Transgenic protein expression in E. coli

Nowadays, genes of interest can be introduced into a large variety of plasmids available from commercial sources, but also from individual scientists who produce them as custom constructs for particularly difficult cloning strategies (if you ask them politely, they might consider sharing them with you). If you are working with a gene that has already been expressed successfully by somebody else, follow their protocol. You can also follow a protocol that has worked for a homologue/orthologue of your gene of interest. However, if you are starting from scratch and have no point of reference, consider your plasmid of choice carefully and, as a matter of routine, clone your gene into several different plasmids.

The most common expression plasmids will add a tag to the expressed protein. There are many different tags to choose from, the most popular include the 6His-tag, GST-tag, MBP-tag and SUMO-tag. The first three types can be used for affinity chromatography and the last three are also employed to aid with solubility of the expressed construct. The tags can be added either at the N-terminus or the C-terminus of the construct and it’s worthwhile trying both as sometimes only one of the two works. Solubility tags are generally large and, even though many proteins will crystallise with these tags, they are best removed after the initial purification steps. Many people don’t use removable 6His-tags as these are quite small and mostly do not interfere with crystallisation. However, removing your 6His-tag after Ni-NTA might help as sometimes protein molecules aggregate via their His-tags, particularly if a little bit of Ni2+ has leached from the beads/column. You can reduce this aggregation by adding 1 mM EDTA to your first dialysis step after Ni-NTA to remove the Ni2+ from your buffer (you can remove the EDTA and the chelated Ni2+ with a second dialysis step). However, the best approach is to remove the 6His-tags from your protein at this point as a matter of routine, particularly since this will give you a product that is more similar to the wild-type protein. Also, sometimes His-tags interfere with crystallisation, so it’s best to remove them if you’re in doubt.

After you have cloned your gene into a suitable plasmid, you can use it to transform competent cells to express the plasmid. Choose E. coli strains that have been engineered for transgenic protein expression, like BL21, Rosetta, etc… Again, different strains are appropriate for different expression strategies, so choose carefully and, just in case, try several at the same time to see which one/s performs best with you construct/s.

Transformed cells are either grown in autoinduction media or, more commonly, in rich media like LB, 2x YT or Terrific broth that will need to be induced for 3 hours (e.g with the addition of 0.1 to 1 mM IPTG) after the cells achieve an O.D.600nm of 0.6. This initial protocol works for most proteins but might need tweaking for optimum results. If the expression is poor or the product is insoluble, try growing your cells at lower temperatures (e.g. between 30°C and 12°C) and increase the time after induction from 3 hours (e.g. 6, 8, 12 hours or simply “over night”). Trial and error will give you the best combination. If the protein is toxic to the cells, it is best to grow the culture to a fairly high initial O.D.600nm (e.g. 1 or 1.2) before induction and then harvest the cells after 30 mins to 2 hrs instead of after 3 hours.

The presence of contaminating proteins can also lead to precipitation after Ni-NTA and your protein might actually become stable if you simply add a second purification step such as ion exchange or size exclusion chromatography. Since you mention you want to carry out structural studies with the protein, you will almost certainly need further purification steps anyway.

Some proteins are temperature sensitive and will aggregate if kept too warm or too cold. My current proteins come from thermophilic bacteria and they quickly precipitate if I keep them on ice or at 4°C for too long.


This guide was authored by:

Dr. Antonio Ariza
ORCID ID: 0000-0003-4364-823X
University of Oxford
Sir William Dunn School of Pathology
South Parks Road