Hi Vishwanath, I'm not sure if I understand your question and I'm assuming expression in E. coli here but I will answer the main issues I believe you are dealing with.

The first issue in separating inclusion bodies from soluble protein can be easily accomplished. The easiest method to separate inclusion bodies from soluble protein would be to centrifuge the lysed cell at 12,000 - 16,000 xg. This should easily separate soluble and insoluble. From there I would check the plasmid you have used for your recombinant protein. In the production of your recombinant protein protein you may have a tag, most often a HIS tag, allowing for a hopefully simple and efficient purification and separation via affinity chromatography. If not then salting out or another form of chromatography should get you fairly pure protein in one or two more steps.

However I believe your central question lies in the fact that your recombinant protein is being expressed as an inclusion body and you are either A. Trying to prevent this or B. Trying to refold.

I'll start with B as it's a little easier to answer. If you have no other alternative but to refold from inclusion bodies then I would check the literature for similar proteins to see if there is a already proven method you can adapt/begin with. Most of these will use some type of denaturing chemical (Guanadine or Urea) and then, depending on the protein, a slow or fast dialysis period. If your protein has disulfide bonds then you may need to use reduced and oxidized glutathione (I try 10:1, 5:1, 3:1 ratio) usually in the 1 - 5mM range to try and produce an environment capable of reforming the correct bonds. The refolding buffer will often contain Lysine or Arginine as they have been shown to aid in protein refolding. Bottom line though is if there is no method you can adapt for your protein a refolding kit may be your best bet to find conditions capable, if there are any, of refolding your protein. Refolding can be great and incredibly efficient, it can also be a crapshoot and a waste of time and resources.

As for reducing or stopping the formation of inclusion bodies again the options are varied and diverse. I would try the following first as a means of producing a soluble product.

1. Reduce temperature of growth and induction, try at 30C, 25C, 20C, 15C. Most likely you can grow the culture at 37C till an OD of 0.5 or 1. I would then swap to the cooled environment for 10 - 20 min to allow the culture to equilibrate. Induce and take time point samples

2. Try different concentrations of IPTG. If you are using 1mM try 0.5, 0.2, 0.1. Lowered levels can aid in reducing 'stress' resulting in a properly folded active protein

3. Look at your vector, can you add a solublization tag? Does your protein need a thioredoxin tag for disulfides?

4. Look at your media, often times changes to MM or pH can have a drastic effect on your expression and solubility of the protein. This includes aeration!! If the culture is not aerated it can lead to a drastic drop off in soluble protein expression.

5. Look in literature, if your protein is common do others use this same system? If not it may indicate that it may not be worth it ie. move on to yeast, insect, or mammalian expression

Hope that helps,

I'm guessing that you mean soluble cytoplasmic expression in E. coli. There isn't a one size fits all solution, you have to work on a case by case basis. Some things you may try:

- First, it is realistic to expect your protein to fold spontaneously in E. coli? Some proteins are produced as a propeptide that first folds and then is matured proteolytically (hence a recombinant mature variant may never refold), others are truncated variants or subunits of oligomeric complexes that expose hydrophobic surface patches, others have hydrophobic surface patches that are masked by glycosylation...you get the idea.

- If your protein has disulfide bridges, try the Origami strains from Novagen or the SHuffle strains from NEB, or use vectors for secretion to the periplasm (although the latter comes with its own set of problems).

- Try decreasing the rate of synthesis of your protein. You may grow your bugs at lower temperature (25-28*C is a good starting point), or switch to a less strong, regulatable promoter. Also, I have anecdotic evidence that the popular autoinducing recipes, while convenient, are not the best choice if you are looking for soluble expression (I guess because expression is raised to brutal levels in a relatively short period of time); in these cases the old LB+IPTG recipes has worked better in my experience.

- You may try co-expressing chaperones. Takara sells a convenient chaperone plasmid set. Beware that there's plenty of cases in the literature where apparently soluble expression is nothing but the formation of soluble aggregates aided by chaperone coexpression.

