Posts Tagged ‘gene’

Question by David: Trying to make a GFP-actin gene fusion for transfection into fibroblasts. Help!?
Do you know how I might construct my plasmid vector for transfection? I’ve been looking up protocols online but am confused about how I should make my primers and PCR product to be inserted into the plasmid vector already containing GFP…. do I insert the gene of interest (for example… the sequence for actin) into it? Do I select part of the sequence? All of it? I’m sort of stuck with this initial step, though everything after making the plasmid seems to be pretty straightforward. The way I understand it, I’m creating a plasmid that will incorporate into some region of the fibroblast genome, and the cell will express both its existing wildtype actin and the newly incorporated actin-GFP fusion. I’ve done PCR plenty of times, but figuring out the right PCR product to make for this type of assay is entirely new to me.

Best answer:

Answer by bradlepe
I’m not sure how big the cDNA for actin is (beta-actin, presumably?), but doing PCR for cloning is kind of a pain (since the polymerases used in PCR tend to be fairly error-prone compared to using bacteria to amplify a plasmid, for example). If you do end up using PCR, make sure you use a high-fidelity polymerase (Vent or Deep Vent are, but New England Biolabs sells one called Phusion and it’s really good — it’s worked every time for me without altering the vendor’s protocol at all and the mutation rates are extremely small).

If you already have a plasmid that has the gene for actin, I would probably just use restriction enzymes to cut out the actin cDNA and clone it into a GFP fusion vector (you just have to make sure the GFP and actin cDNAs are in-frame with each other).

Otherwise, you’d want to amplify actin cDNA by PCR. This is a lot more work, however.

Your best bet would be to harvest RNA from a cell (probably whatever cell type you’re eventually going to transfect later), do first-strand (cDNA) synthesis to get a DNA/RNA hybrid duplex (since you can’t do PCR on a single-stranded RNA or DNA, obviously) — for this, it’s probably best that you use oligo dT primers; if you just use random hexamers, you may not get any full-length template.

Next, you can do one of two things: direct amplification of actin cDNA or nested PCR followed by amplification of just actin cDNA. Since actin mRNA is very abundant, I’d probably just try to the direct strategy and if it goes wrong, try nested PCR.

To design primers, you’ll need to look up actin’s sequence for the appropriate species online (this site is useful: http://genome.ucsc.edu/). You want the primers to include the entire ORF. The forward primer should start at the translational start site — ATG — and continue on into the first several codons of the ORF. For the reverse primer, it depends on where the GFP is with respect to actin; if the GFP is 3′ to the actin cDNA, you want to reverse primer to stop just before the stop codon, otherwise it’s OK to include the stop codon in the primer.

The primers should also contain restriction enzyme sites (that don’t need to be complementary to target sequence) so it’s easy to sub-clone the PCR product later.

If you need to do nested PCR, you’ll have to design primers that are in actin mRNA’s 5′ and 3′ UTRs. Do a primary PCR reaction with them (using the DNA/RNA hybrid duplexes as before as the template) and then a secondary PCR reaction with the primers used for just the actin ORF (using the primary PCR reaction as the template).

In any case, once you’ve amplified actin cDNA, I’d suggest sub-cloning into a cloning vector designed for use with PCR products (something like pGEM-T from Promega). That way, you can easily screen (by a blue/white screening) plasmids that got a PCR fragment inserted and then you can take them and sequence the insert to make sure (1) that the sequence is actin cDNA and (2) that there are no mutations (except silent mutations, obviously). Once you’ve found a clone that has no mutations, you can then cut it out of pGEM-T with the restriction enzymes you engineered into the primer sequences and subclone the cDNA fragment into the GFP fusion vector.

It’s a lot of work, but it can be done. I had to make a bunch of cyclin-GFP/RFP fusion constructs a few years ago; it took a few months and a considerable amount of frustration, but you’ll get it if you keep trying. If you have any follow-up questions, feel free to email/PM me.

Know better? Leave your own answer in the comments!