Excluding one of my biology professors, everyone that I’ve tried to explain this experiment to seems to think that I’m insane...
So I’m determined to prove that I can explain it properly.
Anyhow, here’s the lowdown, which I partially explained in the last post. My preceptor screwed up part of the experiment. It wasn’t a “live-action” screw-up, like accidentally leaving an irreplaceable volume of virus-laden fluid out in the open for the lizard people to enjoy (they work in a separate building, but they do come over sometimes).
Not to beat up on my preceptor more than I already have, but it was due to an oversight on his part that the latter half of the experiment failed. I get that hindsight is 20/20, as they say, but still, you’d think that what a junior scientist would be able to recognize post scriptum, a much more “seasoned” scientist ought to be able to foresee. Am I just that good of a scientist? Doubtful. The fact remains though, the experiment failed, and my preceptor was not able to foresee said outcome.
But let’s talk about what worked first, so you’ll have a better chance of guessing what went wrong later. There were two experiments in our project, the first one worked splendidly, and the second one failed outright.
To Recap
The objective of our project was to test the new “viral stamping” technique of infecting individual cells. The technique involved coating the tips of the hyper-sharp pipettes (made specifically for this purpose) with a special polymer, then dipping the tips of the treated pipettes into a high-concentration solution of viruses. The polymer allowed the viruses to stick to the pipette, they wouldn’t have stuck reliably otherwise. These pipettes would be maneuvered via a mechanical arm into contact with individual cells, infecting them.
[source]
My preceptor didn’t come up with this technique, mind you. Instead, we were working off of the protocol provided by a previously published scientific paper.
Experiment One: The Test Drive
Our first experiment was really as simple as can be.
Our objective: use the virus stamping technique to infect a single Mouse Embryonic Fibroblast cell.
We’d be using my old friend: modified HIV. The HIV virions would be carrying a gene that hopefully is familiar to some of you, that is, Green Fluorescent Protein or GFP. For those who don’t recall or remember what it does, it’s just a gene that encodes for a protein which glows green when exposed to UV light. Anyhow, we’d be able to tell which cells we’d successfully infected (with some caveats), by seeing which cells glowed green.
As the cells only could have gotten the GFP from the modified HIV, a green cell must have been one that was infected successfully.
Even if you had successfully infected a cell though, you wouldn’t immediately be able to see green. GFP must be present in sufficient quantities in a cell for it to visibly glow green upon being exposed to the appropriate frequencies of light. And it takes time for a cell to produce that quantity, which is why we waited 48 hours post-infection to check cells for fluorescence. Don’t want to fool yourself into thinking there were no successful infections when in reality there were, and the evidence of those infections was simply still forthcoming.
“Aha!” some of you might be thinking. “You said the objective of your experiment was to infect an individual cell, yet the image you showed me shows four green cells! Doesn’t this mean you infected not one, but four cells initially?”
Here’s that picture again.
Yes, you are in fact correct in saying that there are not one, but four green cells in this picture (you can see the distinct outline of all four). And based on what I’ve told you previously, it is good logic to assume that the multiplicity of green cells here can only mean a multiplicity of initially infected cells. However, there’d be a couple of related details that you’d be neglecting to consider.
Firstly, cells divide. “Yeah I know that, so what?”
So this: secondly, HIV can create stable genetic changes in infected cells across generations.
I hope you can now see where I’m going with this. Not only would the initial singly infected cell glow green, but the daughter cells it produces through asexual mitosis (where produced cells are genetically identical to the single “parent”) would also glow green, the genetic chances wrought by HIV infection staying with them. Same deal with the “grandchildren” of the first infected cell, and so on, for quite a while.
So our explanation for this deviation from expectation is that what we’re seeing here is not the result of multiple initially infected cells, but a single infected cell surrounded by its glowing children.
This is also supported by the fact that the “doubling time” of MEF cells is 24 hours. Remember that this picture was taken 48 hours post-infection. That means the hypothetical singly infected cell would have time to divide twice.
The first time resulting in 2 green cells, then the second division resulting in 4 green cells as the first two cells divide. See? What we see in this picture is consistent with the infection of only an intial single cell. If we’d infected, say 2 cells, the number of green cells present would be much higher at 8 cells.
So the first experiment? Flawless. But now we get to the tricky stuff.
Experiment 2: The Crash and Burn
My preceptor was way too ambitious with this one. He made it unnecessarily complicated , and as a result, we couldn’t verify whether we’d accomplished any of the objectives we had in mind for this experiment.
What were these objectives? They were twofold:
1. Test the ability of the viral stamping system to simultaneously infect a cell with two distinct viruses (although not necessarily just one of each type)
2. Use viruses that encode the components of the CRISPR-Cas9 system
[Credit]
“Crisper what?”
