close

Вход

Забыли?

вход по аккаунту

?

2311661

код для вставки
Susceptibility of Mouse Minute Virus to Inactivation by Heat in Two Cell Culture
Media Types
Marc Schleh and Peter Romanowski
Biosafety Development Laboratory, Amgen, Seattle, WA 98119
Prince Bhebe and Li Zhang
Cell Sciences and Technology, Amgen, Seattle, WA 98119
Shivanthi Chinniah, Bill Lawrence, Houman Bashiri, Asri Gaduh, Viveka Rajurs, and Brian Rasmussen
Biosafety Development Laboratory, Amgen, Seattle, WA 98119
Alice Chuck
Cell Sciences and Technology, Amgen, Seattle, WA 98119
Houman Dehghani
Biosafety Development Laboratory, Amgen, Seattle, WA 98119
DOI 10.1002/bp.181
Published online April 29, 2009 in Wiley InterScience (www.interscience.wiley.com).
Viral contaminations of biopharmaceutical manufacturing cell culture facilities are a significant threat and one for which having a risk mitigation strategy is highly desirable. High
temperature, short time (HTST) mammalian cell media treatment may potentially safeguard
manufacturing facilities from such contaminations. HTST is thought to inactivate virions by
denaturing proteins of the viral capsid, and there is evidence that HTST provides ample
virucidal efficacy against nonenveloped or naked viruses such as mouse minute virus
(MMV), a parvovirus. The aim of the studies presented herein was to further delineate the
susceptibility of MMV, known to have contaminated mammalian cell manufacturing facilities, to heat by exposing virus-spiked cell culture media to a broad range of temperatures
and for various times of exposure. The results of these studies show that HTST is capable of
inactivating MMV by three orders of magnitude or more. Thus, we believe that HTST is a
useful technology for the purposes of providing a barrier to adventitious contamination
C 2009 American
of mammalian cell culture processes in the biopharmaceutical industry. V
Institute of Chemical Engineers Biotechnol. Prog., 25: 854–860, 2009
Keywords: high temperature short time, MMV, viral contamination, media treatment,
parvovirus
Introduction
Previous contaminations of large-scale cell culture processes with MMV have been reported in the biopharmaceutical industry.1 To safeguard against future contamination
events, it is necessary to evaluate technologies that can provide a measure of safety against potential introduction of adventitious contaminants through raw materials. Theoretically,
such technologies would act as a barrier to introduction of
adventitious contaminants into the cell culture process. One
such safeguard is media treatment with high temperature,
short time (HTST) technology, which is essentially pasteurization. It involves processing cell culture media through a
thin tube at high pressure and holding the media for a specified amount of time at a specified temperature.
Correspondence concerning this article should be addressed to H.
Dehghani at [email protected]
Current address of Peter Romanowski: Rosetta Pharmaceutics LLC,
SWA-3175W, 401 Terry Ave N, Seattle, WA 98109.
854
Pasteurization or HTST has been evaluated for use in the
manufacturing of biopharmaceuticals for the purpose of inactivating adventitious viral contaminants.2 Enveloped viruses
are generally regarded as highly sensitive to infectivity inactivation by heat while nonenveloped viruses are less so. In one
study, a 105 CCID50/mL load of the enveloped Murine Leukemia Virus (MuLV) was completely inactivated (no residual
active virus was detected) in a 1 mg/1 mL protein solution after a 15 min warm-up to 65 C.3 Cell culture infection dose
(CCID50) represents the titer obtained from a 50% cell culture
infectious dose serial dilution end-point assay.
In contrast to MuLV, MMV particles are highly stable and
resistant to many commonly used infectivity inactivation
methods such as low pH, heating at 56 C for 60 min, exposure to high salt concentrations and extraction with lipid solvents.4 This resilience is primarily due to its lack of a lipid
envelope which, in contrast, enables the relatively easy
infectivity inactivation of enveloped viruses such as MuLV
using the same methods. MMV is a member of the autonomous parvoviridae family of viruses, which are classically
C 2009 American Institute of Chemical Engineers
V
Biotechnol. Prog., 2009, Vol. 25, No. 3
855
Figure 1. Cytopathic Effect of MMV on 324K Cells.
