News from the Sternberg lab at Columbia Univ., HHMI. Posts are from lab members and not Samuel Sternberg unless signed SHS. Posts represent personal views only.
New York, NY
- Tweets 128
- Following 23
- Followers 1,579
- Likes 10
Link to post with preprint
New preprint from the Sternberg lab!
biorxiv.org/content/10.64898…
We used pooled oligonucleotide library mutagenesis + high-throughput sequencing to systematically dissect the molecular determinants of IStron transposition, RNA-guided DNA cleavage, and self-splicing.
1
842
Link to paper!
academic.oup.com/nar/article…
1
2
18
1,132
Just published! IStrons are a unique class of transposable elements that pack three distinct functions in one sequence. Here, we used pooled library mutagenesis and high-throughput sequencing to systematically map the sequence and structural features that govern these functions.
1
10
46
4,101
Out now! In collaboration with Hiroshi Nishimasu's lab, we uncover how a dual reverse transcriptase immune system builds double-stranded DNA from both RNA and protein templates.
sciencedirect.com/science/ar…
Previous thread for the preprint:
1/9 New preprint from the Sternberg Lab in collaboration with the Nishimasu Lab! We uncover how the DRT3 antiphage immune system pairs two reverse transcriptases, one RNA-templated and one protein-templated, to build a double-stranded DNA effector. doi.org/10.64898/2026.05.04.…
1
24
119
15,006
11/11 Many thanks to all authors for their collective efforts in bringing this story to life! And a special thanks to @shsternberg and @IsraelF96135088 for their mentorship and support throughout this project.
2
286
10/11 This suggests a shared architectural logic underlying RNA-guided, tandem-repeat DNA synthesis across diverse bacterial Class 2 DRT systems & suggests that DRT10 is a structurally minimal version of the same strategy used by eukaryotic telomerase to achieve repeat addition.
1
1
314
9/11 And because the boundaries are geometric rather than sequence-specific, we could swap in templates from DRT3 and DRT9 and reprogram DRT10 to synthesize their signature repeats.
1
1
165
8/11 This anchored-boundary logic is structurally conserved across bacterial Class 2 DRT systems (DRT2, DRT3, DRT9), despite substantial divergence in protein sequence, ncRNA sequence, and oligomeric states.
1
1
165
7/11 How does DRT10 RT know where each repeat starts & ends? 2 RNA stem-loop anchors physically define the template boundaries. Geometry, not sequence, sets the template window, w/ microhomology on ncRNA sequence directing iterative template resetting between cycles of extension.
1
2
186
6/11 Symmetric structure, asymmetric function: both RT monomers are structurally identical and catalytically competent, but only one monomer produces the kilobase-length repeat DNA products, while the other generates short abortive products.
1
2
193
5/11 Two identical RT monomers bind opposite sides of a single, pseudo-symmetric, figure-eight-shaped ncRNA. The RNA is both the template and the scaffold holding the whole complex together.
1
1
4
362
4/11 How does DRT10 RT achieve such elegant tandem-repeat synthesis? We determined cryo-EM structures of two evolutionarily diverse DRT10 systems & found a surprise: the RT enzyme doesn't work alone...
1
1
241
3/11 Last year we reported that DRT10 RT repeatedly copies a short internal RNA template to generate long tandem-repeat DNA, which is conceptually similar to how telomerase extends chromosome ends. More details in our earlier preprint: biorxiv.org/content/10.1101/…
1
2
497
2/11 Reverse transcriptases (RTs) are typically thought of as molecular copy machines: they convert RNA template into DNA, with product length defined by template length. But DRT10 is different.
1
3
650
1/11 New preprint from the Sternberg lab in collaboration with the Fernández lab! We are excited to share our structure-function study of DRT10, a bacterial defense-associated reverse transcriptase that synthesizes long tandem-repeat DNA.🧵
More here: biorxiv.org/content/10.64898…
2
24
87
9,460
8/8 Want more? The preprint has it all-- subcomplex structures, mutagenesis validations & a clear path toward engineering next-gen gene insertion tools. Huge congrats to co-first authors Giada and Seraina, and the whole Jinek & Sternberg lab teams! 🎉 tinyurl.com/typeI-CAST
1
12
1,213
7/8 The full mechanistic model 🏛️: every step gated, every checkpoint structural. A complete blueprint for engineering better gene insertion tools. 🛠️
1
1
14
1,359
6/8 But hooks alone aren't enough. TnsB's beta-barrel domain must physically dock onto TnsC and grab the target DNA, triggering an allosteric relay to order disordered domains, remodel the catalytic site by ~11 Å, & switch TnsB on for DNA integration. No target = no chemistry.
1
2
5
874
5/8 This "super-ring" clamps the targeting complex and fixes the integration site 49 nt away. The transposase (TnsAB) arrives next, dangling by tiny C-terminal "hooks" 🪝 off the TnsC ring.
1
1
5
926