Getting a grip on prophage

by Dr. Ray Nims

In a previous post, we discussed bacteriophage as a risk for the manufacture of biopharmaceuticals by bacterial fermentation. We mentioned briefly that bacteriophage may integrate within the genome of bacterial cells and that this may also represent a problem. Now we will explain why.

Bacteriophage are viruses that infect bacteria, and they have evolved two mutually exclusive strategies for survival. One involves a lytic growth cycle leading to death (lysis) of the host cell and release of progeny phage that may then infect additional host cells (so-called horizontal transmission). The other strategy is called lysogeny and involves integration of phage coding sequences into the host (bacterial) cell genome. The integrated phage is termed a prophage. This strategy for phage survival is referred to as vertical transmission since the phage genomic material is reproduced along with that of the host cell as the latter proliferates. Under certain circumstances, however, the integrated prophage can excise itself from the host cell chromosome in a process referred to as induction. The excised phage then may initiate a lytic infection of the host cell, causing all of the problems discussed in the previous post.

Illustration of a T4 phage infecting E. coli by Jonathan Heras

The relative success (i.e., from the perspective of the phage!) of the lytic vs. lysogenic survival strategies changes with the probability of host cell survival. Lysogeny appears to be a strategy that allows phage to persist during periods of low host cell availability or poor environmental (e.g., nutrient) conditions. Induction of prophage is an adaptation of the phage to host cell damage. This damage usually takes the form of a major stress to the host cell.

If stess can lead to prophage induction, the worry then becomes that some manipulation of a bacterial production cell during biopharmaceutical manufacture could lead to induction and initiation of a lytic phage infection. How can we assess and mitigate the potential for this to occur? There are two approaches: first, we can perform chemical or physical induction studies to determine the likelihood of encountering a prophage in a given production cell; and second, we can engineer the conditions of bacterial growth such that induction of a prophage is discouraged.

Phage induction studies may be performed on the bacterial production cell following initial engineering of the cell or during characterization of the cell bank. The inducing agent most often employed is mitomycin C. Other types of inducing agents (conditions) include carcinogens (such as the N-nitrosamines), hydrogen peroxide, high temperature, starvation, and UV radiation. The cells are treated with the inducing agent or condition, then one of various endpoints is used to detect the initiation of a lytic phage infection. These could include culture assays as well as molecular techniques such as PCR, microarray, or DNA chips.

Suppose you have an E. coli production cell harboring a problematic prophage. What can be done to discourage phage induction? Certain growth procedures have been shown to reduce spontaneous phage induction in E. coli cultures. These include using lower bacterial growth rates, replacement of glucose in growth medium with glycerol, and engineering the production cell through introduction of a plasmid conferring over-expression of the phage cI gene.

In summary, there are approaches that can identify the likelihood of encountering prophage induction from a bacterial production cell. The time to perform this type of testing is during development of the fermentation process (following the engineering of the production cell), or following banking of the production cell. If prophage induction appears to be a problem, bacterial growth procedures can help to reduce the potential. If this is not sufficient, the production cell may need to be re-engineered to produce a phage-resistant mutant.

Cold Facts About Dissolved Oxygen

By Dr. Scott Rudge

What’s the solubility of oxygen in water? Everyone knows that the answer to this question is “low”, and that’s enough to know for many practical applications. But it’s high enough to rust unprotected metal surfaces, and high enough to grow cells, provided that it’s replenished at some rate. It’s easy to find a number on the internet, at a temperature and pressure that the author of the internet resource thinks is interesting. But my interesting condition always varies from the internet’s, and finding the constants that I need to calculate the actual value is always difficult.


Most references that you can find say that the solubility of oxygen in water follows Henry’s Law. Henry’s Law is a very simple expression that says the concentration of a substance in a liquid phase is related to the partial pressure of that substance in the gas phase by a “constant”. Constant is a relative term in thermodynamics, because the constant in this and most cases varies with temperature and other components in both phases. But we’ll go with it.

