Work in a lab & there are some “boring” experiments you have to do A LOT in order to do the “exciting” experiments that require more thought. An example many people might be overly-familiar with if they’ve ever taken a biochemistry or molecular biology course or joined one of those types of labs is the “miniprep” (aka alkaline lysis). This is a technique in which we separate and purify plasmid DNA we put into bacteria (a circular piece of DNA often containing the genetic instructions for a protein we want to study) from all the stuff that was already in the bacteria. Once pure, we can then do things like check that the plasmid has the gene we think we put in there. And, if so, stick the plasmid into cells that’ll make protein from it. So, when you have a lot to molecularly clone, this is a technique you’re gonna hone!
note: originally posted 11/16/20, updated & added video 10/27/21 & 5/20/22
The first times you do an experiment it’s always kinda stressful & it’s hard to imagine it will ever seem “boring” but, for things you end up having to do a lot that don’t require a lot of “mind thought” thanks to muscle memory, you can get bored – if you let yourself. But if you imagine what the molecules are doing as you carry out each step, it makes it a lot more fun. And it helps you really appreciate why you’re doing what – which is especially important when you’re working from a kit as is often the case with minipreps. So let the bumbling biochemist help you beat boredom & bring some pep to the miniprep!
Aka alkaline lysis, “minipreps” is a technique in which we separate and purify the plasmid DNA we put into bacteria (the DNA we want) from all the stuff that was already in the bacteria (the DNA we don’t want, proteins, sugars, etc.). You can’t just break the cells open (lyse them) & pull out *all* the DNA because the bacteria has its own DNA which you aren’t interested in. But some cool chemistry comes to the rescue! Of course, we can’t just jump to the climax of the story, so let’s step back and set the scene…
Say you want to study a protein. You can use molecular cloning to stick the gene for that protein into a circular piece of DNA called a vector plasmid (using methods such as restriction cloning where you use sequence-specific DNA scissors called restriction enzymes to cut open the plasmid and cut around the gene you want to insert, generating complementary sticky ends that you can stitch together using DNA ligase). http://bit.ly/molecularcloningguide
Once you have your cloned plasmid, you can use “transformation” to stick that plasmid into bacterial cells. Common methods for this include chemically competent cells + heat shock and electroporation, techniques which temporarily open pores in the bacterial membranes which allow the DNA to slip in. http://bit.ly/transfectionmethods
Not all of the cells will take in the plasmid, but we’ll weed out the ones that don’t in the next step… plating on agar plates (think Petri dishes) filled with bacteria food (media) spiked with an antibiotic. In addition to (hopefully) our gene of interest, the plasmid contains an antibiotic resistance gene. If we plate the bacteria on media containing the matching antibiotic, only cells that took in the plasmid will be able to grow.
note: by “plating” it’s not some 5-star-restaurant display thing – all we’re doing is trying to get the bacteria as spread out as possible, which we can do using cell spreaders or glass beads.
Each cell that took in the plasmid will start growing and dividing. Even though we spread the cells out, instead of getting a fuzzy lawn of bacteria we (hopefully) get some scattered globby dots. Each of these dots is called a colony, and represents 1 original cell taking in the plasmid and replicating asexually (just copying its DNA & splitting itself in two). Thus, each cell within a colony should have the same plasmid (though the cells might have different numbers of copies of it) but cells in different colonies might have different plasmids.
If all went well, all of the plasmids taken in will be “perfect” – the gene really got in there okay, there aren’t any typos etc. But all doesn’t always go perfectly… so we want to check.
In order to check, we’ll need more DNA.* So we choose a few colonies and stick a bit of them in some liquid media (typically 5mL LB) and let them grow overnight (at 37°C with shaking). Then we can isolate the plasmid DNA and send it for sequencing (and/or do an analytical restriction enzyme digest) to make sure all is A-OK before moving on.
*note: you can do a pre-check directly from the colony (no purifying) using a method called colony PCR – take a teeny bit of the colony swirl it around in a PCR reaction mix and try to amplify regions in and around your insert. Sorry for all that jargon but it isn’t relevant right now and didn’t want to stray too far from the mini prep, just wanted to let you know this technique exists and you can learn more about it here: http://bit.ly/colonychecking
Even if you do colony PCR, you often do it in tandem with starting the overnight cultures (so use part of the colony for PCR and part for an overnight culture (and make sure you cross-reference them!) because in addition to just checking, you’ll also want purified plasmid for using if it checks out okay!
