Bovine Viruses:Integrating a Quality by Design (QbD)

Approach into the Quality Control

of Bovine Raw Materials

David Onions PhD MRCVS FMedSci DVMS(hon) FRSE

Biologics Safety Testing

O-0800812

www.bioreliance.com

About BioRelianceBioReliance Corporation is a leading provider of cost-effective contract services to the pharmaceutical and biopharmaceutical industries, offering more than 1,000 tests or services related to biologics safety testing, specialized toxicology and animal health diagnostics. Founded in 1947, BioReliance is headquartered in Rockville, Maryland, with laboratory operations in Rockville and Scotland and offices in Tokyo, Japan, and Mumbai, India. The Company employs more than 650 people globally. For more information, visit www.bioreliance.com.

Key Services:• Custom Assay Development to fulfill your exact requirements• Biosafety Testing of biologicals for viruses, bacteria, mycoplasma, fungi• Cell Line Characterization including identity testing, genetic stability, EM, sequencing • Final Product Testing including biopotency testing, residual DNA, host cell proteins, cross-reactivity• Virus/TSE Validation Studies for all biological products• Contract GMP Production and Testing of viral vectors and cell banks • Veterinary Vaccine Services including characterization/identity, extraneous agent testing• Regulatory and Consulting Services

All work undertaken by BioReliance is in compliance with appropriate GLP or GMP standards.

Further information is available by visiting our website at www.bioreliance.com

Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials 1

www.bioreliance.com

1.1 Risk categoriesIn evaluating the viral risk of bovine products several factors have

to be evaluated. These include the risks associated with individual

viruses, the probability of the viruses being present in the mate-

rial and the procedures used to inactivate or clear contaminating

viruses.

An important aspect of the risk assessment is that viruses weak-

ly pathogenic in their host may be highly pathogenic in heter-

ologous hosts. Ovine Herpes virus 2 is non-pathogenic in sheep,

but causes high mortality when transmitted to cattle. Similarly

Herpes-B, a herpesvirus of monkeys, causes minimal disease in its

host species, but is a fatal infection in humans. Moreover, paren-

teral introduction of viruses in a therapeutic may overcome some

of the barriers normally limiting cross species transmission. Sev-

eral categories of viruses are of concern:

• Zoonotic viruses known to be transmissible from animals to

humans. Amongst those viruses considered as zoonotic in cat-

tle are: Rabies virus, Reovirus 3, Parainfluenza virus 3, rotaviruses,

noroviruses, Cache Valley virus, West Nile virus, and tick borne en-

cephalitis viruses.

• Animal viruses that replicate in human cells in vitro. Some

animal viruses replicate in human cells in vitro but evidence

of their zoonotic potential is weak or absent. For instance, the

human population is seronegative for infection by Bluetongue

virus, but the virus replicates efficiently in human cells and has

been proposed as an oncolytic vector.

• Viruses that infect human cells, fail to undergo productive

replication but may initiate disease. Experimental transmis-

sion of certain members of the herpesvirus, adenovirus and

polyomavirus families can result in tumour formation outside

their natural host. These events are associated with abortive

replication and often with integration of part of the viral ge-

nome into chromosomal DNA.

• Virus families that have shown a propensity to change

host range. For instance, a single mutation in Feline parvovirus

1 may have been responsible for its initial transmission to the

dog population, an event that is believed to have resulted in

the death of 25% of the young puppies in the first phase of the

outbreak. One of the most disturbing findings in recent years

has been the discovery of new parvoviruses in bovine serum,

whose properties and pathogenicity are only just beginning

to be evaluated. Retroviruses are also included in this group

because of the potential severity of the diseases associated

with this family and their recent history of cross species trans-

mission.

1.2 Probability of contaminationSome zoonotic viruses are geographically restricted by vectors or

other factors and pose a negligible risk to bovine materials har-

vested from the US, Europe, Australia and New Zealand. Never-

theless, each of these regions has their own viruses of concern,

such as Ross River virus, a zoonotic alphavirus found in Australia.

