Today I wanted to look at small amounts of RNA – so I went for the gold! SYBR gold, that is. It’s a fluorescent nucleic acid stain, meaning that it binds to nucleic acids (RNA & DNA) and gives off light at a specific wavelength when you shine light of a different specific wavelength at it. Other fluorescent nucleic acid stains include the rest of the proprietary SYBR family (e.g. SYBR Green, SYBR Safe) as well as the non-proprietary Ethidium Bromide (EtBr) and DAPI. I’ve used many of these, but was curious as to how they work and how they compare to one another, so, in addition to looking for RNA, I went looking for information! And I want to share with you some of what I learned. There are a lot of different stains out there and I’m not going to try to compare them all, but what I will do is tell you about how they work and some qualities you should look for when choosing which to be using!
First, let me make sure we’re all on the same page regarding terminology. Nucleic acids refer to DNA & RNA (there are also some different lab-made nucleic acids which I won’t get into here). Both DNA & RNA are made up of building blocks or “letters” called nucleotides, which consist of a sugar (ribose for RNA and deoxyribose for dNA) off of which sticks a phosphate-containing group (used for linking letters together into chains) and a nitrogen-containing ringy-thing called a nucleobase (“base” for short). These bases are what make different letters unique – DNA has A, G, C, & T and RNA has A, G, C, & U – and they’re important for allowing nucleic acids to form specific base pairs to other nucleic acids (A:T or A:U) and (G:C). This specific base pairing allows for DNA & RNA to store genetic information and be copied and stuff, but it’s not that relevant for our discussion today because a good fluorescent stain shouldn’t care about what letters are present – we want a stain that stains any sequence!
One consequence of base pairing that does matter for our discussion, however, is that it allows single-stranded (ss) nucleic acids (ssRNA & ssDNA) to bind to strands of complementary nucleic acids to form double-stranded (ds) helical structures (dsRNA & dsDNA) in which the bases are “stacked” in the center like rungs on a ladder. note: for DNA, this typically happens between 2 separate DNA strands, but for RNA, which is more commonly single-stranded, regions of the same strand of RNA often fold back upon themselves to form double-stranded regions, such as hairpins, & stem-loop structures, within a single strand. Why am I telling you all this? Because the stranded-ness can affect how well some stains work. So now let’s talk about how these stains work.
There are 3 things we need in our stain. They need to be able to:
- bind what we want them to bind (have high affinity for DNA and/or RNA)
- not bind what we don’t want them to bind (be selective)
- show us where they’re bound (this is where the fluorescence comes in)
Different dyes bind slightly differently, but there are 2 major modes of binding
- intercalation (wedging in between the bases in the helix ladder, or between stacked bases in other “shapes”)
- groove binding (binding on the “outside” in the helix’s grooves)
there are also other modes of binding, and some dyes bind multiple ways and/or unknown ways and I’ve gone down way too many rabbit holes to try to find out! So I’m going to focus on these 2.
Before I discuss their differences, let me get everyone up to speed on why any molecules bind, and what molecules even are. Molecules (e.g. DNA & RNA) are groups of atoms (individual carbons, hydrogens, oxygens, etc.) joined together by sharing negatively-charged subatomic particles called electrons in strong “covalent bonds.” These electrons whizz about in “electron clouds” surrounding the atoms’ dense central atomic cores. Those cores are called atomic nuclei and they contain positively-charged protons (which help reign in the negatively-charged electrons) and neutral neutrons (which help hold everything together). If there’s an imbalance of electrons & protons, you’ll get a fully-charged molecule (an ion). But you can also get partial charges in neutral molecules. This is because, when there’s more “electron density” in one area of a molecule (more electrons hanging out in that area of a cloud), that area will be partly negative and somewhere else will be partly negative. And opposites attract, so you can get molecules binding through temporary, weaker “attractions” than the strong covalent bonds which involve actual electron sharing/cloud merging.
