I would like to know: how many restriction sites does a restriction enzyme use on a DNA molecule? In other words: If a sequence on a plasmid contains the following bases:


and I want to have only the bases in bold as a new, separate molecule, will I need two restriction enzymes (One to cut the molecule between the T and G and another one to cut it between the T and C)? Or one restriction enzyme (that will cut them both)?

It wasn't clear enough in Wikipedia and in my biology book.

  • 2
    $\begingroup$ Welcome to Biology SE. Do I understand correctly that you wish to remove the bolded-out sequence, i.e., blunt-end cuts resulting in a oligo-nucleotide of 5 base pairs and the rest? The plasmid question - please google... $\endgroup$ – AliceD May 27 '15 at 14:43
  • $\begingroup$ Yes. I have a question about a plasmide being cut by only one restriction digest, and I want to know if it will remain the same size... $\endgroup$ – Benjli May 27 '15 at 15:00
  • $\begingroup$ This doesn't make sense... Your comment is a different question altogether. Please be specific in your question. $\endgroup$ – AliceD May 27 '15 at 15:02
  • 1
    $\begingroup$ I have a plasmide (as much as I know - a round form of DNA sequence in prokaryotes). One restriction digest is cutting it - in one or two places. If it will be cut in just one place - the plasmide's size won't change, it will just open, beacuse it's round. If it will be cut in two places - there will be a cut plasmide and a new oligo-nucleotide. $\endgroup$ – Benjli May 27 '15 at 15:05
  • 1
    $\begingroup$ please add the restriction enzyme to your question, that would help. It is all too abstract $\endgroup$ – AliceD May 27 '15 at 15:08

Most restriction enzymes have a 6bp restriction site (some have 8bp site also). So two restriction sites generally have to span 12bp. However some restriction sites can overlap. For example BamHI and SmaI:


In this case the total length has reduced to 10bp. However such combinations are not that common and at max you will achieve a 2-3bp overlap. A restriction endonuclease only makes a single cut. To make double cuts you need two restriction sites.

Even if the sites are non-overlapping there has to be a minimum length of separation otherwise the binding of one enzyme will hinder the binding of the other enzyme in a double digest. A 6 nucleotide gap between the restriction sites should be fine. If there is no gap then double digest may not be very efficient.

You can also have two restriction sites of the same enzyme so that you need not use two separate enzymes.

BTW the sequence in your question does not have any restriction site.

Hope this answers your question.


Just to bring another light to the question: a restriction enzyme cuts a double-stranded DNA two times, but it only cuts it once on each strand.

So yes, you would need two different restriction enzymes to excise the sequence in bold (assuming you want that sequence in double-stranded-form): one enzyme that would cut between the T and the G on the strand that you have represented, and one that would cut between the T and the C on this same strand.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.