⚙ How is this computed? Methods: Aging Hallmarks
Aging / Hallmarks

The 12 Hallmarks of Aging

An evidence-aware guide to the framework updated by López-Otín and colleagues in Cell (2023). Use the hallmarks to organize observations across molecules, cells, tissues, and organisms—not as interchangeable biomarkers or a clinical score.

2023 Cell framework5 primary hallmarks3 antagonistic hallmarks4 integrative hallmarksMechanism-aware interpretation
Explore the 12 hallmarks โ†’ Read the evidence guide โ†’ Review scientific limits โ†’ Explore LACA data โ†’
12 Hallmarks in the 2023 framework
3 Interconnected conceptual layers
4 Evidence dimensions to record
0 Universal diagnostic cut-offs
Conceptual mechanism map

Twelve hallmarks, three interconnected layers

The 2023 framework organizes relatively upstream damage and maintenance failure, context-dependent stress responses, and system-level decline. The relationships are conceptual and highly interconnectedโ€”not a strict one-way causal sequence.

Layer 01

Primary

Damage and maintenance failure

GIGenomic instability TLTelomere attrition EPEpigenetic alterations PRLoss of proteostasis AUDisabled macroautophagy
Layer 02

Antagonistic

Adaptive responses that can become maladaptive

NSDeregulated nutrient sensing MTMitochondrial dysfunction SNCellular senescence
Layer 03

Integrative

Cell, tissue and organism-level decline

SCStem cell exhaustion ICAltered intercellular communication INChronic inflammation DYDysbiosis
LACA data integration

Explore hallmarks across the SP121 single-cell atlas

Explore twelve versioned transcriptomic proxy programs across 1,246,365 Human cells and compare descriptive patterns by cell type, tissue, and system.

Open full-screen explorer โ†’
Conceptual map

Three layers of an interconnected system

The categories offer a conceptual organization of relatively upstream damage, context-dependent compensatory responses, and system-level decline. They are highly interconnected rather than strictly sequential.

Primary

Damage and maintenance failure

Genomic instability, telomere attrition, epigenetic alterations, loss of proteostasis, and disabled macroautophagy compromise the fidelity and renewal of cellular components.

Interpretation: measure both the lesion and the maintenance pathway. A static abundance measurement rarely establishes impaired repair or flux.
View primary hallmarks โ†’
Antagonistic

Stress responses with context-dependent effects

Deregulated nutrient sensing, mitochondrial dysfunction, and cellular senescence can protect the organism acutely yet contribute to aging when excessive, persistent, or mistimed.

Interpretation: direction matters. More or less pathway activity is not automatically beneficial across every cell type, dose, or life stage.
View antagonistic hallmarks โ†’
Integrative

System-level functional decline

Stem cell exhaustion, altered intercellular communication, chronic inflammation, and dysbiosis connect cell-intrinsic changes to impaired regeneration and organism-wide phenotypes.

Interpretation: tissue composition, environment, infection, medication, and co-morbidity can produce similar readouts and must be modeled explicitly.
View integrative hallmarks โ†’
Update

What changed in 2023

Disabled macroautophagy, chronic inflammation, and dysbiosis were elevated to distinct hallmarks, expanding the original 2013 framework from nine to twelve.

These additions do not imply independence: autophagy, inflammation, microbiota, proteostasis, metabolism, and immune-cell state are strongly coupled.
Layer 1 / Primary

Sources of damage and impaired cellular maintenance

Primary hallmarks are proposed as relatively upstream damage and maintenance processes. For atlas analyses, pair molecular readouts with cell identity, tissue context, and a functional or perturbational assay whenever possible.

Primary 1

Genomic instability

Age-associated DNA lesions, somatic mutations, chromosomal abnormalities, transposable-element activity, and impaired nuclear or mitochondrial DNA repair can erode cellular fitness.

  • Orthogonal assay examples: somatic variant burden, mutational signatures, copy-number change, micronuclei, DNA-damage foci, clonal hematopoiesis.
  • Key limit: mutation burden is assay-, depth-, lineage-, and selection-dependent; a DNA-damage marker alone does not identify the lesion or prove causal aging.
Primary 2

Telomere attrition

Progressive shortening or uncapping of chromosome ends can activate DNA-damage responses, replicative arrest, apoptosis, or genome rearrangement.

