Bridging the Gap: Neuroscience and Mental Health

The mere mention of the phrase “translational gap” can bring shivers down the spine of basic researchers and clinicians alike. The use of this term is meant to capture the magnitude of the task of transforming any understanding of how the nervous system functions to its realization as a therapeutic intervention or available technology in a human population. No investigator is more vulnerable to the sleep-depriving capabilities of this proverbial ‘boogey-man’ than one who focuses on neuropsychiatric disorders and, more broadly, mental health.

As of today, the level of resolution and specificity with regards to tools stand in opposition with the model organisms that may provide the most neurobiological similarity. Working with rodents offers the incredible advantages of genetic and viral manipulation, single cell monitoring and imaging capabilities, a modest behavioral repertoire, and (maybe most importantly) the access to all of these techniques in a mammalian brain. However in spite of all these attractions, the cognitive character of neuropsychiatric disorders and mental health make the translational gap separating insights gained from the rodent nervous system and their application to clinical populations dauntingly vast.translational_gap                                                                                                        Credit: B. MELLOR

Dr. Jyoti Mishra, founder of the Neural Engineering and Translation Labs (NEATLabs) at UCSD, employs a highly collaborative and interdisciplinary approach to help bridge the gap. By leveraging the latest monitoring and neural-interfacing technologies, Dr. Mishra examines the relationship between the neural activity of human subjects (measured using mobile cognitive assessment tools, EEG and fMRI) with more invasive and small-scale circuit manipulation in rodents in response to homologous sets of behavioral assays designed to assess neural and cognitive function. By this ‘translational transitive property’ a link can be drawn between circuit manipulations which give rise to any given cognitive or neural change as measured by a particular test and their accompanying neural fingerprint present in human subjects displaying similar responses to that test.

In her 2014 paper Dr. Mishra examines the phenomenon of aging and its association with deficits in the ability to ignore distractions. In this study she developed a targeted cognitive training intervention designed to train older rats and older humans to be able to filter out increasingly disruptive distractions. This is distinct from older tests formulated in efforts to recovery this age related decline in attention in that instead of modeling the task around training the subject to focus on increasingly subtle task-relevant stimuli, Dr. Mishra focused exclusively on the ability to ignore distractions. This may seem like a case of semantics, but the neurobiology says otherwise. There has been work to show that the circuits underlying neural enhancement and neural suppression are distinct (Chadick and Gazzaley, 2011). It follows that, unsurprisingly, cognitive training regiments targeted at stimuli-identification did not improve the deficiencies in distraction suppression displayed by older populations.


Mishra, Gazzaley, et al. 2014

Shown above are the differences in responsiveness of neurons in the Auditory Cortex of rats both trained (ADT) and untrained(UT). The right graph shows a marked suppression of neural activity correlated with distractors, while the response to the task-relevant auditory cue (Oddballs) is unaffected. Shown below is the corresponding data from a cohort of older humans, showing a similarly selective suppression of neural activity in response to distractors. The dotted line shows neural responses in the auditory cortex at the beginning of training and the solid line represents a performance at the end of training regiment. figure_human.PNG

                                                                                            Mishra, Gazzaley, et al. 2014

This inter-species approach to developing potential therapeutic interventions as well as elucidating the circuitry at work is an essential characteristic of Dr. Mishra’s approach to her work. This conceptual framework also serves as the foundation on which NEATLabs operates as they continue to make strides towards bridging the translational gap.

To hear more about the work being done in Dr. Mishra’s lab, please join us at 4:00pm, Tuesday 10/15/2019 at the Marilyn G. Farquhar Seminar Room.

Blogpost written by Peter Fenton: 1st year Neurograd Ph.D student currently rotating in the Spitzer Lab studying neurotransmitter switching.


The Social Brain: Correlated Neural Activity During Social Interactions

Although the brain is involved in powerful computations that construct our reality as well as control our motor and visual systems, many of these computations can take place in isolation from other animals. While this circuitry is interesting, it lacks the profound influence our social lives impact our biology. Because our world is largely socially constructed, we are fundamentally influenced by differing social interactions. Thus, it calls into question the neural basis of socially interacting brains and the specific computations that underlie a social interaction.

Classically, the way of understanding social behavior is through testing paradigms with which one animal is the subject. However, Dr. Weizhe Hong, an Assistant Professor in Neurobiology and Biological Chemistry at UCLA, joins us this week to uncover what is happening at the neuronal level between two socially interacting mice through simultaneous recordings of their brains.