Ah, well I need to quickly explain to you what the CRISPR-Cas9 system is. CRISPR stands for “Clustered Regularly Interspaced Palindromic Repeats.” Knowing that isn’t really essential to understanding what the CRISPR-Cas9 system is, and how it works, but it’s guaranteed to make your IQ seem at least 10 points higher to other people when you spit that out like it’s some hot rhymes. Cas9 (for CRISPR associated protein 9)I suppose is more important to know, it’s the name of an enzyme (a nuclease, which cuts DNA in two) that’s one half of the CRISPR-Cas9 system.
More to the point, what is the CRISPR-Cas9 system? In nature, it’s found as part of a bacteria’s immune system. Bacteria capture bits and pieces of viral DNA, then create the corresponding RNA which attaches to the Cas9 nuclease. The attached RNA allows the nuclease to selectively “cut” DNA that matches the RNA.
With CRISPR-Cas9, the lab version of this bacterial immune system, scientists can choose what RNA (otherwise known as a guide-RNA) is put with the Cas9 nuclease. This is very useful, as using the system, scientists are able to selectively cut virtually any sequence of DNA that they wish.
That’ll have to do. So what was my preceptor’s grand plan for accomplishing these objectives?
It’s a bit complicated.
We’d be using two modified HIVs with completely different genetic content in the form of plasmids.
As the diagram says, plasmid B is “chemically activated.” This means that the contents of this plasmid won’t be made until doxycycline, the chemical, is added to the solution of cell media.
That's how my preceptor thought things would play out. Unlike the first experiment, 48 hours after infection, we expected to see red cells.
Red cells, not green cells?
Bear in mind that Plasmid B, housing the GFP and Cas9 genes requires the presence of doxycycline to be transcribed, which had not been added yet. Plasmid A with RFP and the GFP gene-complementary guide RNA on the other hand, was not dependent on the presence of doxycycline to be transcribed into protein, hence why it was present in sufficient amounts to produce visible red fluorescence.
Right. So after checking fluorescence, we added doxycycline to the cell culture and waited until it was around 120 hours post-infection. When we checked the fluorescence of the cells again at that point, my preceptor expected us to see yellow, the result of combined green and red fluorescence. Later, as the CRISPR-Cas9 system came online and the GFP gene was inactivating through CRISPR-Cas9s cutting action, later cells would be only red as the result of GFP deactivation. So in theory, we should have seen some yellow cells, surrounded by red cells. To clarify, the first few cells would have had their GFP gene deactivated as well, but my preceptor supposed that enough GFP would have been produced at that point to create visible fluorescence, so the first cells would remain yellow. The later ones would have inherited the defunct GFP gene from the start, so they would only be red.
Except we didn’t see that...
There was an awkward half-hour as my preceptor hunched over the fluorescence microscope, attempting to locate any speckle of green or yellow using both light filters among the stubbornly dark cells. That’s what that dark image was, a “picture” of the culture taken through the green fluorescence filter, displaying the green fluorescence, or the conspicuous lack of it. On my poster, it served as the bleak announcement that the latter half of our experiment had failed.
It wasn't so much the lack of yellow that meant the experiment had failed, it was our inability to verify whether or not the objectives for our experiment had been accomplished or not. We needed to see the transition from yellow back to red to prove that CRISPR-Cas9 (which should have "cut" the GFP gene, deactivating it) had done its work, and by extension that both viruses had infected a single cell, as the CRISPR-Cas9 required both components to work and as there was one component for each virus.
Anyhow, for what it was worth, we could see red. It was faint, but it was there. That meant we knew for sure that at least Virus A had successfully infected the cell.
Pretty colors are all well and good, but you might still be unclear on what the Cas9 and the GFP-sequence guide RNA were supposed to be doing inside the infected cells. After I clear that up, you’ve got all the information you need to unravel the mystery of the failure of this experiment.
To clarify/reiterate:
1. The guide-RNA and Cas9 element can spontaneously join together upon bumping into each other, forming a complete CRISPR-Cas9 complex capable of cutting
2. CRISPR-Cas9 can deactivate genes by cutting their sequences, in this case, the GFP gene
Aaand, lastly, take a look at this (hopefully elucidating) diagram
Again, this diagram shows what my preceptor expected to happen in terms of the CRISPR-Cas9 stuff. However, you now know that instead of observing yellow fluorescence, we only saw red. The million-dollar question is: why?
I’ll give you a few hints. Firstly, recall that GFP (and RFP) needs to be produced in large enough quantities for visible fluorescence to occur. Again, this takes time. Secondly, the GFP-sequence guide RNA was present in very high concentration by the time Plasmid B was activated, as it was given a head start.
Given how the guide RNA was given time to be made in such high concentrations, how fast do you think the Cas9 protein took to bump into a guide RNA once the Cas9 started to be made? Do you think it took a long time, or only a few minutes? Major hint there.
