Increased REST to Optimize Life Span?

Gene Expression Profiles Associated with Human Longevity

Unsurprisingly, clear differences in gene expression develop in the brains of aging versus young humans. ¹ To investigate these changes, Zullo et al. ² used transcriptome analysis (RNA-seq/microarray) of cells obtained from the frontal cortex of several cohorts of healthy aged humans (i.e., ROSMAP, CommonMind Consortium). ³ Extended longevity of cognitively intact individuals was associated with downregulation of genes related to neural excitation and synaptic function, and upregulation of genes involved in immune function. ²

A Model System Instructs Mechanisms

Studies of the model organism Caenorhabditis elegans dating back two decades support a critical role of the nervous system in regulation of aging. ^{4 –6} For example, specific neurosecretory or sensory neurons modulate life span, ^4,5,7 with direct effects on the insulin/IGF pathway central to the beneficial effects of calorie restriction on longevity. ^7,8

Seeking mechanisms of neuromodulation of longevity, Zullo et al. ² first monitored neural excitation in C. elegans using GCaMP calcium imaging in glutaminergic ASH neurons. ⁹ C. elegans contains at least 11 bilaterally symmetric pairs of sensory neurons in the head and two in the tail that function primarily as chemosensors. ASH, one of these bilateral pairs in the head, plays a critical role in mediating avoidance behaviors in response to chemical repellents as well as several different noxious stimuli. ASH is required for behavioral avoidance of water-soluble (e.g., copper, quinine, and SDS) and volatile (octanol) chemical repellents, osmotic shock, and mechanical stimulation on the tip of the animal's nose.

Calcium influx in ASH neurons was shown to increase during normal aging (i.e., days 1–2 vs. days 12–16). Administration of nemadipine (an inhibitor of calcium flux via L-type calcium channels) or ivermectin (an agonist of invertebrate glutamate-gated channels) suppressed neural excitation and extended the lives of treated worms. Approximately 50% of worms survive ∼20 days with an additional 5-day survival with treatments (∼20%). Importantly, administration of these drugs AFTER cessation of feeding activity (beyond day 8) mediated the same health extension effects independently of calorie restriction, suggesting that global inhibition of excitatory neural activity mediates an independent prolongevity effect in C. elegans.

Several clever experiments supported these results. Zullo et al. ² generated transgenic worms expressing the Drosophila histamine-gated chloride channel (HisCl1) in different populations of neurons. ¹⁰ Exogenous histamine (which is not produced or utilized by worms) activates this channel to mediate inhibition of neural excitation. Exogenous histamine (10 mM) extended the life span of worms expressing the HisCl1 under the control of a pan-neuronal promoter (all neurons) as well as specific promoters for glutaminergic and cholinergic neurons known to be the major C. elegans excitatory neurons. Histamine had no effects on the life span of wild-type worms. Conversely, RNAi inhibition ¹¹ of GABAergic neurotransmission increased excitation in ASH neurons with a concomitant reduction in life span. Thus, a reciprocal relationship between neuronal excitation and life span was demonstrated: more excitation shorter life span, less excitation longer life span. In most of these experiments, the median survival was significantly increased by ∼25%–30%.

Digging Deeper into the Mechanism: Repressor Element 1-Silencing Transcription Factor

The repressor element 1-silencing transcription factor (REST) increases with normal aging in human cortical and hippocampal neurons. ¹² Levels of REST correlate with preservation of cognition during aging as well as individual longevity. REST protects neurons from oxidative stress and amyloid β-protein toxicity, while conditional deletion of REST in the mouse brain leads to age-related neurodegeneration. In Alzheimer's disease, frontotemporal dementia, and dementia with Lewy bodies, REST is lost from the nucleus and appears in autophagosomes together with pathological misfolded proteins. Chromatin immunoprecipitation with deep sequencing and expression analysis shows that REST represses genes that promote cell death and the pathology of Alzheimer's disease, and induces the expression of stress response genes. Thus, REST is a candidate mediator of neuroprotection versus neurodegeneration in the aging brain.

Based on the earlier work from the Yankner laboratory at Harvard, ¹² it was logical for Zullo et al. ² to question whether REST might modulate neuronal excitation in the aging brain. Not only did they confirm that REST mRNA was upregulated in the brain during aging but elevated levels of REST were found in prefrontal cortical neurons in centenarians as well, compared with individuals 20–30 years younger. Perhaps REST mediates downregulation of neuronal excitation?