In July of this year, the Hong lab published a paper in Cell displaying their findings: “when two animals interact, neural activity across their brains synchronize in a way that predicts how they will behave and how they form social dominance relationships” (Kingsbury et. al 2019, Cell). To get their findings, the lab utilizes a new microendoscopic calcium imaging technique.  They injected mice with an adeno-associated virus that selectively fluoresces neuronal cells after their release of calcium. Above the site of the viral injection, they implanted a gradient refractive index (GRIN) lens that gives higher resolution to the fluorescently-labeled cells during recordings. The GRIN lens and camera setup on a mouse is pictured below (NeuroNex):


Specifically, Kingsbury et. al decided to record neuronal populations from the medial prefrontal cortex (mPFC), where it is known as the central hub that processes both social information as well as representations of social status (Utevsky and Platt, 2014; Wang et al., 2011; Zhou et al., 2017). From here, their experimental design was elegantly simple. They placed two mice in an open arena where they recorded both their behavior and neuronal firing in the mPFC. To understand the social interactions between the two mice, Kingsbury et. al annotated the videos for both non-social and social behaviors. They found that 66% of the behavior that the mice conducted was social behavior directed at their paired mouse. During these social interactions, the neural activity of the two animals were analyzed and showed a significant correlation compared to a randomized control, displayed below:  (Figure 1K,L)


Utilizing this significant correlated neural activity between socially interacting mice, Kingsbury et. al placed mice of differing social rank to freely behave together. They found that in differing social rank interactions, there was a higher correlation of neural activity compared to similar ranked littermate controls. These and other findings from the paper have interestingly shown interbrain synchrony and “sets the groundwork for a more incisive investigation of the emergent neural properties of multi-individual systems, which may yet reveal a richer and deeper understanding of the social brain within this truly social world” (Kingsbury et. al, 2019, Cell).

To hear more about the work being done in Dr. Weizhe Hong’s lab, please join us at 4:00pm, Tuesday 10/1/2019 at the Marilyn G. Farquhar Seminar Room.


Chiaki Santiago is a 1st year neurosciences PhD student currently rotating in Dr. Olivier George’s lab. You can find her on Twitter and LinkedIn.

Directly changing disordered brain activity in depression with Deep Brain Stimulation

Dr. Helen Mayberg, MD is a professor and neurologist at Mount Sinai in New York City. She has gained a reputation as an authority in the study and treatment of depression. Focusing at the circuit level on the etiology of depression, Dr. Mayberg formulated a novel theory of the systemic dysfunctions involved in depression. Her work showed how many disparate areas of the brain interact with each other in depression, eventually discovering evidence pointing toward a region of the brain, Brodmann Area 25 (BA25), whose (dys)function is of central importance in the condition. This area’s activity was demonstrated to be connected to the experience of extreme negative affect. In the context of the brain as a whole, BA25 represents a good candidate for a network hub that coordinates the different regions involved in bringing about the unhealthy mental states associated with major depression.

Having uncovered the central role of BA 25, Dr. Mayberg subsequently turned to the goal of translating her discovery into an effective therapeutic. This is something which is sorely needed, depression being one of the most significant causes of morbidity and lost productivity throughout the world. Dr. Mayberg moved forward on the hypothesis that if BA 25 was an important coordinating region in depression-related pathological states, modulating its activity may have an overall salutary effect on mental health. This modulation can be achieved by the use of Deep Brain Stimulation, wherein a pacemaker-like device is surgically implanted deep within the brain in order to deliver electrical current to precisely alter the activity of a tiny target region of the brain. DBS is an effective treatment for the neural degeneration associated with Parkinson’s Disease, but it also represents a promising topic of research in the development of therapeutics for a wide variety of psychiatric and neurological conditions. Dr. Mayberg and her collaborators thus became pioneers in the use of DBS in depression; beginning nearly two decades ago, the team began to successfully treat depressed patients who had previously failed to respond to many other treatments. DBS remains an important emerging technology in the alleviation of depression, with more research coming out every year.

