One of my favorite activities is talking about enzyme activities. But one of my least favorite activities is tracking down what people are using to define their enzyme’s specific activity. So, biochemists – I call upon you to unite against undefined units! I hope “U” will permit a little rant about a pet peeve of mine that’s peeving me that has to do with enzyme activity. Say you want to tile a floor – so you buy a bunch of tiles but you need help laying them down. So you want to hire some tilers. But how many to hire? Depends on how many tiles you need to lay, how quickly you want the job done, & how quickly the tilers tile.

note: refreshed & video added 2/7/22. note: 17:43 should say 69.7 kg/mol – sorry!

So you check out the ads. One company says that their tilers can tile a 100 sq. ft room in 1 hour. Another tiler claims he can lay down 100 tiles in 1 hour. And another company says they’ll send you 3 workers for the price of 1. What’s the better deal? We need more info!!!!!! Even if they are all telling the truth, there are still a lot of unanswered questions. Here are some questions I’d like to ask them…

Company 1: How big are those tiles you used? The tiler could have laid down 1 giant tile – 10 ft by 10ft – and have filled the whole room. But what if you have smaller tiles? If he can only lay 1 tile an hour it’s gonna take a long time to tile the same size room.

In that case, the second tile company’s looking a lot better! But how many tilers will they send? The more tilers there are working, the faster the job can get done – assuming that all the workers are working at that same rate…

How about that 3 for 1 deal? Are all 3 tilers of equal tile-laying-skillness? What if they send you 1 tiler and 2 dog-walkers? Even if that tiler’s pretty good, those dog-walkers aren’t gonna be much help. 

Just as we need more information about the tilers before deciding which to use and how many to hire, we need more information about the enzyme “workers” we use before we decide how much to use. 

Enzymes are biochemical reaction speeder-uppers (catalysts). They’re usually proteins, sometimes RNA or protein/RNA complexes. They bring together reactants, hold them in prime position for reacting, etc. to do things like get DNA letters or protein letters to link up (polymerization) or break them apart or move them, etc.

Just like the tiler doesn’t do the actual tile-sticking, just takes it to the grouted slot, gets it oriented correctly, and lets the grout do the actual work, enzymes speed up reactions but they don’t do the actual reacting part. It’s just the mediator – so it doesn’t get used up – and can keep doing it over and over

But different tilers are more efficient than others. So instead of just knowing how many tilers we have we often care more about how good they are. One way we can report on an enzyme’s goodness is in terms of the enzyme activity unit, or simply unit, which gets abbreviated as a capital U. 

It tells you how much of that enzyme is required to do a certain amount of stuff in a certain amount of time. (e.g. how many tilers from each company would you need to lay down 100 tiles in 1 hour. eg. maybe you’d need 100 of the company A tilers but only 50 of the company B tilers.)

“1 enzyme unit” is often the amount of enzyme needed to convert 1 umol substrate/min or 1 nmol/min – but it can be whatever the heck someone decides it is – and that’s fine – they just have to tell you what they’re defining a unit as! And you have to actually go look at the definition. (more on the mol in a moment, but for now just know it’s a way to report how many copies of a molecule there are)

And that gives you an idea about how “good” the enzyme is. If, for example, less enzyme A is needed to hit that 1 U mark than enzyme B, or, put another way, if the same “amount” of enzyme A has more activity units than that amount of B, then enzyme is “better” than enzyme A.

But better at what? Not all enzymes are “tilers” – and just like you wouldn’t judge a demolisher by how well it lays tiles, you wouldn’t judge a protease (protein cutter) by how well it sticks together DNA letters or even how well it cuts DNA. 

So U is specific to each type of worker. And not just the type of worker but the actual task it’s doing. And not just the actual task but working conditions. Just like it wouldn’t be fair to compare the tile-laying rates of a worker laying big tiles to one laying small tiles, it also wouldn’t be fair to compare a worker working in a walk-in freezer to one working on a beach. 

Speaking of which, enzymes often are more active at warmer temperatures where they have more energy (but not so warm that they unfold!) But a lot of the time we work with proteins we want to keep the proteins we’re working on cold (in part to protect them from unwanted activity of contaminating enzymes… And sometimes the conditions that the companies measure activity are at warmer temperatures, so if you want to use it in the cold room, on ice, etc. expect them to work slower (and still hopefully work).

