The Molecular Biology Code

“You’re pirates. Hang the code. Hang the rules. They’re more like guidelines anyway.”              Elizabeth, The Pirates of the Caribbean: The Curse of the Black Pearl


Molecular biology, the overall practice of manipulating DNA sequences using biological scissors and glue, is a major component of any modern research lab. Its techniques have been revolutionary with respect to expanding the types of questions scientists can answer. While it may be grounded in the fundamentals of biology’s central dogma, molecular biology is often practiced with rituals and superstitions characteristic of pirates and other Caribbean stereotypes. And, like the pirate code, molecular biology contains as many exceptions as rules. What one molecular biologist may swear by is probably not so strictly enforced by another. After all, with most molecular biology construction projects, you only need it to work once and hope the sequence is right. So what I write here today are more along the lines of guidelines for cloning success.

  1. The enzyme is the last thing to add to any reaction.
  2. Do not vortex enzyme reactions.
  3. Use a new tip for your pipettor each time.
  4. Prepare a master mix for multiple reactions.
  5. Avoid gel purification of DNA fragments if at all possible.
  6. I’m not convinced that the enzyme ligase does anything, but add it anyway.
  7. Use restriction enzymes until you have evidence they don’t work anymore.
  8. Do the controls.
  9. Phosphatase treat your vector or insert but not both.
  10. Selection > Screen > Hope
  11. Sequence your finished construct to verify it is error free.
  12. Make freezer stocks.

See below for more explanations…

The enzyme is the last thing to add to any reaction.

This is basically another way of using Biochemistry Rule #4 (Use a buffer). Your expensive enzyme will likely be ruined if you put it in a tube of water or other dilute solution with no buffering capacity. So, pipet everything else first (DNA, nucleotides, water, concentrated buffer solution etc) and add the enzyme last. I would even argue that you shouldn’t even take the enzyme out of the freezer until you are ready to pipet it. Always keep it on ice or a cold block container. NEVER let it sit around on your lab bench at room temperature. If you were thinking of taking your reactions right away to the necessary water bath or instrument, think again. PUT THE ENZYME BACK IN THE FREEZER FIRST!!! That way, there’s no chance of you moving onto the next experiment without your precious enzyme safely stored away. Now, whether or not you choose to dilute your reactions with Holy Water or otherwise blessed dH2O is up to the individual scientist.

Do not vortex enzyme reactions.

After you’ve taken the care to ensure your enzyme is gently pipetted into an appropriately buffered reaction, don’t kill it by mechanically ruining it. You will need to physically mix the enzyme into the reaction because the enzyme exists in a glycerol solution that will sink to the bottom of your reaction. Do this by gently pipetting up and down or flicking the tube with your fingers. You may develop your own distinct style of mixing akin to a secret handshake. This routine is usually the superstitious result of a single instance that ‘the experiment worked when I did it this way.’

Use a new tip for your pipettor each time.

Reagents are precious and must not be contaminated. Use a fresh tip every time for each reagent and sample. Contaminated reagents may not ruin the experiment you are doing today, but they will ruin yours tomorrow and all of your labmates’ experiments. Be mindful when pipetting in general. If you can’t remember whether you added a reagent or not, just start over. Soon enough you will have your own Rain Man-like process for pipetting your reactions. On top of this, you may also develop your own style for using tips out of the box. Some of you may be a strict left-to-right or right-to-left row user, others prefer a diagonal strategy, and others may prefer to introduce designs in the tip boxes (initials or emoticons) with their tip use. Just go with whatever you find the most comforting or the technique with the highest rate of success on your reactions.

Prepare a master mix for multiple reactions.

The more times you have to pipet anything, the more potential error you introduce. Reactions become inconsistent across different samples. The way to avoid wearing out your opposable thumb doing tedious, pointless and downright erroneous pipetting is to prepare a master mix. This is just a scaled-up version of your reaction to accommodate as many samples as you have. Multiply the reagents in a single reaction by (x + 1), where x is the number of reactions you really need. Again, there is always pipetting error and even with the master mix, you will end up short on volume if you only use ‘just enough’. Once the master mix is prepared and gently mixed (very important in this scenario or some of your reactions to do not get enzyme), pipet out equal amounts of your master mix into individual reactions. In this case it is sort of OK to break rule #1. Usually the only difference among your many samples is the DNA or other reagent that’s not the enzyme. In this case, add the enzyme second-to-last, divide the master mix among individual reactions and then add the variable reagent. This variable reagent is usually such a small fraction of the total reaction volume that your enzyme is still under safe buffer conditions. Feel free to come up with your own ‘lucky number’ for scaling up your reactions with more than enough volume to accommodate the reactions you really need.