One of Dr. Mayberg’s more recent publications, published in April of 2017 in the Nature journal Molecular Psychiatry, focused on a connectomic approach to precisely target the placement of DBS devices in interventions for depression. Before surgery, the researchers used high-resolution MRI scans to perform Diffusion Tensor Imaging (DTI), an imaging method that can reveal the connections (white matter fiber tracts) between different areas of the brain in three dimensions. Using a technique called Determinisic Tractography, they used these images to determine the strength of the connections between those regions identified as pathological in depression, including those networked together with BA 25. The image included here (Figure 2 from the paper) provides a visual representation of these connections: the panel on the left is a global view of the four main fiber bundles that converge on BA 25 which are important in the mechanism of successful DBS treatment, and on the right a view of one patient’s network connections (tractography) is depicted. They used this connectomic knowledge to tailor the placement of DBS contacts within the subcallossal cingulate gyrus (a region which includes BA 25). After surgery and the initiation of DBS treatment, they performed further tractography to identify connectomic changes associated with the therapeutic and further tailor future patient’s surgeries. Overall they achieved a positive response in 9 out of 11 patients after one year, 6 being in remission. This study represents a significant step forward in the refinement of DBS into a precise and effective treatment for a devastating disease.


Zarek Siegel is a first-year PhD student working in Eric Halgren’s research group.

“Jump” from the album Totally Tau-ed Out

Dr. Karen Duff, Professor of Pathology and Cell Biology at Columbia University and recipient of the Potamkin Prize in 2006, has done notable work focusing on the development and characterization of Alzheimer’s Disease (AD) mouse models and amyloid deposition. One hallmark of AD is the misfolded and hyperphosphorylated form of a microtubule stabilizing protein called tau. In early stages of the disease, tau is found in the entorhinal cortex (EC), but then spreads medially and enters limbic and neocortical areas at later stages. Lately, she has been investigating tauopathies and mechanisms of how tau spreads, aiming to identify therapeutic approaches for prevention of these neurodegenerative disease.

In a recent 2016 Nature Neuroscience paper, Duff and her lab found that tau protein can spread through the brain by jumping from one neuron to another via the extracellular space that surrounds the brain’s neurons. Using iPSCs and a combination of fluorescent reporters in donor cells (containing or not containing tau) and recipient cells, they show that physiological tau from donor cells can get released and transfers to recipient cells. Tau aggregation occurs in a time dependent manner, and they found that tau aggregates are transferred to recipients as well.

With a novel microfluidic system, they demonstrate that the tau generated in recipients can then be transcellularly propagated to other distant recipient cells. Tau can travel long distances within the neuron before it gets released, migrating to other brain regions. This may explain why only one area of the brain is affected during early stages, but multiple areas are affected later on. But what is the mechanism of this propagation? Through a slew of molecular techniques of western blots, immunoprecipitating, immunofluorescence staining, and more, their experiments reveal that tau is released into the medium from cells expressing it at physiological levels. By transferring nonmutant tau in conditioned media from human iPSCs or wild type mice to primary neuronal cultures, tau was successfully internalized from the extracellular space. They showed that tau release and internalization occur at physiologically relevant levels, without requiring overexpression of tau.

Through stimulation of neurons with optogenetic and chemogenetic DREADD techniques, increases in neuronal activity leads to increased tau levels in the media and also increased and accelerated tau spread in vitro. Their in vivo experiments reveal that tau pathology and neurodegeneration get exacerbated when neurons are stimulated in hippocampus (Figure 8). This finding suggests treatments that increase brain activity, like deep brain stimulation, should be administered cautiously in people with neurodegenerative diseases and tauopathies. Other clinical implications include targeting therapies to tau when it is present in the extracellular space instead of within neurons.

Screen Shot 2019-05-08 at 1.53.55 PM

Figure 8. Increased neuronal activity through chemogenetic stimulation for 6 weeks shows accelerated tau pathology in EC. Brain tissues of stimulated and non-stimulated hemispheres in mice are labeled with hTau (green), NeuN (magenta), and DAPI (blue).

To hear more about the work being done in Dr. Duff’s lab be sure to join us at 4 PM, Tuesday 5/14/2019 at the Marilyn G. Farquhar Seminar room in CNCB.

To read the paper, visit:

Vivian Ko is a first-year PhD student.

How do neurons ‘see’ objects? Understanding computations used by face cells

This week we’re excited to hear from Dr. Doris Tsao on computations underlying object recognition in visual cortex. Her research focuses on computations used to identify and encode objects, specifically in the inferotemporal cortex (IT), a higher order area of the ventral visual pathway, which is colloquially known as the ‘what’ pathway for processing the content of visual stimuli. Dr. Tsao carried out hugely influential studies on how face patches, specialized regions of IT, encode faces in macaque monkeys.


