Faculty Advisor

Clark, William M.


Under the direction of Professor William M. Clark, past MQP groups in the WPI Chemical Engineering Department constructed several two-phase electrophoretic protein separation devices. These devices use a mix of Dextran/polyethylene glycol to form two aqueous phases with a clearly defined, stable interface. The two-phase system provides initial separation of different chemical solutes into either phase. An electric field is applied perpendicularly to the interface to further separate charged species into either phase. The purpose of these two-phase electrophoretic devices is to recover and begin the purification of recombinant proteins produced by Escherichia coli. With additional research, these devices can potentially replace several existing unit operations in protein purification, and to reduce operational cost. Some of the unit operatiions to be replaced serve a secondary purpose of endotoxin removal. It is therefore of interest to know if the two-phase system is capable of removing endotoxin from desired protein. In this project the endotoxin removal capability of a batch-wise aqueous two-phase electrophoretic protein separation device was demonstrated. Escherichia coli, strain K-12 W3110, was grown and lysed as the source of endotoxin. Dextran and polyethylene glycol were used to form the two-phase medium. A factorial experiment design bracket was used with varying cell contentration, system voltage, and run time. However, not every sample could be assayed for endotoxin concentration due to limited number of assays available. Lymulus amoebacyte lysate was used as the assay reagent. The project demonstrated that the soluble endotoxin favors the bottom phase over the top phase 10 to 1, or if accounting for the cell sediments that accumulated on the bottom, 20 to 1. This proves the electrophoretic device is capable of removing endotoxin from the desired protein. However, the desired top phase still contained endotoxin concentration of 6000 EU/mL. Typical release limits in pharmaceutical preparations range anywhere between 0.02 EU/mL to 1666 EU/mL. More work is needed to determine whether or not the two-phase system, by increasing the run time, and/or increasing the system voltage, could further reduce endotoxin concentration to clinically acceptable levels.


Worcester Polytechnic Institute

Date Accepted

January 1999


Chemical Engineering

Project Type

Major Qualifying Project


Restricted-WPI community only

Advisor Department

Chemical Engineering