Analyzing Associations: Differential Amygdalar Activity to Conditioned Stimuli

Associative learning, the process by which an organism forms a link between a cue or behavior and an intrinsically valenced stimulus, is a critical feature of adaptive behavior and survival. Learning to run upon hearing a roar associated with a predator allows one to avoid getting maimed or eaten; learning to rush to the kitchen when you smell freshly baked cookies enables you to snag a warm, tasty treat. As demonstrated by these two examples, associations can occur with both negatively valenced and positively valenced stimuli, with the former often invoking withdrawal and avoidance behaviors and the latter enforcing active approaching behaviors. Because associative memory acts as such a cornerstone in shaping flexible and appropriate behaviors, when this process goes awry, various negative consequences can ensue: depression, substance abuse, and anxiety disorder are only a few of such maladies tied to disturbances in circuitry underlying associative and motivational functionality. Consequently, studying this subject is of paramount therapeutic, as well as intellectual, interest.

Dr. Kay Tye’s Lab at the Massachusetts Institute of Technology aims to understand these very processes, employing a variety of methods from optogenetics and electrophysiological recordings to pharmacological and imaging techniques. A recent paper published by Dr. Tye and colleagues provides significant insight into the routing of positive and negative information from the amygdala during memory retrieval, suggesting that specific cell populations in the basolateral amygdala (BLA) encode different types of emotional valence depending upon the region to which they project. While the BLA as a whole is crucial in forming both positively and negatively valenced associations, subpopulations within the BLA demonstrate distinct activity patterns to rewarding versus aversive conditioned stimuli.

The main targets of the BLA investigated in Beyeler et al.’s (2016) study included the nucleus accumbens (NAc), central amygdala (CeA), and ventral hippocampus (vHPC). While the NAc is generally involved in positive reinforcement, the CeA is linked to learning that mediates aversive behaviors, and the vHPC presumably promotes anxiogenic actions. To examine the neural codes of BLA neurons targeting the NAc, CeA, and vHPC, Beyeler et al. (2016) utilized a dual-virus approach entailing optogenetic-mediated phototagging and in vivo electrophysiological recordings in mice trained to associate a particular tone with sucrose (a rewarding outcome) delivery and another tone with quinine delivery (an aversive outcome). After the mice learned the associations between each stimulus and its respective outcome—assessed as showing anticipatory licking selectively after hearing the sucrose-predictive tone for 70% of the trials—Beyeler et al. (2016) performed acute recordings in the BLA.

Half of the neurons recorded in the BLA were found to respond to sucrose and/or quinine, with 28% of neurons demonstrating selective responses to the tone associated with sucrose and a mere 9% altering firing selectively to the tone associated with quinine. In addition, 13% of BLA neurons responded to both tones similarly, and less than 1% responded to both tones in qualitatively different manners.



Nevertheless, the most notable finding relates to the analysis of the subpopulations in the BLA that responded to both or to either of these auditory cues. Specifically, the subpopulation of BLA neurons that projected to the NAc exhibited a starkly different response profile than that of the BLA-CeA neurons. Zero BLA-NAc neurons were found to be excited by the tone associated with quinine, but nearly all (77%) were excited by the tone tied to sucrose; Any BLA-NAc neurons that reacted to the quinine-associated tone demonstrated inhibition by the conditioned stimulus. In contrast, all BLA-CeA projectors recorded were excited by the tone paired with quinine.

Classifying BLA projector populations using “valence bias”—considering excitation and inhibition as qualitatively different—reinforced the differential activity of the three subpopulations examined. Approximately half of the BLA population encoding valence showed a positive bias—increasing firing for the sucrose-associated tone and/or decreasing activity to the quinine-associated tone—while the other half appeared to encode negative valence—increasing firing to the quinine cue and decreasing firing to the sucrose cue. Within the BLA, a greater percentage of BLA-NAc projectors were observed to be encode positive valence (80%, compared to 31% in the BLA-CeA); whereas more BLA-CeA cells encoded negative valence (69%). Interestingly, the BLA-vHPC subpopulation responded evenly to both tones, with 43% encoding positive valence.


These findings clearly suggest that BLA-NAc neurons preferentially encode cues associated with rewarding outcomes, consistent with the known role of the NAc in positive reinforcement; and that BLA-CeA neurons preferentially encode cues linked to aversive outcomes, also consistent with its acknowledged role in assigning negative valence to stimuli. The observation of BLA-vHPC responses evenly distributed between positive and negative valence is somewhat surprising due to its supposed role in anxiogenic behavior and requires further investigation.

If you find this experiment intriguing and would like to learn more about the nuances of and mechanisms underlying neural systems supporting associative learning, motivation,  and emotional processing, attend Dr. Kay Tye’s talk this Tuesday, February 21, at 4:00 pm at the Center for Neural Circuits and Behavior.


Gina D’Andrea-Penna is a first-year neurosciences Ph.D. student whose interests largely reside in the domain of cognitive neuroscience, particularly in investigating perceptual awareness, attention, and working memory.