Dear Hothpet,

Before answering your question, may I know which vector system you used to clone your gene of interest? Because, the tag coded by the vector you choose mainly determines the solubility of your protein of interest.

I would recommend you the pMAL vector series (New England Biolabs) that allows you to express your recombinant protein tagged with maltose binding protein (MBP) which could be purified by amylose resin in a simple chromatography approach. This helps in getting a higher fraction (>85%) of your protein in a highly soluble form, without forming inclusion bodies.

However, careful handling is a key factor in avoiding the formation of inclusion bodies. For instance, you should maintain your bacterial culture on ice throughout the purification procedure. During the cell disruption and other steps, heat generation should be mitigated. These precautions might help you in getting the recombinant protein in soluble form. Good luck.

You can try to decrease temperature (18 °C overnight induction), change expression vector to some less active, also switching places of your purification tags might help.

Firstly, you could try tagging the protein to a solubility enhancing molecule like MBP or GST. This has an added advantage tht the tag could be used for purification of the molecule later. Secondly, in case you do not want to use protein tags, try induction of the protein at lower temperatures i.e. 16ºC or 20ºC could help resolve the problem (it usually helps especially when trying to express proteins of eukaryotic origin in bacteria)

You can use a kind of hybrid system (combination of denaturing and native conditions). Use guanidinium lysis buffer for lysis and then urea-based binding and washing buffer. Finally use native washing and elution buffer to regain the activity. You would get related protocol in Invitrogen or GE life Sciences website.

Which expression system do you use? If you use pET vectors, optimizing IPTG concentration and lower the temperature during induction may help (try with 0.1mM to 0.5mM concentration of IPTG, overnight at 17oC). You also can clone it to vectors such as MBP-, SUMO- fusion. You may have high chance to obtain a soluble fusion protein.

Hi-

You can try to induce with less IPTG (0.1 to 0.5mM) and at a lower temperature (18°C) O/N, this usually helps.

Simon

You may try to fuse your target proteins with some solubility enhancers. Traditional way is to use glutathione S-transferase (GST), or thioredoxin (TRX), or N utilization substance A (NusA), or maltose-binding protein (MBP) as translational fusion partners. However, the number of available solubilization tags is very large. In addition to the more traditional tags, entropic bristle domains were recently proposed (see reference below) as soluble expression enhancing fusions based on intrinsically disordered proteins (IDPs).

Santner AA, Croy CH, Vasanwala FH, Uversky VN, Van YY, Dunker AK. Sweeping away protein aggregation with entropic bristles: intrinsically disordered protein fusions enhance soluble expression.Biochemistry. 2012 Sep 18;51(37):7250-62. doi: 10.1021/bi300653m. Epub 2012 Sep 5. PMID: 22924672

After Dialysis, concentrate the protein by using centricon tubes or After induction and protein expression, go for sonication and purify protein.

This can be a difficult problem. Often, a heterologous protein is simply not soluble enough, or gets trapped in aggregated intermediates, at the high concentrations produced in a bacterial expression system. Here are four strategies that may or may not work:

1. Co-expression with GroEL and GroES (I think there maybe be a commercial vector system to do this now. There is plenty of literature on this approach.)

2. Tandem gene insertion. Sometime, multiple copies of the same gene in the same plasmid will slow down transcription/translation. 4 copies of the alpha subunuit of trp synthase in tandem expresses satisfacatorily, whereas one copy produces aggregates.

3. Expression at a lower temperature. Sometimes, lowering the rate of transcription/translation will discourage aggregate formation

4. Use "leaky expression" from a pET vector. Don't induce with IPTG. You will get a lower rate of expression that might discourage aggregate formaiton.

If these approaches don't work, you might consider moving to another expression system (yeast, mammalian cell, etc.) or choosing another protein homolog, if available.