So we want to purify our plasmid DNA (pDNA), but that plasmid is in bacterial cells which have their own genomic DNA (gDNA) as well as a bunch of proteins and stuff. So how do we isolate the pDNA?
Clearly, we’ll need to get it out of the cells, which we can do by breaking the cells open (LYSIS). But before you break the cells open, you want to make sure they have an ideal environment to come out into. The cell membrane provides a nice natural barrier to allow you to swap out the external environment – so take advantage of it!
Right now, the cells are in the liquid media you grew them in (often Lysogeny Broth, LB) – in addition to the antibiotic for selection, the media contains salts, peptides (from the tryptone – trypsin-digested casein protein), vitamins etc. (from the yeast extract), all sorts of stuff the bacteria want/need to survive and thrive (and make lots of copies of your plasmid). http://bit.ly/bacterialmedia
If our goal is to purify the pDNA, we might as well get rid of as much of that stuff we don’t want now, while the cell membrane provides a convenient natural barrier (especially because some of that stuff could interfere with future steps). We also want to minimize the volume we’re working with. Right now, the cells are “spread out” in excess media but if we “spin them down” by centrifugation, we can separate the cells (heavy) from the media (light).
Now we resuspend them in a smaller volume of a more ideal solution. We’re going to be using a series of different solutions we refer to as “buffers.” “Buffer” is in reference to their pH-stabilizing ability which comes from component(s) that can both give protons (act as an acid) and take protons (act as a base). But we often use the term “buffer” pretty broadly to refer to different pH-stabilized “salt waters” http://bit.ly/phbuffers
The first buffer we’ll use in the miniprep is resuspension buffer (“P1” in this QIAprep kit (this isn’t a paid endorsement or anything, just what our lab uses) has:
∙ Tris – this acts as the actual pH buffer (pH stabilizer) – it’s here because we’re not ready to change the pH yet
∙ glycerol – this is an osmotic balancer – osmosis refers to the movement of water from where water is more concentrated (there’s less other stuff around) to where water is less concentrated (there’s more other stuff in between the water molecules). Since there’s a lot of stuff in the cells, if there isn’t a lot of stuff outside the cells, water will flood in. So we add glycerol to increase the amount of stuff outside the cells and thus keep the cells from bursting before we’re ready. Later we’ll burst them with detergent (synthetic soap), not pressure differences
∙ RNaseA – this degrades RNA (and is why P1 has to be stored in the fridge) – we don’t want the RNA, so might as well break it down so it doesn’t clog things up or try to tag along (a lot of times in the lab we’re trying to protect from RNaseA, because this RNA chewer is all over the place (for example, bacteria often secrete it to destroy foreign RNA (like from viruses) before they can invade), but here our usual enemy is our friend because we’re both on team DNA! (RNaseA won’t chew DNA because it works in a sneaky way that gets the RNA to attack itself using its 2’ oxygen (which DNA doesn’t have – hence the d in DeoxyriboNucleic Acid) https://bit.ly/rnaseadepc
note: with this kit you have to add the RNaseA to the buffer (and then check the check box on top of the bottle to signal you did)
∙ EDTA – this is a chelating agent (metal-biter) -> binds divalent cations (molecules with a +2 charge) like Mg²⁺ or Ca²⁺ which DNases (DNA chewers) need – prevents DNases from degrading the plasmid (RNaseA doesn’t need cation cofactors so it can still work and chew away the RNA).
EDTA also helps destabilize the cell membrane. This membrane is made up of a “sandwich” of phospholipids, which are kinda soap-like molecules that have long fatty tales that form the hydrophobic (water-excluded) “peanut butter” of the sandwich, and hydrophilic (water-loving) heads that form the “bread” facing the watery environments of the inside of the cell (cytoplasm) and the outside of the cell. These phospholipids have negatively-charged phosphate groups that bind to positively-charged ions that helps keep them from repelling one another. So, by stealing those cations, the EDTA sets the scene for some head-to-head repulsive action that’ll make the next step easier
What’s this next step? Lysis! We didn’t want to break the cells open before, when the outside of the cell was so “dirty”, but now that we’ve resuspended it in this “safer” & “cleaner” environment, we can go ahead lyse them. And for this we use another buffer
Lysis buffer has a detergent called Sodium Dodecyl Sulfate (SDS) – this is the same detergent we use to denature proteins (unfold them and keep them unfolded) when running SDS-PAGE gels to separate proteins by size. http://bit.ly/sdspageruler
And SDS will denature the proteins in the cells (which is fine (even good) because we don’t want them anyway), but before it does that, it has to get into the cells. And when it does, it breaks down the door so everything comes spilling out.