To directly investigate this question, a functional ortholog of REST, C. elegans spr-4, was studied. Spr-4 protects against oxidative stress and amyloid β-protein toxicity in worms. Augmented expression of spr-4, using a modified CRISPR-cas9 method, ¹³ generated worms with increased mean life spans and reduced ASH neuronal excitation. What about mammals?

To address this question, CNS uptake of fluorodeoxyglucose (FDG) by positron emission tomography and computerized tomography was used to evaluate neuronal activity in the brains of REST conditional knockout (REST-cKO) mice versus littermate controls. The REST cKO mice (age 18 months) demonstrated an elevated ¹⁸ F-FDG cortical uptake characteristic of elevated neuronal activity. In summary, these initial data supported the hypothesis that REST plays a central role in the downregulation of neuronal excitation, a correlate of successful aging of the nervous system. How does this finding play into a critical hallmark of aging the energy metabolism pathway orchestrated by DAF-2-insulin/IGF signaling activated by calorie restriction?

Connection to DAF-2-Insulin/IGF-Like Signaling Pathway

The DAF-2-insulin/IGF-like signaling pathway regulates the life span of C. elegans via downstream forkhead transcription factor DAF-16. ^14,15 Nemadipine and ivermectin both elevated total and nuclear levels of DAF-16. RNAi-mediated knockdown of daf-16 prevents the extension of life span by spr-4 described above. These results support the hypothesis that DAF-16 mediates health span extension via spr-4 (REST analog), which mediated neural inhibition.

Inhibition of Daf-2 by RNAi augments worm life span by about 50%, a significant effect size. ¹⁶ Mutations in spr-4 (or a related gene, spr-3) abolished the effect of knocking down Daf-2, implicating that the REST homologue life span extension was dependent on the insulin/IGF-like pathway. Additional experiments using longer lived Daf-2 mutant worms showed that RNAi knockdown of Spr-3/4 in neurons was sufficient to reduce the life span. RNAi studies (vs. eat-4 [glutaminergic], cha-1/unc-17 [cholinergic], and cat-2 [monoaminergic] neurons as well as certain neuropeptides) supported the contention that SPR-3 and SPR-4 suppress excitatory neurotransmitter systems to extend the life span of the daf-2 mutant worms.

Reduced daf-2 extends life span of worms via activation of DAF-16. ¹⁴ Zullo et al. ² showed that the loss-of-function spr-3/4 worms treated with daf-2 RNAi exhibited reduced DAF-16 levels, suggesting a role for the activation of ortholog forkhead transcription factor DAF-16 in the protective effects of SPR-3/4 to suppress neural excitation.

To extend this work to mammals, Zullo et al. ² showed that REST and FOXO1 (DAF-16 ortholog), but not other FOXO family members, were positively correlated throughout aging of the human prefrontal cortex. Direct regulation of FOXO1 by REST was shown in the Rest cKO mice. Age-dependent induction of nuclear FOXO1 was abolished in the cKO mice versus normal littermate controls. Other pharmacological studies demonstrated that FOXO1 is regulated (like DAF-16) by glutaminergic signaling in mammalian cortical neurons.

These studies may also explain the benefits of anticonvulsants (e.g., ethosuximide and valproic acid) to extend worm life spans. ^{17 –19} Valproic acid is a histone deacetylase (HDAC) inhibitor that is needed for REST-mediated transcriptional inhibition. Active HDAC promotes transcriptional repression and chromatin condensation.

In summary, persistent activity of transcriptional repressor REST (and the C. elegans orthologs SPR-3/4) to attenuate neuronal activity correlates with health span and longevity at least in worms. The mechanism of this activity involves the forkhead transcription factors (FOXO1 in mammals and DAF-16 in worms) and provides a link between reduced excitatory neural activity and beneficial metabolism engendered by calorie restriction leading to prolongevity. It remains an open question whether the neuronal activity/spr-4/DAF-16 pro-life span/health span effect holds for mammals through analogous neuronal activity/REST/FOXO1.

