Produce large quantities of difficult-to-express proteins
Tamara Kleber-Janke Wolf-Meinhard Becker
Biochemical & Molecular Allergology, Research Center Borstel, Germany
By using Stratagenes BL21-CodonPlus(DE3)-RIL cells * with extra copies of E. coli, argU, ileY, and leuW tRNA genes, it is possible to attain high-level expression of difficult proteins affected by poor codon usage: IPTG-induced expression of several recombinant peanut allergen genes, such as Ara h 2, were greatly increased in these special cells compared to the expression yield achieved by conventional E. coli cells like M15 or BL21-Gold(DE3) cells.
E. coli remains a popular host for the expression of heterologous proteins. Although the E. coli cell has a tremendous capacity to produce large quantities of protein, there are limits when the composition of the mRNA or protein is not typical. Within E. coli, a clear bias exists among the 61 amino acid codons found within the population of mRNA molecules, and the level of cognate tRNA appears directly proportional to the frequency of codon usage.1
A subset of codons, namely AGG/AGA, AUA, CUA, CGA, and CCC, are the least used codons in E. coli that are encoded by rare tRNAs. Recent studies suggest that an excess of any of these codons creates problems during translation, leading to a reduction in quantity of the protein synthesized.2 Errors are detectable as frameshifts and decreased levels of expression. More subtle changes, such as mistranslation, may also be occurring.
Until recently, three alternatives were usually recommended to solve these problems: Use a host containing a plasmid with the appropriate tRNA,3 synthesize the gene to replace rare codons with frequently used codons, and use aeukaryotic expression system.
Several results from expression experiments suggest that arginine codons AGG and AGA can have pronounced effects on the translation efficiency of cloned genes in E. coli.2 These codons are the least frequently used in E. coli, and the tRNAs that recognize them are among the least abundant. Given this situation, we predicted translational problems with the mRNA of some of the recently identified peanut allergens 4 that contain an excess of rare, low-abundance tRNA codons. Hence, to achieve high-level expression of the peanut allergen Ara h 2, we tested different expression systems and compared the final yield of recombinant protein.
Ara h 2 cDNA was first subcloned into an expression vector of the pQE series (Qiagen) to produce (His)6-tagged fusion proteins in E. coli M15 cells. In Figure 1A, expression of Ara h 2 protein was examined following 0, 2, and 4 hours as well as an overnight incubation once induction was completed. The Ara h 2 protein was visualized by immunoblot experiments (Figure 1A). No additional protein band of the appropriate molecular weight was detected after IPTG induction, demonstrating that Ara h 2 is only marginally expressed in these conventional E. coli cells.
Examination of the codons used in this sequence indicated that Ara h 2 contains 12% of the least used codon AGG/AGA in E. coli and, additionally, 2% of other rare codons read by minor tRNAs within a message encoding a protein made up of 157 amino acids. This might explain the inefficient translation (also found by other authors) when truncated translation products from peanut allergen Ara h 1 are present.5
The Ara h 2 cDNA was subcloned into an expression vector of the pET series (Novagen) that uses the T7 RNA polymerase responsive promoter to produce (His)10 -tagged fusion proteins. Two E. coli strains were transformed with this plasmid and selected on LB broth supplemented with ampicillin: BL21-Gold(DE3) cells and the recently generated BL21-CodonPlus (DE3)-RIL cells 6 carrying extra copies of the argU, ileY, and leuW tRNA genes. Recombinant Ara h 2 expression was induced by adding 1 mM IPTG when the cells reached an optical densitiy of 0.6 to 0.8 at 600 nm. Incubation was continued at three different temperatures (22C, 30C, 37C) for an additional 2, 4, and 20 hours.
The production of Ara h 2 was enhanced significantly with the new BL21-Codon Plus (DE3)-RIL cells (Figure 1C), compared to the pQE/M15 system (Figure 1A) and to the conventional BL21-Gold(DE3) cell expression system (Figure 1B). The highest amount of recombinant protein was achieved by selecting an overnight incubation at 22C after IPTG induction.
The final yield of soluble recombinant Ara h 2 was estimated as 5 mg/L of culture, which is 100 times greater when compared to the 50 g achieved by the pQE/M15 expression system.
It is possible to make large quantities of high-quality heterologous proteins that are otherwise difficult to express in conventional E.coli hosts. When low-expression yields or low-quality recombinant protein is due to rare codon usage, BL21-CodonPlus (DE3)-RIL cells are the hosts of choice. We used these special cells and greatly increased the expression yield of the major peanut allergen Ara h 2, which contains 12% of the rare AGG/AGA codons in its message.
A 10-ml culture of each of the expression experiments was induced by adding 1 mM IPTG, then incubation commenced at three different temperatures (22C, 30C, 37C). Of the total E. coli lysates, 1 ml each was harvested after 2, 4, and 20 hours (overnight incubation) by centrifugation, then resuspended in 100 l of reducing sample buffer. The samples were analyzed on SDS-15% PAGE and visualized by staining with Coomassie Brilliant Blue stain. Proteins were transferred to a nitrocellulose membrane for Western blot experiments. A serum of a peanut allergen was diluted 1:50 and used as the first antibody. The specific binding of the IgE to the recombinant Ara h 2 was detected by incubating the membrane with alkaline-phophatase-conjugated antihuman IgE as the second antibody, and bands were visualized with nitroblue tetrazolium chloride including 5-bromo-4-chloro-3-indolyl phosphate potassium salt.
Wada, K., et al. (1992) Nuc. Acid Res. 20: 2111-2118.
Kane, J.F. (1993) Curr. Opin. Biotechnol. 6: 494-500.
Schenk, P.M., et al. (1995) Biotechniques 13: 14-18.
Kleber-Janke, T., et al. (1999) Int. Arch. Allergy Immunol. 119: 265-274.
Burks, A.W., et al. (1995) J. Clin. Invest. 96: 1715-1721.
Carstens C.-P. and Waesche A. (1999) Strategies 12: 49-51.
* Patents pending