Speaking of electron sharing, something cool is going on in the rings of nucleic acids’ bases, and in the rings of the dyes we’ll see. This cool thing is called resonance stabilization or electron delocalization or conjugation and it’s basically a bunch atoms sharing “extra” electrons with one another in a sort of electronic commune. It takes 2 electrons to make a single covalent bond and 4 to form a double covalent bond, but if atoms have extra, under certain circumstances like alternative single and double bonds, they can share them if all of the atoms agree to buy in. We’ll talk more about this later because it’s crucial for light absorption. Now let’s get back to modes of binding.
A classical intercalating stain is Ethidium Bromide (EtBr), which works best for dsRNA (it can bind ssDNA & RNA, but not as well). The ethidium part of EtBr is positively charged, which makes it attractive to negativity-seeking Bromine (Br). But when EtBr sees DNA coming by, it ditches that bromide in favor of the super negative DNA (negative because of the phosphates in the DNA backbone). DNA’s positive charge draws Et in and Et has nice flat rings kinda like DNA bases which stick themselves in nicely in between DNA’s bases. And it will stick there because all those rings sync up their electron clouds (they’re not merging them, just syncing up where the electrons are hanging out within them) – it’s the biochemical version of doing that thing where you take 2 phone books and interleaf their pages and then try to pull them apart.
Having EtBr super stuck in DNA is great in a gel but not in a cell! If there’s EtBr in DNA and a polymerase goes to copy that DNA, it can make mistakes. This can lead to mutations. So EtBr IS a “mutagen.” It CAN cause mutations to DNA and in some cases, mutations can cause problems like the uncontrolled growth of abnormal cells (cancer). BUT in order to cause those mutations, EtBr would have to get past all your body’s defenses and into your cells. And apparently it’s not very good at this.
All the data to date that EtBr’s so dangerous comes from tests done on cells to test if something’s a mutagen. Check out this post on the Ames Test http://bit.ly/2TkzbKR
Tests done on mice tell a different story. They’d have to eat or breathe a lot of it (way more than you’d ever be exposed to if you’re staining a gel properly – especially if you’re also taking proper safety precautions). And it’s actually used to treat African sleeping sickness in cattle. Here’s a link to a great article by Dr. Rosie Redfield, who calculates that “A 50kg researcher would need to drink 50 liters of gel-staining solution to get even the non-toxic dose used in cattle.” http://bit.ly/2SqJAzK
I’d always heard that EtBr was super dangerous, but looks like scientists themselves can buy into the chemophobia fear-mongering. Perhaps the worst part is that some of the “safer” alternatives companies market may actually be MORE toxic (but they haven’t been tested as much) and more expensive. And a lot of the articles you’ll find online about the dangers of EtBr and the advantages of newer stains are written by the companies selling the newer stains. I’m not here to tell anyone what they should or shouldn’t use but I do encourage you to look into the data (like I should have done a long time ago).
That’s not to say there aren’t benefits of some newer stains. Especially when it comes to sensitivity (how little a quantity it can detect). Sensitivity is especially important if you want to extract the DNA out of the gel and use it after you get a look because you don’t need to shine as much UV light on it to get the fluorescence as we’ll talk about -> UV light is a mutagen you should be much more afraid of! https://bit.ly/uvdnadamage
So that was an intercalating stain, and many of the newer stains are too. But let’s look at that other mode of binding – binding to the groove. When DNA &/or RNA form a double helix, the helix has a more open “major groove” and a narrower “minor groove” which offer binding potential. An example of a minor groove binder is 4′,6-diamidino-2-phenylindole, or as its friends like to call it, DAPI. It’s the basis of the EZ-vision stain our lab uses. It binds in the minor groove, specifically in AT-rich regions (G has an -NH₂ group sticking out into the minor groove, making it uncomfortable for DAPI).