Within these regions, the probability of an endemic virus being in

a particular material, like foetal bovine serum (FBS), is dependent

on a number of factors; the frequency of infection, the ability of a

virus to cross the placenta and establish a viraemia in the foetus,

the length of the viraemic period and the size of the serum pool.

Other products like collagen, thyroid hormones and trypsin re-

quire their own unique, but similar assessments.

1.2.1 Sporadic virus infectionsSome virus infections, particularly those transmitted by arthro-

pods, are only sporadic. This is well illustrated by Cache Valley virus

contamination of US origin FBS. BioReliance have recorded four

major episodes of fermenter contamination by this virus (Onions

2004; Nims et al. 2008). While contaminations by this virus are

uncommon they are very serious. Cache Valley virus is a zoonotic

virus associated with fatal encephalitis. It grows to high titre in

CHO fermenters and, therefore, poses a major biohazard and a

formidable decontamination problem.

Contamination by this virus reveals the limitations of standard

serum testing conducted by some suppliers of serum. Typically

only 50ml to 100ml may be tested in an infectivity assay out of

a commercial lot of 1000 litres. A pool of that size may contain

samples from a thousand individual foetuses. Given that at peak

viraemia there may be only 103 to 104 iu/ml, and for most of the

short viraemic period the titre will be much lower, the final titre in

the pool may be below the detection limit in a 50ml sample. This

was confirmed by detailed analysis of one fermenter crash that

occurred several days after initiation. No change in cell viability

2 Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials

www.bioreliance.com

was noted until a few hours before the crash when all the cells

died. The replication rate of the virus was determined and by

back calculation it was shown that only 10 to 100 virions entered

the fermenter in 20 litres of serum (Onions et al 2004). Applying

this very low level of virus input into our model predicted a crash

within a few hours of the actual crash (Figure 1)

The recent contamination of Genzyme’s manufacturing pro-

cesses by Vesivirus 2117 is another example of a rare sporadic

contaminant that caused significant economic loss. Only two

examples of this calicivirus contaminating fermenters have been

recorded (Kurth et al 2006). Vesiviruses are of concern because of

their ability to replicate in the cells of a wide range of species and

their known ability to jump species. One notable example was

the development of a new disease in pigs, vesicular exanthema,

following the feeding of contaminated sealion meat to pigs. The

sealion virus spread from the US to Europe and also infected the

human population, before expensive and rigorous procedures

eradicated the disease.

The vesivirus that contaminated the commercial processes almost

certainly entered in bovine serum. Antibody to vesiviruses in

cattle is variable with some herds being seronegative and others

having high seropositivity; it is possible that cattle are sporadically

infected from some other reservoir (Kurth et al. 2006).

0

5E + 11

1E + 12

1.5E + 12

2E + 12

2.5E + 12

3E + 12

3.5E + 12

4E + 12

4.5E + 12

1 8 15 22 29 36 43 50 57 64 71 78 85 92 99 106 113 120 127 134 141 148 155 162 169 176 183 190 197

Total Cell Number in Fermenter

Tota

l Num

ber o

f Cel

ls

Time (hours)

Normal Cell GrowthCell number predicted by model of infection

Fermenter Crashed here: our model indicated only 10 to 100 viruses entered the fermenter in 20 litres of serum.

Figure 1 Modelled cell population dynamics in a fermenter following introduction of 1 to 100 infectious units of Cache Valley virus.

Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials 3

www.bioreliance.com

1.2.2 Common virus infectionsBovine viral diarrhoea virus (BVDV) and HoBi virus

In contrast to Cache Valley virus and Vesivirus 2117, a very high

proportion of FBS lots may contain Bovine viral diarrhoea virus

(BVDV). The reasons for the high prevalence are related to the

biology of the virus. Transplacental transmission of BVDV can lead

to foetuses and calves that are immunotolerant and persistently

viraemic. Estimates of foetal infection in unvaccinated herds

vary from 1 in 100 to 1 in 1000 so that every serum pool is likely

to contain one or more contaminated samples. The viruses

present in foetuses are non-cytopathic and can go unnoticed

as contaminants in cell cultures unless PCR or immuno-staining

are employed to detect the virus. FBS lots are screened by serum

producers using an infectivity assay that detects non-cytopathic

virus but the inherent statistical uncertainty in sampling a large

pool means that positives are missed. A solution is to institute

sub-pool testing and rejection of positive sub-pools but this adds

to material costs. For donor calf-serum, it has been possible to

develop closed herds that are free of BVDV but again this is an

expensive procedure. Given these uncertainties it is prudent for

manufacturers to re-test the serum before use.

