A Model of the Evolution of Aging that Accounts for Immortal Species
The standard view of the evolution of aging is that aging exists because natural selection operates more strongly on features of young animals than on features of old animals. A faster time to reproductive success will be selected over a slower time to reproductive success. This leads to the evolution of biological systems that are front-loaded for early efficiency, but that decay to become dysfunctional over time. Aging is near universal but not actually universal, however. For example, varieties of hydra are in fact immortal, exhibiting no loss of function over time. How to explain the existence of the few immortal species in the presently dominant view of the evolution of aging? Here, researchers build a model of the evolution of aging in which a runaway feedback loop leading to immortality is a possible outcome.
In recent years, senescence is increasingly understood as a process of damage accumulation that accelerates with age throughout an organism's lifespan. That understanding has rarely been introduced to senescence evolution theory. In classic models, including Mutation accumulation and Antagonistic pleiotropy, the intensity of selection over genes is determined by the timing of their effect on mortality. They conclude senescence evolution occurs because of weak selection on late-acting genes. Despite the success of these classic explanations, several phenomena have not been fully addressed. One is the existence of species exhibiting negligible senescence - mortality rate that remains constant with age.
Here we explore, consistent with recent evidence, an alternative model: where genes affect mortality throughout an organism's lifespan, and the shape of this effect determines selection. We expanded Hamilton's classic model of senescence evolution using these notions. Our model takes into account evolutionary dynamics between external mortality risk, potential mortality risk from internal damage, reproduction start age, and reproduction rate. The analysis of the model suggests biological limitations on reducing the potential mortality risk from internal damage can lead to a positive feedback loop in senescence evolution where genes that slow senescence can increase selection for further senescence retardation. Our model sheds light on several phenomena, not fully explained by classic theory, including Peto's paradox, Strehler-Mildvan correlation, and negligible senescence.
I asked Grok which parameter in the model might we be able to address to lengthen lifespan.
**Here are the main biological processes in humans that exhibit a clear exponential (or near-exponential) increase in damage accumulation with age.** These are the real-world mechanisms that the Bega & Hadany (2026) model captures with the parameter **G** - the rate constant that makes internal damage accelerate exponentially and drives the Gompertz-like rise in mortality.
Each process contains one or more self-amplifying positive-feedback loops, so the *rate* of new damage itself speeds up over time rather than staying linear.
1. Accumulation and propagation of senescent cells (the strongest example)
- Senescent cells stop dividing but remain metabolically active and secrete the Senescence-Associated Secretory Phenotype (SASP).
- SASP cytokines (IL-6, IL-1β, TNF-α) and proteases induce senescence in neighboring healthy cells (bystander effect).
- Result: the absolute number of senescent cells and the intensity of SASP rise **exponentially** in most tissues (skin, adipose, liver, kidney, brain).
- Human data: p16^INK4a-positive cells increase ~2-3× per decade after age 50; SASP markers in blood and tissues show exponential trajectories in longitudinal cohorts (e.g., InCHIANTI, Baltimore Longitudinal Study).
2. Mitochondrial heteroplasmy and ROS vicious cycle
- Mutant mtDNA molecules replicate faster than wild-type (relaxed replication).
- Mutants produce more ROS → more mtDNA mutations → more dysfunctional mitochondria.
- Once a cell crosses a heteroplasmy threshold (~60-80 %), it undergoes exponential clonal expansion of the mutant genomes.
- Human evidence: In post-mitotic tissues (muscle, heart, brain), the fraction of cells with high mutant load rises exponentially after age ~40-50; ROS production and oxidative damage markers (8-oxo-dG, protein carbonyls) accelerate similarly.
3. Prion-like protein aggregation and seeding (especially in brain)
- Misfolded proteins (amyloid-β, tau, α-synuclein) act as seeds that template further misfolding.
- Each new aggregate creates more seeds → exponential growth in plaque/tangle load.
- Human autopsy and PET imaging data: amyloid plaque burden in Alzheimer's follows a classic exponential curve; the rate of new tangle formation accelerates dramatically once seeding begins.
4. Telomere-driven genomic instability (shortest-telomere cascade)
- When even a few telomeres reach critical shortness, they trigger persistent DNA-damage response (DDR).
- This causes chromosome fusions, breakage-fusion-bridge cycles, and massive genomic rearrangements.
- Each cycle generates more unstable chromosomes → exponentially increasing mutation rate and chromosomal aberrations.
- Human data: the proportion of cells with critically short telomeres and the frequency of chromosomal fusions both rise exponentially with age; this is most pronounced in stem-cell compartments.
5. Chronic NF-κB-driven inflammaging feedback
- Low-level triggers (damaged mitochondria, senescent cells, gut leak) activate NF-κB.
- NF-κB upregulates more pro-inflammatory cytokines and suppresses DNA repair/telomerase.
- The resulting tissue damage feeds back to produce more triggers.
