keen on reverse transcription | also keen on spliceosomes | cryoEM dabbler Assistant Member @MSKCC Post-doc @MIT PhD @MRC_LMB BSc @Otago
Manhattan, NY
- Tweets 158
- Following 188
- Followers 3,001
- Likes 280
My favourite discovery ever has just come online. Can I please tell you about some seriously wacky molecular biology? The story starts with a reverse transcriptase that SOMEHOW defends bacteria from viruses.
(👇 I recommend sound ON for the video 🎹)
1/
101
1,214
4,790
751,771
How do LINE family retrotransposons jump around?
Can we reprogram retrotransposons?
What does this have to do with Rum and Raisin?
Answers here and below!
32
367
1,431
497,519
1/5
My first postdoc paper! We found a new viral defence system in bacteria that looks suspiciously similar to inflammasome and apoptosome proteins from fancy eukaryotes – and indeed it acts as a true pattern-recognition receptor to detect phage proteins
science.org/doi/10.1126/scie…
23
233
1,339
Have you seen all these recent spliceosome #cryoEM structures and just wished you could see the bigger picture beyond all the snRNPs and RNAs and a million protein names? Check out our review! annualreviews.org/doi/abs/10…
Minimum detail, many pretty pictures, and LOTS of MOVIES
10
179
476
The Wilkinson Lab is open for science! @MSKCancerCenter
🧬We'll be finding funky new RNA biology, mainly by looking at reverse transcriptases (i.e. the Best Enzymes In The World)🧬
11
44
367
34,755
Lots more in the paper (#cryoEM etc!)
doi.org/10.1126/science.adq3…
Big thanks to coauthors @_David_Li @algao8 Rhiannon Macrae and @zhangf
Please also check out @shsternberg @stephentang23 related work - we come up with similar models, but it'll be exciting to resolve the differences!
15
31
309
20,988
(Incidentally, we first named this activity DARCARTs – for Defence-Associated Rolling Circle Amplification by Reverse Transcription – because of the similarity to RCA. Sadly, the journals/reviewers/editors/PIs cut this name but NO ONE can stop me using it here :P)
3/
6
13
278
20,689
What does this gene do? It produces a repetitive protein that is super toxic 💀. In fact, the more repetitive it is, the more toxic it is 💀💀💀. It stops bacterial growth, and this probably stops the virus from replicating. The lone cell dies, but the pack survives.
5/
9
14
249
17,387
I know it's old news in #cryoEM these days, but it's my first time seeing it and it's kind of really really exciting 🤓
5
3
214
covercovercovercovercovercovercovercovercovercovercover this is terrifically exciting
(Also please dw, the central dogma is still very much intact, this is just a bit of a funky spin on it)
The central dogma of molecular biology states that genetic information flows from DNA and RNA to protein, with reverse transcription converting RNA to DNA.
In the pursuit of understanding how bacteria defend themselves from viral infection, two groups have found alternative pathways to making genes from RNA that did not previously encode proteins. Learn more this week in Science: bit.ly/3ZO7tI1
3
27
216
20,995
Why make repetitive DNA? It turns out the juxtaposed repeats reconstitute a perfect promoter and continuous open reading frame (a mildly mind-blowing thing to realise). In other words: the bacterium is physically MAKING A GENE in response to viral infection.
4/
2
18
211
20,169
Fellow RNA geeks might see parallels to two eukaryotic systems. 1) Splicing – gene construction from non-contiguous RNA segments. 2) Telomerase – repetitive DNA synthesis from an RNA template. All use, or evolved from, reverse transcriptases. RTs rule!
6/
1
10
193
17,505
The system consists of a reverse transcriptase and a beautifully pseudoknotted non-coding RNA. We found that during viral infection, the reverse transcriptase goes rampant and makes REPETITIVE DNA, templated by a middle bit of the non-coding RNA.
2/
2
21
181
22,481
Apparently all I want for christmas is phage in my E coli, but at least they look realllly pretty. Any #bacteriophage people recognise these guys? (@PeterFineran?)
13
10
161
(P.S. eternal thanks as ever to @bradyajohnston for his #MolecularNodes plugin to #Blender3D. It makes molecular animation so much fun)
3
2
138
17,949
the inevitable forgotten link to the actual paper!
science.org/doi/10.1126/scie…
5
25
129
9,728
The spliceosome is pretty huge and likes to completely bury its intron substrates. So how does the pre-mRNA get in there in the first place? Excited to finally share our answer! (And for the cryoEM fans: how to solve structures from <1% of particles) science.sciencemag.org/conte…
4
38
108
(Just kidding, one more movie)
Thank you my lovely co-authors @FrangiehChris @zhangf and Rhi Macrae
Also @bradyajohnston's #MolecularNodes Blender plugin is great. Everyone should use it. #b3d
6
6
107
13,746
Happy World Phage Day!
