Many types of commercially valuable recombinant proteins produced by fermentation are expressed at high levels in Escherichia coli. Often, high-level expression in the host results in the formation of insoluble inclusion bodies. The release of these intracellular inclusion bodies from E. coli following cell disruption is a requirement for further downstream recovery. The ability to discern between intact unruptured cells and granules released from broken cells can provide valuable information for improving recovery yields in downstream purification. This paper describes a rapid and sensitive cytometry-based method that allows the simultaneous measurement of intact heat-killed E. coli and inclusion bodies using staining with nucleic acid binding fluorochromes.

Introduction

The production of therapeutic proteins using Escherichia coli as a host organism typically involves growth of the cells in a bioreactor, induction of gene expression and protein production, recovery of cells containing inclusion bodies, cell disruption, inclusion body recovery, and subsequent protein purification via chromatography (1,2). The ability to maximize productivity of the manufacturing process depends upon reliable, sensitive, and rapid methods to assess the efficiency of these unit operations. Although a multitude of methods exist to assess product accumulation during fermentation, relatively few methods measure the disruption of the cells and subsequent recovery of inclusion bodies.

Several biophysical methods including electrical sensing zone (ESZ) methods, photon correlation spectroscopy (PCS), electron microscopy, and centrifugal disc photosedimentation (CDS) have been used by investigators in attempts to characterize and quantify these biological particles (3). These methods, however, lack sensitivity, suffer from lysed cell component interference, require significant sample pretreatment, or are not conducive to obtaining statistically significant numbers of events.

Phase contrast microscopy, which enables whole, intact-cell counting relative to free inclusion bodies, serves as the current gold standard for measuring cell disruption efficiency. The small size of both host cells (nominal 3–5 µm × 1 µm) and inclusion bodies (∼1 µm) requires operators to observe several microscope fields containing <100 objects per field at high power (1000×) to produce satisfactory results. Given these constraints, this microscopic method is time consuming, tedious, and subject to a high degree of error.

Flow cytometric methods designed to assess bacterial cell viability, including E. coli (4,5,6,7,8,9) have also been described. These live/dead cell assays often rely upon fluorochromes with various membrane permeability properties that often target bacterial DNA. The list includes 4′, 6-diamidino-2-phenylindole (DAPI), propidium iodide (PI), ethidium bromide, BacLight Green and Red, and acridine orange staining (10,11). More recently, cell-impermeant cyanine dyes, such as SYTOX Green, that have a higher affinity for nucleic acids while providing exceptionally bright signals, have been developed (12,13,14). These dyes easily penetrate cells with compromised plasma membranes, yet will not cross the membranes of live cells. Binding of SYTOX Green stain to nucleic acids results in a 500-fold enhancement in fluorescence emission (absorption and emission maxima at 502 and 523 nm, respectively), rendering bacteria with compromised plasma membranes brightly green fluorescent. SYTOX Green enabled superior detection and discrimination of intact versus permeabilized cells as compared with PI in flow cytometric analysis (14).

While several of the staining methods described above have been used to enumerate live/dead E. coli, there have been relatively few flow cytometric methods described for the enumeration of inclusion bodies (15,16,17,18). These methods have been used to assess inclusion body formation during fermentation and are largely based upon light scattering (15,16). The lack of methods may be due to the physical size of the inclusion bodies (∼1 µm diameter), which is near the threshold for detection based upon light scattering in modern flow cytometers. Light scattering methods are complex, require precise tuning of the optical systems, and for the reason stated above are prone to error. A flow cytometric method using fluorescently labeled antibodies for detection of inclusion bodies has also been described (18). This method, however, is both time consuming and tedious and requires the generation of highly specific antibodies for the expressed proteins. We recently developed a rapid and sensitive flow cytometric method, based upon nucleic acid binding fluorochromes, with the capability to resolve intact heat-killed cells from free inclusion bodies in a homogenized preparation.

Materials and Methods

Microorganisms

Recombinant E. coli produced by fed-batch fermentation were heat-killed at 70°C for 5 min, harvested, and concentrated by continuous disc centrifugation at 8000 × g. The centrifuged material was then passed through a high-pressure homogenizer at 8000 psig (pound-force per square inch gauge) for the indicated number of passes. Inclusion bodies were harvested from the broken cells and separated from cell debris via continuous disc centrifugation at 8000 × g. The null strain consisted of the host strain of E. coli K12, which does not express the recombinant protein.