  • Orthogonal assay examples: telomere length distribution, critically short telomeres, telomere dysfunction-induced foci, telomerase activity.
  • Key limit: mean leukocyte telomere length is composition-sensitive and is neither a universal clock nor a direct measure of telomere dysfunction in every tissue.
Primary 3

Epigenetic alterations

Changes in DNA methylation, histone marks, chromatin accessibility, nucleosome organization, and transcriptional control can destabilize cell identity and stress responses.

  • Orthogonal assay examples: methylation drift, chromatin accessibility, histone marks, heterochromatin loss, enhancer and transcription-factor activity.
  • Key limit: an epigenetic-clock value is a trained predictor, not a complete measurement of this hallmark or proof that epigenetic change caused the phenotype.
Primary 4

Loss of proteostasis

Declining control of protein synthesis, folding, trafficking, quality control, and degradation permits damaged or aggregation-prone proteins to accumulate.

  • Orthogonal assay examples: chaperone response, ubiquitin-proteasome activity, aggregate load, unfolded-protein response, proteome insolubility.
  • Key limit: transcript abundance poorly approximates protein folding, turnover, localization, or aggregate toxicity without orthogonal measurements.
Primary 5 / New in 2023

Disabled macroautophagy

Reduced formation, maturation, or lysosomal clearance of autophagic cargo impairs recycling of proteins, lipids, glycogen, and organelles.

  • Orthogonal assay examples: autophagic flux, LC3 turnover, p62/SQSTM1, lysosomal function, mitophagy or other selective-autophagy assays.
  • Key limit: a single steady-state LC3 or autophagosome measurement cannot distinguish increased induction from blocked downstream clearance.
Layer 2 / Antagonistic

Adaptive programs that can become maladaptive

These processes often have non-linear effects. The same pathway may support survival during transient stress but promote dysfunction when chronically activated or suppressed.

Antagonistic 1

Deregulated nutrient sensing

Age remodels insulin/IGF-1 signaling, mTOR, AMPK, sirtuins, and nutrient-responsive transcription, altering growth, repair, autophagy, and metabolic flexibility.

  • Orthogonal assay examples: pathway phosphorylation, metabolite availability, fasting responses, mTOR/AMPK target activity, insulin sensitivity.
  • Key limit: pathway effects depend on tissue, timing, diet, sex, dose, and health state; cross-sectional expression is not pathway flux.
Antagonistic 2

Mitochondrial dysfunction

Altered electron transport, dynamics, biogenesis, mitophagy, metabolite signaling, and mitochondrial stress responses can impair energy homeostasis and cell communication.

  • Orthogonal assay examples: respiration, membrane potential, ATP, mtDNA integrity, NAD redox state, mitochondrial morphology, mitophagy.
  • Key limit: reactive oxygen species can signal adaptively as well as damage cells; mitochondrial abundance is not equivalent to mitochondrial function.
Antagonistic 3

Cellular senescence

A durable stress-associated state can combine cell-cycle exit, resistance to apoptosis, chromatin remodeling, metabolic change, and a context-specific secretory phenotype.

  • Orthogonal assay examples: multi-marker panels spanning arrest, damage, lysosomal activity, anti-apoptotic programs, and SASP factors.
  • Key limit: no single universal senescence marker exists; p16, p21, SA-β-gal, or a SASP transcript alone can be positive in non-senescent states.
Layer 3 / Integrative

From cell-state change to organismal decline

Integrative hallmarks are best evaluated with multi-scale data: cell-state measurements, tissue architecture, systemic exposures, and functional outcomes.

Integrative 1

Stem cell exhaustion

Loss of stem or progenitor number, fitness, lineage balance, or niche support reduces regenerative capacity and may favor dysfunctional or clonally selected populations.

  • Orthogonal assay examples: lineage tracing, clonogenicity, repopulation, differentiation bias, clonal diversity, niche-state measurements.
  • Key limit: marker-defined abundance is not regenerative function; lineage markers and stem-cell states differ by tissue and species.
Integrative 2

Altered intercellular communication

Endocrine, neuronal, immune, extracellular-vesicle, and local ligand-receptor signals are remodeled, disrupting coordination between cells and organs.

