What is Adult Hippocampal Neurogenesis?
In the 1990s Elizabeth Gould discovered that adult mammalian brains could engender new neurons, confined primarily to the dentate gyrus of the hippocampus – a brain region known to be a substrate for learning and memory. (New neurons have also been reported in the neocortex (ref, ref, ref) and hypothalamus (ref) although the extent of neurogenesis in these areas remains controversial (ref).) Histological marker studies have demonstrated that the daily production of several thousand (e.g. 5000-10,000) new dentate hippocampal neurons, a process termed neurogenesis, occurs in a variety of adult mammals including mice, rats, tree-shrews, marmoset and macaque monkeys, and (as studies in the late 1990s showed) humans (ref). The total number of dentate hippocampal neurons is estimated to be between 1.5–2.0 million in the adult rat (ref). Turnover rate of mature dentate gyrus cells in humans has been estimated at ∼1.75% a year (ref).
The subgranular zone of the dentate gyrus (SGZ) in the adult brain contains neural stem cells capable of producing thousands of new granule cells per day. At the onset, stem cells give birth to dentate gyrus progenitor cells in a process called proliferation. During their first week of life, these new-born progenitor cells migrate from the edge of the dentate gyrus (subgranular zone) to a deeper (granule cell) layer where they mature, developing signal detecting dendrites and action potential conducting axons in a process of differentiation. Some progenitor cells will survive and integrate into functional neural circuits as dentate gyrus granule cells while within just a few weeks some – perhaps most – will die through programmed cell death (ref, ref). For instance in lab control rats, the majority of new neurons degenerate within two weeks of their production (ref).
Heightened Neuroplasticity of Newborn Neurons
While the number of newly produced dentate hippocampal neurons is small compared to total number (estimated to be between 1.5–2.0 million in the adult rat), the new neurons could have a disproportionate functional effect due to distinct structural and functional characteristics which enhance neuroplasticity such as the ability to form synapses quickly or extend axons rapidly (ref, ref). Adult dentate gyrus, immature neurons may differ electrophysiologically from mature neurons, with larger action potentials, or by having greater effects on their targets via other mechanisms enhancing LTP (synapse strengthening) (ref, ref). Adult hippocampal neurogenesis can be modulated by artificial induction of network activity. -e.g. induction of LTP in the dentate gyrus by high-frequency stimulation can increase the proliferation of progenitor cells in the subgranular zone and enhance the survival of adult-born dentate gyrus cells at 1–2 weeks of age (ref, ref).
How is Adult Neurogenesis Measured?
The most common method for labeling dividing cells involves incorporating a traceable molecule (e.g. radioactive nucleoside, 3H-thymidine) into the cells’ DNA. Because DNA synthesis is generally limited to mitosis this has been used as a marker of hippocampal neurogenesis (ref).
Another method used frequently is immunohistology staining – imaging specific antigen-antibody interactions. Various immunohistological markers are expressed during specific stages of adult neurogenesis – proliferation, differentiation, migration, targeting, and synaptic integration. The availability of such markers allows the time-course and fate of newly born cells to be followed within the DG in a detailed and precise fashion (ref).
Turn-over rate of human granule cells has been estimated with a technique using atmospheric radioactive carbon detection in DNA (ref).
Extrinsic Modulators of Hippocampal Neurogenesis
Neurons are not created at a constant rate, and their survival rate is not constant. Numerous studies have shown that the production, differentiation and survival of neurons is a highly plastic process, depending on experience.
Note that because the process of proliferation, differentiation and survival of hippocampal cells occurs over several (e.g. 4-6) weeks (ref), these factors might have to act in a chronic / continuous, rather than an acute / single shot, way in order to have significant functional effects on learning and memory.
For some of these regulators, there is a critical window in which the intervention can save new-born neurons: 7-10 days after cells arise. 7-10 days roughly corresponds to the period when the newborn cells differentiate into functioning neurons with signal detecting dendrites, and electrically conducting axons.
Some researchers dissociate between factors that modulate (increase or decrease) the proliferation of precursor hippocampal neurons and those that modulate the survival (and functional integration as granule neurons) of those precursor neurons (ref). Some of these factors are shown in Gould and colleagues’ Table below.