Many studies on object recognition presented isolated objects to macaques. However, naturalistic behaviors involve accurately identifying multiple objects in a complex scene. A recent study published by the Tsao lab tried to explain how the visual system is able to effectively identify multiple objects in the presence of visual clutter, by examining the response of IT face cells presented with faces in isolation and with irrelevant objects. Previous studies on this subject disagree on what types of computations are used by these neurons to reliably fire in the presence of a preferred object despite visual clutter. While some observed ‘winner-take-all’ behavior—where activity patterns only depended on the presence of the preferred object, with no effect of other objects—others hypothesized that cells respond according to an averaged preference of the objects presented.


These conflicting models may have arisen because previous studies recorded from random IT neurons, selected without prior knowledge of their object preference. The Tsao lab approached this question by focusing on neurons in the macaque ML face patch, presenting several combinations, arrangements, and contrasts of the known preferred object (a face) and a non-face object. Their recordings from face cells showed a remarkable result: depending on the arrangement of the stimuli—specifically, whether the faces were shown on the contralateral or ipsilateral side, with the object on the opposite side—the cells displayed different neural computations. If a face appears contralaterally and an object appears ipsilaterally, cells ignore the object—a ‘winner-take-all’ response. But if the face and object switch sides, the neural response appeared to be a weighted average of responses to the face and object, with weights dependent on the contrast between the two stimuli.


The investigators also recorded face cells while presenting two faces to the macaque. The Tsao lab has developed a routine for quantifying face stimuli by using a face database to distill realistic variation between faces into 6 dimensions. This provides a way to both generate sets of face stimuli that evenly cover the space of natural variation and numerically describe the tuning of face cells to particular types of faces. As expected from their previous work on face cells, the face cells exhibited tuning to particular dimensions of face variation; this tuning was consistent between faces presented on the contralateral and ipsilateral sides. However, they were surprised to see that when two faces were presented together, cells exhibited a ‘contralateral-take-all’ response, where tuning to the contralateral face was unchanged and absent for the ipsilateral face.


In order to reconcile these observed integration methods, the Tsao lab then proposed a model where IT cells perform a weighted normalization computation. Normalization scales a neuron’s response by the overall activity level of its neighboring cells; when cells are arranged in patches of object category preference, this integration method can generate ‘winner-take-all’, ‘contralateral-take-all’, and weighted average responses. A model of the normalization computation using parameters estimated from LFP measurements from face patches was able to explain the data from the above experiments with relatively high accuracy. These integration patterns were also found in analogous experiments on macaque body-selective patches, suggesting that normalization computations are a generalizable strategy in IT patches. And because this integration strategy is dependent on cells residing in a neighborhood of similarly-selective cells, they must be observed within object-preferring patches, hence why it has been difficult to observe this integration pattern in the past.


For more insight on the function and organization of object identification areas, check out Dr. Doris Tsao’s talk on Monday 4/30/19 at 4pm in the Marilyn G. Farquhar Seminar room in CNCB.


Jessica Du is a neurosciences PhD student, currently a first-year working with Dr. Sung Han at the Salk Institute.


What’s the role of motor cortex in learning and executing motor skills?


Bence Ölveczky studies the learning and execution of motor skills in his lab at Harvard. His lab uses both songbirds and rodents to understand complex sequence and timing learning and execution.

In a study published in Neuron in 2015, the Ölveczky lab used a rodent model to investigate the role of motor cortex in learning and executing a motor skill. They specifically sought to address the degree of autonomy that subcortical circuits have during movement execution.

Motor cortex has previously been shown to be necessary for executing dexterous movements, so they developed a task that did not involve dexterity. Rats were trained to press a lever twice with a distinct inter-press interval. After weeks of training, rats developed stereotyped movements to complete the task. (See supplemental video of stereotyped rat movements here)  

Figure 1 alternative

After rats had been trained to criterion, Ölveczky and colleagues lesioned primary motor cortex specifically in the region controlling the forelimb the animal used for lever-pressing. Initially, rats were impaired on the contralesional side, but after 10 days, animals moved around normally.

When they tested the recovered rats on the spatiotemporal task, they found that the rats maintained the same trajectory and speed of their movements to solve the task. This provides evidence that motor cortex is not necessary for executing complex motor sequences.