  • Orthogonal assay examples: paired ligand-receptor expression, spatial proximity, secreted proteins, receptor activity, perturbation and co-culture assays.
  • Key limit: computational ligand-receptor scores are hypotheses; co-expression does not establish secretion, binding, directionality, or functional signaling.
Integrative 3 / New in 2023

Chronic inflammation

Persistent, low-grade, often sterile immune activation can amplify tissue damage, senescence, metabolic dysfunction, and impaired regeneration.

  • Orthogonal assay examples: longitudinal cytokines, acute-phase proteins, immune-cell states, inflammasome activity, tissue immune infiltration.
  • Key limit: infection, obesity, medication, smoking, and disease can confound age associations; one cytokine is not a validated inflammaging index.
Integrative 4 / New in 2023

Dysbiosis

Age-associated shifts in microbial communities and host-microbe interactions may alter barrier integrity, metabolites, immunity, and inter-organ signaling.

  • Orthogonal assay examples: strain-resolved composition, functional genes, metabolites, barrier measures, host response, transfer or perturbation studies.
  • Key limit: diet, geography, medication, frailty, and care setting strongly shape microbiomes; taxonomic association alone does not establish causality. The LACA transcriptomic explorer contains host-response proxies only, not microbiome measurements.
Atlas evidence model

How to map an observation to a hallmark

A defensible annotation separates what was measured from what is inferred. LACA recommends recording four complementary evidence dimensions and preserving the exact tissue, cell type, species, age range, and assay.

Molecular observation

A marker, pathway score, molecule, lesion, or molecular complex changes with age.

  • Report assay, normalization, direction, effect size, uncertainty, and multiple-testing control.
  • Label as association unless temporal order or perturbation is shown.

Cell-state evidence

The signal localizes to a defined cell population and is supported by multiple compatible features.

  • Distinguish within-cell change from altered cell-type abundance.
  • Use donor-level replication; cells are not independent biological replicates.

Tissue or organism function

The molecular or cellular state co-varies with a relevant physiological, regenerative, disease, or survival outcome.

  • Preserve tissue and species scope.
  • Adjust for study design and major confounders without over-claiming mediation.

Perturbational evidence

Manipulating the candidate mechanism changes both the hallmark readout and an aging-relevant phenotype.

  • Require target engagement, suitable controls, and adverse-effect assessment.
  • Separate lifespan, healthspan, disease treatment, and biomarker movement.
Scientific guardrails

What the hallmark framework does not establish

The framework is a powerful organizing model, but it is not a substitute for validated endpoints, causal design, or clinical evidence.

Not twelve independent variables

Hallmarks overlap and form feedback loops. Assigning one observation to several hallmarks may be biologically appropriate, provided the evidence for each link is stated.

Example: impaired mitophagy can connect disabled macroautophagy, mitochondrial dysfunction, inflammation, and altered communication.

Not a universal aging score

There is no accepted formula that weights the twelve hallmarks into a diagnostic age, nor universal thresholds that classify an individual as biologically old.

Use validated clocks or functional outcomes for their intended estimand; do not convert hallmark counts into a clinical claim.

Human causality remains uneven

Much mechanistic and lifespan evidence comes from yeast, worms, flies, and rodents. Human studies are often observational, tissue-limited, or disease-enriched.

Cross-species conservation strengthens plausibility but does not guarantee the same intervention effect or safety in humans.

Biomarker movement is not rejuvenation

A shifted marker, pathway score, or clock can reflect composition, acute stress, assay artifacts, or an on-target response without improved long-term function.

Clinical interpretation requires reproducibility, durable functional benefit, and safety in an appropriately controlled study.
Provenance

References & source resources

Primary literature and official resource pages used to define this research guide.

Lรณpez-Otรญn C, Blasco MA, Partridge L, Serrano M, Kroemer G. Hallmarks of aging: An expanding universe. Cell. 2023;186(2):243-278. โ€” DOI
Lรณpez-Otรญn C, Blasco MA, Partridge L, Serrano M, Kroemer G. The hallmarks of aging. Cell. 2013;153(6):1194-1217. โ€” DOI
Schmauck-Medina T et al. New hallmarks of ageing: a 2022 Copenhagen ageing meeting summary. Aging. 2022;14(16):6829-6839. โ€” DOI
Gorgoulis V et al. Cellular Senescence: Defining a Path Forward. Cell. 2019;179(4):813-827. โ€” DOI