A pretty #cryoEM structure:
#WorldPhageDay
7
96
Super excited to finally share this! With @Seb_Fica and @WojtekGalej we found a new spliceosome intermediate that explains the equilibrium between the two steps of catalysis - branching and exon ligation. 1/n for more movies and #cryoEM!
biorxiv.org/content/10.1101/…
5
15
85
and with 78 asymmetric units per particle… you can get pretty nice Vault reconstructions from the #amyloid micrographs 😉
3
4
77
This is such a lovely piece by Ben and Chris about Kiyoshi
nature.com/articles/s41594-0…
1
25
61
#Phage update! Had spare afternoon on a Glacios so calculated cryoEM map of tail, to dock in the *presumably* known structure and ID the phage. But surprisingly there are few phage tail structures and none fitted. But the sequence of the T1 phage tail fits the map quite nicely!
4
2
59
We had a bunch of questions about R2, so solved its structure #CryoEM
Q: How does R2 find its target?
A: We found two sequence motifs, we call them the RUM and RASIN 🥃+🍇
RUM is bound by accessory DNA binding domains on R2.
2
6
62
7,724
cries in miRNA/snoRNA/spliceosome... 😭
Anyway, it's a eukaryotic CRISPR* Congrats to authors!
*first confirmed RNA-guided ✨DNA nuclease✨ in eukaryotes
In a new study out today, @zhangf has uncovered the first programmable RNA-guided system in eukaryotes.
mcgovern.mit.edu/2023/06/28/…
@ScienceMIT @MITEngineering @MITdeptofBE @broadinstitute @HHMINEWS @MITEngineering
1
6
57
11,221
Calling on #cryoEM twitter and all protein detectives! Anyone recognise this protein? It comes from yeast, is 120 Angstrom diameter, and seems to have 3-fold symmetry.
9
4
56
Nils's incredible work is finally out, amazing work on a multi-talented protein by a multi-talented postdoc! It was a joy to use Dari @kimanius 's new Blush algorithm in #RELION to solve the cryoEM structure of this 40 kDa RNA complex
👋🏻 I'm absolutely glee-ridden to finally see our study on the RNA- and DNA-binding anti-CRISPR repressor Aca2 published in @Nature. 🥳 This unassuming little helix–turn–helix protein packs quite the punch! 🧶👇🏻
rdcu.be/dNl45
4
52
10,118
A #CRISPR that cuts a sigma factor inhibitor to unleash a transcriptional response?? Wild stuff from @thestrecker, Esra, @_David_Li et al. Such a joy to watch this story come together, and to help a little with some beautifully presented #cryoEM structures science.org/doi/10.1126/scie…
7
49
Our paper found such a pretty home in @MolecularCell !
My prev tweetorial has highlights, but check out the final version for awesome new experiment by @Seb_Fica: with power of K+ 🍌 he made the spliceosome go forward... then back... then forward again!
cell.com/molecular-cell/full…
1
5
41
When R2 binds the RUM, it unzips the RASIN, then cuts one strand of the RASIN.
The cut strand is used as a primer for reverse transcription of the R2 RNA - so this is called TPRT (target-primed reverse transcription)
1
4
35
6,702
Now it gets really fun: based on the structure, we can devise the ✨R2-tag✨ - 43 nt that you can attach to any RNA, and R2 will integrate it into the ribosomal DNA.
(sorry, I ran out of movies)
4
4
35
6,613
4/5
In a particularly devious manner, one of these receptors even recognises the ATP ligand bound by the phage terminase protein – good luck to the phage mutating that ATP to escape detection!
1
1
36
3/5
I had so much fun doing the cryoEM part of this project – the receptors form beautiful flowers 🍀, and can bind phage proteins with as little as 5% sequence identity through recognising folds, active site residues, and (my favourite) even ligands!
1
1
32
ALSO R2 doesn't have to use the ribosomal DNA target, it can be retargeted: fused to Cas9, it can instead reverse transcribe at the Cas9 nick site. Maybe this could one day be really useful for gene therapy.
