Blackwell Publishing IncMalden, USATRFTransfusion0041-11322006 American Association of Blood BanksNovember 2006461119511958Original ArticlePCV2 AND CAV RESISTANCE TO INACTIVATIONWELCH ET AL.

BLOOD COMPONENTS

Resistance of porcine circovirus and chicken anemia virus to virus

inactivation procedures used for blood products

Jon Welch, Carol Bienek, Edward Gomperts, and Peter Simmonds

I
nadvertent transmission of viral infections via
BACKGROUND: Virus inactivation procedures are used human blood or fractionated blood components has
to prevent contamination of plasma-derived blood reduced dramatically over the past 20 years and
products with viruses. Pasteurization or prolonged dry posttransfusion infection with identified viral (and
heat has proven effective against several enveloped and other) pathogens is now rare. This advance in the safety
nonenveloped viruses and provides an additional layer of of blood and blood components is largely attributable to
safety for plasma products. the introduction of microbiologic screening of donated
STUDY DESIGN AND METHODS: The resistance of blood and the development of virus inactivation proce-
porcine circovirus 2 (PCV2) and chicken anemia virus dures suitable for the treatment of plasma products. Cur-
(CAV), two small, nonenveloped viruses, to standard rently used inactivation procedures seek to remove all
(pasteurization, 10 hr at 60°C; dry heating, 80°C for 72 hr) infectious agents in plasma products, i.e., both recog-
and more extreme heat inactivation procedures nized pathogens for which there is serologic screening,
(temperatures up to 120°C) was determined. The ability such as human immunodeficiency virus, hepatitis C virus
of these procedures to inactivate PCV2 and CAV was (HCV), and hepatitis B virus, together with a wide range
measured by comparison of in vitro infectivity before and of other unscreened viruses that may be present in
after treatment. donated plasma, including West Nile virus, GB-virus C,
RESULTS: Infectivity of PCV2 and CAV was reduced by human herpesviruses, enteroviruses, and torquetenovi-
approximately 1.6 and 1.4 log by pasteurization and by rus (TTV).
0.75 and 1.25 log by dry-heat treatment, both Currently used virus inactivation steps for plasma
substantially more resistant than other viruses previously products represent a balance between effectiveness in
investigated. PCV2 and CAV were additionally almost removing viral infectivity and retaining the biologic activ-
completely resistant to dry-heat treatment up to 120°C for ity of the components. The most widely used virus inacti-
30 minutes (mean log infectivity reductions, 1.25 and 0.6), vation methods are based on incubation in solvent/
although both were more effectively inactivated when the detergents (S/Ds) or heating, either in solution (pasteur-
temperature of wet-heat treatment was increased to 80°C
(>3.2 and >3.6 log infectivity reduction).
CONCLUSION: Although neither PCV2 nor CAV are
known to infect humans, their inactivation properties may
ABBREVIATIONS: CAV = chicken anemia virus; PCV2 = porcine
represent those of other small DNA viruses known to be
circovirus 2; TTV = torquetenovirus.
present (e.g., TT virus, small anellovirus) or potentially
present in human plasma. Findings of extreme thermal From the Virus Evolution Group, Center for Infectious Diseases,
resistance demonstrate that recipients of plasma-derived University of Edinburgh, Summerhall, Edinburgh, UK; the
therapeutics may potentially still be exposed to small DNA Protein Fractionation Center, Scottish National Blood
viruses, despite the implementation of viral inactivation Transfusion Service, Edinburgh UK; and the Children’s Center for
steps. Cancer and Blood Diseases, Children’s Hospital Los Angeles, Los
Angeles, California.
Address reprint requests to: Peter Simmonds, Center for
Infectious Diseases, University of Edinburgh, Summerhall,
Edinburgh, EH9 1QH, UK; e-mail: Peter.Simmonds@ed.ac.uk.
JW was supported by a grant from Baxter Healthcare.
Received for publication February 3, 2006; revision received
April 21, 2006, and accepted April 22, 2006.
doi: 10.1111/j.1537-2995.2006.01003.x
TRANSFUSION 2006;46:1951-1958.

