symptoms of
schizophrenia and the trend in the reconceptualization of the disease nature and treatment
modalities. The NMDA receptor model plays a critical role in the revolution of pharmaceutical
industry as a new set of drug targets in addition to those based on the traditional monoaminergic
models is proposed. The evidence of NMDA receptor hypofunction in schizophrenia is
accumulating from the investigations on the modulation of glutamatergic system, particularly
the intrinsic NMDA/glycine site, through genetic research and various clinical trials. A group of
“NMDA-enhancing agents,” being used either as adjuncts to typical/atypical antipsychotics or
as monotherapy, in schizophrenic patients, particularly those with refractory negative and
cognitive symptoms, offer efficacy in preclinical and early clinical trials. Novel therapeutic
agents acting as NMDA enhancers show promise as the next wave of drug development for
schizophrenia.
Key words: schizophrenia, NMDA receptor hypofunction, negative symptoms, cognitive
function (Taiwanese Journal of Psychiatry [Taipei] 2012; 26:
147-61)
1
2
Institute of Clinical Medical Science, China Medical University, Taichung, Taiwan Department of Psychiatry, Chiayi and Wanqiao
Branch, Taichung Veterans General Hospital, Chiayi, Taiwan Department of Psychiatry, China Medical University Hospital, Taichung,
Taiwan Department of Psychiatry, Harbor-UCLA Medical Center, Torrance, CA, U.S.A. Received: August 21, 2012; revised: August 23,
2012; accepted: August 24, 2012 *Corresponding author. No. 1000, West Carson Street, Torrance, CA 90509, U.SA. E-mail address:
Guochuan E. Tsai <etsai@labiomed.org>
3
4
• 148 • The
NMDA Hypofunction in Schizophrenia
Introduction
Schizophrenia, affecting about 1% of the
population worldwide, is a devastating and costly
illness due to its resistance to treatment, the consequences of relapse, and substantial economic
burden. Clinical symptoms of schizophrenia have
three (positive symptoms, negative symptoms, and
neurocognitive defi cits) main categories. The
latter two possess high predictive value for clinical
outcomes [1] and account for much of the
long-term morbidity of this illness. Optimizing the
treatment of schizophrenia will be an important
goal in the early era of 21st century, particularly
on negative symptoms, and neurocognitive
deficits.
Drug models have been applied extensively to study
the pathophysiology of schizophrenia and thus
provide a better insight into the neurobiology of the
disorder. Hypofunction of N-methyl-Daspartate
receptor (NMDAR) mediated neurotransmission is
implicated in the critical deficits associated with
many brain disorders, especially schizophrenia [2].
This is evidenced by observations of the clinical
simulation exerted by the noncompetitive
antagonists of NMDAR, phencyclidine (PCP) and
ketamine on nonpsychotic individuals and
schizophrenic patients [3]. Therefore, in addition to
the dopamine and serotonin hypotheses, the NMDA
hypofunction model of schizophrenia has recently
gained extensive attention.
Enhancing NMDAR neurotransmission has been
considered as a novel treatment approach, in
particular through the glycine “modulatory” component (the coagonist site) at these receptors to
avoid the excitotoxicity mediated through the glutamate binding site [4]. Recent advances in understanding the function, pharmacology, genetics and
structure of NMDAR have promoted a search for
new compounds that could be therapeutically beneficial. These compounds act on the coagonist
binding sites, either directly or indirectly (Figure
1). Various NMDA-enhancing agents have been
proposed and they have been or currently are under
extensive studies. Many clinical trials on
NMDA-enhancing agents have revealed encouraging results.
Glutamatergic Hypothesis
The classical dopamine hypothesis of schizophrenia [5] postulates that dopaminergic hyperactivity is responsible for the psychotic symptoms of
this disorder. First-generation antipsychotic drugs
([FGAs], typical antipsychotics), which were
developed in the 1950s with blockade of D2
receptor as a necessary therapeutic action [5], treat
positive symptoms of schizophrenia effectively. In
addition to D2 receptor blockade, the 5-HT
receptor blockade plays a contributory role in the
actions of the second-generation antipsychotic
drugs ([SGAs], atypical antipsychotics), thus has
advantages over typical antipsychotics in terms of
greater efficacy for improving positive and
negative symptoms, equivocal beneficial effects on
cognitive function, and less extrapyramidal side
effects and tardive dyskinesia. SGAs have been
gradually replacing the FGAs since 1990s and
became the first line treatments for schizophrenia.
Nevertheless, these medications still show limited
efficacy on negative and cognitive symptoms of
schizophrenia as well as qualities of life, and they
are associated with severe side effects, including
agranulocytosis, sudden cardiac death, stroke,
diabetes mellitus, hypercholesterolemia and
significant weight gain.
2A
Carlsson et al. [6] have pointed out that it is likely
the dopaminergic system is not the only
Figure 1.
Subunits and binding sites of NMDAR receptors. The NMDARs form heteromeric
receptor-ion channels comprising two copies of NR1 and NR2 (A-D) subunits. The subunits contain
binding sites for the agonist glutamate and the co-agonist glycine (or D-serine, D-alanine, D-cycloserine).
Mg , magnesium, is a channel blocker which blocks the channel at resting membrane potential and is
released upon depolarization. The NMDAR channel is highly permeable to Ca and Na , and its opening
requires simultaneous binding of glutamate/glycine and postsynaptic membrane depolarization. Zn , zinc,
blocks the channel in a voltage-independent manner. The polyamine site binds compounds, which either
potentiate or inhibit the activity of the receptor, depending on the combination of subunits forming each
NMDAR. The PCP receptor site, where non-competitive antagonists such as PCP and PCP-like
compounds (ketamine, and MK-801) bind.