2
1
34
6,931
Super exciting to share my PhD work - ever wondered how the cell knows where an intron stops? Read on to find out...
science.sciencemag.org/conte…
1
11
30
R2 is an exquisite LINE-like retrotransposon that inserts in ribosomal DNA repeats. Thomas Eickbush et al have done so much awesome work, showing that the R2 protein is INSANE: by itself, it finds a DNA target, cuts it, and reverse transcribes its own RNA right there and then.
1
1
32
7,886
R2 is really specific for reverse transcribing its own RNA (i.e. its 5'UTR, ORF, 3'UTR)
Q: What makes it so specific?
A: A surprisingly small part of the 3'UTR: two hairpins and a linker. Only 40 nt!
1
2
31
6,259
5/5
This was an awesome collaboration with the hyper-talented Linyi Alex Gao and Jonathan Strecker @thestrecker, and for me, such a cool first project with @zhangf lab @broadinstitute @mit_nano
1
31
2/5
Combining lots of experiments: bioinformatics, genetics, biochemistry, and of course, #cryoEM, we found that phage terminase and portal proteins bind to the receptor, trigger tetramerisation, and activate an effector domain like a nuclease to trigger cell death.
1
1
27
Blender! It has awesome rigid body simulations, I just made a giant ramp for ~100 T4 phage to slide down. Then some geometry nodes and shape keys for the bacterium and some sunset-style shaders
2
26
Thanks so much @RNASociety, and congratulations to the other winners!
Congratulations to our 2019 Scaring Award winners, Xuebing Wu (@wu_xuebing), Michael Chen, and Max Wilkinson!!! 🥳🥳🥳
2
1
19
Mystery solved thanks to @DongchunNi! It is "NMNAT", this crystal structure projection looks pretty convincing.
Twitter is wonderful 😀
1
3
16
I'd recommend watching all the movies here mrc-lmb.cam.ac.uk/groups/nag…
There's some extra explanation in non-review-speak and they're nice and big and easy to see
Many thanks to our very own Chris Norman for narrating these movies and sparing everyone my kiwi accent
1
8
19
PS if someone in the Boston area spots a spare printed copy in their department tearoom could they please steal it for me 😊😊🙏🙏 (steal one for yourself first of course)
1
18
2,561
Chris was an inspiring scientist, and such a hilarious and kind friend to everyone. He always had time to help in the lab, or to talk about his zany music tastes and geek out about silly hypothetical situations at teatime. We'll miss him a lot
1
13
As well as giving *amazing* geometry scores, @CrollTristan's ISOLDE provides a very rare chance to blend structural biology geekiness with classical music geekiness 🤓
2
2
18
annnnd: I'm hiring - come join! Especially postdocs and PhD students - please get in touch (NYC is great)
wilkinsonlab.bio
21
2,178
omg
Did you enjoy @crispr2023 or want to come to the next one! The #crispr conference will be in @Christchurch_NZ organised by @PeterFineran, @dsashital @gavinjknott and @drbridgetwatson. Check out the website crispr2025.org
17
5,401
Thinking about making a Twitter bot that periodically reminds people that Vaults exist and no one knows why
Then maybe, someday, someone will figure out what they do!
1
14
oh no now I have to work out how to install cryoSPARC again
Continuous flexibility is the next frontier for #cryoEM.
We are thrilled to introduce 3D Flexible Refinement, a new deep generative model that solves both detailed non-rigid motion, and high-res structure of flexible proteins!!
biorxiv.org/content/10.1101/…
cryosparc.com/3dflex
1/
1
9
very gratifying to see the importance of this part of U4 snRNA, but also kinda sad to learn that exome sequencing has historically excluded such essential small RNAs by default
But what is this RNA? RNU4-2 encodes the U4 snRNA component of the major spliceosome. The 18 bp region maps to two critical structures thought to control the positioning of the U6 ACAGAGA sequence (image credit to @Seb_Fica - SNV positions in red, insertions as arrows) 7/9
1
1
12
4,520
this ~200kDa complex is on Quantifoil 2nm carbon and gives 3.1A (once you chuck away the damaged 95% of particles) science.org/doi/10.1126/scie…
10
675
this is so so insane 😍
Hot from the press! We (@plaschka_lab @bpachecofiallos) dissected how the transcription-export complex recognizes mRNPs (=mRNA and associated proteins) in the nucleus, and we made a few surprising observations: here is the paper rdcu.be/c9e0f and here the tweetorial👇
7
2,735
go R2!! 🐦⬛🐦⬛
this is a very lovely study 😍
Bird bird bird, bird is the word!
