Optimising cell culture

30-May-2012

Microcarriers have become an essential tool in the scale-up and manufacture of many cell-based processes, but creating the optimal environment for growth for each individual cell type and culture to ensure good cell yields and quality remains a challenge. Microcarriers can be created as either microporous or macroporous substrates; the former contain very tiny pores and the latter pores that are sufficiently large for the cells to grow into them. While macroporous microcarriers have a larger surface area, harvesting the cells can be difficult. But with microporous microcarriers the cells cannot grow into the pores, which facilitates cell harvest.

With the growth in biologicals and vaccines, production of cell cultures in large volumes can be a rate limiting step in the manufacturing process. This recent study by ATMI and SoloHill illustrates the beneficial use of microcarriers in conjunction with disposable bioreactors.

Microcarriers are small particles, typically 70µm to 1000µm in size, and come in a variety of shapes, from simple spheres to elongated rods. Since their introduction in the late 1960s, microcarriers have progressed to become an essential tool in the scale-up and manufacture of many cell-based processes. They are made from a range of materials, such as alginates, collagen, dextrin or polystyrene, and used in a wide array of applications with a diverse range of cell types.

Microcarriers can be used for the production of biologicals and vaccines or for expanding cells used in drug and toxicity testing. They can also be used to generate stem cells for cell therapy and regenerative medicines. However, these applications require very different conditions and pose a serious challenge: creating the optimal environment and conditions for cell growth for each individual cell type and culture to ensure good cell yields and quality.

While there have been attempts to modify cells so that they can be grown in suspension rather than on a surface, the degree of success is variable

Most mammalian cells require a surface on which to grow. While there have been attempts to modify cells so that they can be grown in suspension rather than on a surface, the degree of success is variable; a number of cells will still require a surface on which to adhere. This poses a problem for growing large numbers of cells because they require a large surface area for expansion. Microcarriers could provide an answer as they provide a large surface area for the growth. Moreover, microcarriers enable the use of bioreactors, which can produce large numbers of these adherent cells under cGMP conditions.

tailored carriers

Michigan-based microcarrier manufacturer SoloHill Engineering supplies spherical micro-carriers for use with different cell types. The firm’s standard microcarrier is 125–200µm in diameter, a size that works well as a substrate for growing mammalian cells in stirred tank bioreactors. These microcarriers are made from modified polystyrene, a material similar to standard cell culture platforms, such as cell culture dishes or roller bottles used for propagating adherent cells. Surface modifications on the microcarriers can be tailored to support growth of different cell types.

Microcarriers compact a large surface area into a small volume by moving from a flat surface to a sphere, and going from 2-D to 3-D, so the surface area within the same reactor volume increases dramatically. This enables large cell culture expansion to be carried out in a space that is much smaller than would be the case for the same style of equipment used in the lab, such as roller bottles. But harvesting the cells from the microcarriers can also pose a challenge. Microcarriers can be created as either microporous or macroporous substrates; the former contain very tiny pores and the latter pores that are sufficiently large for the cells to grow into them.

While macroporous microcarriers have a larger surface area, harvesting the cells is a significant challenge. For products such as stem cells, this has the additional drawback of making it much harder to control the environment; stem cells may start to interact with each other within local niches created by the microcarrier pores, leading to potential problems with homogeneity and undifferentiated cell population.

With microporous microcarriers, such as those made by SoloHill, the cells cannot grow into the pores. This design facilitates cell harvest and the same enzymatic methodologies employed in standard cell culture can be used to harvest cells efficiently. Cells can be separated from microcarriers either by gravity or by passing the microcarriers/cell suspension over a screen. This process captures the microcarriers but allows the cells to pass through.

