STING mediates immune responses in a unicellular choanoflagellate

Arielle Woznica, Ashwani Kumar, Carolyn R. Sturge, Chao Xing, Nicole King, Julie K. Pfeiffer

Preprint posted on 14 May 2021

Article now published in eLife at


The STING ligand 2’3’-cGAMP induces an NF-κB-dependent anti-bacterial innate immune response in the starlet sea anemone Nematostella vectensis

Shally R. Margolis, Peter A. Dietzen, Beth M. Hayes, Stephen C. Wilson, Brenna C. Remick, Seemay Chou, Russell E. Vance

Preprint posted on 14 May 2021

Article now published in Proceedings of the National Academy of Sciences at

Evolution of STING – Ancient Functions of a Critical Immune Effector

Selected by Connor Rosen


Animal immune systems include numerous specialized immune features, such as the vertebrate-specific interferon cascade and adaptive immune system.However, other aspects of animal immune systems, including various pathogen sensors, are evolutionarily ancient and common to most branches of animal life. An outstanding question in the evolution of the immune system is whether these sensors always carried immune function, and if not, how those functions were acquired. Efforts to understand these evolutionary questions are aided by studies of early-branching taxa that may carry signatures of functions that were already present in early common ancestors.

The closest relatives of all animals are the choanoflagellates, which are unicellular marine eukaryotes. Within animals, many well-studied model organisms are Bilaterians (having bilateral symmetry during development), including classical vertebrate and invertebrate models. The Cnidaria, aquatic animals such as jellyfish, sponges, and anemones, are the “sister group” of Bilateria. Therefore, studies of choanoflagellates and Cnidaria provide critical context for the origins of animals and Bilaterians, respectively.

In vertebrates, the stimulator of interferon genes (STING), senses cyclic dinucleotides (CDNs) produced by cyclic GMP-AMP synthase (cGAS) to trigger interferon signaling and an anti-viral immune response. Although the downstream interferon response pathway is a vertebrate innovation, the cGAS-STING pathway is detectable in a broad range of metazoans, including the Cnidarian sea anemone Nematostella vectensis. It has previously been shown that STING binding to CDNs is evolutionarily conserved in Cnidaria [Kranzusch and Wilson et al, Mol Cell 2015], and that it can substitute for mammalian STING in at least some functions [Gui and Yang et al, Nature 2019]. In a pair of new preprints, Woznica et al and Margolis et al explore the immune functions of STING in choanoflagellates and Nematostella vectensis, clarifying the ancient anti-bacterial roles of STING.

Key findings:

  • Woznica et al characterize the immune activity of the choanoflagellate Monosiga brevicollis, and establish it as a new model system for studying the evolution of animal innate immunity. By screening numerous bacteria in co-culture with brevicollis, they characterize Pseudomonas aeruginosa as a novel pathogen for M. brevicollis. They show that it induces cell death and drives the upregulation of STING transcriptionally, as measured by RNA-seq, and at the protein level, using a newly developed and validated antibody for Western blotting. The authors developed transfection and CRISPR-based knockout protocols to enable expression of tagged proteins for microscopy and isolated a clonal STING knockout linease. Using these overexpression and knockout techniques, they showed that 2’3’cGAMP (the CDN ligand of mammalian STING) drives STING upregulation, cell death, and the induction of autophagy, and that these activities depend on the expression of endogenous STING.
  • Margolis et al study the endogenous function of STING in vectensis, using RNA-seq to show that treatment with CDNs drives activation of an anti-bacterial gene program (such as LBP and Dae4) in addition to expression of anti-viral genes (OAS, Viperin, and STING itself). While they did not formally demonstrate a dependence on STING, they were unable to generate full knockdown or knockout of STING, and therefore this remains an untested question. They were able to show effects of knockdown on the N. vectensis homolog of NFkb on this gene response, rather than the interferon-system-like IRFs or STAT homologs present in N. vectensis. These genes were additionally activated during bacterial challenge with P. aeruginosa. Furthermore, they biochemically characterized anti-bacterial homologs of Dae4 (a peptidoglycan cleaving enzyme capable of direct bactericidal activity on gram-positive bacteria) and LBP (a lipopolysaccharide binding protein that targets gram-negative bacterial outer membranes) to demonstrate the expected anti-bacterial activities.


These two preprints firmly establish the evolutionarily ancient role of STING / CDN sensing as a mediator of anti-bacterial immunity through clear studies of endogenous activity in choanoflagellates and the cnidarian N. vectensis. This reinforces the early role of STING in immune activation, prior to its integration into anti-viral sensing and the interferon system more recently in the evolution of animal immunity. These reports, and particularly the development of a new model system in M. brevicollis, are critical to advance our understanding of the evolution of innate immunity through animal evolution.

Moving forward:

  • Both preprints comment on the role of cGAS in activating STING responses, and the need for further investigation to clarify the role of cGAS activation upstream of STING in brevicollis and N. vectensis. These will be extremely interesting developments to follow to understand how an immune circuit evolved – and whether STING ever operated “independently” before co-opting cGAS activity, or whether the “minimal unit” of this antiviral circuit required both genes.
  • Both studies use treatment with CDNs as a key experimental technique to trigger immune responses. While pathogen-derived CDNs showed no activation of brevicollis STING, this may be due to differences in uptake or transport. Recognizing that P. aeruginosa is the first identified pathogen for M. brevicollis, what might be the relative “dangers” from extracellular vs intracellular pathogens (which may deliver CDNs at higher concentrations)? Could STING even be acting as a kind of “food quality control” mechanism to enhance autophagy when certain bacteria are ingested? Some of these questions could be informed by the contribution of cGAS to STING activation, as described above, or biochemical studies with purified STING assessing the activity of different CDNs in closer detail.
  • STING is generally described as an anti-viral protein in mammalian cells – there was a clear antiviral response in N. vectensis, and further examination of the genes induced in M. brevicollis will be needed to clarify if putative anti-viral homologs are induced there. What kinds of pressures might drive the broadening/narrowing of STING functions in the innate immune system through animal evolution?


Posted on: 1 July 2021 , updated on: 2 July 2021


(No Ratings Yet)

Have your say

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Sign up to customise the site to your preferences and to receive alerts

Register here

preLists in the evolutionary biology category: