|
The Hippocampus
and Stress-related Disorders
XXIInd Congress of the Collegium Internationale Neuro-Psychopharmacologicum
Takuya Saito, MD, PhD
Multiple lines of studies indicate that stress is associated
with and often triggers psychiatric disorders. It is important
to understand how stress affects the brain in order to understand
psychiatric disorder.
Hormones and Experience Modulate Adult Neurogenesis
Recently, cortical neurons were found to be generated in
adult mammals.[1] However, it was more than 30
years ago that the dentate gyrus neurons in the hippocampus
were discovered to be generated. The dentate gyrus of the
hippocampal formation is formed during an extended period
that begins in gestation and continues well into the postnatal
period. In all mammalian species, the production of granule
neurons, the principal neuron type of the dentate gyrus, begins
during the embryonic period. Thereafter, the production of
new granule neurons tapers off, but it never ceases. The new
cells are incorporated into the granule cell layer and have
the nuclear morphology of mature granule neurons. It has been
shown that adult-generated cells in the dentate gyrus of rats
develop the same morphologic characteristics as the prenatally
generated granule cells.
They receive synaptic input, extend axons into the mossy
fiber pathway to the CA3A area in the hippocampus, and express
a number of markers of mature neurons. Until recently, these
findings were controversial, but in the past few years, evidence
has been growing to support the view that adult neurogenesis
in the dentate gyrus is a feature of all mammalian species.
Using 5-bromo-2'-deoxyuridine (BrdU) labeling combined with
cell type-specific markers, recent studies have shown that
several thousand new neurons are produced every day in the
dentate gyrus of adult rats. Gould and colleagues[2]
speculated that the high rate of regenesis of the neurons
implicates a high demand for the new neurons in the hippocampus,
and the late-generated cells play an important role in hippocampal
function. This is also because the neurons in the dentate
gyrus are unusually sensitive to experience-dependent structural
changes and are easily damaged. The generation and survival
of the dentate gyrus neurons in the hippocampus are regulated
by many factors, including endocrine, neural, and experiential
factors. The factors facilitating regeneration are: (1) ovarian
steroid hormones, (2) learning, (3) environment complexity,
and (4) running. The factors repressing the regeneration are:
(1) adrenal steroid hormones, (2) stress, and (3) deprivation.
Environmental complexity is especially important for survival
of the neurons, and often laboratory animals show low survival
rate of the neurons because of the lack of complexity in their
environment.
Throughout postnatal life, glucocorticoids appear to exert
suppressive effects on cell proliferation in the dentate gyrus.
Basal levels of adrenal steroids are negatively correlated
with the rate of cell proliferation in the dentate gyrus.
Treatment of adult rats with corticosterone also diminishes
the proliferation of granule cell precursors. The suppressive
effects of glucocorticoids on cell genesis suggest that stressful
experiences, which are known to elevate levels of circulating
glucocorticoids and stimulate hippocampal glutamate release
in adulthood, naturally inhibit cell proliferation in the
dentate gyrus. Previously, it was shown that an acute stressful
experience decreases the number of adult-generated neurons
produced in the adult dentate gyrus in a number of different
mammalian species. In adult rats, exposure to fox odor (predator
odor) elicits a stress response characterized by increased
adrenal steroid levels and suppresses the proliferation of
cells in the dentate gyrus. This suppression is reversed by
adrenalectomy (ADX). This reversed suppression was observed
only in the neurons in the dentate gyrus. In primates, social
stress affects the neuron genesis, and subordinate animals
produce fewer new hippocampus cells than dominant animals
do.
A recent study[3] has reported a gender difference
favoring females in the production of new cells in the dentate
gyrus of adult rats. Moreover, female rats exhibit naturally
occurring fluctuations in the number of new cells that are
produced across the estrogen cycle, with maximal cell proliferation
occurring during proestrus, the stage when estrogen levels
are highest. Pregnancy and aging are also factors affecting
the production of neurons in the female.
Although the exact functional significance of late-generated
neurons is not known, several lines of evidence suggest that
these new cells play an important role in learning. In a recent
study, it was demonstrated that training in a task that requires
hippocampal formation for acquisition results in an increase
in the number of adult-generated granule cells. Even though
thousands of neurons were generated every day, the majority
of adult-generated cells degenerated within a few weeks of
production; however, learning or complexity of environment
resulted in the rescue of a significant proportion of these
cells. Neurogenesis in the dentate gyrus seems to be associated
with learning and experience. There have been reports demonstrating
that stress has an impact on human memory. It is speculated
that survival and neuron genesis in the dentate gyrus may
be associated with stress-induced memory impairment.
