Is Membrane Chromatography the Answer?

by Dr. Scott Rudge

Membrane chromatography gets a fair amount of hype.  It’s supposed to be faster, cheaper, it can be made disposable.  But is it the real answer to the “bottleneck” in downstream processing?  Was Allen Iverson the answer to the Nugget’s basketball dilemma?  I’m still skeptical.

The idea to add ligand functionality to membranes was not new at the time, but the idea really got some traction when it was endorsed by Ed Lightfoot in 1986.  Lightfoot’s paper pointed out that the hydrodynamic price paid for averaging of flow paths in a packed bed might not be worth it.  If thousands of parallel hollow fibers of identical length and diameter could be placed in a bundle, and the diameter of these fibers could be small enough to make the diffusion path length comparable to that in a bed of packed spheres, or smaller, then performance would be equivalent or superior at a fraction of the pressure drop.  This is undoubtedly true; there is no reason to have a random packing if flowpaths can be guaranteed to be exactly equivalent.  However, every single defect in this kind of system works against its success.  For example, hollow fibers that are slightly more hollow will have lower pressure drop, lower surface to volume ratio, lower binding capacity and higher proportional flow.  Slightly longer fibers will have slightly higher pressure drop, slightly higher binding capacity, carry proportionally less of the flow.  Length acts linearly on pressure drop and flow rate, but internal diameter acts to the fourth power, so minor variations in internal diameter would dominate performance of such systems. 

Indeed, according to Mark Etzel, these systems were abandoned as impractical for membrane chromatography based on conventional membrane formats that have been derivatized to add binding functionality.  As this technology has been developed, its application and scale up has begun to look very much like packed bed chromatography.  Here are some particulars:

1.       Development and scale up is based on membrane volume.   However, breakthrough curves are measured in 10’s, or even 70’s of equivalent volumes (see Etzel, 2007) instead of 2’s or 3’s as found in packed beds

2.       Binding capacities are less in membrane chromatography.  In a recent publication by Sartorious, the ligand density in Sartobind Q is listed as 50 mM, while for Sepharose Q-HP it is 140 mM.  In theory, the membrane format has a higher relative dynamic binding capacity, but this has yet to be demonstrated (see above)

3.       The void volume in membranes is surprisingly high, at 70%, compared to packed beds at 30%.  This is a reason for the low relative binding capacity.

4.       Disposable is all the rage, but there’s no evidence that, on a volume basis, derivatized membranes are cheaper than chromatography resins.  In fact, economic comparisons published by Gottshalk always have to make the assumption that the packed bed will de facto be loaded 100 times less efficiently than membranes, just to make the numbers work.  The cost per volume per binding event goes down dramatically during the first 10 reuses of chromatography resins.

It turns out that membrane chromatography has a niche, and that is for flow-through operations in which some trace contaminant, like the residual endotoxin or DNA in a product is removed.  This too can be done efficiently with column chromatography when operated in a high capacity (for the contaminant) mode.  But there is a mental block among chromatographers who want to operate adsorption steps in chromatographic, resolution preserving modes. This block has not yet affected membraners.  A small, high-capacity column operated at an equivalent flowrate to a membrane (volumes per bed or membrane volume) will work as well, and in my opinion more cheaply if regenerated.

These factors should be considered when choosing between membrane and packed bed chromatography.

Assessing rapid viral enumeration/detection systems

By Dr. Ray Nims

In a previous posting, we alluded to the recent availability of rapid methods for identification of viruses. These technologies, together with rapid methods for enumerating viruses, should greatly expedite the quantification and identification of viruses (and bacteriophage) as compared with the existing cell culture-based approaches.

Rapid enumeration technologies are intended to replace the cell-based infectivity endpoints such as plaque assays or tissue-culture infectious dose assays, which typically require 7-10 days for completion. The use of the rapid methods may be appropriate in cases where it is not necessary to determine the infectious titer of a virus stock. An example of this might be for monitoring the amplification of viruses for preparation of live or subunit vaccines. The particle enumeration technologies include those that specifically measure viral particles and those that measure particles in general. As shown in Table 1, the particle enumeration technologies are not specific to any given virus. These are not generally useful, therefore, for viral identification, although the particle detection method associated with the NanoSight system does allow for sizing of the particles. Viral particle size is a key attribute to be aware of when, for instance, attempting to identify an unknown viral contaminant.