There are multiple ways and most are not particularly efficient. The most likely to work is testing several different expression temperatures. This can either be done by growing the entire time at the target temperatures (for E. coli say 18-37C, we usually do 18, 25, 30 and 37 but that is pretty arbitrary). Alternatively, we often grow at the 37 and just before induction switch to the target temperature (this speeds things up). If you are using autoinduction media you would do the switch at a point in the growth curve where little protein has been expressed. Alternatively, you can express your protein with chaperones (some reports that it works but our experience is that at best we get the target protein stuck to the chaperone and upon release with ATP it aggregates). There are some recent reports that you can grow in the presence of stabilizing agents (not sure which ones but TMAO comes to mind). Finally, and perhaps the most likely way to succeed is to try to find an optimal expression construct via refining the domain boundaries, removing potentially aggregation prone sequences at the N- and C-terminus. If you are really ambitious you can try random mutagenesis of the protein fused to a c-terminal tag (like GFP) and look for better folded variants (increased green fluorescence).

Dear Hothpet,

Before answering your question, may I know which vector system you used to clone your gene of interest? Because, the tag coded by the vector you choose mainly determines the solubility of your protein of interest.

I would recommend you the pMAL vector series (New England Biolabs) that allows you to express your recombinant protein tagged with maltose binding protein (MBP) which could be purified by amylose resin in a simple chromatographic approach. This helps in getting a higher fraction (>85%) of your protein in a highly soluble form, without forming inclusion bodies.

However, careful handling is a key factor in avoiding the formation of inclusion bodies. For instance, you should maintain your bacterial culture on ice throughout the purification procedure. During the cell disruption and other steps, heat generation should be mitigated. These precautions might help you in getting the recombinant protein in soluble form. Good luck.

Generally, if you slowdown the protein synthesis rate, you may find less inclusion bodies or aggregates. Try with reduced concentration of IPTG and/or reduced temperature after IPTG treatment. These two simple modifications may help you to get soluble active proteins.

Try to grow bacteria at lower temperatures (20°C-25°C) and induce using a lower concentration of your inducer. You can also try to use different strains of E.coli like ORIGAMI (DE3) or fuse a chaperone to your protein: DsbA, DsbC...

Take also in consideration that in recent years IB's produced at low temperature (25°C) show traces of correct folded proteins inside. It's possible to solubilize this protein with mild detergents (0,2% N-lauryl Sarkosyl) or 2M L-arginine.

If you are interested in this last approach I can give you some more advices.

Hi, just some thoughts: You could try to co-express appropriate chaperones / HSPs, optimize solubilization / renaturation / refolding. Do you have an opportunity to confirm that your protein sequence is expressed correctly? Any posttranslational modifications required for activity?

What conditions are you using? sometimes lowering the temperature can help, e.g. expression at 18ºC over night. alternatively you could use a gluc/lac medium for self-induction of the expression.

Try to grow bacteria at lower temperatures (20°C-25°C) and induce using a lower concentration of your inducer. You can also try to use different strains of E.coli like ORIGAMI (DE3) or fuse a chaperone to your protein: DsbA, DsbC...

Take also in consideration that in recent years IB's produced at low temperature (25°C) show traces of correct folded proteins inside. It's possible to solubilize this protein with mild detergents (0,2% N-lauryl Sarkosyl) or 2M L-arginine.

If you are interested in this last approach I can give you some more advices.

Try to induce at lower temperatures and inducer concentrations. Set up trails using different conditions in smaller cultures (3-5ml).

If you find a way out then you will be the scientist of the year ;-)!

First I suggest you to check several conditions on your expression system like: type of cells, temperature, final concentration of inducer; also you can check over time the expression: sometimes some proteins are expressed as soluble molecules but after a certain time, the accumulation inside the cell can lead to the accumulation of the protein in inclusion bodies.

After that I would use the tag approach: some tags or protein domains increases the chance to obtain a soluble protein. MaltoseBindingProtein (MBP) is a typical example of solubility tag but is relatively large (42kDa). GB1 domain in my opininion is much better since is small (56 residues, 6kDa) and enhances the solubility greatly (in my experience up to 98% of the protein becomes soluble) but sometimes only one GB1 domain is not sufficient (you need to add more than one). Other tags like Histag or GST do not change the solubility as far as I know.

