I mean, you don’t just use a tweezers, right?
So, how do you gene-sequence something that is undoubtably mixed with all kinds of other DNA?
What, no microbiologists here?
OK, since I’m bumping - I read about using micro filtration to separate viruses from other garbage, but - aren’t we full of viruses? How do you get the one you want isolated?
So… I think the challenge to your question is two-fold: 1. there are multiple different ways of isolating/identifying it 2. one doesn’t actually isolate it (until later in research in which case it is often not isolated but rather produced). This is not in my area of expertise so I’ll do it from a non-pharmacological knowledge into the pharma lab setting.
You can identify it through genetic analysis of sick people:
A. These people are all sick
B. Looks viral and perhaps related to known/previous coronavirus symptoms
C. broad sequencing of all genetic material (particularly non-human and perhaps searching more heavily for coronavirus like DNA)
D. Yep, looks like there are a lot of DNA/RNA related to coronavirus capsid proteins in sick patients but nothing substantial in healthy controls
E. Deeper sequencing and analysis of coronavirus DNA until we get teh full sequence and then we can classify it and start to understand it.
Or one can use antibodies (known from previous outbreaks) to similarly figure out and jump to C. This can be done in either direction: 1. sampling patient blood for proteins which stick to known antibodies or 2. giving known virus particles to the (recovering) patients and seeing if their antibodies recognize and bind.
After getting the full genetic sequence, DNA tests (like Abbott’s fast one that is launching imminently) as well as production of the virus (components) in a controlled setting can be done. From the genetics we can figure out strucutre, likely binding targets, etc.
We do this all the time in my lab.
(here’s the simplified version)
You homogenize the sample (or elute the sample off a swab etc…) in buffer, apply it in various dilutions to a monolayer of cultured cells of the appropriate type, and wait. The virus will infect the cells and start killing them. At the proper dilution you’ll have a small enough number of infectious viral particles in the sample such that instead of killing all the cells you’ll get a series of holes (or “plaques”) in the cell monolayer; each of these plaques represents an event where a single virus particle infected a cell, multiplied, burst out and infected the neighboring cells, ultimately making a hole in the cell layer.
You then swab the hole (which is full of virus) and apply that to a new cell culture. Bingo, you’ve isolated the virus (really a population of billions that all descended from a single viral particle. Then you sequence it to know what you actually isolated.
In practice, in the case of someone shedding COVID19, most of the viruses in the sample are probably COVID19. But you’d isolate and sequence many clones to make sure you got the one you were looking for.
In practice, there are a lot of complexities in this, but that’s the short version.
Thanks - that makes sense.
In this case, “cultured cells of the appropriate type” would have to be human cells, right? Where do the appropriate cell cultures come from? HeLa, embryonic cells, what?
I had understood that, with a few exceptions like HeLa, human cells were difficult to culture.
In my lab, we use cells from humans (cultured and primary), monkeys, hamsters, multiple mosquito species, fruit flies, and moths. “Appropriate” means a cell line the virus will replicate in. If we’re trying to isolate something new we might try a dozen different cell lines. They may not have to be human. Depends on the virus.
There are lots of human cell lines available. HeLa are definitely the most common, but I don’t like them, they tend to contaminate everything. There are tissue-specific cell lines, and we do a lot with primary cells from blood (which we get from blood banks) or other tissues (which we get from medical centers).
But we do more with Vero (monkey kidney) and BHK (hamster kidney) cells. And lots of insect cells of course.
As to where we get immortalized cells from, there are depositories that provide them. ATCC is one of the main ones.
Oh, I do remember that there are a lot of lines of cultured human cells. That doesn’t mean that it’s not difficult, though, especially if one wants a specific sort of tissue.
They’re not “easy” to culture. But it’s what we do for a living 
This was the part I was wondering about. Say you wanted to do x-ray crystallography. You’d have to produce enough virus, then you’d have to ??? purify/concentrate/isolate ?? enough of it to make a crystal out of?
How would you do that? How much would you need?
Um…Yes?
Couple ways, depending on the virus. You ca get a LOT of virus in a well-established culture system, enough to crystalize and get an x-ray structure.
More and more, people are using cryo-electron microscopy these days to make viral structures. Basically a special electron microscope that takes thousands of images of viral particles (produced in cell culture and specially prepared), then uses a computer to “average” all of the images to get very high resolution structures. Forget viruses, you can get structures of proteins this way.
For simple viruses you can just express the capsid protein and they’ll self-assemble into empty virus particles that you can get structures from.
Finally, if you have other similar viruses where a structure of the capsid in known, you can do theoretical prediction of the structure of your new virus protein by “mapping” the sequence onto the known structure. For viral capsid proteins, they need to be pretty similar but it can be done.
Scientists are starting to realize that you can visualize the structure of viruses by shooting them with powerful lasers (reminding them why they got into science in the first place.)
How would they have isolated the virus from the culture system?
And would it have required purification? How would they have done that?
How big a crystal was required?
(the basic simplified procedure)
You grow the virus in cell culture, lyse (or burst) the cells, and use a combination of differential filtration and gradient centrifugation to isolate pure virus particles.
If you’re trying to crystalize actual virus, yes it needs to be purified. See the isolation procedure above - properly done you can get extremely pure virus preperations.
See above.
No idea, i’m not a crystallographer. But i’ve seen crystals that were used for structural determination of various proteins, they’re not very big (fraction of a millimeter).
My guess is that size is secondary to crystal homogeneity anyway (i.e. it doesn’t have to be big, but it does have to be very regular - this will be easier to do for a small crystal) but this is a guess. In normal times i’d walk down the hall to ask a colleague, i’ll send an email and see what she says and report back