A useful representation of some extrinsic factors modulating hippocampal neurogenesis is found in Aimone and colleagues’ 2014 review of adult neurogenesis (ref), shown below:
Chronic estrogen treatment in female rats increases the pool of progenitor cells capable of continuous proliferation and neuronal differentiation (ref). Chronic estrogen treatment has been associated with improved performance on spatial navigation tasks (ref), while low levels of estrogen have been linked to cognitive decline (ref).
Oxytocin but not vasopressin, stimulates both cell proliferation and adult neurogenesis in the hippocampus of rats (ref).
Aerobic exercise (e.g. running)
Running can result in a dramatic (e.g. 50%) increase in the rate of dentate gyrus cell proliferation in just 2 weeks and four-fold over two months – but has less direct impact on cell survival (ref, ref). (Running may also have an influence on synaptic plasticity, specifically enhancing LTP in the dentate gyrus of the hippocampus (ref).) This influx of circulating growth factors due to aerobic exercise may mask the negative effects of stress-induced glucocorticoids (ref).
Diet – like exercise – can have a profound impact on many aspects of brain function, including mood, energy metabolism, behavior, cognitive function, and hippocampal neurogenesis.
Flavanols (found in berries, parsley, green tea, and cocoa) have been shown to increase cell proliferation in chronically stressed rats (ref), and this enhancement might be mediated by IGF-I signaling and (indirectly) BDNF (ref) Flavanol rich diets increase hippocampal vascularization, and neuronal spine density – but not new cell survival (ref). Cerebral blood flow, neuronal spine density is considered important for learning and memory (ref). Flavanol-rich green tea (ref), blueberries (ref), pomegranates (ref), and strawberries enhance memory and synaptic plasticity (ref), and attenuate age-related cognitive decline in both working memory and motor skill in rats (ref). A 10-yr study assessed the cognitive function of humans subjects aged 65 or older; those with the greatest amount of flavonoid intake performed better at both the beginning and end of the study compared with those with the lowest intake (ref).
Synergy. When epicatechin is combined with running in rodents, the cognitive improvement is greater, and this combination is associated with an increase in spine density (ref). This suggests that a synergy exists between diet and exercise for neurogenesis.
Appetite / hunger
Sexual experience not only buffers against the early elevated glucocorticoids (see below) but but promotes neuronal growth and reduces anxiety (ref).
Cannabinoids are able to strongly promote embryonic and adult hippocampal proliferation via the CB1 receptors in the progenitor cells – and CB1 receptor–knockout mice display profound suppression of hippocampal neurogenesis.This may underlie the mechanism of anxiolytic- and antidepressant-like effects produced by a high dose of chronic cannabinoid treatment. The anxiolytic/antidepressant effects are dose-dependent (ref).
Increasing glucocorticoid levels (which are adrenal steroids derived from cholesterol) lowers the number of proliferating cells while removal of adrenal steroids stimulates cell proliferation (ref, ref). In humans the main glucocorticoid hormone is cortisol.
Nitric Oxide (NO)
In the normal brain, NO, synthesized in nitrergic neurons, is a negative regulator of precursor cell proliferation. (Although after brain damage, NO overproduction in different neural and non-neural cell types promotes neurogenesis.) (ref)
Stress responses to aversive or threatening environments (e.g. physical or social stressors) have a substantial negative effect on DG cell proliferation (ref, ref). Stress is associated with elevated adrenal hormones (consistent with 1 above).
There is a decline in neurogenesis that occurs throughout the lifespan of an animal related to a decreasing proliferation of granule cell precursors and not attributable to a general aged-related metabolic impairment (ref). In one study, numbers of newborn neurons decreased roughly eight- to ninefold from middle-aged to aged rats (ref). Note that when older rats are given running exercise over several weeks, there is a remarkable increase in cell survival and neuronal differentiation, greatly reducing the aging effect (ref).
Lack of Sleep
Whilst disruption of sleep for a period shorter than 1 day has little effect on the basal rate of cell proliferation, multiple studies clearly show that prolonged sleep loss can inhibit hippocampal proliferation and survival – independently of adrenal stress hormones (ref, ref).
Drugs of Abuse
Major drugs of abuse including opiates, alcohol, nicotine, and cocaine suppress hippocampal neurogenesis in adult rats (ref, ref), suggesting a potential role of hippocampal neurogenesis in the treatment of drug addiction (ref).