Another important consideration when scaling up cell growth on microcarriers is the choice of bioreactor

Another important consideration when scaling up cell growth on microcarriers is the choice of bioreactor. Traditionally, bioreactors are cylindrical systems with top or bottom-mounted impellers and the design remains unchanged whether the bioreactor has a volume of a few litres or thousands of litres. This can cause difficulty when looking to achieve efficient mixing in cylindrical reactors, particularly when the mixing must be sufficiently gentle to avoid damaging cells. Typically, in this situation, baffles are incorporated into the reactor design to improve the mixing and avoid a deleterious situation.

A novel solution for this problem is to use a cubic bioreactor with a paddle mixer, which moves around the vessel in a cylindrical fashion. The corners provided by the cubic design naturally function as efficient baffles, which facilitate the gentle mixing required for successful microcarrier processes.

The recent emergence of this type of technology in a disposable bioreactor format also offers a cost-effective solution to the challenges. Standard stainless steel bioreactors require cleaning between bioprocesses, which extends time between production runs and increases costs. Single-use reactors arrive at the manufacturing site sterile, so no preparation is needed prior to use and no cleaning afterwards as the vessel is disposed of rather than reused.

Single-use reactors arrive at the manufacturing site sterile, so no preparation is needed prior to use and no cleaning afterwards as the vessel is disposed of rather than reused

When exploring market options to address these challenges, SoloHill discovered that the ATMI LifeSciences Integrity PadReactor (see Figure 1) offered a solution to many bioreactor shortcomings. The PadReactor single-use system features a cube-shaped bag that integrates an internal paddle mixer with a sparging system. The sparger’s location on the end of the paddle enables sparging and mixing to be carried out concurrently. This design provides an environment for efficient cell growth on microcarriers as it is possible to achieve very slow mixing speeds and efficient gas transfer.

ATMI and SoloHill partnered to complete formal studies that demonstrate the successful use of SoloHill’s collagen-coated microcarriers in the ATMI system. The single-use, cube-shaped bioreactor vessel offers an open-architecture controller platform that provides flexibility to use either the companion control system offered by ATMI or another controller of the user’s choice. With a specific gravity of 1.03 and a diameter of 125–212µm, the microcarriers were maintained in suspension with a gentle, low stirring speed and provided an excellent substrate for cell growth. Importantly, these solid microcarriers can be sterilised either by autoclaving at temperatures up to 131°C, or via gamma irradiation at doses of 25–40kGy with no deleterious effects on function.

In the test procedure, a 25L, single-use PadReactor bioreactor retrofitted with a perfusion filter, was used in conjunction with SoloHill collagen-coated microcarriers at a final concentration of 30g/L. Glass spinner flasks containing 200ml of media were used for serial passaging (the virus attenuation technique) and the test cells were Vero monkey kidney cells.

Figure 2: PadReactor Vero cells on Microcarriers

Figure 2: PadReactor Vero cells on Microcarriers

To evaluate the mixing capabilities of the single-use bioreactor system with these microcarriers, the microcarrier distribution was mapped at various impeller speeds by collecting samples from multiple positions and depths in the bioreactor vessel. Samples from the top, middle and bottom of the bag at all four corners were retrieved and dried overnight in an oven, and the resultant microcarrier weight for each sample was measured on a material balance. To determine % mixing, the empirically determined mass of microcarriers in each sample was compared with the microcarrier concentration that would be predicted for 100% mixing.

The temperature was maintained at 37°C, with a continuous agitation rate of 35–45rpm, dissolved oxygen (DO) concentration was maintained at 30–40% of air saturation, and pH was controlled at 7.0–7.4. During the culture period, DO was controlled by oxygen overlay at 100ml/min, as well as through a 2mm macrosparger that was fixed on the mixing paddle and running at 100–200 ml/min. The pH of the culture was controlled by air overlay and the addition of 2.5N NaOH. Media perfusion began two days after culture initiation, and continued through to day six at about 150% volume changed per day, with a total utilisation of 92L.

