February 15, 2014 (Vol. 34, No. 4)

Transfecting Mammalian Cells without Endotoxin Cleanup Steps

In drug discovery research, E. coli is commonly used for plasmid production for downstream mammalian transfection and protein expression experiments. However, this method usually requires endotoxin cleanup steps prior to mammalian cell experiments to avoid cell toxicity and poor transfection efficiencies.

Lipopolysaccharide (LPS) from E. coli, also known as endotoxin, is a potent activator of the TLR4/MD-2 cascade in mammalian cells, and extensive cleanup of proteins or plasmids for downstream use is required. Failure to remove LPS can compromise research that involves human cells as regulatory response experiments may be altered due to the presence of endotoxin. Current methods for endotoxin removal from plasmids can cost $10 or more per small-scale prep, take longer to perform, and can produce lower yields.

Instead of removing LPS contamination from plasmid DNA preparations, it is now possible to eliminate the LPS at the source. A novel E. coli strain produces clean plasmids without the need for expensive endotoxin removal methods. ClearColi™ K-12 cells are the first commercially available competent cells with a modified LPS (called lipid IVA) that does not trigger the endotoxic response in human cells.

Strains of the ClearColi series have been engineered by Uwe Mamat and Kathleen Wilke at the Research Center Borstel, Germany, in collaboration with Ronald W. Woodard at the University of Michigan College of Pharmacy and Research Corporation Technologies to remove outer membrane agonists for hTLR4/MD-2 activation.

Activation of hTLR4/MD-2 signaling by ClearColi plasmid DNA is orders of magnitude lower compared with E. coli wild-type cells. Importantly, plasmids prepared from ClearColi are innately free of endotoxic activity. The cell line is recA-, endA- for optimal plasmid recovery, and after minimal purification from ClearColi cells, plasmids can be used directly in downstream applications without eliciting an endotoxic response in human cells.

Figure 1. Comparison of post-plasmid purif-ication endotoxin unit equivalents detected from ClearColi K-12 (red bar) and DH10B (grey bar) competent cells. Plasmids were purified using a standard Qiagen Plasmid Maxi Kit. Plasmid DNA from ClearColi demonstrates sig-nificant reduction in EU/mg without the need for endotoxin-free plasmid prep kits.

Endotoxin Reduction via Competent Cell Choice

To determine the biological activity of plasmid DNA derived from ClearColi K-12 competent cells vs. normal DH10B cells, plasmid DNA was isolated via standard maxi prep method (Qiagen) from the two cell lines. No endo-free plasmid purification or cleanup methods were used. Plasmid yields for ClearColi K-12 and DH10B were very similar, with 0.4 to 0.5 μg of DNA recovered per mL culture per OD600 unit. The resulting DNA was then analyzed using the Lonza Pyrogene fluorescent assay, which is based on a recombinant form of the Limulus amoebocyte lysate assay.

Endotoxin unit equivalents detected in ClearColi-derived plasmids were less than 1% of those found in DH10B-derived plasmids (Figure 1) and were equal or less than the levels obtained from plasmid DNA prepped with Qiagen’s EndoFree Plasmid Maxi Kit (data not shown).

Direct Transfection without Endotoxin Cleanup

With the original source of endotoxin eliminated, it is now possible to transfect plasmid DNA prepared with standard Qiagen Maxiprep kits directly into human or other mammalian cell lines without concern for cell viability, altered cellular responses or poor protein expression.

To illustrate this, a plasmid (pME-HA) containing a fluorescent protein was transformed into both ClearColi K-12 and E. coli DH10B. The plasmid was then isolated from ClearColi using the standard Qiagen Plasmid Maxi Kit method, or from DH10B using Qiagen’s EndoFree Plasmid Maxi Kit. The resulting plasmids were transfected into HEK293T cells and evaluated for fluorescent protein expression (Figure 2). Using a standard (nonendo-free) plasmid prep method in combination with ClearColi, no difference in cell viability or protein expression levels are seen in comparison to endo-free preps from DH10B cells.

These results are especially significant in light of recent efforts to scale up mammalian protein expression via transient transfection. The large-scale transfection of mammalian cells allows moderate (milligram to gram) amounts of recombinant proteins to be obtained for fundamental or clinical research. The method is based on a DNA delivery step performed at high cell density by direct addition of DNA and polyethylenimine to the culture.

One of the major bottlenecks in this procedure is the production of large amounts of endotoxin-free DNA. Isolation of plasmid DNA from ClearColi can alleviate this limitation, making it more economical and producing higher yields. It should be noted that while lipid IVA is known not to induce an endotoxic response in human LPS-responsive cells, species-specific differences in the structures of TLR4 and MD-2 may allow varying degrees of sensitivity to lipid IVA in other mammalian cells.

Rapid protein expression experiments are critical to speeding the overall process of drug discovery, as is the need to produce endotoxin-free plasmids and proteins. The ability to achieve this without the need for endotoxin-removal procedures will enable researchers to reduce false positives in toxicity assays while increasing confidence in their results. Simultaneously, plasmid purification costs can be reduced by 90% or more by eliminating the need for endo-free maxiprep kits.

In addition to the plasmid production strain described here, a ClearColi BL21(DE3) strain is also available for E. coli-based protein expression, which is ideal for producing low endotoxin proteins for downstream toxicity testing. Both of these strains allow scientists to remove time-consuming and expensive cleanup steps that may affect yield, solubility, and functionality of the end product.

Figure 2. Comparison of expression of a green fluorescent protein in HEK293T cells from ClearColi-derived plasmids and standard Plamid Maxi prep (left) versus DH10B-derived plasmids and EndoFree Plasmid Maxi prep (right). The upper panels show fluorescence; the lower panels show a combined fluorescence and bright field image.

Eric Steinmetz, Ph.D. ([email protected]), is senior scientist at Lucigen.

Previous articleAGBT Report: Improving Analysis and Interpretation of NGS Data
Next articleSix Websites You Need to Bookmark: February Picks