Learning may suppressing the survival of earlier-born neurons (ref).
Learning can rescue more than double the newly born hippocampal granule cells in rats from death compared to controls (ref). Collectively, there is evidence (ref, ref, ref, ref, ref) that the following types of learning that promote cell survival in rodents:
- spatial-navigation learning
- long-term spatial memory retention
- spatial pattern discrimination
- trace conditioning (associative learning)
- contextual fear conditioning (association learning)
- physical skill learning
Elizabeth Gould and Tracey Shors and colleagues’ work indicates that the following conditions are need to be met for optimal cell survival:
- The learning experience should be between one and two weeks after mitosis, a time when adult-generated granule cells might be forming connections with the CA3 region ‘downstream’, and integrating with functional hippocampal circuits (ref).
- The learning task must be cognitively challenging. Easier tasks do not result in cell survival (ref. ref).
- The learning process must be successful – that granule cell survival does not occur with only the training in the absence of learning the task There is a strong positive correlations between how well an animal learns the skill and the number of surviving cells in the animal’s dentate gyrus (ref, ref).
- The task-learning must involve sustained effortful / concentration, and that more effort results in more cell survival (ref). neurogenesis was increased by allowing animals to exercise in activity wheels for approximately two months (Clark et al., 2008). The animals learned a new task faster (the Morris water maze). However, when neurogenesis was disrupted the exercise no longer facilitated learning.
(Environments with opportunities for learning, social interaction, exploration, and physical activity.)
The effect of enriched environments on neurogenesis is pronounced. Some studies have found an approx. two-fold increase in survival of hippocampal granule neurons in mice living in ‘enriched environment’ (EE) conditions compared to laboratory caged controls (ref, ref). In another study, progenitor cells were more than fivefold more likely to differentiate into a neuron in mice that lived in an EE long-term compared with mice that were housed in a standard cage (ref). In multiple instances, these enrichment-induced increases in neurogenesis levels were correlated with improved performance on hippocampus-dependent behavior such as in the Morris water maze (MWM).
Combining enrichment with aerobic exercise can result in a synergistic effect on neurogenesis.In rodents, a combination of exercise and environmental enrichment results in a greater increase in neurogenesis than either exercise or enrichment alone (ref, ref).
Environmentally induced neurogenesis requires the neurotrophin BDNF (ref).
Caloric Restriction (CR) & Intermittent Fasting (IF)
Similar to the effects of enriched environments (ref), caloric restriction does not increase the proliferation of neural stem cells, but does increase survival of their neuronal progeny (ref). For instance, a reduction in calorie intake of 30–40% increases adult hippocampal neurons in rodents (ref). Without reducing calorie intake, extending the time between meals (IF) increases adult hippocampal neurogenesis and hippocampal gene expression and correlates with performance in hippocampus-dependent tasks and mood (ref).
The effect of CR is mediated in part by BDNF production, which may also enhance the synaptic plasticity of the mature hippocampal neurons (ref). Intermittent fasting may also enhance learning and memory (ref).
Mounting evidence suggests that sleep may contribute to hippocampal functions by promoting neurogenesis (ref). Promoting adult hippocampal neurogenesis may be a mechanism by which sleep supports learning and memory processes. Sleep deprivation disturbs memory formation and negatively modulates cell dentate gyrus granule cell survival (ref). Even mild sleep restriction may interfere with the increase in neurogenesis that normally occurs with hippocampus-dependent learning (ref).
Periods of nighttime sleep after learning enhance the consolidation of declarative and procedural memory – that is, there is sleep-dependent performance enhancement. This consolidation depends on synaptic long-term plasticity in a hippocampal-neocortical network (ref) .The effect of sleep on motor learning has been demonstrated to be particularly high within the first 4–6 hrs after initial training, a period of heightened susceptibility to disruptive interference by other learning (ref).
Brain-Derived Neurotrophic Factor (BDNF)
The neurotrophin BDNF (that plays a causal role in both enriched environment and fasting experiments) promotes the differentiation and maturation of adult progenitor cells in culture. The expression of BDNF is regulated by neural activity and plasticity. Increasing BDNF mRNA and protein is associated with memory acquisition and consolidation (ref, ref). Interestingly, the signaling of BDNF promotes the survival and maturation of GABAergic inhibitory neurons, potentially linking neural activity and GABA-mediated effects on neurogenesis (see below) (ref).
