PSI Structural Biology Knowledgebase

PSI | Structural Biology Knowledgebase
Header Icons

Related Articles
Membrane Proteome: Tuning Membrane Protein Expression
December 2012
Membrane Proteome: Every Protein Has Its Tag
November 2012
Insert Here
October 2012
Overexpressed problems
February 2012
Gentle membrane protein extraction
January 2012
Five good reasons to use single protein production for membrane proteins
January 2010
Crystallizing glycoproteins
September 2009
Tips for crystallizing membrane proteins
June 2009

Technology Topics Purification

Membrane Proteome: Tuning Membrane Protein Expression

SBKB [doi:10.1038/sbkb.2012.113]
Technical Highlight - December 2012
Short description: A tunable E. coli strain results in less protein aggregation in vivo and more functional, membrane-inserted protein.

Schematic representation of the Lemo21(DE3) protein overexpression system. Figure courtesy of Susan Schlegel.

For many biochemists, the workhorse of protein production remains the Escherichia coli strain BL21(DE3). However, this strain is not well suited for the production of most membrane proteins because of a limited ability to tune protein overexpression. This frequently leads to in vivo protein aggregation and toxicity caused by saturation of the membrane biogenesis and protein secretion machinery.

de Gier and colleagues now demonstrate that the Lemo21(DE3) strain may provide a better alternative for producing functional, properly folded membrane proteins. Lemo21(DE3) is a derivative of BL21(DE3). As in BL21(DE3), protein expression is driven by phage T7 RNA polymerase (RNAP) recognizing the T7 promoter that controls expression of the target gene. Unlike BL21(DE3), whose T7 RNAP expression is governed by the isopropyl-β-D-thiogalactopyranoside-responsive (and not well-titratable) lacUV5 promoter, Lemo21(DE3) allows control of T7 RNAP activity via tuning expression of T7 lysozyme, its natural inhibitor. In this system, the T7 lysozyme gene is located on a plasmid under a titratable rhamnose promoter, creating an indirect but very effective way to control the amount of protein produced. This titratable system, also compatible with popular autoinduction media, provides distinct advantages for cases in which expression of the target protein is toxic, as one can increase rhamnose amounts until one reaches a level of target protein that can be tolerated by cells.

As shown for several green fluorescent protein fusions of membrane proteins such as YidC and GltP, by optimizing rhamnose levels, one can maximize the amount of functional membrane-inserted protein and minimize the accumulation of protein aggregated in inclusion bodies. Furthermore, protein can be semicontinuously expressed by back dilution in rhamnose-containing fresh media, suggesting that under optimal rhamnose levels, no pressure exists to evade membrane protein overexpression. Temperature and codon optimization did not appear to significantly increase protein production to levels exceeding those obtained by optimizing rhamnose levels, which indicates that, for the targets at hand, simple optimization of the rhamnose concentration would suffice. Two membrane proteins purified from this strain, Mhp1 and NhaA, were not only monodisperse, but also produced crystals, suggesting that Lemo21(DE3) has the potential to become a membrane protein production platform for functional and structural studies.

Alexandra M. Deaconescu


  1. S. Schlegel et al. Optimizing Membrane Protein Overexpression in the Escherichia coli strain Lemo21(DE3).
    J Mol Biol. 423, 648-659 (2012). doi:10.1016/j.jmb.2012.07.019

Structural Biology Knowledgebase ISSN: 1758-1338
Funded by a grant from the National Institute of General Medical Sciences of the National Institutes of Health