Using Optogenetics to Shed Light on Deep Brain Stimulation

Medical interventions often come about by the application of current theory surrounding a disease, but ultimately whether one is approved and used in patients depends not on whether it fits with current dogma, but simply on whether it works. Consequently, as the theories surrounding how a disease develops change with new research, studying why certain treatments are effective can provide further information about the disease’s pathophysiology. In addition, research into how a treatment exerts its effect can reveal avenues to refine and improve the intervention, as well as generate new and better strategies to tease apart and understand complex biological systems.

Dr. Viviana Gradinaru’s lab at Caltech uses new techniques in neuroscience such as optogenetics, CLARITY, and 2-photon imaging to study deep brain stimulation (DBS), which is an effective treatment for various neurological disorders. As a postdoctoral researcher working with Dr. Karl Deisseroth at Stanford, Dr. Gradinaru used the then-developing technique of optogenetics to study how DBS of the subthalamic nucleus (STN) improves the symptoms of Parkinson’s disease. Optogenetics is a technique where light-activated channels are expressed in neurons and used to influence the behavior of those neurons. These channels can also be targeted to specific populations of neurons using those neurons’ genetic markers, making this technique especially powerful.

The researchers used optogenetics to attempt to identify which populations of cells contribute to the efficacy of DBS in the STN. First, they created mice that model Parkinsonian symptoms by using a chemical called 6-OHDA to kill most of the dopamine neurons in the substantia nigra pars compacta (neurons which are lost in Parkinson’s disease). Unlike in Parkinson’s disease, however, the neurons were killed only on one side of the brain, causing deficits only on the opposite side of the body. The severity of the mice’s parkinsonism could then be assessed by measuring a characteristic rotating behavior, which is produced by a bias toward movement on one side that results from the asymmetrical deficit.

The researchers then set about exciting and inhibiting different populations of cells in the STN of the hemiparkinsonian mice, as may happen to these cells during DBS, and measuring the effect of these manipulations on the turning behavior, among other measures of the parkinsonian phenotype. After ruling out local excitatory neurons and glia, they found that high-frequency stimulation of axons in the STN that carry signals from other brain regions improved the mice’s phenotype. Like with DBS in humans, low-frequency stimulation was not effective, and by some measures even worsened the phenotype.


Further, high-frequency stimulation of a population of cells in the primary motor cortex (M1) from which some of the axons in the STN originated also improved the mice’s symptoms. Given that stimulation of the neurons influenced by these axons could not account for the improvement in phenotype, the researchers suggest that signals could be traveling backwards along the axon from the site of stimulation in the STN, relieving the parkinsonism by shifting activity in M1 away from low frequency bursting.


Though other mechanisms may still account for the efficacy of STN DBS, this research revealed a surprising mechanism by which DBS could improve Parkinson’s disease in humans, which was not intended at the time that STN DBS was originally developed. In addition, these results show the power of techniques for targeted dissection of neural circuits like optogenetics. Such methods have the potential to help researchers understand diseases and their treatments, and even provide insight in how to better target such treatments to improve their efficacy or reduce side effects.

To learn about Dr. Gradinaru’s current research, attend her talk on Tuesday, February 7th at 4 pm in the Marilyn G. Farquhar Seminar Room of the Center for Neural Circuits and Behavior.


Jacob Garrett is a first-year graduate student in the UCSD Neurosciences program, whose main interests within neuroscience include modeling and primary sensory systems.

Can the “dynome” succeed where the connectome cannot?

Recently, the big research push in the neurosciences has been to characterize the complete wiring diagram of the brain – to describe the Connectome.  Though a lofty goal that will likely span another decade or more of work, no one doubts the enormous contribution that it will provide to the understanding of brain functioning.  However, even with the connectome in hand, researchers will still be at a loss to explain the generation, coordination, and consequences of dynamic rhythmic processes that characterize so much of the cognitive neuroscience literature.  To bridge this explanatory gap, Dr. Nancy Kopell, a professor of mathematics and statistics at Boston University, has suggested a new large-scale research undertaking aimed at characterizing the dynamic interaction of brain regions – to describe the “Dynome”.

Dr. Kopell’s work mainly focuses on mathematically characterizing the types of dynamics interactions that take place between cells, within networks, and between brain regions, as well as constructing biophysical models of how these dynamics may occur.  Her research, and that of her contemporaries, has demonstrated that dynamic brain rhythms can be involved in phenomena such as network synchronization, attentional gain, and recruiting brain regions per cognitive demands.  For example, the well-known pyramidal interneuronal gamma (PING) mechanism leading to higher frequency gamma rhythms relies on feedback inhibition.  This inhibition leads to a winner-take-all type scenario whereby the most activated cells end up suppressing more weakly activated ones, and potentially serves as a mechanism of focused attention.

