Where do restriction enzymes come from?
April 6, 2021 10:51 PM   Subscribe

When doing DNA recombination, how do people select what restriction enzymes to use? Is there like, a catalog of restriction enzymes I can look at and pick the right ones?

I'm trying to understand how insulin is produced via recombinant DNA, but I can't find anything that goes into the level of "here are the actual physical steps that are involved in producing insulin". Particularly mysterious to me is how the restriction enzymes are selected. I'm curious both what restriction enzymes are used to produce insulin (is it the same for all the different types of insulin? is it a trade secret, or something?), how those particular enzymes were decided on, and how those enzymes are actually produced.
posted by wesleyac to Science & Nature (3 answers total)
Best answer: There is indeed a catalog of restriction enzymes.

You select the restriction enzymes based on the sequence you want to cut--traditional restriction enzymes recognize specific sequences and cut at those specific sequences. There are newer types of enzymes called Type IIS restriction enzymes that cleave outside their recognition sites, but let's not get into that.

When you express proteins like insulin in other organisms (heterologous expression, often done in E coli or yeast), the main process is:
1) Cut out the sequence of the gene that encodes the protein (ie, the nucleotides that get transcribed and then translated into amino acids). This part is called the coding sequence.
2) Stick that into a piece of DNA that can be reproduced (either the genome of an organism, or more often, a round piece of DNA called a plasmid). You paste the coding sequence in after a DNA sequence called a promoter, which tells the RNA polymerase gene to transcribe the gene into RNA (which will then get translated by the ribosome).
3) Stick that piece of DNA into the organism that will be expressing your protein.
4) You grow the organism in large batches (this is an oversimplification, sometimes you have specific promoters that only express your gene under certain nutritional/drug conditions) and then harvest your protein of interest.

You need restriction enzymes for both steps 1 and 2. You're somewhat limited by what sequences are present in your gene--you need to select a restriction enzyme that cuts around your coding sequence, but not within. Same for your plasmid and promoter--you want to cut around the parts that you're going to be gluing together, but not cut them up inside. So there are tools like NEB Cutter where you can paste in your DNA sequence and then it will tell you what enzymes cut, and where, and how many times.
posted by katecholamine at 12:19 AM on April 7 [9 favorites]

As far as making recombinant DNA for expressing a protein though, it's gotten a lot cheaper in the past 10 years to just order a very long stretch of DNA that contains your gene of interest plus some bookend sequences capable of recombining into specific target sequences. This has made molecular cloning by cut & pasting with restriction enzymes a lot less common than it used to be. I've broken out our lab's restrictions enzymes only a few times when making plasmids during my molecular biology PhD, but done Gateway cloning every 3-6 months. It's not dependent on having the right restriction sites and fragment sizes available in your sequences of interest, which is super handy.
posted by deludingmyself at 6:44 AM on April 7 [1 favorite]

Here's the sequence for human insulin [genbank]:
If you want to cut this chunk of DNA from the genome to insert it into a plasmid [as explained by katecolamine above] to be grown up in E.coli, then you want to choose a restriction enzyme that will make its specific cuts conveniently upstream and downstream from the working part of the gene. I have highlighted the start and stop codons for the gene. I suspect that the reason why the first 4 bases and the last 4 bases in the sequence are reverse complements to each other is that they are part of the cutting sites of one such restriction site. If you want to clone a different gene you'd run the sequence through NEB cutter or similar, again as katecholamine says. Hope helps.
posted by BobTheScientist at 8:27 AM on April 7 [1 favorite]

« Older What did you put on your wedding registry?   |   Can I complain to Deliveroo UK without the driver... Newer »

You are not logged in, either login or create an account to post comments