Sterilization of Culture Media and Supplement Solutions 14221 Sterilization by filtration

Most culture media and supplement solutions for mammalian cell culture are heat-sensitive and must therefore be sterilized by filtration, as described below. Autoclaving and steam sterilization in place are also discussed briefly in Section, for the few cases where this method is applied to solutions in mammalian cell culture.

A sterilizing filter is defined as a filter that, when challenged with the microorganism Brevundi-monas diminuta, formerly called Pseudomonas diminuta (ATCC 19146), at a minimum concentration of 107 organisms per cm2 of filter surface, will produce a sterile effluent (FDA 1987). Viruses are not included in the definition. It should be noted that this definition represents a worst case, since no filter is expected to be exposed to this level of challenge during an actual cell culture process. Additional regulatory requirements are that the membrane should be sterilizable by steam or y-rays, and should not affect the original product attributes. All filter materials in contact with the product or medium should therefore have minimal particle shedding during operation, a low level of extractables, pass USP Class VI toxicity testing and should not adsorb any component to a significant extent (Docksey et al. 1999).

Sterilizing filters for liquids are made of a hydrophilic membrane that retains, on its surface, mostly particles that are larger than the pore size, as a result of a sieving (or size exclusion) mechanism. Additionally, particles can be retained by non-specific adsorption and bridging due to accumulation of particles around an open pore (Raju & Cooney 1993). Membrane filters are often called 'absolute'; however, due to the fact that sieving is not the only retention mechanism and that these filters have a pore size distribution that is often unknown, the term 'absolute' is considered improper (Stinavage 2003).

Originally, sterilizing-grade filters were 0.45 |im-rated membranes until B. diminuta was identified in the 1960s and found to be small enough to penetrate these membranes at high challenge levels (Jornitz et al. 2002a); 0.2 or 0.22 |im-rated membranes then became the standard for sterilizing filters. More recently, several studies have shown that microorganisms such as B. diminuta can actually penetrate even these membranes. For example Sundaram et al. (2001a, b) found that several bacteria from 'natural' water sources could penetrate 0.2/0.22 |im-rated membranes at challenge levels lower than 107 CFU/cm2. Bacteria with a width smaller than 0.3 |im were actually not limited in their ability to penetrate the filter; as the bacterial width increased above this value, penetration depth into the membranes decreased almost exponentially with size. Interestingly, these penetrating bacteria were on average 20-40 % wider and 40-70 % longer than B. diminuta (length 0.88 |im, width 0.31 |im, as determined under the specific experimental conditions). These findings confirm that sieving is not the sole mechanism governing bacterial retention. Most of these bacteria were Gram-negative, oxidase positive, and closely related to B. diminuta. They can be considered to be common environmental or ubiquitous organisms. Interestingly, in the latest FDA guidance on sterile products (FDA 2004), the new proposed definition of sterilizing filters does not mention the type and concentration of challenge microorganisms any more; the filter should simply 'remove all microorganisms from a fluid stream, producing a sterile effluent'. For challenge tests, the use of B. diminuta at a concentration of at least 107 CFU/cm2 is only suggested.

Similar experiments with 0.1 |im-rated filters revealed significant performance differences between different types, in terms of microbial removal efficiency (Sundaram et al. 2001c); actually only those filters that had been qualified with both B. diminuta and Acholeplasma laidlawii (related to mycoplasma) consistently produced sterile effluents, whereas the other 0.1 |im filters could not prevent the penetration of bacteria. The authors therefore recommended that, apart from knowing the viable bioburden level in the process, one should also try to identify the possible microorganisms present. If there is evidence that the bioburden does not contain any microorganisms that can penetrate the filter more easily than B. diminuta, then the use of 0.2/0.22 |im-rated filters may be sufficient. Otherwise, when the presence of mycoplasma or of bacteria with a similar or even slightly larger width than B. diminuta cannot be excluded (this is likely to be the case with culture media), tighter, i.e. 0.1 |im-rated, filters should be used. Since there is currently no industry-wide or regulatory standard for rating 0.1 |im filters, those that have been functionally qualified are recommended for instance with, A. laidlawii (Meeker et al. 1992) or Hydrogenophagapseudoflava, formerly Pseudomonas pseudoflava (ATCC 700892) (Sundaram et al. 2001d), a bacterium that is slightly longer but narrower than B. diminuta. Currently there is, however, no consensus on the benefits of systematically replacing for sterilization all 0.2/0.22 |im-rated filters by 0.1 |im-rated ones (Jornitz et al. 2002a).