BVDV infects a wide range of cell lines including bovine, ovine,

canine, feline, equine, lapine, simian and insect cells (Potts et al.

1989; Bolin et al. 1994) and has contaminated veterinary vaccines

with devastating results (Barkema et al. 2001). Caution is required

in extrapolating species susceptibility from a few cell lines as

in some cases resistant variants can be selected (Dezengrini et

al 2006). Human cells have been regarded as resistant to BVDV

(Potts et al 1989) but recently there has been a report of human

cell infection (Uryvaev et al 2012). It is generally assumed that

BVDV does not infect widely used production lines like CHO and

NS0, however, recorded host ranges should never be taken as

absolute, as even minor mutagenic change can profoundly alter

host range and pathogenicity.

BVDV-1 & 2 are members of the genus Pestivirus within the

family Flaviviridae, which also contains classical Swine fever virus,

and Border disease virus. Recently a new virus HoBi, related to but

distinct from, BVDV-1 & 2 has been isolated from contaminated

cells. The source of the contamination was serum that had been

screened for BVDV. HoBi virus, which may be renamed BVDV-3

(Liu et al 2009), should be detected in serological assays for BVDV

using polyclonal antisera but several monoclonal antibodies used

in BVDV diagnosis do not cross react with HoBi. Similarly, standard

PCR tests for BVDV cells miss this virus and a specific PCR should

be employed if cells banks are being screened (Schirrmeier et al.

2004; Ståhl et al 2010).

In 2007 a new pestivirus, Bungowannah virus, was identified in

Australian pigs with myocarditis (Kirkland et al. 2007). This virus

appears to be an emerging infection and is currently restricted to

Australia. It is known that pestiviruses can cross species barriers,

for instance BVDV can infect pigs. As Australia is a major supplier

of FBS, it is important that a watching brief is maintained for the

presence of this virus in bovine serum.

BVDV enters the serum pool from immunotolerant, persistently

infected foetuses. However, if a foetus is infected late in the

third trimester it can mount an immune response and develop

antibody to BVDV. This antibody can neutralise BVDV present in

the pool and may mask infectious virus. Consequently in Europe,

tests for “inhibiting antibody levels” need to be conducted before

determining if infectious BVDV virions present. High levels of

inhibiting antibody can result in rejection of the pool. There are

some differences between the CVMP and CPMP requirements but

these are encompassed in the tests described below (see Table

1). These mandated tests are based on the incorrect assumption

that BVDV is monotypic for neutralisation whereas, in practice,

the use of different BVDV strains can profoundly alter the result

4 Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials

www.bioreliance.com

agent. Enormous amounts of data are developed in this process,

usually between 100 to 400 Mbp. Consequently, sophisticated

bioinformatic algorithms are required to analyse and verify virus

targets (the combination of MPS and interpretative algorithms

being termed MP-Seq™).

The basis of the method is to optimise the isolation of putative

viral sequences from serum or tissue while decreasing cellular

genomic nucleic acids, to vastly reduce the complexity of the

system. Typically this involves nuclease treatment to remove

cellular sequences followed by nuclease inactivation and

capsid dissociation in the presence of chaotropic agents and

finally, recovery of undamaged, encapsidated nucleic acid. The

remaining DNA and RNA targets are randomly primed and

sequenced.

