Hypothesis

The epigenetic age of the gut microbiome, quantified by a microbiome‑specific clock, predicts circulating phenylacetic acid (PAA) levels and induces endothelial senescence in cerebral vessels even when host age is held constant. Reversing microbial epigenetic age through targeted phage depletion of PAA‑producing strains will lower PAA, protect brain endothelium, and improve cognitive performance in aged animals.

Mechanistic Basis

Age‑associated shifts enrich phenylalanine‑catabolizing bacteria that raise PAA, which directly triggers endothelial senescence (1). Microbiota also exhibit an epigenetic drift analogous to host aging, measurable by microbiome‑specific CpG methylation patterns (2). We propose that this microbial epigenetic age, rather than host age, determines the functional output of phenylalanine pathways and thus PAA production.

Predictions

1. Older mice with a youthful microbiome epigenetic profile (low microbiome‑age score) will exhibit lower plasma PAA, reduced cerebral endothelial senescence markers (p16^INK4a^, SA‑β‑gal), and better performance in hippocampal‑dependent tasks than age‑matched controls with an aged microbiome.
2. Transplanting fecal microbiota from epigenetically aged donors into young germ‑free recipients will accelerate microbiome epigenetic aging, elevate PAa, induce endothelial senescence, and impair cognition despite the recipients’ young host age.
3. Phage‑mediated depletion of the PAA‑producing clade (e.g.

Hypothesis

Hypothesis: It's predicted that sustained expression of telomerase together with piRNA‑Piwi pathway components in adult somatic cells creates a germline‑like surveillance system that selectively removes genomically damaged cells via apoptosis, thereby delaying age‑related tissue decline.

Germline cells avoid mutational load by maintaining high telomerase activity Telomerase activity, silencing transposons through the piRNA‑Piwi axis piRNA‑Piwi pathway, and eliminating damaged progeny by rapid apoptosis piRNA‑Piwi pathway. Somatic tissues, in contrast, downregulate these mechanisms after early life, allowing telomere shortening, transposon reactivation Transposon reactivation with age, and the accumulation of senescent cells that secrete pro‑inflammatory factors. If the germline’s “cheating” strategy is a set of inducible modules rather than a fixed germline‑only program, then re‑activating them in somatic cells should impose a similar selection pressure.

Experimental Design

We propose to drive telomerase (TERT) and a core piRNA biogenesis factor (e.g., MIWI2) in a tissue‑specific, inducible manner in mice aged 12 months. Concurrently, we will transiently sensitize the intrinsic apoptotic pathway using a low‑dose BH3 mimetic to mimic the germline’s low threshold for culling defective cells.

Readouts

• Telomere length measured by fluorescence in situ hybr

Hypothesis

Allotopic expression of nuclear-encoded mitochondrial genes shifts the NAD+/NADH balance in senescent cells, activating SIRT3‑dependent deacetylation of superoxide dismutase 2 (SOD2) and thereby lowering mitochondrial ROS below the threshold required to sustain the senescence‑associated secretory phenotype (SASP). This metabolic re‑programming primes senescent cells for rapid clearance by senolytic agents.

Mechanistic Rationale

• Somatic mtDNA mutations accumulate with age and impair OXPHOS, raising ROS that drives SASP via p53/p21 and ATM/ATR pathways[2].
• Low‑level heteroplasmy at OriL already shortens lifespan and disrupts glucose metabolism[4].
• Allotopic expression of essential ETC subunits (e.g., ND4) restores electron flow, increases ATP production, and raises the NAD+/NADH ratio[5].
• Elevated NAD+ activates the mitochondrial deacetylase SIRT3, which deacetylates and activates SOD2, converting superoxide to hydrogen peroxide and reducing oxidative stress.
• Senescent cells maintain a bioenergetic sweet spot: enough ROS to sustain SASP but not so much as to trigger outright mitochondrial collapse; complete mtDNA depletion (ρ0) suppresses senescence[6].
• By tuning ROS to a lower, non‑signaling level, allotopic expression removes the metabolic cue that keeps the SASP transcriptional program active, making the cells dependent on a

Hypothesis

Oral NAD+ precursors (NMN or NR) will raise SIRT1 deacetylase activity in human peripheral blood mononuclear cells (PBMCs) only when administered alongside a fasting‑mimicking diet (FMD) that induces the NAD+ salvage enzyme NAMPT.