So “enzyme activity” tells us about the “goodness” of the individual workers at doing the task – but also the concentration, because it’s commonly given as U/mL or U/μL or something like that. Because the # of tilers matters!

But U/mL etc. doesn’t tell you about the # of tilers! Just about the output ability contained in a certain volume. You could have a little bit of super-active enzyme in there or a lot of not-so-active enzyme…

That’s where the “specific enzyme activity” (aka specific activity) comes in. It says if you send me a certain amount of workers (defined by mass), how many tiles can they lay in an hour (or the enzymatic equivalent). 

note: we’re still not talking in terms of actual # of tilers because we’re discussing mass not mol, but often that’s because it’s hard to know true molarity

note: the concept of “enzyme activity” can get really confusing because although when talking about units and stuff we define it as U/volume, but when we’re thinking about individual enzyme molecules, we can talk about enzymes being more or less active. So, it doesn’t really seem fair to the enzyme that you can half its activity by diluting it 2-fold but, you can… See it all has to do with how we define “amount of enzyme.” With “enzyme activity” we talk in terms of volume but with “specific activity” we talk in terms of mass. But in day-to-day talk we often talk about “activity” in more qualitative terms.

The specific enzyme activity (specific activity): (# of enzyme units per mL)/concentration of protein (mg/mL) -> units/mg or mol/min/mg

Specific activity increases if you have better tilers, but even if you have really good tilers if most of the workers you send aren’t tilers but instead you have dog-walkers, demolishers, remodelers, etc. in the crew the company sends, specific activity will be lower.

With proteins, our “other workers” are other contaminating proteins that add to the total mass but don’t help do our task. A really pure protein company will only send you tilers. They “weed out” the other workers using things like protein chromatography (which separates proteins based on differences in how they interact with little beads (resin) in a column you send them traveling through)- as the purification progresses, the specific activity will increase, even if the “enzyme activity” (goodness of an individual tiler) doesn’t change. So specific activity is good for measuring enzyme purity 

Say you did a serial dilution of an enzyme and you measured its activity at different concentrations

  • the specific activity should be the same since both the numerator & denominator take the concentration into account
  • the enzyme activity will be different because you have fewer units per ml

Specific activity is also good for comparing preparations of the same enzyme. But it’s important that you use the same conditions. Unlike with chemistry gas law stuff where we have really strict “standard conditions” called standard temperature and pressure (STP): 0°C & 1 atm. In biochemistry, “standard conditions” can vary from lab to lab, so it’s important to be as specific as possible when reporting specific activity – and enzyme activity

even so, measurements may vary slightly from lab to lab and batches are checked internally (i.e. when a lab makes a new batch, they compare it to the old batch) and if you decide to make your own, you want to test it against the commercial stuff rather than just going by what the label says it is. 

Which brings me to a little rant… One place you’ll find enzyme activity listed is on tubes of endoproteases, which are proteins that recognize & cut specific “code words” in proteins. And we can stick that code word between the end of our protein and the beginning of an affinity purification tag to untag our protein after we use the tag to bind a column to help us purify it.  More here:

One of the awesome things about being a protein biochemist is you can save a lot of money on enzymes you use a lot and that other people usually view as “reagents” that they buy by expressing & purifying it yourself and cutting out the significantly-swipey middleman. 

A few months ago, I purified TEV Protease. It recognizes the sequence ENLYFQ(G/S) and cuts between the Q & G(or S). So I stick that sequence into a linker connecting my protein and a tag – I use a strep-SUMO tag which has a StrepTag sequence that mimics biotin and binds to a modified srreptavadin-coated resin, and a SUMO “fusion partner.” Other common fusion partners include GST & MBP & they can help with expression and solubility. more here:

We have to purify it fairly frequently because we purify a lot of protein. And, while for us it’s most useful to know how each prep compares to the one we had been using previously, I was curious in how my prep’s activity compared to that of the stuff you can buy commercially. Which brought me to a big pet peeve of mine…

One thing that bugs me is that often the data sheet will give information about enzyme activity in terms of mass (g, mg, ug, etc.) of substrate (note –  we often use “mass” and “weight” interchangeably, but technically weight takes into account the extra “push on the scale” from gravity, but mass doesn’t). 

