I am aware that each enzyme generate a certain amount of misproducts. This is well documented, for example, for the DNA polymerase.

I am interested in enzyme involved in biochemical processes, so for example methytransferase, oxidoreductase, carboxylase and so on. I have been looking for any reference that could report the error rate or the amount of misproduct for any of these enzymes, but I was not successful.

Does anyone know a reference reporting the error rate, or any other information about the misproduct generate by these enzymes?


I will try to clarify a bit better my question. Let's assume that we are considering an enzyme that add a methyl group to a molecule, we can choose propionic acid for example. In theory the correct product would derive from the addition of a methyl to the C3. For the biochemistry study we know that the activation energy is decreased by an enzyme for example by orienting the molecules in the right way to react with the second molecule. Therefore, if the enzyme in question will not be able to orient the molecule in the right way there is a certain probability that there will be a mis-product. For example the methyl group could be added to the C2 and not the C3. Now. the probability for this process is quite low, but it cannot be excluded from happening. My question is, if there is anything known about this type of error rates in the different type of enzymes.

I hope this make things a bit more clear


Lets consider the imagine below that was taken from here

enter image description here

In B you can see the interaction of the active site of the enzyme with a substrate (for the example it does not matter if it is the real substrate or an inhibitor). I would say that among al possible interaction that this active site can make with substrates there are some much more likely than others. An hypothetical chart could be created considering the free energy of the complex enzyme-substrate, higher the free energy lower the stability. Therefore among all possible enzyme-substrate combination we will have just few possibilities that have a low free energy. In the first place we will find the "correct" substrate, then we will find eventual competitors and other molecules that have similar chemical features as the substrate. Even though with low probability these interaction, enzyme-competitors, will happen. This might led to the formation of "unwanted product" that I refereed to in my question as "misproduct".

I am interested in reading some paper that are considering thie phenomenon and also quantify its rate. I am also curious if this phenomenon can happen with different rates in different enzymes.

  • 3
    $\begingroup$ I think these enzymes are different. There is one chemical reaction that they catalyze and thus this will take place or not. This is different for the DNA polymerase where 4 different nucleotides can be incorporated (which raises the possibility for some errors). $\endgroup$
    – Chris
    Commented Feb 15, 2015 at 21:24
  • $\begingroup$ There can certainly be enzymes other than polymerases that catalyze errant reactions. Aminoacyl tRNA synthetases immediately come to mind as an example. Some even have proofreading mechanisms to correct errors. $\endgroup$
    – canadianer
    Commented Feb 22, 2015 at 17:17
  • $\begingroup$ DNA pol only appear different because enzyme is influenced by template strand, in a sense. Chemical reaction catalyzed is still a chemical reaction $\endgroup$ Commented Feb 22, 2015 at 23:37
  • $\begingroup$ please let us know if you find answers sufficient. I think that question is important $\endgroup$ Commented Feb 23, 2015 at 11:33
  • 1
    $\begingroup$ If you ask references for all the enzyme then I am afraid that this question would be closed as broad. I am giving you one example and the basic idea behind the problem. You can search for the other cases. $\endgroup$
    Commented Mar 1, 2015 at 9:20

4 Answers 4


First it is important to define what actually an error is. What you call misproduct is actually a right product for a different substrate. The enzyme has a low specificity; but you won't say hexokinase is error prone because it has wide substrate specificity. The question is — is specificity critical?

Coming to the example of DNA polymerase, you should note that DNA polymerase only adds a nucleotide to the DNA strand. Here the template DNA strand acts like a co-enzyme and it is the template DNA that provides specificity to the reaction. In this case the specificity is really critical and there are mechanisms that help in ensuring that.

Other enzymes also make errors and the probability of making errors depends on different parameters. Take the example of restriction enzymes that are supposed to cut at a specific site. When this specificity is lost we say that the enzyme is showing a "star activity" i.e. non-specific cleavage. This depends on different parameters such as buffer composition, enzyme concentration etc.

In general specificity is provided because of high affinity of binding towards one molecule compared to the other. These binding reactions are second order — a high affinity substrate will bind to the enzyme even at a low concentration, however if you increase the concentration of the low affinity substrate or the enzyme, then at some point the rate of binding may be significant enough to create a "wrong" product.

Sometimes there are cases where the enzyme can bind to a wrong substrate but cannot create a product. This is the typical case of a competitive inhibitor. It is also possible that a certain molecule is good at binding but the conversion rate is slower.

To conclude, these kinds of "errors" happen with many enzymes but it is very critical in case of DNA polymerization. You can get information on star activity of restriction enzymes (just google it).


Rubisco is an enzyme that is involved in the biochemical process of carbon fixation in photosynthesis, but also has an economically important error (photorespiration) which incorporates the wrong molecule (oxygen) into RuBP, therefore wasting the energy captured by the plant and producing toxic downstream products.

Rubisco's error rate in carbon fixation ranges from approximately 1:10 to 1:80.

enter image description here

  • 1
    $\begingroup$ Thank you very much for your answer. However, the paper you refer to neither contains the information about the error rate nor the table. Did I miss something or you took these data somewhere else? $\endgroup$
    – efrem
    Commented Jun 29, 2015 at 14:28
  • 1
    $\begingroup$ @efrem Rubisco makes an error when it incorporates oxygen instead of carbon, and therefore the error rate is Vc/Vo. $\endgroup$
    – March Ho
    Commented Jun 29, 2015 at 19:50

Chris is right here, polymerases are generating sequences, the enzymes your asking about do a single type of reaction, so you would be assessing its activity not its fidelity. A good resource I've found for looking at qc data for enzymes like these are just the product inserts from the manufacturer. NEB for example generally seems to provide sufficient qc data in my opinion. For example:


  • $\begingroup$ DNA polymerase does single type of reaction too, namely, creates covalent bond between dNTP and DNA, releasing diphospate group $PP_i$ $\endgroup$ Commented Feb 23, 2015 at 11:31

Since chemical reactions are governed by quantum physics, there is a whole lot of products that can come out of mixing together several multi-atom molecules. And all of those will have non-zero probability of occurrence. But this probability is extremely low at normal temperatures, so you don't see those products.

Enzymes lower activation energy and so increase chance of reaction, but only for certain substances and products (spatial position of reagents can influence outcome). In case of DNA replication, when wrong nucleotide is inserted, it happens because enzyme/template nucleotide complex has not enough specificity. All 4 nucleotides have rather high chance of participating in the reaction (but errors are still on order of $10^{-6}$ and lower).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .