Chat with us, powered by LiveChat Summary and reflect on chosen scientific articles | All Paper

Attached below are two papers that you need to read, do a short summary and reflection. You don’t need to be detailed, just the main points are needed. You will do a 1 page summary and reflection from each article. SummaryIn ½ pages, briefly state the purpose of the article and summarize the findings/results. You may also mention the method used for collecting data as well.Paragraph may be double spaced. Make sure you have set your margins at one inch! Paragraphs must be a minimum of at least 5 sentences. ReflectionIn ½pages, provide a brief reflection of your thoughts regarding the article. You can cover such topics as: Were the results surprising to you? How do you think this study adds to the field of psychology? What was your overall opinion of the article? Did you like it? Dislike it?Paragraph may be double-spaced. Make sure you have set your margins at one inch! Paragraphs must be a minimum of at least 5 sentences.


Unformatted Attachment Preview

Disinhibition, a Circuit Mechanism
for Associative Learning and Memory
Johannes J. Letzkus,1,* Steffen B.E. Wolff,2,3,4 and Andreas Lüthi2,*
Planck Institute for Brain Research, 60438 Frankfurt, Germany
Miescher Institute for Biomedical Research, Maulbeerstrasse 66, CH-4058 Basel, Switzerland
3University of Basel, 4000 Basel, Switzerland
4Harvard University, Center for Brain Science, Cambridge, MA 02138, USA
*Correspondence: (J.J.L.), (A.L.)
Although a wealth of data have elucidated the structure and physiology of neuronal circuits, we still only have
a very limited understanding of how behavioral learning is implemented at the network level. An emerging
crucial player in this implementation is disinhibition—a transient break in the balance of excitation and inhibition. In contrast to the widely held view that the excitation/inhibition balance is highly stereotyped in cortical
circuits, recent findings from behaving animals demonstrate that salient events often elicit disinhibition of
projection neurons that favors excitation and thereby enhances their activity. Behavioral functions ranging
from auditory fear learning, for which most data are available to date, to spatial navigation are causally linked
to disinhibition in different compartments of projection neurons, in diverse cortical areas and at timescales
ranging from milliseconds to days, suggesting that disinhibition is a conserved circuit mechanism contributing to learning and memory expression.
The neuronal mechanisms of associative learning have been under intense investigation for many decades. At the macroscopic
level, this work has been very successful in identifying the brain
areas involved in the acquisition, consolidation, and expression
of different learning tasks (Fanselow and Poulos, 2005; Frankland and Bontempi, 2005; LeDoux, 2000). A parallel line of
research has provided an in-depth understanding of the cellular
and molecular mechanisms of plasticity at excitatory synapses
as a necessary component of memory formation (Bliss and Collingridge, 1993; Malenka and Bear, 2004; Martin and Clark, 2007;
Martin et al., 2000; Nabavi et al., 2014; Neves et al., 2008; Nicoll
and Malenka, 1995; Sah et al., 2008; Sjöström et al., 2008; Whitlock et al., 2006). In contrast to these well-understood levels of
organization, the most important unresolved issues today relate
to the events in local neuronal microcircuits during learning and
memory (Neves et al., 2008). At this mesoscopic level, associative learning manifests as a change in information processing by
neuronal circuits. But due to the great complexity of these networks and their emergent properties, several fundamental questions have remained open: how is plasticity induced during
learning (i.e., which afferent pathways put the circuit into a plastic state) and which local circuit elements are affected by these
signals? And how does learning-related plasticity in turn alter
the function of the local circuit (i.e., which aspects of stimulus encoding are changed to mediate memory expression), and how
does previous experience affect the circuit’s plasticity state?
While these topics will keep the community busy for years to
come, a strong body of recent data indicates that disinhibition—transient and selective breaks in the excitation/inhibition
balance—is causally involved in all these functions.
One of the most robust and ubiquitous findings in neuroscience is that projection neurons process information at a tightly
264 Neuron 88, October 21, 2015 ª2015 Elsevier Inc.
controlled balance of excitation and inhibition. For instance, during sensory processing in neocortex, presentation of a sensory
stimulus invariably recruits inhibition in addition to excitation,
leaving the excitation/inhibition balance approximately unperturbed (Gabernet et al., 2005; Isaacson and Scanziani, 2011;
Wehr and Zador, 2003; Xue et al., 2014). This is at first glance
a paradoxical situation that also claims a major part of the brain’s
energy consumption (Buzsáki et al., 2007) but has important
functions. First, it ensures temporally precise firing of projection
neurons: afferent excitation to cortical structures is transmitted
to both projection cells and interneurons, which mediate feedforward inhibition of projection neurons after a delay of a few milliseconds, thus creating a very brief window of opportunity for
firing (Gabernet et al., 2005; Pouille and Scanziani, 2001; Wehr
and Zador, 2003; Wilent and Contreras, 2005). Second, different
inhibitory interneuron types selectively connect to different subcellular compartments of projection neurons such as the axon
initial segment, the perisomatic, and different dendritic regions
(Fishell and Rudy, 2011; Klausberger and Somogyi, 2008), and
inhibition thus has the power to dynamically regulate the processing of specific inputs, their interactions, and plasticity (Gidon and Segev, 2012; Miles et al., 1996; Pouille et al., 2013).