Medical Implications

REST (aka neuron-restrictive silencer factor, NRSF) acts as a gene silencing transcription factor playing a major role in neuron development from embryogenesis to eventual differentiation. ^{20 –22} REST acts via epigenetic remodeling to regulate ∼2000 coding and noncoding neuron-specific genes involved in synaptic plasticity and structural remodeling, including synaptic vesicle proteins, channels, receptors, transporters, and neuron-specific microRNAs. ^{23 –26} Loss of REST is essential for acquisition of the differentiated neuronal phenotype. ²⁷

Thus, it comes as no surprise that dysregulation of REST and REST-dependent epigenetic remodeling contributes to the pathogenesis of various neurodegenerative diseases such as Alzheimer's disease as well as stroke, epilepsy, and Huntington's disease. However, regulation of REST and processes regulated by REST are quite complex and even counterintuitive. For example, ischemic insult ^{28 –32} and seizures ^33,34 activate REST in susceptible neurons, where it may then contribute to pathology, whereas in the aging brain, loss of REST is associated with the development of Alzheimer's disease. ¹²

Levels of neural activity and REST connect to insulin-IGF signaling to regulate the activity of forkhead transcription factors known to be central to the regulation of life span. ^14,35 Metabolism may be integrated with neuronal activity via daf-16 and REST orthologs spr-3/4 in worms and FOXO1 and REST in humans. Artificial activation of REST with subsequent curtailment of excitatory neural activity or perhaps balancing of neuronal activity may promote health span in humans.

The levels of the REST transcription factor are highest in the brains of long-lived people with intact cognition. Levels stayed high specifically in the brain regions vulnerable to Alzheimer's, suggesting that they might be protected from dementia. ^12,26 It is assumed that REST represses genes that promote cell death and Alzheimer's disease pathology, and induces the expression of stress response genes. Moreover, REST potently protects neurons from oxidative stress and amyloid β-protein toxicity. ²¹

On the contrary, REST is also responsible for ischemia-induced neuronal cell death in mouse models of brain ischemia. Ischemia, resulting from reduced blood perfusion of tissues, decreases nutrient and oxygen supply, which induces REST transcription and nuclear accumulation, leading to the epigenetic repression of neuronal genes leading to cell death. ³⁰ The mechanism beyond REST induction in ischemia might be tightly linked to its oxygen-dependent nuclear translocation and repression of target genes in hypoxia (low oxygen) where REST fulfils the functions of a master regulator of gene repression in hypoxia. ³⁶

It's as if increased REST may be protective up to a point and then problematic while reduced REST is problematic. In both cases, the end result is neuronal death. The idea is that increasing REST reduces excitotoxic neuronal activity, but too much REST is damaging to neuronal function. Alternatively, too little REST increases excitotoxic neuronal activity.

The excitation/inhibition (E-I) balance develops and is maintained by critical homeostatic mechanisms in the healthy brain. E-I balance controls excitability, dynamic range, and input information gating in many brain circuits, while E/I imbalances have been implicated in nearly every major neurological and psychiatric disorder, including autism spectrum disorders, schizophrenia, and epilepsy. ^{37 –39} An E-I imbalance can result from dysregulation of the formation or maturation of either excitatory glutaminergic or inhibitory GABAergic synapses. ⁴⁰ Recent work by Kang et al. ⁴¹ provides insight into the mechanism by which a susceptibility gene for major mental illness (e.g., DISC1, Disrupted-in-schizophrenia 1) can facilitate the development of an asymmetry in excitatory and inhibitory synapse formation to disrupt neural circuit homeostasis.

With regard to aging, morphological studies of prefrontal layer I (where inhibitory GABAergic neurons predominate) ⁴² showed a 30%–60% reduction in the density of synapses in older compared with younger monkeys. This loss of inhibition correlated with cognitive impairment. ⁴³

E-I controls neuronal excitation and, in turn, neuronal homeostasis participates in the maintenance of organismal homeostasis and aging.

Conclusion

The importance of neural activity level in aging needs to be more fully investigated. Results suggesting that lower levels of nervous system activation associated with increased levels of neural master inhibitor REST result in greater life span need to be confirmed in other animals. That increased REST activity is associated with neurological pathology in epilepsy and Huntington's disease, and the subject of development of inhibitory drugs should raise caution in developing approaches that seek to increase REST activity to promote health- or life span.