Conference Abstracts 2010 – 2014

2014

G. N. PHO, M. J. GOARD, B. CRAWFORD, M. SUR,
Active engagement induces stimulus-specific modulation of population activity in visual and parietal cortex of mice
Soc. Neurosci., 2014.

Two individuals presented with similar sensory input may process the incoming material in very different ways depending on their level of engagement with the presented information. Attentional engagement is thought to improve encoding of sensory inputs, but the underlying neural mechanisms are not well understood. Studies in non-human primates have shown that attention can modulate neural response amplitude and between-cell correlations, with stronger attentional effects in higher cortical regions compared to primary sensory areas. However, the specific circuit mechanisms by which stimulus-selective enhancement is achieved have been difficult to elucidate in primates. Here we investigate the effects of active behavioral engagement using large-scale two-photon calcium imaging in mice trained on a go/no-go visual discrimination task. Population activity from hundreds to thousands of neurons in primary visual cortex (V1) and posterior parietal cortex (PPC) were measured using the genetically encoded calcium indicator GCaMP6s and a novel volumetric imaging approach. Using a retractable lick spout, we alternated blocks of “engaged” behavior with blocks of “passive” viewing of identical stimuli, and compared the neural responses across the two conditions. We find that V1 neurons show subtle but significant modulation with engagement in a stimulus-specific manner, with enhanced responses to target (go) stimuli and suppressed responses to nontarget (no-go) stimuli. Additionally, engagement results in a decrease in between-cell noise correlations for both target and nontarget-selective neurons. In PPC, however, we find a dramatic enhancement of responses to target stimuli during engagement, reminiscent of the enhanced attentional effects observed in higher cortical areas in primates. Lastly, we have used the VGAT-EYFP-ChR2 mouse line (which express ChR2 in all inhibitory neurons), to test the causal role of each region during the task with high spatiotemporal precision. Optogenetic inhibition of either V1 or PPC during the stimulus period of the task completely disrupts performance. Ongoing work will investigate the underlying circuit connectivity that enables behavioral modulation of information flow from lower to higher cortex.

M. GOARD, G. PHO, M. SUR,
Optical dissection of cortical circuits underlying short-term memory
Soc. Neurosci., 2014.

Short-term memory, the ability to hold information in mind over short timescales, is a fundamental cognitive process underlying an array of complex abilities. Short-term memory is associated with sustained neural activity in cortical (and subcortical) structures, but it is not established how mnemonic information is maintained within distributed cortical regions. Here we developed a memory-guided visual discrimination task (with stimulus, delay, and choice epochs) for head-fixed mice, enabling large-scale 2-photon calcium imaging and targeted optogenetic perturbation of sensory, association, and motor cortices during STM. High-speed volume scanning was used to simultaneously image the neural activity of large populations (hundreds to thousands of cells) of GCaMP6s-expressing cortical neurons in visual, parietal, and frontal motor cortices. Visual cortex (V1) neurons primarily responded during the stimulus epoch whereas posterior parietal (PPC) and frontal motor cortical (fMC) neurons exhibited complex responses across all task epochs, including steady-state delay-period activity. To test the functional role of each region during task performance, we used VGAT-ChR2-EYFP mice (expressing channelrhodopsin in inhibitory interneuron types) to inactivate defined regions of cortex in a spatially and temporally restricted manner. Optogenetic inhibition of bilateral V1 or PPC disrupted behavioral performance only during the stimulus epoch, revealing that PPC was not necessary for STM maintenance despite exhibiting prominent delay-period activity. Surprisingly, photoinhibition of fMC disrupted performance not only during choice, but also during stimulus and delay epochs. These results indicate that in a memory-guided behavior, task-relevant information can be rapidly transmitted from sensory to association to frontal motor cortices, where steady-state activity is essential for STM maintenance.

D. A. FELDMAN, N. MELLIOS, S. D. SHERIDAN, S. KWOK, B. ROSEN, B. CRAWFORD, S. HAGGARTY, M. SUR,
An isogenic human induced pluripotent stem cell model of Rett Syndrome reveals early alterations in microRNA expression patterns and downstream neuronal maturation
Soc. Neurosci., 2014.

Rett Syndrome (RTT) is an X-linked developmental disorder that is predominantly caused by mutations in methyl CpG-binding protein 2 (MECP2). Clinical features of RTT include a period of normal development lasting up to 6-18 months followed by stagnation of both neurological and general growth development. As this is a disorder of early development, it has proven difficult to elucidate phenotypes and/or molecular signatures at a pre-symptomatic stage. We have generated human induced pluripotent stem cell (iPSC) lines in which to study molecular and functional phenotypes in vitro at various stages of neural development. We took advantage of clonal X-inactivation from RTT patient lines in addition to shRNA knockdown of MECP2 in control lines to generate pairs of isogenic cell lines. Neural progenitors were generated from iPSCs via dual-SMAD inhibition and subsequently differentiated into cortical neurons, which exhibited intrinsic membrane excitability and spontaneous post-synaptic activity as observed via whole-cell patch clamp and functional calcium imaging. In this study, we aimed to identify key miRNA and molecular signatures at early stages of neuronal differentiation and in neural progenitors using these isogenic patient-derived and MECP2 knockdown cell lines. We screened for affected miRNAs and discovered a distinct miRNA family of interest that is robustly augmented in RTT patient-derived cell lines. Our results were subsequently confirmed in MECP2 knockdown cell lines and embryonic mouse brain tissue. In parallel, components of the mitogen-activated protein kinase (MAPK) pathway that are known targets of the upregulated miRNA family were found to be misregulated in RTT lines. Such pathways are known to converge on key processes of neuronal development, thus leading to observed maturation deficits in early RTT neurons. Downregulation of aforementioned miRNAs in MECP2 knockdown neural progenitors ameliorated downstream neuronal maturation. Ongoing experiments are focused on elucidating the mechanisms of disease-related impairments of neurogenesis in a mouse model of RTT. Our data highly supports the hypothesis that the expression of a key miRNA family is altered in a human model of RTT and has significant effects on downstream signaling pathways implicated in RTT pathogenesis

R. GARCIA, M. GOARD, J. PETRAVICZ, M. SUR,
Ca2+ responses in astrocytes of unanesthetized mouse visual cortex
Soc. Neurosci., 2014.

Rett Syndrome (RTT) is an X-linked developmental disorder that is predominantly caused by mutations in methyl CpG-binding protein 2 (MECP2). Clinical features of RTT include a period of normal development lasting up to 6-18 months followed by stagnation of both neurological and general growth development. As this is a disorder of early development, it has proven difficult to elucidate phenotypes and/or molecular signatures at a pre-symptomatic stage. We have generated human induced pluripotent stem cell (iPSC) lines in which to study molecular and functional phenotypes in vitro at various stages of neural development. We took advantage of clonal X-inactivation from RTT patient lines in addition to shRNA knockdown of MECP2 in control lines to generate pairs of isogenic cell lines. Neural progenitors were generated from iPSCs via dual-SMAD inhibition and subsequently differentiated into cortical neurons, which exhibited intrinsic membrane excitability and spontaneous post-synaptic activity as observed via whole-cell patch clamp and functional calcium imaging. In this study, we aimed to identify key miRNA and molecular signatures at early stages of neuronal differentiation and in neural progenitors using these isogenic patient-derived and MECP2 knockdown cell lines. We screened for affected miRNAs and discovered a distinct miRNA family of interest that is robustly augmented in RTT patient-derived cell lines. Our results were subsequently confirmed in MECP2 knockdown cell lines and embryonic mouse brain tissue. In parallel, components of the mitogen-activated protein kinase (MAPK) pathway that are known targets of the upregulated miRNA family were found to be misregulated in RTT lines. Such pathways are known to converge on key processes of neuronal development, thus leading to observed maturation deficits in early RTT neurons. Downregulation of aforementioned miRNAs in MECP2 knockdown neural progenitors ameliorated downstream neuronal maturation. Ongoing experiments are focused on elucidating the mechanisms of disease-related impairments of neurogenesis in a mouse model of RTT. Our data highly supports the hypothesis that the expression of a key miRNA family is altered in a human model of RTT and has significant effects on downstream signaling pathways implicated in RTT pathogenesis

J. SHARMA, R. LANDMAN, M. SUR, R. DESIMONE,
Emotional distracters and emotional state both influence spatial attention in macaque monkeys
Soc. Neurosci., 2014.

Faces with emotional expression, even when irrelevant to the ongoing task, are known to act as potent distracters. The influence of affective stimuli is further modified by the emotional state. However it remains debatable whether this process is automatic and reflexive or dependent on the duration of exposure. Here we varied emotional content in face pictures used as distracters to test their influence on behavioral performance on a spatial attention task in monkeys. We sought to vary monkeys”™ emotional state by intranasal administration of the neuropeptides, Oxytocin (OT) and Vasopressin (VP). OT and VP are key emotional regulators implicated in producing opposing effect on anxiety and social stress. The task required monkeys to monitor one of two gratings for a subtle color change and were rewarded for making a saccade to target grating when the change occurred. Two irrelevant images of monkey faces with threatening, fearful or neutral expression appeared between fixation-spot and gratings. Presence of affective faces for duration greater than 200 ms significantly reduced accuracy (d”™) and shortened reaction time (RT). Both OT and VP slowed RT and improved accuracy in trials with emotional faces, however in VP trials this effect was less dependent on exposure duration, particularly in presence of threatening faces. We also tested whether monkeys looking behavior showed particular bias towards emotional expression in a free choice paradigm. When presented with faces of conspecifics with or without emotional expression, in 2/3rds of the trials, they made the first saccade towards neutral face, but on average spent significantly greater time scanning the threatening or fearful faces. When all three expressions were presented, they chose to scan threatening faces for far longer duration. Thus while early saccades showed tendency for active avoidance, persistent presence of emotional images required redirection of attentional resources, possibly an adaptive mechanism for active scanning of the perceived threat.

K. MERGENTHALER, D. ROY, C. RUNYAN, J. PETRAVICZ, M. SUR, K. OBERMAYER,
A class of computational models for orientation selectivity in mouse primary visual cortex
Soc. Neurosci., 2014.

Neurons in primary visual cortex of mouse respond selective to oriented gratings. Several recent experimental studies investigated orientation selectivity for firing rates of pyramidal cells and interneurons [1], or for subthreshold membrane potential, excitatory and inhibitory conductances of pyramidal cells [2,3]. In addition several structural properties in primary visual cortex of mouse were identified: salt-and-pepper organization, preferred orientation specific sparce connections between pyramidal cells, distinct afferent tuning, subpopulations of interneurons. However, how these structural properties comprise orientation tuning in mouse is unclear. Here we develop a computational model which investigates how salt-and-pepper organization, specific lateral connections and variations in afferent tuning width shape orientation selectivity. It uses laterally connected Hodgkin-Huxley type point neurons projecting either via excitatory or inhibitory synapses to nearby neighboring neurons organized on a 2D salt-and-pepper map. Neurons are stimulated by afferent inputs representing differently oriented gratings. It is evaluated in contrast to other models not only on means and variances but on the entire experimentally observed distributions (2-way Kolmogorov-Smirnov-test) and can assess how far experimental data [1,2,3] constrains the model with respect to structural elements. It turned out that models which only consider salt-and-pepper organization as well as models with additional lateral specific connections between pyramidal cells generate significantly different distributions and are therefore rejected. If afferent tuning is drawn from specific unimodal distributions for excitatory and inhibitory cells organized in a salt-and-pepper map the model becomes rich enough to explain the data distributions. However, as distributions across the experimental datasets can be distinguished different parameterizations are required. A common feature identified for non-rejected models are strong lateral connections which dominate afferent input strength. [1] C. Runyan & M. Sur, 3[28],11724ff, J. Neurosci, 2013 [2] A. Tan et al., 31[34],12339ff, J. Neurosci, 2011 [3] B. Liu et al., 71[3],542ff, Neuron, 2011

Soc. Neurosci., 2014.

J.C. PETRAVICZ, N. MELLIOS, S. EL-BOUSTANI, C. LE, M. SUR,
Role of astrocyte glutamate transporters in ocular dominance plasticity and response properties of visual cortex
Soc. Neurosci., 2014.