Mechanistic Rationale

• NAD+ precursors increase circulating NAD+ pools, but SIRT1 activation requires locally generated NAD+ within nuclear or mitochondrial microdomains where the enzyme resides.
• NAMPT catalyzes the rate‑limiting step of the NAD+ salvage pathway, producing NAD+ from nicotinamide; its activity is acutely upregulated by AMP‑activated protein kinase (AMPK) during energy stress such as fasting.
• In mice, AMPK‑driven NAMPT expression creates a permissive NAD+ milieu that allows SIRT1 to engage targets like PGC‑1α; without this boost, excess NAD+ remains largely cytosolic and fails to stimulate SIRT1.
• Human data show that NMN/NR reliably elevate blood NAD+ (1, 6) but no study has measured SIRT1 activity in accessible tissues after precursor supplementation alone.
• FMDs (5‑day low‑calorie cycles) activate AMPK and increase NAMPT transcription in human immune cells (3), providing a testable condition to couple precursor availability with enzyme‑driven NAD+ production.

Testable Predictions

1. Baseline: PBMC NAD+ levels rise ~2‑fold after 2 weeks of NMN or NR monotherapy; SIRT1 activity (measured by PGC‑1α dea

We hypothesize that age‑related loss of vagal efferent output directly drives vocal fold extracellular matrix (ECM) dysfunction by disabling the cholinergic anti‑inflammatory pathway, thereby permitting unchecked NF‑κB signaling in local fibroblasts and immune cells. This leads to excess matrix metalloproteinase activity, reduced tissue inhibitor of metalloproteinases, abnormal collagen cross‑linking, elastin fragmentation and hyaluronan depolymerization, which together raise vocal fold stiffness and impair vibratory function. Because the vocal fold receives dense vagal innervation that controls muscle tone, vascular perfusion and immune modulation, its ECM state serves as a read‑out of peripheral autonomic health. Consequently, measurable changes in vocal fold viscoelasticity (e.g., via shear‑wave elastography) should parallel systemic inflammaging markers such as circulating IL‑6, TNF‑α and nitro‑tyrosine, and predict mortality risk independent of traditional frailty indices. Mechanistically, vagal motor neurons release acetylcholine onto α7 nicotinic acetylcholine receptors on macrophages and fibroblasts, suppressing NF‑κB translocation and limiting pro‑inflammatory cytokine transcription. With age, myelinated vagal axon area declines (r = –0.74)【2】, decreasing ACh release and weakening this brake. Simultaneously, age‑related Schwann cell atrophy and delayed Wallerian degeneration impair axonal regeneration【6】, compounding the loss of efferent tone. The resulting cytokine surge activates fibroblast‑derived TGF‑β1/Smad signaling, driving collagen I/III overproduction and

Hypothesis

Age‑associated decline of TET enzyme activity reduces 5hmC at promoters of proteostasis genes, triggering a shift from soluble chaperone‑mediated refolding to the formation of structured, amyloid‑like aggregates that sequester misfolded proteins and limit proteotoxic stress.

Mechanistic Rationale

TET‑mediated hydroxymethylation maintains an open chromatin state at loci encoding HSP70, HSP90 and components of the ubiquitin‑proteasome system [1]. In aging T‑cells, global 5hmC erosion diminishes transcription of these genes, weakening dynamic protein quality control. When refolding capacity falls below a threshold, the cell redirects excess misfolded proteins into ordered, β‑sheet‑rich deposits. These aggregates are not random garbage; they arise because the remaining proteostasis network favors thermodynamic minimization via amyloid formation, converting toxic oligomers into inert, sequestered matrices. Restoring 5hmC (e.g., via TET overexpression or Vitamin C) should re‑activate chaperone expression, dissolve the aggregates, and restore proteasome flux.