Marked Alterations in Peptidergic Systems Induced by Convulsive
Stimuli
Stimuli in the hippocampus modulate the peptide system in
animal models. In a study by Vezzani and colleagues,[4]
electrical stimuli were used to produce stress. Classic kindling
stimuli (subconvulsive electrical stimulation 12 +/- 1 trains,
50 Hz, 2 msec pulses) applied via intrahippocampal electrodes
increased the expression of somatostatin, neuropeptide Y (NPY),
and neurokinin B (NKB) in GABA interneurons. In granule cells
and their axons, mossy fibers and elevated NPY and NKB concentrations
were observed 2 days after final stimulation. Further repeated
stimulation (27 +/- 2.5 trains) caused losses of 21% to 27%
of deep hilar neurons. Rapid kindling (subconvulsive electrical
stimulation, 48 trains within 1 week, 50 Hz, 1 msec pulses,
10 s/train), yielded loss of hilar neurons and some CA3 pyramidal
cells near the electrode and resulted in a reduction of somatostatin/NPY
co-localizing interneurons in the deep hilus. Thus, a loss
of GABAergic transmission may be involved in the increased
seizure susceptibility of these rats. One week after classical
and 1 month after rapid kindling, peptide levels in interneurons
were still high, while NKB and NPY in mossy fibers were markedly
reduced when compared with levels measured 2 days after treatment.
These data suggest that the mechanism of increased expression
of NKB and NPY in granule cells is different from that in
interneurons. NPY expression can be induced by group I metabotropic
receptors, but not by ionotropic glutamate receptor agonists.
Status epilepticus induced by kainic acid (10 mg/kg, intraperitoneally)
resulted in moderate to severe loss of hilar and pyramidal
neurons and recurrent, spontaneous seizures. In relation to
neuronal loss, the number of somatostatin/NPY interneurons
in the deep hilus was reduced, while NPY, NKB, and cholecystokinin-octapeptide
(CCK)-containing basket cells survived. Electrically stimulated
rats with moderate neurodegeneration (50% hilar neurons, 30%
dorsal CA3, 40% ventral CA1) exhibited minor levels of NPY
and NKB; rats with severe neurodegeneration (80% to 90% hilar
neurons, 50% to 60% CA3, 40% to 60% CA1) showed chronically
high levels of NPY and NKB in granule cells. In the ventral
hippocampus, sprouting of mossy fibers to the inner molecular
layer was observed in severe cases. The lasting changes in
the pattern of various peptides in the hippocampus may reflect
functional modifications in the corresponding peptide-containing
neurons. These changes may be involved in chronic epileptogenesis,
which evolves in response to limbic seizures.
Kainic acid-treated rats exposed even more neuronal loss
in either CA1 or CA3. Such rats showed mossy fibers sprouting
in the entire hippocampus. Both phenomena, lasting expression
of NKB/NPY and sprouting of mossy fibers, appear to be related
to convulsive stimuli and marked neuronal loss. Together with
the increased expression of NPY, changes in affinity and number
of receptors were evident. While the number of presumably
proconvulsive NPY-Y1 receptors was reduced, the
transmission via presumably anticonvulsive Y2 receptors
was facilitated in a biphasic manner. Soon after injection
of kainic acid (4-12 hours), the affinity of Y2
receptors in the hippocampus proper was increased 2-fold.
Subsequently, Y2 receptors were established in
mossy fibers, involving group I metabotropic glutamate receptors.
Electrophysiologic and glutamate release studies suggest that
these changes may contribute to reduced excitatory transmission,
thus representing an endogenous anticonvulsant mechanism.
FK506: A Novel Cytoprotective Agent in the Hippocampus
A recent study by Sharkey and Butcher[5] has reviewed
the action of FK506 (tacrolimus), a powerful immunosuppressant
whose mechanism of action involves inhibition of calcineurin
in T-lymphocytes by a complex of FK506 and FK506-binding protein
12. FK506 is used for prevention of allograft rejection and
treatment of dermatitis. A phase 3 trial is currently being
conducted on FK506 for the treatment of rheumatoid arthritis.
In animal studies, FK506 shows anti-inflammatory effect, inhibits
nitric oxide synthase and apoptosis, and stabilizes mitochondria.
Sharkey reported on an in vivo model that demonstrated that
FK506 is a powerful neuroprotective agent of focal cerebral
ischemia when administered up to 2 hours postocclusion and induced
by endothelin-1. FK506 may be of use as a cytoprotective agent
for the central nervous system. The precise mechanism of action
remains uncertain. FK506 prevents oxidative stress from MTPT,
3-NP, and methylenedioxymethamphetamine [MDMA]. FK506 is protective
for serotonergic neurons in the hippocampus by MDMA; mitochondrial
stabilization is considered to be the mechanism of action. His
studies have shown that FK506 is a potentially useful neuroprotective
agent.