Monocular deprivation (MD) during a critical period induces ocular dominance plasticity of responses in primary visual cortex (V1), by reducing responses from the closed eye and increasing responses from the open eye. The mechanism underlying this scaling of responses in OD plasticity involve synaptic potentiation and depression, both of which have been shown to be influenced by glutamate transporter activity. Further, astrocyte glutamate transporter expression is developmentally regulated and can be dynamically tuned in a neuronal activity-dependent manner. Astrocytic Glt-1 is responsible for the majority of glutamate reuptake in the cortex, and blockade of glutamate transport induces changes in cortical transmission and in vivo processing. Monocular deprivation may affect the expression and function of astrocyte glutamate transporters, as a basis for plastic changes in the visual cortex. We have used a combination of molecular and in vivo recording techniques, as well as transgenic mouse models, to begin to investigate how reduced glutamate clearance in the visual cortex affects the response properties associated with OD plasticity and the expression of glutamate transporters. We find that with both 4 days and 7 days of MD there is an alteration in Glt-1 expression in the binocular region of V1. Glt-1 show dynamic changes in both gene and protein expression levels, as measured by mRNA and western blot analysis. Additionally, using in vivo optical imaging we find a significant reduction in contralateral eye dominance in V1 of Glt-1 Het mice, prior to MD. Contralateral eye MD for 4 days produces a significant shift towards ipsilateral eye responses; however this shift is absent with 7 days of MD. Examining cell-specific responses using in vivo two-photon imaging of calcium responses and targeted recordings, we find that reduced glutamate clearance alters excitatory and inhibitory cell firing rates differentially, resulting in alteration to orientation tuning of V1 neurons. These findings indicate that a critical threshold of glutamate reuptake exists for the establishment of contralateral bias in the binocular region of the visual cortex and for ocular dominance plasticity. Furthermore, our findings reveal a novel regulation of both inhibitory and excitatory cell response properties that is reliant on synaptic glutamate reuptake by astrocytes. Our results provide new insights into how astrocyte glutamate reuptake influences development of synapses, cortical plasticity and response properties of intact visual cortex circuits.

Soc. Neurosci., 2014.

2013

A. BANERJEE, J. CASTRO, C. RUNYAN, M. SUR,
Role of Cortical Interneuron Subtypes in Rett syndrome
Soc. Neurosci., 2013.

Maturation of Parvalbumin (PV+) and Somatostatin (SOM+)-expressing interneuron-mediated inhibition plays an important role in shaping cortical plasticity and directly affects integration of sensory inputs and excitatory-inhibitory balance – features that are disturbed in Rett syndrome (RTT). RTT is a neurological disorder of genetic origin, caused by mutations in the X-linked gene methyl-CpG binding protein 2 (MeCP2). Effects of global loss of MeCP2 from forebrain excitatory and inhibitory neuronal population has previously been studied, however, little is known about how MeCP2 regulates the development and plasticity of specific subpopulation of GABAergic inhibitory circuits and how they are altered in RTT. To identify cortical circuit abnormalities that are specifically associated with MeCP2 deficiency, we have analysed the role of PV+ and SOM+ expressing inhibitory interneurons in neural circuit plasticity and visual information processing, taking mouse visual cortex (V1) as a model system.

To study the role of MeCP2 in experience-dependent inhibitory circuit maturation and OD plasticity, we specifically removed MeCP2 from PV+ and SOM+ interneurons and compared with global MeCP2 KO Bird mice. Using optical imaging of intrinsic signals, ipsi- and contralateral eye-specific responses were recorded from layer 2/3 of V1 during postnatal day (P) 28-30. Deletion of MeCP2 did not significantly alter OD plasticity at P28, upon 3-4 days of monocular deprivation. Optogenetic activation of MeCP2-deficient PV+ and SOM+ cells while in vivo loose-patch recording from their target neurons revealed no change in their inhibition profile. Using two-photon guided targeted recordings from RFP+ MeCP2 deleted PV+ and SOM+ interneurons in layer 2/3 of mouse V1, we studied visual response properties of these interneurons and found their response properties are significantly altered compared to control PV+ and SOM+ cells.

Overall, we found that loss of MeCP2 in specific subsets of interneurons does not significantly alter plasticity during critical periods of mouse visual cortex development. Their inhibitory synaptic input also remains unaffected, whereas, there is a significant alteration of excitatory visual drive onto these interneuron subtypes that is sufficient to induce a pathological synaptic defect in RTT mice. Taken together, these observations imply profound abnormalities in selective cortical circuits due to MeCP2 deficiency.

Soc. Neurosci., 2013.

Caggiano V., Sur M., Bizzi E.,
Optogenetic rostro-caudal inhibition of movement without inhibition of afference in the spinal cord of mice.
Soc. Neurosci., 2013.

The γ-aminobutyric acid (GABA) and glycine are the main inhibitory neurotransmitters in the adult mammalian spinal cord. Several line of research have supported the idea that inhibitory interneurons have a local organization defined by sensory terminals which contribute to control motor neurons for limb alternation/coordination and for regulation of sensory-motor reflexes.
In order to understand the role and the organization of inhibitory neurons in the control of motor and sensory information within the spinal cord with both higher precision in space and in time, we developed a new method to optogenetically manipulate the spinal circuits in anesthetized and in awake freely moving animals using transgenic mice expressing ChR2 (Channelrhodopsine-2) in GABAergic/glycinergic neuronal populations (VGAT-mhChR2).
In anesthetized animals, light was shined at different levels of the spinal cord (from middle thoracic to middle lumbar with about 0.5 – 2 mm steps) by means of a movable optical fiber. Movements produced by electrical stimulation of the motor cortex were strongly reduced when light stimulation was applied (about 90% reduction, p < 0.05 sign-rank test, n = 10 mice). Maximum suppression was obtained at the middle-lower thoracic sector. These observation were confirmed by testing motor behavior in freely moving animals when light was shined above T12. Brief periods of optical stimulation of the spinal cord produced a loss of muscular tone (median latency of EMG suppression 6.55 ms) in all of the muscles i.e. agonist and antagonist, caudal to the stimulated point.
Some of the animals from the previous experiment were tested to evaluate the sensory consequences of the inhibition of the motor behavior. Once we identified the muscle(s) suppressed during light stimulation, we anesthetized the animal and we recorded single neuron activity in the somatosensory cortex evoked by electrical stimulation of one of the muscles affected. In addition, stimulation of the hind-paw/foot was performed. Overall, the cortical sensory responses evoked by peripheral electrical stimulation showed the same pattern (time and amplitude) with only electrical and electrical-couple-optical stimulation of the spinal cord.
In conclusion, we developed a new technique for investigating the spinal inhibitory network in both anesthetized and awake freely moving mice. These experiments showed a new global rostrocaudal inhibitory effect without modifying the sensory information transmitted to the cortex at the middle/lower thoracic level in addition to the well-established local inhibition for postural control and locomotion which is present at the lumbar and sacral levels.

Soc. Neurosci., 2013.

N. CHEN, H. SUGIHARA, M. SUR,
Cholinergic activation of somatostatin-expressing interneurons enhances cortical processing and changes brain states
Soc. Neurosci., 2013.

Cholinergic activation of the cortex has been shown to modulate brain function at multiple timescales. This includes neuromodulation of information processing through decorrelation of neuronal responses, cortical plasticity, attention and arousal. To understand the physiological basis of cholinergic modulation of these functions, it is critical to identify the cortical circuit elements involved and define how their interactions can contribute to the cortical network dynamics. We have shown in previous work that cholinergic activation of astrocytes and their facilitatory influences on excitatory neurons in visual cortex are crucial for the induction of stimulus-specific plasticity over prolonged timescales of minutes (Chen N, Sugihara H et al., PNAS 2012). In the present work we aimed to dissect the circuit involved in cholinergic modulation of cortical sensory processing, a phenomenon that occurs on a rapid timescale. To induce endogenous release of acetylcholine (ACh) in vivo, we performed photostimulation of cholinergic axons in the primary visual cortex of ChAT-ChR2 animals. During cholinergic activation, we observed a dynamic change in brain state through robust desynchronization of local field potential (LFP) as well as changes in neuronal population visual responses through in between cell decorrelation. Cell-attached recordings further revealed that cholinergic axon activation evoked facilitatory responses in somatostatin-expressing interneurons (SOM), in contrast to the suppressive responses in parvalbumin expressing interneurons (PV) and absence of change in excitatory neurons (Exc) in vivo. The mechanisms were dissected in visual cortical slices where we observed that ACh evoked transient action potentials in SOM neurons and IPSCs in both PV and Exc neurons. With selective expression of Arch in SOM neurons in slices, we further demonstrated that blocking cholinergic activation of SOM neurons abolished the IPSCs in both PV and Exc neurons. Finally, when cholinergic activation of SOM neurons was blocked using Arch in vivo, we observed that the LFP desynchronization and decorrelation were both absent during ChAT axon stimulation. Collectively, these findings reveal the powerful role of SOM neurons in dynamically changing both brain states and cortical information processing during cholinergic modulation.

Soc. Neurosci., 2013.

D.A. FELDMAN, N. MELLIOS, S.D. SHERIDAN, S. KWOK, B. CRAWFORD, V. DANI, C. LE, S. HAGGARTY, M. SUR,
An isogenic human induced pluripotent stem cell model of Rett Syndrome reveals novel alterations in microRNA processing and downstream molecular targets
Soc. Neurosci., 2013.

Rett Syndrome (RTT) is an X-linked monogenic form of autism spectrum disorder that is predominantly caused by mutations in methyl CpG­-binding protein 2 (MECP2). Clinical features of RTT include a period of normal development lasting up to 6-18 months followed by stagnation of both neurological and general growth development. As this is a disorder of early development, it has proven difficult to elucidate phenotypes and/or molecular signatures at a pre-symptomatic stage. We have generated human induced pluripotent stem cell (iPSC) lines in which to study molecular and functional phenotypes in vitro at various stages of neural development. By utilizing X-inactivation, we were able to generate isogenic cell lines that originate from the same patient. Additionally, we knocked down MECP2 expression in control cell lines using viral shRNA constructs. Neural progenitors were generated from iPSCs via dual-SMAD inhibition and subsequently differentiated into cortical neurons, which were capable of firing action potentials and exhibited spontaneous synaptic activity as observed via whole­-cell patch clamp and functional calcium imaging. In this study, we aimed to identify key miRNA and molecular signatures at early stages of neuronal differentiation and in neural progenitors using these isogenic patient-derived and MECP2 knockdown cell lines. We screened for affected miRNAs and identified one miRNA family that is robustly augmented in RTT at both time points in patient-derived cell lines, as well as in control cell lines following MECP2 knockdown via shRNA. Measurements of primary miRNA precursor (pri-miRNA) expression revealed a patient-specific alteration in pri-miRNA processing. In parallel, components of the mitogen-activated protein kinase (MAPK) pathway that are known targets of the upregulated miRNA family were found to be misregulated in RTT lines. Ongoing experiments are focused on elucidating the molecular mechanisms of disease-related changes in miRNA processing as well as determining the effects of MAPK deficiencies on brain­derived neurotrophic factor (BDNF) expression. As such, we have shown that MeCP2-­mediated miRNA processing is altered in a human model of RTT and has significant effects on downstream signaling pathways implicated in RTT pathogenesis.

Soc. Neurosci., 2013.

R.I. GARCIA, M. GOARD, J. PETRAVICZ, M. SUR,
Awake in vivo Ca2+ responses in astrocytic processes of mouse visual cortex.
Soc. Neurosci., 2013.

Astrocytic intracellular Ca2+ signaling has become a prominent feature in neuronal-glial interactions. The majority of data concerning astrocyte Ca2+ signaling come from either culture or in situ brain slices, approaches that rely on electrical stimulation or pharmacological methods to examine the spatial and temporal coding of astrocyte Ca2+ signals. Recently, several studies have utilized in vivo Ca2+ imaging in response to physiologically relevant stimuli or combined with electrical stimulation of brain nuclei to examine the role of astrocyte Ca2+ transients in intact circuits. Further, in situ data shows that localized Ca2+ elevations in distal processes of astrocytes occur at a higher frequency than somatic increases. This is difficult to examine in vivo; bulk loading of SR101 and OGB require anesthesia and allow for imaging primarily of somatic response, with limited detection of Ca2+ activity in processes. Additionally, anesthesia has been shown to influence both neuronal and astrocytic activity, which may alter the spatial and temporal coding of Ca2+ transients in response to stimuli. We are currently investigating visually evoked Ca2+ responses in visual cortex astrocytes of an awake, head-fixed animal using two-photon microscopy. To achieve this, we have generated a new astrocyte reporter line that expresses tdTomato in cortical astrocytes driven by the shortened human GFAP promoter, together with viral mediated delivery of a membrane-bound genetically encoded Ca2+ indicator, Lck-GCamp5G, to specifically target cortical astrocytes. We present evidence that Ca2+ transients in distal processes of cortical astrocytes are more frequent than has been observed for anesthetized preparation, with variable relationship to somatic responses. Furthermore, we are able to identify structurally identifiable regions of distal processes from single astrocytes that are responsive to visual stimuli and display orientation tuning. The combination of these technologies will allow us to further explore the functional role of astrocytes in the primary visual cortical circuit.

Soc. Neurosci., 2013.

R. LANDMAN, J. SHARMA, M. SUR, R. DESIMONE,
The effect of emotional distractors on spatial attention in macaque monkeys
Soc. Neurosci., 2013.

