I'm planning a gene synthesis project. The goal is to express the cDNA of a human-derived sequence in E.coli for protein production purposes. The company I submitted the sequence to has detected 2 short palindromic sequences of 12bp each (which, as they say, will obviously add some difficulties to them during the synthesis, but they could handle it). As these sequences are within the coding region, I am curious if they might affect the expression. Any ideas?
Hello Wolfgang,
it should be possible to rid of these sequences by using different codons coding the same aminoacids in the desired area.
Good luck!
Christoph Leder
Hello Wolfgang,
it should be possible to rid of these sequences by using different codons coding the same aminoacids in the desired area.
Good luck!
Christoph Leder
Yes, changing the palindromes might have an effect, but it is hard to predict what (better or worse expression levels), if any, and how large. I'd go ahead and change them using synonymous codons, or ask the company to do it for you (most have some sort of gene optimization service incorporating several prediction algorithms; I have good memories of GeneArt, but there are others).
As to how the palindromes might change expression levels, there's a number of (a bit bizarre) possibilities. If they sit close to the 5' end of the gene, changing their sequence might change the secondary structure of the mRNA around the RBS and hence, translational efficiency (See e.g. Pubmed 24072823, 19801975 and 23774758). If the change creates or removes an SD-like sequence, it might cause/eliminate a ribosomal pausing site and also change influence the kinetics of mRNA decay (See Pubmed 22456704). They might be involved in some sort of translational switch at the mRNA level (which will be totally irrelevant once you transplant the CDS to an E. coli expression vector). They might even create a cryptic translational terminator.
Hope it helps.
Hi Wolfgang:
Before you start a big expression project, I recommend you to express the protein using a “Cell-free” Expression System, either Ecoli-based or a eukaryotic-based expression system. By using this way, the protein will be in just one (1) day and then you could solve those doubts that concern you most.
I hope this advice could give you some help.
Good luck,
Arnubio.
Based on experience, I agree with Alejandro. Getting a hairpin can completely ruin your expression, while changing a few codons is much less risky. I would just change a minimal number of codons to whatever is most biased toward E.coli expression. Generally, when making a synthetic gene, I optimize for the species I will express in, then search through and eliminate inconvenient palindromic repeats, poly A regions longer than 5 bases and common restriction sites.
It seems like a lot, but I have found that you end up doing one or more of these steps when expressing a protein in a foreign organism anyway. Getting it out of the way can save a lot of troubleshooting and it makes downstream engineering a lot easier!
Good luck!
I would be concerned to keep the inverted repeats stable in E. coli. If they get deleted, then you may have to switch to E. coli strains that lack, say SbcCD, or other resolvases. Good luck.
I have carried out similar experiments with highly repetetive DNA and the main problem was unusual deletions occurred in the DNA preps from E.coli. Sections of the repeat element in the DNA had been removed. But, by using using a different strain of E. coli (Sure cells), this problem was overcome.. Now for the actual expression, there was never a problem in E. coli expression. - Cheers, Stephen
Dear Wolfgang
As Christoph said it would be an excellent idea to use synonymous codons. However, it may interrupt downstream interactions or intractum.
Good luck,
The presence of short palindromic sequences within a coding mRNA should not represent a major problem in proteins expression, unless they involve the starting codon and the Shine and Dalgarno RBS sequence. In my experienece, hairpin structures involving such regulatory elements may signifiicantly affect the ability of ribosomes to bind to mRNA and frequently are between the most important causes of failure in protein expression. In contrast, internal hairpins are usually easily unwound by the ribosomes.
Keep in mind that mRNA degradation in bacteria usually proceeds in the 3'-5' direction and that stem-loop structures may slow down degradation (3' stabilizing elements are palindromic sequences). This may sometimes allow production of small amounts of truncated proteins.
In any case, unless the stem loop involves the translation initiation region, the easiest thing to do is to try and see if the protein is expressed. Only in case of failure I would turn to changes in codons.
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology