This site uses cookies.
Some of these cookies are essential to the operation of the site,
while others help to improve your experience by providing insights into how the site is being used.
For more information, please see the ProZ.com privacy policy.
Indépendant et donneur d'ordre, Utilisateur confirmé du site
Data security
This person has a SecurePRO™ card. Because this person is not a ProZ.com Plus subscriber, to view his or her SecurePRO™ card you must be a ProZ.com Business member or Plus subscriber.
Affiliations
This person is not affiliated with any business or Blue Board record at ProZ.com.
Rates per language pair: français vers anglais - Tarif : 0.11 - 0.12 EUR par mot / 28 - 30 EUR de l'heure anglais vers français - Tarif : 0.11 - 0.12 EUR par mot / 28 - 30 EUR de l'heure espagnol vers anglais - Tarif : 0.11 - 0.12 EUR par mot / 28 - 30 EUR de l'heure
All accepted currencies
Euro (eur)
Entrées de Blueboard effectuées par cet utilisateur
anglais vers français: DBI Induces Neurogenesis via GABAA Receptors Cell Stem Cell 10, 76–87, January 6, 2012 General field: Médecine Detailed field: Médecine : médicaments
Texte source - anglais Diazepam Binding Inhibitor Promotes
Progenitor Proliferation in the Postnatal SVZ
by Reducing GABA Signaling
SUMMARY
The subventricular zone (SVZ) of the lateral ventricles
is the largest neurogenic niche of the postnatal brain.
New SVZ-generated neurons migrate via the rostral
migratory stream to the olfactory bulb (OB) where
they functionally integrate into preexisting neuronal
circuits. Nonsynaptic GABA signaling was previously
shown to inhibit SVZ-derived neurogenesis. Here we
identify the endogenous protein diazepam binding
inhibitor (DBI) as a positive modulator of SVZ postnatal
neurogenesis by regulating GABA activity in
transit-amplifying cells. We performed DBI lossand
gain-of-function experiments in vivo at the peak
of postnatal OB neuron generation in mice and
demonstrate that DBI enhances proliferation by
preventing SVZ progenitors to exit the cell cycle.
Furthermore, we provide evidence that DBI exerts
its effect on SVZ progenitors via its octadecaneuropeptide
proteolytic product (ODN) by inhibiting
GABA-induced currents. Together our data reveal
a regulatory mechanism by which DBI counteracts
the inhibitory effect of nonsynaptic GABA signaling
on subventricular neuronal proliferation.
INTRODUCTION
The majority of neurons are born during embryonic development
in restricted progenitor zones and neuroblasts migrate along
specific pathways to reach their final destination in the brain.
There is ample evidence, however, that neurogenesis, migration,
and subsequent integration of new neurons into preexisting
networks persists postnatally in several brain structures. In
rodents there are three neurogenic zones in the early postnatal
period: the subventricular zone (SVZ) of the lateral ventricles,
the subgranular zone of the hippocampal dentate gyrus (DG),
and the cerebellum (Hatten, 1999). In the SVZ and DG, neurogenesis
continues throughout adulthood, albeit at a lower level
(Abrous et al., 2005). The largest fraction of SVZ-derived neurons
migrate anteriorly via the rostral migratory stream (RMS) to their
destination, the olfactory bulb (OB), where they mature into local
GABAergic interneurons (Lois and Alvarez-Buylla, 1994; Luskin,
1993) and, to a minor extent, into glutamatergic neurons (Brill
et al., 2009). In young animals, a small fraction of neuroblasts
from the SVZ migrate also to other cortical areas (Bandeira
et al., 2009; Inta et al., 2008). Postnatal neurogenesis contributes
to structural plasticity in mature cortical circuits and affects OBand
hippocampus-dependent learning and memory (Deng et al.,
2010; Lledo et al., 2008).
The SVZ comprises different cell types. Astrocyte-like stem
cells give rise to transit-amplifying cells (fast proliferating progenitors)
that in turn give rise to immature neurons (neuroblasts)
(Doetsch et al., 1999). Newly generated neurons migrate
in chains in the RMS that is ensheathed by processes of
astrocyte-like cells (Lois and Alvarez-Buylla, 1994). Postnatal
progenitors have specific intrinsic properties that in conjunction
with distinct environmental cues govern their proliferation,
migration, and differentiation into defined neuronal subtypes.
Over the last years, a number of molecular signals involved in
postnatal neurogenesis have been identified (Abrous et al.,
2005; Cayre et al., 2009). Of these, neurotransmitters received
special attention (Platel et al., 2010). For instance, nonsynaptic
GABA signaling via GABAA receptors was shown to regulate
in vitro and in vivo neural stem cell proliferation in the SVZ
(Fernando et al., 2011; Liu et al., 2005). However, little is known
about how GABA signaling is modulated by intrinsic or extrinsic
factors that might differentially affect distinct cell types in the
neurogenic niche.
GABAA receptors are heteropentameric ion channels that are
prevalently but not exclusively expressed at the cell surface on
multiple cell types of the CNS (D’Hulst et al., 2009). The activity
of GABAA receptors can be altered by a plethora of intrinsic
modulators including neurosteroids, endozepines, and zinc
(Bormann, 1991; Lambert et al., 2003; Smart et al., 2004). We
found it striking that the endozepine diazepam binding inhibitor
(DBI) is highly expressed in the SVZ and RMS (http://mouse.
brain-map.org/experiment/show/68632649). DBI is a small cytosolic
protein (10 kDa) that is secreted to the extracellular space
via a nonconventional secretory pathway (Abrahamsen and
Stenmark, 2010) and binds to both central and peripheral-type
benzodiazepine receptors (CBRs and PBRs, respectively) (Costa
and Guidotti, 1991). The cloning and molecular characterization
of GABAA receptor subunits revealed that the CBR is in fact part
of the receptor. The CBR is identical to the benzodiazepine
binding site of the GABAA receptor and is located extracellularly between the a and g2 subunits (D’Hulst et al., 2009). The PBR
complex, on the other hand, is localized mainly in the mitochondrial
membrane in cells of the CNS and peripheral organs.
Expression in the brain is restricted to glial and ependymal cells
(Bormann, 1988; Papadopoulos et al., 2006). The full-length DBI
is the precursor of octadecaneuropeptide (ODN) and triakontatetraneuropeptide
(TTN), which also have a high affinity for diazepam
receptors (Slobodyansky et al., 1989).
In this study we investigated the function of the endozepines in
the SVZ-RMS in vivo at the peak time of OB-neuron generation
(Bayer, 1983; Hinds, 1968) and identified a mechanism by which
they modify GABA signaling thereby affecting postnatal subventricular neurogenesis.
RESULTS
DBI Expression in the SVZ/RMS
Expression of DBI and of its processing product ODN was found
in several brain areas and was reported in previous studies (Alho
et al., 1988; Tong et al., 1991; Tonon et al., 1990; Yanase et al.,
2002). Accordingly, we detected DBI expression in glial cells
from cortical areas and ependymal cells lining the ventricle but
not in mature neurons (not shown). Focusing on the SVZ and
RMS, we detected DBI in astrocyte-like GFAP/Nestin stem cells
and in Mash1 fast dividing progenitors, but not in DCX
migrating neuroblasts in young mice (Figures 1A–1D). It thus
appears that DBI is expressed at early stages of neurogenesis,
subsiding already at the neuroblast stage. A similar expression
pattern for DBI was found in the adult brain (Figure S1 available
online). Expression levels slightly decreased over time but the
cell type-specific expression pattern of DBI was not altered
during postnatal development. Although the majority of
Mash1 cells were positive for DBI during the first postnatal
week (67% ± 1.2%, mean value ± SEM, n = 4 mice), we detected
DBI in a smaller subset of Mash1 cells in adult animals (32% ±
2.2%, mean value ± SEM, n = 6 mice).