PROLIFERATION & SURVIVAL
Mental and Physical (MAP) Training
In answering the question: ‘Can we train our brains to learn better?’ Curlik and Shors propose that we can through a combination of aerobic exercise and sustained, rigorous mental training on tasks that show training-based ‘transfer’ to performance on related tasks or even unrelated non-practiced tasks (such as the dual n-back). Their combination may have an additive effect on neurogenesis: “a combination of these two forms of training may result in greater cognitive gains than either form alone.” (Curlik and Shors).
Neurotransmitters in Local Circuits
Each of the three stages of neurogenesis (proliferation, differentiation, and survival) is a target of regulation by network neurotransmitter factors – as shown in the diagram below (ref).
This mood and emotion-regulating neurotransmitter induces proliferation. Serotonin-selective reuptake inhibitor (SSRI) antidepressants (e.g. fluoxetine) directly induce increased levels of proliferation Subsequent studies have shown a direct effect of serotonin on proliferation (ref, ref).
GABA is the primary inhibitory neurotransmitter in the brain, appears to suppress precursor cell proliferation while inducing differentiation and survival. There is evidence for an inverse correlation of proliferation with overall network activity (and tonic GABA levels) (ref). Newborn granule cells are tonically activated by ambient GABA, before they are sequentially innervated by GABAergic and glutamatergic synaptic inputs. GABA has an essential role in the synaptic integration of newly generated neurons in the adult brain (ref).
This is the primary excitatory neurotransmitter and is necessary for proper survival of newborn neurons. GABA remains depolarizing in young neurons until roughly 2 weeks of age corresponding to the time in which glutamatergic spines develop on the developing dendrites. Glutamatergic inputs (through NMDA receptors) become increasingly important as immature neurons develop (ref). NMDA activation is critical in the survival of young neurons passing through the 2- to 3-wk-old “critical period” (ref).
An excitatory neurotransmitter and neuromodulator – this is necessary for proper maturation and survival. Depletions of ACh (through lesions of basal forebrain cholinergic regions) lead to a reduction in neurogenesis (ref). Acetylcholine receptors are needed for the proper maturation of adult-born hippocampal neurons, with decreased survival of new neurons and reduced synaptic development in those young neurons that survive (ref).
Dopamine (DA) is involved in the regulation of motivation, interest/pleasure, and attention/ concentration (likely to be impaired in depressed patients). Evidence indicates that DA promotes both the proliferation of neural precursor cells and survival of newborn cells in the adult hippocampus via D1-like receptors. and D1-like receptor agonist promoted the survival of newborn cells in the adult hippocampus (ref).
Dopamine (D1 receptor binding) is enhanced with working memory training (ref).
Functional Role of Hippocampal Dentate Gyrus Neurogenesis in Cognition & Behavior
General Level Theories
Neurogenesis is a substrate in the service of adapting to rewarding environments that promote exploration, learning and reproduction. Reduced neurogenesis – in threatening, chronic aversive environments that release glucocorticoids aid aid ‘survival mode’ (associated more with anxious behaviors). Glucocorticoids are catabolic in nature (for energy release) growth inhibition is adaptive where survival may depend on shunting energy for immediate use in stressful ‘fight or flight’ type situations. This inter-play allows for the modification of behavioral responses throughout adulthood, fine-tuning the hippocampus to the predicted environment (e.g. Glasper, Shoenfeld & Gould at Princeton, ref, ref),
“Reduced neurogenesis that occurs with persistent exposure to a high threat environment produces a hippocampus that is more likely to respond with behavior that maximizes the chance of survival. Conversely, enhanced neurogenesis that occurs with continual exposure to a rewarding environment leads to behavior that optimizes the chances of successful reproduction.” (Glasper et al, 2012).
Experiences as diverse as running, learning (under relatively low stress conditions, sexual experience, living in an enriched environments, intracranial self-stimulation and even acute cocaine stimulation, promote neuron proliferation, and can all be classified as ‘rewarding’. Thus a bigger pool of new new neurons is associated with behaviors that maximize chances for reproduction. Conversely, a smaller pool of new neurons is associated with more anxiety-like behavior – e.g. more caution, less exploration – that ensure survival under aversive conditions. Moreover, rewarding experience is associated with improved performance on hippocampus -dependent, low stress cognitive tasks such as navigation and object recognition where there is novelty (ref).