Is the secret to safe gene addition a retroelement in birds?
We repurposed the avian R2 element to insert transgenes into human cells (and more!) using RNA-only delivery 🪶
a thread 🧵
nature.com/articles/s41587-0…
7
2,600
This paper was in the works for ages, and I'm really thankful that Kiyoshi Nagai got to read an early draft and was also excited by the findings. It only took a global pandemic to make us actually write the paper! 7/7
1
8
Over the years we had collected lots of cryoEM datasets of the branching spliceosome (C-complex), which kept appearing in all our preps! I gathered all these datasets, producing a lovely highres #cryoEM structure and allowing lots of focused refinements. 2/n
3
2
9
had an ambiguous density that could be two possible proteins based on prior knowledge. One by eye was probably a better fit. The HMMs generated by ModelAngelo in no-seq mode gave dramatically different E-values for each protein and supported the "by eye" assessment 👌
1
8
3,389
(although not perfectly! In a couple of places the T1 sequence does not fit, like Tyr117. So this must be a T1-related tunavirus 🐟, not T1 itself. BLAST tells me e.g. a phage with the fantastic name Rogue1 would fit the map a little better.)
3
8
and of course thanks @Seb_Fica for the best spliceosome cake in the world
9
everyone go join this lab please
Excited to announce the opening of the Strecker lab at MGH Molecular Biology/CCIB and HMS Genetics. I am tremendously grateful to all of my labmates who have helped me over the years, my past mentors @durocher1 and @zhangf, and my new colleagues @MGHMolBio for the opportunity!
8
3,638
structural biology twitter! I have a fun case where the N-formyl group of an initiator Met makes very pretty hydrogen bonds with another chain. Has anyone seen similar structural roles for N-formylmethionine? #cryoEM
1
7
thanks Wojtek! It's what happens when a reverse transcriptase gets jealous of its cousin Prp8 having all the fun with the RNA
3
425
These are great! In case they're helpful for finishing this off, I have some spliceosome PyMOL scripts for all yeast and human structures. You can get them here github.com/maxewilkinson/Spl…
2
5
@rnaErin We've been working on it! Here's a little sneak preview showing the transition from Bact to C complexes, the final videos will be on the Nagai lab website along with various PyMOL sessions that help exploring spliceosome structures.
1
1
4
(if you happen to be working on the structure of this protein, I'm very sorry, and I can send you the particles)
2
4
Hey #RNA2019 twitterers, I've mislaid my abstract book 😢, it's got my name inside the cover and on the edge.
Don't suppose anyone has picked it up? #desperateplea
1
2
5
Wowwwww this is so cool! Congrats @SuyangZhang2
Our RNA polymerase II-U1 snRNP structure elucidates co-transcriptional splicing.
It suggest a nascent RNA loop on the Pol II surface enables efficient splice site scanning.
Fabulous work by Suyang Zhang in collaboration with the Lührmann lab.
Manuscript: biorxiv.org/content/10.1101/…
1
5
😍 so cool!
How do CRISPR immune systems make molecular memories? Joy, Petr and I characterized an all-in-one trimmer-integrase to ask questions about CRISPR adaptation. Really excited to share this one - my first preprint @doudna_lab @igisci. Check it out! biorxiv.org/content/10.1101/…
4
1,591
tl;dr - you thought splicing was complicated? It's actually more complicated! Factors required for step two bind the step one spliceosome. Factors required for step one bind the step two spliceosome. 6/n
1
4
Finally clear how spacers are inserted into CRISPR loci - amazingly beautiful structures! science.sciencemag.org/conte…
1
4
Thank you! I should say, all the raw movies (compressed to TIFF!), the polished particles, and a dozen star files for all the EMDB maps will be on EMPIAR soon
1
4
really good question! At the moment, unclear. We haven't figured out how the protein is toxic yet
3
89
thank you and wonderful work! It's great we can both (hopefully 🤞) get our stories out
*returns to frantic manuscript writing*
3
158
(also Flotillin proteins in lipid rafts are basically just mini-Vaults. If someone can explain that, I'd be very grateful)
2
This increased the detail in the spliceosome periphery, revealing a surprise guest: the exon-ligation factors! These should only bind and stabilise the exon-ligation conformation of the spliceosome. That's their job after all. Why are they binding the branching spliceosome? 3/n
1
2
(also having second thoughts about calling that region "quasi-pseudoknot", although it seemed like a good idea at the time)
1
2
1,544