In Vivo Knockdown of DBI in the SVZ Reduces Neuronal
Proliferation
To investigate the functional role of DBI in the SVZ, we performed
in vivo knockdown experiments. We used two shRNAs (short
hairpin RNA) sequences (a and b) for which the efficiency of
DBI silencing was first confirmed by western blot analysis of
transfected cell cultures (Figures S2A and S2B). To specifically
target newborn neurons, we performed retroviral injections in
the SVZ and established that DBI knockdown silenced DBI
expression in vivo (Figures S2C and S2D). The nature of the
retrovirus guarantees the exclusive infection of dividing cells
(mainly fast dividing progenitors) that give rise to neurons
(Ming and Song, 2005). Infected cells expressing knockdown
shRNA were positive for the Tomato protein whereas control
cells expressing scrambled shRNA were EGFP positive. To
control for differences resulting from individual variability and
differences in the injection site, P4-old mice were injected with
a mix of control (green) and DBI-knockdown (red) viruses (Figure 2A). We determined the total number of infected cells at 2, 4, and 7 days postinjection (dpi) and found that the ratio of red/green cells decreased over time (Figures 2A and 2B). Two
scenarios could account for these results: DBI knockdown resulted in either increased cell death or reduced cell proliferation.
The former could be excluded because active caspase-3
stainings revealed no difference in the number of apoptotic cells
between the DBI-knockdown and control population when
measured at 2 and 7 dpi (Figures S2E and S2F).
To confirm that DBI knockdown impaired proliferation in the
SVZ niche, we injected the control and DBI-knockdown viral
mix into the SVZ of P4-old mice followed by intraperitoneal injections of the thymidine analog BrdU, a marker for dividing cells.
Twelve days postinjection we determined the number of BrdUpositive neurons in the OB in DBI knockdown and control neurons (Figure 2C). The percentage of BrdU cells was significantly
lower in the DBI-knockdown neuronal population
compared to controls, indicating that DBI expression in dividing
cells promotes proliferation. To further verify the specificity of the
DBI knockdown effect, we rescued DBI expression in shRNADBI-
infected cells with a retroviral vector containing the DBI
coding sequence linked to a fluorescent marker via a T2A selfcleaving
peptide (Figure S3A; Szymczak et al., 2004). Given
that the shRNA-DBIb sequence matches the 30UTR in DBI
mRNA, the exogenous DBI coding sequence expressed by the
virus was not a target for silencing (Figure S3B). Importantly,
neuronal proliferation measured by BrdU uptake was significantly
higher in double-infected cells coexpressing the knockdown
and rescue construct compared to knockdown only
(Figures S3C–S3E), indicating that a restoration of DBI levels
sufficed to rescue impaired proliferation.
Quantitative evaluation of DBI knockdown at 7 dpi revealed
differences along the SVZ-RMS-OB axis. Thus, the proportion
of cells that had reached the OB was higher in DBI-knockdown
cells compared to control (Figures 2D and 2E). Because the
majority of retrovirally infected cells in the SVZ are transitamplifying
cells and neuroblasts (Rogelius et al., 2005) and DBI
was not detected in neuroblasts, it is unlikely that the higher
proportion of DBI-knocked down cells in the OB was the result
of faster neuroblast migration. Indeed, live imaging experiments
indicated that neuroblasts infected with control and shRNA-DBI
virus migrated at comparable speed (Figure S4 and Movie S1).
Alternatively, knockdown of DBI in neuronal progenitors may
reduce the time they spend in proliferation by promoting faster
differentiation to migrating neuroblast. Thus, DBI knockdown
cells would be detected in the OB before the controls. Such
a scenario is supported by the finding that 3 days postinjection,
the proportion of Mash1 progenitor cells was significantly lower
upon DBI knockdown compared to control (Figures 2F and 2G).
DBI In Vivo Overexpression Promotes Proliferation
by Expanding Mash1 Population
To further substantiate that DBI affects proliferation, we
performed gain-of-function experiments in vivo. To compare
overexpressing with control-infected cells, we mixed the DBIoverexpressing
virus with the retrovirus coding only for EGFP
(Figure 3A). We injected the viral mix into the SVZ of P4-old
mice and analyzed the ratio of red/green cells at three time points
(2, 4, and 7 dpi). The results showed that the proportion of red
cells increased over time, suggesting that DBI-overexpressing
cells proliferate more than control cells (Figure 3B). Thus, DBI
gain- and loss-of-function experiments had opposite effects.
To directly test that DBI promotes neuronal proliferation, we
analyzed the number of BrdU-containing cells in the virusinfected
population. As shown in Figure 3C, the percentage of
BrdU neurons was higher upon DBI overexpression, indicating
that DBI expression stimulates the proliferation of neural SVZ
progenitor cells.
We next analyzed the distribution of infected cells in the
RMS/OB at 7 dpi and found that the ratio of red/green cells
was significantly higher in the RMS than in the OB (Figures 3D
and 3E). The possibility that neuroblasts overexpressing DBI
might die before reaching the OB was ruled out because the
number of apoptotic cells was comparable to controls at 4 dpi
(1.08% ± 0.27% and 1.11% ± 0.43% activated caspase-3-
positive cells, mean ± SEM from 2 animals, 667 and 421 control
and DBI-overexpression, respectively). Instead, a likely explanation
for this result is that DBI-overexpressing progenitors remain
in division for a longer time in comparison to controls, and hence
more time is required for these cells to differentiate to neuroblasts
and to eventually reach the OB. Such a scenario is in
line with our findings that the number of Mash1-positive cells
was augmented after DBI overexpression (Figures 3F and 3G).
In summary, overexpressing and knockdown experiments
support the conclusion that DBI is a positive regulator of subventricular
neurogenesis.
DBI Knockdown Induces Cell Cycle Exit without
Affecting Cell Cycle Length in Dividing SVZ Progenitors
We subsequently investigated whether the effects of DBI on
proliferation resulted from an alteration of the cell number reentering
the cell cycle and/or the cell cycle length. We carried out
in vivo experiments by using two analogs of BrdU (IdU and
CldU) and measured the cell fraction remaining in division after
DBI knockdown. We first injected IdU to label the virus-infected
cells in S phase, and 12–20 hr later we performed multiple CldU
injections (every 2 hr) to label cells that divided again in the next
cycle. Thus, double-labeled cells reentered a new cell cycle
continuing to divide, whereas cells labeled only with the first
marker exited the cell cycle and stopped dividing (Figure 4A).
The results showed that among actively cycling cells, fewer
DBI-knockdown cells were double labeled with IdU/CldU in
comparison to control cells, indicating that DBI silencing induces
cell cycle exit (Figures 4B and 4C). This result was confirmed by
an alternative approach, where virus-infected cells were labeled
with BrdU, and 22 hr later the brains were fixed and stained for
the proliferation marker Ki67 (Figure S5). Thus, after DBI knockdown,
the proportion of cells reentering the cell cycle was
reduced by about 40%.