Unfortunately, a difficulty that arises, even within a type brain rhythm, is that the generative mechanisms and modulation of rhythms can vary widely by brain region, cortical layer, and even cell type.  Thus, it becomes critical to outline not only the anatomical structure of a region, as connectomics is attempting, but also the cellular physiology, functional synaptic connectivity, and neuromodulatory profiles present.  Fortunately, technologies that have been emerging over the last few years in experimental neuroscience seem well-suited to provide answers to these exact types of questions.  Specifically, high density electrode recordings, optogenetic manipulation, and large scale 3D imaging of neuronal activity have allowed more in-depth analysis of network activity, small circuit functioning, and cell-type specific physiology.  Additionally, new data analytic techniques are providing ways to characterize activity and understand correlations, while data-driven computational models allow researchers to test potential generative mechanisms.

The neuronal heterogeneity involved in implementing these brain rhythms might seem intimidating, but it is likely that it is necessary for the emergence of many cognitive phenomena.  For example, if the same rhythms are being differentially generated in two brain regions, then the similar operating frequency will facilitate stronger interactions, while the dissimilar implementation may mean distinct computations are being performed.  Moreover, the heterogeneity allows these various regions to be differentially regulated by the same neuromodulators, possibly leading to the diverse set of changes that occur across different cognitive states.  Perhaps the most compelling evidence for this is in the effectiveness of deep brain stimulation in treating various mental illnesses.  Cognitive and behavioral abnormalities exhibited by patients with Parkinson’s disease, depression, or obsessive compulsive disorder all experience dramatic improvements simply by a perturbation of the underlying pathological network dynamics.

Importantly, elucidation of the Dynome can and should occur in concert with that of the Connectome.  The highly plastic nature of the brain means that connections constantly change according to the dynamic activity present.  With that in mind, it seems a static Connectome can never completely capture its architecture.  However, this type of broad framework, large-scale synthesis, and open-ended goals may allow neuroscience to continue its rapid progression towards explaining brain functioning.  As data is continually compiled across various levels of brain functioning, the contributions of both connectomics and dynomics will both be needed to move us from genes, to physiology, to network dynamics and interactions, eventually to cognition and pathology.

To hear more about Dr. Kopell’s specific contributions to the dynome, please attend her talk on Tuesday, January 31 at 4pm in the CNCB Marilyn G. Farquhar Seminar Room. To learn more about her lab and for a list of recent publications, please visit her website:

Ryan Golden is a first-year in the Neurosciences Graduate Program, and is currently rotating in Maxim Bazhenov’s lab.  His interests broadly fall under computational and theoretical neuroscience, and is specifically interested in the biophysical implementation and modulation of reinforcement learning.



Juvenile mouse pheromone inhibits sexual be

With the past few decades bringing exponential growth in the fields of neuroscience and cell biology, we have begun to understand physiology in a way that our predecessors could only hypothesize. We have identified countless proteins, genes, receptors and channels, traced their interactions, manipulated their properties and even developed drugs that can interact to treat diseases. However, there are still many areas that lack understanding of the players involved in some of the most basic biological processes. Stephen Liberles, PhD, is an associate professor in Cell Biology at Harvard Medical School. His research involves internal and external sensory systems, mainly olfaction and pheromone signaling and sensory modalities of the vagus nerve. Here I will discuss the findings from a 2013 Nature paper regarding juvenile mouse pheromone signaling and inhibition of sexual behavior.

Pheromones come in many forms: urinary volatiles, steroid derivatives and proteins in urine, tears and saliva. However, the function of many of these pheromones is unknown. qPCR using cDNA from a range of ages in mice demonstrated stark age-dependent differences in some peptide expression levels in the extraorbital lacrimal glad (LG). Amongst these peptides was exocrine-gland secreting peptide 22 (ESP22) which was produced in juveniles, but not in adults (see attached figure). The expression levels were highest at 2-3 weeks, corresponding to the time just before puberty. The levels were similar in males and females, and was isolated to the LG. RNA in situ hybridization localized ESP22 to a set of acinar lacrimal cells, the cells responsible for releasing contents into tears. Western blot analysis supported that ESP22 was a component of juvenile tears, using mass spectrometry the Liberles lab was able to identify the primary structure of the peptide.

Once ESP22 had been identified in juvenile mice, the sensory perception pathway was examined. Electrophysiological recordings determined that neurons in the vomeronasal organ (VNO) respond to ESP22, and more specifically require the ion channel TRPC2. This was affirmed by juvenile tears activating VNO neurons, while adult tears produced no recordings. Immunohistochemical staining (cFos) was used to trace the neural circuitry. It was demonstrated that upon activation with ESP22 cFos expression was increased in the medial amygdala (MeA) which receives input from the VNO by the accessory olfactory bulb and sends projetions to the hypothalamic areas controlling reproductive responses. In order to further examine the signaling, TRPC2 knockout (KO) mice were used and demonstrated no activation of the amygdala.

Once the signaling pathway was determined, implicating the need for both ESP22 and TRPC2 receptor, behavioral studies were performed in mice lacking ESP22 (C3H strain) and TRPC2 KO mice. While wild type male mice have a mating preference for adult females, they tend to not mount juveniles. However, when ESP22 was not present, the adult mice had an increased time spent mounting juveniles, and decreased latency between mountings. Similarly, the TRPC2 KO mice also spent more time mounting juveniles. When exogenous ESP22 was added to the C3H juvenile mice, the males with TRPC2 stopped the mating behavior.