Volume 46, November 2006 TRANSFUSION 1951

WELCH ET AL.
ization, wet heat) or by heating lyophilized blood products
(dry heat). These methods effectively inactivate a broad
range of human pathogens and are particularly effective
against enveloped viruses and viruses with larger virions.
Nonenveloped, small viruses, however, often show some
resistance to these physicochemical treatments, with doc-
umented cases of transmission of hepatitis A virus and
parvovirus B19 by S/D-inactivated blood components.1-4
Investigation of the effectiveness of virus inactivation
procedures has been facilitated by measurement of their
effect on a range of (cell culture–derived) viruses spiked
into plasma or protein carrier and then inactivated in
model processes. For many human viruses that until
recently have lacked culture methods, such as HCV and
parvovirus B19, inactivation data have been obtained
from various “model” viruses, generally members of the
same virus family, which frequently, although not always,
possess similar physicochemical properties. The recent
development of a cell culture method for parvovirus B19
has indeed revealed an much greater susceptibility to
inactivation5,6 than found in animals parvoviruses previ-
ously used as models for B19.5-8
Although currently used inactivation procedures
show efficacy for almost all well-characterized viruses that
might be present in human plasma, the recent discovery
of a highly prevalent, and extremely heterogeneous, group
of small DNA viruses in humans and other mammals (TTV
and related viruses),9-11 as well as several pathogenic cir-
coviruses infecting other mammals, have led to concerns
about the potential for transmission and pathogenicity in
humans treated with plasma-derived blood products and
other biopharmaceuticals.
In this study, we have investigated the inactivation
characteristics of two animal circoviruses, porcine
circovirus 2 (PCV2) and chicken anemia virus (CAV). Both
viruses have small particle sizes (17 and 28 nm12-14), short,
covalently closed single-stranded genomes (1800 and
2500 bases), and are currently classified into two different
virus families, the Circoviridae and the Gyroviridae.
PCV was first detected as a natural contaminant of
cultured porcine kidney cell line PK15,12 and infection is
widely distributed in farmed pigs. A subsequently iden-
tified, related virus (PCV2) has been strongly implicated
in postweaning multisystemic wasting syndrome, which
affects young pigs and inhibits their growth.15-17 CAV vir-
ions are larger and morphologically distinct from
PCVs13,14 with a genome organization and size more
similar to that of the human virus TTV and related
viruses in other mammals.18 CAV is pathogenic in young
chicks, targeting hematopoietic precursor cells in the
marrow.19-21
PCVs and CAV represent useful models for the inacti-
vation processes used in blood product manufacture.
First, they represent structurally distinct examples of a
number of small DNA viruses that circulate in mammals
1952 TRANSFUSION Volume 46, November 2006
and potentially reflect their biophysical properties. Sec-
ond, both viruses can be cultured, allowing comparison of
their inactivation characteristics to previously investi-
gated viruses, including human and animal parvoviruses.
Third, there is evidence that both viruses are resistant to
a variety of physical and chemical inactivation proce-
dures, such as disinfectants, chloroform, and short-term
exposure to wet heat.21-23 It is therefore important to deter-
mine the extent to which such resistant viruses survive
established virus inactivation procedures.
MATERIALS AND METHODS
Cell culture
PCV2 was grown in porcine PK15 cells. CAV was grown in
the MSB1 cell line (MDCC-MSB1). PK15 cells were grown
as an adherent monolayer in cell culture with Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with
2 mmol per L L-glutamine, 100 U per mL penicillin, 100 U
per mL streptomycin, and 10 percent fetal bovine serum
(DMEM/FBS). The PK15 cell line was maintained in cul-
ture by regular passage, performing a 1:8 split every 3 to
4 days or when cells were confluent. MSB1 cells were
grown in suspension in RPMI supplemented with
2 mmol per L L-glutamine, 100 U per mL penicillin, 100 U
per mL streptomycin, and 10 percent FBS. The MSB1 cell
line was maintained in culture by regular passage, per-
forming a 1:8 split every 4 to 5 days or when cells were at
high density.
Preparation of virus stocks
Isolates of PCV2 and CAV were provided by F. McNeilly
(Department of Agriculture and Rural Development for
Northern Ireland, Stormont, Belfast, Ireland). PCV2 stock
for use in infectivity assays was prepared by infecting sub-
confluent monolayers of PK15 cells with PCV2; 4 hours
after infection, inoculum was removed and cells were
incubated for 30 minutes with a 300 mmol per L solution
of D-glucosamine in Hanks’ balanced salt solution (HBSS);
after this incubation, cells were washed twice and covered
with fresh DMEM/FBS. CAV stock for use in infectivity
assays was prepared my inoculating MSB1 cells with 0.5-
mL high-titer virus stock per 40 mL of cells. Virus was
recovered 5 days after infection by harvesting cells and
supernatant together.
Cells were lysed to release the virus by three rounds
of freeze-thawing. Cell debris from lysed cells was then
removed by centrifugation for 10 minutes at 1500 rpm
(450 × g). This virus stock was concentrated by ultracen-
trifugation at 103,000 × g in a SW28 rotor for
150 minutes, resuspending the pellet in 1/10 the original
volume. The resuspended pellet was then filtered
through a 0.8-µm pore filter to remove residual cell
debris.