To investigate whether DBI expression might affect also the
cell cycle length, we used a short-interval IdU/CldU double
labeling approach (Figure 4D; Fukumitsu et al., 2006; Martynoga
et al., 2005; Quinn et al., 2007; Teissier et al., 2011) that allows
the estimation of the S phase (Ts) and the total cell cycle (Tc)
length (see Experimental Procedures). The Ts (4.78 ± 0.28 and
4.83 ± 0.24 hr, mean values ± SEM for control and DBI-knockdown,
respectively) and Tc (15.59 ± 0.79 and 15.36 ± 1.07 hr,
mean values ± SEM for control and DBI-knockdown, respectively)
values for the two groups were similar (Figures 4E
and 4F). In summary, DBI-knockdown does not affect the duration
of the cell cycle in proliferating cells but dramatically reduces
the proportion of cells reentering the cell cycle.
DBI Acts via Central Benzodiazepine Receptors
The enhanced postnatal neuronal proliferation induced by DBI
raised the question whether this effect was mediated via the
central or peripheral benzodiazepine receptor. Hence, we employed
SVZ primary cultures in which neural progenitors from
postnatal brain were allowed to proliferate and form neurospheres.
We expressed either DBI-RFP or only a control gene
(RFP) and analyzed the number of dividing (BrdU ) cells within
primary neurospheres (Figure 5A). In agreement with the in vivo
experiments, DBI enhanced cell proliferation in SVZ-derived
neurospheres (Figure 5B). We next evaluated the effect of the
CBR or PBR antagonists flumazenil and PK-11195, respectively,
and found that DBI-induced proliferation could be blocked by flumazenil but not by PK-11195, suggesting
that DBI acts mainly via the CBR,
hence the GABAA receptor (Figure 5B).
The presence of both GABA and GABAA
receptors in SVZ-neurospheres was previously
established (Fernando et al.,
2011). Interestingly, treatment of RFPinfected
neurospheres with flumazenil
reduced the number of BrdU cells in
comparison to RFP untreated cells,
whereas DBI-infected cells treated with
flumazenil showed no difference to RFP
untreated cells. These results suggest that the effect of
endogenous DBI might be blocked by flumazenil and that this
reduction can be reversed with the addition of exogenous DBI. ODN Replicates the DBI Effects In Vivo
Earlier studies showed that the DBI proteolytic fragments ODN
and TTN prefer different binding sites. Whereas ODN has
a high affinity for the benzodiazepine site of GABAA receptors
(sensitive to flumazenil), TTN preferentially binds to PBRs and
can displace PK-11195, but not flumazenil (Slobodyansky
et al., 1989). Our pharmacological results suggest that the short
peptide ODN might be the mediator of the above-described DBI
effect on neuroblast proliferation. Therefore, we tested whether
ODN has an effect on SVZ neurogenesis. To this end we generated
an ODN-overexpression retrovirus that expressed also RFP
(Figures 5C and 5D). Mice were injected into the SVZ with a mix
of control (green) and ODN-overexpressing virus (red), and we
analyzed in vivo proliferation of SVZ-generated neurons by performing
BrdU labeling experiments. Overexpression of ODN
induced neuronal proliferation as reflected by the higher number
of BrdU neurons in the OB in comparison to control (Figure 5E).
Furthermore, we examined the cell distribution of red and green
cells along the SVZ-RMS-OB axis at 7 dpi. This analysis revealed
that ODN-overexpressing cells required longer time to reach the
OB than control-infected cells (Figure 5F). In conclusion, ODN
replicated the results obtained when overexpressing full-lengthDBI, suggesting that ODN mediates DBI effects on subventricular
progenitors.
ODN Inhibits GABA-Induced Currents in Transit-
Amplifying Cells
Depending on the receptor subunit composition, endozepines
can act as positive or negative modulators of GABAA receptors
(reviewed in D’Hulst et al., 2009). Whereas the presence of
GABAA receptors was demonstrated in stem-like cells and
neuroblasts (Liu et al., 2005; Wang et al., 2003), there is no information
regarding their expression in transit-amplifying cells.
Therefore, we investigated first whether fast dividing progenitors
have functional GABAA receptors, and second, whether ODN
can modulate GABA-induced responses. We injected a retrovirus
expressing RFP into the SVZ of transgenic mice in which
migrating neuroblasts express EGFP, using the 5-HT3A-EGFP
mouse line described before (Figure 6A; Inta et al., 2008). Fast
dividing progenitors were identified based on the presence
of red fluorescence (newborn cells infected with retrovirus)
and the absence of green signal (neuroblast) as well as cell
morphology and location. Indeed, immunostainings on brain
sections from these experiments confirmed that virtually every
virally infected cell negative for EGFP in the SVZ expressed
Mash1 (Figure 6B). There were other infected cells also negative
for EGFP that did not express Mash1; however, these were glial
cells close to the injection site and clearly distinguishable from
transit-amplifying cells based on their morphology (not shown).
Two to five days after injection, we selected fast dividing
progenitors based on the above-mentioned criteria for electrophysiological
recordings. We recorded from these cells via
the whole-cell patch-clamp configuration (Figure 6C). EGFP/
RFP cells were hyperpolarized at rest (Vrest = 86.8 ± 5.9 mV,
n = 7). Their input resistance was 1.79 ± 0.46 GU (Figure 6D),
distinct from the input resistance of stem cells and ependymal
cells reported previously (30–40 MU) and from the input resistance
of neuroblasts in the SVZ and RMS (4 GU) (Lacar
et al., 2010). GABAA receptor-mediated currents in transitamplifying
cells were studied in voltage-clamped nucleated
outside-out patches. Brief applications of GABA at a nondesensitizing
concentration (50 mM) via a piezo-driven two-barrelled
fast application system elicited inward currents with an amplitude
of 76.1 ± 20.6 pA. In the presence of ODN (20 mM),
GABAAR-mediated currents were significantly decreased, and
after washing out ODN, currents recovered to nearly control
values (Figures 6E and 6F). These results demonstrate that
ODN inhibits GABAA receptors in transit-amplifying cells. A
scheme showing the proposed link between DBI-modulated
GABA signaling and progenitor proliferation is presented in
Figure 7.
DISCUSSION
In this study we have identified DBI as a key modulator of postnatal
neurogenesis in the largest proliferative area of the rodent
brain, the SVZ. We provide evidence that this protein acts in vivo
as a positive regulator for neuronal proliferation by reducing
GABA signaling. Ambient GABA in the SVZ niche was shown
to derive from neuroblasts and it was proposed that it provides
a feedback mechanism acting as a paracrine stop signal for
neuronal proliferation (Nguyen et al., 2003; Wang et al., 2003).
We show here that DBI produced by neuronal progenitors counterbalances
this effect by reducing GABA activity. Furthermore,
our results indicate that the effect of DBI on GABAA receptors
is exerted via its processing peptide ODN.
Immunostainings on sagittal brain sections containing the
RMS showed that DBI is synthesized in stem cells and fast
dividing progenitors but not in neuroblasts or mature neurons.