Avoid gel purification of DNA fragments if at all possible.

Sometimes you may need to isolate a DNA fragment for molecular cloning purposes. Agarose gel purification is a way of doing that. Companies will sell you easy-to-use kits to do this. In my experience, the sample losses are so great that it’s not even worth it. It is very difficult to obtain the quantities of DNA you need for subsequent steps from gel purification. At that point, you can either resort to faith-based cloning, in which, you can’t see your fragments on analytical gels with the human eye nor with help from the imaging camera, but you use it for ligation any way. Some times this works, but usually you are just disappointed when you check your transformation results. There are other tricks to avoiding this technique and if you are a clever cloner, you can get around having to gel purify anything. I swore it off several years ago and have never looked back. Sometimes, gel purification is unavoidable and I would recommend invoking some kind of Voodoo incantation to help success along.

I’m not convinced that the enzyme ligase does anything, but add it anyway.

Ligase is the enzyme responsible for gluing two desired DNA fragments together. The stitching together of these molecules in vitro is not the most efficient process, but we’ve been told that ligase ultimately seals the deal between our pieces of interest. Let’s just say, I’ve done enough positive and negative control reactions (more on that below) and I’m not sure ligase really does anything. However, I add it anyway because… protocols. Really, at this point in the cloning process, you’ve done so many purely superstitious acts, it doesn’t matter if you do one more. UPDATE: As per the comments below, Yes, ligase really does do something. I’m just convinced that it knows which pieces I want glued together and which ones I don’t. It always seems to favor the ones I don’t want, but just enough so that I still check through quite a few possible clones before I find the right one.

Use restriction enzymes until you have evidence they don’t work anymore.

These enzymes are the molecular scissors that cut DNA. They expensive, but they can last decades past their expiration dates when they are properly taken care of. Y’know, when you keep them cold, don’t vortex them or contaminate them with other reagents. You will also have to keep them in a frost-free freezer so they do not endure the temperature fluctuations of a self-defrosting freezer. So until you have evidence confirming a failed digest, keep using the enzyme.

Do the controls.

This rule is true across all scientific disciplines, but in the case of molecular biology work, it can save you lots of time, headaches and wasted reagents. For anything that you are doing, make sure you do a negative control that you know shouldn’t work and a positive control that should work. If both of these types of reactions give the expected result, then you know how to interpret all of your other samples. If only one or the other or neither of these controls work, then you will have difficulty saying what is going on with your experiment. It usually means there is a problem with user error or some other fatal flaw in your construction plan. You may try to interpret faulty experiments and even hope against hope that your experiment worked, but this relies more on faith and superstition than scientific probabilities.

Phosphatase treat your vector or insert but not both.

In many molecular cloning experiments, you are trying to combine two separate pieces of DNA with one another (a vector and an insert) in a useful way. However, sometimes there’s nothing or not much to stop the ligase enzyme from gluing together the vector with itself or the insert with itself, giving you useless byproducts. This is what the phosphatase enzyme does. It can remove the reactive chemical group from a piece of DNA such that ligase can’t use it. So if you treat one piece of DNA but not the other, it eliminates the possibility that ligase will glue any piece to itself, but instead glue the two pieces together. However, if you ligate both pieces, they all become useless to ligase. Sure, I didn’t think it was doing anything anyway, but now those reactions are guaranteed not to work.

Selection > Screen > Hope

When you are trying to get organisms to produce the DNA construct that you have engineered, it is better to select than to screen and better to screen than hope with blind faith. When you are selecting for a construct, all of the cells with the wrong thing will die and only cells with the right DNA will live. Thus, anything living at the end of that experiment is likely to be correct. When you can’t do this, there are ways of screening either based on color or replica plating onto a special medium. The most widely-used example is blue/white screening. If your bacterial colonies are white, they have the correct DNA. If they are blue, they do not. This color gives you a visual clue as to which colonies are most likely to give a positive result based on another experiment. If you can do neither of these things and can only hope to find the correct clone in a plate with hundreds of colonies to choose from, then you have more work ahead of you. In any event, doing the controls is still important. In the case where neither selection nor screening is possible, it may not even be worth looking through the colonies with subsequent experiments to verify a positive DNA sequence. There may just be too many false clones to sort through. Nevertheless, you may try it anyway. 99% of the time you will just end up wasting reagents and time. There does exist a possibility that the correct one can be found, and if you find it, you should probably buy a lottery ticket on the way home.