The most common membrane materials for sterilizing filters for culture media are PVDF, nylon-66 and polyethersulfone; these materials have been modified chemically to become hydrophilic. They typically show steam sterilization resistance up to 134°C. Only a few supplement solutions, such as lipids or vitamins dissolved in ethanol, require a hydrophobic filter in PVDF or PTFE. In this case, special care should be taken to ensure that these chemical compounds do not adsorb significantly to the membrane; saturating the membrane with the solution prior to filtration or flushing the membrane with adequate amounts of ethanol after filtration are recommended.

Sterile filtration is usually performed in a flow-through mode; at small scale, flat disk membranes with a diameter in the range of 13-293 mm (1.3-670 cm2) in a disposable capsule are used. At intermediate scales, cartridges made of a stack of membrane disks in a pre-assembled disposable capsule are commonly used, providing a filtration area in the range of 100-2000 cm2. At larger scales, pleated membranes are used; the membranes are supported on a non-woven polyester support, folded to form pleats, wrapped around an inner core, and sealed by two end caps. Support cages are usually placed around the membrane to protect it from mechanical damage (Jornitz et al. 2002b). Cartridges are available in a variety of sizes (from 2 to 40 inches nominal length), one

Figure 14.2 Filter cartidges and housings for sterization of gases and liquids. From left to right: 10" cartridge for gases with a collector for condensates; 20" cartridge for liquids (e.g. culture media); 10" cartidge for utilities (e.g. WFI), to be mounted horizontally in line (courtesy of Millipore).

Figure 14.2 Filter cartidges and housings for sterization of gases and liquids. From left to right: 10" cartridge for gases with a collector for condensates; 20" cartridge for liquids (e.g. culture media); 10" cartidge for utilities (e.g. WFI), to be mounted horizontally in line (courtesy of Millipore).

10-inch cartridge providing an effective filtration area of around 0.7 m2. Cartridges are usually installed in a fixed stainless steel housing (316L grade, Dillon et al. 1992) prior to sterilization, as illustrated in Figure 14.2. Nowadays, they can also be purchased pre-assembled and sterilized in a disposable capsule, eliminating the need for cleaning and sterilization of the housing, as well as for the associated validation studies. Filtration can be performed either at a constant differential pressure (typically 1-1.5 bar) or at a constant flow rate. In the first case, the feed solution is pressurized with a compressed gas; in the second case, a pump is connected on the upstream side of the filter. During filtration, pressure or flow rate as well as temperature are commonly measured continuously.

The use of a pre-filter to remove larger particles is recommended, particularly if the medium contains protein hydrolysate or serum, and if 0.1 |im sterilizing filters are used; the required surface area of the sterilizing filter can thus be minimized. Depth filters are typically used for this purpose, thanks to their ability to retain large loads of particles before clogging; they consist of a porous structure of fibres, which form an irregular three-dimensional net. Typical filter materials are polypropylene or cellulose acetate ester fibres, sometimes combined with glass microfibres in a double-layer structure; the membrane is supported by a polypropylene cage. Electrostatic interactions, adsorption, diffusion and impaction of the particles on the filter material are the main retention mechanisms; consequently, particles smaller than the characteristic pore size or dimension of the filter can be captured. These are actually retained not only at the surface (as with membrane filters) but also inside the filter (Raju & Cooney 1993).

Was this article helpful?

0 0
Healthy Chemistry For Optimal Health

Healthy Chemistry For Optimal Health

Thousands Have Used Chemicals To Improve Their Medical Condition. This Book Is one Of The Most Valuable Resources In The World When It Comes To Chemicals. Not All Chemicals Are Harmful For Your Body – Find Out Those That Helps To Maintain Your Health.

Get My Free Ebook

Post a comment