CDKN2BAS

ANRIL and Functional Aging — The 9p21.3 Longevity Locus

A single stretch of chromosome 9 — the 9p21.3 locus — is the most consistently
replicated region in genome-wide association studies for age-related disease.
Coronary artery disease, type 2 diabetes, glioma, melanoma, and now physical
aging have all been linked to variation here. At the heart of the locus sits
ANRIL (antisense non-coding RNA in the INK4 locus) | Also designated CDKN2BAS or CDKN2B-AS1 — a long non-coding RNA transcribed antisense to the CDKN2A/CDKN2B tumor suppressor genes,
a molecular rheostat for cellular senescence.

rs2811712 is a tag SNP within ANRIL that was identified in a landmark study as
associated with physical function in older people | Melzer D et al. A common variant of the p16INK4a genetic region is associated with physical function in older people. Mech Ageing Dev. 2007.
The G allele — carried by roughly one in five people of European ancestry — is
associated with substantially better preserved physical function in old age, while
the common A allele is the risk-conferring genotype.

The Mechanism

The 9p21.3 locus contains three protein-coding genes — CDKN2A (encoding
p16INK4a and p14ARF), CDKN2B (encoding p15INK4b), and MTAP — all flanked and
overlapped by ANRIL. p16INK4a and p15INK4b are
cyclin-dependent kinase inhibitors | CDK inhibitors block CDK4/6, preventing phosphorylation of Rb and halting cell-cycle progression from G1 to S phase — the canonical senescence checkpoint
that drive cellular senescence: as cells age and accumulate damage, p16 levels
rise, arresting the cell cycle and converting cells into
senescent "zombie" cells | Senescent cells stop dividing but remain metabolically active, secreting pro-inflammatory cytokines (the SASP — senescence-associated secretory phenotype) that damage neighboring tissue
that fuel the chronic inflammation underlying age-related functional decline.

ANRIL regulates CDKN2A/CDKN2B expression in cis through Polycomb group protein
recruitment, particularly the PRC2 complex, which methylates histone H3K27 to
silence the INK4 locus. In proliferating cells, ANRIL keeps senescence genes
suppressed; as ANRIL expression shifts with aging or genetic variation, the
balance tips toward elevated p16 and p15, accelerating the senescent phenotype.
rs2811712 is an eQTL for CDKN2B expression: studies show that variation here
correlates with altered ANRIL isoform abundance and downstream CDKN2B levels,
providing a molecular pathway from genotype to functional aging phenotype.

The Evidence

The primary association was established by
Melzer et al. (2007) | A common variant of the p16INK4a genetic region is associated with physical function in older people. Mechanisms of Ageing and Development.
across three independent European cohorts totaling 3,372 elderly individuals (EPIC-Norfolk sets 1 and 2 plus InCHIANTI). Severely limited physical function — defined by performance-based tests of walking speed, chair stand, and balance — was present in 15.0% of AA homozygotes versus 7.0% of GG homozygotes, nearly half the rate. The per-A-allele odds ratio was 1.48 (95% CI 1.17–1.88, p = 0.001), adjusted for age, sex, and study. The association held across all three cohorts tested independently, including a trend in the Iowa-EPESE cohort (n=419, p=0.079 one-sided), making this one of the first robust genetic associations for functional aging specifically.

A Han Chinese case-control study
Wu et al. (2012) | Heterozygote genotypes at rs2222823 and rs2811712 SNP loci are associated with cerebral small vessel disease in Han Chinese population
found the heterozygote genotype at rs2811712 was associated with OR 1.75 (CI 1.13–2.71, p=0.004) for cerebral small vessel disease — consistent with the broader pattern that the A (risk) allele at this locus impairs vascular and neurological aging beyond musculoskeletal function.

The broader 9p21.3 locus literature confirms that ANRIL variants influence
CDKN2B expression and cellular senescence | SNPs here show inverse effects on ANRIL and CDKN2B expression, supporting a role of antisense transcription in regulating senescence pathways,
with the risk alleles generally increasing ANRIL expression in ways that paradoxically reduce appropriate p16/p15 upregulation in vascular cells, impairing the normal senescence response to damage. This locus is the strongest genetic signal for coronary artery disease in Europeans (OR ~1.3 per risk allele), though rs2811712 itself does not independently associate with CAD — its primary documented phenotype is functional physical aging.

Practical Actions

For carriers of two A alleles (AA), the evidence supports a targeted approach to
slowing the accumulation of senescent cells and their inflammatory secretome.
The most directly relevant interventions for the CDKN2B/p16 axis include:

• Senolytic compounds: Fisetin (a flavonoid concentrated in strawberries and onions) and quercetin have demonstrated senolytic activity in human studies — selectively promoting apoptosis in p16-expressing senescent cells. Intermittent high-dose fisetin (≥500 mg/day for 2–3 days per month) reduced senescent cell burden and SASP markers in published clinical data.
• Cardiovascular monitoring: The 9p21.3 region is the strongest common genetic locus for coronary artery disease. While rs2811712 itself does not independently associate with CAD, it tags a haplotype block where other risk variants reside. AA carriers should ensure thorough cardiovascular risk assessment.
• Strength and resistance training: p16-driven muscle senescence contributes directly to sarcopenia (age-related muscle loss) — the leading mediator of physical impairment in older adults. Resistance training specifically reduces senescent cell burden in skeletal muscle and preserves muscle fiber function through mechanisms that parallel the CDKN2B pathway.

AG carriers carry one protective G allele: physical function risk is intermediate,
and the same principles apply with moderated urgency.

Interactions

rs10757278 and rs1333049 (9p21.3 haplotype): These are the primary
CAD-associated SNPs at the 9p21 locus, in moderate-to-high LD with rs2811712.
If you carry the risk haplotype at rs10757278 (G allele) in addition to AA at
rs2811712, the combined burden on ANRIL/CDKN2B regulation is likely greater
than either marker alone. The 9p21.3 block spans ~58 kb and multiple SNPs
contribute to its overall phenotypic effect.

rs564398 (upstream of CDKN2A): Another 9p21 tag SNP associated with type 2
diabetes and glioma risk, which shares the ANRIL regulatory context. Combined
risk at rs2811712 and rs564398 may implicate broader ANRIL dysregulation
beyond the physical aging phenotype.