Virus inactivation by heat treatments
Inactivation of PCV2 at a constant temperature was car-
ried out with virus inactivation procedures, which were
scaled down to accurately represent the manufacturing
process. Pasteurization was carried out at 60 ± 0.5°C for
24 hours in a water bath on aliquots of PCV2 resuspended
in human albumin (Alba, 20% Scottish National Blood
Transfusion Service (SNBTS) UK) at a virus:product ratio
of 1:10.4. The time course of viral inactivation was fol-
lowed by removing aliquots at 0, 2, 5, 10, and 24 hours after
the point when the spiked material reached 59.5°C. A pre-
pasteurization sample was collected before the material
was placed into the waterbath. For dry heat treatment,
PCV2 was first spiked into human Factor VIII concentrate
(FVIII; Liberate HT, SNBTS, UK) at a virus:product ratio of
1:21.8, freeze-dried in an Edwards Supermodulyo freeze-
drier, and then selected vials partially rehydrated by expo-
sure to the atmosphere. Dry-heat treatment was then
performed at 80-25 ± 0.75°C for 8, 24, 48, or 72 hours.
Inactivation of CAV at constant temperature was per-
formed by heating aliquots of virus in a dri-block. Pasteur-
ization was carried out at 60°C for 24 hours on aliquots of
CAV resuspended in heat-treated human FVIII (Liberate
HT). The time course of viral inactivation was followed by
removing aliquots at 6, 18, and 24 hours. Dry-heat treat-
ment was performed at 80°C for 72 hours on aliquots of
CAV-spiked human FVIII freeze-dried in an freeze drier
(Edwards Modulyo, Crawley, UK). The time course of viral
inactivation by dry heat was followed by removing ali-
quots at 24, 48, and 72 hours. Additional control samples
were removed from storage at the same time as samples
to be inactivated but not heated.
Virus inactivation at a range of temperatures was
carried out by heating aliquots of virus in screw-top
microcentrifuge tubes in a dri-block for 30 minutes. For
dry-heat treatments, samples were first frozen to −40°C
and then freeze-dried as before. To prevent virus inactiva-
tion caused by an abrupt change in temperature, tubes
were placed in the dri-block at room temperature before
heating to the target temperature; at the end of the 30-
minute incubation the heated block was removed and
allowed to cool to ambient temperature. The temperature
of the heated block was monitored regularly with a digital
thermometer.
Infectivity assays
Susceptible cells were infected with serial half-log dilu-
tions of heat-treated test virus preparations to determine
the endpoint of infection. This assay was performed in
sterile 24-well cell culture plates with six replicates per
dilution. The carrier proteins used (human albumin and
FVIII) were tested and had no cytotoxic effect on the cell
lines used and were found not to interfere with infection.
For PCV, each well was inoculated with PK15 cells
equivalent to 0.25 cm2 of monolayer in a total volume of
PCV2 AND CAV RESISTANCE TO INACTIVATION
500 µL. Plates were then incubated at 37°C in an atmo-
sphere of 4 percent CO2 until subconfluent (24 hr). Dilu-
tions of stock PCV2-spiked protein were prepared in
DMEM/FBS. Culture supernatant was aspirated, and cells
were infected with 500 µL of diluted virus per well of the
24-well plate. Cells were then incubated at 37°C in an
atmosphere of 4 percent CO2 for 4 hours. After this incu-
bation, 100 µL of 300 mmol per L D-glucosamine in HBSS
was pipetted onto the cells, which were then incubated at
room temperature for 30 minutes. The glucosamine was
then removed and the cells were washed with HBSS and
then PBS. Cells were covered with fresh DMEM/FBS and
incubated at 37°C in an atmosphere of 4 percent CO2 for
30 minutes. This medium was then collected as a “zero”
sample (see below). Cells were then covered with 330 µL
of fresh DMEM/FBS and incubated at 37°C in an atmo-
sphere of 4 percent CO2 for 5 days.
At the end of the assay, 40 µL of 20 percent Triton X-
100 was added to each well to disrupt cell membranes and
release virus into the culture supernatant. This super-
natant was collected in screw-cap vials and stored frozen
(−20°C) until required for DNA extraction. Alternatively,
an equal volume of TNE lysis buffer for DNA extraction
was added directly to wells and incubated for 30 minutes
before transferring the lysate to microcentrifuge tubes for
immediate use.
For CAV, serial half-log dilutions of CAV-spiked pro-
tein were prepared in RPMI. For each replicate, 500 µL of
diluted virus was pipetted into a sterile microcentrifuge
tube with 0.2 mL of confluent MSB1 cell suspension. Cells
were then incubated at 37°C in an atmosphere of
4 percent CO2 for 1 hour. After this incubation, inoculum
was washed from cells by two rounds of centrifugation at
1600 rpm (approx. 1000 × g) for 1 minute to collect cells
followed by removal of supernatant and resuspension of
the cell pellet in 1 mL of sterile PBS. After being washed,
cells were resuspended in 400 µL of RPMI and transferred
to 1 well of a sterile 24-well cell culture plate. A zero time
point sample was stored frozen until needed. Infected
cells were incubated at 37°C in an atmosphere of 4 percent
CO2 for 5 days.
As neither PCV2 nor CAV were cytopathic in in vitro
culture, polymerase chain reaction (PCR) was used to
detect virus growth in replicate wells at each test dilution.
DNA was extracted from 250 µL of supernatant from each
well in the infectivity titration as previously described.24
Extracted nucleic acids were diluted 1:25 before amplifi-
cation. Amplification of PCV2 DNA was performed by
nested PCR with one round of conventional PCR and a
second round of real-time PCR, with the following prim-
ers: outer sense, 5′-CCGCTGCCACATCGAGAAAGC; outer