Visual attention involves selection of what is relevant while ignoring what is irrelevant. However, some irrelevant stimuli are harder to ignore than others. In primates, social and emotional stimuli such as faces of conspecifics may attract attention even when “irrelevant” to the task at hand. The amount of resources directed to those stimuli and whether they attract or repel direct gaze can be measured by the way in which they affect a primary task. Here we varied emotional content in face pictures used as distracters to test their effect on behavioral performance on a spatial attention task in monkeys. In addition we sought to vary monkeys”™ sensitivity using intranasal administration of the hormone Oxytocin. Two monkeys were trained to perform an attention task in which they monitored one of two gratings for a subtle color change. The monkeys were rewarded for making a saccade to the target grating when it changed. At random times before the target distractor changes (50, 200 or 500ms before the change), two irrelevant images would appear. One of the images could be an image of a monkey face, with a neutral, threatening or a fearful facial expression. The images were located between the fixation spot and the gratings. Reaction times, accuracy and eye movements were analyzed. We find that face images negatively affected accuracy while decreasing reaction time. There was a significant effect of image duration on accuracy with lower accuracy for larger image duration (ANOVA F(3,16)=75.79, p<0.001). Among trials with faces, reaction time was slower when the face was located near the target grating than when it was located near the distractor grating (Kruskal-Wallis test H(1)=5.26, p=0.02). Accuracy was more strongly affected when the facial expression was fearful or threatening than when the expression was neutral. Intranasal Oxytocin slowed reaction times while improving the accuracy in trials with emotional faces. The data support the notion that emotional distractors compete with learned attentional tasks even when they are irrelevant, most likely because they draw away available attentional resources. Oxytocin reduces these effects in line with the prevalent view that it reduces social fear.

Soc. Neurosci., 2013.

N. MELLIOS, S. SHERIDAN, S. KWOK, D. FELDMAN, B. CRAWFORD, J. WOODSON, S. HAGGARTY, M. SUR,
Novel microRNA-mediated mechanisms regulate brain growth factor expression in Rett Syndrome – Implications for therapeutics
Soc. Neurosci., 2013.

Rett Syndrome is a debilitating childhood-onset neurodevelopmental disorder that is predominantly caused by mutations in methyl-CpG-binding protein 2 (MEPC2). Using the MeCP2 knockout (KO) mouse model of the disease we uncovered a novel miRNA-mediated molecular pathway that bridges the observed alterations in Brain-derived neurotrophic factor (BDNF) and Insulin-like growth factor 1 (IGF1) expression in the brain of MeCP2 KO mice. Importantly, chronic treatment with a β-2 adrenergic receptor agonist completely normalized the expression of the components of the affected molecular pathway in the cerebellum of Mecp2 KO mice, and resulted in greatly increased survival, improved respiratory function, social recognition, and motor coordination; all cardinal symptoms of Rett syndrome. Notably, coadministration of the same β-2 adrenergic receptor agonist with recombinant human IGF1, further ameliorated the phenotype of MeCP2 KO mice, and resulted in a notable increase in survival. In parallel we used patient-derived induced pluripotent stem cells (IPSCs) to screen more effectively for miRNAs that are affected in Rett Syndrome. Our results from IPSC-derived neuronal cultures revealed among others an additional miRNA family that is robustly increased in two different patient-derived samples, at two different developmental stages, and following viral-mediated knockdown of WT samples. Protein and RNA expression analysis uncovered altered levels of known targets of the affected miRNA family, which are upstream regulators of BDNF expression. In summary we show using both a mouse and IPSC-model of Rett syndrome that a subset of Mecp2-regulated miRNAs are important effectors of complex regulatory networks related to brain growth factor expression, and reveal novel therapeutic alternatives for the treatment of Rett syndrome.

Soc. Neurosci., 2013.

Mergenthaler K., Jungnickel E., Petravicz J., Roy D., Sur M., Obermayer K.,
Stimulus driven calcium response in astrocytes is likely to depend on concurrent sodium co-transport by glutamate transporters and glutamate receptor activation: A computational study.
Soc. Neurosci., 2013.

Astrocytes play an important role in controlling extracellular glutamate concentration in cortex. In an in-vivo experiment in ferrets Schummers et al. [1] investigated simultaneously the calcium-response to oriented gratings in neurons and astrocytes in primary visual cortex under normal condition and with impeded astrocytic glutamate uptake. Under normal condition neurons and nearby astrocytes displayed prominent orientation tuning with same preferred orientation. When astrocytic glutamate transporters (GluT) are blocked using TBOA the neuronal response was increased and broadened, suggesting a higher efficacy of glutamate. However the calcium response in astrocytes was absent, which contradicts earlier computational models [2] of astrocytic calcium dynamics driven by receptors activated by extracellular glutamate. Here we present a computational model for the astrocyte calcium dynamics which reproduces the suppressed calcium response with blocked glutamate transporters. It extends the model by De Pitta et al [2] by incorporating GluTs and a model for the sodium-calcium exchanger (NCX) [3]. Furthermore, it requires an explicit description of astrocytic sodium concentration. The astrocyte model is driven by realistic glutamate dynamics as generated during synaptic transmission [4] as response to visual stimulation. It reproduces the magnitude of observed change as well as the observed temporal calcium dynamics. It furthermore provides insight into the mechanism underlying the calcium response. It turns out to elicit a strong response under normal conditions concurrent activation of metabotropic glutamate receptors and co-transport of sodium by GluTs is required. Without the sodium co-transport by GluT the NCX would rapidly reduce the internal calcium concentration disabling positive feedback mechanisms in calcium and IP3-concentrations and therefore hinders the built-up of a strong calcium response.

[1] J. Schummers, H. Yu, & M. Sur, 320,1638ff, Science, 2008
[2] M. De Pitta, M Goldberg, V Volman, H Berry, & E Ben-Jacob, 35, 383ff, J. Biol. Phys., 2009
[3] C Luo & Y Rudy, 74(6), 1071ff, Circulation Research, 1994
[4] J. Diamond, 15, 2906ff, J. Neurosci., 2005

Soc. Neurosci., 2013.

G. PEREA, A. YANG, E. S. BOYDEN, M. SUR,
Astrocytes regulate synaptic information in visual cortex
Soc. Neurosci., 2013.

Astrocytes play important roles in synaptic transmission and plasticity. Despite in vitro evidence, their contribution to cortical network activity and sensory information processing in vivo remains unresolved. Here, we use optogenetic tools to selectively stimulate astrocytes and investigate their consequences on neuronal activity in primary visual cortex (V1). Astrocytes were selectively targeted with adeno-associated viral vector with light-sensitive channelrhodopsin-2 (ChR2) under GFAP promoter and neuronal activity was recorded from layer 2/3 neurons of V1. Photostimulation of astrocytes in vivo increases the spontaneous firing rate of parvalbumin-positive (PV+) inhibitory neurons and excitatory neurons, but does not modify the spontaneous firing rate of somatostatine-positive (SOM+) inhibitory interneurons. Moreover, PV+ neurons show increased baseline visual responses and reduced orientation selectivity to visual stimuli, whereas excitatory neurons show either increased or decreased baseline visual responses together with complementary changes in orientation selectivity. On the other hand, unlike PV+ neurons, SOM+ interneurons show either increased or decreased baseline visual responses, with changes in orientation selectivity, similar to excitatory neurons. Excitatory and inhibitory drive was measured in V1 slices by recording whole-cell currents in excitatory neurons, PV+ and SOM+ interneurons. Optogenetic activation of astrocytes evoked a transient potentiation of spontaneous excitatory postsynaptic currents (sEPSCs) and miniature excitatory postsynaptic currents (mEPSCs), increasing the frequency of the synaptic currents but not their amplitude, that was induced by stimulation of metabotropic glutamate receptors (mGluRs) at presynaptic terminals. Therefore, astrocyte activation, through the dual control of excitatory drive and of inhibitory drive, modulates excitation/inhibition balance in cortical circuits, and thus influences integrative response features of neurons critical for sensory information processing.

Soc. Neurosci., 2013.

R. V. RIKHYE, M. SUR,
Feature-specific processing of natural information in mouse primary visual cortex
Soc. Neurosci., 2013.

Classical theories of sensory processing propose that natural images are efficiently represented in the visual cortex due to their novel statistical structure. A predominant statistical feature of natural images is their unique pattern of spatial correlations, which results in a 1/k Fourier amplitude spectrum and a unique phase structure. In particular, edges arise from the alignment of phase of different spatial frequency (SF) components. However, how these edges are extracted from natural images and processed in V1 remains unclear. The aim of this work is to understand how populations of neurons in V1 represent the unique statistical features of natural images. To this end, we performed two-photon calcium imaging in anesthetized mice while presenting an ensemble of natural movies that had specific perturbations to their SF and phase spectra. This method allowed us to simultaneously monitor the activities of hundreds of neurons in layer 2/3 with single cell resolution. Using a Canny edge detector, we selectively removed all SF and pixel intensity information while preserving the edge structure of natural movies. We find that these edge movies evoked reliable and synchronized responses from a large fraction of the network. Using a nearest means classifier, we were not able to distinguish between responses to different edge movies. Edge movies also had significantly higher signal but not noise correlations compared to raw natural movies. This suggests that completely removing SF information alters functional connectivity patterns and impairs the ability of neurons to extract orientation-specific information. Next, we used a band-pass SF filter to selectively alter the SF spectra and a Gabor filter to enhance certain orientation components. We find that responses to both Gabor and SF filtered movies were spatially and temporally sparse. Furthermore, neurons remained highly selective to the orientation structure of the original movie. Finally, we compared responses to phase-randomized movies and synthetic stimuli with different amplitude spectra. These movies had significantly decorrelated population codes, implying that different amplitude spectra are also differentially processed by neurons in V1.

Our results show that changing the phase structure of images markedly changes the response dynamics of a population of V1 neurons. Thus, coding strategies in V1 change depending on the statistics of the stimulus. In summary, the processing of orientation specific information from natural images is a nonlinear computation that appears to critically depend on the local image statistics such as contrast, luminance and spatial correlations.

Soc. Neurosci., 2013.

Garcia R.I., Goard M., Petravicz J., Sur M.,
Awake in vivo Ca2+responses in astrocytic processes of mouse visual cortex.
Gordon Research Conference on Glial Biology, Ventura, CA, 2013.

Astrocytic intracellular Ca2+ signaling has become a prominent feature in neuronal-glial interactions. The majority of data concerning astrocyte Ca2+ calcium signaling comes from either culture or in situ brain slices, an approach that relies on electrical stimulation or pharmacological methods to examine the spatial and temporal coding of astrocyte Ca2+ calcium signals. Recently, several studies have utilized in vivo Ca2+ calcium imaging in response to physiologically relevant stimuli or combined with electrical stimulation of brain nuclei to examine the role of astrocyte Ca2+ calcium imaging transients in intact circuits. Further, in situ data shows that localized Ca2+ elevations in distal processes of astrocytes occur at a higher frequency than somatic increases. This is difficult to examine in vivo; bulk loading of SR101 and OGB require anesthesia and allow for imaging primarily of somatic response, with limited detection of Ca2+ transient activity in processes. Additionally, anesthesia has been shown to influence both neuronal and astrocytic activity, which may alter the spatial and temporal coding of Ca2+ calcium transients in response to stimuli. We are currently investigating visually evoked Ca2+ responses in cortical astrocytes in the visual cortex astrocytes of an awake, head-fixed animal using two-photon microscopy. To achieve this, we have generated a new astrocyte reporter line that expresses tdTomato in cortical astrocytes driven by the shortened human GFAP promoter, together with viral mediated delivery of a membrane-bound genetically encoded Ca2+ indicator, Lck-GCamp5G, to specifically target cortical astrocytes. We present evidence that Ca2+ transients in distal processes of cortical astrocytes are more frequent than has been reported for anesthetized preparation, and are not correlated to somatic responses. Furthermore, we are able to identify structurally identifiable regions of distal processes from single astrocytes that are responsive to visual stimulus and display orientation tuning; continued experiments will determine if this will reflect the tuning of surrounding neurons. The combination of these technologies will allow us to further explore the functional role of astrocytes in the primary visual cortical circuit.

Gordon Research Conference on Glial Biology, Ventura, CA, 2013.

Petravicz J, Mellios N, Le C, Sur M,
Role of Astrocyte Glutamate Transporters in Ocular Dominance Plasticity
Gordon Research Conference on Glial Biology, Ventura, CA, 2013.