After the purifcation you can cleave the domain using an appropriate protease like TEV (works better but if you have to produce it is quite expensive or you can buy it) or thrombin (you can buy it but it doesn't work smoothly).

If the tag approach will not work then I'm afraid you need to refold the protein.

I agree with most of the approaches that have been suggested including lower temperature induction and the use of fusion proteins. In particular, you may want to try using the MBP fusions and then cleave your passenger (target) protein from MBP in vivo by putting a TEV protease site between the the two and then coexpressing TEV protease. Often the fusion protein is only needed to assist the passenger in getting to the folded state. The in vivo cleavage approach will let you know if your target is soluble in the absence of MBP. David Waugh has published a number of papers on this approach (e.g. Nallamsetty et al 2004).

Alternatively it may be that your protein is aggregating during lysis and you should look to add stabilizing agents during this step. Look at: Churion and Bondos, 2012, Methods Mol Biol. and Leibly et al. 2012. PLoS One. for some ideas.

Hi Vishwanath,

Can you be a bit more specific about your tag, fusion protein, expression system etc?

As you already have a construct that expresses then there are a few things that you can try (I am assuming you are using T7, LB and IPTG at 37°C so far with a globular protein and not an IDP).

Firstly if you want to continue to use IPTG for induction use a richer, buffered medium , terrific broth is simple to make but can improve soluble yields dramatically over LB.

Try reducing temperatures post-induction to 16-20°

Try reducing IPTG concentration-although in most e.coli expression strains induction is not really titratable.

Try auto-induction media instead of IPTG

If none of these work then think of changing your fusion protein or tag, I particularly like SUMO, thioredoxin and Z-tag (and much prefer them to GST and MBP) and/or changing the boundaries of your construct to avoid e.g. unstructured domains. There is no sure way of getting soluble protein that stays soluble after cleavage from the fusion but these should help.

Of course the effects of all of these are dependent on your protein and it may be that you ultimately have to accept re-folding from inclusion bodies.

Good Luck

Hi Vishwanath, I'm not sure if I understand your question and I'm assuming expression in E. coli here but I will answer the main issues I believe you are dealing with.

The first issue in separating inclusion bodies from soluble protein can be easily accomplished. The easiest method to separate inclusion bodies from soluble protein would be to centrifuge the lysed cell at 12,000 - 16,000 xg. This should easily separate soluble and insoluble. From there I would check the plasmid you have used for your recombinant protein. In the production of your recombinant protein protein you may have a tag, most often a HIS tag, allowing for a hopefully simple and efficient purification and separation via affinity chromatography. If not then salting out or another form of chromatography should get you fairly pure protein in one or two more steps.

However I believe your central question lies in the fact that your recombinant protein is being expressed as an inclusion body and you are either A. Trying to prevent this or B. Trying to refold.

I'll start with B as it's a little easier to answer. If you have no other alternative but to refold from inclusion bodies then I would check the literature for similar proteins to see if there is a already proven method you can adapt/begin with. Most of these will use some type of denaturing chemical (Guanadine or Urea) and then, depending on the protein, a slow or fast dialysis period. If your protein has disulfide bonds then you may need to use reduced and oxidized glutathione (I try 10:1, 5:1, 3:1 ratio) usually in the 1 - 5mM range to try and produce an environment capable of reforming the correct bonds. The refolding buffer will often contain Lysine or Arginine as they have been shown to aid in protein refolding. Bottom line though is if there is no method you can adapt for your protein a refolding kit may be your best bet to find conditions capable, if there are any, of refolding your protein. Refolding can be great and incredibly efficient, it can also be a crapshoot and a waste of time and resources.

As for reducing or stopping the formation of inclusion bodies again the options are varied and diverse. I would try the following first as a means of producing a soluble product.

