Hypothesis

Aging‑associated shifts in cfDNA fragment size distribution bias epigenetic clock readouts by preferentially releasing short, hypomethylated fragments from open chromatin regions.

Rationale

1. cfDNA fragment size reflects nucleosome protection patterns; shorter fragments arise from less protected DNA, often at promoters and enhancers.[1]
2. With age, global hypomethylation and inflammaging increase accessibility of immune‑gene loci, making them more prone to nuclease digestion and thus over‑represented in the short cfDNA pool.[2]
3. Epigenetic clocks (Horvath, PhenoAge, GrimAge) weight CpG sites that are themselves enriched in promoter‑associated regions, many of which show age‑related hypomethylation (e.g., ELOVL2 promoter).
4. Therefore, an age‑dependent increase in the proportion of short cfDNA fragments will amplify the signal from hypomethylated clock CpGs, leading to an overestimation of biological age when total cfDNA is analysed without size fractionation.

Mechanistic Insight

• Oxidative stress and reduced DNMT1 activity promote loss of methylation at nucleosome‑depleted regions.
• Increased caspase‑3 and DNASE1L3 activity in aging drives apoptosis that preferentially cuts linker DNA, generating ~90‑bp mononucleosome fragments from open chromatin.
• These fragments retain the hypomethylation signature of their source loci, while longer fragments (>200 bp) derive from more protected heterochromatin where methylation is relatively preserved.
• Consequently, the short‑fraction methylome mirrors a ‘transcription‑active, inflammaging’ epigenotype, whereas the long‑fraction reflects a ‘structural’ epigenotype.

Testable Predictions

| Prediction | Experimental Approach | Expected Outcome if Hypothesis Holds |------------|----------------------|--------------------------------------| | P1: Short cfDNA fraction yields higher epigenetic age estimates than long fraction in individuals >60 y. | Isolate plasma cfDNA, size‑select using SPRI beads (short <150 bp, long >150 bp), perform bisulfite sequencing, compute Horvath/PhenoAge/GrimAge. | Short fraction age > long fraction age; difference grows with chronological age. | P2: The age‑related increase in short‑fraction proportion correlates with circulating inflammaging markers (IL‑6, CRP). | Measure fragment size distribution via low‑pass sequencing or fragment analyzer, correlate % short fragments with plasma cytokine levels. | Positive correlation (r > 0.4, p < 0.01). | P3: Enrichment of hypomethylated clock CpGs (e.g., cg16867657 in ELOVL2) in short fraction relative to long fraction. | Targeted bisulfite amplicon sequencing of clock CpGs in each fraction. | Short fraction shows significantly lower β‑values (more hypomethylated) at promoter‑associated clock CpGs. | P4: Artificially shifting fragment size toward longer fragments (e.g., by inhibiting DNASE1L3) reduces the apparent epigenetic age of total cfDNA. | Treat ex‑vivo plasma with DNASE1L3 inhibitor, re‑measure size distribution and epigenetic age. | Reduction in % short fragments accompanied by decreased epigenetic age estimate.

Falsifiability

If short and long fractions produce indistinguishable epigenetic age estimates, or if the short fraction shows hypermethylation at clock CpGs, the hypothesis would be refuted. Similarly, lack of correlation between short‑fragment proportion and inflammaging markers would undermine the mechanistic link.

Implications

• Confirms that fragment‑size bias contributes to the ‘inflammaging’ signal captured by current cfDNA clocks.
• Suggests that size‑fractionated cfDNA methylation could yield two complementary clocks: a ‘short‑fraction inflammaging clock’ and a ‘long‑fraction structural clock’.
• Guides clinical cfDNA assays to incorporate fragment‑size sorting for improved specificity.

References

[1] https://pmc.ncbi.nlm.nih.gov/articles/PMC12833446/ [2] https://pubmed.ncbi.nlm.nih.gov/40164165/ [3] https://www.openpr.com/news/4406053/united-states-dna-methylation-sequencing-market-2026-growth