Given it a couple minutes of thought? Think you’ve got an explanation?
Alright, well ready or not, it’s time to reveal the answer. Well, my answer and my biology professor’s answer. After a couple of seconds of consideration, she came up with the same explanation as me, thus proving without a doubt, that I was not a madman to question the validity of my preceptor’s experimental design.
Q: Why did we only observe red fluorescence?
A: Because the CRISPR-Cas9 complex formed and deactivated the GFP gene too quickly for our cells to produce the amount of GFP required for green fluorescence to be visible!
A2: Alternatively, it might have been because virus B had outright failed to infect the target cell, although this is very unlikely, due to the high number of viruses that infect a cell upon pipette contact, and the even distribution of both viruses on the pipette (as we had mixed the virus solution containing both viruses before pipette immersion)
Hold up, isn't it still possible that seeing only red cells means that both objectives had been accomplished? As pointed out by your first explanation, couldn't it have been the case that both viruses had infected the cell, only GFP wasn't produced in sufficient quantities to produce fluorescence before CRISPR-Cas9 deactivation, meaning both objectives had actually been accomplished?
This is true, however, the issue is that it's also unverifiable whether or not this was the case. You see, the second explanation, with only virus A infecting the cell, is also a possibility, however slight. It's actually more than likely that the objectives were accomplished, it's just we can't tell for sure. Because of that, it's pretty much the same as failing outright.
Anyhow, my preceptor reluctantly agreed that the former was the most likely explanation. Think about it, because one half of CRISPR-Cas9, the GFP guide-RNA, was given so much time to be produced, once the Cas9 popped out, it wouldn’t have taken long at all for a functioning complex to be formed. And there definitely wouldn’t have been enough time for enough GFP to be produced before the CRISPR-Cas9 complex cut and deactivated it. Case closed.
There also was another contributing factor though, a very low titer (concentration) of virus B used. My preceptor decided to use a significantly lower titer for virus B than the one used for the singular virus in the first experiment. The reason he did so was to prevent “off-target” infection, as higher virus titers led to more of that. Off-target infection just means viruses floating off from the pipette tip and wandering off to infect some cell that we didn’t intend to infect. Only ideally, we wanted many viruses of each type to infect a cell, and not just one of each. More viruses infecting means more GFP and RFP produced in parallel as each virus would integrate their own copy of their genetic materials into the host genome, each integrated gene being transcribed. With lower virus titer, and thus less successful infections, GFP and RFP would be made slower, and thus they would fluoresce fainter. Avoiding off-target infections was a solid rationale for lowering the used virus titer, as we had to account for the titer of virus A as well, but regardless, it had some very drastic consequences for the strength of fluorescence. Take a look at this micrograph here.
It’s one thing to complain about piss-poor experimental design, it’s another to offer an alternative design that is not only an improvement upon the latter, but also simpler.
Is this even possible in this instance? Complicated problems require complicated solutions...right?
Nope, not necessarily, and certainly not in this instance. Allow me to briefly present an alternative experiment that would work and accomplish the objectives. See below for diagrams describing the plasmids I would use, and what would go on in a cell simultaneously infected with two viruses bearing these plasmids.
As promised, both objectives are fulfilled by this experimental setup. As an added plus, you don’t even need to bother with doxycycline.
As you can see, plasmid A has GFP, RFP, and a GFP guide-RNA. The other just encodes the Cas9 nuclease. When you infect a cell with both viruses, everything is made at once. However, the GFP would be inactivated fairly quickly, as the components of the CRISPR-Cas9 complex would be made from the start. So a short while later, you’d see the cells would fluoresce red and red only. [Edit 6/03/21] You might see yellow if something failed (see below), but you would never see green on its own. If you do, fire your lab tech. They probably screwed up the plasmid and only put GFP on plasmid A.
How do you know both objectives are fulfilled?
You know that objective 1 (simultaneous infection with two distinct viruses) is fulfilled if you see only red. Let me explain.
[Edit 6/03/21]: RFP is only encoded by plasmid A. So if you see red, that means plasmid A was successfully delivered. But wait, GFP is also on plasmid A, how is only red visible? This could only be the result of GFP being successfully knocked out. The only way this could have happened is if a complete CRISPR-Cas9 complex was formed, which requires BOTH viruses to infect the cell and deliver their plasmids.
Some contingencies...
If you see a yellow cell, that means virus B failed to infect the cell, as GFP is still active. A non-fluorescing cell could mean anything pretty much, as it could be a cell only infected with virus B, or perhaps more likely, it could just be a cell not infected by either virus.
For objective 2 (successful usage of the CRISPR-Cas9 system) well, again, if you see a cell that just fluoresces red, it means that GFP has been deactivated, by the CRISPR-Cas9 system. Which just means CRISPR-Cas9 worked successfully.
Q E friggin D.
Thanks for reading!!
- BJW
댓글