antisense, 5′-AGCTTCTACAGCTGGAGCAG; inner sense,
5′-GACCTGTCTACTGCTGTGAGTACC; and inner anti-
sense, 5′-CACACAGTCTCACCCA. The second-round am-
plicon was 337 bp long and corresponds to nucleotides
Volume 46, November 2006 TRANSFUSION 1953
WELCH ET AL.
749 to 1086 in the PCV2 sequence, AF055392. DNA from
5 µL of PCV2 DNA was amplified with outer sense and
outer antisense primers. Target sequences were amplified
with the following thermal profile: 2 minutes at 94°C, 18
cycles of 30 seconds at 94°C, 21 seconds at 53°C, and
90 seconds at 72°C in a reaction buffer containing
10 mmol per L Tris-HCl (pH 9), 30 µmol per L each dNTP,
1.5 mmol per L MgCl2, 0.1 percent Triton X-100, 50 mmol
per L potassium chloride, 120 nmol per L each primer, and
0.8 U Taq polymerase in a final volume of 25 µL. Second-
round PCR procedures were performed with inner sense
and antisense primers in a real-time PCR system (Lightcy-
cler, Roche, Indianapolis, IN) with detection of amplified
product by SYBR green I (DNA master SYBR green I kit,
Roche), with a magnesium concentration of 3 mmol per
L. Secondary PCRs used 2 µL of primary product with the
following thermal profile: 30 seconds at 94°C, 23 cycles of
5 seconds at 94°C, 12 seconds at 53°C, and 70 seconds at
72°C. Amplified DNA was identified by melting curve
analysis. Standard curves were run in each assay with 10-
fold dilutions of stock PCV2 or CAV DNA to ensure assay
repeatability and sensitivity.
CAV DNA was detected by nested PCR with the fol-
lowing primers: outer sense, 5′-GCAGTGAATCGGCGCT
TAGC; outer antisense, 5′-GCTGGGAGTAGTGGTAATC
AAGC; inner sense, 5′-ATCACTTGTGGCTG; and inner anti-
sense, 5′-TCGCAGGATCGCTTCTTCGAGG. Amplified DNA
from the second-round PCR was 149 bp in length (posi-
tions 615-763 in M81223). First-round PCR was carried out
as for PCV2 for 20 cycles, with 2.5 mmol per L MgCl2. Sec-
ond-round PCR was carried out as described for PCV2.
After infection of replicate wells, productive viral
growth in each replicate was determined by real-time
PCR, based on the observation that viral DNA produced
by infected cells even at limiting dilution was orders of
magnitude greater than the signal from residual inocu-
lum. In practice, all dilutions used for infectivity measure-
ments were assayed in parallel by PCR by collection of a
zero-time-point sample, where supernatant was collected
after washing with HBSS and PBS. Zero-time-point sam-
ples were generally negative or very weakly positive in the
nested PCR, with Ct values greater than 21. In the real-time
PCR assays, replicates were considered to contain repli-
cating virus if amplified sequences showed melting peak
of the correct temperature, contained substantially more
DNA than the corresponding zero sample, and showed a
crossing-point value of cycle 21 or earlier, corresponding
to more than 500 DNA copies in the original well. The
infectious titers are expressed as log TCID50 values, calcu-
lated with the Spearman-Kärber formula, for a 0.5-log
dilution series:
log TCID50 = Xp = 1 − 0.5 (Σp − 0.5),
where Xp = 1 is the highest dilution at which all replicates
were positive by PCR and Σp is the sum of the fraction of
1954 TRANSFUSION Volume 46, November 2006
positives at this and greater dilutions. The upper and
lower 95 percent confidence limits were set at ±1 SD and
calculated from the formula
SD2 = d2 × (Σp(1 − Σp)/nr − 1),
where SD is the standard deviation, d is the log dilution
step (i.e., 0.5), and nr is the number of replicates (i.e., 6).
Infectivity reduction was calculated by subtracting the ini-
tial titer log TCID50 value from the titer at each time or
temperature point.
RESULTS
The heat resistance of PCV2 and CAV was investigated by
reduction in infectious titer after wet- and dry-heat treat-
ment for a range of incubation times and temperatures.
Inactivation by pasteurization
An infectious stock of PCV2 was prepared by infection of
large cultures of PK15 cells and purifying virus from lysed
cells. PCV2 was resuspended in 20 percent human albu-
min solution and subjected to pasteurization at 60°C for
up to 24 hours. The infectivity of each heat-treated virus
preparation was compared with that of an unheated con-
trol, and an infectivity reduction factor was calculated.
Infectivity reduced gradually over the pasteurization
period (Fig. 1), showing a final reduction of 1.33 log at
24 hours, corresponding to a final infectivity of approxi-
mately 5 percent of the spiked starting material. Infectious
titers of CAV also reduced steadily over the 24-hour pas-
teurization period (Fig. 1), with a final log reduction of
1.42 after 24 hours, corresponding to a final infectivity of
approximately 4 percent of the spiked starting material.
Inactivation by dry-heat treatment
The infectious titer of freeze-dried preparations of PCV2
subjected to dry-heat treatment at 80°C reduced slowly
over the 72-hour heating period (Fig. 2), showing a final
reduction of 0.75 log (18% of the post–freeze-dried, non-
heated FVIII sample). Partial rehydration of freeze-dried
FVIII from approximately 0.1 percent to approximately
1.5 percent (w/w) before heat treatment had little effect on
overall virus inactivation, with a log reduction of 1.0 (68%
of the infectivity of nonrehydrated samples and 10% of the
unheated control). CAV resistance to dry-heat treatment
was similar to that of PCV2. After heating freeze-dried CAV
for 72 hours at 80°C, infectious titer was only reduced by
1.25 log (Fig. 2).
Effect of higher temperatures on inactivation
Because of the observed failure of standard virus inactiva-
tion conditions to substantially inactivate PCV2 and CAV,
we investigated whether great reductions in infectivity
 PCV2 AND CAV RESISTANCE TO INACTIVATION

