How do the ancestors of CRISPR-Cas unwind DNA and how can this lead to better genome editing? With our collaboration between @doudna-lab.bsky.social x @jacobsenucla.bsky.social x Zev Bryant's lab x @savagecatsonly.bsky.social we've uncovered the secrets behind TnpB's dynamics!
Together, these findings showcase the potential of engineered TnpB editors delivered via viral vectors for efficient, heritable genome editing in plants 🧬.
(7/7)
They demonstrated enhanced activity of two engineered variants, eTnpBc (TnpB-KYLI) and eTnpBe (TnpB-VGIRL), across multiple plant species (N. benthamiana, rice🌾, and pepper🌶️).(read more: doi.org/10.1101/2025...)
(6/7)
These findings build on prior work by @brittneywthornton.bsky.social and @rfw.bsky.social, who engineered highly active TnpB variants by conducting and leveraging deep mutational scans of the TnpB ribonucleoprotein. (5/7)
These edits are transmitted to the germline without retaining the eTnpBc transgene – enabling a highly efficient path to heritable, transgene-free editing in plants in a single generation 🧬🌱 4/7) Read more: doi.org/10.64898/202...
In a new preprint, Nagalakshmi and Rodriguez demonstrate that an engineered, compact RNA-guided nuclease, eTnpBc (TnpB-KYLI), can be efficiently delivered via tobacco rattle virus (TRV) to achieve heritable editing efficiencies of up to 90% in Nicotiana benthamiana. (3/7)
Efficient delivery of genome editing reagents to plants remains a major challenge. Although viral vectors offer a promising solution, their limited cargo capacity has constrained their impact. (2/7)
New from Ugrappa Nagalakshmi and Jorge Rodriguez 🌱✂️– Gene editing in plants just got a lot easier! Using engineered TnpB, they demonstrate high-efficiency, heritable, transgene-free editing in plants. (1/7)
Congrats to our savage lab undergrads on their research and presentations in our inaugural undergrad research symposium 🎉🥳
‼️ New pre-print from co-leads @owentuck.bsky.social and Jason Hu! Check out this fascinating example of how coevolution enables defense system innovation.
We love generating data! But what should we do with it? We collaborated w/ the Listgarten lab on ProteinGuide - a new method for using experimental data to guide protein generative models. Congrats to @junhaobearxiong.bsky.social, @hnisonoff.bsky.social, @marialukarska.bsky.social, Ishan, & Luke! ⬇️
Excited to share our new preprint on mutational scanning of the TnpB RNP! We identify several highly activating, single-position mutations in TnpB’s protein and reRNA. These datasets enabled us to engineer enhanced TnpB protein variants (>50x plant editing levels compared to WT).
It is finally out!
Muntathar Al-Shimary, @doudna-lab.bsky.social , @cresslab.bsky.social and I present a gene knockdown tool that works in diverse bacteriophages: CRISPRi through antisense RNA Targeting (CRISPRi-ART)!
Check it out @naturemicrobiol.bsky.social
www.nature.com/articles/s41...
Congrats to the whole team!! (Whole team not pictured)
(9/9) In addition to laying the groundwork for engineering, the data could provide valuable insights into TnpB evolution and function in transposable elements. If you want to read more check out the pre-print here: www.biorxiv.org/content/10.1...
(8/9) They then chose 33 enriched muts across 19 AA positions & made & assayed a combinatorial protein library. 6 enriched variants were followed up on in HEK293T cells & N. benthamiana editing. All variants increased editing, with a max of 50-fold increase over WT in benthi!
(7/9) Enrichment varied by domain - many muts in RuvC and WED were enriched, most ZnF muts were depleted, muts in the unstructured C-terminal tail were ~neutral. Additionally, positive AAs were enriched w/i the vicinity of nucleic acids, particularly near the heteroduplex
(6/9) For the protein library, the authors made all single AA mutations and all stop codons. Surprisingly, ~20% of all mutations were enriched over WT!
(5/9) The hinge forms a sharp bend in the reRNA in stem 2, which is hypothesized to act as a regulatory switch for TnpB. Multiple groups have shown that stem 2 truncations increase TnpB editing, and our data suggests that hinge mutations may activate TnpB by a similar mechanism
(4/9) Starting with the reRNA– the library contained all single substitutions, 1-2 nt deletions, and larger secondary structure truncations. Many mutations were enriched over WT. Particularly, some (especially 1-2 nt dels) in the “hinge” region were highly enriched
(3/9) In the native genomic context, TnpB’s RNA scaffold (reRNA) is encoded in the protein’s mRNA. To map the mutational landscapes of the RNP components separately, they assayed two DMS libraries for the protein and RNA in an in vivo selection for endonuclease activity in yeast
(2/9) TnpB is an RNA-guided endonuclease found in prokaryotic transposons. Given its compact size, complex scaffold RNA, and putative ancestral relationship to Cas12s, they thought the TnpB ribonucleoprotein would be an ideal model for DMS
What's better than 1 deep mutational scanning (DMS) library? 2! In a new pre-print @brittneywthornton.bsky.social and @rfw.bsky.social et al. map the mutational landscape of ISDra2 TnpB protein and reRNA and leverage these datasets to engineer highly active variants (1/9)
Photo of Noam Prywes standing in from of bamboo plants
New research from IGI Investigator Dave Savage @savagecatsonly.bsky.social and first author Noam Prywes @prywes.bsky.social (pictured) details the landscape of possible Rubisco variants, suggesting ways to make the world's most abundant enzyme better at its job. 🌱🧪🧬
Read more: ow.ly/t2JO50ULh10
Check out the summary and "behind the paper" here doi.org/10.1038/d415...
Small changes in sequence can have big consequences in function, and not only negatively! We hope that this data is useful to other enzyme engineers and that this result expands our view of what is possible with even small changes. (7/7)
Several single AA mutations increased the CO2 affinity! Given sequence divergence and different oligomeric states of Form I & II, we were surprised to find single muts led to CO2 affinity outside of the range of Form II & at the edge of the distribution of plants and algae (6/7)
Rubisco is, after all, an enzyme, so we wanted to biochemically characterize each mutation. By repeating the assay over CO2 titration, key biochemical parameters (e.g. Vmax, Kc) are estimated for each mutant! (5/7)
This assay neatly identified known structural, functional, and evolutionary effects, confirming the robustness of the screen. For example, key residues in Loop 6 and the active site, are mutation-intolerant (4/7)
To address this, @prywes.bsky.social et. al developed an assay to estimate the fitness of 99% of single mutations in a high-throughput E. coli screen. E. coli growth rate is linked to enzyme behavior! (3/7)