Taking together this information supports the findings that ESP22 is a pheromone produced by the acinar cells in the lacrimal gland of juvenile mice before puberty. It activates TRPC2 in adult male mice, which acts on the medial amygdala to inhibit mating behavior. Once the mice reach puberty, the production of ESP22 decreases from the females, thus removing the inhibition from the males, and allowing for regular mating behavior.

For more information and publications from the Liberles lab, please visit the website at

Amy Taylor is a third year MSTP student in the Neuroscience PhD program at University of California San Diego. Her research focus is on the identification and manipulation of EEG biomarkers in schizophrenia and their relation to functional outcomes.

Role of Sox4 and Sox11 in Early Neuronal Differentiation

Corticogenesis is a complex process requiring proliferation of progenitor cells in the ventricular and subventricular zones, neuronal migration, and synaptogenesis. Early in neural development, radial glia divide to produce excitatory neurons of the cortex. Later on in development, radial glia produce intermediate progenitor cells (IPCs) that detach from the surface of the ventricular zone (VZ) to relocate to the subventricular zone (SVZ). IPCs have limited mitotic potential relative to radial glia, and also produce cells that migrate into the cortex. The timing of activation of molecular pathways involved in proliferation and neural differentiation is critical during corticogenesis, and relies on various transcriptional regulators.

The Sox family of transcriptional regulators is known to alter gene expression and determine cell fate, and they partner with other proteins to exert their effects. The SoxC subfamily of transcriptional regulators (Sox4, Sox11, and Sox12) is thought to be functionally identical and expressed in all neuron subtypes. Chen et al. (2015) assessed the expression levels, protein localization, and putative function of both Sox11 and Sox4 in the developing mouse cortex. Sox4 and Sox11 had similar expression levels across cortical development from E10.5-P10 measured by qRT-PCR, peaking between E12.5-E16.5, a time corresponding to early neuronal differentiation.

While expression levels were similar between Sox11 and Sox4 across corticogenesis, immunohistochemical analysis showed distinct protein localizations. Sox11 localized to differentiated neurons, as expected, and also co-localized with Neurogenin1 (Ngn1), a transient marker of early differentiated neurons, at E12.5. Sox4 was localized to both differentiated neurons as well as IPCs in the SVZ. Given the difference in protein expression patterns seen via IHC, Chen et al. examined the role of Sox4 and Sox11 in neuronal differentiation separately. In vitro analysis of differentiated neuron cultures transfected to either overexpress (GOF) or underexpress (LOF) Sox11 was performed. GOF cultures showed increased neuronal polarization and increased neurite length, whereas LOF cultures showed less mature neurons and decreased neurite length. These results support a role of Sox11 in early neuronal differentiation.

So what might Sox11 be doing functionally in these early differentiating neurons? In order to answer this question, Chen et al. used mice with Sox11 coding sequence flanked by loxP sites, and crossed them with a Cre line expressed from the Emx1 promoter, knocking out Sox11 in all cells derived from cortical germinal zones. At early stages of corticogenesis, Sox11 conditional mutants showed reduced immunohistochemical staining for the immature neuronal marker Tuj1. At later stages, mutants showed reduced levels of the deep cortical layer markers Tbr1 and Ctip2. In vivo electroporation to create LOF Sox11 mutants further illustrated that Sox11 promotes the creation of early born neurons by sacrificing apical progenitor cells, as LOF mutants showed an increased proportion of Sox9+ progenitor cells compared to controls.

Sox11 co-localized with Ngn1, a known transcription factor critical for neuronal maturation, and co-immunoprecipitation experiments were performed using E14.5 cortex to assess the ability of Sox11 to bind Ngn1. Sox11 selectively bound Ngn1, suggesting it may be a binding partner with Sox11. Chen et al. also demonstrated that Sox11 could bind regulatory sequences in NeuroD1, an early neuronal differentiation marker in the forebrain regulated by Ngn1, via ChIP analysis. Follow up luciferase assays using an expression vector containing the NeuroD1 promoter linked to luciferase showed that Sox11, when transfected with Ngn1, had a synergistic effect on the activation of the NeuroD1 promoter. These results suggest that Ngn1 and Sox11 act as binding partners to regulate the expression of NeuroD1.

Chen et al. next assessed the function of Sox4 in IPCs in the SVZ. IHC showed that Sox4 co-localized with Ngn2, a transcription factor previously shown to be present in IPCs, as well as Tbr2, another marker of IPCs. In vivo manipulation of Sox4 levels was performed to ascertain the function of Sox4 in IPCs. Cells were transfected with GFP, along with either a Sox4 GOF or Sox4 LOF vector. Cortex was then stained for Tbr2. GOF cortex had significantly more Tbr2+ cells, and LOF cortex had significantly less Tbr2+ cells compared to control cells. At E12.5, conditional Sox4 mutants had a reduction in the number of Tbr2+ IPCs, indicating Sox4 involvement in IPC specification.