This theory may relate to the hypothesized Behavioral Approach System (BAS) and Behavioral Inhibition System (BIS).
Supports Effortful Learning
The theory proposed by Tracey Shors (ref) is that hippocampal dentate gyrus neurogenesis supports effortful learning of cognitively challenging tasks. New cells are produced ‘just in case’ on a use it or lose it basis.
“If the animals are cognitively challenged, the cells will linger. If not, they will fade away.” (Shors)
The learning-based cell-survival studies (above) support this theory. In other relevant studies: Enriched environments (see below) increased the number of new hippocampal cells in rodents by ~70%. After enrichment animals were able to remember a novel object for a longer period of time. This long-term memory enhancement did not occur if the new cells were not produced (ref). In another study, neurogenesis was increased through wheel running exercise for two months, and the animals learned a new spatial navigation task faster. When neurogenesis was disrupted, the exercise no longer facilitated the learning (ref).
Hypothesis 1: Pattern Separation in Memory Formation
In terms of neural networks, pattern separation is a network process whose outputs are less similar, or less overlapping to one another, than its inputs. It is hypothesized that events, consisting of highly similar features and configurations can be learned to be distinguished in memory (using ‘sparse coding’) if neurogenesis is present in the dentate gyrus – either through having extra neurons to encode the differences in contexts, or reducing interference between baseline activity of mature granule cells (ref, ref). Conditions inducing prolonged (chronic) inhibition of neurogenesis, such as glucocorticoid hormone treatment, chronic stress an+d aging, is associated with diminished performance on hippocampal-dependent tasks such as spatial discrimination (ref). Neurogenesis knockdown studies looking at spatial and context discrimination support this hypothesis (ref). Molecular and physiological studies demonstrated that the dentate gyrus can achieve pattern separation by using distinct population codes and different firing patterns to represent different environmental inputs (ref).
Hypothesis 2: Encoding Temporal Context
New neurons encode time. Memories formed at different times are separated via populations of young neurons at different time-slices (although this separation would not apply for memories for novel events experienced at or around the same time). temporally close but distinct events would activate separate populations of mature granule cells (pattern separation) as well as overlapping populations of immature neurons (temporal association in episodic memory). Distance in time is encoded through the ongoing maturation process (ref). Trace condition tasks provide evidence that require integration of events that occur close in time require neurogenesis (ref) and there is evidence that the DG as a whole is necessary for temporal associations over longer time scales (ref).
Hypothesis 3: Memory Resolution
Mature neurons and young neurons encode different types of information. Mature sparsely coding, strongly inhibiting ‘fine tuned’ neurons are powerful at representing what they have encoded in the past with high specificity and information content. Immature neurons – by contrast – have ‘broad tuning’ with less information content, that can form a powerful distributed code that can uniquely represent featureless specificity in response to novel, unfamiliar feature inputs (ref). Low-information immature neurons can later become high-information mature neurons via their high level of synaptic plasticity, while mature neurons can ‘fix’ their coding on familiar features indefinitely, with little interference over time (ref). Rodent performance on some tasks is unaffected by experimental manipulations of neurogenesis – tasks that have been interpreted as more familiar or naturalistic. This may be taken as evidence for this hypothesis. And one study showed that ablating adult-born neurons that were thought of as previously specializing to the trained event, showed that the loss did impair subsequent performance on those tasks (ref).
Hypothesis 4: Non-Contiguous Associative Encoding
Hippocampus is involved most critically in learning and memory tasks in which discontiguous items must be associated, in terms of their temporal or spatial positioning, or both – i.e. the hippocampus is necessary to overcome temporal or spatial “discontiguity” (ref). Event-related fMRI binding of delay presented words was associated with left hippocampal activity, compared to simultaneously presented words, supporting this hypothesis (ref).The associations that activate the hippocampus the most may be those that are task related (ref).