Allander et al (2001) first applied this approach to bovine

serum resulting in the surprising discovery of two new bovine

parvoviruses BPV-2 and BPV-3. Studies by BioReliance scientists

using MP-SeqTM confirmed these findings and resulted in

the finding of a new parvovirus BAAV-2, a member of the

dependovirus genus (Onions and Kolman 2010). As discussed

below, these are very frequent contaminants of serum and

parvoviruses are amongst the most resistant viruses known,

posing a challenge for inactivating procedures. Little is known of

the tropism of BPV-2 & 3, even within their host species, but this

family of viruses have shown major changes in host range. The

onset of the canine parvovirus pandemic around 1979 is believed

to have followed cross transmission of a feline virus following

a single mutation in the capsid gene. In contrast, BAAV-2 and

possibly the other bovine dependovirus BAAV-1, has a wide host

range with BAAV-2 able to infect human cells.

New parvoviruses were not the only surprising discoveries. In a

survey of four different FBS serum lots from major manufacturers,

2 out of 4 batches had complete sequences of bovine noroviruses

and 2 also had sequences of kobuviruses (Onions 2011). In both

cases it was possible to reconstruct the complete genomes of

these viruses and, as the samples had been nuclease treated,

these genomes were contained within capsids and therefore

potentially infectious (see Figure 2).

(BioReliance unpublished; Patel et al. 2005; Kalaycioglu et al. 2012).

This area needs discussion and revision by the European regulatory

groups particularly in the light of the discovery of HoBi virus.

Bovine polyomavirus (BPyV)Bovine polyomavirus is another virus that is extremely common in

virus pools when assessed by PCR (Schuurman et al 1991) and, in

this case, there are significant concerns for human therapeutics.

The virus had originally been discovered as a productive infection

in primate cells and was initially thought to be a monkey virus

until it was determined that it was a contaminant from serum

used in cell culture. BPyV is of particular concern because, like

other members of the polyomavirus family, it is can oncogenically

transform cells in culture (Schuurman et al 1992). There is good

serological evidence that it is a zoonotic virus; Parry and Gardener

(1986) demonstrated that seroconversion was common in people

with occupational exposure to cattle (71% in veterinarians) but

essentially absent in the general population. No detailed follow

up molecular study has been conducted to determine whether

BPyV sequences are found in cancers of at risk populations.

The high prevalence of the virus by PCR poses a particular

problem as a PCR positive result does not necessarily indicate the

presence of an infectious virus. This has led to the development

of an infectivity assay that should be conducted on unirradiated

serum where a PCR positive result is recorded (Nairn et al. 2003).

As discussed below, where the PCR signal is above a threshold it

is also advisable to conduct infectivity assays on irradiated serum

to ensure no infectious virus has survived. It should be noted

that standard infectivity assays for bovine viruses do not detect

BPyV and a specific assay involving multiple passages with PCR

endpoints is required to detect low level contamination (Nairn

et al 2003).

1.3 Recent developments in testing raw materials Many of the assumptions about the frequency of particular

viruses in serum have had to be radically revised following the

introduction of massively parallel sequencing, sometimes referred

to as deep sequencing. Massively parallel sequencing (MPS) is a

powerful new method for the identification of viruses and other

adventitious agents, without prior knowledge of the nature of the

Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials 5

www.bioreliance.com

Noroviruses are known causes of diarrhoea in humans and there

is evidence to suggest cross species transfer of bovine viruses to

the human population (Widowson et al 2005). Kobuviruses are

particularly interesting and important contaminants. This member

of the Picornaviridae was unknown to veterinary science until it

was detected as a contaminant of HeLa and Vero cells (Yamishita

et al 2003). Since then it has been recorded as an important cause

of diarrhoea in young calves. The MP-SeqTM data regarding its

presence in FBS suggests this is not a rare contaminant and it has

the potential to be as serious a contamination agent as Vesivirus

2117 or Cache Valley virus. The virus can be cytolytic in cell cultures

but until more is known about the virus, there is a case for adding

specific PCR endpoints to in vitro cultures. This is supported by

the data on the detection of vesivirus in fermenters, where the

virus only became detectable towards the end of 14 days culture

period (BioReliance personal communication).