Monocular deprivation (MD) during a critical period induces ocular dominance plasticity of V1 responses, by reducing responses from the closed eye and increasing responses from the open eye (1). The mechanism underlying this scaling of responses in OD plasticity involve synaptic potentiation and depression, both of which have been shown to be influenced by glutamate transporter activity (2, 3). Further, astrocyte glutamate transporter expression is developmentally regulated and can be dynamically tuned in a neuronal activity-dependent manner (3, 4). Astrocytic Glt-1 is responsible for the majority of glutamate reuptake in the cortex, and blockade of glutamate transport induces changes in cortical transmission and in vivo processing (5, 6). Monocular deprivation is a paradigm to measure plastic changes in the visual cortex, which may affect the expression and function of astrocyte glutamate transporters. We have used a combination of molecular and in vivo recording techniques, as well as transgenic mouse models, to begin to investigate how reduced glutamate clearance in the visual cortex affects the response properties associated with OD plasticity and the expression of glutamate transporters. We find that with both 4 days and 7 days of monocular deprivation there is an alteration in Glt-1 expression in the binocular region of the primary visual cortex. Glt-1 show dynamic changes in both gene and protein expression levels, as measured by mRNA and western blot analysis. Additionally, using in vivo optical imaging we find a significant reduction in contralateral eye dominance in Glt-1 Het mice with no monocular deprivation. Monocular deprivation for 4 days produces a significant shift towards ipsilateral eye responses; however this shift is absent with 7 days of MD. These findings indicate that a critical threshold of glutamate reuptake exists for the establishment of contralateral bias in the binocular region of the visual cortex and for ocular dominance plasticity. Our results provide novel insights into how astrocyte glutamate reuptake plays a role in cortical development and plasticity using the well-established model system of the primary visual cortex.
1. G. B. Smith, A. J. Heynen, M. F. Bear, Bidirectional synaptic mechanisms of ocular dominance plasticity in visual cortex. Philos Trans R Soc Lond B Biol Sci 364, 357 (Feb 12, 2009).
2. A. Filosa et al., Neuron-glia communication via EphA4/ephrin-A3 modulates LTP through glial glutamate transport. Nat Neurosci 12, 1285 (Oct, 2009).
3. A. Omrani et al., Up-regulation of GLT-1 severely impairs LTD at mossy fibre–CA3 synapses. J Physiol 587, 4575 (Oct 1, 2009).
4. Y. Yang et al., Presynaptic regulation of astroglial excitatory neurotransmitter transporter GLT1. Neuron 61, 880 (Mar 26, 2009).
5. J. Schummers, H. Yu, M. Sur, Tuned responses of astrocytes and their influence on hemodynamic signals in the visual cortex. Science 320, 1638 (Jun 20, 2008).

Gordon Research Conference on Glial Biology, Ventura, CA, 2013.

Konstantin Mergenthaler, Dipanjan Roy, Jeremy Petravicz, Mriganka Sur, Klaus Obermayer,
Changes in V1 orientation tuning when blocking astrocytic glutamate transporters: Models for extra- and intrasynaptic mechanisms
CNS Paris 2013, 2013.

Several recent studies reveal a close interplay of neurons and astrocytes in information processing [1]. Astrocytes affect neural transmission by their control of glutamate concentrations by glutamate transporters [2] in direct vicinity to the synaptic cleft as well as extrasynaptically. An in-vivo study in primary visual cortex of ferrets [3] demonstrated the impact of astrocytic glutamate transport on neuronal response by blocking glutamate uptake pharmacologically with TBOA. A severe effect on neuronal orientation tuning curves to a full-field stationary grating, tilted every second by additional 10 degrees, was observed. While glutamate transporters were blocked tuning curves were broadened (HWHM increased from 28 deg. to 39 deg.) and response at preferred orientation was enhanced. However, it is unclear if the change in response originates from a prolongation of synaptic glutamate clearance or locally increased ambient glutamate concentration. Here we investigate in a computational model of ferret V1 how the intra- and extrasynaptic mechanisms affect orientation tuning.
We implemented both mechanisms in a highly recurrent single layer 2d map model based on [4]. Neurons are placed on a 50 x 50 grid and connected to their neighbors by randomly drawn connections from a Gaussian. On every grid point an excitatory neuron is placed, additionally every third grid point is occupied by an inhibitory neuron. Peak conductivities for the four types of connections are set to be in the recurrent regime, exceeding the conductivity of the additional afferent input. The afferent input specific to each grid location induces a pinwheel-domain organization of neurons in the network. For the implementation of the synaptic mechanism the glutamate concentration follows a bi-exponential description with a decay time prolonged if glutamate uptake is reduced [2]. Synaptic glutamate then activates complex kinetic NMDA and AMPA-receptors [5]. The synaptic mechanism can however have a stronger influence on excitatory-to-excitatory or excitatory-to-inhibitory connections, as the different synapse geometries also affects glutamate clearance. Extrasynaptically ambient glutamate provides a constant NMDA-receptor mediated somatic current [6]. Here different sensitivities (different NMDA-receptor densities) of excitatory and inhibitory neurons to ambient glutamate may affects tuning. For both mechanism possible parameter combinations are assessed in a grid search.
We observed that a selective increase in modalities towards inhibitory neurons even leads to a sharpening of orientation tuning. While selective enhancement in the modalities towards excitatory neurons leads first to a drop in orientation tuning and very fast for further increase to pathological firing rates. For both mechanisms the closest fit in orientation tuning (HWHM) to the experimental observation with TBOA was found for a stronger effect on excitatory neurons along with a simultaneous but weaker effect on the inhibitory population. While both models can explain the current data, they, however, provide different predictions for sub-threshold properties and for neurons close to pinwheels or domain centers.
1. References
[1] De Pittà M, Volman V, Berry H, Parpura V, Volterra A, Ben-Jacob E, Computational quest for understanding the role of astrocyte signaling in synaptic transmission and plasticity. Frontiers in Computational Neuroscience, 2012, 6:1-25
[2] Diamond JS, Deriving the glutamate clearance time course from transporter currents in CA1 hippocampal astrocytes: transmitter uptake gets faster during development. Journal of neuroscience, 2005, 25:2906-2916
[3] Schummers J, Yu H, Sur M, Tuned responses of astrocytes and their influence on hemodynamic signals in the visual cortex. Science (New York, N.Y.) 2008, 320:1638-1643
[4] Stimberg M, Wimmer K, Martin R, Schwabe L, Mariño J, Schummers J, Lyon D, Sur M, Obermayer K, The Operating Regime of Local Computations in Primary Visual Cortex. Cerebral Cortex, 2009, 19:2166-2180
[5] Destexhe A, Mainen ZF, Sejnowski TJ, Kinetic models of synaptic transmission, Eds.: Koch C, Segev I, Methods in neuronal modeling, MIT Press, 1998, 1-25
[6] Bentzen NCK, Zhabotinsky AM, Laugesen JL, Modeling of glutamate-induced dynamical patterns. International journal of neural systems, 2009, 19:395″“407

CNS Paris 2013, 2013.

2012

K. MERGENTHALER, D. ROY, J. PETRAVICZ, M. SUR, K. OBERMAYER,
Synaptic and extrasynaptic influence of astrocytic glutamate uptake on orientation selectivity in primary visual cortex
Bernstein Conference, Munich, Germany, 2012.

One of the most prominent response features in primary visual cortex is orientation selectivty. In an in-vivo study in Ferrets Schummers et al. [1] investigated the importance of rapid glutamate uptake by glutamate tansporters situated on the astrocytic membrane in close vicinity to the synpatic cleft for sharp orientation tuning. With TBOA they blocked the astrocytic glutamate uptake and observed an increased response for preferred orientation along with broadend orientation tuning. The experiment, however, was unable to determine wether the broadend orientation tuning derives from changes in synaptic transmission via slowed down gluamate clearance or if the underlying mechanism is a change in neuronal excitability due to increased amibent glutamate concentration. In the computational study presented here we investigate which of the two possible mechanisms better explains the observed data. The synaptic as well as the somatic mechanism are incorporated independently in a recurrent 2d-model for ferret primary visual cortex [2]. The model consists of two populations, one excitatory and one inhibitory, of Hodgkin-Huxley point neurons. The total recurrent input received by these cells are stronger than the total afferent input, which leaves the network poised to a balanced regime. In the 2d-model orientation tuning is organized in a pinwheel-domain structure. The somatic mechanism is incorporated as an additional gluamate driven receptor current. Here the level of ambient glutamate represents how strong gluamate uptake is reduced. For the synaptic mechanism, synaptic NMDA and AMPA receptors were described by detailed models with several conformational receptor stages [3]. Theses receptors are driven by a bi-exponential glutamate concentration. Its decay constant is modified by the astrocytic gluamate uptake and decay is prolonged in the presence of TBOA [4]. Motivated by the significant difference in synaptic geometry for excitatory to excitatory and excitatory to inhibitory synapses as well as the difference in glutamate transporter density in the vicinity of these two types of synapses, we selectively vary the glutamate decay constant along these two types of connections. We find that both mechanism can achieve an increase in the preferred response along with changes in orientation tuning. However, the synaptic mechanism with the differentiated increase in decay times for both types of connections yields a better agreement with the experimental observations.
References
[1] J. Schummers, H. Yu, & M. Sur, 320,1638ff, Science, 2008
[2] M. Stimberg et al., 19,2166ff, Cerebr. Cort., 2009
[3] A. Destexhe et al.,Edt.: Koch & Segev, Methods in Neuronal Modeling (Book), 1998
[4] J. Diamond, 15,2906ff, J. Neurosci., 2005

Bernstein Conference, Munich, Germany, 2012.

S. EL BOUSTANI, M. SUR,
Role of parvalbumin-expressing interneurons in shaping visual discrimination and coding by mouse V1 pyramidal cells
Soc. Neurosci., 2012.

Neuronal circuits of the primary visual cortex (V1) encode multiple properties of the visual image. In particular,responses of V1 neurons to increase or decrease in light intensity in the visual field are encoded by two subfields -ON- and OFF – that together constitute the cells”™ receptive field. Cells with segregated ON- and OFF- subfields have been termed “Simple” due to their linear coding property whereas cells with a substantial overlap between these subfields have been named “Complex. It has recently been shown that simple cells display significant overlap between their ON- and OFF- subfields at the subthreshold level that does not manifest at the suprathreshold level due to inhibitory synaptic input. However, it remains unclear which interneuron subtypes participate to this process and what their specific impact is on pyramidal cells receptive field structure. To address these questions, we performed high-speed calcium imaging in V1 of PV-Cre mice injected with an adenoassociated virus containing a lox-flanked STOP segment followed by mCherry and Channelrhodopsin. Using a sparse noise analysis, we assessed the receptive field properties of multiple cells simultaneously with and without optogenetic activation of parvalbumin-expressing (PV+) interneurons. We found that increase in PV+ interneurons activity linearizes complex cells responses by reducing the overlap between ON- and OFFsubfields. Surprisingly, we also found in many cells that PV+ activation resulted in subfield-specific suppression, thus suggesting that PV+ interneurons can strongly affect sensory integration through localized projections targeting a pyramidal cell”™s dendritic tree. Overall, the reduction in receptive field size and overlap between ON- and OFF- fields increases the ability of neuron populations to discriminate between different visual scenes in a linear encoder scheme. We hypothesize that this inhibition-mediated mechanism homeostatically adapts V1 cells”™ coding properties to the ongoing visual statistics, in accordance with previous results reported in the cat visual cortex.

Soc. Neurosci., 2012.

N. CHEN, H. SUGIHARA, J. SHARMA, G. PEREA, J. PETRAVICZ, C. LE, M. SUR,
Astrocytes mediate nucleus basalis enabled, orientation stimulus specific plasticity in V1
Soc. Neurosci., 2012.

While sensory experience associated with nucleus basalis (NB) driven, cholinergic activation of the cortex has been shown to instruct cortical plasticity, little is known about the underlying cellular mechanisms. To understand how cortical responses and representations can be altered by experience during cholinergic modulation, it is critical to identify the circuit elements involved and elucidate how their interactions can alter cortical network dynamics. In this work, we show that electrical stimulation of NB, paired with visual stimulation in vivo, can induce significant potentiation of visual responses in excitatory neurons recorded by cell-attached patch in mouse primary visual cortex (V1). We further demonstrate with in vivo two-photon calcium imaging, ex vivo calcium imaging and whole-cell recordings that this pairing induced potentiation is mediated by direct cholinergic activation of V1 astrocytes via muscarinic acetylcholine receptors. The astrocyte-mediated potentiation is stimulus specific, as pairing NB stimulation with a specific visual orientation reveals a highly selective potentiation of responses to the paired orientation compared to unpaired orientations. Collectively, these findings reveal a novel and surprising role for astrocytes in NB-induced stimulus-specific plasticity in the cerebral cortex and suggest an intimate organization of astrocytes with the synapses that convey and generate orientation specific responses.

Soc. Neurosci., 2012.

N. R. WILSON, C. A. RUNYAN, F. L. WANG, M. SUR,
Two-way targeted optical dissection of cortical circuits during visual computations
Soc. Neurosci., 2012.