Given the co-localization of Sox4 with both Ngn2 and Tbr2, co-immunoprecipitation experiments using E14.5 cortex were performed, showing that Sox4 binds to both Ngn2 and Tbr2. Tbr2 is a known target of Ngn2, and ChIP analysis revealed that Sox4 is able to bind a promoter region on Tbr2. Transactivation experiments showed that Sox4, but not Ngn2, can increase Tbr2 expression, and that cotransfection of Sox4 and Ngn2 produces similar Tbr2 levels to Sox4 alone. These results suggest that Sox4 and Ngn2 may act via distinct molecular pathways to influence Tbr2 levels in IPCs. Luciferase assays showed that Tbr2 cannot increase Tbr2 levels alone, but when cotransfected with Sox4, Tbr2 levels were elevated over Sox4 alone. These results taken together suggest that Ngn2 and Tbr2 are involved in pathways used by Sox4 to induce IPC specification.

Taken together, these results suggest that Sox4 and Sox11 have distinct roles during corticogenesis, and suggest a model by which temporal regulation of corticogenesis may occur (see figure below). To hear more about Dr. Maria Donoghue’s research, please attend her talk on Tuesday, January 17 at 4pm in the CNCB Marilyn G. Farquhar Seminar Room. To learn more about her lab and for a list of recent publications, please visit her lab website:

Molly Kwiatkowski is a third year MD/PhD candidate, currently working in the Consortium for Translational Research in Neuropsychopharmacology.


What can invertebrates tell us about our brains?

With its hundred billion neurons and quadrillion synapses, the human central nervous system(CNS) can seem intractably complex. Fortunately, there is a class of animals whose nervous systems and behaviors are much more easily understood.  Invertebrates, such as sea slugs and worms, have on the order of only hundreds or thousands of neurons and their connections are extremely well stereotyped. This simplicity makes them amenable to experimentation and modeling, and has allowed scientists to understand the structure and function of their neural circuits.

In his review, Allen I. Selverston, Professor Emeritus at UCSD, asks if information gained from the study of invertebrates can be translated to our understanding of the human CNS.  He focuses on a particularly well characterized type of circuit called Central Pattern Generators (CPG).  CPGs are networks of neurons which produce rhythmic outputs in the absence of sensory feedback, and often control simple motor actions such as feeding or swimming. CPGs are not only found in invertebrates but vertebrates as well, where they control certain low level functions.  An example of a CPG is the leech heartbeat network which is shown in the diagram below.


Leech heartbeat neuronal network

The study CPGs using electrical and chemical manipulation of their constituent neurons has led to three primary types of discoveries.  First, it has revealed how a complex array of ion channels contributes to the distinct activity properties of individual neurons. Second, it has shed light on the types of synapses and how they are modulated and third, how circuits produce functional outputs.

Selverston uses these three types of analysis to explain how many different CPGs from the invertebrate world work. Unfortunately, he concludes that there are very few general principles for the design of these circuits that are transferable from model to model. Each CPG has its own evolutionary history that has crafted it into a bespoke circuit for the unique function that it serves. Moreover, the experimental methods used to study CPGs are unlikely to be effective in more complicated vertebrate systems because they cannot be probed with single cell techniques. This means that while the cellular and synapse level data may broadly applicable, the further study of invertebrate CPGs is unlikely to give us much insight into the human CNS.

Selverston’s review can be found here.

Leo Breston is a first year student in the Neuroscience Graduate Program. He is currently rotating in the Navlaka lab. 

Visual Action: Instructive Timing in the Primary Visual Cortex

By Tunmise Olayinka

Behavior in in the visual world necessitates reciprocal feedback between the environment and the observer. To act, an animal need to 1) sense the outside world, 2) compute upon this percept, and 3) generate an optimal response.

The long-held canonical view of the primary visual cortex (V1) is that its major role lay within only 1 and 2; that is, it functions as the initial computing gateway for the processing of visual sensory stimuli. This has been supported by the correlated spatial distribution of neurons in V1: they share a topographic mapping of response that in turn reflects the spatiotemporal structure of the visual data they process.

However Vijan Mohan K. Namboodiri and others in the audacious lab of Marshall G. Hussain Shuler have now posited that the V1 may play a broader, more instructive function; viz., in the direction of visually-responsive actions. They knew that, in visually-cued tasks, V1 predicated the learned interval between the stimulus and reward, in turn correlating with the action-response. The temporal consistency of this correlation led them to ask: is the learned timing they see in V1 used for solely sensory processing, or does it play a governing role in making actions and directing behavior?

Namboodiri et. al specifically ask these questions in rats, using a visually-cued interval timing task. In the task, the rats attempt to optimally time an action—when to a lick on a spout-in order to receive the maximal reward: water. The longer the rat patiently waits after the visual stimulus, the more reward it gets. However, this is up to a point: delays longer than the maximal delay, i.e. exceeding the target stimuli-lick interval, receive no reward. Thus within this task, the rats must compute an optimal timing for their licks, instead of simply waiting arbitrarily longer. With this task, Namoboodiri can now attempt to answer their focal questions 1) can we see representations of the the timing-delay between stimulus and reward across V1 neurons, and 2), does this representations instruct the action itself, by computing the prediction of the lapse of reward from the stimulus.