In this equation, k_H is Henry’s constant, p is the partial pressure of the substance in the gas phase and c is the corresponding concentration of that substance in the liquid phase. Partial pressure is simply the amount of the total pressure due to that substance. In a room full of air at sea level, the partial pressure of oxygen is approximately 21% of the atmospheric pressure, or 0.21 atmospheres (159.6 torr or mm Hg, or 3.1 psi). At 25°C, Henry’s constant for oxygen in pure water is 769.2 L*atm/mol.

To correct for temperature, an approximation using a reference temperature can be used, although the results will not be exact. The correlation is:

Where T_r is the reference temperature. Again, the use of these equations will give approximate values.

Adding salt to the water further decreases the solubility of oxygen, such that in sea water the solubility of oxygen is about 80% of that in pure water. In fermentation medium, the saturation concentration is probably even less.

For those who don’t like math, here’s the table for the dissolved oxygen concentration under air atmospheres containing 21% oxygen (dissolved oxygen concentrations in moles/L):

Lesson learned: Outsource but remain in control!

By Dr. Ray Nims

In a previous posting, we described the responsibilities of the contract giver (contractee) and the contract acceptor (contractor) in outsourced pharmaceutical quality control testing. Our blog title: "Outsource it, and fuggedaboutit?" somewhat facetiously suggested that the outsourcing of quality control testing does not transfer quality control responsibility from the contract giver to the contract acceptor.

Elizabeth Meyers and I expanded upon this theme in a recent article in BioProcess International. Our conclusion in that article was more direct: “The use, by a pharma organization, of a contract testing lab to fulfill some or all of its Quality Control testing obligations does not absolve the contractee of its overall Quality responsibility of ensuring the safety, purity, identity, efficacy, and potency of its products.”

This point was illustrated nicely in a recent warning letter published on the FDA Website. The name of the firm involved is not important to this discussion. Among the other findings was the following:

“Your firm failed to properly evaluate a contract laboratory to ensure GMP compliance of operations occurring at the contract site.”


The FDA then provided the following detail: "...we are concerned about your firm’s fundamental understanding of what is required by your Quality Unit and the regulatory expectations for a firm that enters into agreements with contract testing laboratories. Although you have agreements with other firms that may delineate specific responsibilities to each party, you are ultimately responsible for the quality of your products and the reliability of test results. Regardless of who tests your products or the agreements in place, you are required to manufacture these products in accordance with section 501(a)(2)(B) of the Act to assure their identity, strength, quality, purity, and safety."

The take-home message from this is that in the outsourcing of quality control testing, responsibility for the outsourced testing is retained by the contract giver. Responsibilities of the contract giver include the following: selecting and qualifying the contract lab, ensuring the suitability of methods used (through qualification, transfer, verification, or validation); putting in place a Quality and business agreement; scheduling and submitting samples (including communicating expectations for sample results); providing in-life guidance; and monitoring of contract lab performance.

There is no denying that fulfilling these responsibilities requires a significant and ongoing effort on the part of the contract giver. In this respect, outsourcing of quality control testing is not so different from doing that testing in-house.

What's up with USP Chapter 1050?

by Dr. Ray Nims

United States Pharmacopeia (USP) general chapter <1050> Viral Safety Evaluation of Biotechnology Products Derived from Cell Lines of Human or Animal Origin originally appeared in supplement 10 to USP23-NF18 in May 1999 and was at that time essentially a verbatim adoption of the International Conference of Harmonization (ICH) Guideline Q5A (R1) having the same title.

The chapter describes the methods of evaluating the viral safety of biotechnology pharmaceutical products that are manufactured using cell lines of human or animal origin.

In 2006, an ad hoc advisory panel was assembled by the USP and tasked with revision of this chapter. The goals were to update the chapter and, more specifically, to add greater detail in the viral clearance validation section. The hope was that a user following the recommendations set forth in the general chapter would have greater confidence that viral clearance validation data generated would prove acceptable to the regulatory agencies.