1. Reduce temperature of growth and induction, try at 30C, 25C, 20C, 15C. Most likely you can grow the culture at 37C till an OD of 0.5 or 1. I would then swap to the cooled environment for 10 - 20 min to allow the culture to equilibrate. Induce and take time point samples

2. Try different concentrations of IPTG. If you are using 1mM try 0.5, 0.2, 0.1. Lowered levels can aid in reducing 'stress' resulting in a properly folded active protein

3. Look at your vector, can you add a solublization tag? Does your protein need a thioredoxin tag for disulfides?

4. Look at your media, often times changes to MM or pH can have a drastic effect on your expression and solubility of the protein. This includes aeration!! If the culture is not aerated it can lead to a drastic drop off in soluble protein expression.

5. Look in literature, if your protein is common do others use this same system? If not it may indicate that it may not be worth it ie. move on to yeast, insect, or mammalian expression

Hope that helps,

For some protein with a high tendency to aggregate and/or for minclusion bodies, we induced protein expression at 15°C ON with very low concentration of IPTG (0,1 uM). that helps a bit.

I think most of the approaches have already been suggested here. I just want to suggest you take a look at this paper, "Purifying natively folded proteins from inclusion bodies using sarkosyl, Triton X-100, and CHAPS" (Tao et al, 2010 Biotechniques) which I have used and it may work you your case.

Good Luck

Many good suggestions have already been made. I'd certainly agree with Nick Berrow's suggestion of autoinduction medium. that has cured the problem for us on several occasions , presumably because it is a less sudden turn-on of protein production.

Other things that have worked for us are:

1. co-expression of GroEL and GroES in the expression host.

2. Delivering a heat shock by a few minutes incubation at 40C before adding IPTG and growing as many people have suggested at low temperature. We have occasionally gone as low as 8C - slow growth but soluble protein!

This is a nice paper as to stabilization and folding of inclusion body protein.

http://www.sciencedirect.com/science/article/pii/S1389172305703728

Hi, the first and fore most thing to do, in order to avoid inclusion bodies is reducing the expression temperature from 37 to 30 or 25 or even 20 degrees. Of course it take longer time for the bacteria to grow at lower temperatures but mostly it works. There are lot of other things that can be done for example to co-express another protein that helps for the folding of the proteins, there are several commercial plasmid vectors available for such co-expression. hope this helps, good luck, -Narasimha.

Although many good replies have been given, I will elaborate along the lines of Zibin Jiang's recent reply. It is an alternative tack to avoiding the misfolded protein in inclusion bodies: in fact, it is usually easier to purify proteins from inclusion bodies, but of course reconstituting biological activity requires refolding. We did this with isoforms of hepatocyte growth factor (Stahl et al., Biochem J, 1997 - happy to provide the full text) which has quite complex folding and disulfide-bind rich kringle domains. Basically, once the inclusion bodies are isolated (by French press or similar lysis method and centrifugation), they are solubilzed in 8M guanidine HCL and 100 mM dithiothreitol. Soluble proteins can then be purified by sizing chromatography and selected fractions can then be dialyzed in a "folding buffer" containing urea, reduced glutathione and oxidized glutathione overnight, then against Tris-HCl (7.5) and 100 mM NaCl before a second round of sizing chromatography to separate folded from unfolded species. The method provided good yield, excellent purity, as well as bioactivity in the refolded protein.

Hi Vishwanath,

Lots of great ideas have already been posted, but I just wanted to reinforce the earlier post that your protein may be aggregating during the lysis. To determine, whether the protein is really in inclusion bodies, you need to wash the lysis pellet in buffer plus a 1% Tween and 1-2M guanidine. If your protein is in the inclusion bodies it will still be in the pellet after the wash ( Inclusion bodies typically need 4-6M guanidine to dissolve). Purifying inclusion bodies can actually be a great way to purify your protein- if you can refold it-since most of the protein in the inclusion bodies should be your expressed protein. If your protein is not in the inclusion bodies than its worth screening different lysis conditions varying salt, pH, presence of detergent etc. We have one protein that seems to stick to DNA and is not soluble in NaCl over 100mM during the lysis step. Adding sufficient benzonase to fully digest the DNA , solubilizes the protein. I've also had great success with induction at low temperature. Good Luck.