Brain circuits process information via specialized neuronal subclasses interacting within a network. Revealing their interplay requires activating specific cells while monitoring others in a functioning circuit. Here, using a new platform for two-way light-based circuit interrogation in visual cortex in vivo, we show the computational implications of modulating different subclasses of inhibitory neurons during sensory processing. While soma-targeting, parvalbumin-expressing interneurons (PV) principally divide responses (E/I) but preserve stimulus selectivity, dendrite-targeting, somatostatin-expressing inhibitory neurons (SOM) principally subtract from excitatory responses (E-I) and sharpen selectivity. Visualized in vivo cell-attached recordings demonstrate that division by PV neurons alters response gain whereas subtraction by SOM neurons shifts response levels. Finally, stimulating identified neurons while scanning many target cells reveals that single PV and SOM neurons functionally impact only specific subsets of neurons in their projection fields. These findings provide direct evidence that inhibitory neuronal subclasses have distinct and complementary roles in cortical computations. Further possibilities for circuit dissection via targeted two-way optical interrogation will be discussed.

Soc. Neurosci., 2012.

J. CASTRO, R. GARCIA, S. KWOK, J. PETRAVICZ, D. TROPEA, M. SUR,
Functional recovery with recombinant human IGF1 treatment in a mouse model of Rett Syndrome
Soc. Neurosci., 2012.

Rett Syndrome (RTT) is a neurodevelopmental disorder that primarily affects girls. It is characterized by stagnation of developmental milestones in communication, social and motor abilities as well as autonomic irregularities with episodes of palpitations and long apneas that can cause sudden death. Over 90% of cases have been linked to sporadic mutations in the gene coding for methyl-CpG binding protein 2 (MeCP2), a transcriptional factor modulator. Loss of functional Mecp2 leads to immature synaptic and neuronal development in the RTT brain but this has been proved to be reversible by genetic manipulation through reactivation of Mecp2 or overexpression of one of its target gene: Brain derived neurotrophic factor (BDNF). BDNF is known to be critical for neuronal and synaptic maturation exerting its influence via the signaling pathways of phosphoinositide 3-kinase (PI3K) and the extracellular signal-regulated kinase (ERK) which have been shown to be downregulated in the RTT mouse model. Another major activator of these pathways is insulin-like growth factor-1 (IGF-1). As opposed to BDNF, IGF-1 is actively transported through the blood brain barrier therefore making it a better therapeutic. Previous work from our lab demonstrated the ability of a cleaved form of IGF-1 (1-3) IGF1 to ameliorate some of the symptoms observed in a RTT model, but the full length recombinant human IGF-1 (rhIGF-1, Mecasermin) has been prescribed for more than 20 years in some growth disorder conditions, holding more potential as a clinical treatment. In this study we test for the first time the therapeutic effects of rhGF-1 in Mecp2 KO mice at an FDA approved dose to assess its translational applicability. We observed that increased concentration of rhIGF-1 and activation of its cognate receptor enhanced the signaling pathways PI3K and ERK. These pathways activation led to downstream effects on synaptic protein levels such as PSD95. In line with this result, synaptic strength and plasticity improved as demonstrated by optical imaging and whole-cell patch recording measurements. The enhancement in synaptic function explains the improvements in other metrics like lifespan, basal locomotor activity, as well as respiratory and cardiac functions and restoration of normal social and anxiety behavior. The provided data mechanistically explains and supports the therapeutic role of rhIGF-1 in preventing the organismal deterioration observed in non-treated RTT animals and the general improvement in the majority of symptoms associated with RTT.

Soc. Neurosci., 2012.

G. PEREA, A. YANG, B. CHOW, E. BOYDEN, M. SUR,
Astrocytes modulate synaptic information processing in visual cortex
Soc. Neurosci., 2012.

Astrocytes have recently been considered to play active roles in synaptic trans¬mission and plasticity in different brain areas. However, the impact of astrocyte activity on cortical networks and on information processing of external stimuli is still poorly known. Here, we use optogenetic tools to selectively stimulate astrocytes and investigate their consequences on neuronal activity in primary visual cortex. Astrocytes were selectively targeted with adeno-associated viral vector with lightsensitive channelrhodopsin-2 (ChR2) under GFAP promoter and neuronal activity was recorded from layer 2/3 neurons of V1. The selective expression and the ability of ChR2 to evoke Ca2+ responses in cortical astrocytes were confirmed in V1 slices. The analysis of neuronal activity showed that optogenetic activation of astrocytes evoked a transient potentiation of spontaneous postsynaptic currents, increasing the frequency of both excitatory and inhibitory currents without modifying their amplitudes. The increase of synaptic activity was prevented in the presence of MCPG, an antagonist of metabotropic glutamate receptors (mGluRs), suggesting that AAV-sGFAP-ChR2-expressed astrocytes can release glutamate that activates mGluRs at presynaptic terminals. Optogenetic astrocyte stimulation also evoked an increase of slow inward current (SIC) frequency, that was abolished by antagonists of NMDA glutamate receptors, indicating the ability of ChR2-expressed astrocytes to release glutamate. The functional significance of Ca2+ signaling in astrocytes was investigated in vivo. Cell-attached recordings from neurons in layer 2/3 were performed in anesthetized mice expressing ChR2 in V1 astrocytes. The ChR2-astrocyte stimulation evoked distinct changes in V1 network activity, including a range of cell-specific responses in layer 2/3 neurons along with a decrease in their spontaneous firing rate. Taken together, these results show that astrocytes, when activated through selective optogenetic stimulation, modulate synaptic transmission in V1 cortical neuronal networks, and can influence visual information processing in primary visual cortex.

Soc. Neurosci., 2012.

J. PETRAVICZ, C. LE, M. SUR,
Role of Astrocyte Glutamate Transporters in Ocular Dominance Plasticity
Cold Spring Harbor Glia Conference, 2012.

Monocular deprivation (MD) during a critical period induces ocular dominance plasticity of V1 responses, by reducing responses from the closed eye and increasing responses from the open eye. The mechanism underlying this scaling of responses in OD plasticity involve synaptic potentiation and depression, both of which have been shown to be influenced by glutamate transporter activity. Further, astrocyte glutamate transporter expression is developmentally regulated and can be dynamically tuned in a neuronal activity-dependent manner. MD induces plastic changes in the visual cortex, which may affect the expression and function of astrocyte glutamate transporters. We have used a combination of molecular and in vivo recording techniques, as well as transgenic mouse models, to begin to investigate how reduced glutamate clearance in the visual cortex affects the response properties associated with OD plasticity and the expression of glutamate transporters. We find that both 4 days and 7 days of monocular deprivation induces alterations in glutamate transporter expression in the binocular region of the primary visual cortex. Both GLAST and Glt-1 show dynamic changes in protein expression levels, as measured by proteomic and western blot analysis. Additionally, using the Glt-1 Het mouse model, we find alterations in the magnitude of the ocular dominance shift in the binocular region of V1 following 4 days and 7 days of MD, as measured by in vivo intrinsic optical imaging. These findings provide novel insights into how astrocyte glutamate reuptake plays a role in cortical development and plasticity using the well-established model system of the primary visual cortex.

Cold Spring Harbor Glia Conference, 2012.

I. NAGAKURA, A. VAN WART, D. TROPEA, B. CRAWFORD, M. SUR,
Novel role of STAT1 immune signaling in cortical plasticity and autism
Soc. Neurosci., 2012.

Emergent evidence suggests an involvement of immune signaling in idiopathic autism; e.g. proinflammatory cytokines such as TNF alpha and IFN gamma are upregulated in the brain and CSF of a subset of individuals with autism. IFN gamma mediates a variety of immune responses through activation of JAK/STAT signaling, particularly through STAT1 activation. In this study, we examined a role of STAT1 immune signaling in cortical plasticity that may contribute to autism pathology. Ocular dominance (OD) plasticity, referred to as the change in the strength of eye-specific inputs in the primary visual cortex (V1) following monocular deprivation (MD), was used as a model for activity-dependent cortical plasticity. Using in vivo optical imaging, we examined the intrinsic signal strength of the deprived (closed) vs. non-deprived (open) eye in the binocular region of V1 following short- (4 days) or long-term (7-8 days) MD. In critical period wild-type (WT) mice, we observed a decreased response from the closed eye following short-term MD, while long-term MD elicited a homeostatic increase in the open-eye response. In STAT1 knockout (KO) mice, the increase in the open-eye response was observed earlier following short-term MD, suggesting that STAT1 negatively regulates a homeostatic component of OD plasticity. The accelerated increase in the open-eye response was accompanied by increased AMPA receptor expression and cell surface trafficking, and was blocked by an inhibitor of TNF alpha signaling, suggesting an antagonistic interaction between STAT1 and TNF alpha signaling. Interestingly, STAT1 KO mice exhibited autism-like behaviors; they spent significantly less amount of time in the open arm of a plus maze compared to WT mice, suggesting an increased level of anxiety in these mice, and also showed impaired social novelty recognition in a three-chamber social test. These results indicate that abnormal cortical plasticity in STAT1 KO mice may underlie autism pathophysiology and suggest OD plasticity as a useful tool to understand the cellular and molecular mechanisms underlying autistic behaviors. We suggest that abnormal immune signaling involving STAT1 and TNF alpha may contribute to autism pathology.

Soc. Neurosci., 2012.

K. MERGENTHALER, J. PETRAVICZ, D. ROY, M. SUR, K. OBERMAYER,
Astrocytic glutamate transporters influence orientation selectivity: A computational study in primary visual cortex
Soc. Neurosci., 2012.

In primary visual cortex, one of the most prominent stimulus specific responses in neurons is orientation selectivity. Schummers et al. [1] investigated the impact of a drug (TBOA), which blocks the glutamate transporters on the astrocytic membrane, on orientation tuning in an in-vivo experiment in ferrets. They observed increased responses for preferred orientation along with broadened orientation tuning. However, it is unknown whether the effect derives from changes in the synaptic transmission via slowed-down glutamate clearance or from changes in neuronal excitability due to increased ambient glutamate concentration.
Here we investigate in a computational study which of the two mechanisms provides the better fit to the data. The synaptic and somatic mechanisms are incorporated independently in a recurrent network model of ferret primary visual cortex [2] which included the astrocytic components. The network consists of an excitatory and an inhibitory population of Hodgkin-Huxley point neurons spread on a 2d-grid, with orientation tuning organized in a pinwheel-domain structure. The somatic mechanism is described by an additional somatic glutamate dependent current. For the synaptic mechanism, detailed models with several conformational receptor stages [3] were incorporated for NMDA and AMPA receptors. Theses receptors are driven by a bi-exponential glutamate concentration, with a decay constant which is modified by astrocytic glutamate uptake and which increases by TBOA-concentration [4]. Furthermore, the decay constant is selectively varied for the excitatory to excitatory and excitatory to inhibitory synapses motivated by the their difference in geometry and receptor concentration. We find that both mechanism can achieve an increase in the preferred response along with changes in orientation tuning. However, the synaptic mechanism with the differentiated increase in decay times for both types of connections is in better agreement with the experimental observations.
[1] J. Schummers, H. Yu, & M. Sur, 320,1638ff, Science, 2008
[2] M. Stimberg et al., 19,2166ff, Cerebr. Cort., 2009
[3] A. Destexhe et al.,Edt.: Koch & Segev, Methods in Neuronal Modeling (Book), 1998
[4] J. Diamond, 15,2906ff, J. Neurosci., 2005

Soc. Neurosci., 2012.

D. ROY, Y. TJANDRA, K. MERGENTHALER, C. RUNYAN, N. WILSON, J. PETRAVICZ, M. SUR, K. OBERMAYER,
Influence of inhibitory subtypes in sensory information coding in a recurrent network model of mouse primary visual cortex
Soc. Neurosci., 2012.

Primary visual cortex (V1) is the first site in visual information processing which can provide crucial insights into the selectivity and emergence of specific output features such as orientation tuning. Several recent in-vivo experimental studies on mouse visual cortex have found that the inhibitory cells of all subtypes are broadly tuned for orientation [1, 2], contradicting the findings of many other studies in higher mammals and rodents, which have shown the existence of inhibitory neurons that are as sharply tuned as excitatory neurons [3, 4]. Two very critical questions naturally emerge as a result of these contradictory findings: (1) How similar is the generation of response properties, such as orientation selectivity compare with that in previously described species? (2) What is the network mechanism behind the sharpening of orientation selectivity in the mouse visual cortex? Here, we investigate the above questions in a computational framework with a recurrent network model of rodent primary visual cortex. This two-dimensional large scale Hodgkin-Huxley network model is adapted from a previously described map model in [5] and consists of an excitatory and inhibitory populations with orientation tuning organized in a “salt-and-pepper” manner. Further, we incorporate here differential input to inhibitory cells motivated from new experimental findings of differential output responses of soma-targeting subtypes [4]. Layer 2/3 excitatory cells are connected preferentially based on recent studies [2]. Network simulation reveals combined feedforward drive with precise fine scale lateral excitation and inhibition predicts a range of orientation tuning for both excitatory and inhibitory populations placed in layer 2/3 of primary visual cortex. In order to constrain the network parameters we estimate the p-values using Kolmogorov-Smirnov test (K-S test) over the entire range of recurrent excitation and inhibition values. Based on the estimated p-Values we infer that there are several points in different operational regimes of this network under sensory drive which commensurates well with several recent experimental observations. In particular, there are several points in the recurrent regime of this network which gives significant p-values, an operational regime, where network parameters most likely generate sharp orientation tuning particularly within orientation representations with diverse local neighborhoods.