Left: MMV-induced CPE on the indicator cell line 324K. Right: Negative control 324K cells on the same day.
characterized as icosahedral, nonenveloped, single-stranded
DNA viruses with approximate sizes of 20–25 nm.5
Despite the resilient nature of MMV, there is evidence it
too is susceptible to infectivity inactivation with sufficiently
high levels of heat.3,6 Boschetti et al. showed that temperatures above 90 C can lead to 4 logs of reduction in MMV
infectivity.6 This loss of infectivity is thought to result from
disintegration of viral capsid proteins.6 Heat treatment denaturation prevents the MMV capsid proteins from interacting
with their cognate cell-surface receptors that facilitate their
entry into the cell. Boschetti et al. provided evidence for the
heat-induced destruction of viral capsids by heating samples
of MMV to 70 C or 90 C for 10 min and then titrating each
sample before and after treating each with DNA nuclease.
Treatment of MMV at 90 C for 10 min or longer resulted in
a reduction of active MMV by 4.7 logs (below the limit of
detection by a CCID50 assay), but MMV maintained its titer
within 1.0 log of the starting titer after being treated with
70 C heat for 10 min. When the samples were treated with
nuclease, the sample treated at 70 C maintained its levels of
viral DNA, whereas the sample treated at 90 C showed significantly fewer MMV genomes post-DNase treatment. The
reduction in MMV genomes in the latter sample suggests
that viral nucleic acid was externalized as a result of heatinduced capsid disintegration at 90 C.
The goal of the studies presented herein was to further
characterize the susceptibility of MMV to infectivity inactivation by HTST treatment and to show that such treatment
is a practical method of preventing contaminations in the
manufacturing of biotechnology-derived therapeutics. To our
knowledge, our study is the first to assess the heat-sensitivity
of MMV at a variety of treatment times on the order of seconds rather than minutes or hours.
Methods and Materials
Preparation of MMV stocks
MMV was prepared in-house for this study. Briefly, an
SV-40 large T-antigen transformed human newborn kidney
cell line known as 324K (source: Peter Tattersall, Yale University)7 was seeded into a 10-chamber cell factory at 0.8 104cells/cm2 and allowed to incubate overnight in a humidified 37 C/5% CO2 incubator. The 10-chamber cell factory
was rinsed once with 1 DPBS and cells were inoculated
with 0.01 MOI of MMV. When 100% CPE was observed
the cell factory was frozen and thawed three times to induce
cell lysis. The cell lysate was then harvested, clarified, and
concentrated. Virus pellets were resuspended in a virus
resuspension buffer (10 mM Tris, 100 mM NaCl, and 1 mM
EDTA 4Na2H2O at pH 7.5). The virus suspension was sonicated for 2 min, 0.22 lm filtered, and stored at 70 C.8
MMV stock titer determination
Cell-based infectivity assays, described earlier,8 were used
for MMV quantification. These assays require an indicator
cell line that exhibits morphological alterations known as
cytopathic effect (CPE), which appears after a specified incubation period post-infection. For MMV, the most commonly
used indicator cell line for such assays is 324K and CPE is
assessed on day 10 post-inoculation with a specific virus
sample. Figure 1 shows a picture of MMV-induced CPE on
324K cells when compared with a negative control inoculated with 324K seed media.
HTST experimental design
A stainless steel pasteurization apparatus was designed
and constructed in-house. During the conduct of the studies,
a peristaltic pump was used to process the spiked media
through this pasteurization system at different flow rates.
Before treatment, 10 L each of media 1 or media 2 was spiked
at 1% (v/v) with MMV in a carboy. Medias 1 and 2 had
slightly different compositions. Each media was at the standard cellular osmolarity (1) of 300 0.15 mOsm/Kg.