demonstrated much greater resistance

of PCV2 and CAV to standard pasteuriza-
tion and dry-heat methods than found
in other viruses that have been previ-
ously investigated (Fig. 1). Although the
infectivity of PCV2 and CAV remained
within 1.5 log of the starting titer even
with prolonged heating (up to 24 and
72 hr, respectively), all of the previously
investigated viruses were substantially
inactivated under similar conditions
with shorter incubation times5,7,25-29
(Fig. 1). This comparison includes sev-
eral viruses, such as canine and bovine
parvoviruses, which had been previ-
ously considered to be highly resistant
to virus inactivation. Inactivation of
PCV2 and CAV was not influenced by
partial rehydration of the virus-spiked
Fig. 1. Effect of pasteurization on the infectivity of PCV2 and CAV. PCV2 (
, graph samples before heating, despite the
series 1) and CAV (, series 2), resuspended in human albumin and FVIII, respectively, evidence that increasing the moisture
were pasteurized at 60∞C for up to 24 hours. Data for the inactivation of other viruses content substantially increases virus kill
() were obtained from the following references: parvovirus B19 (graph series 7) and of other nonenveloped viruses in previ-
canine parvovirus (CPV; series 5) treatment in human intravenous immunoglobulin, ous model inactivation studies.29,30
Yunoki et al.;7 CPV (series 3) and B19 (graph series 8) in 25 percent human serum On finding that PCV2 and CAV were
albumin, Yunoki et al.;5 human hepatitis A virus (HAV; series 4) and human poliovirus almost completely resistant to standard
1 (PV; series 6) in 1500 mg/mL sucrose, Blumel et al.26 The data used to plot data series inactivation methods, we investigated
3 through 7 were estimated from graph plots in the cited references. Data points where whether thermal stability was also evi-
no residual infectivity was detected are indicated schematically by dashed lines. Where dent at higher temperatures. Remark-
necessary to avoid superimposing data points, CAV data have been plotted offset by ably, both viruses were also almost
0.5 hour. completely resistant to dry heat to at
least 120°C (Fig. 3B). These elevated
could be obtained by increasing the temperatures used for temperatures are well beyond those that are used for viral
heat treatment. These experiments were carried out on inactivation procedures for blood products, as they would
virus preparations made in a FVIII excipient because albu- be highly damaging for the constituent active proteins.31
min coagulated at temperatures of 70°C or greater. Dry heat during manufacturing would therefore be
Pasteurization at elevated temperatures resulted in unlikely to be able to inactivate viruses with stabilities
more rapid inactivation of PCV2 and CAV (Fig. 3A). Heat- similar to PCV2 and CAV from blood products.
ing PCV2 at 65°C for 30 minutes led to no significant In contrast to dry heat, pasteurization at tempera-
reduction in infectivity (0.25 log), whereas heating to 70 tures above 70°C was more effective at removing PCV2 and
or 75°C for 30 minutes reduced infectivity by 1.59 and CAV infectivity, with log reductions of 2 and more than 3.5
1.92 log, respectively (Fig. 3A). CAV was similarly labile to at 80°C for PCV2 and CAV, respectively. As described for
increased heat with no significant infectivity reduction at dry-heat, however, available evidence indicates that these
60°C for 30 minutes (0.16 log), but 0.91 log at 65°C, 2.58 at elevated wet-heat temperatures would have a substantial
70°C, and more than 3.5 log at 75°C. In contrast, both damaging effect on the biologic activities of blood prod-
PCV2 and CAV were remarkably resistant to dry-heat treat- ucts and on their antigenicity (possibly leading to inhibi-
ment; heating to 120°C for 30 minutes (the highest tem- tor formation).32,33
perature tested) led to approximately 1-log reductions in The underlying basis for the inactivation resistance of
infectivity of either virus (Fig. 3B). PCV2 and CAV is unclear. Both viruses, particularly PCV2,
have small virion sizes and presumably a simplicity of
organization that potentially confers stability to thermal
DISCUSSION
disruption, even though they are structurally dissimilar
The study describes the use of two new model viruses to from each other.13 Other factors potentially conferring
investigate the effectiveness of inactivation processes used resistance include possession of a DNA genome and
for blood product manufacture. In this study, we have perhaps the existence of covalently closed DNA sequence