Third, appropriately timed inhibition is crucially involved in the
generation and maintenance of network oscillations, which serve
to organize information processing temporally and to coordinate
communication between different brain areas (Buzsáki and
Wang, 2012; Fries, 2009). A fourth important functional consequence of synaptic inhibition is response selectivity: projection
neurons in sensory neocortex for instance are often highly selective for certain features of external stimuli, and stimulus discrimination is also a crucial factor during learning and memory
expression. GABAergic inhibition in general (Chen and Jen,
2000; Katzner et al., 2011; Kyriazi et al., 1996; Wang et al.,
2000) as well as defined interneuron types (Hamilton et al., 2013;
Lee et al., 2012; Wilson et al., 2012) are crucial for this tuning, suggesting that inhibition is a dominant factor mediating the selectivity of projection neuron responses. Finally, inhibition is crucially
involved in adjusting the input-output relationship or gain of the
projection neuron network. Gain control is a general attribute of
neuronal circuits (Salinas and Thier, 2000) and in its simplest
form can normalize the average activity of neurons across a
wide range of input strengths, with little or no effect on stimulus
selectivity. In cortex, one mechanism producing this form of
gain control is that stronger stimuli recruit more feedforward inhibition, making it harder for excitation to fire projection cells (Isaacson and Scanziani, 2011; Pouille et al., 2009). Recent studies suggest that parvalbumin-positive (PV) interneurons targeting the
perisomatic domain of projection neurons can fulfill this function
in visual cortex (Atallah et al., 2012; Wilson et al., 2012).
It is important to note that the concept of tightly regulated excitation/inhibition balance in cortical networks has largely been
derived from experiments on sensory physiology, where inhibition and excitation are recruited by the same sensory stimulus
while the behavioral state of the animal is kept constant. In
contrast, gating by disinhibition has long been recognized as a
central processing motif in the basal ganglia, which perform action selection by relieving defined target neurons from ongoing
inhibition (Chevalier and Deniau, 1990; Goldberg et al., 2013).
While simulations suggest that regulation of inhibition independent of and by other factors than excitation can selectively
gate the processing of specific signals also in cortical circuits
(Kremkow et al., 2010; Vogels and Abbott, 2009), only recent evidence accumulated using cell-type-specific recordings (Kerr
and Denk, 2008; Lima et al., 2009; Royer et al., 2010) and activity
perturbations (Sternson and Roth, 2014; Zhang et al., 2007)
under behaviorally relevant conditions has demonstrated that
cortical inhibition can indeed be modulated independent of excitation. A fundamental discovery in circuit neuroscience is thus
that inhibition dynamically orchestrates circuit activity according
to the current processing requirements of the animal (Kepecs
and Fishell, 2014; Poorthuis et al., 2014; Roux and Buzsaki,
2014), and this capacity may be another important reason for
the ubiquity of inhibition. Consistent with the fact that inhibition
is strong during baseline circuit function, a recurring observation
in these recent studies has been that salient stimuli, experience,
and the animal’s internal state can cause disinhibition in cortex, a
selective and transient reduction of synaptic inhibition received
by projection neurons that significantly changes their computations. Reductions in inhibition can be mediated by a variety
of subcellular, cellular, and network mechanisms (Froemke,
2015). Here, we focus mainly on forms of disinhibition caused
by reduced firing of different interneuron types, since this mechanism has been most extensively studied in relation to the
animals’ behavior. Cortical disinhibition has now been linked to
brain functions ranging from sensorimotor integration (Lee
et al., 2013; Xu et al., 2012) to social behavior (Cohen and Mizrahi, 2015; Marlin et al., 2015; Owen et al., 2013; Yizhar et al.,
2011) and attention (Sridharan and Knudsen, 2014; Vogels and
Abbott, 2009; Zhang et al., 2014b). However, the strongest and
most comprehensive evidence to date has been obtained in experiments addressing learning and memory, in particular audi-
tory fear conditioning. The emerging picture from this work is
that disinhibition occurs in diverse cortical areas at timescales
ranging from milliseconds to days and that different disinhibitory
circuits targeting distinct sub-cellular domains of projection neurons are causally related to learning, memory expression, and
regulation of the circuit’s plasticity state. While disinhibition
has thus been firmly established as an essential processing
motif, a full understanding of its mechanisms, consequences,
and behavioral relevance requires much further investigation,
and some of these open questions will be discussed in the
concluding section of the review.