This expression pattern is consistent with the role we propose
for DBI on neuronal generation. Neural stem cells constitute
a small proportion of subventricular cells (0.4%) that have
a very slow turnover (few weeks) in contrast to transit-amplifying
cells, whose cell cycle lasts only few hours (Morshead et al.,
1998; Morshead and van der Kooy, 1992). It is therefore reasonable
to assume that the majority of retrovirus-infected cells
correspond to the fast dividing cell type. There was a significant difference (40%) of BrdU-labeled cells
between control and DBI-knockdown or
overexpression cells, indicating that a
large proportion of cells were affected
by the virus expression. Hence, our results
point toward an effect on fast proliferating
cells. In addition, possible effects on DBI manipulation in
stem cells would not be observed after such short time periods
that were used in this study (i.e., 3 to 12 days). However, it is likely
that DBI/ODN acts also on neural stem cells because they
express both DBI and GABAA receptors. Although neuroblasts
do not express DBI, they contain functional GABAA receptors
whose activity can also be inhibited by ODN (data not shown).
Given that stem cells, transit-amplifying cells, and astrocytes
are in close proximity to neuroblasts in the RMS and can secrete
DBI/ODN to the extracellular space, it is possible that paracrine
ODN affects migrating cells as well.
The cell cycle experiments demonstrated that DBI keeps
progenitor cells in division preventing cell cycle exit. Symmetric
division of transit-amplifying cells is a key determinant in regulating
the pool size of neural progenitors. After several rounds
of symmetric divisions, progenitors undergo asymmetric division
and eventually develop into neuroblasts (Takahashi et al., 1996).
A plethora of extracellular and intracellular factors have been
identified that regulate the cell cycle during development, tipping
the balance from proliferation to differentiation and thus ultimately
controlling the production of neurons (Caviness et al.,
2009; Mitsuhashi and Takahashi, 2009). In contrast, the literature
pertaining to factors modulating cell cycle kinetics during
postnatal neurogenesis is scarce (Doetsch et al., 2002; Zhang
et al., 2008). Overexpression and knockdown experiments provided the first indirect evidence that DBI might regulate proliferation
by affecting the mode of cell division. Thus, low levels of
DBI in progenitor cells favored the differentiation to migrating
neuroblasts, whereas high DBI expression expanded the population
of proliferating progenitors. Altered cell division would also
explain the early arrival to the OB of DBI-knockdown cells as
a result of reduced proliferation, and conversely delayed arrival
of DBI-overexpressing cells resulting from enhanced proliferation.
More direct evidence, however, and the mechanism by
which DBI affects proliferation is provided by our in vivo cell
cycle experiments, demonstrating that DBI favors progenitor
cell division without affecting cell cycle length.
Previous in vitro studies investigated the involvement of GABA
in neuronal proliferation and showed that it negatively regulates
cell division in embryonic and postnatal progenitors. Thus,
studies on embryonic neocortical explants revealed that GABA
decreases the proliferation of cortical progenitors (LoTurco
et al., 1995) and SVZ cells (Haydar et al., 2000). Furthermore,
in organotypic slices from neonatal rats, activation of GABAA
receptors inhibits cell cycle progression of neuroblasts (Nguyen
et al., 2003). Finally, in organotypic slices from adult mice, GABA
exerts a tonic inhibitory control on the proliferation of SVZ astrocyte-
like cells (Liu et al., 2005). Interestingly, Tozuka and colleagues
described that GABA promotes differentiation of hippocampal
progenitor cells (Tozuka et al., 2005), raising the question
whether similar GABA-mediated mechanisms might regulate
proliferation in the two postnatal neurogenic niches. This is of
note, given that SVZ-derived neurons develop an inhibitory
phenotype while those in the hippocampus become excitatory
neurons. Here we demonstrate that proliferative progenitor cells
in the SVZ express GABAA receptors. Furthermore, we show
in vivo that GABA signaling controls proliferation of transitamplifying
cells. There is recent in vivo evidence based on pharmacological
approaches that GABA signaling also controls
proliferation of neural stem cells (Fernando et al., 2011). Here
we used a virus-mediated approach that allowed us to affect
GABAA receptor-mediated activity selectively in progenitors,
thereby circumventing indirect effects that are obtained upon
systemic pharmacological intervention. Most importantly, we
identify a mechanism and show that GABA signaling is modulated
by endozepines in the postnatal SVZ. Specifically, we
demonstrate that the DBI-proteolytic product ODN inhibits
GABAA receptor currents recorded from transit-amplifying cells
in acute brain slices. The electrophysiological results most probably
account for the observed effect on increased proliferation
in vivo upon ODN delivery onto neural progenitors.
The downstream signaling mechanism by which GABA
reduces proliferation of transit-amplifying cells remains to be
elucidated. The GABA effect on proliferation in other cell types
has not been deciphered entirely yet, either. It was proposed
that reduced proliferation of embryonic progenitors and striatal
neuroblasts may be induced by intracellular calcium release
triggered by GABAA receptor activation (LoTurco et al., 1995;
Nguyen et al., 2003). In a recent study, downstream GABA
signaling in stem cells has been linked to histone H2AX phosphorylation
that limited proliferation (Fernando et al., 2011).
However, further investigations are warranted to identify the
intracellular pathways of GABAA receptor signaling in the distinct
cell types of the neurogenic niche.
A role for DBI on cellular proliferation has been proposed for
other cell types in the central nervous system both under normal
and pathological conditions. In vitro, TTN and ODN were shown
to enhance cellular division of cultured rat astrocytes (Gandolfo
et al., 1999, 2000). Interestingly, Alho and colleagues found a
significant increase in DBI expression in different human brain
tumors such as astrocytomas, glioblastomas, and medulloblastomas
(Alho et al., 1995; Miettinen et al., 1995), suggesting that
endozepines might be involved in the highly proliferative rate of
neoplastic cells. The same authors reported that astrocytic
tumors express high levels of PBRs (Miettinen et al., 1995) and
also that GABAA receptors were detected in human gliomas
and medulloblastoma cell lines (Codina et al., 2000; Labrakakis
et al., 1998). Hence, both scenarios regarding DBI mechanism
of action through PBRs and CBRs are possible for a putative
role of endozepines on brain tumor proliferation.
The presence of GABA receptors on neuronal progenitors and
the here-described modulation by endozepines imply that exogenous
ligands may also be effective in altering proliferation.
Indeed, recent in vivo studies on neonatal brain development reported
that the GABAA receptor agonists diazepam and phenobarbital
suppress cell proliferation in several cortical areas and
neurogenesis in the dentate gyrus (Stefovska et al., 2008). Drugs
acting as GABAA receptor modulators are frequently administered
as tranquilizers, sedatives, or anticonvulsants in pediatric
medicine. However, different response pathways may be activated
in distinct cell types upon systemic delivery of therapeutic
compounds acting on GABAA receptors. Thus, from a clinical
point of view, it is important to understand the multiple effects
of these drugs on postnatal development in order to design a
rational therapeutic treatment. In addition, manipulation of neural
cell proliferation as is desirable in conditions requiring brain
repair or treatment of tumors demands better knowledge of
the cell type-specific action of the employed drugs.
EXPERIMENTAL PROCEDURES
Animals
We used wild-type C57BL/6 mice (Charles River) and the transgenic
5-HT3A-EGFP mouse line described before (Inta et al., 2008). All animal procedures
were according to the Heidelberg University Animal Care Committee
regulations.