2+
2+
+
2+
dysfunctional system in schizophrenia as post-might be involved in the pathophysiology of mortem
examination of the brains of patients with schizophrenia. Kim et al. [7] were among the pioschizophrenia
did not reveal alterations in the lev-neers in proposing a glutamatergic hypothesis of els of dopamine or
dopamine receptors. schizophrenia, based on findings of significantly Interactions between dopamine and
several other reduced cerebrospinal fluid (CSF) levels of glutaneurotransmitters in complex neural
networks mate in patients with schizophrenia compared have been revealed, thus other neurotransmitter
with normal controls. Studies of post-mortem systems are likely involved in the pathophysiolo-brain and
CSF have revealed a lower density of gy of schizophrenia. glutamatergic receptors and lower level of
glutaSeveral studies have provided evidence that mate in schizophrenic patients than in healthy a dysfunction in
glutamatergic neurotransmission comparison subjects [8]. Lower indices of gluta
• 150 • The
NMDA Hypofunction in Schizophrenia
matergic neurotransmission are in correlation to
more prominent thought disorder as well as ventricular enlargement [9]. Among these findings, the
most compelling and more direct evidence is
provided by the psychomimetic effects of the
NMDA antagonists, phencyclidine (PCP) and ketamine [8].
Glutamate and dopamine have been reported to
exhibit reciprocal actions at subcortical structures.
The mechanisms underlying hyperdopaminergic
function in schizophrenia may involve cortical
glutamatergic projections to dopamine neurons in
the midbrain [6]. Investigators further raised the
hypothesis that NMDARs which regulate
mesolimbic and mesocortical dopamine pathways
may be hypoactive in schizophrenia [10].
Dopaminergic neurons are mastered either directly
by corticofugal glutamatergic neurons or indirectly
through γ-aminobutyric acid (GABA)
interneurons, which act as accelerators or brakes,
respectively [11]. A descending glutamatergic
pathway projecting from cortical pyramidal neurons
to dopaminergic neurons in the ventral tegmental
area (VTA) normally functions as a brake on the
mesolimbic dopaminergic pathway through an
inhibitory GABAergic interneuron in the VTA,
resulting in tonic inhibition of dopamine release
from the mesolimbic pathway. Hypoactive
NMDAR, as in schizophrenics or induced by ketamine, in the VTA may fail to tonically inhibit mesolimbic dopaminergic neurons; this would cause
mesolimbic dopamine hyperactivity and thus the
positive symptoms. Different from the actions of
cortico-brainstem glutamatergic neurons on mesolimbic dopaminergic neurons, separate corticobrainstem glutamatergic neurons synapse directly
upon those dopaminergic neurons in the VTA that
project to the cortex, so-called mesocortical dopaminergic neurons, and normally act to tonically
excite them [10]. Therefore, hypoactive NMDAR
in the VTA would cause mesocortical dopamine
hypoactivity and thus the cognitive defi cits and/or
negative symptoms.
Although a focus on cognitive defi cits and/or
negative symptoms has come up with a more recent hypothesis of schizophrenia as a “glutamate
disorder” [12], the glutamatergic hypofunction
hypothesis is not in conflict with a role for dopamine in the pathogenesis of schizophrenia or with
the action of currently available antipsychotics.
This notion is supported by a rodent study which
demonstrated that persistent NMDAR blockade
produced a rapid and profound decrease in the
levels of D2 receptor mRNA and receptor density,
which suggests that NMDAR play an important
role in the expression of D2 receptors in basal
ganglia. Therefore, the interaction between glutamate and dopamine regulate the functions of the
basal ganglia [13]. It was further proposed that
dopamine receptor blockade might act secondarily
to balance glutamatergic insuffi ciency [6].
Consequently, enhancement of NMDAR-mediated
neurotransmission has been proposed as having the
therapeutic
potential
at
a
fundamental
pathophysiological level. Supporting the hypothesis, accumulating evidences from ongoing and
forthcoming clinical trials, using drugs acting on
NMDAR, give rise to optimism.
NMDA hypofunction model of schizophrenia
complements the dopamine and serotonin hypotheses and its role has gained extensive attention.
Over the last two decades, the relationship of
NMDA function and schizophrenia is supported
by the evidences of the effects of the noncompetitive antagonists of NMDAR. Anis et al. [14] firstly
reported that PCP and ketamine blocked the action
of NMDA in ion flow through the NMDAR in the
brain. A PCP receptor site was soon identified and
characterized [15], where non-competitive
antagonists, such as PCP, ketamine and MK
801 bind, within the NMDA ion channel which is
composed of both NR1 and NR2 subunits. These
NMDA antagonists induce psychiatric and physiological changes that closely resemble schizophrenia [3]. In contrast to amphetamine/dopamine
model, which implies increased dopaminergic activity in the brain, PCP induces not only positive
symptoms similar to amphetamine, but also negative symptoms and cognitive deficits seen in
schizophrenia [3]. The physiologic manifestations
of schizophrenia such as hypofrontality, disruption
of prepulse inhibition (PPI), enhanced subcortical
dopamine release, and increased metabolism and
extracellular glutamate levels in defined limbic
circuits [16] are demonstrated by these antagonists
as well. These findings suggest that schizophrenic
symptoms could possibly arise from attenuated
NMDAR-mediated neurotransmission. It was
further postulated that the psychotomimetic effects
may not be exerted by noncompetitive antagonists
alone but could be an outcome of any
dysfunctional attenuation of the NMDAR-mediated
neurotransmission [17].
NMDAR, a subtype of the ionotropic glutamate
receptor with a family of subunits identified thus
far including NR1, NR2A, NR2B, NR2C, and
NR2D [18], plays an important role in neurodevelopment and cognition. A functional NMDAR
is composed of multiple subunits including NR1
and one of four NR2 subunits (A-D), which contain binding sites for glycine and glutamate, respectively, to form heteromeric receptor-ion channels [18] (Figure 1). Some studies offer the
evidence of developmental regulation of specific
components of the NMDAR unit [18]. An animal
study of mRNAs encoding NMDA receptor subunits in the developing rat CNS provided evidence
that the NR1 gene is expressed in virtually all neocortical neurons at all stages, whereas the four
NR2 transcripts display dissimilar developmental
expression pattern [18]. Therefore, defi cits in
NMDA neurotransmission can potentially account
for developmental risk factors and cognitive
impairments in schizophrenia.
