Can CRISPR edits enable precise tuning of plant gene expression? We think: yes.

In our newest manuscript, we measured the effects of >30,000 CRISPR-like promoter mutations in sorghum protoplasts.

Thank you! Excited to see how to field evolves in understanding cyclic nucleotide transport!

This was a collaborative effort between the Bondy-Denomy lab at UCSF and Gitai lab at Princeton. I deeply appreciate all the contributions from my co-authors Bruce Wang, Esther Shmidov, Harshu Musunuri. Stay tuned for more at SISB 2026!

Altogether, this study establishes that CBASS does not have to culminate in cell death, but instead, the CapV effector antagonizes late-stage phage replication while preserving cell viability. This broadens the field’s current perspective of CBASS, and we hope it motivates future studies!

While this mechanism is still under investigation, it opens up many exciting avenues of investigation for CBASS and several other anti-phage immune systems with membrane-acting effectors.

Leading hypothesis: CapV interferes with phage capsid assembly at the inner membrane of bacterial cells, limiting virion production and leaving phage DNA unpackaged and vulnerable to degradation.

Our leading hypothesis is that CapV interferes with phage capsid assembly at the inner membrane of bacterial cells, limiting virion production and leaving phage DNA unpackaged and vulnerable to degradation.

Single-step growth curve of PaMx41∆acb2 phage infecting Pa011 WT or ∆CBASS strains (n=3 +/- s.d.). PFU/ml were quantified at the indicated times after chloroform treatment.

In this and earlier studies, we also observed that phages fail to produce maximal levels of mature virions in cells with active CBASS, suggesting that CBASS stops phage at the late-stage of infection. How does CapV come into play?

DNA levels of PaMx41∆acb2 phage gp1 (early-stage gene, unknown protein), gp11 (middle-late stage; major capsid protein), and gp23 (late-late stage, unknown structural protein) during infection of Pa011 WT or ∆CBASS strains. Values were normalized to each respective Pa011 strains at 30 minutes post-infection.

How does CBASS stop phage infection? Or, in other words, at what stage of the replication cycle does the phage ‘fail’? We observed that phage transcription and DNA replication proceed normally, but in cells with active CBASS, phage DNA fails to reach maximal levels.

Top: Structural comparison of CapV phospholipase effectors from P. aeruginosa BWHPSA011 (NCBI protein ID: ERW743131) or V. cholerae El Tor N16961 (NCBI: WP_001133548.1). Protein structures predicted using AlphaFold3. Bottom: Matrix representing the order of magnitude change in phage titer on bacterial strains expressing full or partial Type II-A CBASS operons, and their relative expression levels of CBASS or synthase genes from Pseudomonas aeruginosa (Pa) or Vibrio cholerae (Vc). Titer changes were quantified by comparing the number of spots (with plaques, or clearings if plaques were not visible).

Upon constitutive activation of CapV, cells grow well and block the replication of many, but not all, phages. We tested both P. aeruginosa and V. cholerae CapV, and further show that, revealing that CapV possess a degree of phage specificity that has been previously overlooked.

OD600 measurements of Pa011 strains with catalytically dead CdnA or CapV, which are overexpressing CdnA or EV plasmids infected with an increasing multiplicity of infection (MOI) of PaMx41∆acb2 phage over 720 minutes (0.1% arabinose, n=1 biological replicate, n=3 technical replicates +/- s.d.).

Mutating the phospholipase catalytic residues of CapV leads to cell lysis due to phage replication, as expected, and demonstrates that CapV is required to block phage and maintain cell viability.

So far, the main take home is that high, constitutive 3’,3’-cGAMP signaling is safe for cells with CapV effectors at endogenous expression levels.

Left: OD600 measurements of Pa011 WT Type II-A CBASS, CdnAD87A/D89A (dCdnA), and CapVS48A (dCapV) strains infected with or without PaMx41∆acb2 phage (MOI:2) and with or without 3’,3’-cGAMP (~150 µM) over 600 minutes (n=3 +/- s.d). Right: Model of putative cyclic nucleotide transporters in gram-negative bacteria. 

Surprisingly, we observed that spiking 3’,3’-cGAMP into the media can also induce CBASS anti-phage activity and support cell growth during phage infection. How 3’,3’-cGAMP – or other diverse CBASS signals – are imported (and potentially exported) remain an exciting area of future study.

OD600 measurements of Pa011 strains expressing CdnA or empty vector (EV) plasmids infected with an increasing multiplicity of infection (MOI) of PaMx41∆acb2 phage over 720 minutes (0.05% arabinose, n=3 +/- s.d.).

Using Type II-A CBASS in P. aeruginosa, we showed that the endogenous activity of a cGAS homolog (CdnA) stops phage infection while maintaining cell viability. High, constitutive cGAS activity (3’,3’-cGAMP signaling), with endogenous CapV levels, can resist high phage doses and cells keep growing!

