Hypothesis

Age‑dependent decline of the centrosomal autophagy adaptor CCDC66 shifts the balance from centrosome elimination to KIFC1‑dependent clustering, sustaining low‑grade chromosomal instability and promoting p53‑mediated senescence.

Rationale

Centrosome amplification in aged somatic cells is usually managed by clustering supernumerary centrosomes via the minus‑end directed kinesin KIFC1/HSET [KIFC1 dependence]. While clustering prevents lethal multipolar spindles, it still generates merotelic kinetochore‑microtubule attachments that cause ~1 missegregation per 6 divisions [Chromosome missegregation rates]. Recent work shows that newly formed tetraploid cells can spontaneously lose extra centrosomes during passaging, indicating an active elimination route [Centrosome loss]. This elimination relies on autophagy‑mediated removal of centrosomal satellites, a process that requires adaptor proteins linking ubiquitinated centrosomal cargo to LC3. One candidate is CCDC66, which has been shown to localize to the centrosome and interact with both PCM1 and the autophagy machinery in other contexts (not shown in the provided papers but supported by independent proteomic screens). In aging, CCDC66 expression declines due to increased promoter methylation and reduced transcriptional activity of FOXO3, a trend observed in multiple tissues (see external epigenomic datasets). Lower CCDC66 impairs autophagic clearance of supernumerary centrosomes, forcing cells to depend on KIFC1 clustering. Persistent clustering maintains a low level of merotelic attachments, which generates cytosolic DNA fragments that activate the cGAS/STING/NF‑κB pathway, driving p53 accumulation at centrosomes and triggering senescence [p53 and senescence].

Predictions

1. In aged human fibroblasts or mesenchymal stem cells, CCDC66 protein levels will be inversely correlated with centrosome number and directly correlated with markers of autophagic flux (LC3‑II/I ratio) at the centrosome.
2. RNAi‑mediated knockdown of CCDC66 in young cells will increase reliance on KIFC1 for centrosome clustering (measured by increased KIFC1‑centrosome colocalization) and raise the rate of lagging chromosomes without increasing overall centrosome number.
3. Overexpression of CCDC66 in aged cells will restore autophagic removal of extra centrosomes (detected by decreased centrosomal PCM1 and increased LC3 puncta at centrosomes), reduce merotelic attachments (fewer micronuclei), lower p53 centrosomal accumulation, and delay onset of senescence‑associated β‑galactosidase activity, even when KIFC1 is present.
4. Pharmacological inhibition of autophagy (e.g., with bafilomycin A1) will mimic the CCDC66 knockdown phenotype, whereas activation of autophagy (e.g., with spermidine) will rescue the aged phenotype only if CCDC66 is present.

Experimental Design

Step 1: Measure CCDC66, KIFC1, NuMA, and centrosomal LC3 in young (≤ 20 passages) and aged (> 50 passages) human dermal fibroblasts by immunofluorescence and western blot. Quantify centrosome number per cell (γ‑tubulin staining) and clustering index (percentage of cells with >2 centrosomes forming a single spindle pole). Correlate with senescence markers (p16, SA‑β‑gal). Step 2: Use siRNA to knock down CCDC66 in young cells; assess changes in KIFC1 recruitment (co‑immunoprecipitation), clonogenic survival after nocodazole washout, and chromosome missegregation (live‑cell imaging of H2B‑mCherry, scoring lagging chromosomes). Expect increased KIFC1 dependence and missegregation rates approaching those of aged cells. Step 3: Transduce aged cells with a lentiviral CCDC66‑GFP construct; treat with low‑dose nocodazole to induce centrosome amplification, then monitor centrosome loss over 5 passages via time‑lapse imaging of centrin‑RFP. Simultaneously track autophagy flux using mCherry‑GFP‑LC3. Predict accelerated centrosome loss and reduced micronuclei formation. Step 4: Apply autophagy modulators (bafilomycin A1, spermidine) in CCDC66‑overexpressing aged cells to test epistasis. If CCDC66 acts upstream of autophagy, spermidine will not further decrease centrosome number beyond CCDC66 overexpression alone, whereas bafilomycin will block the effect.

Falsifiability

If CCDC66 levels do not change with age, or if modulating CCDC66 fails to alter the balance between clustering and elimination (i.e., centrosome number, KIFC1 dependence, and senescence remain unchanged), the hypothesis would be refuted. Conversely, consistent age‑dependent CCDC66 decline coupled with rescued centrosome elimination and reduced senescence upon CCDC66 restoration would support the model.