Volume 46, November 2006 TRANSFUSION 1955

WELCH ET AL.

that protects the genome from thermal,

chemical, or other forms of damage
more active against terminal nucle-
otides. The circular configuration of
PCV2 and CAV DNA may conceivably
account for their markedly greater sta-
bility compared to parvoviruses with
linear genomes.
Although not specifically investi-
gated in this study, the PCV/CAV model
system we describe would be of value in
the evaluation of other methods of virus
removal or inactivation, such as such as
nanofiltration, partitioning, or irradia-
tion.34-38 Of potential concern for nano-
filtration methods is the extremely small
size of the PCV2 virion; unless immune-
complexed, it is likely that PCV2 would
Fig. 2. Effect of dry-heat treatment on the infectivity of PCV2 and CAV. PCV2 and CAV be inefficiently removed by many cur-
(, series 3) were resuspended in human FVIII, freeze-dried, and then heated to 80∞C rently developed filters. Pore sizes of
for up to 72 hours. For PCV, in addition to freeze-dried samples (
, graph series 1), substantially less than 20 nm may be
additional samples partially rehydrated after freeze-drying to approximately 1.5 required to remove PCV, in the size
percent (w/w) were tested (series 2; ). Inactivation data for other viruses () were range that may impair the recovery of
obtained from the following sources: canine parvovirus (CPV; series 4) and bovine some the larger therapeutically active
parvovirus (BoPV; series 5) heat-treated in human FVIII data, Roberts and Hart;27 CPV proteins such as FVIII.39
(series 6), Hart et al.;30 human parvovirus B19 in fibrinogen with a residual moisture The finding of extreme inactivation
content of 0.5 to 0.7 percent (series 7) and 1.3-1.7 percent (series 8), Prikhod’ko et al.29 resistance in PCV2 and CAV described

A B

Fig. 3. Effect of elevated temperatures on PCV2 and CAV infectivity. Carrier proteins (human albumin and FVIII) were inoculated with
PCV2 and CAV and subjected to pasteurization or dry-heat inactivation at increasing temperature. (A) Wet heat: PCV2 was suspended
in human FVIII (, graph series 1) or albumin (
, series 2) and subjected to pasteurization at 60 to 80∞C for 30 minutes. CAV (,
series 3) was suspended in human FVIII and subjected to pasteurization at up to 80∞C for 30 minutes. Data points for PCV2 and CAV
staggered as in Fig. 1. (B) Dry heat: PCV2 (, series 1) and CAV (, series 3) were resuspended in human FVIII, freeze-dried, and heated
up to 120∞C for 30 minutes.

