Disposable bioreactors

One of the functions of the Laboratory of Chemical Engineering of the University of Liege is to make its expertise available to industry. The laboratory recently published the results of a hydrodynamic study inside a disposable bioreactor used for animal cell cultures in the pharmaceutical industry. Because they enable companies to avoid the rigorous process involved in cleaning and sterilizing traditional stainless steel tanks, these disposable bioreactors are being progressively viewed as serious alternatives from an economical point of view. The performance of these bioreactors still needs to be assessed, particularly considering the unusual shapes they are designed in. This is a comprehensive study which bridges the gap between fundamental research and the economic requirements of the industrial world.  

In order to develop animal cell cultures, the pharmaceutical industry uses stainless steel tanks which require fastidious and costly cleaning and sterilization. More and more manufacturers are proposing an alternative which may seem crazy: plastic disposable bioreactors. The University community is asked to quantify the performance of these bioreactors and verify that they are as efficient as traditional bioreactors. Researchers at the Chemical Engineering Laboratory of the University of Liege were recently asked by the pharmaceutical giant GlaxoSmithKline (GSK) to characterize the liquid flow inside Nucleo bioreactor which is one of the bioreactors supplied by the company ATMI LifeSciences.

GSK had good reason to exercise a certain caution. As well as being disposable, these bioreactors are parallelepiped in shape but in the pharmaceutical industry it is widely believed that a cylindrically-shaped reactor is the best solution for fluid-mixture consistency and optimal cell-development conditions. The task given to the researchers, led by Marie-Laure Collignon and Sébastien Calvo, under the direction of Dominique Toye, was to verify the hydrodynamics inside the bioreactor. The analysis was intended to establish whether, for a similar amount of energy use, the mixture of fluids was as efficient as in traditional bioreactors. The results were published in January 2015 in the Biochemical Engineering Journal(1). The results spoke for themselves. Moreover, they illustrated a methodology and a development of useful tools available for a large number of applications both in fundamental research and in applied sciences.

Bioreactors for growing cells

This is a story that has its roots in a very precise stage of industrial vaccine production. “Traditionally, stainless steel reactors are used for animal cell cultures”, continues Dominique Toye, a lecturer at the Chemical Engineering Laboratory of the University of Liege. “Clearly, these are not jet engines. It would be more accurate to compare them to tanks or to large saucepans with a capacity of between 50 liters to several cubic meters. We call them “reactors” simply because they are containers in which we perform “reactions” on an industrial scale. These reactions can be chemical, biochemical or biological”.

The animal cells are then infected by a virus to allow them to develop antibodies used in the manufacture of vaccines. In order to reproduce, the cells need a food source (glucose, mineral salts…) and oxygen, present in the culture medium which the bioreactor is filled with. “There are two imperatives for optimal growth. Firstly, these reactors must remain sterile. Otherwise other stronger microorganisms could grow more rapidly in the same conditions to the detriment of the required cells. Then the system must be as homogenous as possible. The fluid must therefore be mixed with the aid of an agitator placed in the middle of the bioreactor. This agitator, which turns on itself like a propeller, enables the liquid to have sufficient speed to mix. However, its rotation speed must also be limited for at least two reasons. Excessive agitation could damage the cells, and the other reason is down to economics. More energy is required. There is therefore an optimal balance to be found”. The cylindrical tanks with concave bottoms that are generally used enable a good quality of mixing and are easy to clean because they do not have any angles that could lead to a build-up of dirt.

“Kleenex” bioreactors: an economic alternative to stainless steel

The end of the 1990s saw the emergence of “Kleenex” bioreactors as an alternative. In the words of Dominique Toye. “The advantage of these reactors was mainly economical. The absolute necessity for a sterile environment is a real constraint. The cleaning processes for the tanks between two reactions are costly in terms of products, water, energy and time consumption. A lengthy verification process must also be put in place to ensure the absence of debris or lack of sterility in even the tiniest areas… A bioreactor that can be discarded after one use could avoid these kind of complications”. The procedure is relatively simple. There are flexible plastic pockets which are sterilized by radiation, placed in stainless steel supports and are used only once. They have the same peripheral equipment as traditional bioreactors with connections for the measurement probes for the oxygen levels, pH, entry points for regulating the environment inside, sample-taking, and finally, an agitator. These reactors, light and detachable, can then be used to move the cell culture without having to transfer the contents into another container.  