Testable Predictions

1. Aged human CD4+ T‑cells will show a significant inverse correlation between 5hmC levels at HSP70/HSP90 promoters and the abundance of Thioflavin‑T‑positive aggregates.
2. Pharmacological inhibition of TET activity in young T‑cells will recapitulate the aged epigenetic profile and increase ordered aggregate formation without raising total ubiquitinated protein load.
3. Genetic rescue of TET2 or treatment with a 5hmC‑enhancing agent will reduce aggregate burden, increase soluble HSP

Hypothesis

We hypothesize that pharmacologically boosting NAD+ at the circadian trough (subjective night) synergizes with tissue‑specific BMAL1 overexpression to restore circadian amplitude, thereby re‑gating mitophagy and suppressing the NFκB‑driven inflammaging loop more effectively than constant NAD+ supplementation.

Mechanistic rationale

1. Circadian gating of mitophagy – BMAL1‑CLOCK drives rhythmic expression of PINK1 and Parkin; peak transcription occurs during the early active phase. When amplitude collapses, mitophagy becomes arrhythmic, allowing damaged mitochondria to accumulate and fuel ROS‑NFκB signaling.
2. NAD+‑SIRT1 axis – SIRT1 deacetylates BMAL1, enhancing its DNA‑binding affinity; NAD+ troughs reduce SIRT1 activity, weakening the clock and increasing acetylated p65 NFκB activity. Timed NAD+ repletion at the nadir should rescue SIRT1‑BMAL1 interaction without causing chronic SIRT1 overactivation that can blunt stress responses.
3. Feed‑forward inflammaging – Elevated NFκB drives CD38 expression, further consuming NAD+ and deepening the trough. Breaking this loop requires simultaneous NAD+ replenishment and BMAL1 reinforcement to restore proper temporal gating of both antioxidant and anti‑inflammatory programs.

Testable predictions

• Prediction 1: In aged mice, timed NAD+ (e.g., NR) administered at ZT14 (subjective night) + AAV‑mediated BMAL1 overexpression in liver will increase circadian amplitude of Bmal1 mRNA (measured by bioluminescence) >2‑fold versus constant NAD+ or BMAL1 alone.
• Prediction 2: The combined inte

Active autophagy suppression in aged cells preserves a damaged mitochondrial pool that fuels senescence‑associated secretory phenotype (SASP) through sustained mtROS production. We hypothesize that age‑related lipid remodeling within lysosomes generates a specific phospholipid species that directly binds the ULK1 kinase, blocking its activation irrespective of nutrient status. This lipid‑ULK1 interaction doesn't just block ULK1; it adds a layer to the known mTORC1‑dependent brake (phosphorylation of ULK1‑Ser758) and it's known that autophagy remains resistant to rapamycin or young plasma in certain tissues.

Testable predictions: (1) Lysosomal extracts from old mouse liver will contain elevated levels of bis(monoacylglycero)phosphate (BMP) or lysobisphosphatidic acid (LBPA) that co‑immunoprecipitate with ULK1; (2) Chemical depletion of BMP/LBPA using siRNA against the lipid‑synthesizing enzyme PLD3 will restore ULK1 phosphorylation at activating sites and increase LC3‑II flux even when mTORC1 is inhibited; (3) Expressing a ULK1 mutant incapable of binding the lipid will rescue autophagy flux and reduce mtROS, SASP cytokines (IL‑6, IL‑1β) and senescence markers (p16, SA‑β‑gal) in aged fibroblasts; (4) Conversely, augmenting BMP/LBPA in young cells will suppress autophagy and induce a senescence‑like state despite nutrient starvation. Moreover, we'll predict that enhancing lysosomal acidity via V‑ATPase overexpression will reduce BMP/LBPA accumulation, thereby relieving ULK1 inhibition and restoring autophagic clearance of damaged mitochondria. This links the lipid hypoth

Hypothesis: Stress granule metastability determines the switch from protective aggregation to pathogenic seeding

The proteostasis network responds to chronic misfolding stress by sequestering risky proteins into stress granules (SGs) that act as metastable sinks, steering clients away from irreversible fibrils[2]. When SG material properties shift toward increased stability—due to mutations, post‑translational modifications, or age‑related decline in chaperone activity—the same condensates begin to nucleate rather than inhibit amyloid formation. This predicts that manipulating SG material state will toggle aggregation outcomes without altering overall protein load.

Testable prediction: In a cellular model expressing an aggregation‑prone reporter (e.g., 'PolyQ‑GFP'), pharmacological or genetic stiffening of SGs will increase the rate of fibrillar seed formation measured by fluorescence recovery after photobleaching (FRAP) and Thioflavin‑T uptake, whereas SG fluidization will reduce seed burden despite unchanged total aggregate mass.

