Read on Twitter
Ladies and germs, it’s time for the thread you didn’t realize you were waiting for: my ride-along narration of @aewatkins6 doing a SalivaDirect run! (1/??) https://twitter.com/nathangrubaugh/status/1295813821798535168
Some front matter: SalivaDirect method development has been led by @VogelsChantal, @awyllie13, @DougBrackney, & @NathanGrubaugh, with vital support from @robbysikka. @ChaneyKalinich did extensive early validation; NBA validations have been led by @XtinaHarden and @aewatkins6. 2/
All this to say: I’ve been around, I’ve helped out (as our most locally infamous tweeter, writing a walk-through thread is part of that), but I can’t claim much credit for the development of this method, and I don’t want this thread to be interpreted otherwise. 3/
With that out of the way: time to start walking through the process with our latest batch of samples from the NBA bubble! I’ll be tweeting fairly casually; our formal protocol is available here if you have technical questions 4/ https://dx.doi.org/10.17504/protocols.io.bjswknfe
The first experimental step is adding a small volume of Proteinase K, an enzyme that digests proteins, to a bunch of 8-strip tubes. This helps break down 1) proteins in saliva that make it goopy + 2) viral proteins that stick to viral RNA and might make it harder to detect. 5/
Next, raw saliva gets mixed by vortexing, and then added to the PK tube.
[cap: working in a biosafety cabinet, Annie uses a pipette to pull up liquid from a clear plastic tube & transfer it to a smaller tube, mixes by moving liquid up and down, & then disposes of the pipette tip]
[cap: working in a biosafety cabinet, Annie uses a pipette to pull up liquid from a clear plastic tube & transfer it to a smaller tube, mixes by moving liquid up and down, & then disposes of the pipette tip]
Once a set of samples is added, the rack gets mixed by vortexing for 30 seconds (will have to catch video on the next plate), and then it’s ready for the heat step!
At this point, we put samples into a machine (in our case, a PCR cycler) that can heat everything to 95°C for 5 minutes, which inactivates the PK (it’s already done its job, and it’d mess with later reactions otherwise)
(brb, I gotta set up the PCR plate)
The two distinguishing features of SalivaDirect have already come into play: first, we’re just using plain saliva, collected in sterile, nuclease-free lab tubes. You don’t need fancy collection devices or buffers, which cuts costs and makes materials easier to source.
And, as work from my first first-author preprint showed, viral RNA remains stable in standard plastic lab tubes for up to a week, even at room temperature! (had to fit the shameless self-promotion in here somewhere) https://twitter.com/awyllie13/status/1290736416406867974
Second: we replace the whole expensive, labor-intensive, time-consuming RNA extraction process (which I unpacked in June in a thread linked here) with a proteinase K vortex and a 5-minute heat step. https://twitter.com/isabelott/status/1271914078911574018?s=21
A quick revision, which I realized was necessary when I filmed Annie doing the vortex step and it lasted twice as long as I expected: the Proteinase K vortex step is a minute, not 30 seconds 
[cap: Annie pressing a plate full of tubes down on a vortex mixer, shaking the plate]
[cap: Annie pressing a plate full of tubes down on a vortex mixer, shaking the plate]
(we’re running two plates today, so timelines are getting a bit crossed, but I’ll do my best to catch up while this first PCR runs)
So, to catch up: while samples are in the PK-inactivating heat step, the simple PCR mix gets prepped and loaded into a plate, usually by the secondary person (today, that was me; action shot courtesy of @tdalpert)
Then, samples + the PCR mix plate get brought back to the biosafety cabinet, where a small volume of the “processed” spit is added to the PCR plate.
[cap: Annie uses a multi-channel pipette to mix and transfer processed sample from the tubes to a set of wells in a 96-well plate.]
[cap: Annie uses a multi-channel pipette to mix and transfer processed sample from the tubes to a set of wells in a 96-well plate.]
Next, the plate gets sealed, spun down, and loaded into the quantitative PCR cycler.
[cap: a PCR machine lid swings shut robotically. the front of the machine is decorated with an exasperated LeBron James meme cutout, gesturing at the control/readout screen.]
[cap: a PCR machine lid swings shut robotically. the front of the machine is decorated with an exasperated LeBron James meme cutout, gesturing at the control/readout screen.]
So, what happens in the PCR reaction? First, an enzyme called reverse transcriptase makes DNA copies of all the RNA (viral and otherwise) in the sample. Then, chunks of DNA that match up to our SARS-CoV-2-targeting "primer" DNA fragments get selectively replicated.
(got caught up dealing with other work, but I have the base footage and will return to wrap this up tomorrow - or later today, I guess?)