1.4 Integrating Testing and Risk Mitigation with Quality by DesignAs the biotechnology industry matures there is increasing em-

phasis on Quality by Design (QbD) principles as formulated in

ICH guidance document Q8 (R2) Pharmaceutical Development

(2009). Encompassed within QbD is a control strategy designed

to ensure that a product of required quality will be produced con-

sistently. Elements of the control strategy focus on input materials

and the “design space” that affects control of those materials. The

difference between the traditional approach and a QbD approach

to raw materials is worth examining; FBS is used in the example

below but the principles apply to all bovine materials.

1.4.1 Required, traditional screening methods for FBS and

bovine materials

The CPMP note for guidance on the use of bovine serum, and the

European Pharmacopoeia recommend that serum producers test

serum before inactivation. A combination of specific and general

tests are used to detect: Bovine viral diarrhoea virus (BVDV), Bovine

adenovirus (BAV), Bovine parvovirus (BPV), Bovine respiratory

syncitial virus (BSRV), Reovirus type 3 (Reo 3), parainfluenza virus

type 3, (PI3V), infectious bovine rhinotracheitis virus (BHV-1), Rabies

virus (RV) and Bluetongue virus (BTV). Virus infection is determined

by a combination of cytopathic effect, haemadsorbtion and

specific immunofluorescence assays.

The US Code of Federal Regulations 9 CFR section 113.53

requires a similar approach using Vero cells and bovine

cells. Bovine turbinate cells are employed because of their

high susceptibility to BVDV. Although PI3V and BHV-1 are

not specifically mentioned they are detectable in standard

9 CFR tests for bovine viruses. The tests are prescriptive in their

requirements but it is possible to design protocols that meet

both European and US regulatory requirements. (Table 1)

Figure 2 Detection of Bovine Kobuvirus by MP-SeqTM.

6 Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials

www.bioreliance.com

These in vitro assays achieve the aim of screening for the viruses

that were considered of main concern over two decades ago.

However as MP-SeqTM technology has shown, these are not

necessarily the commonest virus infections and they specifically

miss detecting critical viruses like bovine polyomaviruses.

Vesivirus 2117 is also likely to missed in these assays as it replicates

more efficiently in CHO cells than in standard bovine indicator

cells (BioReliance unpublished). At a minimum it is now usual to

add screening for bovine polyomavirus initially by PCR, followed

by infectivity assays for positive material. Some manufacturers

have also put in place specific screening for Cache Valley virus and

Vesivirus 2117. One advantage of PCR in the latter cases is that the

presence of non-infectious virus increases the sensitivity over an

infectivity assay, in the case of Cache valley virus the PCR assay

was about 400 times more sensitive (Onions 2004).

1.4.2 A new approach to raw material quality control

An inherent part of traditional testing strategies was the belief that

it was possible, with a high degree of certainty, to select sera free of

adventitious agents and that if a material passed a 9 CFR or CPMP

test it was safe to use. The greater understanding of viruses pres-

ent in serum that has come from new technologies like MP-SeqTM,

emphasises the need for a quantitative risk based approach.

A new approach to raw material quality control involves 3 or 4

steps:

1. Understanding the universe of potential contaminants in the

raw material.

2. Developing specific, quantitative assays for those viruses,

taking account of the statistical limitations of sampling from

the raw material pool.

TitleDuration (weeks)

Assay description CVMP assay CPMP assayCombined CVMP/CPMP assay

US assay

Determination of inhibition levels of bovine serum on multiplication of BVDV

64 passages, then 2 week titration

032940 032920 032931

Determination of inhibition levels of bovine serum on detection of BVDV

2 2 week titration 032941 032921 032932

Detection of viral contaminants in bovine serum

64 passages, 8-spot slides

032942 032922 032933

Detection of BVDV in bovine serum 64 passasges, 2-chamber slides for BVDV

032943 032923 032934

Combined CPMP/9 CFR 64 passages, 2-chamber slides, plates

n/a 032930 n/a 032930

Bovine 9 CFR 42 passages, 2-chamber slides, plates

032910 n/a n/a032900 (7 viruses)032901 (9 viruses)

Table 1 BioReliance assays available to meet Regulatory requirements

Note for “Guidance on the use of bovine serum in the Manufacture of Human Biological Medicinal Products”. Committee for Proprietary Medicinal Products 2003. CPMP/BWP/1793/02.