Experimental design: 1) Use CRISPR to introduce point mutations in SG nucleators (G3BP1) known to alter material properties[2]. 2) Treat parallel cultures with low doses of 1,6‑hexanediol or with overexpress HSP70 to promote fluidity. 3) Quantify SG dynamics via FRAP, seed competence via seeding assay in naïve cells, and cell viability over time. 4) Correlate SG half‑time of recovery with seeding efficiency.

Falsifiable outcome: If SG stiffening does not change seeding propensity, or fluidizing SGs fails to lower seed

Hypothesis

Intermittent, low‑dose GH pulses paired with liver‑specific knockdown of IGF‑1 production will preserve muscle and bone anabolism while maintaining circulating IGF‑1 in the range associated with improved insulin sensitivity and longevity.

Rationale

• Somatopause drives sarcopenia and metabolic dysfunction, yet centenarians exhibit low IGF‑1 bioactivity alongside preserved insulin sensitivity (2).
• GH replacement improves lean mass and bone density but provokes insulin resistance, hyperglycemia, and oncogenic risk (3).
• Hepatic IGF‑1 accounts for ~75 % of circulating IGF‑1; muscle‑derived IGF‑1 acts locally and is less tied to systemic metabolic effects (1).
• An inverse U‑shaped relationship links IGF‑1 levels to metabolic syndrome, with both excess and deficiency worsening risk (5).
• Sex‑specific signaling shows IGF‑1 dominates female aging while insulin signaling is more influential in males (6).

By suppressing hepatic IGF‑1 we aim to shift the systemic IGF‑1 curve leftward into the favorable low‑normal zone, while pulsatile GH spikes provide transient activation of muscle and bone IGF‑1 receptors, delivering anabolic signals without sustaining high systemic IGF‑1 that drives insulin resistance and tumorigenesis.

Predictions

1. Circulating IGF‑1 will r

Hypothesis

In aged tissues, autophagy is not merely declining; it is actively restrained by a self‑reinforcing circuit in which DNMT2‑driven promoter methylation, Rubicon accumulation, and mTORC1 hyperactivity each reinforce the others, creating a stable epigenetic‑signaling lock that preserves damaged macromolecules as a short‑term survival strategy.

Mechanistic Model

1. DNMT2‑mediated methylation silences Atg5 and LC3 promoters in macrophages and fibroblasts, reducing basal autophagic flux【https://pmc.ncbi.nlm.nih.gov/articles/PMC4889231/】.
2. Loss of autophagy leads to accumulation of damaged mitochondria and lysosomal stress, which activates the Rubicon protein. Rubicon binds the VPS34 complex, inhibiting autophagosome maturation【https://pmc.ncbi.nlm.nih.gov/articles/PMC11352966/】.
3. Rubicon‑mediated blockade further elevates cytosolic amino acids, sustaining mTORC1 activity even under nutrient deprivation【https://doi.org/10.1083/jcb.201610113】.
4. Persistent mTORC1 phosphorylates and inhibits ULK1, while also promoting DNMT2 expression via S6K‑dependent transcriptional feedback, closing the loop.

This triad creates a bistable switch: once triggered by age‑related stress, the system settles into a low‑autophagy, high‑damage state that resists reversal unless multiple nodes are perturbed simultaneously.

Novel Insight

The circuit explains why single‑intervention approaches (e.g., mTOR inhibition alone) often yield only transient autophagy rescue in aged models: suppressing one node leaves the others capable of re‑establishing the lock. Effecti

Hypothesis

Lysosomal cholesterol accumulation locks mTORC1 on the lysosomal surface, rendering it insensitive to starvation and actively suppressing autophagy in aged cells.