Immunostainings
Young pups were killed by decapitation and their brain was removed and fixed
overnight in 4% PFA. 70 mmsections were cut in a vibratome (Leica VT1000S).
Free-floating sections were permeabilized in 0.2% Triton-PBS for 30 min and
blocked in3% BSA-PBS for at least 30 min before incubation with the first antibody
in 3% BSA-PBS overnight at 4C. After three washes in PBS, sections
were incubated with the secondary antibody in 3% BSA-PBS for 1–2 hr,
washed again in PBS, and mounted. For BrdU/IdU/CldU stainings, sections
were incubated in HCl 1M at 45C for 45 min followed by 15 min in Tris-HCl
10 mM (pH 8) prior to permeabilization and blocking. Primary antibodies
were: rabbit anti-DBI (1:50, Santa Cruz), mouse anti-GFAP (1:500, Sigma),
chicken anti-Nestin (1:200, Novus), mouse anti-Mash1 (1:500, BD PharMingen),
chicken anti-EGFP (1:1000, Abcam), rabbit anti-DsRed (1:1000, Clontech
Living Colors), rat anti-BrdU (1:500, Accurate), rat anti-BrdU/CldU
(1:250, Abcam), mouse anti-BrdU/IdU (1:2000, BD PharMingen), rabbit anti-
Active Caspase-3 (1:1500, BD PharMingen), and rabbit anti-Ki67 (1:250,
Abcam). Secondary antibodies were: anti-rabbit Cy3 (1:1000, Jackson
Immuno Research Laboratories), anti-rabbit Alexa 647 (1:1000, Invitrogen),
anti-chicken Alexa 488 (1:1000, Invitrogen), anti-mouse Alexa 488 (1:1000,
Invitrogen), anti-mouse Alexa 647 (1:1000, Invitrogen), anti-mouse Alexa 647
highly cross adsorbed (1:2000, Invitrogen), anti-rat Alexa 488 (1:1000, Invitrogen),
anti-rat Alexa 647 (1:1000, Invitrogen), and anti-rat DyLight 405 (1:500
Jackson Immuno Research Laboratories). The sections were analyzed by
confocal microscopy (Zeiss LSM 700). The specificity of the DBI antibody
was confirmed by the lack of signal in brain sections from DBI knockout
mice (Neess et al., 2011).
Plasmid Construction
shRNA viral vectors were generated as follows: we first used the pSUPER RNAi
vector system (OligoEngine) to clone specific shRNA sequences against
mouse DBI. DBIa: GCTGTTCATCTACAGTCACTT, DBIb: CCTGTGAGGACA
TAATGC. We next subcloned the shRNA sequence together with H1 promoter
from pSUPERinto the retroviral backbone pFB-tdTomato (Stratagene) and into
a murine Moloney leukemia virus-based retroviral vector containing the RSV
promoter driving the expression of EGFP. As a control we used pFB-EGFP
backbone (Stratagene) containing an shRNAscrambled sequence. For overexpression
constructs, we amplified by PCR the CDS of DBI from a mouse cDNA
brain library and cloned it in-frame downstream of an RFP-T2A sequence from
a retroviral vector containing the RSV promoter. The sequence for ODN was
amplified by PCR with DBI clone as a template, adding a stop codon after
the last codon, and cloned in-frame downstream the RFP-T2A sequence as
before. As a control, we used a similar viral vector expressing EGFP instead
of RFP-T2A. For electrophysiological recordings we used a murine Moloney
leukemia virus-based retroviral vector containing RFP driven by the CAG
promoter, kindly provided by Dr. Fred Gage (Salk Institute, San Diego).
Viral Production and Injections
The packaging cell line HEK293 was transfected with the viral backbone vector
together with the helper plasmids and the viral particles were purified by ultracentrifugation.
The concentrated viral solutions (106 108 cfu/ml) were titrated
and mixed. P4 pups were injected into the SVZ with 1 ml of viral solution
through a glass micropipette with the following coordinates from bregma:
0.5 anterior, 1.5 lateral, 1.5 ventral. The animals were returned to their mothers
and killed by decapitation after 3–12 days. For proliferation studies, the
animals were injected twice per day with BrdU (Sigma), 30 mg/kg body weight
on the second and third dpi.
Cell Cycle Experiments
For cell cycle exit experiments, mice were injected with IdU (Sigma) (30 mg/kg
body weight) 50 hr after viral injections, and with CldU (Sigma) (30 mg/kg body
weight) at 12, 14, 16, 18, and 20 hr after IdU injection. With this protocol, we
ensured that, independently of any variation in cell cycle length, all cells would
be labeled with CldU if they divided in the next cycle after IdU injection.
For cell cycle length experiments, mice were injected with IdU (Sigma)
(30 mg/kg body weight) 64 hr after viral injections, followed by a CldU (Sigma)
(30 mg/kg body weight) injection 1.5 hr later, and sacrificed 0.5 hr after the last
injection. This double labeling approach is based on the assumption that all
dividing cells have similar cell cycle length and that their progression through
the cell cycle is asynchronous. Consequently, the ratio of the number of cells in
any two phases of the cell cycle equals the ratio of the length of the two phases
(Nowakowski et al., 1989). Therefore, (the number of cells in S phase)/(the
number of cells leaving S phase) = Ts/ΔT, where ΔT = time during which cells
can incorporate only the first marker. Here ΔT = 1.5 hr, so (double-labeled
IdU CldU cells)/(single-labeled IdU cells) = Ts/1.5 hr. For Tc estimations
we performed the same analysis with the following equation: Ts/Tc = Ns/Nc,
where Ns is the number of cells in S phase and Nc is the total number of proliferating
cells. Therefore, Ts/Tc = (IdU CldU cells)/(all virally infected mitotically
active cells). To calculate the number of mitotically active cells, we performed
Ki67 stainings (a marker of proliferating cells in any active phase of the cell
cycle) and quantified the proportion of Ki67 cells for each group (green and
red infected cells) in six mice. A total of 16 mice were virally infected and
subsequently injected with IdU and CldU. Of these, ten mice were processed
for IdU/CldU stainings and six for Ki67 stainings.
Neurosphere Cultures
The SVZ was microdissected from 500 mm-thick coronal sections of P8–P12
mouse brains. Tissues were treated with papain (0.08%)/DNase I (0.001%) for
3 min at 37C to obtain single-cells suspensions, which were cultured at
a density of 100,000 cells/ml in B27 serum-free/Neurobasal media (Invitrogen)
supplemented with glutamine (2 mM), EGF (20 ng/ml), and FGF (20 ng/ml)
(Sigma). Neurospheres were infected with either RFP or RFP-T2A-DBIexpressing
retrovirus at 1 DIV. Six days postinfection, the cultures were incubated
with either only media, Flumazenil 1 mM (Sigma), or PK-11195 1 mM
(Sigma) for 48 hr, followed by the addition of BrdU 10 mM (Sigma) to the media
for the last 3 hr prior fixation. Cells were fixed in 4% paraformaldehyde-PBS
and immunostained. For quantifications, the number of BrdU-positive cells
was counted and the neurosphere area was measured with ImageJ software.
Electrophysiology
Brains were removed from deeply anesthetized (isoflurane) P6–P10 mice.
