SFN 2017 Abstracts

This year’s SFN abstracts, for your anticipatory pleasure!


Alyssa Ash, Jack Clemans-Gibbon, Timothy O’Leary, Erin Chahley, Desiree Seib, Jason Snyder

One feature that sets the dentate gyrus (DG) of the hippocampus apart from other regions of the brain is its ability to produce new neurons throughout adulthood. This results in a heterogeneous population of cells of various ages and diverse properties. For example, immature neurons have a net inhibitory effect on downstream populations, and there is evidence that new and old neurons are differentially modulated by feedback circuitry in the DG in vitro. Recent reports indicate that feedback inhibition plays a powerful role in selecting which DG neuron are recruited during memory formation. This raises the question of whether developmentally-born and adult-born neurons have distinct roles in feedback inhibition, particularly in vivo when neuronal ensembles are selected during memory encoding. To address this we combined chemogenetics and immunohistochemistry for BrdU+Fos to silence, and measure activity in, developmentally and adult-born neurons as rats learned a spatial water maze task. Specifically, retrovirus was injected into the DG of male rats at postnatal day 1 or 6 weeks of age to express the inhibitory DREADD receptor, HM4Di, in neurons born in early development or adulthood. The same rats were also injected with BrdU to label developmentally or adult-born neurons. At 10 weeks of age rats were injected with either the HM4Di agonist CNO or vehicle and then trained in the water maze (8 trials). One hour after water maze training brains were collected and processed immunohistochemically for BrdU, GAD67, GFP and c-Fos to identify neurons that were recruited during learning. We expect that silencing adult-born neurons will reduce activity (c-Fos) in inhibitory interneurons and, in turn, increase overall levels of neuronal recruitment in the DG, particularly in developmentally-born neurons. Since adult-born neurons receive less GABAergic inhibition, we expect their recruitment to be less impacted when populations of DG neurons are silenced. By manipulating and measuring activity in discrete populations of DG neurons, our findings will provide novel insights into the mechanisms by which subcircuits within the DG contribute to memory formation.


Timothy P. O’Leary, Bonnie Lee, Delane Espinueva, Jason S. Snyder

Adult neurogenesis in the hippocampus is involved in visuo-spatial learning and memory, and also regulation of the hypothamic-pituitary-adrenal axis and stress-related behaviour. Although, neurogenesis has been shown to buffer the effects of chronic stress on expression of depressive-like behaviours, it is not yet known if adult neurogenesis acts in a similar way to buffer negative effects of chronic stress on learning and memory function. Sex differences exist in the effects of chronic stress on learning and memory, and thus the role of adult neurogenesis during chronic stress likely differs between male and females. To address this possibility, we used a transgenic rat that expresses the herpes simplex virus thymidine kinase, under the glial fibrillary acidic protein promoter (GFAP-TK). Administration of the anti-viral drug valganciclovir leads to death of mitotic neural precursor cells in the GFAP-TK rat. Beginning at 6 weeks of age, GFAP-TK rats were given valganciclovir orally for 6 weeks (twice per week), which leads to a near-complete ablation of adult-neurogenesis in the dentate gyrus. At 13 weeks of age, rats then completed 21 days of chronic restraint stress, with 6 hours of daily restraint stress. Following restraint stress, visuo-spatial learning and memory was assessed with a modified Morris water maze procedure, in which stress during testing was increased using cold 16 °C water. Rats completed acquisition training for 3 days (4 trials per day) with a probe trial (60 sec) on the following day to assess memory for the escape platform location. Typical effects of chronic stress were observed, with reduced body-weight gain in restrained rats, and also increased adrenal weight in restrained male, but not female rats. Restraint stress did not impair learning and memory ability in either GFAP-TK or wild-type male rats. In female rats, however, restraint stress improved learning performance in wild-type rats. In female GFAP-TK rats, learning performance of restrained and control rats was similar, but in the probe trial restrained GFAP-TK rats performed worse than controls. These results indicate that the chronic restraint stress paradigm used was not sufficient to impair performance in wild-type male and female rats. In GFAP-TK rats, however, restraint stress impaired memory performance in female rats, suggesting that neurogenesis may buffer the deleterious effects of chronic stress on memory ability specifically in females. These findings may indicate a sexually dimorphic role for adult neurogenesis in the etiology of cognitive dysfunction within depression and stress-related disorders.