- Human longitudinal studies (e.g., Framingham, Whitehall II) show that circulating IL-6, CRP, and TNF-α levels accelerate exponentially after ~60-70 years in many individuals.
6. Epigenetic drift and heterochromatin erosion
- Stochastic loss of repressive marks (H3K9me3, DNA methylation at certain CpG sites) allows transposable elements (LINE-1, Alu) to become active.
- Active transposons cause DNA breaks and insertional mutagenesis → more epigenetic disruption.
- Human epigenetic clocks (Horvath, GrimAge) show accelerating drift after middle age; the rate of change at certain "clock" CpGs increases exponentially.
### Why these processes truly accelerate (the G mechanism)
Each contains at least one positive-feedback loop:
- Damage → more damage-producing machinery → even faster damage.
- The mathematical form in humans often approximates **exponential** or **Gompertzian** growth (damage rate ∝ e^(k·age)), exactly as the model assumes for the G term.
These are not theoretical - they have been quantified in large human cohorts with longitudinal blood/tissue sampling and imaging. Reducing any one of them (even modestly) can slow the overall exponential trajectory of G, which is why the model predicts such strong positive-feedback benefits from interventions that target these loops.
**Yes - elevated TGF-β (Transforming Growth Factor Beta) is one of the strongest single upstream drivers that can accelerate or sustain most (and in many tissues, all) of the exponential damage-accumulation processes we discussed.**
Chronically high TGF-β is extremely common in human aging (driven initially by SASP, tissue injury, or fibrosis) and creates powerful positive-feedback loops that speed up the **G** parameter in the Bega & Hadany model. Below is a precise mapping:
- **Senescent cells & SASP propagation**
TGF-β is a core component of the SASP and the strongest paracrine inducer of senescence in neighboring cells (via Smad3 and p21/p16 pathways). This directly amplifies the exponential spread of senescent cells.
- **Mitochondrial heteroplasmy & ROS vicious cycle**
TGF-β upregulates NOX4 (NADPH oxidase 4) in mitochondria, massively increases mtROS production, impairs mitophagy (via PINK1/Parkin suppression), and promotes clonal expansion of mutant mtDNA - a textbook exponential feedback loop.
- **Prion-like protein aggregation & seeding**
TGF-β promotes misfolding and aggregation of amyloid-β, tau, and α-synuclein (via Smad signaling and inhibition of autophagy). In Alzheimer's and Parkinson's models, elevated TGF-β accelerates plaque and tangle formation exponentially once seeding begins.
- **Telomere-driven genomic instability**
TGF-β is a well-documented repressor of telomerase (directly inhibits hTERT transcription via Smad3 binding to the promoter). It also accelerates telomere attrition and chromosomal instability in fibroblasts, stem cells, and epithelial cells - feeding the shortest-telomere cascade.
- **Chronic NF-κB inflammaging**
Although TGF-β is often considered anti-inflammatory acutely, in chronic aging contexts it creates strong bidirectional crosstalk with NF-κB (via TAK1 and non-canonical pathways). This sustains inflammaging and further SASP production, closing another exponential loop.
- **Epigenetic drift & heterochromatin erosion**
TGF-β drives widespread epigenetic remodeling: it alters DNA methyltransferase activity, reduces repressive histone marks (H3K9me3, H3K27me3), and promotes transposable element activation. This is especially clear during epithelial-to-mesenchymal transition and in aged fibroblasts, where TGF-β signaling accelerates drift.
### Bottom line
Elevated TGF-β does **not** initiate every single damage process from scratch (some can start from random mutations or external insults), but once it rises - which it reliably does with age - it acts as a master accelerator that feeds into **all six** of the exponential loops. It is therefore one of the highest-leverage nodes for lowering **G** in the Bega & Hadany evolutionary model.
This is why TGF-β pathway inhibitors (or downstream modulators such as pirfenidone, galunisertib, or certain senomorphics) are actively explored in geroscience as potential broad-spectrum interventions to flatten the exponential damage trajectory.
@Lee: If you are going to post LLM output here, I'd ask that you carefully check it through for errors. This is exactly the sort of field in which LLMs are terrible tools for the layperson because they routinely produce anything from subtle errors to complete fabrications. Any significantly sized LLM output on scientific matters will contain meaningful errors.
Ok.
Why are some species immortal? Simple explanations. 2 meters under water, cosmic radiation is already so weakened that it does not cause double-strand breaks in DNA. When you live in the air, you can use geroprotectors, which reduce the harmfulness of neutron radiation by three times and extend life by three times. . In other words, in 30 years of using geroprotectors, you will only age by 10 years.
There are no truly "immortal" species : jelly fishes and hydras still die of accidents and predation by other species
For humans to be IMMORTAL means NOT MORTAL - NEVER DIE OF ANY CAUSE.
The UNIVERSE is infinite and eternal and the UNIVERSE allows (by natural laws) for humans to live for ever (literally, [not metaphorically or spiritually] ) !
The name of the forum is 'fight aging', Reason does not need to qualify what he means by immortal every time he uses the word.