What you are observing is not an uncommon phenomenon. In a lot of cases where you have a very stable, strongly expressed tag protein, this will tend to accumulate in the cell, without your protein of interest. This can either occur because the ribosome stalls after making the tag protein, and the tag alone is released; or because the protein of interest is degraded in the cell, but not the tag protein. The same is commonly observed with tags like maltose binding protein (very common) and NusA. My lab recently undertook a large scale experiment with a set of ten different tags and a nasty protein, and most of the constructs delivered just the tag protein. A couple did give the protein of interest, though.

It would be good sense to do the reality checks that your clone is correct, and that the cells you think are expressing do indeed have the gene in. However, after this, you will probably need to optimise all the parameters of the experiment (growth media, induction temperature, IPTG concentration/autoinduction, cell line, induction time) to get the best results. As a rule, dropping the temperature to 20 C, and rolling the IPTG right back (1 mM is almost always too much: 0.2 M is a lot, and often 10-20 uM is optimal for tricky proteins) are usually good starts.

You should also consider where the protein comes from. Check the number of rare codons and their distribution (http://people.mbi.ucla.edu/sumchan/caltor.html is a good tool) - if you have a lot, or several in a row, this often leads to ribosome stalling. Either switching to Rosetta, or getting the gene synthesised with E. coli codon usage, might help.

Also, if you really have the unusual sequence that you suggest, remember that your protein of interest is going to run really oddly on SDS-PAGE. SDS adds about one negative charge per two amino acids to your protein: if your protein is already highly negatively charged, then the mass to charge ratio may well be highly skewed from the average protein. The result is that your protein will run much faster on the gel than you expect it to.

Strange proteins like this often have interesting biology. Good luck getting it working!

According to your description that the sample which you are calling as induced construct actually has only empty vector. There is no reason otherwise why you can see GST band but not the fusion.

You might want to retransform, check your construct and if necessary sequence your construct

Have you checked if you cloned your GOI in frame?

I agree it is strange that you would only get a GST-tag showing up. Even with an unstable fusion protein you would expect a degradation profile. Also it sounds like a relatively large protein for bacterial over-expression purposes. Assuming it is is in frame you might want to try small scale induction conditions of varying IPTG (0, 0.01 and 0.1mM) as well as different temperatures (20, 25 and 30oC) for comparison (and confirm you are not inadvertanly cleaving the fusion). Also you may want to consider overnight expression (16hrs).

Where does your gene originate from? it may need to be codon optimised for bacterial expression.....

You've to optimize IPTG concentration, growth Temperature and try at different time intervals.

Checkout the promoter strength.

Else, change the host system like BL21 (DE3) pLys or Rosetta.

Use LBG or 2X-YT medium for better result.

Alternative suggestion: Yeast expression system is far better than bacterial system..

All the Best.

What you are observing is not an uncommon phenomenon. In a lot of cases where you have a very stable, strongly expressed tag protein, this will tend to accumulate in the cell, without your protein of interest. This can either occur because the ribosome stalls after making the tag protein, and the tag alone is released; or because the protein of interest is degraded in the cell, but not the tag protein. The same is commonly observed with tags like maltose binding protein (very common) and NusA. My lab recently undertook a large scale experiment with a set of ten different tags and a nasty protein, and most of the constructs delivered just the tag protein. A couple did give the protein of interest, though.

It would be good sense to do the reality checks that your clone is correct, and that the cells you think are expressing do indeed have the gene in. However, after this, you will probably need to optimise all the parameters of the experiment (growth media, induction temperature, IPTG concentration/autoinduction, cell line, induction time) to get the best results. As a rule, dropping the temperature to 20 C, and rolling the IPTG right back (1 mM is almost always too much: 0.2 M is a lot, and often 10-20 uM is optimal for tricky proteins) are usually good starts.

You should also consider where the protein comes from. Check the number of rare codons and their distribution (http://people.mbi.ucla.edu/sumchan/caltor.html is a good tool) - if you have a lot, or several in a row, this often leads to ribosome stalling. Either switching to Rosetta, or getting the gene synthesised with E. coli codon usage, might help.

Also, if you really have the unusual sequence that you suggest, remember that your protein of interest is going to run really oddly on SDS-PAGE. SDS adds about one negative charge per two amino acids to your protein: if your protein is already highly negatively charged, then the mass to charge ratio may well be highly skewed from the average protein. The result is that your protein will run much faster on the gel than you expect it to.

Strange proteins like this often have interesting biology. Good luck getting it working!

Thanks to all for your kind instant ideas and reply

Hi Vanita Uppada

i already checked the construct by sequencing my gene is in-frame and when i did the induction i used the freshly transformed colonies on the medium with respective antibiotics.

Dear Nicholas

i checked the occurrence of rare codons in my gene of interest (a human protein) it contains the rare arginine, glycine, proline, isolucine codons, so i want to try the expression in Rosetta strains. i will also try to optimize the codons for bacterial expression

Hi Prasad Muthu

i will going to try the expression in 2-YT medium with low temperature and low IPTG concentration.