Finally, they tested whether or not rats could learn the task if they had motor cortex lesions. Another cohort of rats were lesioned and subsequently recovered for 6-21 days. Then, the rats were introduced to the lever-pressing task.

They found no difference between lesioned and control rats in the early stages of learning the task (during the first 1000 trials). Both groups of rats had similar means and distributions of inter-press intervals, similar reward rates, and similar motivation for the task as measured by number of total lever presses in a session. These data suggest that early exploratory lever-pressing is not affected by motor cortex lesions.

However, none of the lesioned animals learned the task to criterion, even when trained 3 times longer than the control animals (figure D). This is seen in the mean inter-press interval and variability in the figure B and C. Then, when they added a needed delay between trials, only 1/11 lesioned animals learned to wait. This indicates that the animals had difficulty learning the larger structure of the task.

Figure 1 alternative

To conclude, Kawai and colleagues probed the role of motor cortex in non-dextrous motor learning and execution, and found that motor cortex is only necessary for learning the task but not executing a learned task.

Please join us at 4pm in the Marilyn G. Farquhar Seminar room in the Center for Neural Circuits and Behavior on Tuesday April 9, 2019. Bence Ölveczky will present on neural circuits underlying motor skill learning and execution.

Talk description:

“I will introduce a motor skill learning paradigm that trains stereotyped complex motor sequences in rodents. By recording and manipulating neural activity in the basal ganglia, motor cortex and thalamus, we delineate the logic by which these circuits work together to promote the acquisition and control of task-specific motor sequences.”


Emily Baltz is a 1st year neurosciences PhD student interested in how internal states influence action selection. You can find her on Twitter, Google Scholar, and LinkedIn.

Founder’s Day Lecture in honor of Dr. Robert Galambos: Dr. Richard Palmiter on tracking the roots of conditioned taste aversion

Founder’s Day Lecture in honor of Dr. Robert Galambos

Dr. Richard Palmiter from the University of Washington

Tuesday, April 2, 4pm in CNCB Large Conference Room


This Tuesday afternoon we have the Founder’s Day Lecture in honor of Dr. Robert Galambos, which will be given by Dr. Richard Palmiter.


First a few words on Dr. Galambos – Dr. Galambos (1914-2010) was a pioneer neuroscientist who is famous for his discovery of echolocation used by bats for navigation. His research in this field paved the way to further understanding of auditory processing. His work exemplifies how even though groundbreaking work that goes against entrenched dogma can sometimes find trouble establishing a foothold, good, sound scientific research prevails in the end. Please see his obituary in the NY Times:


Dr. Richard Palmiter contributed to a technique that is giant in its own right: the development of transgenic animals. In 1981, his group along with some others showed that injection of a fusion gene into the pronucleus of an egg could result in genetically altered animals. His group went on to develop the methods for not only integrating genes into an animal’s cells, but into its germline in a way that would stably pass the gene to all their progeny. For a historical perspective, please see:


On Tuesday, Dr. Palmiter will be speaking about his recent work published in Neuron last year titled “Parabrachial CGRP Neurons Establish and Sustain Aversive Taste Memories” ( In brief, they found that a group of neurons expressing calcitonin-gene-related peptide (CGRP) located in the parabrachial nucleus (PBN) were sufficient and necessary to establish conditioned taste aversion (CTA).


In their paper they show CTA caused by LiCl and LPS paired with a novel taste activates CGRP neurons in PBN. They then use CNO with AAV1-DIO-hM3Dq:mCherry injected mice to activate the same CGRP-expressing neurons of PBN and find that they can cause CTA (see figure). Furthermore, they show that KO out Arc or Grin1 in these neurons before aversive taste conditioning prohibited formation of CTA, while only Grin1 KO after conditioning impaired CTA.


They then silenced the PBN CGRP-expressing neurons with tetanus toxin after establishing CTA. This abolished the CTA phenotype.


CGRP neurons in the PBN have robust projections to the CeA, BNST, and VPMpc. They then tested whether stimulation of these projections could induce CTA. They found that stimulation of projections to CeA or BNST decreased food intake for a short time while stimulation of projections to VPMpc had no effect.

Screen Shot 2019-03-31 at 11.10.51 PM

Figure from Chen et al. 2018 Neuron. Drive of CGRP expressing neurons in PBN is sufficient to cause CTA. Using the construct in Figure C, Palmiter’s group induced activity of CGRP neurons in PBN. Figure D shows how induction of CGRP neurons in PBN causes profound CTA (drastically decreased sucrose preference).