Code of Federal Regulations Title 9 (9 CFR). Animal and Animal Products (2008). Part 113.53. Requirements for Ingredients of Animal Origin used for Production of Biologics.

Code of Federal Regulations Title 9 (9 CFR). Animal and Animal Products (2008). Part 113.46. Detection of Cytopathogenic and/or Haemadsorbing Agents.

Code of Federal Regulations Title 9 (9 CFR). Animal and Animal Products (2008). Part 113.47. Detection of Extraneous Viruses by the Fluorescent Antibody Technique.

European Pharmacopoeia, 6th Edition, supplement 6.0, 01/2008:2262 Bovine Serum

Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials 7

www.bioreliance.com

3. Relating the potential viral load in a given batch of raw

materials to inactivating procedures like gamma-irradiation.

4. Where no inactivating steps are in place for the raw material,

adding monitoring assays later in the process to ensure the

viruses are eliminated.

Understanding the range of contaminants that may be present is

best determined through the use of new technology like MP-SeqTM

that makes no assumptions about the nature of the virus (or other

biological contaminant), or its ability to replicate in a set of pre-

determined indicator cells. MP-SeqTM is not likely to become a

routine batch by batch quality control tool until sequencing costs

fall further. However, several manufactures are now embracing the

concept of reviewing the data from MP-SeqTM on several batches

of raw materials from a given supplier. This approach should be

linked to agreements that tightly specify the geographical source

of the materials so that the MP-SeqTM data are reflective of the

universe of contaminants from that supply source.

As discussed above this technology provided indications that new

viruses like BPV-2, BPV-3, BAAV-2, Bovine norovirus and Bovine

kobuvirus were frequent, and often high level, contaminants

of serum. The next stage is to develop specific assays for these

viruses. In the case of BPV-2 and 3 permissive cell systems have

not been identified and therefore specific PCR assays have been

used to determine the frequency and level of viral genomes

in serum. As shown in the figure below, these specific PCR

tests confirmed the high frequency and the very high levels of

circulating genomes present.

Finally the viral load should be linked to inactivating procedures.

In Europe it is now a requirement to use gamma-irradiated serum

in vaccine manufacture, but in a QbD approach it is important

to understand the limitations of inactivation by irradiation.

Standard irradiation involves treatment with 35Gy, but where

batch irradiation is used, outer parts of the batch may receive

higher doses impairing the quality of the serum. The kinetic

inactivation curves for gamma irradiation are essentially first

order. The dose required to produce a 1 log10 inactivation of the

virus, or D value, varies between viruses but lies in the range

of 3.9 to 5.3kGy for several major groups (Sullivan et al. 1971).

Protection in a serum environment is likely to increase protection

for viruses and, as Plavsic & Bolin (2001) demonstrated, ssDNA

viruses like circoviruses and parvoviruses are remarkably resistant

to irradiation. This has important consequences for analysing FBS

which may contain BPV-2 and BPV-3 genomes at levels above the

capacity of irradiation to inactivate. An appropriate approach is to

screen batches by quantitative PCR using only those batches with

a low level of genomes. For instance, the control might specify an

inactivation capacity 3 log10 greater than the virus load.

Where no serum inactivating steps are in place then, as part of the

QbD approach, appropriate in process tests should be conducted.

An evaluation of the capacity of a downstream purification

process to inactivate or remove the contaminants identified

in serum should also be undertaken. Implementation of this

approach would have avoided the catastrophic contamination

of rotavirus vaccines by porcine circoviruses introduced in

contaminated trypsin.