Table 1. Characteristics of rapid viral enumeration/identification technologies

The quantitative polymerase chain reaction (Q-PCR) and universal biosensor (Ibis T5000) technologies represent approaches that are capable of providing information both on the relative quantity of a virus in a sample and its identity. The important difference between the two is that in the former case (Q-PCR), the user is typically evaluating the identity and or quantity of a virus which is reactive with the specific primers and probes used in the assay. From an identification standpoint then, the Q-PCR technique has typically been used to confirm whether an unknown virus is related to the virus for which the assay primers and probes was designed. The degree of relatedness required is determined by the specific primers and probes used in the assay, and may be either to the genus level or the species level. Efforts are being made to incorporate primers for more highly conserved sequences to allow for more broad coverage in Q-PCR assays intended for viral screening. In the case of the universal biosensor (Ibis T5000), an unknown virus in a sample may be simultaneously identified and quantified, as long as the virus is or is closely related to one for which mass spectrometry information is present in the software used for assay analysis. Quantification in either case is in genomic units, and as with the particle enumeration methods, the readout of the quantitative nucleic acid methods does not indicate whether the virus detected is infectious. An additional nucleic acid-based method that may prove useful, in cases where relatively rapid identification of an unknown viral contaminant is needed, is deep (massively parallel) sequencing. This method is more labor intensive (and perhaps costly) then the other quantitative nucleic acid methods described above, but has the advantage that it can provide information regarding the completeness (partial vs. full-length) of the viral genomic sequences detected. This approach has displayed utility in identifying a novel picornavirus in harbor seal samples, porcine circovirus in rotavirus vaccines, and a new parvovirus in bovine serum.

Microarray screening is a technology that may be used to rapidly identify (but not enumerate) an unknown virus in a sample, provided that a probe for the virus is part of the microarray chip. Some microarray chips intended for viral identification also contain probes for conserved viral genomic sequences. In this case, the microarray may identify a novel unknown virus, at least to the genus level. As with the other rapid methods that are based on presence of specific genomic material, the assay cannot discriminate between infectious and non-infectious virus.

See Table 1 for some of the important characteristics and limitations of each method. The use of the rapid methods discussed above and in Table 1 should reduce the time needed for viral quantitation from weeks to hours, and for identification of an unknown contaminant in a sample from months (or years) to one or more days. This should greatly facilitate the monitoring of viral proliferation in manufacturing processes and the investigation of viral contamination events.

The inactivation literature for circoviruses

by Dr. Ray Nims

The Circoviridae family of viruses represent an extreme case for small, non-enveloped viruses. We have posted previously that the latter group constitutes a high risk for manufacturers of biologicals due to the difficulty of eradicating the viruses from raw materials or from a contaminated facility.  At 17-25 nm particle size, the circoviruses are among the smallest of the animal viruses. These viruses represent more of an economic threat than a threat to human health. The porcine circoviruses (PCV-1 and PCV-2) and the chicken anemia virus are well known examples of the circoviruses and these have been studied extensively due to their impact on the swine and poultry industries. So why care about the circoviruses in the biologics industry?

We begin being concerned about porcine circoviruses in the context of xenotransplantation of porcine tissues (e.g., islet cells) into humans. The worry was that a porcine circovirus might be transmitted to a patient via the porcine donor tissue. More recently, PCV genomic sequences were discovered in rotavirus vaccines manufactured by GlaxoSmithKline and Merck. The presence of the genomic material was attributed to the use of porcine trypsin during the culture of the cell substrates in which the vaccines were manufactured (see previous post).

As a result of the heightened awareness of the contamination threat posed by the porcine circoviruses, infectivity assays for these viruses are now being offered at contract testing organizations (e.g., BioReliance, MICROBIOTEST, and WuXi Apptec), for raw material testing, cleaning efficacy testing, and for evaluating the clearance of the circoviruses by purification processes. As might be expected based on our experience with other small, non-enveloped viruses, inactivation approaches that are effective against many larger non-enveloped  or enveloped viruses have little efficacy for the circoviruses.

