Telomeric proteins may have an unexpected connection to the mitotic spindle. On page 819, Xu and Blackburn provide evidence that the human homologue of a yeast telomere-binding protein is found on anaphase midzone microtubules.

The telomeric functions of the human and yeast versions of this protein, Rif1, have diverged during evolution. In budding yeast, Rif1 is a telomere-bound negative regulator of telomere length. But recent results have shown that the human version (hRif1) only binds to damaged telomeres, as well as sites of damage elsewhere in the genome (Silverman et al. Genes Dev. 18:2108–2119).

Xu and Blackburn find that hRif1 also binds to microtubules. In early anaphase, hRif1 was found on spindle microtubules between separating chromosomes. This position is shared with chromosomal passenger proteins, which monitor DNA–microtubule attachments. The microtubule-localized hRif1 may thus affect mitotic functions, potentially including spindle stabilization or anaphase progression. If so, HRIF1 knock-out cells should have segregation defects. Such defects were seen in budding yeast lacking Rif1, although the problems were attributed to faulty telomeres.

Two other telomeric proteins, TRF1 and tankyrase, have been linked to mitosis through their mitotic spindle and centrosomal localizations, respectively. If perhaps sister chromatids were once connected at telomeres as well as centromeres, it might explain why telomere-associated proteins are used during mitosis.