Thus with this design, they evaluated neurons in the V1 one at a time (i.e. via single unit recordings) finding that they had a variety of receptivities. Some neurons were entrained to this <i>mean expected interval between the stimulus and the reward, as predicted, while others instead represented the interval between the stimulus and the rat’s response itself, i.e. they were timed to the actual action (nosepoke entry), rather than the prediction of the delay-from-stimulus of the impending reward. Importantly, these visuotemporal representations correlated with the rats behavior: they only noted ‘interval-timing’ neurons in V1 when the rat successfully responded in visually-timed manner. In contrast, in non-visually timed trials, V1 neurons showed a consistent delay from nosepoke entry, independent of the visual stimulus. Only 2% (7/351) neurons in early training show significant action-timing (on the order of their false positive rate), helping to support that this ‘action-timing’ is indeed a computation on the interval itself; viz. the wait time-reward contingency, and not just timed to the action, and which <i>a priori, would not require visual feedback from V1.

If activity in V1 was entirely top-down processive, and driven by the action itself, their hypothesis is that it would present throughout V1, with no selectivity between the visually timed and non-visually timed neurons. However, if V1 played in a role in specifically generated and instructing the action, one would expect such corresponding visual task-based selectivity, with activity in V1 correlating with action on the visually-timed trials and not on the non-visually timed trials.

In addition, the intervals represented by these visually-timed, directive neurons were expected to demonstrate a trial-by-trial correlation with the neural representation of the interval and the action. The interval-representing neurons would reflect this timing in their firing profile, modulating their firing rate in correlation with the mean expected delay. So to instruct a delayed lick, for example, an interval neuron would simply increase the duration of its firing response. Similarly, the responding population—the action-timed neurons decoding the activity of these interval neurons, would in turn modulate their population firing rate with respect to the timing of the lick: the later they fire, the later the lick.

Namboodiri and his colleagues not only observed this behavior, they furthermore found that they could even predict the response of the action-timed neurons using the visually timed neurons, and only in that direction. Altogether, these results seem to demonstrate V1’s role in instructing timed action.

However, while suggestive, these results were only that—merely suggestive. The denouement of the experiment was to see if they could validate their hypothesis on the instructive nature of visually-timed neurons, by seeing if they could modulate this very instruction. Using the glorious power of optogenetics, they were able to consistently shift the firing rate, and thus the timing, of the action. To further delve more into a mechanistic explanation, they simulated their model using a reduced computational model. Therein, they demonstrated that this interval—the average delay between the predictive cues and the reward—could be locally generated and represented in V1.

So does V1 play an instructive role for stereotyping timing behavior on visual-cued tasks? This paper definitely motivates that idea. For visually-cued tasks, they could show that some neurons seemed to encode the interval, while others correlated to the action. They then were able to show that the antecedent firing response of the interval neurons could predict the population firing of the action-timed neurons. Then they not only showed they could directly perturb this effect—by modulating the visually timed interval neurons—but could validate it mechanistically within a computational model. Though one might argue that further elucidation into the nature of the interval representation is warranted (i.e., do these neurons represent the entire duration of the interval, or the endpoints / expiry?), altogether their results heavily suggest an excitingly novel and instructive role for the primary visual cortex.

Tunmise Olayinka is a third-year MD-PhD candidate at UCSD, currently in the labs of Bradley Voytek & Alysson Muotri.

Fragile X Syndrome: When translational regulation goes awry

Fragile X syndrome (FXS) is the most common hereditary form of intellectual disability affecting approximately 1 in 4,000 males and 1 in 6,000 females. The syndrome develops from a mutation in fmr1 on the q arm of the X chromosome, resulting in loss of RNA-binding protein FMRP, the fragile X mental retardation protein. The end result is a constellation of physical phenotypes, cognitive dysfunction, autistic behaviors, childhood seizures, and on a molecular leve, abnormal dendritic spines. Previous mouse models aimed to identify the deficits in neuronal plasticity have identified an increase in long, thin dendritic spines with an increased turnover rate and decreased response to input. Rapid protein synthesis is necessary for synaptic plasticity, which relies on the translation of existing mRNAs. Normal synaptic communication is dependent on spine dynamics and plasticity. Varying dysfunctions in circuits have been attributed to the loss of FMRP, however the mechanism by which FMRP affects plasticity, circuits and ultimately behavior was primarily unknown until recently.

Dr. Jennifer Darnell at the Laboratory of Molecular Neuro-Oncology, Rockefeller University is a leading expert on FXS. Her laboratory has recently used a new technique to begin to understand the pathophysiology of FXS and the molecular function of FMRP. High throughput sequencing cross-linking immunoprecipitation (HITS-CLIP) uses ultraviolet irradiation to create covalent bonds between proteins and RNA molecules that are in direct contact. By using this method, Dr. Darnell was able to identify 842 FMRP target mRNAs in mouse brain, which were increased in pre- and postsynaptic proteins: NMDA receptor subunits and metabotropic GluR5 receptor were among the several post synaptic mRNAs that interact with FMRP. This supports previous studies that show the increased turnover rate, increased vesicle recycling and increased vesicle pools in Fmr1 KO mice.