On the day of the culture initiation, 450g of the collagen microcarriers (equivalent to a surface area of 162,000cm2) were pre-autoclaved in 900ml of deionised water, and added to the disposable bioreactor bag containing 12.5L of complete media without foetal bovine serum (FBS). The bioreactor was then seeded at 40rpm with a single cell suspension of 12.23 x 106, at a final concentration of 0.2 x 106 cells/ml, which equates to about 16 cells per microcarrier, when the DO content of the medium was 80% of air saturation, and the pH was 7.45.

Once cells were attached to more than 90% of the collagen microcarriers, FBS was added to the culture to a final concentration of 5%, and the final volume of culture, comprising microcarrier plus media, was adjusted to 15L by the addition of complete medium without the serum. After the cell seeding, both the DO concentration and pH drifted down within the bioreactor operating range, the DO concentration to 30–40% of air saturation within 13 hours, and the pH to 7.0–7.4 within 15 minutes. These conditions were maintained for the remainder of the run.

Bioreactor culture samples were retrieved daily at 40–50rpm to evaluate cell growth on the microcarriers. Prior to sampling, the sample line was purged with culture, and a 20ml sample was then immediately collected for counts. The immobilised cell density was estimated by cell lysis with an aqueous Triton X-100 (0.5%) and 0.1M citric acid solution. The released intact cell nuclei were counted with a Cellometer Auto T4 cell counter.

SoloHill microcarriers with a relative density of 1.03 exhibited an optimal distribution between 30–50rpm. At these speeds, the PadReactor bioreactor’s mixing performance was very high, and these conditions were chosen for subsequent cell growth studies. Microscopic examination of a representative sample retrieved from the bioreactor one hour after culture initiation revealed that cells had attached and begun to spread onto more than 90% of the microcarriers. At day one, it was apparent that cells had begun to grow on about 90% of the microcarriers. By 72 hours most microcarriers were confluent, and at day seven (Figure 2) the Vero cells had grown 5.9 generations, reaching a cell concentration of 11.2 x 106 cells/ml.

To demonstrate the feasibility of serial passaging high density cultures, cells harvested from an aliquot of the entire bioreactor culture were used to seed a 250ml spinner containing fresh collagen microcarriers. The viability of this cell suspension was 98%, and cells efficiently attached and spread on microcarriers when seeded into the spinner. A conservative estimate using cell numbers obtained from the bioreactor sample indicate that this 15L culture could be used to seed at least one 300L bioreactor containing an equivalent microcarrier concentration of 30g/l, providing a spit ratio of 1:20. This estimate assumes an 80% cell recovery obtained from the entire bioreactor contents during harvesting.

Monitoring spinner cultures seeded with bioreactor-derived cells revealed that the cells attached and spread on more than 90% of the collagen microcarriers one day after seeding, and by 72 hrs had grown to near confluence (see Figure 3).

Figure 3: Regulation of dissolved oxygen (blue) and pH (red) during cultivation

Figure 3: Regulation of dissolved oxygen (blue) and pH (red) during cultivation

The success of this trial demonstrated the compatibility and performance of SoloHill’s collagen microcarriers in ATMI’s disposable bioreactor. In addition, the data show that cells derived from the high-density culture can be serially passaged to fresh microcarriers for cell expansion. The microcarrier and disposable bioreactor combination represents an attractive platform for generating large numbers of cells in a small footprint. The resultant cells could be harvested and used for subsequent passage into a larger, independent reactor containing microcarriers, or with proper developmental efforts for cell expansion in a single reactor through addition of fresh microcarriers and media.

Additionally, the efficient attachment obtained with serum is not present in the medium. The rapid growth profile observed upon subsequent cultures with serum indicates that the cells could be infected for virus production after as little as three days of culture.

These results have laid the foundation for further studies exploring virus production in this platform, in both serum-containing and animal component free systems. It also provides an attractive platform for the expansion of cells required for seeding large bioreactors used for vaccine production.

Companies