Some stains work much better on certain types of nucleic acid. For example, SYBR Green. It works much better for dsDNA, which makes it super useful for real-time quantitative PCR (RT-qPCR) where you measure the formation of dsDNA products in real time. Here you want something that will only fluorescence when dsDNA is present, not the single-stranded DNA you start with. http://bit.ly/rtrtqpcrprimer
EtBr also works best on dsDNA, but it’s much less sensitive, so it can’t detect the low levels of DNA present in these reactions. Sensitivity is usually reported as the detection limit and they usually report it for each type of nucleic acid (ssDNA, dsDNA, or RNA). The SYBR Green I mentioned is part of a family of proprietary “SYBR” dyes and the SYBR Gold I used today for RNA staining is the most sensitive. It can detect as little as 1 nanogram (ng) (a billionth of a gram) of RNA! The dark band in the picture is ~30ng.
But sensitivity isn’t the only thing that matters – for instance, a stain could be super sensitive but not specific, so it’d bind to anything and fluoresce which wouldn’t tell you anything about what there was to bind to. Or it could fluoresce without binding to anything, so you’d have mega high background unless you washed out all the unbound stuff. What’s really cool about all of these stains (I think) is that they’re pretty much only able to fluoresce once bound because the binding alters their electron accommodations just right… What am I talking about? Sit tight! Because we need to go into some more detail on fluorescence…
As I mentioned above, what I mean by fluorescence is that you can shine a light of one wavelength at a fluorophore (such as these stains) and they will absorb that original light and give off light of a different wavelength. Why? To understand this we need to back up and discuss what light is. Light is “just” little packets of energy called photons traveling in waves and different colors of light have photons with different amounts of energy. The more energy the photons have, the more they want to move, but they all travel at the same linear speed (the speed of light), so instead of traveling faster, more energetic photons take a more slalom-y path, up-down-up-down-ing more along the way. As a result, the more energetic the photon, the higher the frequency of the waves, and the tighter packed the peaks are (shorter the wavelength). ⠀
When we think about light, we usually think about “visible light” but that only reflects a small slice of a vast electromagnetic radiation (EMR) spectrum which also includes things like microwaves and infrared on the slow side and ultraviolet (UV) & x-rays on the fast side. Visible light is just the portion of the EMR spectrum that our eyes have receptors for – the least energetic (longest wavelength) light we can see is reddish and the the most energetic (shortest wavelength) light we can see is purplish. A lot of times, things absorb light of wavelengths that we can’t see. So we don’t realize those wavelengths have been stolen. But with fluorescence, they give back light of a different wavelength which we can detect, so we indirectly measure that light being stolen. These stains often absorb light in the UV range, which is convenient for labwork where labs often have UV lightboxes like “GelDocs” which have cameras attached and/or just old-school lightboxes. We just got a new GelDoc and it’s amazing! Our old one we had to manually focus and it took forever so usually I just gave up and took a blurry-banded picture. This one autofocuses though!!! in the pics I have it open so you can see the gel tray but you want it closed when you have UV on!
Sorry for that detour… Now we can revisit why things absorb different wavelengths of light in the first place. This is where it helps to go back to thinking of light as energy packets instead of waves. The electrons in a molecule whizz all around but there are places they most like to hang out, and we call these “orbitals.” You can think of them kinda like electron housing units and electrons have to “pay rent” in the form of energy to live there. The further away from the central atomic nucleus (containing the positively-charged protons charged with reigning them in), the more energy is required to live there. Resonance lowers the difference between these, making aromatic rings great for absorbing light. ⠀
If an electron absorbs a photon with the exact right amount of energy to pay the rent difference between their current orbital and a further-out one, they can “get excited” and temporarily move to a higher-energy, further from the nucleus, orbital (and then usually fall back down with the energy released as heat, although in the case of fluorescence, that energy is released as light of a different wavelength). The amount of energy required for the move depends on the atoms involved, how they’re hooked up, and what other molecules are around, so different molecules absorb & emit different wavelengths of light (and the same molecule might absorb different wavelengths under different circumstances). Some molecules have multiple parts that can absorb light, so they have multiple absorption wavelengths. For example, SYBR gold has 2 choices of excitation wavelength (~300nm (which is in the ultraviolet (UV) range)) & ~495 nm (in the blue range).