There are numerous additional techniques that you can try out to increase soluble protein expression, here are a few suggestions:

- Examine both cytoplasmic and periplasmic expression using various signal sequences as the ideal redox potential varies for each protein.

- Test additional promoter systems such as PhoA.

- Test different vector backbones who's origin of replication generate either high or low cellular copy numbers.

- Co-express with chaperones such as GroES and GroEL.

- If there are numerous cysteine residues, co-express with a disulfide bond isomerases such as DsbC.

Unfortunately no one technique works for all proteins, so it is a matter of trial-and-error. Hopefully one or a combination of a couple of these suggested techniques works out for you.

I agree the tag (MBP, TRX) is critical to promote solubility. If you don't have vectors to create such translational fusions, you can try a denaturing protocol (to destroy any potential aggregation, inclusion bodies...) and refolding of the protein.

I kindly recomend the protocol published by Feilner and coworkers (Molecular & Cellular Proteomics, 4(10), 1558-1568.).

http://www.mcponline.org/content/4/10/1558.short

Aggregation would usually happen when the recombinant proteins get expressed higher amount, which may result in out of range of chaperone capacity of host cells (say, E. coli). Most of the answers suggested here therefore, focus on lowering the expression level of the recombinant proteins; lowering IPTG, temperature, induction time, less dense host cells, shaker rpm, etc. Also tagging soluble proteins such as GST would help a lot.

As for me, I grow E. coli at 37 degree up to OD less than 0.4 and induce with 0.4 mM IPTG for 1.5 hr at 30 degree in 150 rpm of shaker for standard, well-solubilized protein production. You can change each parameter for adjustment. Good luck !

Hi you can try lower temp , like 30. its better than 37C.some vectors that have preplasmic secretory signal like pet21a, can produce solube protein.

you have to use Ni affinity chromatography

to try to precipitate by salting out using ammonium sulfate and the apply to affinity chromatography, it will maintain your protein in 3d structure and active.

it can be depend on host strain or vector. and /or in vivo in vitro protein. in addition may depend induction time. If you give more information about the study can get more specific answers

Decreasing IPTG has only a small effect since the transporter allows accumulation of the compound. Still some differences can be seen. Lowering the temp has the largest effect, although some proteins are not expressed at low temperatures.

We've worked with some insoluble proteins before. What has worked best for us is to use an N-terminal tag (either MBP or GST, need to test which works best empirically) and co-express with the chaperones GroES/GroEL. The chaperone co-expression really seems to help keep these things folded and increases the yield of soluble protein. Our expression strain also contains pLysS and we use a French Press to lyse the cells (E. coli), which gives efficient lysis and the most soluble protein. Good luck!

You can possibly change the affinity tag. Sometimes, it helps in better folding and increases the solubility of your protein. For example, MBP is a very good tag.

Hi

One note. Some of the proteins solubilized by expression with GroEL/ES and DnaK/J/E are still not folded. Several of the high throughput groups have looked at this issue in one way or another. Many are actually stuck to these chaperonins and either don't release with ATP or aggregate upon release. When using them you need to determine whether they are actually folded soluble or just solubilized (presumably in an unfolded or partially unfolded state) by the chaperonin. When it works correctly its great, unfortunately many eukaryotic proteins can't productively use the E. coli chaperones.

Since you have tried the IPTG concentration and low temperature conditions. I would suggest two things. First you can add glycerol in your sonication buffer after growing the culture at 18 degrees. Secondly if you don't have much problem in subcloning you can try arabinose promoter or tac promoter vectors lik pBAD etc. for soluble proteins. T7 is a strong promoter so most of the times protein goes in inclusion body.

Hi,

Although it may sound weird, sometimes one has to de-optimize expression so it is not that efficient and then inclusion bodies are not formed and/or expressed proteins are easier to re-solubilize. Is your sequence codon-optimized? Have you tried a non-codon-optimized version? Have you tried different temperatures and incubation times?

Godd luck!