J. SHARMA, H. SUGIHARA, M. SUR,
Time course of neuronal modulation with spatial and temporal attention follow distinct trajectories in monkey V1 and V4
Soc. Neurosci., 2012.

Attention in the temporal domain impacts behavioral responses leading to faster response times. Human studies have shown that expectation of an impending behavioral response is reflected in progressively shorter reaction times with increasing trial duration. In this ongoing series of experiments conducted in behaving monkeys, we are investigating whether neurons in the early visual pathway signal impending events in time that require a behavioral response. We have previously shown that V1 neurons in an attention task show robust late modulation that correlates with monkey”™s temporal expectation. Specifically, late attentional modulation of responses shows significant negative correlation with monkey”™s reaction time (RT). Here we extend our investigations by simultaneously recording from V1 and V4. The behavioral task required monkeys to maintain fixation on a central spot while attentively tracking a peripherally presented attention spot which stayed on for a variable period of time. The task consisted of two distinct epochs: an initial fixed time window in which the attention spot appeared and stayed on, and a subsequent variable time window, within which the attention spot disappeared with constant probability; thus, the conditional probability of attention spot disappearance increased with duration of the second epoch. Recordings were made from V1 and V4 neurons with largely overlapped RFs. As expected, V4 neurons showed robust early modulation, which was essentially sustained at the same level for entire duration, including fixed and variable epochs. In contrast, V1 had two distinct phases, an early modest modulation during the initial, fixed epoch that tapered-off within 500-600 ms, followed by a response which started approximately 150 ms before the second, variable epoch and showed a sharp, sustained rise till the attention spot disappeared and the monkey executed a behavioral response. The monkey”™s RT reduced with increasing duration of the variable epoch and showed significant negative correlation with attentional modulation of V1 responses. These changes were at best modest in V4 responses. Our results indicate that neurons in V1 show modulation within the same trial that signal the monkey”™s task dependent behavioral requirements, such as early spatial attention followed by time dependent changes in expectation. V4 neurons on the other hand mainly signal spatial attention that essentially needs to be sustained during the task. This distinction in V1 and V4 responses reveals remarkably distinct temporal dynamics of attention and behavioral expectation, and suggests the involvement of distinct task-dependent networks in the two areas.

2011

D. TROPEA, I. MOLINOS, E. PETIT, M. BAUM, A. AJETNMOBI, S. BELLINI, C. O’TUATHAIGH, K. MITCHELL, J. WADDINGTON, M. SUR , A. CORVIN
Circuit reorganization in disrupted in schizophrenia 1 (DISC1) mutant mice
Soc. Neurosci., 2011.

Neurodevelopmental disorders (NDDs), including autism and schizophrenia, are highly heritable conditions that affect brain function and lead to behavioral, social and cognitive impairments. Recent advances in our ability to study the human genome have prompted significant research breakthroughs in uncovering the risk genes for these disorders, nonetheless, the neurobiology of the candidate genes remains unclear. The function of these genes can be studied in vitro and in vivo using transgenic animal models. We hypothesize that in these mutants there is a major deficit in the activity-dependent reorganization of the neuronal network, and this has an effect on the functional dynamics of synaptic connections. One highly penetrant candidate genes in NDDs is Disrupted in schizophrenia 1 (DISC1), and several mice mutants for this gene have been generated. Among the available models, we studied the one with a missense variation that disrupts the interaction between DISC1 and PDE4B (phosphodiesterase 4B) which is also a risk factor for schizophrenia. In order to test whether DISC1 mutant mice are able to reorganize the connections in response to altered stimulation, we measured the synaptic properties of the circuitry in primary neuronal cultures and in the live animal. In vitro, we find that, induction of long term potentiation in DISC1 mutants fails to induce the synaptic strengthening observed in control WT (PSD95 relative immunostaining (PSD95 RI) wt=0.99 mean ± 0.08 se; wt_ltp=0.13 mean ± 0.113 se; PSD95 RI -/- = 1 mean ± 0.07 se; -/-ltp= 0.83 mean ± 0.08 se). For the analysis of the circuitry in vivo, we studied the reorganization of visual circuitry in response to altered stimulation (ocular dominance plasticity). We find that mice mutant for DISC1 have an abnormal response of the circuitry to sensory deprivation (Ocular Dominance Index (ODI) -/- = 0.075 mean ± 0.2 se; ODI -/- deprived = 0.065 mean ± 0.2 se), contrary to what observed for WT littermates (ODI WT= 0.27 mean ± 0.08 se; ODI WT deprived = -0.01 mean ± 0.1 se). These results suggest that DISC1 plays a role in activity-dependent plasticity.

Society for Neuroscience Abstract, 2011.

N.R. WILSON, C.A. RUNYAN, Y. DENG, F.L. WANG, M. SUR;
Parvalbumin-expressing interneurons implement dynamic gain control in cortical networks as assessed through all-optical interrogation
Soc. Neurosci., 2011.

Inhibition in cortical networks is administered by a diverse array of specialized cell types with distinct intrinsic properties and connectivity patterns. This stark heterogeneity in cellular form and function suggests that different inhibitory subtypes may actually implement distinct computations that rely on their specific morphologies and functional positions within the network. For example, inhibition has been suggested to regulate the global level of input responsiveness (sensitivity/gain), sharpen responses to specific inputs (selectivity), and carry out suppressive interactions between those responses (competition), all of which might be mediated by distinct cellular subclasses. To causally explore this possibility, we optogenetically targeted parvalbumin-expressing interneurons in mouse visual cortex, and combined high speed (50 Hz) two-photon imaging, in vivo electrophysiology, and a novel form of single cell optical stimulation in order to isolate the role of Pv inhibition during the well established cortical computations of the visual system. Specifically, we injected the adenoassociated viral vector double-floxed inverted open reading frame ChR2-mCherry (AAV DIO ChR2-mCherry), with Cre-dependent expression of ChR2, into PV-Cre knock-in mice. We observed robust light-activated responses in mCherry-labeled fast-spiking neurons (>100 Hz) in both brain slices and visually-engaged cortex, and found that optically driving these neurons effected functional shutdown in neighboring pyramidal cells in vitro and in vivo. By targeting this shutdown to specifically timed intervals within the receipt of visual stimuli by the cortical network, we could explore the functional impact of Pv inhibition on the resolution of orientation and contrast responses. For orientation responses measured through cell-attached recordings in vivo, we observed a Pv-driven suppression of visual responses in proportion to the magnitude of those responses, suggesting that Pv neurons implement a “division” of responsiveness equivalent to a network shunting effect. This result was confirmed using high-speed two-photon imaging in conjunction with the optical stimulation. Similarly, when activating visual cortical neurons with increasing levels of network drive, by presenting optimal stimuli of increasing contrast, we found that Pv activation had much stronger attenuation of the higher contrasts (stronger network drive) compared to the lower ones (weaker drive), again suggesting a normalization of network state via divisive gain control. Finally, we present a novel methodology for optically activating localized neurons in vivo at many locations on a rapid timescale, and combine this methodology with simultaneous high-speed two-photon imaging to enable single cells to be toggled during concurrent visualization of distributed network responses. Taken together our results strongly implicate parvalbumin-containing inhibitory interneurons as arbiters of dynamic gain control in the cortex.

Society for Neuroscience Abstract, 2011.

C.A. RUNYAN, M. SUR;
Two distinct subtypes of PV+ inhibitory interneurons in mouse primary visual cortex
Soc. Neurosci., 2011.

The roles that the vast array of inhibitory neuron subtypes play in cortical circuits remain to be elucidated, largely because of the historic difficulty in targeting these cells for physiological recordings or manipulations in vivo. Recently, however, genetic labeling methods have allowed the targeted recording of identified subtypes of inhibitory interneurons. Several different strategies have been used to label individual cell classes, including knock-in mouse lines expressing fluorophores under control of inhibitory-specific promoters, as well as mice expressing Cre under subtype-specific promoters, such as Parvalbumin (PV), which are then injected with a floxed viral fluorescent reporter construct, or crossed with a floxed reporter mouse line. The cells can then be targeted with patch pipettes under two-photon guidance, or loaded with calcium dye, allowing the comparison of these cells’ response properties to those of other cell classes. Conflicting results using subsets of these methods have led to controversy regarding the receptive field properties of inhibitory neurons, and in particular, whether inhibitory neurons in the mouse visual cortex can be selective for orientation. In order to help resolve this issue, we have recorded from a large number of PV+ neurons expressing RFP (n = 93), using two-photon guided loose-patch recording. Briefly, adult PV-Cre mice were injected with an LS2L-RFP AAV 2/9 construct. Two to four weeks later, mice were re-anesthetized, a craniotomy was performed, and a patch pipette containing Alexa 488 dye was guided to RFP+ neurons. Gratings drifting in 18 directions were displayed at the preferred spatial and temporal frequencies for each cell, in order to characterize the orientation tuning. Next, each cell was filled with the Alexa 488, and a z-stack of the cell’s dendritic tree was collected. Finally, the dendritic tree was reconstructed off-line. Although PV+ neurons are less tuned than unlabeled neurons on average, we found a clear bimodal distribution of tuning in the PV+ population, that included a large untuned subgroup, and a second highly tuned subgroup. We then compared the dendritic morphology of the untuned and tuned PV+ neurons, and found that the untuned PV+ neurons have significantly larger dendritic trees than do highly tuned PV+ neurons. Thus, by recording from a large number of PV+ neurons, we have been able to define two subclasses that differ both functionally and morphologically: untuned PV+ neurons with wide dendritic trees, and highly tuned PV+ neurons with smaller dendritic trees, possibly corresponding to large and small basket cells.

Society for Neuroscience Abstract, 2011.

G. PEREA, A. YANG, B. Y. CHOW, E. S. BOYDEN, M. SUR;
Channelrhodopsin-2-stimulated astrocytes increase synaptic transmission in visual cortex
Soc. Neurosci., 2011.

Astrocytes have recently been considered to play active roles in synaptic trans¬mission and plasticity in different brain areas. However, the impact of astrocyte activity on cortical networks and on information processing of external stimuli is still poorly known. Here, we use optogenetic tools to selectively stimulate astrocytes and investigate their consequences on neuronal activity in primary visual cortex. Astrocytes were selectively targeted with adeno-associated viral vector with lightsensitive channelrhodopsin-2 (ChR2) under GFAP promoter and neuronal activity was recorded from layer 2/3 neurons of V1. The selective expression of ChR2 in cortical astrocytes was confirmed by GFAP staining, and the ability of this construct to evoke Ca2+ responses was validated in V1 cortical slices where transduced cells displayed robust increases in intracellular Ca2+ in response to 473 nm light. The analysis of neuronal activity showed that optogenetic activation of astrocytes evoked a transient potentiation of spontaneous excitatory postsynaptic currents (sEPSCs), increasing the frequency of sEPSCs without modifying the amplitude of excitatory currents. The increase of synaptic activity was prevented in the presence of MCPG, an antagonist of metabotropic glutamate receptors (mGluRs), suggesting that AAV-GFAP-ChR2-expressed astrocytes can release glutamate that activates mGluRs at presynaptic terminals. To confirm the ability of ChR2-expressed astrocytes to release glutamate, we recorded slow inward currents (SICs), i.e., NMDA-mediated currents induced by glutamate released from astrocytes. To optimize SIC recordings magnesium was removed from the ACSF, and TTX was added to block action potential-mediated neuronal activity. In those conditions, optogenetic astrocyte stimulation evoked both increase of SIC frequency and potentiation of miniature EPSCs. SICs were abolished by antagonists of NMDA glutamate receptors. Taken together, these results show that astrocytes, when activated through selective optogenetic stimulation, increase excitatory synaptic transmission in cortical neuronal networks. In addition, optogenetic manipulation of astrocytes can be a useful tool to selectively stimulate astrocyte Ca2+ signaling (cf. Kasparov et al., 2010) and to evoke the release of gliotransmitters.

Society for Neuroscience Abstract, 2011.

N. MELLIOS, H. SUGIHARA, J. CASTRO, A. BANERJEE, C. LEE, A. KUMAR, B. CRAWFORD, J. STRATHMANN, D. TROPEA, S. LEVINE, D. EDBAUER, M. SUR
miR-132, an experience-dependent microRNA, is essential for ocular dominance plasticity in visual cortex
Soc. Neurosci., 2011.

