Read on Twitter
Excited to see the project I’ve been working on out on @ScienceMagazine. Thanks to @v_hornung for great guidance, as well as @ScherrMatthias from @DuderstadtLab and @PichlmairL for a wonderful collaboration. Here is a little background information on how this project evolved:
1/n https://twitter.com/v_hornung/status/1332057371099738115
1/n https://twitter.com/v_hornung/status/1332057371099738115
Working on this project was an odyssey I enjoyed a lot. I remember when I started my PhD and got hooked on work from the @DBachovchin Lab. I presented their first Val-boro-Pro pyroptosis paper in journal club ..
2/n
2/n
.. and we quickly embarked on a discussion what the mode of action of this compound could be. I ordered VbP to see myself how it works and to my surprise it turned out to be a quite specific caspase-1 activator. Yet our first guess – NLRP3 activation – turned out to be wrong.
3/n
3/n
Naively thinking that this might be a “niche project”, I geared up to conduct a genome-wide CRISPR screen, to identify positive regulators of VbP-triggered pyroptosis.
But then things started happening very fast …
4/n
But then things started happening very fast …
4/n
The @DBachovchin Lab showed that Nlrp1 and CARD8 are activated upon VbP treatment, the @ReversadeLab found DPP8/9 to be associated with NLRP1 and then the @russellevance and @DBachovchin labs published their seminal ‘functional degradation’ model for Nlrp1 activation
5/n
5/n
So I decided to re-focus my work on human NLRP1, trying to identify a ‘physiological’ trigger beyond DPP8/9-inhibition.
In light of the fact that keratinocytes express high levels of NLRP1, I screened various viruses for potential NLRP1 activation ..
6/n
In light of the fact that keratinocytes express high levels of NLRP1, I screened various viruses for potential NLRP1 activation ..
6/n
.. et voilà, Semliki Forest virus showed a strong NLRP1-dependent inflammasome response.
SFV is a usual suspect when it comes to double-stranded RNA formation, so we went on to test whether dsRNA can do the trick on its own.
7/n
SFV is a usual suspect when it comes to double-stranded RNA formation, so we went on to test whether dsRNA can do the trick on its own.
7/n
And indeed, it could!
I still remember when I developed the ELISA of wildtype and NLRP1 KO cells stimulated with poly(I:C).
8/n
I still remember when I developed the ELISA of wildtype and NLRP1 KO cells stimulated with poly(I:C).
8/n
The next step was to find out whether NLRP1 indeed served as a direct sensor for dsRNA.
This turned out to be a rather tricky task.
We decided to take a biochemical approach, expressing and purifying various recombinant NLRP1 constructs.
9/n
This turned out to be a rather tricky task.
We decided to take a biochemical approach, expressing and purifying various recombinant NLRP1 constructs.
9/n
Doing so we found NLRP1 directly binds to dsRNA with high affinity. Of note, human NLRP1 has a large positively charged patch on its LRR, which we assume to be involved in nucleic acid binding.
However, these in vitro assays also showed that NLRP1 bound double stranded DNA?!
10/n
However, these in vitro assays also showed that NLRP1 bound double stranded DNA?!
10/n
So what is the difference between dsDNA and dsRNA?
When we studied ATP hydrolysis as a function of nucleic acid binding, we observed a stark difference between dsRNA and dsDNA: Only dsRNA triggered ATP hydrolysis, while dsDNA was largely inactive.
11/n
When we studied ATP hydrolysis as a function of nucleic acid binding, we observed a stark difference between dsRNA and dsDNA: Only dsRNA triggered ATP hydrolysis, while dsDNA was largely inactive.
11/n
Altogether, from these experiments we infer that dsRNA-bound NLRP1 undergoes a conformational switch that allows it to gain a signalling competent state. How this exactly translates into its C-terminal part forming an inflammasome is yet unclear.
So more to be done here!
12/n
So more to be done here!
12/n
Anyway super excited this paper is out and looking forward to more science about NLRP1 and friends.
13/n
13/n
By the way: Kudos to @KimSamirah, @ReversadeLab, and Franklin Zhong for identifying how human NLRP1 senses protease activity.
14/n
14/n