Initial Evidence for Disinhibition in Learning
In 1964, Young proposed a model of learning where ‘‘in the untrained condition.pathways are held the action
of small cells.[with] inhibitory collaterals.Learning would
then consist in removal of inhibition from one path’’ (Young,
1964). It had thus been proposed for a long time that inhibition
and disinhibition play key roles in learning and memory formation. While strong and unspecific disinhibition is deleterious to
any brain function and causes epilepsy, a number of early experimental studies using mild pharmacological and genetic manipulations of inhibition provided evidence consistent with the
notion that inhibition restrains learning. Systemic or local interventions that increase GABAergic function during learning can
interfere with the acquisition of aversive (Brioni et al., 1989; Davis, 1979; Harris and Westbrook, 1995; Sanger and Joly, 1985)
and spatial memory (Arolfo and Brioni, 1991; McNaughton and
Morris, 1987). Conversely, manipulations that mildly decrease inhibition often lead to better learning (Brioni et al., 1989; Izquierdo
et al., 1993). While phasic GABAergic transmission likely plays a
major role in these effects, there is also good evidence for a
similar constraining effect on memory by extrasynaptic, tonic inhibition, which is regulated at a much slower timescale and likely
much more global in its action (Farrant and Nusser, 2005;
Semyanov et al., 2004). In the hippocampus, tonic inhibition is
mediated by GABAA receptors containing the alpha-5 subunit
(Caraiscos et al., 2004; Fritschy and Mohler, 1995), and genetic
and pharmacological manipulations decreasing this current
lead to marked enhancement of hippocampus-dependent memory such as spatial learning (Chambers et al., 2003; Collinson
et al., 2006; Collinson et al., 2002) and trace fear conditioning
(Martin et al., 2010), likely due to enhanced memory acquisition
and expression but not consolidation (Collinson et al., 2006).
In addition to these experimental manipulations of GABAergic
function, there is also evidence suggesting that memory acquisition and/or consolidation can be associated with physiologically
reduced inhibition. Fear conditioning for instance causes downregulation of several genes associated with inhibition, such as
glutamic acid decarboxylase (the GABA synthesizing enzyme;
Bergado-Acosta et al., 2008; Heldt and Ressler, 2007) several
types of GABAA-receptors, and the GABAA-receptor scaffolding
protein gephyrin (Chhatwal et al., 2005; Heldt and Ressler, 2007).
Consistently, the amplitude and frequency of spontaneous inhibitory currents in lateral amygdala projection neurons are reduced
after fear conditioning and retrieval (Lin et al., 2009).
Together, these data are consistent with disinhibition as an
important mechanism enabling acquisition and expression of
Neuron 88, October 21, 2015 ª2015 Elsevier Inc. 265
memory. However, more specific approaches were required to
address several crucial questions: is physiological disinhibition
indeed involved in learning and memory? In which tasks does
disinhibition occur, and when is it required for learning? What
is the time course of disinhibition during acquisition and expression of memory? Which afferent pathways produce disinhibition? Which local interneuron types mediate disinhibition, and
which subcellular domain of local projection neurons do they
contact? How do potentially different forms of disinhibition affect
computations in projection neurons? Does disinhibition affect
all local projection neurons equally, or is it specific for certain
groups or even for specific inputs to certain groups?
Disinhibition in Fear Learning
Auditory fear conditioning performed by pairing an initially
neutral tone (conditioned stimulus [CS]) with a mildly aversive,
inescapable stimulus (unconditioned stimulus [US]) is a powerful
model system for investigating the plasticity of neuronal circuits
and the mechanisms of associative learning and memory
expression (Duvarci and Pare, 2014; LeDoux, 2000; Maren and
Quirk, 2004; Pape and Pare, 2010). One key advantage of this
paradigm is that the timing of the CS and the US can be precisely
controlled, enabling dissection of the underlying neuronal mechanisms at high temporal resolution. In addition, the neuronal
circuitry underlying this form of learning has been thoroughly
investigated. This work identified the basolateral amygdala as
a key brain area where associative synaptic plasticity at glutamatergic sensory afferents is induced by convergence of CS- and
US-related input (LeDoux, 2000; Maren and Quirk, 2004; Pape
and Pare, 2010). More recent work suggests that the amygdala
functions as a vital hub integrating information from several
different brain areas (Herry and Johansen, 2014). For instance,
auditory cortex (areas A1 and AuV) contributes to acquisition
of fear conditioning (Herry and Johansen, 2014; LeDoux, 2000;
Romanski and LeDoux, 1992), and recent evidence suggests
that this pathway is essential for fear learning to complex tones
(Letzkus et al., 2011; Peter et al., 2012). In addition, emerging
evidence suggests that the dorso-medial prefrontal cortex is critically involved in the flexible control of fear expression (Herry and
Johansen, 2014; Sotres-Bayon and Quirk, 2010).
