Gene optimization mechanisms: a multi-gene study reveals a high success rate of full-length human proteins expressed in Escherichia coli

Abstract

The genetic code is universal, but recombinant protein expression in heterologous systems is often hampered by divergent codon usage. Here, we demonstrate that reprogramming by standardized multi-parameter gene optimization software and de novo gene synthesis is a suitable general strategy to improve heterologous protein expression. This study compares expression levels of 94 full-length human wt and sequence-optimized genes coding for pharmaceutically important proteins such as kinases and membrane proteins in E. coli. Fluorescence-based quantification revealed increased protein yields for 70% of in vivo expressed optimized genes compared to the wt DNA sequences and also resulted in increased amounts of protein that can be purified. The improvement in transgene expression correlated with higher mRNA levels in our analyzed examples. In all cases tested, expression levels using wt genes in tRNA-supplemented bacterial strains were outperformed by optimized genes expressed in non-supplemented host cells.

Figure 1

Workflow of multi-gene study: 100 wt and sequence-optimized genes were cloned or synthesized into a pQE-T7 E. coli expression vector. PT7: T7 promoter; lac O: lac operator; RBS: ribosome-binding site; ATG: start codon; 6xHis: N-terminal hexahistidine tag; wt/optimized: cloning cassette to receive the gene coding sequence; Amb: amber stop codon; Stop: translational stop; ori: origin of replication; lacI: Lac repressor gene; Kanamycin: kanamycin resistance gene. The N-terminal 6xHis tag is exoproteolytically cleavable using the TAGzyme system. Every QIAgene E. coli contains a universal stop point for the TAGzyme protease. His tag sequences can be deleted by NdeI restriction for generation of a construct for expression of an untagged protein. The amber stop codon (UAG, Amb) can be used to incorporate a label making use of the amber suppression principle. Each wt and optimized construct was expressed in E.coli cells in vivo. The total cell lysate was labeled with the dye Chromeo P503 which only becomes fluorescent upon binding to an amino group of a protein. Lysates were separated on a SDS gel and scanned using an Ettan DIGE™ Fluorescent Scanner. Signals were quantified using the ImageQuant™ TL software. The factor (3.76) displays the ratio of protein expression using optimized (opt) and wild type (wt) sequences. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 2

Optimized sequences increase yield of soluble protein in in vivo E. coli expression system. The expression in E. coli BL21(DE3) and Ni-NTA purification via His tag under native conditions of four wild type (WT) and optimized (OPT) sequences and optimized CAV1 is shown (wt CAV1 cannot be detected). Samples were analyzed on a SDS gel and stained with Coomassie Brilliant blue. Arrows indicate the protein of interest, arrowheads show lysozyme; elution fractions (E) were quantified with a Bradford assay and 3 μg protein was separated in case of sequence-optimized expression in comparison to the same volume of wt protein fraction. TL: total lysate, CL: cleared lysate; 2.5 μL of each fraction were separated R: resolubilized membrane fraction; BT: break through; W: wash. Note that some protein in the cleared lysates is insoluble and purification of the soluble protein results in enrichment in the elution fraction. Marker: Page ruler prestained protein Ladder (Fermentas); for more information on the genes and proteins see Table I. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]

Figure 3

Enhanced codon usage is only one aspect of gene optimization. (A) Human wt genes coding for PLK1, SMARCD1, CSF2, AP-1, YY1, CCL5, and CAV1 were expressed in Rosetta2 (white) and BL21CodonPlusRIPL (grey) E. coli strains, both supplemented with rare tRNAs. Wt and sequence-optimized genes coding for the same proteins were expressed in E. coli BL21(DE3) (black). (B) Lysates were labeled and quantified using the fluorescent dye Chromeo P503, separated on an SDS gel and analyzed with an Ettan DIGE scanner. Proteins bands were evaluated using the ImageQuant TL software. Every expression was done in triplicates. WT: wild type sequences; OPT: optimized sequences; C: control (mock transformation).

Figure 4

mRNA level correlates with amount of recombinant protein. Expressions of CK1 (A), LCK (B) were analyzed at 4 different time points after IPTG induction at mRNA and protein level. 6 × 10⁸ cells were harvested, total RNA was isolated and used for relative quantification by two-step real-time PCR. Real-time PCR measurements were done in triplicate with samples from two independent experiments. The fold changes in mRNA expression relative to the mRNA level at T₀ are plotted against the time after induction. Representative Western blots show the expression levels of the corresponding proteins. Total cell lysates from an identical number of cells at the different time points post induction were subjected to SDS-PAGE and subsequent Western blotting using Penta-His HRP Conjugate.