Evidence of NMDA Receptor
Hypofunction in Schizophrenia
Challenging tests with NMDA receptor
antagonists
PCP and ketamine function primarily by binding to
a site within the ion channel of the NMDAR that
blocks cations influx, thereby acting as
noncompetitive antagonist [19] (Figure 1). Binding
studies in brain tissue from schizophrenics on PCP
sites with PCP ligands (3H-MK-801, 3H-TCP)
show
significant
differences
between
schizophrenics and healthy comparisons. The regional distribution of PCP sites in controls is the
highest in frontal cortex, followed by entorhinal
cortex, hippocampus and amygdale, while the
fewest number of receptors is in the substantia
nigra and the nucleus dentatus [20]. In the postmortem brains of schizophrenia, PCP sites are
more abundant in regions including orbital frontal
cortex, amygdala, entorhinal area, hippocampus
and putamen [20].
Nonpsychotic volunteers
Studies have been conducted to demonstrate that
the administration of subanesthetic doses of PCP
to nonpsychotic subjects induced neuropsychological and behavioral psychopathology similar
to that associated with schizophrenia while acutely
PCP-intoxicated
individuals
were
almost
indistinguishable from symptomatic schizophrenic
patients [21]. But placebo-controlled, double-blind
studies of the effects of NMDA antagonists in
humans have been limited for safety concern as
highly potent PCP can induce pathomorphologi
• 152 • The
NMDA Hypofunction in Schizophrenia
cal changes in rat brain as well as prolonged toxic
psychosis and severe medical problems in humans
[21].
Ketamine, with lower binding affinity to NMDAR
and lower potency than PCP, produces minimal
cardiac and respiratory adverse effects and its
anesthetic and behavioral effects mitigate soon after
administration [22]. When subanesthetic dose
(0.3-0.5 mg/kg) was infused intravenously to
nonpsychotic subjects, ketamine produces an
avolitional state characterized by blunted affect,
withdrawal, and psychomotor retardation [23],
psychotic symptoms in the form of suspiciousness,
disorganization, and perceptual alterations [24], and
cognition deficits demonstrated by impaired
performance on tests of vigilance, verbal memory,
verbal fluency, and the Wisconsin Card Sorting Test
[3, 23]. Significant increase in the scores of Brief
Psychiatric Rating Scale (BPRS), Scale for the
Assessment of Negative Symptoms (SANS) and
Scale for the Assessment of Positive Symptoms
(SAPS) were observed [24]. Dissociative symptoms
were prominent, with depersonalization in particular
similar to the early feature of the schizophrenia
prodrome [3]. In addition, during smooth pursuit eye
tracking task, ketamine induces nystagmus and
oculomotor dysfunction that were similar to some of
the abnormalities seen in schizophrenia [8].
Patients with schizophrenia
Further studies were conducted in patients
with schizophrenia to elucidate the possible
underlying etiology of schizophrenia.
Chronic PCP abusers have commonly been
misdiagnosed as schizophrenics, whereas
PCP administration exacerbates symptoms
in chronic stabilized schizophrenic patients
[3, 21]. Dramatic exacerbation of psychotic
symptoms was observed in chronic
schizophrenics under administration of
subanes
thetic doses of PCP (0.1 mg/kg) [25]. These patients became more assertive, hostile, and unmanageable, and these changes lasted not only a few
hours as in nonpsychotic volunteers but around
four to six weeks [25], suggesting the substantial
NMDA vulnerability of this population that is easily to deteriorate further. Similar to PCP findings,
overall, patients with schizophrenia receiving
subanesthetic dose of ketamine experienced a brief
worsening of positive and negative symptoms as
well as further impairment in recall and recognition
memory [8].
Novel Therapy: Modulation of
NMDA Receptors
NMDA glycine-site agonists
NMDAR is unique as it consists of a co-agonist site
that binds the endogenous full agonists, in addition
to the glutamate recognition site. Glycine, D-serine,
or D-alanine acts as an obligatory endogenous
co-agonist for activation of the NMDAR complex
through a strychnine-insensitive site on the NR1
subunit [26] (Figure 1). Several approaches have
emerged aiming towards modulating this co-agonist
site. D-serine is a more potent agonist than glycine
at the coagonist site and has a greater ability to
penetrate the blood brain barrier [27], indicating
that D-serine administration may be more
efficacious for the symptoms of schizophrenia.
D-serine is synthesized in protoplasmic astrocytes
by SR that reversibly converts L-to D-serine and is
degraded by D-amino acid oxidase (DAAO) into
hydroxylpyruvate [27]. Reduction of central and
peripheral D-serine levels in schizophrenic patients
resulting in impaired D-serine function could
contribute to NMDAR hypofunction [27]. Several
controlled clinical trials have shown that
co-administration of D-serine or glycine with
antipsychotics can ameliorate
154 • The NMDA Hypofunction in Schizophrenia
some symptoms of the disorder [9, 10, 28].
D-alanine, another endogenous full agonist of the
coagonist site, might also have benefi cial effects on
schizophrenia [26]. D-cycloserine, an anti-tuberculosis drug, is a partial agonist at the coagonist
site. Its clinical efficacy is less than the full agonists
[29] as the intrinsic activity of D-cycloserine is only
about half that of full agonists (i.e., glycine,
D-serine) in potentiating NMDAR activation.
D-cycloserine may even function as a partial
antagonist in the presence of high levels of glycine
and D-serine, leading to decreased NMDA
neurotransmission [29].
phenyl)-3-(4′-phenylphenoxy) propyl] sarcosine
(NFPS), a specific inhibitor of GlyT-1, potentiated
NMDAR-mediated responses, such as increased
LTP in the dentate gyrus and enhanced PPI of
acoustic startle, in vivo [31]. GlyT-1 knockdown
mutation further demonstrated that the effects exerted by NFPS were indeed mediated by GlyT-1
[32]. Sarcosine, which is an endogenous inhibitor
of GlyT-1, has shown clinical efficacy while being
administered as add-on therapy to FGAs and SGAs
or as monotherapy, thereby supporting its
NMDA-enhancing and antipsychotic functions [4,
33, 34] ( Figure 2).