MicroRNAs (miRNAs) are small non-coding RNAs conserved throughout evolution that are predicted to regulate more than half of protein coding genes. Although they have been shown to be important for neuronal development, differentiation and connectivity, very little is known about their role in experience-dependent cortical plasticity. We utilized visual deprivation paradigms to elucidate the role of miRNAs in visual cortex plasticity. Through a sequential analysis of miRNA expression with miRNA microarray, mature and pri-miRNA qRT-PCR we identified a subset of abundant miRNAs whose expression in mouse primary visual cortex was altered following dark-rearing (DR) and/or monocular deprivation (MD). Focusing on the most robustly affected miRNA in both DR and MD, miR-132, we further determined that its expression was distinctively increased during the critical period of visual cortex plasticity, and gradually normalized after light exposure following DR. Locked Nucleic Acid (LNA) in situ hybridization revealed that miR-132 reduction after DR and MD was specific to cortical layers 2/3 and 4. Neuronal inhibition of miR-132 availability in vivo, using the technology of miRNA “sponging”, resulted in a dose-dependent loss of ocular dominance plasticity in individual identified superficial layer neurons following short term MD. Notably, changes in dendritic spine density of miR-132 sponge infected neurons accompanied the observed functional deficits, and were in accordance with increased levels of miR-132 target p250GAP, a known inhibitor of spine growth. These data support a critical role for miR-132 in activity-dependent regulation of neuronal structure and function.

Society for Neuroscience Abstract, 2011.

J. SHARMA, M. SUR;
Contextual interactions in space and time in macaque V1
Soc. Neurosci., 2011.

In ongoing vision, the statistical structure of visual inputs plays a central role in perceptual processing. Though natural scenes typically have smoothly varying spatial structure, ongoing saccadic behavior may lead to abrupt change in spatial and temporal contextual statistics. Several past studies, including from our lab, have shown that spatial and temporal context have profound influences on receptive field (RF) properties of single neurons in V1 and provide insights into mechanisms underlying perceptual effects. Placing tilted oriented gratings within and outside the RF such that the two overlap in time but not in space leads to a perceptual tilt-effect (TE), whereas presenting gratings of two different orientations such that they overlap in space but not in time leads to an adaptation induced tilt-after-effect (TAE) whose magnitude depends on the relative orientation of the two stimuli. At the single neuron level TE and TAE lead to marked changes in orientation tuning properties of V1 neurons; however not much is known about contextual interactions when neither stimuli overlap in space or in time. In this study done in awake, fixating monkeys, we recorded from V1 neurons using an adaptation protocol consisting of oriented sinusoidal gratings placed in the RF surround. We used two orientations for surround adaptation: one same as the preferred orientation (iso-oriented surround) of the neuron and the other oriented orthogonally (ortho-surround), and studied the effect of adaptation on gratings of 8 different orientations presented in pseudorandom order within the RF center or RF center+surround. The time course of center responses with and without surround adaptation followed significantly different trajectories: there was an early facilitation of responses that lasted for >200ms when the center stimulus was presented after surround adaptation, whereas this period of facilitation lasted for just 100 ms when surround adaptation was followed by presentation of both center and surround stimuli. In the latter case, the initial facilitation was followed by profound suppression at all orientations. In addition, surround adaptation unmasked center responses to all orientations, revealing significantly greater variability in responses compared to the unadapted center alone responses. In the case of surround adaptation followed by center and surround stimuli, the response variability was confined to nearby orientations and the far flank orientations were minimally affected by adapting context. Our results demonstrate profound and divergent influences of spatial and temporal context on tuning properties and time courses of V1 responses.

Society for Neuroscience Abstract, 2011.

I. NAGAKURA, A. VAN WART, D. TROPEA, M. SUR
STAT1 regulates homeostatic plasticity in visual cortex
Soc. Neurosci., 2011.

A previous mRNA microarray analysis identified components of JAK/STAT immune signaling, in particular STAT1, as being significantly upregulated in visual cortex during the critical period following long-term sensory deprivation. STAT1 is activated by cytokines such as IFN-gamma and induces transcription of immune-related genes. However, little is known about the role of STAT1 in cortical plasticity. We found that STAT1, increased after monocular deprivation (MD), acts to inhibit plasticity, suggesting that STAT1 signaling is a part of a negative feedback loop to regulate plasticity. Ocular dominance plasticity was measured using optical imaging, where we examined the intrinsic signal strength of the deprived vs. non-deprived (open) eye in the binocular region of primary visual cortex. IFN-gamma, which increases the level of STAT1, blocked plasticity induced by MD. On the other hand, STAT1 knockout (STAT1-/-) mice showed enhanced plasticity in adult mice, due to accelerated increase of the response from the open eye. Also during the critical period, STAT1-/- mice showed early increase of the open eye response after short-term (4 day) MD. In control wild-type (WT) mice, plasticity after short-term MD was mediated by a decreased response from the deprived eye and the response from the open eye was increased only after long-term (7-8 day) MD. These results strongly suggest that STAT1 regulates homeostatic upregulation of the open eye response, so that removing STAT1 inhibition accelerates the homeostatic response. In ongoing experiments, we have identified specific molecular signals downstream of STAT1 that implement such plasticity.

Society for Neuroscience Abstract, 2011.

J. Castro, R. Garcia, S. Kwok, M. Sur
Recovery of function by recombinant human IGF1 treatment in a mouse model of Rett Syndrome
Soc. Neurosci., 2011.

Rett Syndrome is a neurodevelopmental disorder that primarily affects girls during early childhood. These individuals are born asymptomatic and develop normally until 6-18 months of age when a variety of symptoms appear gradually such as weight loss, ataxia and gait apraxia, loss of motor coordination, changes in social behavior and autonomic anomalies such as breathing and cardiac perturbations. The majority of Rett Syndrome cases are caused by sporadic mutations in the X-linked gene MECP2, a transcriptional modulator with a great variety of gene targets. One of the best known targets is BDNF, known to trigger neuronal and synaptic maturation through activation of signaling pathways such as PI3K/Akt. In fact studies in mouse models of Rett Syndrome have demonstrated the therapeutic effects of activating the PI3K/Akt pathway with BDNF or a common activator, IGF1. Unlike BDNF, IGF1 is able to cross the blood brain barrier and daily injections of a tri-peptide form of IGF1 is able to partially reverse many of the symptoms (Tropea et al., PNAS 106: 2029, 2009). In this study we have tested the therapeutic effects of recombinant human IGF1 (rhIGF1) in a mouse model of Rett syndrome. This full length, 70 amino acid version of IGF1 is expected to be more effective and holds significant potential as a clinical treatment. Beginning at 2 weeks of age, MeCP2 null mice and wild type littermates were injected daily with rhIGF1 throughout their lifetime. Treated animals showed significant increase in lifespan and improvement in body weight as compared to saline treated MeCP2 null mice. These effects in the treated group were accompanied by the improvement of locomotion at 8 weeks and recovery of cardiac and respiratory rhythms to those of wild type levels. In a social preference test measured by a three-chamber assay, wild type mice showed significant habituation in their social interactions, as demonstrated by reduced interactions with a stimulus mouse on a second encounter. MeCP2 null mouse showed no such habituation in their social interactions, whereas treatment with rhIGF1 restored habituation to wild type levels. These data demonstrate recovery of function at organismal and brain system levels in MeCP2 null mice due to rhIGF1 treatment, and point to widespread impact of IGF1 signaling in this disorder. They are consistent with the hypothesis that deficits in signaling pathways that influence synapse maturation are a key component of MeCP2 dysfunction or deletion, and suggest that rhIGF1 is a potential therapeutic for Rett Syndrome.

Society for Neuroscience Abstract, 2011.

Kwok, Showming, Sheridan, Steven D, Petravicz, Jeremy C, Reis, Surya, Haggarty, Stephen J, Sur, Mriganka
DIFFERENTIATION AND CHARACTERIZATION OF RETT SYNDROME PATIENT-SPECIFIC NEURAL PROGENITOR CELLS
Intl. Soc. for Stem Cell Res. 9th Ann. Mtg., 2011.

Rett syndrome is a neurodevelopmental disorder caused by mutations in a transcriptional modulator gene, methyl CpG-binding protein 2 (MECP2). Mouse models with MeCP2 deletion recapitulate Rett-like symptoms. Reintroduction of MeCP2 in adult mice was shown to rescue the mutant phenotype, pointing to the possibility that Rett syndrome is a treatable disorder even if MeCP2 or its downstream signaling pathway is restored in adulthood. Administration of a tripeptide form of IGF1, a growth factor that is important for neuron survival and synaptic maturation, was found to partially reverse Rett symptoms in a mouse model. However, the detailed disease mechanisms and affected signaling pathways remain elusive, making identification of therapeutic targets for treating Rett patients difficult. To overcome these limitations, we are using pluripotent stem cell approach to derive mature neurons from reprogrammed adult somatic cells from Rett patients for studying disease mechanisms and discovering therapeutics. Using viral-mediated delivery of reprogramming factors, we have generated several lines of patient and control fibroblast-derived induced pluripotent stem cells (iPSCs) that carry various mutations in MECP2. Neural progenitor (NP) cells were successfully derived from these iPSC clones and were characterized using immunocytochemical methods that demonstrated that these cells are homogenous and express stage-appropriate markers. Upon long-term culturing on laminin coated surface up to 10 weeks, these NP cells differentiate into neurons as demonstrated by the detection of lineage specific proteins using immunocytochemistry. We are able to further examine these neurons with patch clamp recording and calcium imaging techniques to assess function and the extent of their maturation. Finally, we are looking into specific signaling pathways that underlie disease mechanisms. Based upon these findings, we are developing high-throughput cellular screening assays using automated microscopy and pathways-selective reporter genes to decipher differences in signaling mechanisms between Rett patient specific neurons and healthy controls.

International Society for Stem Cell Research, 2011.

Slavin, Ileana, Sheridan, Steven D, Kwok, Showming, Morey, Robert, Lynch, Candace, Loring, Jeanne F, Sur, Mriganka, Laurent, Louise C, Haggarty, Stephen J
EPIGENETIC ABNORMALITIES IN RETT SYNDROME FIBROBLASTS ARE PARTIALLY CORRECTED BY REPROGRAMMING
Intl. Soc. for Stem Cell Res. 9th Ann. Mtg., 2011.

Rett syndrome is a neurodevelopmental disorder caused by mutations in MECP2, a gene located on the X chromosome. Since males have only one X chromosome, affected males carry one mutant copy and no normal copies of the MECP2 gene, and generally die prior to birth. Affected females carry one mutant copy and one normal copy in each of their cells; however, in any given cell, either the mutant copy or the normal copy is active, due to X-inactivation. Girls with Rett syndrome often appear normal in early infancy, but experience loss of developmental milestones and onset of stereotypical movements and breathing patterns at 6-18 months. The stochastic nature of X-inactivation results in a range of clinical severity, depending on the fraction of cells in the brain that have inactivated the mutant versus the wild-type copy of MECP2. In order to develop an in vitro model of Rett syndrome where we could study the potential defects in neural differentiation, function, and survival caused by mutations in MECP2, we generated induced pluripotent stem cells (iPSCs) from dermal fibroblasts from 8 patients with Rett syndrome and 8 control individuals. Since the MECP2 gene encodes a methyl-CpG-binding protein, we compared the DNA methylation profiles of both dermal fibroblasts and iPSCs from the affected and control subjects. We used a high-resolution genome-wide microarray (Illumina, Inc.) to interrogate the methylation status of 450,000 individual cytosines distributed across the genome. The DNA methylation profiles of the affected and control fibroblasts were markedly different, while those of the affected and control iPSCs were largely indistinguishable. The reversal of the abnormalities in DNA methylation seen in the Rett syndrome fibroblasts with reprogramming to iPSCs suggests that it may be possible to recapitulate the onset of aberrant DNA methylation by performing directed differentiation of these iPSCs to the neural lineage.

International Society for Stem Cell Research, 2011.

K. Mergenthaler, et al.
Modeling the broadening of orientation tuning in ferret V1 during application of glutamate transporter blocker
Soc. Neurosci., 2011.