The spiked media was mixed with a stir bar on a stir plate
for several minutes to ensure even distribution of the virus
particles. A sample of the spiked load was taken and was
either subjected to 0.22 lm filtration or left unfiltered. An
unspiked sample of each media type was processed through
the HTST system and held at the maximum treatment temperature for the maximum time—115 C for 30 s—to test the
effects of heat on media. The spiked media was then processed continuously through the HTST apparatus and held in
residence at the specified temperatures for the specified times
(see Table 1).
Different holding tubes were used to achieve the required
residence times. Holding tube 1 was used for the 2 and 5 s
treatment times; holding tube 2 was used for the 10, 15, and
30 s treatment times. To achieve the required residence times
with each tube, the pump speed was adjusted. Table 1 shows
these and other technical specifications of the HTST process.
To conserve the spiked media, the HTST system was
856
Biotechnol. Prog., 2009, Vol. 25, No. 3
Table 1. HTST Technical Specifications and Treatment Summary
Process Temperature* ( C)
Residence Time† (s)
Heating/Cooling
Residence Time‡ (s)
Flow Rate (mL/min)
Pump Speed§ (rpm)
Holding Tube{
95
2
5
10
15
30
2
5
10
15
30
2
5
10
15
30
7.90
12.91
7.90
8.17
14.35
7.90
12.91
7.90
8.17
14.35
7.90
12.91
7.90
8.17
14.35
250
100
250
158
90
250
100
250
158
90
250
100
250
158
90
43
18
43
29
17
43
18
43
29
17
43
18
43
29
17
1
1
2
2
2
1
1
2
2
2
1
2
2
2
2
105
115
* Process temperature: one of the three specified treatment temperatures. † Residence time: time the media was held at the treatment temperature.
Heating/cooling residence time: time needed to heat the media to the treatment temperature and cool the media to 17 C. § Pump speed: speed of the
peristaltic pump used to force media through the HTST system. { Holding tube: unit in which the media was held to achieve required residence time.
‡
Scheme 1. HTST Processing Diagram.
Diagram of the HTST treatment and sample collection set-up. Thick arrow denotes stainless steel pasteurization apparatus; thin arrows denote collection tubes. All collected samples were 0.22 lm filtered. Samples collected from points A, B, and C were CCID50 assayed; only samples collected
from point C were large volume assayed. Unfiltered and filtered samples of the spiked media pretreatment were CCID50 assayed to determine filtration’s effect on virus titer. Media was cooled to 17 C in cooling tube prior to collection at point C.
allowed to equilibrate to each treatment temperature with
deionized water. Samples were collected from three collection points, A, B, and C (see Scheme 1).
Point A: Media at collection point A was raised to the
treatment temperature briefly.
Point B: Media at collection point B was held at the specified temperature for the specified time.
Point C: Media at collection point C was held at the specified temperature for the specified time and was then cooled
to 17 C before cooling.
Sample collection points A and B, which consisted of silicon tubing, were immersed in a wet ice bath. At collection
points A and B, 5-mL samples were taken and 0.22 lm filtered for CCID50 assaying; these samples were not tested
with a large volume assay. At collection point C, a 50-mL
sample was also taken and 0.22 lm filtered to accommodate
the volume required for CCID50 and large volume assaying.
The HTST system was cleaned by flushing it with 1 N
NaOH for 30 min. One limitation of the study was that the
amount of time the media spent at collection point A (before
the hold time) was not measured, though it was brief; samples collected at point A were therefore not used to determine the log reductions achieved with HTST treatment.
Titer determination of process samples
A CCID50 assay was performed on samples collected from
points A, B, and C as described earlier for the titer determination of the virus stock used to spike the media. Additionally, the unfiltered and filtered spiked load samples were
titrated as positive controls to establish log reduction values
for each treatment condition and to demonstrate that 0.22
lm filtration has no significant impact on virus titer. These
spiked load samples were frozen and thawed several times
during the course of assaying and were therefore assayed
multiple times. The average filtered load value of these
assays was used to determine the log reduction achieved
with each treatment. Negative control wells were inoculated
with the same media used to dilute the samples.