Hypothesis 5: Temporary Storage & Long Term Memory Consolidation
The hippocampus is a temporary storage site for memory and a conduit for long-term consolidation of contextual memory. Temporary memories are encoded in rapidly changing populations of adult-generated neurons over discrete time periods. These cells might then undergo changes in connectivity, gene expression, or both, in the transition from short term to long-term storage. Neurogenesis may also serve as a mechanism for forgetting in the hippocampus through general retroactive interference, regardless of memory content (ref). Data supports this hypothesis: Hippocampal lesions result in a global impairment in forming new memories and the loss of recent, but not remote,memories for hippocampal-dependent tasks (ref). The recall of contextual fear memory after its formation is initially dependent on the hippocampus but becomes independent of the hippocampus after several weeks (ref). Ablation of neurogenesis by irradiation prolonged the hippocampus dependency of contextual fear memory in mice, suggesting a role for neurogenesis in promoting system memory consolidation (ref).
Hypothesis 6: Emotion/Stress Regulation
The hippocampus is a direct regulator of the hypothalamic-pituitary-adrenocortical (HPA) axis, through regulating HPA sensitivity to corticosteroid feedback (ref). The hippocampus has projections to the hypothalamus, and the hippocampus contains corticosteroid receptors (ref). Suppression of neurogenesis leads to a potentiated hypothalamo-pituitary-adrenal axis response (ref), and stress-induced decreases in neurogenesis are a causal factor in precipitating episodes of depression. Conversely increasing brain levels of serotonin through selective serotonin reuptake inhibitor (SSRI) antidepressants enhances the basal rate of dentate gyrus neurogenesis (ref), thereby promoting recovery from depression. The ability of to restore hippocampal regulation of the HPA axis under chronic stress conditions, has been shown to occur only in the presence of neurogenesis (ref).
Aimone, J. B., Li, Y., Lee, S. W., Clemenson, G. D., Deng, W., & Gage, F. H. (2014). Regulation and Function of Adult Neurogenesis: From Genes to Cognition. Physiological Reviews, 94(4), 991–1026. https://doi.org/10.1152/physrev.00004.2014
Curlik, D. M., & Shors, T. J. (2013). Training your brain: Do mental and physical (MAP) training enhance cognition through the process of neurogenesis in the hippocampus? Neuropharmacology, 64, 506–514. https://doi.org/10.1016/j.neuropharm.2012.07.027
Curlik, Daniel M., & Shors, T. J. (2011). Learning Increases the Survival of Newborn Neurons Provided That Learning Is Difficult to Achieve and Successful. Journal of Cognitive Neuroscience, 23(9), 2159–2170. https://doi.org/10.1162/jocn.2010.21597
Deng, W., Aimone, J. B., & Gage, F. H. (2010). New neurons and new memories: how does adult hippocampal neurogenesis affect learning and memory? Nature Reviews. Neuroscience, 11(5), 339–350. https://doi.org/10.1038/nrn2822
DiFeo, G., & Shors, T. J. (2017). Mental and physical skill training increases neurogenesis via cell survival in the adolescent hippocampus. Brain Research, 1654(Pt B), 95–101. https://doi.org/10.1016/j.brainres.2016.08.015
Glasper, E. R., Schoenfeld, T. J., & Gould, E. (2012). Adult neurogenesis: optimizing hippocampal function to suit the environment. Behavioural Brain Research, 227(2), 380–383. https://doi.org/10.1016/j.bbr.2011.05.013
Gould, null, Tanapat, null, Hastings, null, & Shors, null. (1999). Neurogenesis in adulthood: a possible role in learning. Trends in Cognitive Sciences, 3(5), 186–192.
Leuner, B., & Gould, E. (2010). Structural plasticity and hippocampal function. Annual Review of Psychology, 61, 111–140, C1-3. https://doi.org/10.1146/annurev.psych.093008.100359
Ming, G., & Song, H. (2011). Adult Neurogenesis in the Mammalian Brain: Significant Answers and Significant Questions. Neuron, 70(4), 687–702.
Shors, T. J. (2009). Saving New Brain Cells. Scientific American, 300(3), 40-48.
Shors, T. J., Anderson, M. L., Curlik, D. M., & Nokia, M. S. (2012). Use it or lose it: how neurogenesis keeps the brain fit for learning. Behavioural Brain Research, 227(2), 450–458. https://doi.org/10.1016/j.bbr.2011.04.023