Glycine transport inhibitors
Modulation of the NMDAR function can be done
through a number of sites other than the coagonist
site. Extracellular glycine levels are regulated
through uptake by two types of high affinity
glycine transporters, GlyT-1 and GlyT-2. GlyT-2
has a more limited distribution, predominantly in
brain stem and spinal cord neurons, and is thought
to provide the principal glycine uptake mechanism
at inhibitory glycinergic synapses [30], whereas
GlyT-1 is widely expressed in glial cells of the
hippocampus, cortex, and cerebellum, as well as
the brain stem and spinal cord in association with
NMDAR [30] and it has been proposed to function
mainly at excitatory synapses by regulating glycine
levels at the coagonist binding site of NMDAR.
D-amino acid oxidase inhibitors
DAAO is a flavoenzyme that catalyzes the oxidative
deamination of D-amino acids. D-serine, the
endogenous NMDAR co-agonist, is the predominant
D-amino acid in mammalian CNS. DAAO might
have the potential in modulating NMDAR function
via D-serine degradation [35], contributing to the
reduction in NMDAR-mediated neurotransmission,
thus increased DAAO is proposed to be in
association with susceptibility to schizophrenia.
Consistent with the altered D-serine metabolism in
schizophrenia, lower serum levels of D-serine were
found in schizophrenic patients as compared to
healthy controls [36].
As evidence suggests that synaptic glycine may be
efficiently regulated at a subsaturating level by the
GlyT-1, a rational approach to enhance NMDA
neurotransmission might be through blocking the
reuptake of glycine by GlyT-1. This mechanism is
analogous to that of using a serotonin reuptake
inhibitor to potentiate serotonergic
neurotransmission [17]. In support of this hypothesis, it was demonstrated that N [3-(4′-fluoro
Several lines of evidence support the above
hypothesis: in postmortem studies, the mean
DAAO activity is two-fold higher in the schizophrenia group compared with the control group and
there is increased DAAO expression in the bilateral
hippocampal CA4 of the schizophrenia group [37];
in genetic studies, DAAO shows associations to
schizophrenia in several, though not all, studies
[35]. Collectively, DAAO inhibition and D-serine
elevation in combination is suggestive of potential
therapeutic benefits.
Figure 2.
Schematic illustration of glutamatergic system. The ionotropic glutamate receptors,
including N-methyl-D-aspartate (NMDA), α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
(AMPA) and kainate subtypes, are located in the postsynaptic membrane. They participate in synaptic
transmission by directly opening ion channels upon glutamate binding, allowing ion infl ux (Na , Ca ) and
causing excitatory post-synaptic current (EPSC) and mainly function to mediate fast synaptic transmission.
Among them, NMDARs are the subtype with strongest implication in the pathophysiology of
schizophrenia. Glutamate is agonist, which is synthesized and stored in high concentration within
presynaptic nerve terminals and released from the nerve terminal into the synaptic cleft; glycine and
D-serine are co-agonists of the NMDAR. D-serine is synthesized by serine racemase from L-serine.
D-serine is localized to both neurons and astrocytes and is uptaken by ASC-1 in the presynaptic membrane. D-serine is metabolized by DAAO into hydroxyl pyruvate. Role of DAOA (G72) as an activator or
inhibitor is unclear. Glycine is uptaken by GlyT-1 and metabolized to L-serine by glycine cleavage system
(GCS). Sarcosine inhibits glycine uptake through GlyT-1. DAAO inhibitors act to reduce D-amino acids
degradation. Other potential regulators and drug targets of NMDA synapse include the “glycine”
co-agonist site, serine racemase, D-amino acid oxidase activator (DAOA, G72), and
arginine-serine-cysteine transporter-1(ASC-1). AMPA receptors have the characteristic of being mobile
and they function cooperatively with NMDARs to maintain overall integrity of glutamatergic synapses.
Activation of AMPA receptor depolarizes the synaptic membrane allowing Ca influx through unblocked
NMDA channels in a voltage-dependent manner.
+
2+
2+
Therapeutic Effects Exerted
by NMDA-enhancing Agents
Tsai and Lin [38] performed a meta-analysis which
is the most comprehensive review of
NMDA-enhancing agents. It included all published,
randomized,
double-blind
trials
of
the
NMDA-enhancing agents and reviewed 26 studies
with about 800 patients with schizophrenia. The
clinical
efficacy
among
different
NMDA-enhancing agents as adjuncts to different
concomitant antipsychotic agents on different
symptom domains, efficacy, the dose-response and
the side effects were examined.
Glycine
Glycine added to FGAs
The efficacy of glycine as adjuvant therapy to
FGAs in treating negative symptoms and cognitive
deficits has been investigated in several
small-scale double-blind and open-label clinical
trials since 1990. One study [39] with 6 patients
found no beneficial effects regarding the negative
symptoms of schizophrenia. However, the limited
penetrability of glycine across the blood-brain
barrier and a relative low dose of glycine (10.8 g/
day) administered orally might explain the lack of
effectiveness in this study. Three studies revealed
significant improvement in negative symptoms
[40-42]. In 2007, the Cognitive and Negative
Symptoms in Schizophrenia Trial (CONSIST), a
randomized double-blind study with a duration of
16 weeks and with the participation of 4 sites in
the United States and one site in Israel, was published [43]. A total of 157 patients were randomly
assigned to glycine, D-cycloserine and placebo
groups. This study suggested that glycine is not an
effective therapeutic option for treating negative
symptoms or cognitive deficits as there was no
significant SANS total score differences between
glycine and placebo groups. Moreover, greater reductions in negative symptoms for both the glycine
and D-cycloserine groups in comparison to the
placebo group were detected in inpatients, but not
in outpatients. Furthermore, there was no significant glycine/placebo between-group difference
on the cognitive measure. In terms of adverse effect,
more new or worsened nausea was observed in
glycine subjects than in placebo subjects.
Three of the above mentioned trials [41, 42, 44]
also evaluated the cognitive effects of glycine
based on PANSS cognitive subscale, showing a
beneficial effect in cognitive function. But it is
noteworthy that these studies were conducted on
relatively small sample sizes and over short-term
periods.
Glycine added to SGAs
A 6-week double-blind placebo-controlled
crossover trial with high-dose glycine (0.8 g/kg
body weight per day) added to olanzapine and risperidone [45] revealed signifi cant improvements
in negative symptoms and cognitive function.