Our lab had good experience with SUMO Tag for several proteins as well. SUMO system not only helps protein solubility but also the fusion tag is very easily to be removed if you want native protein w/o tag. We have a SUMO/HIS double tag expression vector and a his-tagged SUMO protease (ULP1). This is a very efficient and cost effective protease to remove the fusion tag (SUMO/HIS tag). In most case, we only use 1ul of ULP1to remove the fusion tags from 1mg purified protein (with>95% cleavage at 4*C for 16 hours).

http://www.enzymax.net/sumo%20protease.htm.

Good luck!

Nick Berrow has covered most of your options. I agree that you need to add a solubility tag. SUMO is good as well as maltose binding protein and Nus-A and a few more that are used less frequently (glutathione -s-transferase, cellulose binding protein, ubiquitin and dsbA, biotin binding protein). Are you certain the protein is in the pellet (see it on gel)? If not you may have very poor expression in which case you may need to do a new transformation with a a recloned plasmid. Again it can't be emphasized enough that this is entirely protein dependent. I would definitely try the auto-induction method. WE've had good success with auto-induction and it eliminates all the trial and error with IPTG. Also there are several different E.coli strains that may work better- Shuffle is one that reduces aggregation. How large is the protein and is it mammalian? If it is more 40 or so Kd it may e expressed better in insect or mammalian cell lines. Over expression is not a problem in these systems.

You seem to have covered most of what I would have done. It may just suggest that your protein is toxic to the bacteria and that's why it's going into inclusion bodies. So, you may not be able to get a soluble version. Or you may need to add a different tag like GST.

Hi Vishwanath,

There are a few tricks you can try to improve the solubility of proteins expressed in E. coli without changing your expression construct. The first potential problem to be dealt with are disulfide bridges. Commercial E. coli are available from NEB ("SHuffle") that express DsbC in the cytoplasm and improve folding of proteins with high oxidation levels - anything with more than 2 disulfides is probably a good candidate for this. You can also alter the intracellular environment of coli by adding osmotic stress agents to the growth media. Common additives include 0.5M sorbitol, 5 w/v % sucrose, 0.2M NaCl. None of these require exogenous glycine betaine be added to the growth media. I have also found success in adding buffering agents to the media which has dramatically improved the solubility of GFP fusions I've made. I have had lots of success with 50 mM phosphate buffer in my media - keep in mind this increases the resistance of coli to kanamycin and thus Km should be used at 100 micrograms/mL in phosphate-buffered media. Finally, several papers have suggested that increasing the ammonia concentration in the growth media can improve the solubility of recombinant products. I have used ammonium chloride at 5 mM with some success.

I hope those suggestions help. If you're willing to work with fusion proteins as others have suggested, I vote for maltose binding protein. I've successfully purified and refolded highly-oxidized human proteins as MBP fusions.

Good luck!

Steve B.

Your major mistake is trying to solubilize the protein in Urea. Replace it with some other substance. It will work like miracle.

there is nothing wrong of getting IBs because yield of IB usually is much higher compare to a soluble protein expression.

For cell rupture it is good to use Microfludizer or French press unit.

After cell rupture is good to do washing IBs before solubilazation, during solubilization of IBs is good to add a reduce agent, instead of Urea you can try a soft detergent (for example, Amisoft).

Refolding should be done with a pair of reduce-oxidize agents in order to increase the yield of refolding, typical yield is 70%, the main problem during refolding is dimerization and there are some techniques to minimize it and you can easy can find them in literature.

After solubilization in urea it is possible to do IEX in urea solution to purify the protein before refolding.

As an alternative to E. coli, you could try Lactococcus lactis, a gram-positive lactic bacterium, that is also widely used in biotechnology for large-scale production of heterologous proteins. One major advantage of Lacto is that it makes no inclusion bodies. See papers of the group of E Kunji, e.g. Monné et al. Protein Science (2005), 14:3048–3056. We also used it with success, see Frelet-Barrand et al. (2010) PLoS ONE 5(1): e8746.

You may also want to try yeast.

Thank you all for your valuable suggestions.