Testing of samples with large volume assays
In cases where no detectable infectivity was observed in
titrations of samples collected at point C, large volume
assays were used to determine a theoretical titer using the
Poisson distribution at the 95% confidence limit:
p ¼ ecv
If p ¼ 0.05, then: c ¼ (ln p)/v, where c is the concentration of infectious particles per milliliter, and v is the tested
sample volume (mL).
The 96-well plates used for large volume studies were
seeded as described for MMV stock titer determination. A
total of 40 mL of each sample collected from point C was
inoculated onto 800 replicate wells of ten 96-well plates at
50 lL per well. For a sample in which no residual active virus was detected in the CCID50 and large volume assays,
these 40 mL large volume assays can increase the assay sensitivity by 2 logs over the CCID50 assay. Media 2 samples
were assayed neat. Undiluted media 1 was found to be toxic
Biotechnol. Prog., 2009, Vol. 25, No. 3
857
to 324K cells, but this cytotoxicity was eliminated with a
1:10 dilution of the test media in 324K seed media; all
media 1 samples were thus diluted 1:10 for large volume
inoculation. As a result of this dilution, the large volume
assay sensitivity for media 1 was reduced to only 1 log more
than that offered by the CCID50 assay. Unspiked, treated and
unspiked, untreated samples of both media types were inoculated neat through the 102 dilution to confirm the toxicity
of media 1 to 324K cells and to show that the toxicity was
not caused by heating.
Calculation of virus load and log reduction values
The virus load of a sample was determined by multiplying
the virus titer (CCID50/mL), total sample volume (mL) and
dilution needed to eliminate cytotoxicity and interference:
Virus Load ¼ ðVirus TiterÞ ðLoad or Eluate VolumeÞ
ðDilution FactorÞ
Table 2. Spiked Load Titers of Media 1
Virus Titer
(log10CCID50/mL
95% CI)
Spiked Load
Unfiltered assay 1
Unfiltered assay 2
Unfiltered assay 3
Unfiltered assay 4
Average unfiltered load
Filtered assay 1
Filtered assay 2
Filtered assay 3
Filtered assay 4
Average filtered load
3.43
2.93
2.80
2.68
3.07
3.43
3.05
2.93
2.80
3.12
0.36
0.25
0.00
0.48
0.33
0.36
0.32
0.25
0.00
0.27
Virus Load
(log10CCID50
95% CI)
7.43
6.93
6.80
6.68
7.07
7.43
7.05
6.93
6.80
7.12
0.36
0.25
0.00
0.48
0.33
0.36
0.32
0.25
0.00
0.27
Process samples were assayed on multipe days and spiked load samples were assayed each time as positive controls. Only filtered spiked
load samples were used to determine the log reductions achieved with
HTST treatment. Spiked load assay numbers refer to titrations performed
on separate days.
This numerical value, expressed as Log10CCID50, was
used to calculate the Log Reduction Value (LRV). The LRV
of a particular process step is the difference between the
input virus load and the output virus load and is the viral
clearance across a particular process.
Input virus load ðCCID50 Þ
LRV ¼ Log10
Output virus load ðCCID50 Þ
The output value used in the calculation of the LRV was
determined as follows: the virus load obtained from the titration assay was used if virus was detected with the titration.
If virus was not detected in the titration, then the virus load
obtained from the large volume assay of that sample was
used to determine the output virus load.
Results and Discussion
The results of the studies are summarized in Tables 2 and 3
for media 1 and Tables 4 and 5 for media 2. The virus load
obtained from each assay of the filtered, spiked load material
(see Tables 2 and 4) was averaged, and this average was used
to calculate the log reduction in virus for treated samples. For
both media types tested, the average virus load of the load
material before and after 0.22 lm filtration was within the
1.0 log assay variability, indicating that the experimental
conditions did not result in formation of large virus aggregates
which would have been removed by filtration and that all
reductions in virus titers can be attributed to the treatment condition. The one exception to this is assay 3 of the unfiltered
media 2 spiked load (see Table 4). Its virus load was 6.68
log10CCID50; the virus load of assay 1 obtained for this unfiltered sample was 7.68 log10CCID50. The reduction in virus
load of this unfiltered sample is most likely the result of repetitive freezing and thawing as well as assay variability.