All fluorophores, even the ones we usually think of as having a single absorption wavelength, actually have a bit of a range, like a bell curve where there’s a peak at the “absorbance maxima” wavelength, but then some absorption at slightly higher or lower wavelengths. This has to do with some wiggle room in the how much energy’s needed to move an electron due to things like “vibrational levels” between states.
The emitted light typically has a longer wavelength, lower-frequency, lower-energy, because some of the absorbed energy from the input light gets lost as heat, vibrations, etc. For example, SYBR Gold emits ~537nm (in the green range). The difference between the absorption maxima and the emission maxima is called the Stokes Shift.
Ideally, there shouldn’t be a lot of overlap between the wavelengths that are absorbed & the wavelengths that are emitted (you want low spectral overlap).
What else do you want? A couple of things to look at when comparing stains to find the best one for the task
extinction coefficient (ε): When light of the ideal wavelength hits, how much can be absorbed? You usually want a HIGH ε, meaning the stain will absorb more light and hopefully emit more light.
But will it *actually* emit more light? For this, you want to look to Quantum Yield (QY): How much of that absorbed light is subsequently emitted as light (as opposed to absorbed energy being dissipated in other ways such as given to other molecules (quenched))? You usually want HIGH QY, meaning you get more light emitted for each photon absorbed.
You also want your fluorophore to be able to do it (absorb->fluoresce->absorb->fluoresce->absorb->fluoresce…) for a long time without whimping out. That is, you want high photo stability: How long does it keep ability to fluoresce despite being exposed to lots of light? i.e. how susceptible is it to photobleaching ( https://bit.ly/2BqnHO0 )? which is a sort of fluorophore exhaustion where the fluorophore gets “stuck” in a non-fluorescent state. The more light you shine on it, the more chances there are to get stuck in one of these “forbidden states” and thus you see photobleaching after high/long light exposure (and why you should store and use the dye shielded from light (such as covered in foil)). You want a fluorophore to have HIGH photostability, meaning it’s less easily “exhausted” and stays fluorescently-active longer.
Other properties depend on what applications you’re using it for. For example, today I was using SYBR-Gold to stain a urea-PAGE gel.
If you’ve read this far, assuming anyone has…, I kinda assume you are familiar with using electrophoretic gels to separate molecules by size. But if not, check out these post: http://bit.ly/ureapage & http://bit.ly/agarosegelrunning
Urea-PAGE gels have tight mesh, which makes them good for separating small pieces of RNA, but it also makes it harder for a dye to get into when you soak it in. So I need a dye that has good penetrance like SYBR Gold. It comes as a 10,000x concentrate, meaning you have to dilute it by a factor of ten thousand before using it (e.g. 5uL to 50mL). Just drop some into buffer (like the buffer you ran the gel in – it’s less stable in plain water), plop in your gel, shield it from light to prevent photobleaching, give it 10-40 min on a shaker and go scan. note: I’m really not getting paid by SYBR Gold makers, it just happens to be what I was using today so what I was most curious about!
Probably the most common place you think about fluorescent staining of nucleic acids is staining DNA separated using an agarose gel. The staining can occur before, during, or after running the gel. Our lab uses EZ-Vision sample-loading buffer when we load DNA onto agarose gels. It’s sold by VWR and this isn’t a paid endorsement or anything – it’s just what we use – because it makes it EZ to see the DNA – we mix our sample DNA with it so the fluorescent DNA-binding dye gets bound to the DNA before we even run it through the gel. Other methods involve putting the stain in the gel itself or staining the gel afterwards by soaking.
If you’re interested in learning more about different dyes, comparing them, etc. I recommend checking out this super technical (and well-referenced) guide from Invitrogen/ThermoFisher (who also isn’t paying me to say/share this!): https://www.thermofisher.com/us/en/home/references/molecular-probes-the-handbook/nucleic-acid-detection-and-genomics-technology.html
more on topics mentioned (& others) #365DaysOfScience All (with topics listed) 👉 http://bit.ly/2OllAB0⠀