The “doors” of e. coli are cell membranes made up of those sandwiches of phospholipids (and lipopolysaccharides (LPS’s) in the outer outer membrane (these have sugar chains sticking off them), with a peptidoglycan wall (made up of peptide-linked sugars) in between (thankfully for us, e. coli is Gram negative so this wall isn’t very thick). SDS knocks down the door by being sneaky. Like other detergents, it’s amphiphilic -> it has a water-loving (hydrophilic) part and a water-avoided (hydrophobic) part
The phospholipid & LPS molecules making up the cell membrane are also amphiphilic, so the SDS can kinda “join their club” -> disrupt the membrane, breaking the cell open (it also denatures the proteins embedded in the membrane
Alkaline is another name for “basic” and is the opposite of acidic. In alkaline conditions, there are more hydroxide ions (OH⁻) than protons (H⁺) and thus the pH is higher. In acidic conditions, there are more H⁺ than OH⁻ and the pH is lower (yeah, it’s an inverse log scale so more protons is lower pH…). The reason we care is that molecules can give or take protons depending on how many there are around, and this changes the giver or taker’s characteristics. http://bit.ly/phacidbase
The “alkaline” part of alkaline lysis comes from sodium hydroxide (NaOH). NaOH is a “base” and a base (at least in one definition of the term) is a molecule that “steals” protons (H⁺) – it can steal protons from DNA “bases.” (egads…) Nitrogenous bases, aka nucleobases, aka “bases” are the unique parts of DNA letters that stick off the individual strands and form specific hydrogen bonds (H-bonds) with bases on the other strand (A:T & C::G). http://bit.ly/nucleicacidstructure
In addition to helping disrupt the cell wall, NaOH disrupts such base pairing between strands of DNA. Both the plasmid and the bacterial genomic DNA (gDNA) are double-stranded and NaOH breaks up the H-bonding between the 2 strands, “unzipping them.” These H-bonds aren’t as strong as the covalent bonds linking the letters up into chains, so you can unzip the strands without unchaining them. Here’s a brief overview of why…
Those strong covalent bonds come from sharing pairs of electrons, and H-bonds can come as a consequence of unfair pair shares. Some atoms (like oxygen and nitrogen) are electron-hoggy (electronegative), so they pull the shared electrons closer to themselves, making them partly negative and their partner partly positive. And opposite charges attract, even partial ones, so you can get hydrogen atoms attached to electronegative atoms becoming partly positive and getting attracted to electronegative atoms with a lone pair of electrons and a partial negative charge. And we call this a hydrogen bond (H-bond). http://bit.ly/frizzandmolecularattractions
These H-bonds rely on having hydrogen, so if NaOH steals that hydrogen, base pairing is disrupted and the strands come apart. This isn’t the final product we want (we want double-stranded pDNA) so we’ll have to let the pDNA strands come back together but there is a point to this step, don’t worry – at the same time we’re unzipping the pDNA we’re also unzipping the gDNA and we won’t let that come back together. And that will let us discriminate between the pDNA (which we want) and the gDNA (which we DON’T want)
You want the lysis buffer to be able to reach all the cells (so hopefully you resuspended the pellet well in P1). You could be vigorous there, because the DNA was still protected by the cell membranes, but once you break the cells open you need to be more careful. When you add P2 (lysis buffer) you want to mix well, BUT gently -> you don’t want to mix too vigorously or you’ll “shear” the gDNA -> break it into pieces that will stay soluble. So mix by inverting the tube *not* by pipetting or (definitely not) vortexing (for those not familiar with the arm-numbing power that is the vortex, it’s a really fast orbital shaker kinda like a small, very aggressive, electronic foot massager.
NEUTRALIZATION – This is where we get the pDNA to come back together by “undoing” the proton stealing. Instead of trying to get the NaOH to give it back, we add a fresh source – acidic potassium acetate.
The potassium acetate acts as an acid, donating protons, so the DNA bases act as bases and take the protons -> now they can form base pairs again. If they can find their binding partners.
