- Is the system working at all? Include a positive control in your expression experiment (i.e. an expression plasmid based on the same vector you're using for hCG, coding for a protein that you know is well expressed under your conditions). If it does not work either, you know it's a problem with your culture conditions, or induction, or analysis.

- If you codon-optimized the construct, there's a good chance you changed the efficiency of translation initiation signals (not necessarily for the better). Compare the efficiency of your codon-optimized plasmid with that of the non-codon-optimized using something like the RBS Calculator at https://salis.psu.edu/software/

cheers,

alejandro

Francesca,

I completely agree regarding the RBS, that is rapidly checked in silico, after that I would consider a few easy controls:

- check the O.D. over time of an induced culture and an non induced, there should be some difference, otherwise

- no difference? Is the selection working properly or (same effect) may your system be leaky, i.e. clones may either loose plasmids, check/compare Ap titer, with the optimized gene the level of the uninduced state may already be harmful to E.c.; alternatively the induction may have failed, since anyway, turning a T7 promotor on means stress to the cell, with an without protein expression there must be a difference to the uninduced culture.

- did your gel loading buffer work, i.e. maybe the inclusion bodies did not dissolve -> clean bands on the gel

Final suggestion, the publication is already giving excellent results, why did you optimize codons at all?

Regards, Michael

If subunits alfa abd beta interact, you could try to purify them both together in plasmids like the pDuet that produces the same amount of both proteins. This has worked well for us when we expressed two subunits of a complex that separatedly expressed tended to go into the inclusion body fraction and expressed together were soluble. If you only want to test activity try to make a translational fusion to a good soluble TAG suchas Maltose Binding Protein or GST and test if its more soluble then. It usually makes the trick. Good luck, good hands and good buffers, you always need them when purifying proteins.

When expressed in E. coli, hCGbeta is not glycosylated and should have a molecular size around 15-16 kD.

I managed to get some expression but I could see it only on a Western blotting, not SDS page... and from a high copy number plasmid, not low copy one.

I checked the rbs strength with RBS calculator but that did not made any difference... even with strong combination I could see it only from the high copy on western.

So, I think it is a matter of expression level.

Thanks all for the help.

Francesca