Another short-term trial [42] also showed a significant reduction in negative symptoms and an
improvement in cognitive function based on
PANSS cognitive subscale. However, these clinical
trials were conducted on small samples under
short-term treatment. In 2007, the CONSIST [43]
suggested that glycine is not an effective therapeutic option for patients with negative symptoms
or cognitive impairments not benefited by SGAs as
no significant difference in change in the SANS
total score and in the average cognitive domain Z
scores between glycine and placebo groups was
observed.
• 156 • The
NMDA Hypofunction in Schizophrenia
Glycine added to clozapine
About one third to two thirds of treatment-resistant
schizophrenic patients fail to benefit from clozapine
therapy or are partial responders despite adequate
dosage and duration. During the last two decades,
several clozapine adjunctive agents have come into
clinical practice in order to maximize the efficacy
of clozapine, including FGAs, SGAs, mood
stabilizers, other anticonvulsants, selective
serotonin reuptake inhibitors (SSRI), and
glycinergic agents. Since 1990’s the efficacy of
glycine as clozapine adjuncts has been evaluated in
several clinical trials. Seven trials [41, 42, 44,
46-49] were published. Five of them [42, 44, 46, 48,
49] did not find a signifi cant improvement in
positive symptoms. One study [47] found
worsening of positive symptoms. Four trials [41, 42,
44, 46] showed significant reductions in negative
symptoms and the improvement persisted after
discontinuation of glycine. Cognitive functioning
was assessed in only four trials [41, 42, 44, 48] and
three of them found positive change based on
cognitive subscale of PANSS.
D-serine
Five clinical trials with D-serine as adjuvant
therapy to antipsychotic treatment [9, 28, 34, 50,
51] were reported. First trial evaluated the addition
of D-serine to either FGAs or SGAs, with significant improvements in positive, negative and
cognitive symptoms [9]. Five trials [4, 28, 34, 50,
51] evaluated the addition of D-serine to SGAs
with three trials showed encouraging results. One
study [51] published in 2010 performed a 4-week,
double-blind investigation of adjunctive D-serine
at dose-escalation (30, 60 or 120 mg/kg body
weight/day) and the findings suggested beneficial
effects in treatment of positive symptoms, negative
symptoms, and cognitive deficits at doses of 60
and 120 mg/kg/day, and signifi cant dose-de
pendent increase of plasma D-serine levels correlated with improved symptomatic and neuropsychological function. Renal side effect was observed
at high D-serine dosage, which resolved upon
D-serine discontinuation. The other study
[28] also demonstrated signifi cant improvements
in positive, negative and cognitive symptoms. But
one study [34] with D-serine and risperidone cotreatment in patients with acute exacerbation of
symptoms and another study [50] with addition of
D-serine to clozapine did not fi nd benefi cial effects in any of the three core symptoms of
schizophrenia.
D-alanine
There is only one clinical trial [26] examining the
efficacy of D-alanine as adjuncts to anti-psychotics.
Thirty-two schizophrenic patients were enrolled in
this 6-week double-blind, placebo-controlled trial,
in which D-alanine (100 mg/ kg/day) was added to
their stable antipsychotic regimens, including
various typical antipsychotics and risperidone.
Significant improvements in positive and negative
symptoms were the findings. Cognitive symptoms
assessed by the cognitive subscale of PANSS also
revealed improvement. D-alanine was, moreover, a
well-tolerated compound, and no significant side
effect was noted [26]. The positive findings of
D-alanine as add-on therapy for schizophrenia,
particularly the negative and cognitive symptoms,
further support the hypothesis that augmentation of
NMDA neurotransmission through the NMDA
coagonist site is a promising approach for
schizophrenia.
D-cycloserine
D-cycloserine added to FGAs
Nine trials evaluated the addition of D-cycloserine
to FGAs in chronic schizophrenic patients [29, 43,
55-58]. The first study [52] re
ported that an exacerbation of positive symptoms
was observed in about 70% of the patients receiving adjunctive high-dose D-cycloserine (dosages
between 500 mg/day and 3,000 mg/day). The second study [53] reported that 250 mg D-cycloserine
daily aggravated psychotic symptoms in four of
seven patients and only one patient exhibited a
slight improvement. Three trials [55-57] exhibit
significant reductions in negative symptoms, while
three others [29, 43, 58] did not fi nd significant
changes in negative symptoms. In addition, one
trial [55] demonstrated equivocal improvement of
one cognitive task at the dose of 50 mg/ day, while
four other trials [29, 43, 56, 58] did not fi nd
significant changes in cognitive symptoms. One
study [54] with doses of 100 mg/day showed
worsening of positive symptoms and failed to improve negative symptoms. These unexpected findings may be explained by the antagonistic effects
of higher-dose D-cycloserine at the coagonist site
of the NMDAR due to competition with the endogenous agonist glycine and/or D-serine or its
interaction with antipsychotics which have endogenous activity on NMDAR-mediated neurotransmission [54].
D-cycloserine added to SGAs
Four trials evaluated the effect of D-cycloserine
added to SGAs [43, 57, 59, 60]. None of them
detected any significant change in positive
symptoms. Three of these studies [57, 59, 60]
demonstrated a significant improvement in negative
symptoms at the dose of 50 mg/day. CONSIST trial
[43] showed no effect of D-cycloserine at 50 mg/day
on negative symptoms. Two trials [43, 60] examined
the effects on cognitive function and revealed no
significant impact.
D-cycloserine added to clozapine
Three trials evaluated the effect of D-cycloserine as
add-on thearpy to clozapine [56, 59, 61]. No
significant effect on cognitive or positive symptoms
was observed in any of the trial. One trial with
D-cycloserine given at a dose of 50 mg/day [59]
found improvement in negative symptoms. However,
the other two trials [56, 61] with D-cycloserine
given at the same dose showed worsened negative
symptoms.
Sarcosine
Four double-blind placebo-controlled clinical trials
investigated the effects of sarcosine as adjuncts to
stable antipsychotic regimens. The first trial
[33]evaluated the addition of sarcosine (2 g/day) to
typical antipsychotics and to risperidone, revealing
significant improvements in the positive, negative
and cognitive symptoms. The second trial [34]
evaluated the addition of sarcosine (2 g/day) or
D-serine (2 g/day) to risperidone in patients with
acute exacerbation of schizophrenia, revealing
significant improvements in positive, negative, and
cognitive symptoms and the therapeutic effects of
sarcosine are superior to those of D-serine. Previous
studies found no beneficial effects of glycine,
D-serine, or D-cycloserine as add-on therapy to
clozapine, Lane et al. [62] thus further examined the
combined effect of sarcosine (2 g/day) and clozapine,
which exhibited no improvement in the core
symptoms.
Addition of sarcosine to an existing antipsychotic
regimen other than clozapine has shown its
efficacy for both chronically stable and acutely ill
patients. However, the efficacy of NMDA agents
as a primary antipsychotic agent has not yet been
demonstrated. Therefore, Lane et al. [63] evaluated
the effect of sarcosine monotherapy on 20 acutely
symptomatic schizophrenic patients, who
• 158 • The
NMDA Hypofunction in Schizophrenia
were randomly assigned to receive either 1 or 2 g
of sarcosine daily. Patients receiving the 2 g daily
dose were more likely to respond, particularly the
antipsychotic-naïve patients. However, in order to
fully assess the therapeutic effect of sarcosine,
placebo-or active-controlled, larger-sized studies
are needed. Recently, Lane et al. [4] conducted a
6-week double-blind, placebo-controlled trial with
enrollment of 60 chronic schizophrenic patients
comparing the effects of sarccosine (2 g/ day) and
D-serine as adjuncts to existing stable
antipsychotics. Greater efficacy of sarcosine therapy than D-serine for all outcome measures was the
finding whereas D-serine treatment has better
efficacy than placebo group.
fore future extensive study is expected to establish
their efficacy, tolerability, and mechanism.
Nevertheless, our recent findings revealed that a
DAAO inhibitor is much more effi cacious than
other NMDA-enhancing agents in improving the
symptoms of schizophrenia. Furthermore, it improves the cognition as measured by a comprehensive MATRICS-like cognitive battery.
Financial Disclosure
Intellectual property right of benzoate treatment is
under US2010/0189818, WO 2010/085452 and
Taiwan patent application, No.100101995.
Fundings and Supports
Perspectives
D-amino acid oxidase inhibitors
Experimental evidence supports that D-serine,
D-cycloserine, and D-alanine as adjuncts to FGAs
or SGAs are efficacious approaches in treating
negative symptoms and cognitive impairments.
Findings have been most encouraging in studies
that have used full agonists of the NMDA receptor
(i.e., glycine, D-serine, D-alanine) as opposed to
the partial antagonist D-cycloserine. Furthermore,
effects have generally been more powerful when
these agents were used in combination with
antipsychotics other than clozapine.
In addition to the coagonist site and the GlyT-1,
another novel drug target to enhance NMDA
function is DAAO which degrades D-amino acids
including D-serine, D-alanine and other D-amino
acids. Overactive DAAO may contribute to NMDA
hypofunction, thus inhibition of DAAO is a
plausible approach to enhance NMDAR function in
schizophrenia. The therapeutic value of DAAO
inhibitors is relatively unexplored and remains at
preclinical stage, there
This work was supported by the National Science
Council, Taiwan (NSC-97-2314-B-039006-MY3,
NSC-100-2627-B-039-001,
NSC-1012314-B-039-030-MY3,
and
NSC-101-2627-B-039-001),
National
Health
Research
Institutes,
Taiwan
(NHRI-EX-101-9904NI), Taiwan Department of
Health Clinical Trial and Research Center of
Excellence (DOH101-TD-B-111-004), and China
Medical University Hospital, Taichung, Taiwan
(DMR-99-153 and DMR-99-117).
References
1
Tamminga CA, Buchanan RW, Gold JM: The
role of negative symptoms and cognitive dysfunction in
schizophrenia outcome. Int Clin Psychopharmacol 1998;
13 Suppl 3: S21-6.
2
Javitt DC, Zukin SR: Recent advances in the
phencyclidine model of schizophrenia. Am J Psychiatry
1991; 148: 1301-8.
3
Krystal JH, Karper LP, Seibyl JP, et al.:
Subanesthetic effects of the noncompetitive NMDA
antagonist, ketamine, in humans: psychotomimetic,
perceptual, cognitive, and neuroendocrine responses. Arch
Gen
Psychiatry 1994; 51: 199-214.
4.
Lane HY, Lin CH, Huang YJ, Liao CH, Chang
YC, Tsai GE: A randomized, double-blind, placebo-controlled comparison study of sarcosine (N-methylglycine)
and D-serine add-on treatment for schizophrenia. Int J
Neuropsychopharmacol 2010; 13: 451-60.
15.
Lodge D, Johnson KM: Noncompetitive
excitatory amino acid receptor antagonists. Trends
Pharmacol Sci 1990;11: 81-6.
16.
Coyle JT, Tsai G, Goff DC: Ionotropic
glutamate receptors as therapeutic targets in schizophrenia.
Curr Drug Targets CNS Neurol Disord 2002; 1: 183-9.
17.
5.
Carlsson A, Lindqvist M: Effect of
chlorpromazine or haloperidol on formation of
3-methoxytyramine and normetanephrine in mouse brain.
Acta Pharmacol Toxicol (Copenh) 1963; 20: 140-4.
6.
Carlsson A, Hansson LO, Waters N, Carlsson
ML: Neurotransmitter aberrations in schizophrenia: new
perspectives and therapeutic implications. Life Sci 1997;
61: 75-94.
Tsai G: A New Class of Antipsychotic Drugs:
enhancing Neurotransmission Mediated by NMDA
Receptors. Psychiatric Times 2008: 25.
18.
Monyer H, Burnashev N, Laurie DJ, Sakmann B,
Seeburg PH: Developmental and regional expression in the
rat brain and functional properties of four NMDA receptors.
Neuron 1994; 12: 529-40.
19.
Kim JS, Kornhuber HH, Holzmüller B,
Schmid-Burgk W, Mergner T, Krzepinski G: Reduction
of cerebrospinal fluid glutamic acid in Huntington’s
chorea and in schizophrenic patients. Arch Psychiatr
Nervenkr 1980; 228: 7-10.