Although the titer was 1.0 log lower for assay 3 than assay 1,
the results of our study are not impacted because only the filtered samples were used to determine the log reductions
Table 3. MMV Inactivation in Media 1: Titration Assays
Virus Titer
(log10CCID50/mL 95% CI)
Treatment
Condition
Collection
Point
Average-filtered
load
95 C, 2 s
–
3.12 0.27
B
C
B
C
B
C
B
C
95 C, 5 s
95 C, 10 s
95 C, 15 s
95 C, 30 s
105 C, 2 s
105 C, 5 s
105 C, 10 s
105 C, 15 s
105 C, 30 s
115 C, 2 s
115 C, 5 s
115 C, 10 s
115 C, 15 s
115 C, 30 s
B
C
Titration
2.18
1.82
0.83
0.83
0.83
0.83
1.30
0.83
Large Volume
0.40
0.58
0.00
0.00
0.00
0.00
0.46
0.00
NT
*NT
‡
NT
*NT
1.12 0.00
*NT
1.12 0.00
*NT
0.96 0.49
Virus Load
(log10CCID50 95% CI)
Titration
Large Volume
7.12 0.27
6.18
5.82
4.83
4.83
4.83
4.83
5.30
4.83
0.40
0.58
0.00
0.00
0.00
0.00
0.46
0.00
Log Reduction
(log10CCID50 95% CI)
Titration
Large Volume
–
–
NT
*NT
‡
NT
*NT
3.88 0.00
*NT
3.88 0.00
*NT
4.04 0.49
†
0.94
1.30
2.29
2.29
2.29
2.29
1.82
2.29
0.48
0.64
0.27
0.27
0.27
0.27
0.53
0.27
*NT
‡
NT
*NT
3.24 0.27
*NT
3.24 0.27
*NT
3.08 0.56
All 0.83 0.00
*NT
All 4.83 0.00
*NT
All 2.29 0.27
*NT
All 0.83 0.00 All 1.12 0.00 All 4.83 0.00 All 3.88 0.00 All 2.29 0.27 All 3.24 0.27
* Samples collected from point B were tested only with a CCID50 assay and not with a large volume assay. † Log reductions of less than 1.0 are not considered significant. ‡ Sample was not tested with a large voume assay because active virus was detected in the CCID50 assay. All media samples were
diluted 1:10 before assaying. 40 mL of each sample from collection point C was assayed in 800 wells (10 96-well plates) for large volume testing.
858
Biotechnol. Prog., 2009, Vol. 25, No. 3
achieved from the HTST treatment. It should be noted that
repetitive freezing and thawing probably has a negative impact
on viral titer, as indicated by the gradual decline over time of
load material titers in both unfiltered and filtered samples
of each media type. For this reason, however, the virus loads
of the spiked material pre-treatment were averaged from multiple assays to determine the log reduction.
HTST impact on test medias
The media types used in this study appeared unaffected by
HTST treatment. Visually, indicator cells inoculated with
untreated or treated unspiked media of either type looked
similar after assaying, indicating that HTST treatment caused
no visually observable changes to the media that were cytotoxic to 324K cells.
Media 1 results
Table 2 shows the titers of the unfiltered and filtered load
samples associated with media 1. Table 3 shows the results
Table 4. Spiked Load Titers of Media 2
Virus Titer
(log10CCID50/mL
95% CI)
Spiked Load
Unfiltered assay 1
Unfiltered assay 2
Unfiltered assay 3
Average unfiltered load
Filtered assay 1
Filtered assay 2
Filtered assay 3
Average filtered load
3.68
3.55
2.68
3.47
3.68
3.43
2.93
3.44
0.25
0.32
0.25
0.28
0.25
0.44
0.25
0.33
Virus Load
(log10CCID50
95% CI)
7.68
7.55
6.68
7.47
7.68
7.43
6.93
7.44
0.25
0.32
0.25
0.28
0.25
0.44
0.25
0.33
Process samples were assayed on multipe days and spiked load samples were assayed each time as positive controls. Only filtered spiked
load samples were used to determine the log reductions achieved with
HTST treatment. Spiked load assay numbers refer to titrations performed
on separate days.