In vivo studies on response properties to differently oriented gratings in neurons and astrocytes in ferret primary visual cortex identified three important features [1]. First, neurons and astrocytes respond to oriented grating selectively with a comparable width in tuning. Second, locations of pinwheels and orientation domains organzied in maps do coincide in neurons and astrocytes. Third, orientation tuning is differently affected in neurons and astrocytes during the application of a glutamate transporter blocker (TBOA): Blocking glutamate transport broadened the orientation tuning in neurons and fully suppressed orientation selectivity in astrocytes. However, the underlying mechanisms generating a broader orientation tuning in neurons and suppressing the tuned response in astrocytes during application of TBOA are unknown. In this context we first investigate how the influence of prolonged TBOA-mediated high glutamate concentrations at the synaptic cleft [2] is expressed in the opening rate of post-synaptic AMPA and NMDA receptors and finally in the resulting combined EPSC. The obtained transfer functions are further used to describe the currents mediated by excitatory synapses in a 2d network model [3] of the primary visual cortex. The network is composed of excitatory and inhibitory populations with Gaussian distributed random recurrent connections and receives weakly tuned afferent input. First simulations show that broadening in neuronal tuning curves during application of TBOA can be explained by changes in the time course of glutamate clearance.This suggests that astrocytic glutamate transporters can influence, the neural response properties to features like orientation by changing the glutamate clearance.
[1] Schummers, J., Hongbo, S., and Sur, M., Science 320,1638 (2008) [2] Diamond, D., J. Neurosci., 25(11),2906 (2005) [3] Stimberg, M., et al., Cerebr. Cort. 19,2166 (2009)

Society for Neuroscience Abstract, 2011

2010

J. Sharma, H. Sugihara, M. Sur
Attention to space and time are separable and depend on behavioral contingencies in macaque V1
Soc. Neurosci., 2010.

In ongoing visual processing, space and time are largely inseparable. However when the temporal flow of events leading to a behavioral response is associated with location of an object, the visual system uses attention as a flexible device to allocate greater resources to space or to time. The key to this dynamic modulation is internal states such as expectation that relate to behavioral contingencies. To test whether neuronal response modulation was indeed a reflection of monkeys’ expectation of a behavioral response, we compared V1 responses in three sets of experiments conducted in two monkeys. In one experiment, the monkeys were simply required to hold fixation at a central spot to earn a juice reward; no behavioral response was required. In the other two experiments, the monkeys covertly attended to a peripheral attention spot and released a lever when it disappeared. The difference between the experiments was the time windows within which the attention spot disappeared: in one case the extinction was randomly varied over the last 300 ms (of a ca 2000 ms presentation), while in the other it was varied during the last 1500ms. Thus hazard rates for attention spot extinction followed very different trajectories. Recordings were made in V1 neurons with similar receptive field (RF) eccentricities, in either an attend-toward or attend-away condition depending on whether the attention spot was situated in the same or opposite visual field as the RF. The stimuli for all the three experiments consisted of sinusoidal gratings centered on neuronal RFs. The population response profiles in the initial phase were remarkably similar in the three experiments; however, they followed very different time courses later in the trial. When attention was not actively engaged and no behavioral response was required, the responses simply reduced monotonically to an arbitrary level. In the attention experiments, on the other hand, there was a significant change in neuronal responses in the late phase, particularly in the attend-towards condition. Here an early stimulus related increase in response was followed by a sustained increase which reflected expectation of a behavioral response. Our data show that the dependence of neuronal dynamics on attention to space and time is dependent on behavioral contingencies. During the initial phase of response, since attention to location of the spot was necessary and no behavioral response was required there was a modest spatial attention-related increase. Later in the trial, the temporal contingency in attention spot extinction dominated the behavior, and neuronal responses strongly reflected the consequent attention to time.

Society for Neuroscience Abstract, 2010.

C.A. Runyan, N. Wilson, J. Schummers, A. Van Wart, S. Kuhlman, M. Carlen, K. Meletis, LH. Tsai, Z.J. Huang, M. Sur
Response features of parvalbumin-expressing interneurons suggest precise roles for subtypes of inhibition in visual cortex
Soc. Neurosci., 2010.

Cortical inhibition is thought to play an important role in shaping the response properties of excitatory neurons; however, the sheer diversity of inhibitory interneuron subtypes has made the study of their specific roles in cortical circuits difficult. Parvalbumin (PV) expression is thought to be limited to inhibitory fast-spiking basket cells and chandelier cells, and thus provides an opportunity to genetically mark and optogenetically manipulate this cell class. We have marked PV-expressing neurons in the mouse primary visual cortex using Cre-dependent recombination of LSL-RFP (Adenoassociated virus) AAV constructs in PV-Cre knock-in mice. Two weeks after the viral injection, we performed either two-photon guided loose-patch recordings or two-photon calcium imaging in the infection site, in order to define the visual receptive field properties of PV+ neurons and compare them to the unlabeled population. Surprisingly, we found that the PV+ population includes a diversity of receptive field properties that is comparable to that of the unlabeled population, including sharp and broad orientation selectivity, small and large receptive fields, and lowpass as well as bandpass spatial frequency tuning characteristics. Our results suggest that the PV+ population includes further subclasses of cells, and furthermore, provide support for the push-pull model of orientation tuning, which relies on highly tuned inhibition to sharpen the orientation selectivity of excitatory cells.

Society for Neuroscience Abstract, 2010.

R. Garcia, N. Mellios, J. D. Jaffe, M. Sur
Monocular deprivation during a developmental critical period alters astrocyte number and astrocyte-specific protein expression in mouse visual cortex
Soc. Neurosci., 2010.

Activity dependent changes in visual cortex involve both structural and functional modifications. Visual experience modulates the maturation of neuronal circuits, including spine structural dynamics. Astrocytes have been shown to ensheath synapses and regulate neuronal activity by the uptake and release of neurotransmitters. They have also been shown to possess highly motile processes and it is likely that these morphological changes play a role in activity dependent plasticity of neuronal circuits. In this study, we examined the effects of monocular deprivation on the expression of astrocyte-specific proteins during postnatal critical period for experience-dependent plasticity in the mouse visual cortex. Mice were monocularly deprived at postnatal day 24 (P24) for 4 days until the peak of critical period. Preliminary analysis from proteomic data showed a reduction of astrocytic-specific proteins, including GFAP protein levels. Immunohistochemical analysis revealed a significant reduction of GFAP-labeled astrocytes in mouse visual cortex as compared to non-deprived control mice. These findings further establish the dynamic nature of astrocyte plasticity in response to visual experience.

Society for Neuroscience Abstract, 2010.

N.R. Wilson, J.M. Schummers, R. Chen, C.A. Runyan, M. Sur
High-speed interrogation of cortical dynamics via conventional two-photon microscopes
Soc. Neurosci., 2010.

Neuronal networks process information in a distributed fashion, yet the emerging tool of choice for their in vivo elucidation is a highly focused laser beam. Investigators attempt to maintain this focus while simultaneously trying to monitor many different areas of interest. This false dichotomy can be resolved by recognizing that most two-photon (2P) imaging sweeps waste the vast majority of their bandwidth (>90%) sampling regions of non-interest (RONIs) as they move monotonously along the same square scan pattern. Bringing the laser under conscious, directed control by the experimenter can allow it to move intelligently between the areas of interest, collect more photons from those areas (greater signal to noise), and/or cycle between those areas more rapidly (higher temporal resolution). Realizing this type of arbitrary, fast laser scanning in vivo has required not only a different type of hardware control (to enable fast laser deflections in precise directions), but also a framework of software control (to ensure that meaningful “smart” scanning happens automatically without distracting the investigator from the experiment). Here, we demonstrate a new in vivo calcium imaging methodology that integrates: 1) automated cell detection to lock onto neurons exhibiting specific criteria, 2) automated laser navigation to traverse the shortest path between those neurons, and 3) automated image stabilization and dynamic adjustment to prevent the scan path from “falling off its axis” upon animal movement. Using this methodology, we observe a 10 to 100-fold improvement in the signal-to-noise collected compared to standard two photon imaging. Importantly, while high speed in vivo calcium imaging has been recently demonstrated using specialized AOD-based hardware, it required modifications to the microscope that involved intricate and expensive additions; we achieve fast imaging in vitro and in vivo simply by running different software on the imaging computer of an unmodified microscope, and validate this “free” augmentation on both an “open” 2P microscope (Sutter Instruments) and a popular commercial one (Prairie Technologies). Finally, we discuss immediate applications of smart laser path technology, including focal imaging of genetically labeled cell types, and patterned optogenetic stimulation of multiple neurons. All in all, the ability to dynamically specify arbitrary laser paths will be an important component of future “network physiology” studies aimed at elucidating interactions between specific neurons and cell types within a cortical circuit.

Society for Neuroscience Abstract, 2010.

H. Sugihara, C. McCurry, M. Sur
Putative inhibitory neurons showed normal plasticity without plasticity of excitatory neurons in visual cortex of mice lacking Arc
Soc. Neurosci., 2010.

Arc (activity regulated cytoskeleton-associated protein), also known as Arg3.1, is an immediate early gene and is considered to play a critical role in synaptic plasticity. However, its contribution to plasticity of specific cell types within intact cortical circuits is unknown. Using optical imaging of intrinsic signals, we have previously reported that Arc knock-out mice show impaired ocular dominance plasticity (McCurry et al., Nature Neuroscience 13: 450, 2010). However, Arc is only expressed in excitatory neurons in the visual cortex. Here, we report on the plasticity of cells which do and do not express Arc, employing Arc-GFP mice, in which GFP is expressed under the control of the Arc promoter. Using GFP expression as a marker for excitatory neurons, we classified neurons into GFP-positive (putative excitatory) and GFP-negative (putative inhibitory) neurons. At around the peak of critical period (P27-29), we performed monocular deprivation (MD) in heterozygote and homozygote Arc-GFP mice. After 5-6 days of MD, we loaded cells with the calcium indicator dye OGB1 and used functional two-photon calcium imaging to measure eye-specific visual responses and calculate the ocular dominance index (ODI) of individual neurons. In heterozygote mice, GFP marks cells that contain Arc; these mice showed a shift of the ODI in both GFP-positive and GFP-negative neurons. In homozygote mice, in which GFP marks precisely those cells that lack Arc, there was no significant ODI shift of GFP-positive neurons, demonstrating impaired plasticity. GFP-negative neurons, however, showed a significant ODI shift, indicating plasticity in inhibitory neurons independent of excitatory neurons. We are currently investigating whether short-duration MD also has differential effects on putative excitatory and inhibitory neurons.

Society for Neuroscience Abstract, 2010.

N. Mellios, H. Sugihara, J. Castro, A. Kumar, R. Garcia, B. Karki, D. Tropea, S. Levine, D. Edbauer, M. Sur
Role of MicroRNAs in experience-dependent cortical plasticity
Soc. Neurosci., 2010.

Deprivation of visual input from one eye (Monocular Deprivation or MD) or both eyes (Dark Rearing or DR) are established paradigms for studying experience-dependent cortical plasticity. MicroRNAs (miRNAs) are small non coding RNAs that have been shown to be abundantly expressed in mammalian cerebral cortex and involved in neuronal development and plasticity. Using a plethora of different techniques including miRNA microarray, qRT-PCR and LNA in situ hybridization, we identified miRNAs whose expression is altered following DR and/or MD in mouse primary visual cortex and examined their laminar and cellular expression patterns. Among the experience-dependent miRNAs were members of the same family of miRNAs, a subset of which has already been shown to be influenced by neuronal activity. In silico analysis revealed that the predicted targets of DR/MD altered miRNAs constitute important pathways previously linked to visual cortex plasticity. We are currently using lentivirus- mediated miRNA overexpression and inhibition approaches to manipulate in-vivo the expression of selected MD/DR altered miRNAs, in order to examine their role in experience-dependent plasticity in mouse visual cortex.

Society for Neuroscience Abstract, 2010.

N. Chen, G. Perea, M. Sur
Cell-type specific cholinergic modulation of the mouse visual cortex
Soc. Neurosci., 2010.

The release of acetylcholine (ACh) into the primary visual cortex (V1) following the activation of nucleus basalis modulates attention and is hypothesized to improve information processing via muscarinic receptor (mAChR) mediated mechanisms. Previously it has been demonstrated that visual cortex astrocytes in vivo respond with calcium elevations to visual stimuli and show tuned responses (Schummers et al., Science 320: 1638, 2008). Although several studies have focused on cholinergic effects on cortical neurons, the consequences of ACh on other cell types, such as astrocytes, are poorly understood. Here, we have investigated possible cholinergic-mediated signaling in cortical astrocytes and the consequential downstream effects on neurons. Using calcium imaging, whole cell patch-clamp and post-experimental immunohistochemistry, we screened for cholinergic responses of both neurons and astrocytes in the superficial layers of mouse V1 cortical slices. Local application of ACh induced intracellular calcium elevations in astrocytes, which were manifested by an increase in the frequency of TTX-insensitive calcium oscillations. These responses were abolished in the presence of atropine, an mAChR antagonist. On the other hand, cortical neurons displayed an increase in spontaneous low-frequency slow inward currents (SICs) that were abolished by AP5, an NMDA glutamatergic receptor antagonist. These results indicate that astrocytes are cellular targets of ACh, where their cholinergic excitation may modulate neuronal excitability via glutamate release.

Society for Neuroscience Abstract, 2010.