Corticosterone and Cytokines in the Hippocampus: Neurotoxicity
vs Neuroprotection
It has been proposed by Kato and colleagues[6] that
post-traumatic stress disorder (PTSD) is a good model to look
at in order to understand the effect of stress on the hippocampus
in humans. He found that the hippocampus in breast cancer patients
with PTSD was smaller on MRI than the hippocampus of breast
cancer patients without PTSD. This finding is consistent with
pyramidal cell atrophy of the CA3 region in the hippocampus
after stress. The effects of corticosterone on the hippocampus
were also studied using trimethyltin (TMT) as a specific neurotoxicant
known to induce a transient increase of plasma corticosterone.
TMT causes widespread behavioral changes in rodents, including
aggression, hyperirritability, seizures, and learning impairment.
Histologically, TMT produces delayed and selective neuronal
damage in the hippocampus, with the pyramidal cells in CA3 predominantly
being affected in conjunction with reactive gliosis and mossy
fiber sprouting. Increases in NPY mRNA and its receptor NPY-Y2
in the dentate gyrus have been noted. Pretreatment with interleukin-1
receptor antagonists partially abolishes the elevation of plasma
corticosterone concentration and decreases cell damage.
TMT was administered to mice with and without adrenalectomy.
Mice with adrenalectomy had the most severe cell damage, and
when corticosterone was replaced to baseline level, the degree
of cell damage was reduced.
Low and steady levels of corticosterone seem to act as a
protective factor. The present study on TMT models may enlighten
us regarding the molecular cascade in stress-induced hippocampal
damage in terms of immuno-endocrine interactions and help
us understand the effects of corticosteroids.
Steroid Hormone Receptor Dynamics in the Hippocampus:
In Vivo and In Vitro Study
The subcellular distribution of the steroid hormone receptors
has not yet been established, and it remains a subject of
controversy. To elucidate the localization of receptors of
adrenal steroids and estrogen, the investigators carried out
an immunocytochemical analysis of specific antibodies and
compared it with imaging analysis using the green fluorescent
protein (GFP) chimera system.[7] By using both
methods, it was found that in the absence of the ligand glucocorticoid
receptor (GR), cytoplasmic and mineralocorticoid receptors
(MR) are both found in the cytoplasm and in the nucleus, whereas
estrogen receptors (alpha and beta) are found in the nucleus.
Dexamethasone was administered, and GR and MR receptors were
accumulated in the nucleus. Simultaneous imaging analysis
of GR and MR translocation showed that in the COS cells, which
are devoid of endogenous steroid receptors, the translocation
of GR and MR into the nucleus differed at the lower concentration
of corticosterone, whereas in the hippocampal cells there
was no significant difference in the translocation rate. Microtubules,
mitochondria, and synaptic vesicles are believed to play important
roles. However, colchicine treatment, which disrupted the
microtubules, did not cause any changes in dexamethasone-induced
GR and MR translocation to the nucleus. The ligand-induced
movement of these nuclear receptors from inactive to active
states is dependent on the type of receptor.
References
- Gould E, Tanapat P. Stress and hippocampal neurogenesis.
Biol Psychiatry. 1999;46:1472-1479.
- Gould E, Tanapat P, Rydel TA, Hastings NB. Hormones and
experience modulate adult neurogenesis. Int J Neuropsychopharmacol.
2000;3(suppl 1):S42. Abstract S.27.1.
- Scharfman HE. Epileptogenesis in the parahippocampal region.
Parallels with the dentate gyrus. Ann N Y Acad Sci. 2000;911:305-327.
- Vezzani A, Schwarzer C, Lothman EW, Williamson J, Sperk
G. Functional changes in somatostatin and neuropeptide Y
containing neurons in the rat hippocampus in chronic models
of limbic seizures. Epilepsy Res. 1996 26:267-279.
- Sharkey J, Butcher SP. Immunophilins mediate the neuroprotective
effects of FK506 in focal cerebral ischaemia. Nature. 1994;371:336-339.
- Kato N, Nishimura T, Imai H, Sadamatsu M, Liu Y. Corticosterone
and cytokines in the hippocampus: neurotoxicity vs. neuroprotection.
Int J Neuropsychopharmacol. 2000;3(suppl 1):S43. Abstract
S.27.4.
- Kawata M, Yuri K, Ozawa H, et al. Steroid hormones and
their receptors in the brain. J Steroid Biochem Mol Biol.
1998;65:273-280.
|
Volver Atrás