Jentsch J, Roth R: The
neuropsychopharmacology of phencyclidine: from NMDA
receptor hypofunction to the dopamine hypothesis of
schizophrenia. Neuropsychopharmacology 1999; 20:
201-25.
20.
Kornhuber J, Mack-Burkhardt F, Riederer P:
Regional distribution of 3H-MK-801 binding sites in the
human brain. Brain Res 1989; 489: 397-9.
8.
21.
7.
Mechri A, Saoud M, Khiari G, d’Amato T,
Dalery J, Gaha L: Glutaminergic hypothesis of
schizophrenia: clinical research studies with ketamine.
Encephale 2001; 27: 53-9.
9.
Tsai G, Yang P, Chung LC, Lange N, Coyle JT:
D-serine added to antipsychotics for the treatment of
schizophrenia. Biol Psychiatry 1998; 44: 1081-9.
10.
Goff DC, Coyle JT: The emerging role of
glutamate in the pathophysiology and treatment of
schizophrenia. Am J Psychiatry 2001; 158: 1367-77.
11.
Tsapakis EM, Travis MJ: Glutamate and
psychiatric disorders. Adv Psychiatr Treatment 2002; 8:
189-97.
12.
Insel TR: Rethinking schizophrenia. Nature
Allen RM, Young SJ: Phencyclidine-induced
psychosis. Am J Psychiatry 1978; 135: 1081-4.
22.
Moretti RJ, Hassan SZ, Goodman LI, Meltzer
HY: Comparison of ketamine and thiopental in healthy
volunteers: Effects on mental status, mood, and personality.
Anesth Analg 1984; 63: 1087-96.
23.
Cohen B, Rosenbaum G, Luby E, Gottlieb J:
Comparison of phencyclidine hydrochloride (sernyl) with
other drugs: simulation of schizophrenic performance with
phencyclidine hydrochloride (sernyl), lysergic acid
diethylamide (LSD-25), and amobarbital (Amytal) sodium,
II: symbolic and sequential thinking. Arch Gen Psychiatry
1962; 6: 79-85.
2010;468: 187-93.
13.
Qin ZH, Zhou LW, Weiss B: D2 dopamine
receptor messenger RNA is altered to a greater extent by
blockade of glutamate receptors than by blockade of
dopamine receptors. Neuroscience 1994; 60: 97-114.
24.
14.
25.
Anis NA, Berry SC, Burton NR, Lodge D: The
dissociative anesthetics ketamine and phencyclidine selectively reduce excitation of central mammalian neurons
by N-methyl-D-aspartate. Br J Pharmacol 1983; 79:
565-75.
van Berckel BN, Oranje B, van Ree JM,
Verbaten MN, Kahn RS: The effects of low dose ketamine
on sensory gating, neuroendocrine secretion and behavior
in healthy human subjects. Psychopharmacology (Berl)
1998; 137: 271-81.
Luby ED, Cohen BD, Rosenbaum G, Gottlieb
JS, Kelly R: Study of a new schizophrenic-like drug:
Sernyl. Sernyl Archives of Neurological Psychiatry 1959;
81: 363-9.
26.
D-alanine
Tsai GE, Yang P, Chang YC, Chong MY:
• 160 • The
NMDA Hypofunction in Schizophrenia
added to antipsychotics for the treatment of schizophrenia. Biol Psychiatry 2006; 59: 230-4.
27.
Hashimoto A, Oka T: Free D-aspartate and
D-serine in the mammalian brain and periphery. Prog
Neurobiol 1997; 52: 325-53.
28.
Heresco-Levy U, Javitt DC, Ebstein R, et al.:
D-serine efficacy as add-on pharmacotherapy to risperidone
and olanzapine for treatment-refractory schizophrenia. Biol
Psychiatry 2005; 57: 577-85.
29.
Goff DC, Herz L, Posever T, et al.: A six-month,
placebo-controlled trial of D-cycloserine co-administered
with conventional antipsychotics in schizophrenia patients.
Psychopharmacology 2005; 179: 144-50.
30.
Zafra F, Aragon C, Olivares L, Danbolt NC,
Gimenez C, Storm-Mathisen J: Glycine transporters are
differentially expressed among CNS cells. J Neurosci 1995;
15: 3952-69.
31.
Kinney GG, Sur C, Burno M, et al.: The glycine
transporter type 1 inhibitor N-[3-(4′-fluorophenyl)-3(4′
-phenylphenoxy) propyl] sarcosine potentiates NMDA
receptor-mediated responses in vivo and produces an
antipsychotic profile in rodent behavior. J Neurosci 2003;
23: 7586-91.
schizophrenia: evidence in support of the N-methylD-aspartate receptor hypofunction hypothesis of
schizophrenia. Arch Gen Psychiatry 2003; 60: 572-6.
37.
Habl G, Zink M, Petroianu G, et al.: Increased
D-amino acid oxidase expression in the bilateral hippocampal CA4 of schizophrenic patients: a postmortem
study. J Neural Transm 2009; 116: 1657-65.
38.
Tsai G, Lin PY: Strategies to enhance
N-methyl-Daspartate receptor-mediated neurotransmission
in schizophrenia, a critical review and meta-analysis. Curr
Pharm Des 2010; 16: 522-37.
39.
Rosse RB, Theut SK, Banay-Schwartz M, et al.:
Glycine adjuvant therapy to conventional neuroleptic
treatment in schizophrenia: an open-label, pilot study. Clin
Neuropharmacol 1989; 12: 416-24.
40.
Javitt DC, Zylberman I, Zukin SR,
Heresco-Levy U, Lindenmayer JP: Amelioration of
negative symptoms in schizophrenia by glycine. Am J
Psychiatry 1994; 151: 1234-6.
41.
Heresco-Levy U, Javitt DC, Ermilov M, Mordel
C, Silipo G, Lichtenstein M: Efficacy of high-dose glycine
in the treatment of enduring negative symptoms of
schizophrenia. Arch Gen Psychiatry 1999; 56: 29-36.
32.
Tsai G, Ralph-Williams RJ, Martina M, et al.:
Gene knockout of glycine transporter 1: characterization of
the behavioral phenotype. Proc Natl Acad Sci USA 2004;
101: 8485-90.
33.