Desiree Seib, Delane Espinueva, Oren Princz-Lebel, Erin Chahley, Ru Qi Yu, Stan B Floresco, Jason Snyder

There is growing evidence that reduced adult neurogenesis plays an important role in depression. However, clear proof from animal models is still missing. We combine a novel rat model in which we can specifically block adult neurogenesis (GFAP-TK) and complex behavioral operant testing paradigms to elucidate if characteristics of depressive behaviors can be mimicked by loss of hippocampal neurogenesis, for example assigning less value to a delayed future reward. We first tested rats on a delay discounting paradigm, where animals must choose between a low reward lever that delivers 1 sugar pellet immediately and a high reward lever that delivers 4 pellets after a delay. Compared to intact wild types (WT), GFAP-TK rats showed a decreased preference for the high reward option with increasing delay times, indicating that adult neurogenesis increases the subjective value of future rewards. We were able to replicate this finding by ablating adult hippocampal neurogenesis with irradiation in WT rats. On the contrary, increasing neurogenesis by voluntary running for four weeks led to increased preference for delayed high rewards. Second, we employed a probabilistic reversal learning task where rats have to show optimal behavior in the face of an uncertain reward. Here, we test for behavioral flexibility and sensitivity towards negative feedback, behaviors that are known to be impaired in depression. Third, we performed an effort-based decision-making task, since depression is often associated with reduced energy and motivation. Here, the number of lever presses required to receive the high reward option increases across blocks. We did not find any difference in the performance of WT and GFAP-TK animals in these two other tasks, indicating that functions for neurogenesis in reward-based decision-making are specific for immediate vs. delayed rewards. Ongoing experiments are investigating cellular mechanisms by which new neurons influence valuation of future rewards. Preliminary results indicate that decreased neuronal activity in the ventral dentate gyrus in the GFAP-TK rat is associated with a lower preference for delayed rewards. This project combines a novel rat model with complex behavioral tasks to study the impact of reduced neurogenesis on various phenotypes observed in psychiatric disorders with a focus on depression. It suggests that increased neurogenesis, for example by exercise, could benefit depressed patients. Results from this project can also be transferred to other psychiatric disorders, where neurogenesis is affected and similar behavioral phenotypes can be observed, such as addiction, Alzheimer’s disease, and schizophrenia.


Shaina Cahill, Angela Martinovic, John-Darby Cole, Jason Snyder

Adult-born neurons contribute significantly to the synaptic circuitry in the dentate gyrus (DG) and play an important role in the cognitive and emotional functions of the hippocampus. An outstanding question is whether postnatal neurogenesis leads to continued growth in the DG or whether the addition of adult-born neurons is balanced by death of developmentally-born neurons. We have recently found that developmentally-born neurons die in early adulthood, unlike adult-born cells, which are known to remain stable (after reaching maturity). While adult-born neurons have enhanced plasticity and unique circuitry during their immature stages, many reports indicate that they may become functionally equivalent to neurons born in early postnatal development once they have reached maturity. These data collectively suggest that adult neurogenesis may serve to directly replace lost developmentally-born cells. Whether there is a causal, homeostatic relationship between numbers of developmentally-born and adult-born neurons remains unknown. We aimed to test this hypothesis by directly manipulating the adult-born neuronal population and examining how this impacts the number of surviving developmentally-born neurons in the DG. Male rats were injected with the mitotic marker BrdU at postnatal day 6, the peak of DG neuron birth, followed by either neurogenesis-promoting or suppressing treatments administered from 2 until 6 months of age. To suppress adult neurogenesis we used transgenic GFAP-TK rats; at the end of treatment adult neurogenesis was reduced by 69%, as measured by doublecortin immunohistochemistry (IHC). To enhance adult neurogenesis, we subjected rats to alternating 4-week blocks of running and memantine, an NMDA antagonist, which we have found produces sustained increases in adult neurogenesis by 28% as measured by doublecortin IHC. According to a homeostatic relationship, we predict that supressing adult neurogenesis will increase survival of developmentally-born cells; conversely, we predict that increasing adult neurogenesis will reduce survival of developmentally-born cells. Finally, independent of changes in cell number, prolonged reductions or enhancements of adult neurogenesis may alter recruitment of DG neurons during learning experiences. To this end, all rats were exposed to a novel environment prior to the end of the experiment, to measure experience-driven expression of immediate-early genes in developmentally-born DG neurons. Collectively, these data will clarify whether there are interactions between neurons born at different stages of development, which may shape how information is retained in the hippocampus.


