Thread by @m_janelee

We have a new preprint out today on biorxiv examining the ability of NK cells to kill SARS-CoV-2-infected cells! tldr; they’re pretty terrible at it.
biorxiv.org/content/10.110…

A tweetorial: (1/11)

To gauge how well healthy NK could specifically kill SARS-CoV-2-infected cells, we infected A549-ACE2s with enough virus to get 50% of cells infected after 48 hours. Then, we performed flow-based killing assays using purified, healthy NK cells. (2/11)

I used SARS-CoV-2 with an mNeonGreen reporter, letting me identify infected cells vs bystander (uninfected) cells & assess the % killing of each group. The results were dramatic: 19 NK cell donors all killed the UNINFECTED cells better than the infected cells! (3/11)

There was no difference in the killing of mock-infected cells (unexposed to virus) vs bystander cells (exposed but uninfected), implying an escape mechanism intrinsic to infected cells. (4/11)

Next, we looked at expression of ligands for NK cell receptors on infected vs uninfected cells. And behold—a winner emerges! % of target cells expressing ligands for the activating receptor NKG2D (“NKG2D-L”) dropped from ~85% in mock cells to ~5% in infected cells. (5/11)

It takes 48 hours for NKG2D-L expression to really drop, so we compared NK cell killing of infected cells after 24 hours (~60% of targets NKG2D-L+) vs 48 hours (~5% NKG2D-L+). As expected, NK cells killed targets much better at 24 hours post-infection than 48 hours. (6/11)

Differences in NK killing of SARS-CoV-2 over time explain why other (very well-done!!) studies show that NKs CAN control SARS-CoV-2. Right after cells get infected, NKs actually kill them really well. Once NKG2D-L is downregulated? Not so much. (7/11)

Next, we looked at how the virus is downregulating NKG2D-L by looking at the impact of each viral protein individually on NKG2D-L expression. Nsp1, a key inhibitor a host translation, had by far the biggest effect. (8/11)

Finally, we tested whether Nsp1 alone was sufficient to confer resistance to NK-mediated killing. Holy crap, it was! Healthy NKs were WAY worse at killing Nsp1-transfected cells than those with GFP or another SARS protein (Nsp14). (9/11)

The big takeaways from this study:
1. It’s all in the timing! SARS-CoV-2 is really good at evading NKs UNLESS the NKs get there early enough
2. Nsp1 is a key driver of NK evasion. Drugs targeting Nsp1 may be even better than we thought bc they could rescue the NK response (10/11)

Check out the full preprint for the full story! Huge shoutout to @blish_lab and all of the amazing authors on this paper @arjunrustagi @aimee_beck888 @ZengLeiping @stanleyqilab @SherlockpHolmes