of titration assays and large volume assays for all media 1
output samples. Titration assays showed that residual active
virus was present in samples treated at the following stringencies: 95 C, 2 s, at points B and C, and 95 C, 15 s, at
point B. Large volume assays detected residual active virus
only in media treated at 95 C for 15 s. The absence of any
detected residual active virus at points B or C in the large
volume assays of media treated at 95 C for 5 or 10 s suggests that treatment at this temperature for 15 s should completely inactivate MMV as well. But active virus was
detected in the large volume assay of media treated at 95 C
for 15 s. This is most likely due to sampling variations that
are inherent to dealing with small quantities of virus particles in a large volume of sample.
The large volume assays of all media 1 samples treated at
115 C or 105 C indicate that treatment at these temperatures
for 2, 5, 10, 15, or 30 s resulted in complete infectivity inactivation of MMV. This is shown by a 3.24 log reduction in
virus for these inactivation stringencies (see Table 3). Complete infectivity inactivation was also achieved by treatment
of MMV-spiked media 1 at a temperature of 95 C for 5, 10,
or 30 s as measured by large volume assays. The maximum
possible LRV of media 1 was limited by the required 1:10
dilution to eliminate cytotoxicity for the large volume
assays. Significant levels of MMV infectivity inactivation
([1.0 log reduction in virus load) were observed with
CCID50 or large volume assays for all stringencies at all collection points, except for media treated at 95 C for 2 s that
was collected at point B.
Media 2 results
Table 4 shows the results of all titration assays of unfiltered and filtered load material. Table 5 shows the results of
the titration and large volume assays of the output samples.
Significant infectivity inactivation was observed at all temperatures, times and collection points except in media treated
at 95 C for 2 s collected at points A and B. No data are
Table 5. MMV Inactivation in Media 2: Titration Assays
Virus Titer
(log10CCID50/mL 95% CI)
Treatment
Condition
Collection
Point
Average Filtered
Load
95 C, 2 s
–
3.44 0.33
B
C
B
C
B
C
95 C, 5 s
95 C, 10 s
95 C, 15 s
95 C, 30 s
105 C, 2 s
105 C, 5 s
‡
105 C, 10 s
105 C, 15 s
105 C, 30 s
115 C, 2 s
115 C, 5 s
115 C, 10 s
115 C, 15 s
115 C, 30 s
B
C
B
C
Titration
2.93
0.83
0.92
0.83
0.59
0.83
Large Volume
0.25
0.00
0.64
0.00
0.91
0.00
–
*NT
1.12 0.00
*NT
1.12 0.00
*NT
1.12 0.00
Virus Load
(log10CCID50 95% CI)
Titration
Large Volume
7.44 0.33
6.93
4.83
4.92
4.83
4.59
4.83
0.25
0.00
0.64
0.00
0.91
0.00
Log Reduction
(log10CCID50 95% CI)
Titration
Large Voume
–
–
–
*NT
2.88 0.00
*NT
2.88 0.00
*NT
2.88 0.00
†
0.51
2.61
2.52
2.61
2.85
2.61
0.41
0.33
0.72
0.33
0.97
0.33
*NT
4.56 0.33
*NT
4.56 0.33
*NT
4.56 0.33
All 0.83 0.00
*NT
All 4.83 0.00
*NT
All 2.61 0.33
*NT
All 0.83 0.00 All 1.12 0.00 All 4.83 0.00 All 2.88 0.00 All 2.61 0.33 All 4.56 0.33
0.83 0.00
0.83 0.00
*NT
1.44 0.85
4.83 0.00
4.83 0.00
*NT
2.56 0.85
2.61 0.33
2.61 0.33
*NT
4.88 0.91
40 mL of each sample from collection point C was assayed in 800 wells (10 96-well plates) for large volume testing.