HI Olga Komarynets

i will try to do the auto induction with various temperature, IPTG con.c

thanks all

Hi, the prior advice is all good. Just to add a bit, mostly in the form of questions. Did you do a test induction and did that express? If it did you are likely losing plasmid. This is a particularly common occurrence with Amp plasmids and even weakly toxic genes. The pGEX constructs are notorious for this. There are however, several things you can do to improve this situation. 1) Make 8% glycerol stocks rather then 20-25 commonly used. The higher concentration stocks tend to thaw everytime you take them out of the freezer to innoculate the culture. Several labs including my own have tested the effects of these on bacterial survival. In general, cells with plasmid fair worse then those without (which also grow faster). Since Ampicillin is rapidly hydrolyzed by about 0.1OD 600nm you have no selection and if you have an overnight you start off with essentially no selection due to the hydrolysis of amp. There is now way to add amp during the growth to keep selection. Carbenicillin is a slight better options and we tend to use these on plates instead of Amp but it is still hydrolyzed. 2) add glucose to your media or use a minimal media such as MDB (see Bill Studier's paper in Protein Expression and purification). This media allows growth to very high levels and is very good at repressing background expression. LB is terrible for this since it contains a small amount of an inducer (lactose or a similar sugar) so you can get background induction.

If you still do not get expression even taking the appropriate precautions you may consider a couple of other sources of problems. Check to see if your C-terminus has a couple of hydrophobic residues (these are tags for degradation see Bob Sauer's and Tanya Bakers work). Alternatively, although less often it could be RNA structure issues. These are harder to debug but some people have found that you can reducing these issues helps in some cases.

A good test for plasmid maintenance is to do serial dilution onto LB plates with Amp, without Amp with IPTG and with Amp+IPTG. Essentially you should see the same number of colonies on with and without AMP and with Amp+IPTG you should not get any growth.

Good luck

dear John

thank for your valuable suggestion. now i will try the plasmid maintenance test, so i can conclude the real problems.

and i do not know about the ampicillin degradation issues during overnight growth, can you suggest me any idea to avoid this selection issues. 

Carbenicillin is significantly better.  However b-lactamase is an enzyme so you are always fighting an up hill battle.  We have moved to Kan based vectors and with care see very little evidence of plasmid loss or scrambling.  As long as your in the pGEX the best that I can suggest is using carbenicillin in plates and overnights with MDG plates and media.  This will keep the background as low as possible.  Second, make sure that you are not losing plasmid.  Best to use 8% glycerol stocks.  Take 900uL of your cells and mix with 100uL of 80% Glycerol (sterile) then freeze and store at -80C.  MDG is a very repressing media (specifically for lac-based promoter systems) so this will limit toxicity.  If you have a good glycerol stock and you then scrape off a bit of the frozen cells with a sterile pipet into a ~10mL culture of MDG and carbenicillin.  You then can seed a large culture the next day at anywhere from 1:1000 to 1:100 dilution (using amp as your antibiotic).  The loss of amp should not be a problem if nearly all of your cells contain a good plasmid since there will be essentially nothing to overgrow the plasmid containing cells.  If you still do not have expression, it may be time to try either an alternative expression plasmid or system.  (NOTE: do you have one of the rare tRNA plasmids co-transformed in your expression host, we routinely use the NEBL, something like pRIL although there is now one with additional ones, we have never found it to hinder expression but it often significantly helps, this plasmid is cam resistant, I think, you do not need to add cam to the expression culture since there seems to be little selective pressure to lose the plasmid).

Dear Behzad Heidari,

i have tried a lot,

After tried the expression in RIL,RILP, pLysS and Rossetta, co expression with a chaperone, with various temperature, IPTG, media. no better results obtained. GOT only faint expression with that i could not purify further due to insoluble nature. That protein i needed for crystallography.

Now i am doing it in yeast expression system.

Hi

I am now facing the same problem, have you done it well in yeast expression system? Could you please tell me what kinds of strains of yeast you are working with?

You should use BL21 not BL21 DE3

https://www.neb.com/faqs/2016/01/21/what-is-the-difference-between-bl21-and-bl21-de3-competent-e-coli-cells2

FAQ: What is the difference between BL21 and BL21(DE3) competent E.coli cells?

Both strains are B strains and thus both are deficient in Lon protease (cytoplasm) and OmpT protease (outer membrane). Accordingly, B strains are generally preferred for recombinant protein expression. The DE3 designation means that respective strains contain the λDE3 lysogen that carries the gene for T7 RNA polymerase under control of the lacUV5 promoter. IPTG is required to maximally induce expression of the T7 RNA polymerase in order to express recombinant genes cloned downstream of a T7 promoter.

BL21(DE3) is suitable for expression from a T7 or T7-lac promoter or promoters recognized by the E.coli RNA polymerase: e.g. lac, tac, trc, ParaBAD, PrhaBAD and also the T5 promoter.

Note that BL21 does not carry the gene for T7 RNA polymerase and thus is only suitable for expression from promoters recognized by the E.coli RNA polymerase: e.g. lac, tac, trc, ParaBAD, PrhaBAD and also the T5 promoter.