While this determined the binding of FMRP directly to mRNAs, it remained to be shown how this affects translation. As expected, Fmr1 KO mice exhibit increased rates of brain protein synthesis. However, the degree of increased synthesis is far more than can be explained simply by the FMRP target mRNAs. This led to the realization that in addition to the direct increase in protein synthesis, there is a global increase in protein synthesis possibly due to the downstream changes in elongation and initiation factors caused by the loss of FMRP.

But how does the presence of FMRP limit translation? Using a brain polyribosome-programmed in vitro translation system it was demonstrated that there is ribosome stalling that occurs at FMRP target transcripts. Thus, the loss of FMRP results in relief of the ribosome stalling and an increase in translation. Several of the proteins affected have been linked to the phenotypes seen in FXS: NMDA and mGluRs affecting synaptic plasticity, ERK and mTORC1 effecting neuronal translation, cAMP and several GTPases which have been shown to alter spine morphology, and SYNGAP1 which has been linked to non-syndromic mental retardation and autism spectrum disorder. The knowledge of the several pathways affected by the loss of FMRP give way to novel therapeutic approaches including most notable the use of antibiotics such as minocycline which repress translation, in order to alleviate some of the increased translational burden in FXS.

The translational role of FMRP both directly and globally, and the significant clinical phenotypes caused by the Fmr1 mutation, is an example of a minor genetic change causing catastrophic downstream effects. While there is still no cure for FXS, understanding the pathophysiology of the disease has allowed researchers to begin to test possible therapeutic approaches, many of which show promise. Dr. Darnell’s work has been integral to the understanding of not only FXS, but synaptic function, autism spectrum disorder, and molecular biology. To learn more about Dr. Darnell’s work and a list of publications, please visit her website at

Amy Taylor is a third year MDPhD candidate at UCSD in the Schizophrenia Research Program.

GCN2 kinase: protector from death by ribosome stalling

Dr. Susan Ackerman at UCSD focuses her research on the molecular mechanisms involved in maintaining homeostasis during development and aging of the mammalian brain. She is particularly interested in how altered translation elongation, caused by ribosome stalling, affects neuronal function and survival.

In her recent paper, “Activation of GCN2 kinase by ribosome stalling links translation elongation with translation initiation” Dr. Ackerman addresses the issue of determining the signaling pathways initiated by ribosome stalling. In order to study these signaling pathways, Dr. Ackerman made use of the mutant mouse line, C57BL/6J-Gtpbp2nmf205-/- , which have stalled neuronal elongation complexes. She first performed gene expression studies on isolated cerebella from control B6J mice and mutant B6J-Gtpbp2nmf205-/- mice at 3- and 5-weeks of age. Dr. Ackerman found a total of 910 and 325 differentially regulated genes in the 5-week old and 3-week old mutant cerebellum, respectively. Since ribosome stalling, as seen in these mutant mice, has been shown to cause neurodegeneration, she next performed Kegg pathway analysis and Ingenuity Pathway analysis on the differentially regulated genes and found enhanced inflammation/immune pathways. Interestingly, when these differentially regulated genes were compared to activated genes in microglia and astrocytes from mice models of amyotrophic lateral sclerosis (ALS) and Alzheimer’s disease, Dr. Ackerman found overlaps with 150 and 60 genes expressed in the activated microglia and astrocytes, respectively.


Further analysis of the differentially expressed genes revealed upregulation of activating transcription factor 4 (ATF4), which is an important component of the integrated stress response. Next, Dr. Ackerman studied the activation of ATF4 in the mutant mice and found ATF4 target genes were upregulated in both cerebellum and hippocampal tissue. Additional analysis revealed that the activation of ATF4 was dependent on GCN2 kinase, which is activated by amino acid deprivation.

Dr. Ackerman then proceeded to study the effects of the GCN2-ATF4 pathway on neuron survival. To do this, she compared the progression of neurodegeneration induced by ribosome stalling in two mutant mice strains. The first was the B6J-Gtpbp2nmf205-/- strain described earlier, and the second was the B6J-Gtpbp2nmf205-/-;Gcn2-/- strain. Dr. Ackerman found that the B6J-Gtpbp2nmf205-/-;Gcn2-/- strain showed increased granule cell death as compared to the B6J-Gtpbp2nmf205-/-. Additionally, the B6J-Gtpbp2nmf205-/-;Gcn2-/- strain showed extensive cell death within the CA1 region of the hippocampus which was not present in the B6J-Gtpbp2nmf205-/- strain.


Dr. Ackerman’s paper, “Activation of GCN2 kinase by ribosome stalling links translation elongation with translation initiation” provides a very detailed examination of the ribosome stalling triggered GCN2-AFT4 signaling pathway.

To hear more about Dr. Susan Ackerman’s research, please attend her talk on Tuesday, November 29, 2016 at 4pm in the CNCB Marilyn G. Farquhar Seminar Room.

