Formation of inclusion bodies is one of the most common complications in heterologous protein expression (1). Despite this complication, the E. coli expression system is still highly used for eukaryote protein expression. Is this practice based on knowledge or historic consequence?

When I was at the final step of my study, attempting to purify my protein of interest, formation of inclusion bodies wreaked havoc on my plans. After several hard years of research in maize paramutation, and screening for mutants, I finally ended up with a few mutants affecting a R-r:std phenotype in a consistent manner. All I had to do was to express/purify my protein of interest and confirm its functionality in vitro. A collaborator handed me a plasmid with a gene of interest already cloned in. A lab colleague handed over a prokaryote protein expression vector containing His tags. He also gave me DH5Alpha and BL21 cells, and a resin for protein purification. My troubles started when I ran my first SDS gel. All of my protein was sequestered away in inclusion bodies instead of in a soluble fraction. I thought to myself “Well, fine, I still have good expression of my protein”. Western Blot had shown that it was the correct size. All I have to do was to make it soluble. I started reading protocols, references, and methods all showing fabulous success in recovery of soluble protein from inclusion bodies. Even so, nothing worked for me.

Meanwhile I spoke with a protein expert who himself was expressing proteins in various systems. He told me how refolding would only work under very specific conditions. For example you have to include the right metal ion during dialysis. Even then, it might not work. There is a codon bias among various organisms. Moreover some proteins need proper modifications and folding that can happen only in a natural environment. Thus, the best way to obtain your functional protein is to change the expression system.

Esposito and Chatterjee (2006) show that only 12–23% of eukaryotic proteins can be successfully expressed in E. coli. If this is such big problem why do researchers even bother to use the E. coli expression system for eukaryote proteins? One of the reasons is that this system has been on the market for a long time. It was one of the first systems available. The technology is simple and cheap.  When it works, it is a great system. That is the reason why it is worth testing. Still, is it really worth spending several months on cloning and expression with such low probabilities for success?

What are some other choices and what decisions do I have to make before choosing a protein expression system?
Table 1 provides a summary of the positive and negative aspects of E. coli and mammalian expression systems. Based on this information I would first take a long hard look at the nature of my protein, material resources and time. If my protein is an intrinsic membrane protein then I would definitely choose a mammalian expression system.

Table 1. Two major production systems for production of recombinant mammalian proteins.

Production of mammalian proteins in E. coli
Advantages	Drawbacks
Potential for high yields	Expression of mammalian proteins often fails
Ease of use	Poor expression levels
Low cost	Inadequate protein function
	Improper folding and aggregation into inclusion bodies
Production of mammalian proteins in mammalian cells
Advantages	Drawbacks
Native environment	Slow growth
Proper folding	High cost
Correct PTM	Laborious
Proteolytic processing	Significant lower expression levels compared to E. coli

Another important decision that a researcher has to make is which protein tag to choose. There are many different tags (2) used for protein purification in various systems (bacterial, yeast, mammalian).  I will briefly explain pros and cons for three common tags: His, FLAG and HaloTag.

The His (polyhistidine) affinity tag is 0.85 kDa in size. It works through metal affinity and gives very high yields due to high resin capacity. However, when using His-tag a researcher can expect impurities due to cross reactivity with numerous endogenous proteins.  In addition to a lack of specificity, this system can leach out metal ions which can inactivate the protein downstream.

FLAG octapeptide is also a small tag (1kDa). It is a chromatography tag based on polyanionic amino acids. When compared to the His-tag, it is much more specific although non specific binding has been observed. However, it has low capacity and requires expensive resins for purification which significantly increases the cost. There are also potential pitfalls associated with elution.  Low pH elution for example can damage the protein of interest.

Evolved HaloTag® is currently one of the most specific tags on the market. It was isolated from a rare bacterium that has no similarities with other organisms. Due to the formation of strong covalent link between HaloTag and ligand attached to the resin, a researcher is not afraid the protein will fall of  during extensive wash with a mild wash solution.  After washing, the tag is removed leaving plenty of clean protein in a correctly folded form.  Resin with high binding capacity is relatively cheap enabling liters of culture to be used if needed. The question researchers often ask is will the HaloTag, being 34 kDa in size, interfere with their proteins of interest. HaloTag protein itself folds so compactly that its small globular structure does not interfere with the protein of interest. A specifically tailored spacer between the protein and tag enables easy separation of the protein from the Halotag whilst also increasing the binding efficiency. One school of thought proposes that small tags like FLAG and His do not influence protein activity.  Nevertheless there is plenty of literature showing that this is not necessarily the case. It therefore comes as no surprise that the Kazusa Research Institute created a Human ORF library using the Halotag backbone. Being evaluated for expression levels, this library recently has been commercialized and available for researchers worldwide.

References:

Esposito D. and Chatterjee D. K .(2006) Enhancement of soluble protein expression through the use of fusion tags. Current Opinion in Biotechnology, 17:  353–358

Young, C. L., Britton, Z. T. and Robinson, A. S. (2012) Recombinant protein expression and purification: A comprehensive review of affinity tags and microbial applications. Biotechnology Journal, 7: 620–634.