John-Darby Cole, Shaina Cahill, Ru Qi Yu, Jack Clemans-Gibbon, Jason Snyder

Immature adult-born neurons are physiologically distinct from older neurons and contribute to the mnemonic and emotional functions of the hippocampus. Methods for increasing neurogenesis therefore have the potential to improve improve mental health in a number of conditions such as Alzheimer’s disease, depression and schizophrenia, all of which are associated with hippocampal structural deficits. Most studies have only examined single methods for short-term elevations of neurogenesis, which may be insufficient to offset major structural changes. The present study therefore examined whether two well known neurogenic treatments, running and the NMDA receptor antagonist memantine, are capable of producing sustained increases in adult neurogenesis. To identify whether there are sex differences in regulation of neurogenesis, both male and female rats were examined. We found that, on their own, both treatments increased adult neurogenesis but levels returned to baseline one month later. To optimize treatments for prolonged elevation of neurogenesis rats were subjected to 8 weeks of running, 8 weeks of memantine, or two alternating 4-week blocks of each treatment and compared to cage controls. PCNA and doublecortin (DCX) were used to quantify proliferating cells and immature neurons, respectively, that were present at the end of treatment. In males, single treatments failed to increase numbers of proliferating cells and immature neurons. However, memantine followed by running increased proliferating cells, and running followed by memantine increased the number of immature neurons. In females, there was a trend for increased numbers of proliferating cells and immature neurons after 8 weeks of running, possibly because they ran significantly more than males. The thymidine analogs CldU and IdU were used to label neurons born at the beginning of each 4-week treatment block. Preliminary analysis of CldU+ neurons born during the first treatment block suggest that in males, but not females, cells born during initial memantine treatments survive to a greater extent when followed up with running. Analyses of IdU+ cells, born during the second treatment block, are currently underway. We are also quantifying putative neural stem cells to determine why some treatment paradigms lead to sustained increases in neurogenesis but others do not. Collectively, our data suggest that there may be limitations in the extent to which single treatments can enhance adult neurogenesis. Instead, a combination of approaches may be more effective and, moreover, may vary between males and females.


Jason Snyder, Jerry Yang, Desiree Seib

Adult neurogenesis is well-characterized in the hippocampus and subventricular zone-olfactory bulb. Despite numerous reports, the idea of widespread adult neurogenesis remains controversial. The striatum is one region where neurogenesis has been identified, but the magnitude and temporal characteristics of neuron addition has not been studied in detail. Here, we characterized the phenotype and survival timecourse of adult-born neurons in the nucleus accumbens, a part of the ventral striatum involved in reward and reinforcement-related behaviors. Young adult male and female rats (8 weeks old) were treated with the BrdU to label newborn cells and then brains were collected 1, 2, 4, 8 or 16 weeks later. BrdU+ cells were quantified in both the core and shell regions of the nucleus accumbens, and were examined immunohistochemically for calbindin and calretinin expression, two markers of GABAergic interneuron populations. The total number of BrdU+ cells was greatest at 1 week post-injection and steadily declined thereafter. No BrdU+ cells showed co-labelling for calbindin but approximately 2-3% expressed calretinin at the 4 and 8 week time points. Few BrdU+calretinin+ neurons remained at 16 weeks, possibly because neurons have died, matured into a different GABAergic neuron phenotype, or migrated elsewhere. In follow up experiments we are examining whether accumbens neurogenesis can be increased by exercise, which robustly increases adult neurogenesis in the hippocampus, and whether newborn accumbens neurons derive from a GFAP+ precursor (likely in the subventricular zone), using GFAP-TK rats. Collectively, our findings suggest that adult neurogenesis is more widespread than is generally appreciated. Specifically, our data identify adult neurogenesis as a novel form of plasticity in the nucleus accumbens that may contribute to reward and reinforcement-related behaviors, many of which are disrupted in psychiatric disorders.

Leave a Reply