250 mm-thick sagittal slices were cut in 4C solution containing (in mM):
125 NaCl, 25 NaHCO3, 1.25 NaH2PO4, 2.5 KCl, 2 CaCl2, 1 MgCl2, 25 glucose;
bubbled with 95%O2/5%CO2 (pH 7.4). Whole-cell recordings were performed
with pipettes pulled from borosilicate glass capillaries with a resistance of
6–8 MU when filled with the following solution (in mM): 127.5 KCl, 11 EGTA,
1 CaCl2, 2 MgCl2, 10 HEPES, 2 Mg-ATP, titrated to pH 7.4 with 35 mM KOH
(final osmolarity 290 mOsm). Liquid junction potentials were not corrected.
Fast (200 ms) applications of GABA onto nucleated outside-out patches
were performed every 20 s with theta glass tubing mounted on a piezo translator.
Applications were repeated 5–10 times for each condition (control, ODN,
wash) at a holding potential of 60mV. Application pipettes were tested by
perfusing solutions with different salt concentrations through the two barrels
onto open patch pipettes and recording current changes with 1 and 500 ms
moves of the application pipette. Only application pipettes with current change
20%–80% rise times below 100 ms and with a reasonable symmetrical on- and
offset were used. The application solution contained (in mM): 135 NaCl, 10
HEPES, 5.4 KCl, 1.8 CaCl2, 1 MgCl2, 5 glucose (pH 7.2). Stimulus delivery
and data acquisition were performed with PatchMaster software (HEKA).
Signals were sampled at 10 kHz and filtered at 3 kHz, and off-line analysis
was performed with Igor Pro (WaveMetrics).
Quantification and Statistical Analysis
For calculations of red/green ratio at different time points, we counted the total
number of infected cells in 7–9 sagittal sections (containing the SVZ) per
animal. All values were normalized to the mean value at 2 dpi. For in vivo proliferation
studies, three pictures from the OBfor each brain section (nine sections
per animal) were taken with a 203objective from a confocal microscope (Zeiss
LSM 700) and the percentage of BrdU-positive cells was calculated for the green, yellow, and red infected neurons. For RMS versus OB quantifications
from virus-infected animals, all sagittal sections where migrating cells could
be observed were used (6–9 brain sections per animal). The number of red
and green cells migrating in the RMS or differentiated in the OB was counted
for each animal. When the total amount of cells per section was less than
500, every cell was counted. When this value was higher (due to a more efficient
virus labeling), three pictures were taken in the RMS and three in the OB per
section with a 203objective lens and the cells from each picture were counted.
Ratios of red/green cells were calculated per animal and these values were
used for the final mean group values and statistical tests applications. All values
were normalized to the mean RMS value. For Mash1 cell quantification,
confocal pictures were taken close to the injection site (one picture per section,
five to eight sections per animal). In all quantifications, double-infected cells
were counted as treated (not control) except for the rescue experiment.
Statistical analyses were performed with GraphPad Prism 5 software. All
data sets were first tested for Gaussian distribution with the Kolmogorov-
Smirnov normality test. We used two-tailed t test or one-way ANOVA (followed
by Tukey’s test or Bonferroni’s Multiple Comparison post hoc analysis) for
parametric data and a Mann-Whitney U test or Kruskal-Wallis test (followed
by post hoc Dunn’s Multiple Comparison test) for nonparametric data.
Traduction - français En 1980, les cellules souches embryonnaires (ES) sont découvertes chez la souris. Moins de deux décennies plus tard, des équipes américaines et israéliennes les découvrent chez l’homme. En 2000, Benjamin Reubinoff transforme des cellules ES en neurones. Depuis plus d’une décennie maintenant c’est un domaine qui trouve application dans bon nombre de branches de la médecine et de la recherche fondamentale.
C’est en août 2005 qu’Angélique Bordey (Faculté de médecine de Yale, USA) publie avec ses collaborateurs une étude dans la section de neurologie de la revue Nature Neuroscience sur la question dont l’étude publiée dans la revue Cell Stem Cell en janvier 2012 a entrepris en partie la résolution. Selon ses propos : « Des cellules souches neurales dans la zone sous-ventriculaire (SVZ) du cerveau provoquerai le gliome, ou des tumeurs de cerveau, quand leur prolifération est hors de commande. (…) Un des buts de cette ligne de recherche est de trouvée des moyens de favoriser la neurogenèse d'une façon ordonnée. Et ce en identifiant ainsi les voies de signalisation, les facteurs et les récepteurs qui bloquent ou favorisent la neurogenèse. Puisque ces facteurs et ces récepteurs fournissent des emplacements additionnels pour des cibles pharmaceutiques favorisant la neurogenèse et le renouvellement automatique des cellules mortes, ils sont considérés comme étant très importants dans la neurongénèse. Par ailleurs, l'identification de cellules GABAergique impliquées dans la prolifération de cellules souches, faites dans cette étude, suggère que toutes les substances qui activeraient les récepteurs GABA vont limiter la neurogenèse. »
Je désir entreprendre l’analyse d’un article issu d’une revue bien connue et prestigieuse dans le milieu scientifique et qui publie les travaux sur les cellules souches, la revue Cell Stem Cell de Cell Press, issu 10, pages 67-87.
La question à laquelle les auteurs répondent est si la présence de récepteurs GABA sur les cellules souches neuronales ainsi que la modulation d’endozépines impliquent que des ligands exogènes, de même famille que les benzodiazépine (Diazepam®), pourraient potentiellement moduler la prolifération des cellules souches neuronales.
La SVZ des ventricules latéraux est connue pour ses caractéristiques de niche importante dans la genèse neuronale au sein du cerveau post natal. Les cellules néo différenciées migrent via le courant de migration rostral (RMS) vers le bulbe olfactif où elles intègrent les circuits neuronaux déjà établis. Dans la présente étude les inhibiteurs de liaison aux diazépines (benzodizépines), en anglais abrégé DBI, sont apparus comme des modulateurs positifs de la prolifération neuronale dans la SVZ par leur effet de régulation des signaux GABA dans les cellules amplifiantes en transit.
Parmi les méthodes utilisées dans l’élaboration du protocole d’expérimentation se retrouvent entre autre l’utilisation d’animaux murins transgéniques, de plasmides, et de marqueurs Ki67 dans les expériences de cycle cellulaire. La mutagenèse est l’inactivation d’un ou plusieurs gènes dans les cellules souches embryonnaires mis en place afin d’étudier les implications de ce ou ces gènes. Par contre deux méthodes révèlent la spécificité de l’article présenté : la création de neurosphères combinées à des manipulations électrophysiologiques. La creation de neurosphères en culture constitue une méthode de production de cellules neuronales dérivées des cellules souches embryonnaires humaines (hESC) qui seront en mesure de former des structures neuronales fonctionnelles à partir d’un certain nombre de lignées. Il a donc été possible d’isoler des cellules de la SVZ de cerveaux murins, et d’en étudier les comportements par électrophysiologie. En effet, cette zone cérébrale chez les mammifères est une niche de cellules souches.
Parmi les résultats obtenus les DBI sont effectivement exprimés à un stade précoce de la neurogenèse régressant significativement au stade du neuroblaste. Parmi les cellules souches testés pour l’expression des DBI dans la SVZ et le RMS, les cellules Mash1 sont majoritairement positives donc présentent lors de la prolifération neuronale, population cellulaire sensible aux DBI.
