The cryo-EM structure of human CST reveals a two-megadalton decameric assembly bound to telomeric DNA

Abstract

The single-stranded DNA-binding CTC1-STN1-TEN1 (CST) complex is essential for telomere maintenance and genome-wide replication recovery, processes that are critical for genome stability. Here, we report the 2.95 Å cryo-EM structure of human CST bound to telomeric single-15 stranded DNA, which unexpectedly assembles as a decameric supercomplex. The atomic model of the 134 kDa CTC1, built almost entirely de novo, reveals the overall architecture of CST and the DNA-binding anchor site. In situ arrangements of STN1 and TEN1 are revealed, with STN1 interacting with CTC1 at two separated sites, allowing allosteric mediation of CST decameric assembly. Surprisingly, CTC1 lacks the anticipated structural homology to yeast Cdc13 but instead 20 shares similarity with a form of Replication Protein A. The atomic-resolution model of human CST provides crucial mechanistic understanding of CST mutations associated with human diseases. Moreover, the decameric form of CST suggests the intriguing possibility of ssDNA architectural organization similar to what the nucleosome provides for dsDNA. One Sentence Summary:25 Human telomeric single-stranded DNA triggers the assembly of a decameric protein supercomplex solved by cryo-EM. Main Text:CST is a heterotrimeric ssDNA-binding complex present in many species (1-3). Although CST 30 was initially discovered as a DNA polymerase-primase cofactor (4) and telomere-associated protein complex (2, 5-7) essential for telomere replication, its role has now expanded genomewide, with CST key in recovering stalled replication forks (2,(8)(9)(10)(11) and facilitating DNA damage repair (12)(13)(14)(15). Consequently, mutations in CST are the basis of human genetic diseases such as Coats Plus Syndrome and Dyskeratosis congenita (16-21). 35 Although CST is similar to the telomere-associated POT1 regarding preferential binding of G-rich ssDNA (22-25), CST is also able to bind non-specific sequences within longer ssDNA (6, 25), analogous to human RPA. Studies have tried to determine the DNA-binding region(s) of CST to

Highly accurate protein structure prediction with AlphaFold

“…We show experimental structures from the PDB with accessions 6Y4F 76 , 6YJ1 77 , 6VR4 78 , 6SK0 79 , 6FES 80 , 6W6W 81 , 6T1Z 82 , and 7JTL 83 .…”