8 Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials

www.bioreliance.com

ReferencesAllander T, Emerson SU, Engle RE, Purcell RH, Bukh J. A virus discovery method incorporating DNase treatment and its application to the identification of two bovine parvovirus species. Proc Natl Acad Sci USA. 2001 Sep 25;98(20):11609-14. Epub 2001 Sep 18

Barkema HW, Bartels CJ, van Wuijckhuise L, Hesselink JW, Holzhauer M, Weber MF, Franken P, Kock PA, Bruschke CJ, Zimmer GM. [Outbreak of bovine virus diarrhea on Dutch dairy farms induced by a bovine herpesvirus 1 marker vaccine contaminated with bovine virus diarrhea virus type 2]. Tijdschr Diergeneeskd. 2001 Mar 15;126(6):158-65.

Bolin SR, Ridpath JF, Black J, Macy M, Roblin R. Survey of cell lines in the American Type Culture Collection for bovine viral diarrhea virus. J Virol Methods. 1994 Jul;48(2-3):211-21.

Dezengrini R, Weiblen R, Flores EF. Selection and characterization of canine, swine and rabbit cell lines resistant to bovine viral diarrhea virus. J Virol Methods. 2006 Oct;137(1):51-7. Epub 2006 Jul 18.

ICH Q8(R2) Pharmaceutical Development. U.S. Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research (CDER) Center for Biologics Evaluation and Research (CBER). November 2009 ICH. Revision

Kalaycioglu AT, Russell PH, Howard CR. The Characterization of the Neutralizing Bovine Viral Diarrhea Virus Monoclonal Antibodies and Antigenic Diversity of E2 Glycoprotein. Vet Med Sci. 2012 May 16. [Epub ahead of print]

Kirkland PD, Frost MJ, Finlaison DS, King KR, Ridpath JF, Gu X. Identification of a novel virus in pigs - Bungowannah virus: a possible new species of pestivirus. Virus Res. 2007 Oct;129(1):26-34. Epub 2007 Jun 11

Kurth A, Evermann JF, Skilling DE, Matson DO, Smith AW. Prevalence of vesivirus in a laboratory-based set of serum samples obtained from dairy and beef cattle. Am J Vet Res. 2006 Jan;67(1):114-9

Liu L, Xia H, Wahlberg N, Belák S, Baule C. Phylogeny, classification and evolutionary insights into pestiviruses. Virology. 2009 Mar 15;385(2):351-7. Epub 2009 Jan 24

Nairn C, Lovatt A, Galbraith DNA Detection of infectious bovine polyomavirus. Biologicals. 2003 Dec;31(4):303-6

Nims, Raymond W.; Dusing, Sandra K.; Wang-Ting Hsieh; Lovatt, Archie; Reid, C. Gordon; Onions, David; Milne, Euan W. Detection of Cache Valley Virus in Biologics Manufactured in CHO Cells. BioPharm International; Oct 2008, Vol. 21 Issue 10, p89

Oehmig A, Büttner M, Weiland F, Werz W, Bergemann K, Pfaff E. Identification of a calicivirus isolate of unknown origin. J Gen Virol. 2003 Oct;84(Pt 10):2837-45.

Onions D. Animal virus contaminants of biotechnology products. Dev Biol (Basel). 2004;118:155-63.

Onions D, Kolman J. Massively parallel sequencing, a new method for detecting adventitious agents. Biologicals. 2010 May;38(3):377-80. Epub 2010 Mar 24.

Onions D. Overview of emerging technologies to detect adventitious agents. PDA J Pharm Sci Technol. 2011 Nov 1;65(6):654-9.

Parry JV, Gardner SD. Human exposure to bovine polyomavirus: a zoonosis? Arch Virol. 1986;87(3-4):287-96.

Patel JR, Didlick S, Quinton J. Variation in immunogenicity of ruminant pestiviruses as determined by the neutralisation assay. Vet J. 2005 May;169(3):468-72.

Plavsic and Bolin. Resistence of porcine circovirus to gamma irradiation. Biopharm Int. 14:32-36, (2001).

