Elsevier

Clinical Biochemistry

Volume 30, Issue 6, August 1997, Pages 455-463
Clinical Biochemistry

Analytical
Removal of Endotoxin from Recombinant Protein Preparations

https://doi.org/10.1016/S0009-9120(97)00049-0Get rights and content

Abstract

Objectives: To develop an effective method to remove endotoxin from large scale E. coli recombinant protein purifications.

Design and Methods: Triton X-114 phase separation, affinity chromatography utilizing immobilized polymyxin B or immobilized histidine, were used to remove endotoxin from purified preparations of recombinant CK-BB, CK-MB, CK-MM, myoglobin, and cardiac troponin I. Endotoxin levels were measured by a Limulus Amebocyte Lysate gel-clot assay. The immunoactivity of these protein preparations was determined by BIAcore™ analysis using a panel of in-house generated monoclonal antibodies and by a Stratus® Fluorometric Analyzer. In the case of troponin I, the BIAcore™ was also utilized to measure troponin C interactions.

Results: Phase separation with Triton X-114 was the most effective method in reducing the amount of endotoxin present in the protein preparations compared to either polymyxin B or histidine affinity chromatography. With Triton X-114, the reduction in endotoxin levels was greater than 99% and recovery of the proteins after endotoxin removal was greater than 90%. All three procedures for removing endotoxin had no deleterious effects on the immunoactivity of majority proteins when tested with a panel of monoclonal antibodies. Troponin I also retained its ability to bind to troponin C in the presence of Ca2+. Recombinant CK-BB and CK-MM which were expressed in the soluble fraction of E. coli cell lysates, contained significantly higher endotoxin levels than recombinant CK-MB, myoglobin and cardiac troponin I which were expressed in the form of inclusion bodies.

Conclusion: Of the three methods tested, Triton X-114 phase separation was the most effective way of removing endotoxin from recombinant proteins.

Introduction

Endotoxin is a lipopolysaccharide (LPS) present in the cell wall of most gram-negative bacteria including Escherichia coli. It has long been known that when gram-negative bacteria gain access to the blood stream of the mammalian host, inflammation and septic shock may occur [1]. These effects were caused by endotoxins that elicited the production and systemic release of inflammatory mediators by LPS responsive cells of the mammalian host [2]. In addition, nanogram quantities of endotoxin can affect the biochemical events of certain cells and cell-free systems [3], and can interfere with in vitro experiments. Endotoxin can readily contaminate water, aqueous solutions, and buffers. Recombinant proteins derived from E. coli are often contaminated with endotoxins since the cell wall is known to contain large quantities of lipopolysaccharides [4]. The presence of even small amounts of endotoxin from preparations of recombinant proteins can cause adverse reactions such as shock when these proteins are used as antigens for immunizing experimental animals 4, 5. Due to those adverse reactions, it is essential to remove endotoxins from drugs, injectabables, and other biological and pharmaceutical products. Some commonly used techniques for removing endotoxin contaminants are ultrafiltration [6]and ion exchange chromatography [7]. In the case of macromolecules such as protein, ultrafiltration would not be applicable while ion exchange is a less specific method. The use of polymyxin B and histidine affinity chromatography, and Triton X-114 phase separation have been more successful in removing endotoxins from protein preparations 8, 9, 10. This article evaluates these three procedures for removing endotoxin contaminants from large scale preparation of recombinant proteins purified from E. coli, in searching an effective way to reduce the endotoxin to a level that experimental animals can tolerate.

Section snippets

Chemicals and Supplies

Pyrotell™ LAL for gel-clot assay, and LAL reagent water (LRW) were purchased from Associates of Cape Cod (Woods Hole, MA, USA). Flint glass reaction tubes and Eppendorf tubes were purchased from Fisher Scientific (Toronto, ON). Sterile pipette tips were purchased from VWR (Toronto, ON). Sterile cryovials were purchased from Simport Plastics (Beloeil, Que). Histidine-sepharose, polymyxin b-sepharose, Triton X-114, and deoxycholate were purchased from Sigma Chemicals (St. Louis, MO, USA).

Reagent Preparation

LAL Gel-Clot Assay and Endotoxin Levels in Protein Buffers

The LAL gel-clot assay determines endotoxin levels by the ability of minute amounts of lipopolysaccharides to induce a gel-clot when in contact with Limulus blood. The clotting reaction is catalyzed by a cascade of enzyme activations occurring in the granules of amebocytes. The amount of endotoxin present in the test solution is directly correlated to the gel formation process.

Using the LAL assay, we found detectable levels of endotoxin in a number of buffers used to store the recombinant

Discussion

We have demonstrated that by performing three cycles of Triton X-114 phase separation, endotoxin levels in recombinant proteins derived from E. coli were reduced by as much as 99%. Further reduction can be obtained if more cycles of phase separation were performed, as shown with rcTnI. Both immunoactivity and biological activity of recombinant cTnI subjected with Triton X-114 phase separation were not affected by this process. Removal of endotoxin from rcTnI solution by affinity chromatography

Acknowledgements

We thank Margaret Dodge, Joan Mangion, and Jianying Yang for demonstrating repeatedly that Triton X-114 phase separation effectively removes endotoxin from other recombinant protein preparations and for their valuable discussions. The authors also wish to thank Julie Wiard for the CK-MB Stratus® assay.

References (22)

  • KJ Sweadner et al.

    Filtration removal of endotoxin (pyrogens) in solution in different states of aggregation

    Appl Environ Microbiol

    (1977)
  • Cited by (0)

    View full text