Par ailleurs des souris Knock Out MASH1- pour les gènes exprimant les DBI a permis de soulever l’hypothèse que lors de la phase de prolifération des cellules souches neuronales celle-ci pourrait être écourtée par l’induction d’une différentiation plus rapide vers des neuroblastes en migration.
Les expériences de surexpression des DBI ont montrés que ces derniers induisaient une augmentation de la population Mash1 , ce qui corrobore la conclusion que les DBI sont des régulateurs positifs de la prolifération neuronale de la SVZ.
Plus avant dans les résultats, le KO des DBI a démontré que ceux-ci n’affectent aucunement la durée du cycle cellulaire dans les cellules en prolifération mais réduit dramatiquement la proportion de cellules s’engageant dans le cycle cellulaire.
L’analyse et la comparaison de plusieurs groupes et données ont montré que les effets des DBI endogènes pourraient être bloqués par le flumazenil et que cette réduction pourrait être réversible par l’addition de DBI exogènes.
D’autre part des fragments protéolytiques des DBI, appelés les octadecaneuropeptides (ODN), surexprimés dans les cellules induisaient une prolongation de la période de migration vers le bulbe olfactif, ce qui pourrait signifier que les ODN interviennent dans les effets des DBI sur les cellules souches dans la SVZ.
Il a été démontré par des méthodes d’électrophysiologies que les ODN inhibaient les flux d’induction des récepteurs GABAa des cellules amplifiantes en transit. Etant donné que les cellules souches, les cellules amplifiantes en transit et les astrocytes sont localisés à proximité des neuroblastes dans le RMS, mais aussi qu’elles sécrètent des DBI/ODN dans l’espace extracellulaire, il est donc possible que les ODN paracrines affectent les cellules en migration également. Résultats qui furent obtenus par de nombreux échantillons neuronaux.
Il est important selon les auteurs que de futures investigations soient menées sur la signalisation des récepteurs GABAa et leurs différents passages intracellulaires dans les différents types cellulaires de la genèse neuronale.
Les conclusions sont claires : la présence de récepteurs GABA dans les cellules souches neuronales ainsi que les différentes modulations d’endozépine (benzodiazepine) décrites dans les expériences impliquent que des ligands exogènes de même nature pourraient être effectivement impliqués dans la dite prolifération.
Au point de vue clinique l’implication dans le développement post natal de futurs découvertes sur ces substances pourraient permettre de concevoir des traitements thérapeutiques plus adéquats, mieux conçus, mais aussi en déterminer les risques. « En effet les benzodiazépines sont sous le scope depuis des années, en France comme ailleurs. En Europe, les benzodiazépines sont 2,5 à 3 fois plus consommées qu'aux Amériques. Au cours des 10 dernières années, leur consommation a augmentée d'environ 25% en Europe, alors qu'elle est restée à peu près stable aux Amériques. L’Afssaps (Agence Française de Sécurité Sanitaire des Produits de Santé) a dressé un état des lieux de la consommation des benzodiazépines en France et de son évolution. Depuis les années 1990, de nombreux travaux ont souligné le niveau élevé de consommation de médicaments psychotropes en France, en particulier les anxiolytiques et les hypnotiques, principalement représentés par les benzodiazépines. En 2009, certaines données européennes plaçaient la France au deuxième rang des pays européens consommateurs d’anxiolytiques (après le Portugal) et d’hypnotiques (après la Suède). De plus 134 millions de boîtes de médicaments contenant des benzodiazépines ou apparentées ont été vendues en France en 2010 (50,2% d’anxiolytiques et 37,6% d’hypnotiques).»
D’autre part dans un contexte il a été rapporté dans la littérature que les cellules souches de la zone sous-ventriculaire pouvaient être à l’origine de glioblastomes, formes très malignes de tumeurs cérébrales, comme le souligne l’extrait ci-dessous.
Le glioblastome multiforme est le plus fréquent et la plus mortel des tumeurs primaires du sytème nerveux central, tumeurs dont la compréhension est insuffisante. Actuellement, existe une controverse sur l'identité de la cellule initiatrice du glioblastome, une modification d'un astrocyte différencié, une cellule souche neurale peut être. Les cellules souches de cancer putatives, lesquels ont les traits des cellules souches normales avec en plus la capacité de récapituler le phénotype de la tumeur in vivo dans de petits nombres de cas, a été identifié dans une grande variété de cancers humains solides, y compris le glioblastome. Ce qui suggère que les régions qui hébergent les cellules souches normales dans le système nerveux central adulte, tel que la SVZ et le gyrus denté, sont plus propices à une oncogenèse virale et chimique, et supportent l'hypothèse que les tumeurs du cerveau proviennent de ces cellules souches. Cependant, il doit encore être déterminé si les cellules souches de tumeur de cerveau sont la cause ou la conséquence d'une initiation de la tumeur à partir de ces cellules souches et son développement.
Cependant d’après un article paru dans la revue Cell Research en 2011 , les différences entre les modèles, murins, primates non-humains et humains divergent en termes de structures dans la zone sous ventriculaire. En effet chez les rongeurs et les primates incluant les humains la niche neuronale de la SVZ existe des similitudes mais aussi des différences, telles que l’organisation anatomique du courant migratoire rostral (RMS). Le débat se situe donc autour de la nature de ce RMS chez les primates humains. Chez les rongeurs ce RMS a pour origine la SVZ, pour se diriger vers le bulbe olfactif ou il remplacera les interneurones. Par contre chez l’homme la mise en évidence d’un tel courant est identifiée par des marqueurs endogènes tels que PCNA, pHH3, Mcm2 and Ki67, les neuroblastes par PSA-NCAM et la doublecortin (DCX), que les neurones différenciés par NeuN et β-III tubulin/TuJ1.
Suite à des expériences mentionnées dans l’étude il n’y a aucune preuve que chez l’humain existe ce RMS. Les phénomènes de différenciation observés démontrent que ceux-ci diffèrent d’avec ceux connus dans les cerveaux des rongeurs et des primates, bien que la présence de cellules de migration soit effectivement présente.
D’autres études, citées dans l’article ont démontrées grâce à la protéine DCX, la présence d’un RMS fœtal confirmant que ce dernier est réellement un courant avec des neurones migrant en chaîne dans le cerveau en développement. Déterminer la présence de neuroblastes en migration au sein d’un RMS dans le cerveau humain est important, car cela impliquerait que les cellules souches de la SVZ dans le cerveau humain sont aptes à générer des neuroblastes capables de migrer et potentiellement de se différencier.
Les expériences effectuées dans un premier temps avec les rongeurs afin de tester l’effet de la protéine DBI, est-elle transposable aux cellules souches neurales d’un cerveau humain ?