Sequence your finished construct to verify it is error free.

Once you think you are done with piecing together the DNA sequences you need, you will want to perform experiments to verify that the pieces have come together as you intended. Sure, you’ll cut them again with enzymes and run them on a gel to make sure it looks as expected, but you need to sequence the DNA to make sure that no point mutations have been introduced somewhere along the line. The enzymes responsible for copying the DNA pieces along the way have error-checking features to maintain sequence integrity, but the course of a typical molecular biology project will involve such a length as to make a sequencing error a formal possibility. To assume correctness is just hubris. Sequence it to make sure. Data always trumps assumptions, and that’s no superstition no matter what your scientific discipline.

Make freezer stocks.

Once you have your precious construct and are sure that the sequence is error free, you will want to get to work on your exciting new experiments so you can get groundbreaking results, publish a paper in a high impact journal, secure your own funding, get a job offer at a top research institution, win the Nobel Prize and ride off into the sunset on a unicorn.* But first, YOU MUST MAKE A FREEZER STOCK OF THE CELLS CONTAINING YOUR PRECIOUS DNA CONSTRUCT. Seriously, if you don’t, I’ll make you walk the plank. Make several tubes; label them with an identifiable name (pNobel2015), include the date and any antibiotic resistances. Then write all that shit down in your notebook. For extra credit, generate a graphic map which notes the important features of the sequence. Put the freezer stock tubes in your own freezer box as well as the lab repository. If you don’t do it now, there is a high probability you will forget to do it. You may have to kiss your Nobel prize goodbye if your peer reviewers ask for additional experiments that may require you to go back and use the DNA and your cells are too old to resurrect. Well, maybe not, but I guarantee there will be some poor graduate student or postdoc will carry on the torch of your research several years after you made the construct and it is nowhere to be found. Thanks to you slacker, they’ll have to remake the whole thing from scratch. You did write down your primer sequences, right?** Just save the world a lot of trouble and make the freezer stock.

I’m sure there are many more guidelines that other cloners could add. Maybe we should all just resort to Gibson cloning methods now anyway. Feel free to add your own guidelines in the comments section.


*Just me? OK then.

**So help me God, if you didn’t… Chickens are being sacrificed to empower Voodoo dolls of you to exact vengeance.

6 thoughts on “The Molecular Biology Code

  1. ihstreet

    Really good list. The Master mix x+1 is a good rule of thumb, but found that x+5 (or more) is sometimes necessary with bubbly buffers (those w/ detergent in them; that I don’t run across often, but always do with qPCR). And these days, there are a lot of alternatives to ligase based cloning, though I still swear by NEB DNA ligase…it just seems to work better. I’m fairly convinced that ligase fails because it requires ATP and that is not easily kept stable. Of course, you may well be right that it actually doesn’t do very much.

  2. Tom Ellis (@DrTomEllis)

    Ligase is not required if you have good R.E. overhangs, E.coli will repair the nicks of plasmids once transformed. It certainly does help though, and is essential for blunt end or short A/T-rich overhang cloning.

  3. Pingback: The Rules of Biochemistry | New Under The Sun Blog

  4. EconomicDisconnect

    Nice list but you are nuts to think ligase does not work. T4 High concentration ligase works unreal and I use it often for genomic library construction. Try the Zymo Gel Purification kit, it helps with the low recovery end of things, or low melt agarose. Also Gibson assembly is cool but when it does not work (often) you can’t really trouble shoot it.

  5. johnnaroose Post author

    Oh, I always use it. I did not mean to besmirch the honor of ligase. My tongue-in-cheek remark about it in the code probably says more about the way I handle my other cloning steps. Lately, I’ve had some cloning steps that must rely on hope (and the fact that I have done everything I could in previous steps to minimize unwanted reactions). Ligase seems to always know which pieces I don’t want put together and proceeds to seal them up anyway. I always do the negative control to see how much background I have to work against. I may have cursed the ligase, but the real culprit in my most recently failed cloning attempt was a restriction enzyme that was finally dead. RIP BclI.

    I’ve been turned off to gel purification for so long now, maybe the new kits are better at recovering decent amounts of DNA. Thanks for the tip on the Zymo kit.

    I’ve never tried Gibson cloning personally. I did one completely in vitro protocol once to anneal three pieces of DNA together (each with complementary ends) and it worked beautifully. It never worked for me again in any other context. That was the lab joke/superstition- it worked for each lab member only once and never again.

  6. Pingback: 2nd Blog-iversary | New Under The Sun Blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s