Oscar Gonzalez is a first-year graduate student in the neurosciences graduate program and a member of Dr. Maxim Bazhenov’s lab. He is interested in the mechanisms leading to hypersynchronous activity in the brain, and the origin of resting state infra-slow fluctuations.

Spatial navigation strategies: where it’s at

When you wake up in the morning and head over to the lab, do you take the scenic route and relish in the San Diego sun, or do you take the shortcut through the library because your cat made you late because he craved attention? Memory experts like Dr. Véronique Bohbot at McGill University have begun using virtual and real environments to probe the distinct navigation strategies recruited by your brain when, for example, you orient yourself within a mental map through your morning commute. These navigation strategies are often conjured spontaneously to adapt to the current environment, such as a closed crosswalk, and vary in the amount of localized brain activity evoked in an individual, which in turn depends heavily on the grey matter volume of those specific areas, hormones, and genetic background. However, when someone’s healthy brain fails to employ optimal navigation strategies, they might only experience the mild inconvenience of being late by a few minutes, a sharp contrast to the underlying protracted spatial memory dysfunction found in Alzheimer’s Disease (AD) patients.

Long-term spatial memory impairments are found early in the development of AD, when diagnostic genotyping often reveals the presence of Apolipoprotein E (APOE), a prominent risk gene associated with the disease. A particular APOE allele, ε4, has considerable ties to the cognitive impairments and hippocampal atrophy associated with aging. Interestingly, a different allele, ε2, is known to be protective against AD neuropsychological symptoms, such as cognitive decline and neuritic plaque formation. However, most of these findings have come as a result of studies on older adults with AD onset or progression, and despite the contrast between the structural, protective or risk increasing qualities of the two alleles, it was Dr. Bohbot’s group who recently proposed that cognitive correlates are sensitive to the genotypes even in young adults.

Lesion studies support the idea that different strategies employed while navigating an environment rely on divergent brain networks. For example, the hippocampus-dependent spatial strategy involves creating relationships between the different landmarks in the environment to incorporate into a cognitive map. On the other hand, the caudate-dependent response strategy incorporates stimulus-response associations to orient yourself in space (“take a left after the second right”). In addition, the neuroanatomical basis for these two different spatial strategies also have a structural inverse relationship, such that greater gray matter volume in one correlates with less gray matter volume in the other, and vice versa. Konishi et al. (2016) used this converging evidence to assess whether recruitment of these strategies would relate to the structural differences found between young adult APOE allele carriers, hypothesizing that APOE ε2 carriers will utilize spatial strategies more, and in turn have greater gray matter volume in the hippocampus, in comparison to ε3/ε3 and ε4 allele carriers.

To test this hypothesis, genotyped participants underwent testing in a computer-based virtual reality navigation task that is akin to the eight-arm radial maze, except with landmarks more common to human environments. A subsequent verbal report of the navigation strategy they employed (i.e. “I used landmarks” vs “I used the patter of open pathways”) allowed the researchers to classify participants between spatial and response learners. Follow-up structural Magnetic Resonance Imaging (MRI) on the participants measured hippocampal structural differences between allele carriers and their most utilized navigation strategy.


Schematic drawings and first person views of the 4-on-8 virtual maze


Interestingly, their hypothesis was a home-run. A higher proportion of ε2 allele participants reported using the hippocampus-dependent spatial strategy throughout the task compared to the other allele carriers.



Apoliprotein E (APOE) e2 carriers used the hippocampus-dependent spatial strategy more than the other genotype groups.


In addition, ε2 allele carriers had greater grey matter in the hippocampus compared to both ε3/ ε3 and ε4 carriers. However, these measurements were conducted only in a subset of the 100+ participant pool from the behavioral task. Despite this, these results aligned with previously published studies looking at hippocampal structure in ε2 allele carriers.


Gray matter contrast of APOE e2 carriers and non-e2 carriers using voxel-based morphometry (VBM)


While the ε4 APOE allele has garnered the most attention due to its association with increased risk for AD onset, pursuing assessments on what makes young ε2 allele carriers cognitively distinct from the others can lead to the creation of early intervention strategies. For example, this particular study implies a future clinical scenario where training ε4 carriers to use spatial navigation strategies might mitigate the spatial learning impairments seen throughout AD progression.

Come check out Dr. Bohbot’s talk, titled “Early detection, sex differences, and intervention in healthy older adults at risk of Alzheimer’s disease”, on Tuesday, November 11th, at 4 P.M. in the CNCB  Marilyn Farquhar Seminar Room.

Christian Cazares is a first-year neuroscience graduate student in the Gremel Lab, where he is looking at the effects of stress on goal-directed and habitual behavior. He can be reached at @fleabrained and

Konishi K, Bhat V, Banner H, Poirier J, Joober R, Bohbot VD. APOE2 Is Associated with Spatial Navigational Strategies and Increased Gray Matter in the Hippocampus. Frontiers in Human Neuroscience. 2016;10:349. doi:10.3389/fnhum.2016.00349.