PREreview of cGLRs are a diverse family of pattern recognition receptors in animal innate immunity

By Mauricio P. Contreras¹*, Daniel Lüdke¹*, Andrés Posbeyikian¹* and AmirAli Toghani¹*

1: The Sainsbury Laboratory, University of East Anglia, Norwich, UK.

*All authors contributed equally to the review, authors are ordered in alphabetical order according to their last names.

This paper reports on the discovery of cGAS-like receptors (cGLRs) as a major family of pattern recognition receptors (PRRs) in animal innate immunity. Through a bioinformatic approach, the authors identified over 3,000 cGLRs present in nearly all metazoan phyla and identified 15 novel active cGLRsusing a forward biochemical screen . These receptors respond to common long double-stranded nucleic acid PAMPs and synthesize nucleotide second messenger immune signals, including novel products that contain pyrimidine bases. The authors describe how diversification of cGLR nucleotide second messengers and STING receptors enables animals to establish complex networks for pathogen detection. Together, these findings establish cGLRs as a widespread family of PRRs that form a network of signaling pathways in animal cells to sense diverse microbial pathogens. Overall, the manuscript is well written and the findings are novel and exciting. The bioinformatic analysis is robust and a comprehensive approach was taken with many genomes being mined. The pipeline is coherent and proper tools were employed. The authors describe the spectrum of cGLRs extensively, resulting in a very complete story.

It should be noted that we are not working in the field of animal innate immunity, but plant immunity and molecular plant-microbe interactions. However, we enjoyed reading this work a lot and it was interesting to see parallels between metazoan and plant innate immunity as the description of cGLR/STING small-molecule based immune receptor networks exhibits many parallels to TIR-NLR/CC_R-NLR signaling networks in plant immunity. There’s no cGAS in plants, so these mechanisms may have convergently evolved. Below, we include some comments and thoughts that arose while discussing this Pre-Print, which we hope the authors may find useful.

General comments

 -In general, the authors may want to consider replacing bar plots with box plots with overlaid scatter-plots when possible, such as in Figure 1B or Figure 2. See Weissgerber et al., 2015 (PLoS Biology) for useful suggestions. This helps the reader see the actual variability in the data and draw conclusions.

-It would also be helpful to include the number of biological/technical replicates in the experiments, possibly in the figure legends?

Comments/Questions by figure

Figure 1

-Regarding the second criterion for the manual curation of the HMM+PSI-BLAST-obtained dataset, was there a threshold for sequence homology? What was the cut-off? Similarly, for the third criterion, is there an RMSD cutoff used? Could the authors elaborate on this criteria some more, perhaps in the methods section?

-Could the authors provide a more specific definition of phylogenetic clusters? Additionally, while it may be difficult to visualize, it would be helpful to have some indication of bootstrap support.

-Some of the branches in the trees do not belong to any of the indicated phyla - where do they belong to?

-Regarding the cGLRs in Bivalvia, why do the authors believe there has been such drastic radiation? Is it possible that they have undergone neo-functionalization and play roles beyond immunity, such as in development? This is of course outside the scope of this study, but perhaps some speculation on this could be added in the discussion?

Figure 4

-While the authors describe a clear preference, it would be interesting to see if the least preferred molecule can still trigger a response. If so, could they speculate what the evolutionary advantage of diversifying in small molecule second messengers is? Perhaps specialization?

Figure S3

-Could the authors please specify the expected size of the bands in the Western Blots or highlight them in the figures?

Additional remarks/thoughts that arose will reading

-Would it be possible to bioinformatically predict the small molecules produced based on differences in the residues in the catalytic domain of cGLRs?

-While this is beyond the scope of this paper, we were wondering if it would be feasible to alter small-molecule binding specificities of the downstream S. pistillata STINGs by making structure-guided mutations in the ligand binding pocket of the STING receptors.

-Is there an evolutionary advantage for having a narrower specificity in downstream S. pistillata STINGs rather than just being able to equally activate both?

-Would it be possible to transfer cGLR/STING systems to plants and engineer it to recognize plant-pathogen derived ligands?

-Considering the prevalence of small-molecule signaling-based systems among metazoans, is it possible that pathogens have enzymes that can manipulate these signals? This is an emerging area of much interest in plant-microbe interactions.

-While this is outside of the scope of this study, the finding that some receptors didn’t respond to dsDNA or dsRNA is important and super interesting, as it suggests there may be new elicitors activating some members of this receptor family.

-Also outside the scope of this study, but we were curious about the extent to which the downstream response is functionally conserved across metazoans and bacteria. Perhaps this could explain why the auto-active cGLR proteins described here did not induce cell death in E. coli.

-Is it possible that the auto-active cGLRs described here do not require a ligand for activation, but are instead constitutively under negative regulation that is absent in the in vitro assays conducted?

Competing interests

The author declares that they have no competing interests.