MadSci Network: Molecular Biology |
Your question is intriguing. Although cloning vectors clearly have to come from _somewhere_, practically speaking, I've usually seen people work with commercially available vectors rather than construct them from scratch. We need to tailor our vector to our host species. If we're making the vector because we're dealing with a new host species, we may face even more difficulties. Step one: First, we need to make a plasmid that contains a host replicon, which is the origin of replication initiation (ORI) and any other sequences necessary for replication. This is our bare-bones requirement; a plasmid will replicate only with the needed ORI. There may be multiple ORIs to choose from. If I recall correctly, the original isolation of an ORI was done by randomly cutting bacterial genomic DNA into fragments, ligating those fragments to DNA that carried a selection marker, and transforming host bacteria with the resulting circular plasmids. Bacteria that survived on selective medium over several generations had to be carrying plasmids with the marker, which implied that they were replicating the plasmid they received. The ligated-in sequences that conferred survival were then cut down until a minimal sequence needed for replication was isolated. We know the replicon sequences for E. coli and other common hosts, but if we want to make a cloning vector for a new host, we'll need to either repeat the above experiment or find a shortcut that lets us quickly find the needed replicon sequences. You mentioned a bioinformatics approach in your letter; this might work if the genome of the host is known and it contains a replicon that has sequence homology to a known replicon. Then the candidate replicon sequences could be identified, PCR-amplified, and put into a marker-carrying plasmid. Step two: Once we have a streamlined plasmid with a selection marker (amp- resistance, etc.) and a replicon, we need to introduce multiple cloning sites, which, ideally, should all be unique. Our sites should be rare enough that they occur only once in the entire plasmid, but common enough that we can work with them conveniently. Three solutions come to mind: 1) Cut the polycloning sites out of an established vector and ligate them into ours. 2) Cut out our replicon region and replace the replicon region in a known vector. We might have to use site-directed mutagenesis to make restriction sites with which to do this. The problem with these first two solutions is that our restriction sites might no longer be unique, since we have a different replicon. We could just use the ones that remain unique, though, and ignore the others. 3) Construct a new multicloning region from scratch by using complementary oligomers to synthesize DNA that carries a series of unique restriction sites. Since this region may be 70 or more bp long, we may need to make multiple sticky-ended fragments and stitch them together. [Bioinformatic relevance: Several computer programs for listing or mapping restriction sites on plasmids are available; a quick Google search should give a few alternatives. This would help choose between possible polycloning sites to incorporate (methods 1 and 2) by letting us pick the ones with the largest number of unique sites. If we make our multicloning region from scratch, then we can use a similar search to find the restriction sites that are absent from our plasmid and can therefore be made in the oligomer.] Step three: Now that we have a workable plasmid, we can insert useful additional sequences, like bacteriophage promoters or reporters. If we've used the second method in step 2, where we swap in our replicon, then this step has already been completed for us. Otherwise, we need to find sites in our plasmid to insert these elements. For reporters, we can just use the endmost restriction sites in our polycloning region. For other elements, we may need to use site-directed mutagenesis to make sites. I hope that my response has been helpful; I'm pretty new at this. Please do not hesitate to contact me at my e-mail address below if you have any questions or comments. REFERENCE: Sambrook, Fritsch, and Maniatis. (1989) "Development of plasmid cloning vectors." In: Molecular Cloning: A Laboratory Manual. 2nd ed. Cold Spring Harbor Laboratory Press, New York. pp. 1.7-1.9 Paul Nagami nagami(AT)its.caltech.edu [address mangled to avoid spam robots.] Undergraduate California Institute of Technology
Try the links in the MadSci Library for more information on Molecular Biology.