Proteins can range greatly in mass depending on how many (and which) amino acids (protein letters) they have. If you have a big protein, the same #g will have fewer actual protein molecules (like that tiler laying down that one big tile), whereas if you have a small protein, a small amount gram-wise can have a large number of protein copies (and a large number of tags needing to be cut off!) And the enzymes can have different sizes too. But whether your tiler’s wearing a hat or not, he can only lay one tile at a time.

So to know how good the tiler really is, you  have to go look up the molar mass of the protein and convert. Because what you care about is how many tags there are to cleave off. And there’s only 1 tag per protein, regardless of how big that protein is. 

A mol is like the biochemist’s “dozen” – it just means Avogadro’s number (6.02×10²³) of something. And the molar mass is how much 1 mol of something weighs. Usually for proteins it’s reported in Daltons (Da) or kiloDaltons (kDa). 1 Da = 1g/mol. So if a protein has a molar mass of 40,000 Da, 1 mol of the protein weights 40,000g (which is the same as 40kg). Alternatively we can write its molar mass as 40 kDa.

That’s a pretty average size for a protein – the average protein is around 300-400 amino acids long (according to and the average weight of an amino acid is ~100 Da). So if you had a 400-amino-acid-long protein, it would have a molecular weight of ~ 400 aa *100 Da/aa = 40,000 Da, or 40 kDa. If you know how long a protein is, you can do one of these back-of-the-envelope estimates, but since each amino acid has a different weight and a protein might have gabobbles of glycines (tiny) or a ton of tryptophan (big), the exact molar mass will depend on their exact sequence and you can use free websites like ExPasy ProtParam to help you get a more accurate number.

But anyways, back to the tale of TEV – You can buy it from New England BioLabs, and they sell it at 10,000 units/ml. What do they define as a unit? Good question – so you look at the product details & it tells you:  “1 unit of TEV Protease will cleave 2 µg of MBP-fusion protein, MBP5-TEV-paramyosin ΔSal, to 95% completion in a total reaction volume of 10 µl in 1 hour at 30°C in 50 mM Tris-HCl (pH 7.5 @ 25°C) with 0.5 mM EDTA and 1 mM DTT.”

They tell you what the substrate is (that MBP-fusion protein), but they tell you how much substrate is cleaved in terms of μg (micrograms (aka millionth-of-a-grams). So you can go look up that protein, find out its molecular weight and then determine the activity. And it’s good that they tell you the name of the substrate because not all proteins are the same “easiness” to cleave – sometimes the tag is hidden, etc.

NEB actually tells you what their substrate is but Sigma defines a unit of TEV as: “One unit of TEV protease cleaves >85% of 3 μg of control substrate in one hour at pH 8.0 at 30 °C.” They get a huge bumbling biochemist side eye for that…

And NEB might have gotten a bumbling biochemist thumbs up for their TEV reporting, but they made me do a bit of digging for EcoRI, a common restriction enzyme – restriction enzymes (site-specific DNA endonucleases) – somewhat analogously to endoproteases like TEV, these recognize and cut specific “code words” in DNA and they’re really useful for things like molecular cloning, where you cut DNA out of one place and stick it somewhere else (such as sticking protein-making instructions into a circular piece of DNA called a plasmid that you can stick in bacteria to have them make protein for you) & analytical digests, where you see if a specific DNA sequence is present based on whether (or how many) a cut gets made.

According to NEB’s EcoRI brochure, “One unit is defined as the amount of enzyme required to digest 1 µg of λ DNA in 1 hour at 37°C in a total reaction volume of 50 µl.” And the same unit definition is given for HindIII. (λ stands for lambda phage, a bacteria-infecting virus). But just like proteins can have different molecular weights, so can DNA. And it can have different numbers of cut sites. I had to look elsewhere to find out that λ DNA has 5 EcoRI cut sites – and it has 7 for HindIII. So HindIII has to do more work per unit, but they cost the same so HindIII kinda gets a raw deal… (though the biochemist gets a steal!)

Another note: when there are lot-to-lot variations you can use the batch # printed on the tube to look up the Certificate of Analysis (COA)

more on topics mentioned (& others) #365DaysOfScience All (with topics listed) 👉 

Leave a Reply

Your email address will not be published.