1956 TRANSFUSION Volume 46, November 2006


in this study provides the basis for further investigation of
the threat posed by small DNA viruses to blood product
safety. The circovirus models developed in this study pro-
vide a new benchmark with which to investigate the
effectiveness of more recently developed inactivation
methods that strive to eliminate all potential infectivity
from blood products and to render them biologically
sterile. Clearly, the methods studied in our investigation
fall well short of the mark. The findings in this study sug-
gest that widespread transmission with small DNA
viruses via human-derived plasma products may be an
ongoing occurrence. With the caveat that inactivation
characteristics can vary substantially between viruses
with similar physicochemical properties (as mentioned
for parvoviruses), the data obtained for PCV2 and CAV
potentially extends to other “circoviruses” such as human
TTV.
TTV is of similar virion size and is genetically related
to CAV, with a similar single-stranded, circular genome.9,10
Infection with TTV and relatives is universal in humans,
and plasma pools contain at least 105 to 106 virions mL,
representing a multitude of different species and geno-
types.40 TTV variants are currently classified into at least
39 genotypes;41 there exist, in addition at least two other
quite different species of TTV-related viruses in humans,
TT minivirus and small anellovirus, each with their own
multitude of variants.11,42,43 If TTV possesses inactivation
characteristics similar to those of PCV2 and CAV, then all
plasma-derived products would likely retain infectivity
and there would be ongoing transmission of a vast spec-
trum of TTV group viruses to recipients. Although patho-
genicity of TTV viruses in humans has not been
substantiated, the consequences of exposure to a wide
spectrum of genetically and antigenically diverse viruses
are not known.
Other DNA viruses, such as JC virus, can spontane-
ously become pathogenic through mutations in the re-
gulatory region and enter a phase of uncontrolled
replication.44 On a more global scale, the very recent emer-
gence and global spread of the highly pathogenic PCV2 in
pigs16,17 provides further evidence for the ability of DNA
viruses to spontaneously change their host relationship,
in this case with serious consequences for the pig-rearing
industry worldwide. If such transitions to pathogenicity
could occur in TTV and related viruses, even if they were
extremely rare (as is the case for JC virus), the practice of
manufacturing blood products from large plasma pools
could increase the prevalence of these putative more
pathogenic forms. Although the risk associated with the
transmission of small DNA viruses such as those investi-
gated in this study is theoretical, the high resistance of
these viruses to thermal inactivation demonstrates the
need for further research aimed at eliminating such agents
from pharmaceutical products derived from human
plasma.
PCV2 AND CAV RESISTANCE TO INACTIVATION
ACKNOWLEDGMENTS
We are grateful to Francis McNeilly from the Department of Agri-
culture and Rural Development for Northern Ireland, Stormont,
Belfast, for providing PCV2 and CAV virus isolates and susceptible
cell lines and to Tony Jones, Marilynn Roff, Judith Campbell, and
Angela Hyde of the SNBTS for technical assistance with the inac-
tivation of PCV2.
REFERENCES
1. Blumel J, Schmidt I, Effenberger W, et al. Parvovirus B19
transmission by heat-treated clotting factor concentrates.
Transfusion 2002;42:1473-81.
2. Santagostino E, Mannucci PM, Gringeri A, et al.
Transmission of parvovirus B19 by coagulation factor
concentrates exposed to 100 degrees C heat after
lyophilization. Transfusion 1997;37:517-22.
3. Robertson BH, Alter MJ, Bell BP, et al. Hepatitis A virus
sequence detected in clotting factor concentrates associated
with disease transmission. Biologicals 1998;26:95-9.
4. Johnson Z, Thornton L, Tobin A, et al. An outbreak of
hepatitis A among Irish haemophiliacs. Int J Epidemiol
1995;24:821-8.
5. Yunoki M, Tsujikawa M, Urayama T, et al. Heat sensitivity of
human parvovirus B19. Vox Sang 2003;84:164-9.
6. Blumel J, Schmidt I, Willkommen H, Lower J. Inactivation of
parvovirus B19 during pasteurization of human serum
albumin. Transfusion 2002;42:1011-8.
7. Yunoki M, Urayama T, Tsujikawa M, et al. Inactivation of
parvovirus B19 by liquid heating incorporated in the
manufacturing process of human intravenous
immunoglobulin preparations. Br J Haematol 2005;128:401-
4.
8. Boschetti N, Wyss K, Mischler A, Hostettler T, Kempf C.
Stability of minute virus of mice against temperature and
sodium hydroxide. Biologicals 2003;31:181-5.
9. Nishizawa T, Okamoto H, Konishi K, et al. A novel DNA virus
(TTV) associated with elevated transaminase levels in
posttransfusion hepatitis of unknown aetiology. Biochem
Biophys Res Commun 1997;241:92-7.
10. Okamoto H, Nishizawa T, Kato N, et al. Molecular cloning
and characterisation of a novel DNA virus (TTV) associated
with posttransfusion hepatitis of unknown aetiology.
Hepatol Res 1998;10:1-16.
11. Biagini P, de Micco P, de Lamballerie X. Identification of a
third member of the Anellovirus genus (“small anellovirus”)
in French blood donors. Arch Virol 2006;151:405-8.
12. Tischer I, Gelderblom H, Vettermann W, Koch MA. A very
small porcine virus with circular single-stranded DNA.
Nature 1982;295:64-6.
13. Crowther RA, Berriman JA, Curran WL, Allan GM, Todd D.
Comparison of the structures of three circoviruses. chicken
anemia virus, porcine circovirus type 2, and beak and
feather disease virus. J Virol 2003;77:13036-41.
Volume 46, November 2006 TRANSFUSION 1957
