Mutations in the dNTPase SAMHD1 cause the interferonopathy Aicardi–Goutières syndrome (AGS) and have been linked to the development of certain cancers, but the underlying mechanisms have been unclear. Coquel et al. have found that SAMHD1 promotes the degradation of nascent DNA strands at stalled replications forks, thereby preventing the intracellular accumulation of single-stranded DNA (ssDNA) fragments that can induce the expression of pro-inflammatory type I interferons (IFNs).

figure a

S. Bradbrook/Macmillan Publishers Limited

SAMHD1 protects quiescent cells against viruses such as HIV, but its dNTPase activity is inhibited in cycling cells owing to phosphorylation by cyclin-dependent kinases (CDKs). As SAMHD1 has been shown to colocalize with DNA repair foci in cells exposed to genotoxic agents, the authors hypothesized that it might have a role in inhibiting ssDNA accumulation at stalled forks. They first showed that knockdown of SAMHD1 expression in a cell line increased the presence of cytosolic ssDNA and induced mRNAs encoding IFNs and TNF. Moreover, ssDNA accumulation and inflammatory gene expression were increased when cells were treated with hydroxyurea, a genotoxic agent that causes replication fork stalling. The authors showed that the induction of IFN genes in the absence of SAMHD1 was driven by the cGAS–STING pathway and also depended on IRF3.

Subsequent studies showed that SAMHD1 colocalizes with replication forks in cells. Notably, SAMHD1-depleted cells grew more slowly than control cells and had a longer S phase owing to slower DNA synthesis. However, dNTP levels were not markedly altered in the absence of SAMHD1, suggesting that the roles of SAMHD1 in the S phase of the cell cycle extend beyond regulation of dNTP pools. To examine whether the phosphorylation of SAMHD1 by CDKs during the S and G2–M phases of the cell cycle affects DNA replication, the authors generated phosphomimetic (T592E) and non-phosphorylatable (T592A) SAMHD1 mutants. Complementing SAMHD1-depleted cells with T592E, but not with T592A, mutants rescued slowed replication forks, whereas both sets of cells had similar dNTP levels. Fork progression was also restored by a dNTPase-deficient SAMHD1 mutant. Therefore, phosphorylation of SAMHD1 by CDKs, but not the dNTPase activity of SAMHD1, seems to be necessary for fork progression.

SAMHD1 promotes the degradation of nascent DNA strands at stalled replications forks

Further detailed experiments showed that SAMHD1 promotes fork resection, a process involving the degradation of nascent DNA. The authors found that SAMHD1 activates the exonuclease activity of the double-strand break repair protein MRE11A, with SAMHD1 and MRE11A cooperating to activate the ATR–CHK1 checkpoint pathway and promote fork restart in cells exposed to genotoxic agents. Finally, the authors hypothesized that in the absence of SAMHD1, helicases and nucleases may promote the displacement of nascent DNA and the accumulation of ssDNA in the cytosol. In support of this, they found that an IFN response was not induced in SAMHD1-depleted cells lacking the ATP-dependent DNA helicase RECQ1.

In summary, SAMHD1 activates the ATR–CHK1 pathway at stalled DNA replication forks and prevents the release and accumulation of ssDNA molecules that can trigger pro-inflammatory responses via the cGAS-STING pathway. These findings provide important insight into why SAMHD1 mutations are linked to AGS and cancer development.