Suite aux résultats qui démontrent des similitudes du RMS et de la SVZ entre rongeurs, primates et humain je peux espérer retrouver dans la littérature suffisamment d’éléments afin de transposer les expériences sur des cellules humaines.
français vers anglais: Behavioral experiment and Bayesian inference with probabilistic population codes General field: Médecine Detailed field: Psychologie
Texte source - français L’élucidation des stratégies de prise décisionnelle ont fait l’objet de nombreuses études depuis plusieurs décennies par des neuroscientifiques, neurophysiologistes, psychocliniciens et médecins. Les études effectuées chez les primates non-humains et chez des sujets humains ont révélées grâce à des tâches psychologiques appliquées de type discriminatoire tout un panel de populations neuronales impliquées dans les processus de prise de décision. Plusieurs investigations invasives et non-invasives ont démontrés que ces populations ne sont pas confinées dans une seule région corticale, mais pourtant qu’un réseau de communication existe entre différentes régions et régit par un code que vient clarifier l’inférence Bayésienne.
L’aptitude des primates humains, et non humains à discriminer les fréquences vibratile ont été étudié (Harris et al., 2001; Sinclair & Burton, 1996; Mountcastle, Steinmetz & Romo, 2000) et mesurée par des expériences psychophysiques. Des conclusions à propos des capacités des deux primates indiquent des similitudes dans leurs aptitudes. Les tâches de discrimination vibratile ont été menés chez des primates non-humains (Hernandez et al., 1997; Hernandez et al., 2007; Romo & Salinas, 2003) mais aussi chez des sujets humains (Kostopoulos et al., 2007; Li Hegner et al., 2007; Pleger et al., 2006; Preuschhof et al., 2006; Harris et al., 2002). Plusieurs études d’imagerie fonctionnelle par résonnance magnétique (fMRI) démontrent que les évènements impliqués dans la prise de décision ont induits l’activation des aires somatosensitives, prémotrices, et du cortex latéral préfrontal (LPFC) (Preuschhof et al., 2006). Cependant cette partie bien spécifique du cortex préfrontal est impliquée activement dans le processus de captation contrôlé nécessaire à l’élucidation de l’information vibratile impliquée dans la mémoire à court-terme (Kostopoulos et al., 2007). Cependant, le processus psychologique de construction de la mémoire de travail (WM) possède des caractéristiques très éphémères quant au temps de rétention et aux capacités de stockage (Fuster et al., 2000; Levy and Goldman-Rakic, 2000). De plus Fuster et al. discutent de l’existence d’une mémoire rétrospective à court-terme, la mémoire de travail sensorielle, ainsi que d’une mémoire prospective attentionnelle, la mémoire motrice. Il semblerait selon ses résultats qu’un processus de rétention non-exhaustif pour cette mémoire de travail surviennent dans les deux exemples par la réintroduction de l’excitation. Quant à Levy et Goldman-Rakic proposent un modèle chez le singe qui permet d’élucider de manière séquentielle et modulaire une spécificité de domaine de l’organisation fonctionnelle du cortex préfrontal (PFC) en lien avec les opérations de la mémoire de travail (Smith et al. 1995, 1996; Courtney et al. 1996, 1997, 1998; McCarthy et al. 1996; Belger et al. 1998; Kelley et al. 1998; Kohler et al. 1998; Smith and Jonides 1999). Des enregistrements neurophysiologiques (Romo et al., 1999) ont démontré qu’un encodage monotone du stimulus pourrait être la représentation opérationnelle d’un stimuli sensoriel unidimensionnel dans la mémoire de travail. Par ailleurs, des études ont démontrées que l’encodage multimodal et les décisions perceptuels qui y sont liées se produisent à l’extérieur des régions sensitives (Lemus et al. 2010). Finalement certaines études ont décrites avec certains modèles (respectivement Machens et al. 2010 ; Barak et al., 2010) que les représentations temporelles et mnésiques sont maintenues par des mécanismes distincts, d’autant plus qu’un partage substantiel et anatomique existe, mais suggèrent aussi qu’un mécanisme existe où l’information à propos du stimuli est impliquée dans un processus de facilitation synaptique sur des connexions récurrentes et de manière activité dépendante.
Traduction - anglais The disentangling decision making strategies have been a matter of study for neuroscientists, neurophysiologists, psychologists and physicians for the past decades. Studies in non-human primates and human subjects have revealed through discrimination tasks a panel of neuronal populations involved in the processes of decision making. Invasive and non-invasive investigations rather demonstrated that these populations were not rendered in unique cortical areas, and the existence of a communication code which is also clarified from Bayesian inference.
The ability of non-human primates and humans to discriminate vibrotactile frequencies has been investigated (Harris et al., 2001; Sinclair & Burton, 1996; Mountcastle, Steinmetz & Romo, 2000) and measured in psychophysical experiments. Relevant conclusions on the capacities of both primates indicate the similarities of this ability. Vibrotactile delayed discrimination tasks have been performed in non-human primates (Hernandez et al., 1997; Hernandez et al., 2007; Romo & Salinas, 2003) but also with human subjects (Kostopoulos et al., 2007; Li Hegner et al., 2007; Pleger et al., 2006; Preuschhof et al., 2006; Harris et al., 2002). Many functional magnetic resonance imagery studies (fMRI) show that decision making events originated activation in the somatosensory, premotor, and the lateral prefrontal cortex (LPFC) areas (Preuschhof et al., 2006). But that specific part of the prefrontal cortex is involved in active controlled retrieval processing necessary for the disambiguation of vibrotactile information in short-term memory (Kostopoulos et al., 2007). However, the psychological construct of working memory (WM) is presumed to have an evanescence characteristic in both time and reminiscence capacities (Fuster et al., 2000; Levy and Goldman-Rakic, 2000). Furthermore Fuster et al. discussed the existence of a retrospective short-term, sensory working memory, and of a prospective attentive set, motor memory. It seemed according to his results that a non-exhaustive process of retention from working memory occurs in either kind by the re-entry of excitability. As for Levy and Goldman-Rakic, they proposed a model in the monkey to assess a modular “domain-specific” of PFC functional organization with respect to WM operations (Smith et al. 1995, 1996; Courtney et al. 1996, 1997, 1998; McCarthy et al. 1996; Belger et al. 1998; Kelley et al. 1998; Kohler et al. 1998; Smith and Jonides 1999). Moreover, in functional-imaging studies of human cognition several sub-regions should in the DLPFC create activation, related primarily on the nature of the information being processed in WM (for a review see; Owen et. 1996a, 1997, 1998, 1999; Ungerleider et al. 1998; Smith and Jonides, 1999; Petrides et al., 1993a, 1993b). Single-unit recordings (Romo et al., 1999) demonstrate that a monotonic stimulus encoding may be the basic representation of one-dimensional sensory stimulus quantities in working memory. Elsewhere (Lemus et al. 2010) studies showed that multimodal encoding and perceptual judgments in these tasks occur outside the sensory cortices. Finally studies elucidated (respectively Machens et al. 2010 ; Barak et al., 2010) with models that the representation of time and memory are maintained by separate mechanisms, even while sharing a common anatomical substrate, but also suggesting that a mechanism exists whereby information about the stimulus is contained in activity-dependent synaptic facilitation of recurrent connections.
More
Less
Études de traduction
Bachelor's degree - University of Montreal
Expérience
Années d'expérience en traduction : 21. Inscrit à ProZ.com : Oct 2012.
In order to present my-self, you will find my cv attached to this email. I have multiple nationalities and speak fluently four languages. My translation experiences are either academic or non professional based in sciences and law fields.
I'm a scientist and publishing in well-renown scientific reviews.
Your post job clarifies your expectations and I am presently available for translations from French to English.
Compétences