Disinhibition Recruited by Unconditioned Stimuli
Auditory fear conditioning as well as several other forms of associative learning causes prominent long-lasting plastic changes of
CS responses in auditory cortex (Quirk et al., 1997; Schreiner
and Polley, 2014; Suga and Ma, 2003; Weinberger, 2007). This
form of plasticity depends on neocortical acetylcholine release
from basal forebrain afferents (Ji et al., 2005; Schreiner and Polley, 2014; Suga and Ma, 2003; Weinberger, 2007), and pairing of
tones with basal forebrain stimulation elicits changes in auditory
cortex that are similar to those observed with fear learning (Kilgard and Merzenich, 1998; Suga and Ma, 2003; Weinberger,
2007). Froemke and colleagues (2007) used whole-cell recordings in anesthetized rats to investigate how pairing of auditory
stimuli with stimulation of the basal forebrain affects toneevoked synaptic input. Acetylcholine release caused a reduction
of tone-evoked inhibition, which was already apparent after a
few seconds. This was followed by an enhancement of excitatory transmission, which took much longer to develop and
266 Neuron 88, October 21, 2015 ª2015 Elsevier Inc.
was specific for the paired tone. Thus, the primary effect of pairing is a break in the excitation/inhibition balance, which leads to
greater action potential firing and higher incidence of burst firing
in the recorded neurons. Given that effective voltage-clamp is
limited to proximal sites in large neurons (Williams and Mitchell,
2008), it appears likely that the observed effects are of perisomatic origin. In contrast to the rapid onset, re-balancing of
excitation and inhibition at the paired tone frequency takes
approximately 2 hr to complete. These data suggest that disinhibition induced by acetylcholine release gates the induction of
LTP at excitatory synapses activated by the paired tone
(Froemke et al., 2007).
How is disinhibition implemented in the local circuit, and how
does it contribute to learning at the behavioral level? Letzkus and
colleagues (2011) investigated the circuit mechanisms of auditory cortex plasticity during acquisition of auditory fear conditioning. They observed that foot-shocks, which drive learning
in this paradigm, elicit strong, time-locked firing in the majority
of layer 1 interneurons of auditory cortex. This response is mediated by acetylcholine release from basal forebrain afferents activating nicotinic receptors on these interneurons (Figure 1B).
Layer 1 interneurons are recruited 50–60 ms after foot-shock
onset, approaching the speed of conventional synaptic transmission and consistent with recent in vitro measurements employing optogenetic stimulation of cholinergic axons (Bennett
et al., 2012; Poorthuis et al., 2014). In line with the diffuse nature
of basal forebrain projections, similar foot-shock responses are
present in visual cortex, suggesting that this signal is widespread
throughout cortex.
Layer 1 contains two main types of interneurons, which are
both depolarized from rest by acetylcholine acting on nicotinic
receptors: interneurons with a simple axonal arbor (sometimes
referred to as single bouquet cells) that preferentially contact
deeper layer interneurons and neurogliaform cells displaying a
dense axonal plexus that supply inhibition to both projection
cells and interneurons (Christophe et al., 2002; Chu et al.,
2003; Jiang et al., 2013; Lee et al., 2014; Letzkus et al., 2011;
Wozny and Williams, 2011). Consistent with inhibition from layer
1 simple axon cells, foot-shocks cause inhibition of fast-spiking
PV interneurons in layer 2/3, which supply strong inhibition to the
perisomatic domain of projection neurons (Hu et al., 2014; Markram et al., 2004). In turn, this leads to a reduction of spontaneous
and feedforward inhibition in projection cells. Notably, a minority
of layer 1 interneurons is inhibited by foot-shocks (Letzkus et al.,
2011), and recent experiments suggest that these may correspond to neurogliaform cells that are inhibited by acetylcholine
during ongoing firing (Brombas et al., 2014). Since neurogliaform
cells directly target deeper layer projection neurons (Jiang et al.,
2013), this mechanism constitutes a potential second source of
Rapid recruitment by aversive foot-shocks and air-puffs has
also been observed for interneurons expressing vasoactive intestinal peptide (VIP) in auditory cortex (Pi et al., 2013). Since
VIP interneuro …
Purchase answer to see full

error: Content is protected !!