Mechanistic reasoning

1. In young cells, amino acid starvation triggers lysosomal cholesterol efflux via NPC1, reducing mTORC1 recruitment and allowing ULK1 dephosphorylation.
2. With age, lysosomal cholesterol builds up due to declined NPC1 activity and increased ACAT1-mediated esterification, creating a rigid lipid environment that sustains Rag GTPase‑mTORC1 interaction even when cytosolic amino acids are low.
3. This cholesterol‑rich platform explains the persistent mTORC1‑Ser758‑ULK1 phosphorylation observed in aged fibroblasts [2] and renders autophagy refractory to rapamycin unless cholesterol is removed.
4. Transient expression of OSK (OCT4, SOX2, KLF4) during reprogramming upregulates LXRα and NPC1 transcription, promoting cholesterol efflux from lysosomes, thereby resetting mTORC1 nutrient sensitivity and enabling the autophagy burst required for pluripotency [4].
5. Consequently, autophagy suppression in aging is not merely a passive decline but an active lipid‑mediated safeguard that preserves a damaged proteome by preventing catastrophic self‑digestion.

Testable predictions

• Prediction 1: Aged human fibroblasts will show a 2‑3‑fold increase in lysosomal free cholesterol (measured by filipin staining) compared with young cells, correlating with persistent mTORC1 signaling under EBSS starvation.
• Prediction 2: Acute depletion of lysosomal cholesterol using methyl‑β‑c

Hypothesis

Senescent cells operate as a bistable 'hostage negotiator' circuit that can toggle between a pro‑repair state and a pro‑fibrotic state depending on autocrine TGF‑β signaling and miRNA‑mediated feedback. In early damage, a transient surge of IL‑6/IL‑8 SASP recruits immune cells and promotes remodeling, while a concurrent low‑level TGF‑β signal keeps the switch poised toward resolution. Persistent stress drives a positive feedback loop where TGF‑β activates SMAD3, which upregulates its own transcription and suppresses miR‑29, leading to a stable high‑TGF‑β, high‑MMP/PDGF SASP that locks neighboring fibroblasts into a senescent, matrix‑producing phenotype. This second state represents the 'negotiation gone awry' where the senescent cell demands excessive matrix deposition, tipping the tissue toward fibrosis. Non‑selective senolytics remove both states, thereby collapsing the bistable switch and either unleashing unchecked progenitor proliferation (if the pro‑repair state is lost) or allowing residual fibrotic senescent cells to dominate (if the pro‑fibrotic state persists). The hypothesis predicts that the transition between states is governed by a measurable threshold of nuclear pSMAD3 and miR‑29 levels, and that pharmacologically biasing the switch toward the low‑TGF‑β state will preserve reparative senescence while preventing fibrosis.

Testable Predictions

1. Single‑cell RNA‑seq of injured skin at 24 h, 72 h, and 7 d will reveal two distinct senescent clusters: one enriched for IL6, IL8, CXCL1 (early SASP) and low pSMAD3/miR‑29; the other enriched for TGFB

Hypothesis

Aged adipose tissue macrophages release extracellular vesicles (EVs) enriched in miR‑21 and TGF‑β1 that reprogram neighboring adipocytes and stromal cells to increase LOX activity and suppress MMP14/TIMP balance, thereby accelerating collagen VI cross‑linking and fibrosis. This EV‑mediated paracrine loop explains the depot‑specific acceleration of fibrosis in visceral fat with age and provides a testable node distinct from cell‑autonomous mechanotransduction.

Mechanistic Model

• Hypoxia in hypertrophic adipocytes (already known) triggers HIF‑1α‑driven EV biogenesis, loading EVs with miR‑21 and latent TGF‑β1 [1]
• EV uptake by stromal pre‑adipocytes and macrophages activates SMAD2/3 signaling, upregulating LOX transcription and secreting active LOX that cross‑links nascent collagen VI fibrils [2]
• Simultaneously, EV‑delivered miR‑21 suppresses MMP14 mRNA translation while inducing TIMP1 expression, tipping the proteolytic balance toward net matrix accumulation [3]
• The increased cross‑linked collagen VI raises pericellular stiffness, which feeds back to adipocytes via integrin‑β1‑FAK‑actin contractility, further boosting HIF‑1α and EV release—a feed‑forward paracrine circuit that is independent of the adipocyte‑autonomous actin‑mediated COL6A3 upregulation described earlier [4]

Predictions & Experimental Approach

1. EV blockade – Trea

Hypothesis

Maintaining NAD+ at youthful concentrations during cyclic OSK expression creates a metabolic checkpoint that biases epigenetic remodeling toward repair rather than oncogenic transformation. NAD+ fuels sirtuin deacetylases (SIRT1, SIRT6) that tighten chromatin, limit c‑Myc–driven transcriptional noise, and promote DNA‑damage response. When NAD+ falls, chromatin becomes hyper‑acetylated and permissive, allowing reprogramming factors to induce aberrant enhancer activation and pre‑cancerous epigenetic drift. Therefore, pharmacologically sustaining NAD+ will lower the Repair Capacity Index (RCI) decline and suppress tumor formation seen with OSK alone.