This paper from Dr. Palmiter’s group identified a group of neurons which express CGRP in the PBN that are necessary and sufficient for formation of CTA. Through a series of experiments they neatly show that driving these neurons leads to CTA, while silencing them impairs CTA formation.


We look forward to hearing more about the work and details from Dr. Palmiter in person on Tuesday, April 2, 2019. The seminar will be held in CNCB Large Conference Room at 4pm.


See you there!

Ben Tsuda


Ben Tsuda is an MD/PhD student in the Neurosciences Graduate Program in Dr. Terry Sejnowski’s group in the Computational Neuroscience Laboratory at the Salk Institute.

Uncovering trail markers for adventurous cortical interneurons

Let’s step back and appreciate our inhibitory interneuronsit’s something many of us don’t do enough, given that around 20% of cortical neurons are inhibitory. But the Fishell lab is working to make sure these cellsRamon y Cajal’s “butterflies of the soul”get their share of the limelight. Dr. Gordon Fishell’s career has explored the mechanisms of cortical development; these days, his lab’s work in Harvard’s Neurobiology Department elucidates interneurons’ transcriptional and activity-dependent regulation of development, their circuit integration, and their function in disease states.


Cortical interneurons are wildly diverse in aspects like morphology, connectivity, gene expression, and intrinsic electrophysiological properties. At the same time, the great majority of these cells arise from the medial and caudal ganglionic eminences (MGE and CGE). From these regions, interneurons must then migrate tangentially to reach the cortical plate, and then migrate radially to integrate into the various layers and circuits of the developing cortex. How is diversity in interneurons generated from a more homogenous progenitor pool? What controls the multi-step process of migration, circuit integration, and maturation for these cells?


With these questions in mind, the Fishell lab observed that Prox1, a known regulator of neural development, is expressed in CGE-derived but not MGE-derived cells, and decided to test its role in differentiation and migration of CGE-derived interneurons. These cells differentiate into 3 main classes of interneurons, identified based on expression of Reelin (RELN), vasoactive intestinal peptide (VIP), or VIP and calretinin (CR). They specifically inhabit the superficial cortical layers. Knocking out Prox1 in cortical interneurons causes deficits in radial migrationCGE-derived interneurons fail to exit tangential migration to enter the cortical plate. By P7, this manifests as a displacement of interneurons from cortical layers I/II to layers V/VI. Further, the number of CGE-derived interneurons is decreased, suggesting that Prox1 is also involved in the survival and maintenance of these cells.


When Prox1 was knocked out postnatally, migration into superficial cortical layers was not deficient. However, the authors did observe decreased cell survival and abnormal morphology, especially in VIP/CR interneurons. To investigate this maturation deficit in isolation from other migration-related deficits, they transplanted immature cells from embryonic brains into postnatal developing brains, allowing the cells to integrate into the cortex and mature without migrating. Whereas VIP/CR cells with functional Prox1 developed the proper bipolar morphology, cells lacking Prox1 again showed severely changed morphology. Thus Prox1 plays an important role in the postnatal development and maturation of CGE-derived interneurons, independent of its function in regulating these cells’ migration during embryonic development.


An external file that holds a picture, illustration, etc. Object name is zns9991577540009.jpg


Read the full publication here. For more insight on the development of inhibitory interneurons as well as cholinergic neurons, check out Dr. Gordon Fishell’s talk on Tuesday 3/12/19 at 4pm in the Marilyn G. Farquhar Seminar room in CNCB.


Jessica Du is a neurosciences PhD student, currently a first-year working in Dr. Byungkook Lim’s lab. 


Friend or Foe? The Innate Immune System’s Responses to self-DNA

With flu season at its peak and our heightened awareness of every cough and sneeze around us, we may begin to wonder how our bodies respond to such microscopic particles. Thankfully our cells have developed complex mechanisms to detect pathogens that sometimes infiltrate our bodies. Dr. Zhijian “James” Chen and his lab at the University of Texas Southwestern Medical Center investigate these pathways and cellular components involved in immune and stress responses. Using cellular and biochemical approaches, the Chen lab studies how the innate immune system senses cytoplasmic nucleic acids and how the immune system triggers downstream signaling cascades resulting in inflammation and eventually restoring homeostasis.