* Samples collected from point B were tested only with a CCID50 assay and not with a large volume assay. † Log reductions of less than 1.0 are not
considered significant. ‡ 105 C, 10s, was not collected at point B because a gasket on the collection tube broke.
Biotechnol. Prog., 2009, Vol. 25, No. 3
available for treatment at 105 C for 10 s at collection point
B because a seal on the collection tube came loose during
the processing of this sample.
In the titration assays of media 2, residual active virus
was detected in samples treated at the following stringencies:
95 C for 2 s at point B; 95 C for 5 s at point B; 95 C for
10 s at point B. All CCID50 assays of media 2 treated at the
remaining stringencies showed no signs of viral infection. In
the three aforementioned samples treated at 95 C (2, 5, and
10 s) that had residual active virus at collection point B, no
such residual active virus was observed in these samples in
either the CCID50 or large volume assays of samples collected at point C.
Residual active virus (one out of eight hundred wells inoculated) was observed only in the large volume assay of
media 2 treated at 115 C for 30 s. As with the large volume assay of media 1 treated at 95 C for 15 s, this result is
most likely due to sampling variations that are inherent to
dealing with small quantities of virus particles in a large
volume of sample. Despite this apparent deviation from the
trend of more infectivity inactivation at higher temperatures
and longer times, HTST treatment of media 2 resulted
in complete infectivity inactivation of MMV at all other
treatment stringencies as measured by large volume assays.
The reduction in active virus in these conditions was
4.56 logs.
Conclusion
The safety to biopharmaceuticals produced in mammalian
cell culture is ensured through orthogonal approaches that
include (i) design of the production process and adherence to
current good manufacturing practices, (ii) extensive characterization and testing of the cell substrates used for production of biopharmaceutical, (iii) lot-to-lot quality control
measures such as testing for adventitious contaminants, and
(iv) evaluation of the purification process unit operations for
their capacity to remove potentially undetected adventitious
viral contaminants. A process design for creation of a barrier
to the introduction of adventitious viruses through raw materials, such as HTST treatment of media, represents an added
measure for the assurance of safety of biopharmaceuticals
derived from mammalian cells. The efficacy of high temperature, short time treatment or dry and wet heat exposure in
the blood and plasma derived therapeutics area, for virus
inactivation, has been well established.9–12 However, the
extent of the use of heat treatment (HTST) as a barrier to
pathogen introduction in the manufacture of biopharmaceuticals produced in mammalian cells is unclear. Various studies
have reported inactivation of parvoviruses using heat,6,13 but
these reports describe effective inactivation of the challenge
virus upon exposure to heat for multiple minutes, not seconds. In one review article,13 inactivation of MMV (5.4
logs) upon exposure to temperatures 98 C was reported,
but no details of the exposure time were given. The use of
high temperature for long treatment times is not an effective
means of media treatment as it may impact the cell culture
performance. In general, the results presented in this report
indicate that heat treatment as low as 95 C and as high as
115 C are effective at inactivating MMV without causing
visually detectable precipitation. For both media types, log
reductions of MMV would have likely been greater had the
stock titer of MMV or the media spike percentage of the
virus been higher. The use of virus spikes higher than 5%,
859
however, is not ideal as it may significantly change the composition of the test matrix. The virucidal efficacy of HTST
at the treatment stringencies tested is thus perhaps significantly greater than indicated by the results. MMV infectivity
inactivation is proportional to treatment at higher temperatures for a given amount of time, especially above 95 C. All
titrations across both media types in which active virus was
detected (5 out of 5) were of media treated at 95 C. Active
MMV was detected in the large volume assays but not the titration assays of samples treated at 95 C for 2 s and 15 s
and at 115 C for 30 s. Neither media type appears to protect
MMV from the damaging effects of heat, as levels of MMV
infectivity inactivation were generally similar in both media
types.