Tsai G, Lane HY, Yang P, Chong MY, Lange N:
Glycine transporter I inhibitor, N-methylglycine (sarcosine)
added to antipsychotics for the treatment of schizophrenia.
Biol Psychiatry 2004; 55: 452-6.
34.
Lane HY, Chang YC, Liu YC, Chiu CC, Tsai
GE: Sarcosine or D-serine add-on treatment for acute exacerbation of schizophrenia: a randomized, double-blind,
placebo-controlled study. Arch Gen Psychiatry 2005; 62:
1196-204.
35.
Verrall L, Burnet PWJ, Betts JF, Harrison PJ:
The neurobiology of D-amino acid oxidase (DAO) and its
involvement in schizophrenia. Mol Psychiatry 2010;
15: 122-37.
36.
Hashimoto K, Fukushima T, Shimizu E, et al.:
Decreased serum levels of D-serine in patients with
42.
Javitt DC, Silipo G, Cienfuegos A, et al.:
Adjunctive high-dose glycine in the treatment of
schizophrenia. Int J Neuropsychopharmacol 2001; 4:
385-91.
43.
Buchanan RW, Javitt DC, Marder SR, et al.:
The cognitive and negative symptoms in schizophrenia
trial (CONSIST): the efficacy of glutamatergic agents for
negative symptoms and cognitive impairments. Am J
Psychiatry 2007; 164: 1593-602.
44.
Heresco-Levy U, Javitt DC, Ermilov M, Mordel
C, Horowitz A, Kelly D: Double-blind, placebo-controlled,
crossover trial of glycine adjuvant therapy for
treatment-resistant schizophrenia. Br J Psychiatry 1996;
169: 610-7.
45.
Heresco-Levy U, Ermilov M, Lichtenberg P,
Bar G, Javitt DC: High-dose glycine added to olanzapine
and risperidone for the treatment of schizophrenia. Biol
Psychiatry 2004; 55: 165-71.
46.
TB,
Leiderman E, Zylberman I, Zukin SR, Cooper
Javitt DC: Preliminary investigation of high-dose oral
finding trial of D-cycloserine added to neuroleptics for
glycine on serum levels and negative symptoms in
negative symptoms in schizophrenia. Am J Psychiatry
schizophrenia: an open label trial. Biol Psychiatry
1995; 152: 1213-5.
56.
Goff DC, Tsai G, Levitt J, et al.: A
1996; 39: 213-5.
placebo-controlled
trial of D-cycloserine added to
47.
Potkin SG, Jin Y, Bunney BG, Costa J, Gulasekaram
conventional neuroleptics in patients with schizophrenia.
B: Effect of clozapine and adjunctive high-dose glyArch Gen Psychiatry 1999; 56: 21-7.
cine in treatment-resistant schizophrenia. Am J
57.
Heresco-Levy U, Ermilov M, Shimoni J, Shapira
Psychiatry 1999; 156: 145-7.
B, Silipo G, Javit DC: Placebo-controlled trial of
48.
Evins AE, Fitzgerald SM, Wine L, Rosselli R,
D-cycloserine added to conventional neuroleptics,
Goff DC: Placebo-controlled trial of glycine added to cloolanzapine, or risperidone in schizophrenia. Am J
zapine in schizophrenia. Am J Psychiatry 2000; 157:
Psychiatry 2002; 159: 480-2.
826-8.
49.
Diaz P, Bhaskara S, Dursun SM, Deakin B:
Double-blind, placebo-controlled, crossover trial of
clozapine plus glycine in refractory schizophrenia negative
results. J Clin Psychopharmacol 2005; 25: 277-8.
50.
Tsai GE, Yang P, Chung LC, Tsai IC, Tsai CW,
Coyle JT: D-serine added to clozapine for the treatment of
schizophrenia. Am J Psychiatry 1999; 156: 1822-5.
51.
Kantrowitz, JT, Malhotra AK, Cornblatt B, et al.:
High dose D-serine in the treatment of schizophrenia.
Schizophr Res 2010; 121: 125-30.
52.
Simeon J, Fink M, Itil TM, Ponce D:
D-Cycloserine therapy of psychosis by symptom
provocation. Comp Psychiatry 1970; 11: 80-8.
53.
Cascella NG, Macciardi F, Cavallini C,
Smeraldi E: D-Cycloserine adjuvant therapy to
conventional neuroleptic treatment in schizophreania: an
open-label study. J Neural Transm-Gen Sect 1994; 95:
105-11.
54.
van Berckel BN, Evenblij CN, van Loon BJ, et
al.: D-cycloserine increases positive symptoms in chronic
schizophrenic patients when administered in addition to
antipsychotics: a double-blind, parallel, placebo-controlled
study. Neuropsychopharmacology 1999; 21: 203-10.
55.
Dose-
Goff DC, Tsai G, Manoach DS, Coyle JT:
58.
Duncan EJ, Szilagyi S, Schwartz MP, et al.:
Effects of D-cycloserine on negative symptoms in schizophrenia. Schizophr Res 2004; 71: 239-48.
59.
Heresco-Levy U, Javitt DC, Ermilov M, Silipo
G, Shimoni J: Double-blind, placebo-controlled, crossover
trial of D-cycloserine adjuvant therapy for treatment-resistant schizophrenia. Int J Neuropsychopharmacol
1998; 1: 131-5.
60.
Evins AE, Amico E, Posever TA, Toker R, Goff
DC: D-cycloserine added to risperidone in patients with
primary negative symptoms of schizophrenia. Schizophr
Res 2002; 56: 19-23.
61.
Goff DC, Tsai G, Manoach DS, Flood J, Darby
DG, Coyle JT: D-cycloserine added to clozapine for patients with schizophrenia. Am J Psychiatry 1996;
153: 1628-30.
62.
Lane HY, Huang CL, Wu PL, et al.: Glycine
transporter I inhibitor, N-methylglycine (sarcosine), added
to clozapine for the treatment of schizophrenia. Biol
Psychiatry 2006; 60: 645-9.
63.
Lane HY, Liu YC, Huang CL, et al.: Sarcosine
(N-methylglycine) treatment for acute schizophrenia: a
randomized, double-blind study. Biol Psychiatry 2008; 63:
9-12.