Through this work, Dr. Chen discovered proteins important for host immune responses, such as the mitochondrial protein MAVS and cGAS-cGAMP pathway components. cGAS detects cytosolic double-stranded DNA by binding to foreign or self-DNA and producing the second messenger, cGAMP. cGAMP then activates the adaptor protein STING which recruits kinases, eventually leading to the production of type I interferons and inflammatory cytokines. Importantly, these mechanisms utilized to detect invading viruses and bacteria could also be activated by other insults or stressors that cause DNA damage. Dr. Chen and his lab are interested in figuring out how activation or suppression of these pathways can lead to diseases such as cancer, autoimmunity, and age-related neurodegeneration.


Figure 1

Recent work from the Chen lab focused on the role of cGAS-cGAMP signaling in cellular senescence (Yang et al., 2017, PNAS). When cells are damaged, they sometimes enter a state of irreversible cell cycle arrest called cellular senescence (Figure 1). In this state, cells secrete inflammatory cytokines, growth factors, and proteases, termed the senescence-associated secretory phenotype (SASP). Activation of DNA damage responses had previously been shown to be required for initiation of cellular senescence, but the exact connections between senescence and the cGAS-cGAMP pathway were still unknown. Using both in vitro and in vivo models, Chen and colleagues discovered that cGAS is required for both spontaneous and stimulated senescence and SASP. Their data also support the hypothesis that cGAS interacts specifically with damaged DNA in the cytoplasm to induce senescence. Of note, they found that cGAS was associated with chromatin during mitosis (see Figure 2). These results suggest that cGAS may play a role in regulating the cell cycle.

Because cellular senescence has been linked to aging and aging-related diseases, cGAS-cGAMP signaling components may be important therapeutic targets. Further investigation is needed to better understand the innate immune system’s role in aging and how these pathways can be targeted.


pnas.1705499114fig05 (2)

Figure 2

The complete study mentioned above can be found here:

A comprehensive review of cellular senescence, cGAS,  and STING can be found here:

To learn more about Dr. Zhijian Chen’s exciting work, join us at 4:00pm, Tuesday 02/12/2019 in the Marilyn G. Farquhar Seminar Room.

Jillybeth Burgado is a Neurosciences PhD student rotating in Dr. Robert Rissman’s lab.

Understanding learned associations through single-neuron activity and population dynamics in the parietal cortex


Dr. Christopher Harvey is an Associate Professor of Neurobiology at Harvard Medical School, where his laboratory investigates how sensory information is integrated and used at the cellular and circuit level to inform decision-related actions. Dr. Harvey’s work largely focuses on the posterior parietal cortex (PPC), a region of the brain thought to be at the interface of sensation and action. The laboratory approaches their research on the PPC through a number of techniques, including in vivo whole-cell patch-clamping to probe the function of neurons at a cellular level and two-photon calcium imaging to chronically monitor the activity of neural ensembles in head-fixed mice performing decision-making tasks in a virtual reality setup.

Recent work in Dr. Harvey’s laboratory published by Driscoll et al. in Cell in 2017 elucidates the dynamic patterns of activity in the mouse PPC throughout the process of learning a virtual-navigation task. The authors show that PPC population activity patterns reorganize through re-exposure of the mouse on the virtual maze in such a way that the relative location of an individual neuron’s peak activity shifts across days. The shifts in representation of the maze encoded by individual PPC neurons can explain the flexibility of this population of cells to encode newly learned behaviors and integrate new information into the existing circuitry.  Importantly, the authors demonstrate that while the activity of individual neurons is highly dynamic throughout this process, the population-level representation of the task features—namely the mouse’s location within the maze and the trial type—in the PPC remains fairly stable. These results support the significant role that PPC ensemble activity plays in maintaining the stable representation of task features.

Adapted from Driscoll et al., 2017


You can read the full paper here.

To hear more about Dr. Harvey’s work, including a new technique employed in his laboratory combining both population activity monitoring and single-neuron perturbations, please join us this Tuesday, February 5, 2019 at 4pm in the Marilyn G. Farquhar Seminar room in the Center for Neural Circuits and Behavior. The title of his talk is “Probing computations in neural circuits using single-neuron perturbations”.

To learn more about the Harvey Lab, visit:


Christopher Lee is a first-year PhD student in the UCSD Neurosciences Graduate Program rotating in the laboratory of Dr. Takaki Komiyama.