The organization of the revised chapter <1050> was not changed. It comprised the following main sections: 1) Introduction; 2)Potential sources of viral contamination; 3) Cell line qualification: testing for viruses; 4) Testing for viruses in unprocessed bulk; 5) Rationale and action plan for viral clearance studies and viruses tests on purified bulk; and 6) Evaluation and characterization of viral clearance procedures. The changes proposed for the initial 5 sections were minor and primarily reflected attempts to update the chapter and to align the chapter more closely with FDA guidance documents. The most extensive changes were to section 6, in keeping with the goals described above.

The revised chapter was published for public comment in Pharmacopeial Forum 36(3) in the fall of 2010. Comments received as a result of the public review apparently suggested that a more extensive update of the chapter was warranted. At any rate, the revised chapter was not made effective during the USP’s 2005-2010 revision cycle. A new ad hoc advisory panel now being assembled as part of USP's 2010-2015 revision cycle will take over the responsibility for moving the revision of this chapter forward.

Remember....bacteriophage are viruses too

by Dr. Ray Nims

Are you using bacterial cells to produce a biologic? Do not make the mistake of thinking that your upstream process is safe from infection by adventitious viruses. True, you are not required to test for the usual viruses of concern using a lot release adventitious virus assay. But bacterial production systems are susceptible to introduction of viruses just as mammalian cell processes are. In this case, the viruses just happen to be referred to as bacteriophage. Other than this, the putative contaminants have the same nasty property exhibited by viruses that can contaminate mammalian cell processes, that is … their small size (24-200 nm) allows them to readily pass through the filters used to “sterilize” process solutions. So media, buffers, induction agents, vitamin mixes, trace metal mixes, etc. that are fed into the fermenter without proper treatment can introduce a bacteriophage. Especially worrisome in this regard are raw materials that are generated through bacterial fermentation (such as amino acids, antibiotics). A fermenter infected with a lytic phage exhibits a clear signal that the bacterial substrate is unhappy. The trick then is to discover where the phage originated and to mitigate the risk of experiencing it again.

How can you mitigate the risk of experiencing a bacteriophage infection? Many of the same strategies used to protect mammalian cell processes may be applicable to the bacterial fermentation world. Raw materials and/or process solutions may be subjected to gamma-irradiation, to ultraviolet light in the C range, to prolonged heating or to high temperature short time treatment, to viral filtration, etc. In addition, mitigation of risk of bacteriophage contamination may require filtration of incoming gasses using appropriate filters. 

A sampling of the data available on inactivation of bacteriophage by various methods is shown in the table below. The literature is extensive, and as with viral inactivation, the inactivation of phage by certain of the methods (e.g., UVC, gamma-irradiation) may be dependent both upon the matrix in which the phage is suspended as well as the physical properties of the phage (e.g., genome or particle size, strandness, etc.). For fairly dilute aqueous solutions, gamma-irradiation, UVC treatment, or parvovirus filtration should represent effective inactivation/removal methods. HTST at temperatures effective for parvoviruses (102°C, 10 seconds) should be effective for most bacteriophage, although this is an area that needs further exploration.

Mitigating the risk of experiencing a bacteriophage contamination of a bacterial fermentation process is possible if one remembers that bacteriophage are similar to mammalian viruses. Strategies that are effective for small-non-enveloped mammalian viruses (i.e., the worst case for mammalian viruses) should also be effective for most bacteriophage.

A possible exception to this is prophage. In analogy with the presence of endogenous retroviruses in certain mammalian cells (i.e., rodent, human, monkey), there is a possibility of encountering integrated bacteriophage (prophage) in certain bacterial cell lines. Like endogenous retroviruses, prophage may result in the production of infectious particles under certain conditions. This phenomenon deserves some discussion, but this will have to be deferred to a future blog.

References: Purtel et al., 2006; Ward, 1979; Sommer et al., 2001.