CBASS is a form of anti-phage immunity that generates cyclic nucleotide signals that bind to and activate downstream ‘effector’ proteins to stop infection. In our study, we focused on a membrane-acting effector – a phospholipase (CapV) – that has been shown to induce cell death upon overexpression.

Cell death is a fundamental mechanism of antiviral immunity across diverse organisms, including bacteria. As my final PhD project with @jbdsf.bsky.social, I was curious whether cell death is required for successful immunity with the ancient cGAS pathway known as ‘CBASS.’

Spoiler – the answer is no!

How can we understand the earliest events in evolution of eukaryotic immunity? @yao-li.bsky.social reports incredible molecular fossils of complete bacterial-like operons in eukaryotes that illuminate how animal immunity was first acquired from anti-phage defense

www.biorxiv.org/content/10.1...

Harnessing synthetic biology to engineer resilient crops

Really excited to join Stanford ChemE this September! We'll engineer proteins in plant immune and hormone pathways, combining high-throughput directed evolution with ML. Our goal is to enhance climate stress resilience and disease resistance in crops: rim.stanford.edu

Must read for all who use Nicotiana benthamiana as a platform: “Causes and consequences of experimental variation in Nicotiana benthamiana
transient expression”
#plantscience

www.biorxiv.org/content/10.1...

END nucleases: Antiphage defense systems targeting multiple hypermodified phage genomes Prokaryotes carry clusters of phage defense systems in defense islands that have been extensively exploited bioinformatically and experimentally for discovery of immune functions. However, little effo...

We've identified a group of nucleases that can target multiple phages with hypermodified DNA, using a sensor domain present across many domains of life. We’ve even used this as a tool to find phages with hypermodified DNA! Really thankful to all the co-authors who have been such great help!

A minimal CRISPR polymerase produces decoy cyclic nucleotides to detect phage anti-defense proteins Bacteria use antiphage systems to combat phages, their ubiquitous competitors, and evolve new defenses through repeated reshuffling of basic functional units into novel reformulations. A common theme ...

Excited to share that my work from the @aaronwhiteley.bsky.social lab is now on bioRxiv! With @benmorehouse.bsky.social, we discovered that the Panoptes defense system—named after the all-seeing watchman of Hera—uses decoy nucleotides to detect phage anti-defense proteins.

Systematic Discovery and Design of Synthetic Immune Receptors in Plants Plants deploy a diverse array of pattern recognition receptors (PRRs) that perceive microbe-associated molecular patterns (MAMPs) to activate immune responses. Leucine-rich repeat receptor-like kinase...

First post on Bluesky!! ☀️

Very happy to share that my main postdoc work ‘Systematic Discovery and Design of Synthetic Immune Receptors in Plants’ is finally out on Biorxiv!

www.biorxiv.org/content/10.1...

The full story is summarized here:

This is excellent

In our latest review, we explore 12 deep-learning tools for metagenomic analysis, covering their strengths, limitations, and key applications. We hope it serves as both a resource and inspiration for new ways to analyze metagenomic data. Great work by Eli Levy Karin!
📄 doi.org/10.1093/nsr/...

New version of DefenseFinder available defensefinder.mdmlab.fr (& cli)

Now encompassing 263 systems (+111 this year..!).

Thanks @ftesson.bsky.social, DefenseFinder grandmaster. Full list of systems here: defense-finder-models/List_system_article.md at master · mdmparis/defense-finder-models

Multi-interface licensing of protein import into a phage nucleus - Nature This study uncovers a highly conserved jumbo phage protein, Imp1, that possesses multiple interfaces to license protein import into a proteinaceous nucleus-like compartment, using a genetic selection that forces the phage to decrease or abolish the import of specific proteins.

Nature research paper: Multi-interface licensing of protein import into a phage nucleus

https://go.nature.com/3QdP5SJ

Design and regulation of engineered bacteria for environmental release - Nature Microbiology Chemla et al. review the release of bacteria into the environment and propose engineering strategies to help improve performance and reduce risk.

REVIEW: Design and regulation of engineered bacteria for environmental release

by Chris Voigt & co

#microsky

www.nature.com/articles/s41...

Combining Directed Evolution with Machine Learning Enables Accurate Genotype-to-Phenotype Predictions Linking sequence variation to phenotypic effects is critical for efficient exploitation of large genomic datasets. Here we present a novel approach combining directed evolution with protein language m...

I'm very excited to share my first graduate research project! Check out our preprint which describes how data generated with directed evolution can be used to model the phenotypic effects of naturally-evolved sequence variants. (1/11) www.biorxiv.org/content/10.1...

I'm thrilled to announce our latest work is now published in Cell! Viruses encode numerous proteins that inhibit host defenses, but identifying immune-modulatory proteins among millions of viral sequences has been nearly impossible - until now! www.cell.com/cell/fulltex...