Research Article

Phosphorylation and chromatin tethering prevent cGAS activation during mitosis

See allHide authors and affiliations

Science  19 Mar 2021:
Vol. 371, Issue 6535, eabc5386
DOI: 10.1126/science.abc5386

You are currently viewing the abstract.

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

Keeping cGAS silent

Cells detect microbial and self-DNA in the cytosol as a danger signal that triggers immune and inflammatory responses. Paradoxically, a large fraction of a DNA-sensing enzyme called cGAS is tightly associated with the chromatin, especially during mitosis. Li et al. uncovered two mechanisms that prevent cGAS from being activated by the chromatin DNA (see the Perspective by Ablasser). First, cGAS is hyperphosphorylated as cells enter mitosis, thereby inhibiting its DNA-binding and liquid-liquid phase separation, which promotes cGAS activation. Second, chromatin-bound cGAS is unable to oligomerize, a process required for its activation. Together, these mechanisms ensure cGAS inactivation during cell division to prevent autoimmune reactions.

Science, this issue p. eabc5386; see also p. 1204

Structured Abstract


Cyclic guanosine monophosphate (GMP)–adenosine monophosphate (AMP) synthase (cGAS) detects microbial DNA in the cytosol to trigger innate immune responses. cGAS also senses self-DNA in the cytoplasm to promote sterile inflammation and cellular senescence. cGAS binds to double-stranded DNA in a sequence-independent manner, and the multivalent interactions between cGAS and DNA lead to their liquid-liquid phase separation and cGAS activation in the resulting liquid droplets. Once activated, cGAS catalyzes the synthesis of cyclic GMP-AMP (cGAMP), a second messenger that activates the adaptor protein stimulator of interferon genes (STING), leading to the induction of type I interferons and other immune mediators. Paradoxically, a large fraction of cGAS is tightly associated with chromatin, especially during mitosis when the nuclear envelope breaks down. How cGAS activity is regulated during mitosis remains poorly understood.


Accumulating evidence reveals both cytoplasmic and nuclear localization of cGAS in cells. In the nucleus, cGAS is tethered to the chromatin through binding to histones that form the nucleosome core; this interaction has been shown to inhibit cGAS activation by chromatin DNA. However, it is unclear whether additional mechanisms exist to regulate cGAS activity, especially during the cell cycle transition. In particular, the nuclear envelope disassembles when cells enter mitosis, allowing cytoplasmic cGAS to mix with the chromosomes. Direct measurements of cGAS activity and posttranslational modifications at different stages of the cell cycle might unravel the mechanism of cGAS regulation and lead to a better understanding of how cells prevent inadvertent activation of cGAS by chromatin DNA, which might otherwise lead to inflammation and autoimmune responses.


Using a biochemical assay, we found that cGAS activity in human cell lines was selectively suppressed during mitosis. We further showed that when cells enter mitosis, cGAS is hyperphosphorylated by Aurora kinase B and other kinases at its disordered and positively charged N terminus, a region important for DNA binding and liquid phase separation. Mutations that mimic N-terminal hyperphosphorylation abolished cGAS activity, whereas those blocking hyperphosphorylation led to enhanced expression of interferon-stimulated genes. We also developed a split green fluorescent protein (GFP) complementation assay to detect cGAS oligomerization. Chromatin-associated cGAS could not oligomerize, which elucidates how chromatin tethering inhibits cGAS activity in live cells. We found that cGAS mutants defective in chromatin tethering were activated by chromatin DNA in a manner dependent on an unphosphorylated N terminus, underscoring the importance of the cGAS N terminus in sensing chromatin DNA. Unexpectedly, deletion of the N terminus from cGAS exposed a cryptic mitochondrial localization sequence that targeted cGAS to the mitochondrial matrix, where mitochondrial DNA activated cGAS. This result explains why cGAS that lacks the N terminus is constitutively active. Together, our results revealed that hyperphosphorylation and chromatin tethering act in parallel to suppress cGAS activity during mitosis.


This work presents direct evidence that cGAS activity is suppressed during mitosis through two mechanisms: hyperphosphorylation at the N terminus and chromatin tethering that inhibits cGAS oligomerization. Both mechanisms prevent cGAS phase separation into liquid droplets where cGAS could efficiently synthesize cGAMP. This fail-safe mechanism of inhibition keeps cGAS silent across the normal cell cycle transition, potentially thereby avoiding autoimmune reactions to nuclear DNA.

Regulation of cGAS activity during cell cycle transition.

When cells enter mitosis, cGAS loses its ability to be activated by DNA (top). During interphase (bottom left), cGAS is poised for activation by DNA in the cytoplasm (blue) but cannot be activated in the nucleus by chromatin (orange). During the G2-M transition (bottom middle), the N terminus of cGAS is hyperphosphorylated by Aurora kinase B (AurB) and other kinases, thereby blocking cGAS liquid phase separation and activation (red). Upon nuclear envelope breakdown, cytoplasmic cGAS is recruited to condense chromosomes and becomes further inhibited by chromatin tethering, which prevents cGAS oligomerization during mitosis (bottom right). +, positive charge; P, phosphorylation.


The cyclic guanosine monophosphate (GMP)–adenosine monophosphate (AMP) synthase (cGAS) detects microbial and self-DNA in the cytosol to activate immune and inflammatory programs. cGAS also associates with chromatin, especially after nuclear envelope breakdown when cells enter mitosis. How cGAS is regulated during cell cycle transition is not clear. Here, we found direct biochemical evidence that cGAS activity was selectively suppressed during mitosis in human cell lines and uncovered two parallel mechanisms underlying this suppression. First, cGAS was hyperphosphorylated at the N terminus by mitotic kinases, including Aurora kinase B. The N terminus of cGAS was critical for sensing nuclear chromatin but not mitochondrial DNA. Chromatin sensing was blocked by hyperphosphorylation. Second, oligomerization of chromatin-bound cGAS, which is required for its activation, was prevented. Together, these mechanisms ensure that cGAS is inactive when associated with chromatin during mitosis, which may help to prevent autoimmune reaction.

View Full Text

Stay Connected to Science