WELCH ET AL.
14. Todd D, Creelan JL, Mackie DP, Rixon F, McNulty MS.
Purification and biochemical characterization of chicken
anaemia agent. J Gen Virol 1990;71:819-23.
15. Allan G, Meehan B, Todd D, et al. Novel porcine circoviruses
from pigs with wasting disease syndromes [letter; comment].
Vet Rec 1998;142:467-8.
16. Allan GM, McNeilly F, Kennedy S, et al. Isolation of porcine
circovirus-like viruses from pigs with a wasting disease in
the USA and Europe. J Vet Diagn Invest 1998;10:3-10.
17. Morozov I, Sirinarumitr T, Sorden SD, et al. Detection of a
novel strain of porcine circovirus in pigs with postweaning
multisystemic wasting syndrome. J Clin Microbiol 1998;36:
2535-41.
18. Kato A, Fujino M, Nakamura T, Ishihama A, Otaki Y. Gene
organization of chicken anemia virus. Virology 1995;209:
480-8.
19. Goryo M, Suwa T, Umemura T, Itakura C, Yamashiro S.
Ultrasctructure of bone marrow in chicks inoculated with
chicken anaemia agent (MSB1-TK5803 strain). Avian Pathol
1989;18:329-43.
20. Goryo M, Suwa T, Umemura T, Itakura C, Yamashiro S.
Histopathology of chicks inoculated with chicken anaemia
agent (MSB1-TK5803 strain). Avian Pathol 1989;18:73-89.
21. Engstrom BE. Blue wing disease of chickens: isolation of
avian reovirus and chicken anaemia agent. Avian Pathol
1988;17:23-32.
22. Royer R, Nawagitgul P, Halbur PG, Paul PS. Susceptibility of
porcine circovirus type 2 to commercial and laboratory
disinfectants. J Swine Health Prod 2001;9:281-4.
23. Goryo M, Sugimura H, Matsumoto S, Umemura T, Itakura
C. Isolation of an agent inducing chicken anaemia. Avian
Pathol 1985;14:483-96.
24. MacDonald DM, Holmes EC, Lewis JC, Simmonds P.
Detection of hepatitis B virus infection in wild-born
chimpanzees (Pan troglodytes verus): phylogenetic
relationships with human and other primate genotypes.
J Virol 2000;74:4253-7.
25. Blumel J, Schmidt I, Willkommen H, Lower J. Inactivation of
parvovirus B19 during pasteurization of human serum
albumin. Transfusion 2002;42:1011-8.
26. Nissen E, Konig P, Feinstone SM, Pauli G. Inactivation of
hepatitis A and other enteroviruses during heat treatment
(pasteurization). Biologicals 1996:339-41.
27. Roberts PL, Hart H. Comparison of the inactivation of
canine and bovine parvovirus by freeze-drying and dry-heat
treatment in two high purity factor VIII concentrates.
Biologicals 2000;28:185-8.
28. Prikhod’ko GG, Vasilyeva I, Reyes H, et al. Evaluation of a
new LightCycler reverse transcription-polymerase chain
reaction infectivity assay for detection of human parvovirus
B19 in dry-heat inactivation studies. Transfusion 2005;45:
1011-9.
29. Hart HF, Hart WG, Crossley J, et al. Effect of terminal (dry)
heat treatment on non-enveloped viruses in coagulation
factor concentrates. Vox Sang 1994;67:345-50.
1958 TRANSFUSION Volume 46, November 2006
30. Savage M, Torres J, Franks L, Masecar B, Hotta J.
Determination of adequate moisture content for efficient
dry-heat viral inactivation in lyophilized factor VIII by loss
on drying and by near infrared spectroscopy. Biologicals
1998;26:119-24.
31. Smales CM, Pepper DS, James DC. Protein modification
during anti-viral heat-treatment bioprocessing of factor VIII
concentrates, factor IX concentrates, and model proteins in
the presence of sucrose. Biotechnol Bioeng 2002;77:37-48.
32. Peerlinck K, Arnout J, Di Giambattista M, et al. Factor VIII
inhibitors in previously treated haemophilia A patients with
a double virus-inactivated plasma derived factor VIII
concentrate. Thromb Haemost 1997;77:80-6.
33. Rosendaal FR, Nieuwenhuis HK, van den Berg HM, et al. A
sudden increase in factor VIII inhibitor development in
multitransfused hemophilia A patients in The Netherlands.
Dutch Hemophilia Study Group. Blood 1993;81:2180-6.
34. Omar A, Kempf C. Removal of neutralized model
parvoviruses and enteroviruses in human IgG solutions by
nanofiltration. Transfusion 2002;42:1005-10.
35. Schulman S, Wallensten R, White B, Smith OP. Efficacy of
a high purity, chemically treated and nanofiltered factor
IX concentrate for continuous infusion in haemophilia
patients undergoing surgery. Haemophilia 1999;5:96-
100.
36. Van Holten RW, Ciavarella D, Oulundsen G, Harmon F,
Riester S. Incorporation of an additional viral-clearance step
into a human immunoglobulin manufacturing process. Vox
Sang 2002;83:227-33.
37. Chin S, Williams B, Gottlieb P, et al. Virucidal short
wavelength ultraviolet light treatment of plasma and factor
VIII concentrate: protection of proteins by antioxidants.
Blood 1995;86:4331-6.
38. Li Q, Macdonald S, Bienek C, Foster PR, Macleod AJ. Design
of a UV-C irradiation process for the inactivation of viruses
in protein solutions. Biologicals 2005;33:101-10.
39. Burnouf T, Radosevich M. Nanofiltration of plasma–derived
biopharmaceutical products. Haemophilia 2003;9:24-37.
40. Simmonds P, Prescott LE, Logue C, et al. TT virus–part of the
normal human flora? J Infect Dis 1999;180:1748-50.
41. Okamoto H, Nishizawa T, Takahashi M. Torque teno virus
(TTV): molecular virology and clinical implications. In:
Mushahwar IK, editor. Viral hepatitis. London: Elsevier;
2005. p. 241-54.
42. Jones MS, Kapoor A, Lukashov VV, et al. New DNA viruses
identified in patients with acute viral infection syndrome.
J Virol 2005;79:8230-6.
43. Takahashi K, Iwasa Y, Hijikata M, Mishiro S. Identification of
a new human DNA virus (TTV-like mini virus, TLMV)
intermediately related to TT virus and chicken anemia virus.
Arch Virol 2000;145:979-93.
44. Sabath BF, Major EO. Traffic of JC virus from sites of initial
infection to the brain: the path to progressive multifocal
leukoencephalopathy. J Infect Dis 2002;186(Suppl 2):S180-
6.