I am cloning five genes of approx. 1.2 kb each into pET-28a(+) vector (separately, i.e., not all in the same vector) by the USER method. The protocol is essentially the same for all constructs, varying only in the template and primers. One of them I successfully cloned from genomic DNA. The other four templates were codon-optimized and synthesized. Two of them were successfully cloned. The remaining two appeared to be properly amplified with Phusion U Hot Start polymerase, since they produced a band with the expected size in the 1% agarose gel electrophoresis. After ligation with the USER enzyme and transformation into E. coli DH5a, I performed the colony PCR for two colonies of each transformation, but they all resulted in bands of 500 bp long, while the expected size was approx. 1190 + 345 of the primers and flanking region. In other words, the products were 3x smaller than expected. No other band was observed in the gel.
For one of the genes, the primers used for cloning were slightly modified, but this didn't change the result much.
I couldn't find any region in the predicted construct with considerable similarity to any of the primers used in the colony PCR.
What could explain these results?
There may not be another primer binding site but I wonder if there are 2 regions of complementary sequence so that once in the vector the insert forms a loop of about 1000 bases and the polymerase reads across the neck of the loop producing a smaller pcr product. It might be worth purifying a colony plasmid dna and doing a restriction digest to see the size of the insert or possibly even cutting the plasmid to change the shape of the molecule then amplifying off the linear dna
In addition to Paul's suggestion that you miniprep and analyze a few to be sure it isn't just a colony PCR artifact, be sure that when you codon optimized you didn't accidentally introduce the restriction site for one of your enzymes! It is easy to forget to check for that.
I sent both samples for Sanger sequencing using both forward and reverse primers that I used for the colony PCR. For both genes, the forward sequencing confirmed 30-ish bp from the start of the genes, and about 140 from the end. Everything in the middle of the genes disappeared, except for these ~170 bases that covered both extremities. The reverse primer covered exactly the same parts for one gene, while for the other it sequenced a region of the vector, far away from the gene.
Summing those numbers up with the size of the primers and everything between primers and the gene, gives about 500 bp, which matches the size of the bands I see in the gel.
It may be that there simply was a deletion. A colleague mentioned something about transposons and bacteria possessing some defense mechanism that recognizes certain DNA sequences, but I don't know much about it. Interestingly, both genes have one thing in common: right after the last successfully sequenced base in the gene, there is a region of 85% GC (14 bases in one gene, 20 in the other), though they occur on opposite ends on each gene. I suppose this GC-rich regions may be related to the deletion, or this loop formation that Paul mentioned.
I'm running some extra colony PCRs right now. Let's see if one of them gives the desired product.
Hi,
This is correct, sometimes a precise deletion of insert is observed. You may screen more clones or try to change the cloning host. There are some strains available where the insert are much more stable then regular cloning hosts.
you might try some of the ideas for high GC . Some companies sell polymerases from ultra hot environments which can cope with high GC like Accuprime because their genome is very high GC. Simple tries are dmso up to 10% final concentartion or betaine up to 1Molar. Raise the melting temperature often helps. Using base analogues like 7-deaza-2′-deoxyguanosine at 25% with your G base can work
Hi all,
I did 6 extra colony PCRs per gene and ran them in the gel. For one gene I got one colony with the desired product, and for the other, I got two. It seems like the problem has been solved. The most likely cause was a deletion made by most E. coli cells during transformation, which clearly doesn't occur for all cells since some colonies carried the correct insert. However, it is out of the scope of my project to investigate this further.
So, if anyone runs into the same issue as me, just try to perform more colony PCRs. As a colleague said: "This usually solves the problem".
I hope this information can help someone.
Best regards,
Gustavo
My default colony screen was about 10 and I rarely saw 100% of the desired clone. I would always recommend screening a variety of colonies - and look for morphological differences to include any that appear unique.
There are several possible explanations for the unexpected shorter products in your colony PCR:
To troubleshoot the issue, you could try optimizing the PCR conditions, checking the template DNA quality and concentration, and performing the PCR with different primer pairs. Additionally, you could consider sequencing the PCR product to confirm its identity and sequence.
These video playlists might be helpful to you:
https://www.youtube.com/playlist?list=PLzBiAPKi8vOPbE7Db9mKwloe34E8rXXEU
https://www.youtube.com/playlist?list=PLzBiAPKi8vONSqjcweFWHMRy13y3484bE
https://www.youtube.com/playlist?list=PLzBiAPKi8vOMheKwb0mu1dcZRaeGUFSlw
- Badges
- Science topic