The results of several samples suggest that the extended
cooling phase that occurred between collection points B and
C may ensure viral infectivity inactivation. This is apparent
from the results of media 1 treated at 95 C for 15 s: residual
active virus was detected with titration assays at point B but
not at point C; large volume assays, however, did detect residual active virus in this sample. This indicates that between
points B and C, some reduction in virus took place, such
that the reduced activity could be detected only with the
expanded sensitivity of a large volume assay.
For media 2, a similar observation was made with samples
treated at 95 C for 2, 5 or 10 s. In these conditions, residual
active virus was detected at collection point B but was not
detected by titration assaying or large volume assaying of
samples collected at point C. Whether the active virus collected at point B following HTST is a random chance event
or a result of virus aggregate formation upon rapid cooling
from high temperatures is undetermined and will require further investigation. In summary, our results show that HTST
provides robust infectivity inactivation of MMV, especially
at the higher temperatures of 105 C and 115 C. Experimentation with other non-enveloped viruses in a broad range of
media types and repeat experimentation with MMV should
be conducted to further evaluate the effectiveness and consistency of heat treatment for the purpose of viral infectivity
inactivation. In addition, it is essential to evaluate the impact
of HTST treatment on the performance of the treated media
for cell culture parameters to ensure absence of any adverse
effects.
Acknowledgments
These studies were supported by Amgen, Inc. We thank Jim
Thomas for his approval of this work; C. Richard ILL for his
technical review; Victor Fung and Joseph Phillips for their support of projects.
Literature Cited
1. Garnick RL. Experience with viral contamination in cell culture.
Dev Biol Stand. 1996;88:49–56.
2. Sofer G. Part 3a: plasma and plasma products (heat and solvent/
detergent treatments). Biopharm Intl. 2002;S15–S23.
3. Lelie PN, Reesink HW, Lucas CJ. Inactivation of 12 viruses by
heating steps applied during manufacture of a hepatitis B vaccine. J Med Virol. 1987;23:297–301.
4. Siegl G, Bates RC, Berns KI, Carter BJ, Kelly DC, Kurstak E,
Tattersall P. Characteristics and taxonomy of Parvoviridae.
Intervirology. 1985;23:61–73.
5. Cotmore SF, Tattersall P. The autonomously replicating parvoviruses of vertebrates. Adv Virus Res. 1987;33:91–174.
860
6. Boschetti N, Wyss K, Mischler A, Hostettler T, Kempf C. Stability of minute virus of mice against temperature and sodium
hydroxide. Biologicals. 2003;31:181–185.
7. Tattersall P, Bratton J. Reciprocal productive and restrictive virus-cell interactions of immunosuppressive and prototype strains
of minute virus of mice. J Virol. 1983;46:944–955.
8. Romanowski P, Schleh M, Rajurs V, Chinniah S, Dehghani H.
Variables affecting titer and long-term stability of xenotropic
murine leukemia and mouse minute virus stocks. Bioprocess
Int. 2008;6:44–52.
9. Charm SE, Landau S, Williams B, Horowitz B, Prince AM, Pascual D. High-temperature short-time heat inactivation of HIV and
other viruses in human blood plasma. Vox Sang. 1992;62:12–20.
Biotechnol. Prog., 2009, Vol. 25, No. 3
10. Horowitz B. Inactivation of viruses found with plasma proteins.
Biotechnology. 1991;19:417–430.
11. Dorner F, Barrett PN, Eibl J. At what stage should virus inactivation be carrried out? Dev Biol Stand. 1993;81:137–143.
12. Goubran HA, Burnouf T, Radosevich M. Virucidal heat-treatment of single plasma units: a potential approach for developing
countries. Haemophilia. 2000;6:597–604.
13. Sofer G. Virus inactivation in the 1990s—and into the 21st century. Part 4: culture media, biotechnology products, and vaccines. Biopharm Int. 2003;50–57.
Manuscript received July 21, 2008, and revision received Nov. 14, 2008.
Документ
Категория
Без категории
Просмотров
2
Размер файла
148 Кб
Теги
1/--страниц
Пожаловаться на содержимое документа