So how much do we actually know about the inactivation of  circoviruses? The literature on the subject is fairly extensive for PCV-2 and for chicken anemia virus, if one is willing to spend time digging deeply. I have done the digging, and have assembled the literature into the following categories of inactivants: heat, irradiation, and disinfectants/chemicals. Keep in mind that the inactivation potential of the physical or chemical agents depends greatly upon the matrix in which the virus is present as well as the temperature and contact time with inactivant. The matrices evaluated varied for the studies described below, and the reader is directed to the individual papers for this critical detail. In addition, some variability in results may be expected because of the relative difficulty in assaying infectivity of the circoviruses.

A number of studies on the thermal stability of the circoviruses have been published. In general, it appears that 15 or more minutes of exposure to wet heat (heating of viruses spiked into solutions) at temperatures ≥80 ◦C should provide extensive inactivation (3-5 log₁₀) of circoviruses. Dry heat (heating of freeze-dried virus and, by implication, viruses on the surfaces of coupon materials) appears to be much less effective, resulting in <1.5 log₁₀ inactivation even at temperatures as high as 120 ◦C . The results of at least one investigator suggest that a temperature of 95 ◦C will be sufficient for high temperature short time (HTST) treatment for mitigating the risk of introducing a circovirus in a process solution.

A number of studies on the inactivation of the circoviruses by disinfectants and other chemicals have appeared in the literature, reflecting the relatively great economic threat of the circoviruses to the swine, poultry, and exotic bird industries. While many of the chemicals/disinfectants had little efficacy for inactivation of the circoviruses (as might be expected for a non-enveloped virus), certain treatments appear to have been highly effective. These included the following: a) glutaraldehyde at 1% or 2% and 10 or more minutes contact time; b) sodium hypochlorite at 6% and 10 or more minutes contact time; c) sodium hydroxide  at 0.1 N or greater; d) Roccal® D Plus at 0.5% and 10 minutes contact time; e) Virkon® S at 1% and 10 minutes contact time; f) β-propiolactone at 0.4% and 24 hours contact time; and g) formaldehyde at 10% and 2 hours contact time. Variable results were obtained for the iodine-containing disinfectants. These ranged from <1.0 log₁₀ inactivation for iodine (10%; 30 minutes contact time; 20 ◦C) to ≥5.5 log₁₀ for Cleanap® (1%; 2 hours contact time; 37 ◦C). A third study involving an iodine-containing disinfectant, FAM®30 (Biocide30) used at 1% or 2%; 30 minutes contact time, and 10 ◦C temperature demonstrated ≥3.5 log₁₀ inactivation.

The literature on inactivation of circoviruses by irradiation is scant, to say the least. Plavsic and Bolin showed that gamma irradiation of PCV-2-spiked fetal bovine serum at the radiation doses normally employed for serum treatment (30 and 45 kGy) resulted in ≤1.0 log₁₀ reduction in virus titer. Gamma irradiation (at the doses normally used) appears to be relatively ineffective for inactivating the very smallest of the viruses (parvoviruses and circoviruses) in serum so this result is perhaps not surprising. One approach that offers hope for inactivation of circoviruses is ultraviolet radiation (specifically UV-C) treatment, as this technology appears to be effective for smaller non-enveloped viruses such as the parvoviruses and caliciviruses. I expect that studies to demonstrate efficacy of UV-C for inactivating circoviruses will be performed in the near future.

In summary, there is ample evidence in the literature that effective inactivation approaches exist for the circoviruses. Careful selection of inactivation technologies that is based on the body of evidence accumulated by workers in the swine and poultry industries should enable appropriate risk mitigation and facility cleaning strategies to be adopted in the biologics industry.

<This material was excerpted in part from Nims and Plavsic, Bioprocessing J, 2012; 11:4-10>

Porcine circoviruses, vaccines, and trypsin

By Dr. Ray Nims

It has now been more than a year since the announcements by GlaxoSmithKline (GSK) and Merck of the presence of porcine circovirus (PCV) genomic material in their rotavirus vaccines.

The presence of the PCV viral sequences was, in both cases, provisionally attributed to the use of porcine trypsin during the culture of the cell substrates used in the manufacture of the vaccines. It has been reported that the genomic sequences were associated with low levels of infectious PCV in the GSK vaccine.     