The plasmid DNA is small (relatively speaking), circular, and supercoiled (really twisted up), so even though the strands unzip under the alkaline conditions, they stay near each other – their partners are right next door, so they can rezip readily.
But the gDNA’s a lot bigger, (e. coli’s genome’s ~4.7 million basepairs (letters) whereas a plasmid’s usually less than ~200 thousand) so rezipping is a lot harder for gDNA, especially since, when it got denatured, hydrophobic bases were exposed that latched onto lipids and proteins to form a kind of tangled up mess.
And the proteins and lipids in that mess are bound to SDS. And when you added potassium acetate you didn’t just add acid, you also added potassium (and a vinegar-y smell). As you might have found out the hard way if you’ve run SDS-PAGE on proteins purified in a buffer with high KCl, unlike sodium dodecyl sulfate (the sodium salt of dodecyl sulfate), the potassium version is insoluble. So that big web of stuff bound to the SDS (all that protein, lipids, – and gDNA!) will precipitate (come out of solution) as a kinda fluffy white precipitate.
You spin this down again and the precipitate will separate from the liquid supernatant containing your plasmid. But this liquid doesn’t only have your plasmid – you’ve gotten rid of gDNA and (most) proteins & cellular debris, but you still have a bunch of salts, EDTA, RNase, etc. to get rid of
Clean up time! At this point, things are basically like a PCR purification (“clean-up”), which I go over here: http://bit.ly/spincolumns
Here’s the gist: We apply the solution to a spin column – these columns have membranes made of silica (amorphous (no-consistent-shape) silicon dioxide, SiO₂), which offers a variety of binding options for DNA. You pipet your sample into the “cuppy” part above the membrane and use centrifugation or vacuum to help pull the liquid through. The DNA binds the column -> we let everything else flow through. We wash it. We unstick it and let it flow through.
To help with the sticking, N3 has guanidinium chloride, a chaotropic salt which disrupts (brings chaos to) the water coat around the DNA and lets it bind the membrane instead. Then comes the wash. Now that you’ve gotten DNA stuck on there tight, you want to wash off any lingering extra salts. You do this using a buffer containing ethanol (buffer PE in Qiagen’s kit). Just add the liquid the same way you added your sample, spin it through & toss it out.
DNA’s not soluble in ethanol, but salts are, so the extra salts are removed with this wash step. After the wash & toss, spin it again to give any residual ethanol another chance to get pulled through. It’s really important to remove the ethanol or the DNA will have problems dissolving &/or when you go to load samples into an agarose gel (more here: https://bit.ly/agarosegelcompare) they’ll float up out of the well cuz ethanol’s less dense than the buffer.
Once we’ve washed off the gunk we can redissolve the pDNA (which is now pure!) so it can flow on through. The Qiagen kit comes with an elution buffer (buffer EB – which is 10 mM Tris·Cl, pH 8.5) (Tris is that buffering agent we saw before) you can use or you can use plain old (but really pure, nuclease-free) water. If you use water, give it a minute to fully dissolve the DNA before you spin it through.
Speaking of which -> once you redissolve it, you spin it through into a NEW TUBE – this one you want to keep!
Then, take your spun-through, pure DNA over to the spectrometer like a NanoDrop & check out its absorbance spectrum (it should have a 260/280 ratio of ~1.8). More on that here: http://bit.ly/beerslawexplained
In the QIAprep kit you have the option of adding “LyseBlue reagent” (thymolphthalein dissolved in ethanol) to P1 It serves as a pH indicator to help you see whether you’ve mixed well (but gently, remember!) in the lysis step and whether you’ve really neutralized in the neutralization step.
Thymolphthalein deprotonates at a pH ~10 which makes it go from colorless to blue (more on pH indicators here: http://bit.ly/2C9bA5J )
P1 has a pH of 8 (thanks Tris!) so it’ll be colorless at this point. But when you add P2 you’re raising the pH, so it’ll turn blue. And you want it to turn blue everywhere, evenly. When you add the neutralization buffer (N3), you lower the pH, so there are more free protons, so it re-protonates and goes colorless again.
Happy prepping! And, if the mini prep isn’t enough fun for you, there’s also bigger versions – maxi & mega & gigapreps! Those can be helpful if you’re trying to purify a lot of template DNA for in-vitro transcription (DNA-to-RNA copying) http://bit.ly/t7rnap