Mechanistic Rationale

1. NAD+‑Sirtuin Axis Controls Chromatin State – SIRT1 deacetylates H3K9ac and H3K56ac, fostering heterochromatin; SIRT6 deacetylates H3K9ac at telomeres and promoters of growth‑associated genes. Low NAD+ reduces sirtuin activity, increasing histone acetylation and chromatin openness, which is known to facilitate oncogenic transcription during reprogramming [3, 4].
2. c‑Myc Activity Is NAD+‑Sensitive – c‑Myc transcription is amplified when SIRT1 is inactive due to increased acetylation of c‑Myc and its co‑activators. NAD+ restoration re‑engages SIRT1‑mediated c‑Myc deacetylation, attenuating its oncogenic potency [5].
3. **Epigenetic Clocks Detect Early

Hypothesis: Intermittent fasting reprograms the brain’s complement‑mediated pruning machinery by simultaneously lowering neuronal C1q expression through SIRT1 activation and boosting astrocytic secretion of the complement inhibitor CD59. This dual action raises the threshold for synapse tagging, preserving weakly active but metabolically adaptable connections while still allowing elimination of irreversibly damaged ones.

Mechanistic

rationale: Fasting elevates β‑hydroxybutyrate, which functions as an endogenous HDAC inhibitor and increases NAD+ levels. The resulting SIRT1 activation deacetylates NF‑κB p65, reducing its transcriptional drive on the C1q gene in neurons and astrocytes [1][2]. Concurrently, fasting stimulates TFEB‑mediated lysosomal biogenesis in astrocytes, enhancing their capacity to load and secrete CD59‑containing exosomes [3]. Elevated extracellular CD59 binds to C8 and C9, blocking membrane attack complex formation and thereby shielding synapses from microglial phagocytosis even when C1q is present [4].

Testable predictions: 1. In young adult mice undergoing alternate‑day fasting (ADF), neuronal C1q mRNA in the hippocampus will decrease by ~30% relative to ad libitum controls, an effect abolished by SIRT1 inhibition (EX‑527) [5]. 2. Astrocyte‑specific CD59 knockout will prevent fasting‑induced rescue of spine density despite normal ketone levels, as measured by two‑photon imaging of dendritic spines after 4 weeks of ADF. 3. CSF exosomes isolated from fasted humans will show higher CD59:C1q ratios and lower synaptosomal C1q deposition in vit

Hypothesis

Chronic mTOR inhibition extends lifespan by mimicking nutrient scarcity, but it simultaneously blocks the mTORC1‑dependent metabolic‑epigenetic rewiring required for trained immunity. We hypothesize that pulsatile (intermittent) mTOR inhibition—alternating short periods of drug exposure with drug‑free windows—will retain the longevity‑promoting stress response (autophagy, AMPK/FOXO activation) while permitting periodic mTORC1 activity sufficient to drive the glycolytic shift and H3K4me3/H3K27ac deposition needed for monocyte/macrophage trained immunity. Consequently, animals receiving intermittent rapamycin will show lifespan extension comparable to continuous treatment and retain or enhance β‑glucan‑induced innate immune memory, whereas continuous rapamycin will impair this memory.

Mechanistic Rationale

• mTORC1 activity is necessary for HIF‑1α stabilization and the glycolytic burst that fuels acetyl‑CoA production for histone acetylation at immune gene promoters (e.g., TNFα, IL6) during trained immunity [3, 4].
• Autophagy and FOXO-mediated stress resistance are activated rapidly upon mTORC1 inhibition and can be sustained with brief inhibitory pulses [1, 2], suggesting that the longevity program does not require continuous suppression.
• Epigenetic marks underlying trained immunity (H3K4me3, H3K27ac