Schuurman R, van Strien A, van Steenis B, van der Noordaa J, Sol C. Bovine polyomavirus, a cell-transforming virus with tumorigenic potential. J Gen Virol. 1992 Nov;73 ( Pt 11):2871-8

Potts BJ, Sawyer M, Shekarchi IC, Wismer T, Huddleston D. Peroxidase-labeled primary antibody method for detection of pestivirus contamination in cell cultures. J Virol Methods. 1989 Oct;26(1):119-24.

Schirrmeier, H., Strebelow, G., Depner, K., Hoffmann, B., Beer, M., 2004. Genetic and antigenic characterization of an atypical pestivirus isolate, a putative member of a novel pestivirus species. J. Gen. Virol. 85, 3647–3652.

Schuurman R, van Steenis B, van Strien A, van der Noordaa J, Sol C. Frequent detection of bovine polyomavirus in commercial batches of calf serum by using the polymerase chain reaction. J Gen Virol. 1991 Nov;72 ( Pt 11):2739-45.

Ståhl K, Beer M, Schirrmeier H, Hoffmann B, Belák S, Alenius S. Atypical ‘HoBi’-like pestiviruses - recent findings and implications thereof. Vet Microbiol. 2010 Apr 21;142(1-2):90-3. Epub 2009 Sep 30.

Sullivan R, Fassolitis AC, Larkin EP, Read RB Jr, Peeler JT Inactivation of thirty viruses by gamma radiation. Appl Microbiol. 1971 Jul;22(1):61-5

Uryvaev LV, Dedova AV, Dedova LV, Ionova KS, Parasjuk NA, Selivanova TK, Bunkova NI, Gushina EA, Grebennikova TV, Podchernjaeva RJ. Contamination of Cell Cultures with Bovine Viral Diarrhea Virus (BVDV). Bull Exp Biol Med. 2012 May;153(1):77-81. English, Russian.

Widdowson MA, Rockx B, Schepp R, van der Poel WH, Vinje J, van Duynhoven YT, Koopmans MP. Detection of serum antibodies to bovine norovirus in veterinarians and the general population in the Netherlands. J Med Virol. 2005 May;76(1):119-28.

Yamashita T, Ito M, Kabashima Y, Tsuzuki H, Fujiura A, Sakae K. Isolation and characterization of a new species of kobuvirus associated with cattle. J Gen Virol. 2003 Nov; 84(Pt 11):3069-77

Bovine Viruses: Integrating a Quality by Design (QbD) Approach into the Quality Control of Bovine Raw Materials 9

www.bioreliance.com

www.bioreliance.comNorth America Toll Free: 800 553 5372 Tel: 301 738 1000

Europe & International Tel: +44 (0)141 946 9999 • Japan Tel: +03 5796 7430Email: info@bioreliance.com

©2012 Sigma-Aldrich Co. LLC. All rights reserved. BioReliance and SAFC are trademarks of Sigma-Aldrich Co. LLC or its Affiliates, registered in the US and other countries. O-0800812

BioReliance Corp.14920 Broschart RoadRockville, Maryland 20850Tel: 800.553.5372 Fax: 301.610.2590Email: info@bioreliance.com

BioReliance Ltd.Todd CampusWest of Scotland Science ParkGlasgow, Scotland G20 0XATel: 44 (0) 141 946.9999 Fax: 44 (0) 141 946.0000Email: info@bioreliance.com

BioReliance Ltd.Innovation ParkHillfoots RoadStirling, Scotland FK9 4NFTel: 44 (0) 141 946.9999 Fax: 44 (0) 141 946.0000Email: info@bioreliance.com

BioReliance, K.K.c/o Sigma-Aldrich Japan K.K.Tennoz Central Tower 4F2-2-24 Higashi-ShinagawaShinagawa-kuTokyo 140-0002, JapanTel: +81 (0)3 5796 7430Fax: +81 (0)3 5796 7435Email: jpinfo@bioreliance.com

BioReliance Ltd. c/o Sigma-Aldrich Chemicals Pvt Ltd102 Alpha BuildingHiranandani GardensPowai, Mumbai 400076Tel: +91 22 40872364Fax: +91 22 25797589Email: indiainfo@bioreliance.com