These so-called “Kleenex” reactors have not met with immediate success. The protocols in place in the production line for pharmaceutical products are very strict for obvious reasons. “Transferring equipment which was designed for the manufacture of a particular medicine to another type of equipment is not easy”, explains the researcher. “You have to be able to demonstrate that the conditions for each step involved in manufacturing are equivalent. In addition, there is always a certain distrust of new products. The industrial environment is reluctant to change procedures that work, especially when this involves replacing equipment”. But today, the disposable bioreactors seem to have proved their worth and more and more companies are turning to them. 

A rectangular bioreactor as opposed to a cylindrical one

The main reservation that GSK had about the bioreactor was its shape. This bioreactor has certainly got two original characteristics. It is parallelepiped in shape, and the agitator which looks more like a paddle or a beaver’s tail makes an elliptical movement rather than rotating around a vertical axis. It was therefore necessary to verify that, for the same energy output, the mixture in the reactor was as homogenous as that used in a cylindrical reactor whether the reactors are of the stainless steel or disposable variety. In other words, the objective was to verify that the fluid did not stagnate in any part of the reactor and that the quantity of energy deployed to agitate the fluid was at least equal to that necessary for traditional reactors. “Of course”, states Dominique Toye, “the company had already conducted trials with cell cultures and these showed promising results. As well as avoiding the disadvantage of cleaning traditional tanks, the Nucleo seemed to consume less electricity for the same performance. However, the company did not have the expertise to objectively demonstrate that the speed of liquid flow and the quality of the mixture were optimal”. In order to characterize the hydrodynamics involved, it was necessary to have a look in the interior.

Lasers applied to a study of velocimetry

In order to characterize the homogeneity of the liquid, the researchers used a velocimetry technique. “Velocity”, explains Dominique Toye, “is a movement divided by a time. To calculate the speed of the liquid in this bioreactor, we filled it with water and fine fluorescent particles which followed the flow. We needed a transparent liquid, and water presents the same flow properties (density, viscosity …) as the culture medium of the cells. Therefore there was no risk of bias. We placed two cameras in the direction of the tank in order to have a stereoscopic view which in turn enabled us to describe our observations in three dimensions. Then we used a laser to lit up a planar surface inside of the tank at very precise time intervals. With each pulsation of the laser, the fluorescent particles emitted light which made it possible for us to locate them and therefore to trace their movement. We were able to calculate the time and distance covered by the particles. We were able to calculate their speed and therefore that of the liquid. By repeating the experiment for different planes we were able to check the variation in speed according to the area of the tank and to see if the liquid had been equally distributed”. 

A satisfactory result

After having characterized the structure of liquid flow and the spatial distribution of the corresponding speeds, the quantitative analysis of the data and the comparison of this with the hydrodynamics of cylindrical tanks, the researchers were able to demonstrate that the speed of the fluid was sufficiently high in all areas of the tank. The conditions were favorable to optimal development of the cells. “Despite its unusual shape, the speeds we observed were comparable to those obtained by a traditional reactor. This rectangular form then became a great advantage as opposed to a limitation. When we are dealing with stainless steel, molding a cylindrical tank with a concave bottom is not difficult. On the other hand, for plastic reactors this is more difficult because it involves soldering the membranes in a certain way. A rectangular shape is a lot easier to manufacture and therefore less costly. But from our point of view, we were happy to be able to show that this system was not iconoclastic and that very different geometrical shapes could correspond with similar flow and mixture performances”.

Between two worlds

The study might seem very particular and to be intended for a very precise target audience but it was carried out without any diktat from the companies involved. This was one of the initial conditions, the results were able to be published freely. Beyond the positive result for this rectangular reactor, the study illustrates the procedure followed by a chemical engineering laboratory which develops basic tools at the cutting edge of science in an industry that faces economic challenges. Dominique Toye and her team are tracing a joint pathway between research and the industrial sector. If the laboratory continues to develop hydrodynamic characterization tools for reactors, service supplies similar to those in this study will become common. “Though they enable us, from an alimentary point of view, to finance our research and therefore to employ researchers, they also impose very clear demands on us. They require us to develop advanced experimental tools which can also be used for technological and practical ends in order to remain close to questions linked to industry. It is pointless to develop methods and tools that are no use to anyone. It is this perspective which drives our work in the laboratory, the overall aim is to be able to understand and characterize what happens in reactors on an industrial scale, so that the microenvironments where cells grow are favorable."

(1) Marie-Laure Collignon, Laurent Droissart, Angélique Delafosse, Sebastien Calvo, Steven Vanhamel, Roman Rodriguez, Tom Claes, Fabien Moncaubeig, Ludovic Peeters, Michel Crine, Dominique Toye, Hydrodynamics in a disposable rectangular parallelepiped stirred bioreactor with elliptic pendulum motion paddle, Biochemical Engineering Journal, Volume 93, 15 January 2015, Pages 212–221