As mentioned in a previous posting, an expected outcome of these disclosures was heightened regulatory expectations, going forward, for PCV screening of porcine raw materials and of Master and Working cell banks which were exposed to porcine ingredients (e.g., trypsin) at some point in their development. In January of 2011, the European Pharmacopoeia (Ph. Eur.) chapter 5.2.3 Cell substrates for production of vaccines for human use was revised to include the following instruction: “Trypsin used for the preparation of cell cultures is examined by suitable methods and shown to be sterile and free from mycoplasmas and viruses, notably pestiviruses, <circoviruses> and parvoviruses.” The addition of circoviruses to the list of viruses of concern (previously, mainly bovine viral diarrhea virus and porcine parvovirus) in Ph. Eur. 7.2 was not unexpected, based on the rotavirus vaccine experience.

A more broad expectation going forward may also be that vaccine and biologics production cell banks be proactively screened for unexpected, perhaps previously undetectable, viruses using detection techniques such as the deep sequencing used initially to detect the PCV in the GSK rotavirus vaccine. A related technique referred to as massively parallel sequencing (Massively Parallel Sequencing (MP-Seq), a New Tool For Adventitious Agent Detection and Virus Discovery) has been adopted for detection of viral contaminants in cells and viral seed stocks and for evaluating vaccine cell substrates by the contract testing organization BioReliance.

The more important sequella of the porcine circovirus disclosures may therefore be the proactive use of these new and powerful virus detection techniques for ensuring the viral safety of production cell banks, going forward.

Ridding serum of viruses with gamma irradiation: part 2

by Dr. Ray Nims

In a previous posting, we described the susceptibility of viruses from various families to inactivation in frozen serum treated with gamma irradiation (data from the literature). Gamma irradiation is a commonly employed risk mitigation strategy for biopharmaceutical manufacture, and indeed the European Agency for the Evaluation of Medicinal Products in its Note for guidance on the use of bovine serum states that some form of inactivation (such as gamma irradiation) is expected. The use of non-treated serum in the production of biologics for human use must be justified, due to the potential for introducing a viral contaminant through use of this animal-derived material.

But how effective is this particular risk mitigation strategy? To answer this, we expressed the susceptibility to inactivation by gamma radiation for a series of 16 viruses in terms of log10 reduction in titer per kiloGray (kGy) of gamma radiation. With this value in hand, one may easily calculate the log10 reduction in titer of a given virus which might be expected following irradiation of frozen serum at any given kGy dose (serum is typically irradiated at a dose of 25-40 kGy).

The next step is to try to understand the results obtain with this relatively limited series of viruses, so that we may extrapolate the results to other viruses. If we take a look at the viral characteristics that might confer susceptibility to inactivation by gamma irradiation, will we find what we are looking for?

Viral characteristics that have been postulated to contribute to susceptibility or resistance to inactivation by gamma irradiation include: 1) radiation target size (genome size, particle size, genome shape, segmentation of the genome); 2) strandedness (in double stranded genomes the genomic information is recorded in duplicate, so loss of information on one strand may not be as damaging as it would be in the case of a single stranded genome); 3) presence or absence of a lipid envelope (non-enveloped viruses are resistant to a variety of chemical and physical inactivation approaches); and 4) genome type (RNA vs. DNA). The characteristics for our series of 16 viruses are displayed in the following table (click on table to enlarge).

For evaluating the contribution of radiation target size, we are able to make use of quantitative values available for each virus for genome size (in nucleotides) and particle size (in nm). Plotting genome size vs. log10 reduction in titer per kGy yields the results shown below. The coefficient of determination obtained is just 0.32, suggesting that factors other than (or in addition to!) genome size in nucleotides must be important..

A somewhat better concordance is obtained between particle size and log10 reduction in titer per kGy as shown below. The fit line in this case is non-linear, with a coefficient of determination of 0.60.

The contributions of genome type (RNA vs. DNA), genome strandedness, and lipid envelope (present or absent) to susceptibility to inactivation by gamma irradiation appear to be minimal, within this limited series of viruses. As a result, we are left with the conclusion that the most clear, albeit incomplete, determinant of susceptibility to inactivation by gamma irradiation appears to be particle size. This probably explains the resistance to inactivation displayed by the extremely small viruses such as circoviruses, parvoviruses, and caliciviruses. It is less clear why the polyomaviruses (e.g., SV-40 at 40-50 nm) are so resistant to gamma irradiation while certain of the picornaviruses (25-30 nm) are less resistant to inactivation. More work in this area is needed to better understand all of the factors that contribute to susceptibility to gamma radiation inactivation in viruses and bacteriophage.

< This information was excerpted in part from Nims, et al. Biologicals (2011) >