Application Note
Cell Culture Media Microfiltration and Mixing:
Reducing Risk with Single-use Technology
Abstract
Cell culture media are complex mixtures of synthetic and
biological components that provide the proper composition
of nutrients for healthy cell propagation and high protein
expression. Key concerns are to ensure product safety and
purity. Dissolution of cell culture media can be facilitated by
single-use mixing systems. Disposable mixers and filtration
assemblies provide operational advantages by eliminating
clean-in-place (CIP) and steam-in-place (SIP) process steps,
resulting in a reduction in water usage and decreased
WFI consumption. In addition, the use of disposables
can reduce the risk of microbial contamination. This case
study presents our recommended methods for thoroughly
mixing media, as well as reducing the risk of mycoplasma
contamination through the use of validated 0.1 µm
sterilizing-grade filters.
Introduction
Cell culture media is typically mixed in bulk and aseptically
transferred to the bioreactor. Careful preparation of cell
culture media is needed to ensure the sterility of the cell
culture batch. Cell culture media spans many media types
and formulations to meet the needs of different cell lines.
As a result, a wide variety of media constituents exist today
at a wide range of concentrations. Prefiltration is commonly
used to remove the bulk of particulate and colloidal
contaminants from the media and to extend the service life
of the downstream sterilizing-grade filters. At production
scale, prefilters are often used to handle batch-to-batch
media variability and to ensure that the sterile media fill
into the bioreactor is completed successfully and on time.
The objective of final sterilizing-grade filtration is to
remove bacteria and mycoplasma to preserve the aseptic
barrier around the bioreactor.
Media and media additives must meet
the following characteristics:
• Sterile / mycoplasma-free
• Regulatory compliant
• Promote proper cell growth and expression
• Free of growth inhibitors and contaminants
– Low in extractables
– Low in endotoxins
Mycoplasma Contamination
A cell culture developer’s objectives when preparing media
for addition into the bioreactor are to ensure that the
filtered medium provides proper cell growth and expression
levels and that the bioreactor and downstream purification
steps remain free from microbial contamination.
Mycoplasma contamination is a tremendous concern for
biopharmaceutical manufacturers, as these small, insidious
bacteria thrive in the nutrient rich environment of the
bioreactor, are capable of passing through a 0.2 µm
sterilizing-grade filter, and can induce changes to cell
metabolism and expression levels. Mycoplasma can be
brought into a cell culture process from plant- or animal-
derived raw materials as well from humans. Acholeplasma
laidlawii, Mycoplasma arginini and M. hyorhinis are isolated
 from hydrolysates, cell lines and sera. M. fermentans,
M. orale, and M. salivarium are isolated from humans,
animals and cell lines. The nutritive composition of cell
culture media supplies the sterols, fatty acids, amino acids,
salts and carbohydrates necessary for the mycoplasma
species to thrive.
Mycoplasma can be difficult to detect in cell culture
because many species do not produce turbidity or cyto-
pathic effects. Due to their diminutive size and deformabil-
ity, mycoplasma readily pass through 0.2 µm-rated devices.
As a result, media preparation increasingly calls for the use
bility performance among 0.1 µm-rated hydrophilic filters
from three different manufacturers, using an animal-free
medium. In this study, membrane samples were tested
simultaneously in three separate experiments using three
separate cultures of A. laidlawii ATCC® 23206. The average
total challenge per membrane was 4.9 x 108 cfu/filter
Figure 1. Mycoplasma Reduction and Permeability
Performance of Millipore Express SHR and Other
Single Layer Membranes
12 360

[log10(Total Challenge/Total Passage)]

of 0.1 µm sterilizing-grade filters to ensure sufficient log 320

= Mean Challenge Permeability

10
removal of mycoplasma. 280

Mean Log Reduction Value

8 240

(LMH/psid)
200

Reducing the risk of 6

160

mycoplasma contamination 4 120

80
For bioreactor feed applications, membrane filter devices 2
40
are designed for various performance characteristics, such
0 0
as capacity, permeability, and mycoplasma removal. Capacity Millipore Express Manufacturer A Manufacturer B
SHR 0.1 µm PES 0.1 µm PVDF 0.1 µm
and microbial removal should be in balance to ensure a safe Single Layer Single Layer Single Layer
and efficient process.
Error Bars = ± 1 standard deviation

Application requirements: (± 5.4 x 107Acholeplasma

) and the laidlawii
average challenge per membrane area
Mean Log Reduction Value and
Process Development was 3.5 xChallenge
107 cfu/cm 2 (± for
Permeability 3.9Various
x 106Single
). Layer Membranes
• Filtration sizing experiments Mycoplasma removal and permeability varied under the
• Validation of sterility / mycoplasma clearance using conditions of this test. The log reduction value among the
validated test methods membranes was different (ANOVA, α 0.05, P=0.000). Millipore
Express® SHR 0.1 µm (n = 10) membrane and Manufacturer A’s
Production PES (n = 9) membrane performed equivalently (Tukey post-test).
• Bioburden monitoring, mycoplasma detection However, Millipore Express SHR 0.1 µm membranes exhibited
• Filter robustness/ease of handling evaluations less variability than those of Manufacturer A. The PVDF
membrane from Manufacturer B (n = 11) exhibited lower
removal of A. laidlawii.
Evaluating filter capacity by modeling filter throughput The flow time was measured using A. laidlawii challenge
over time is a well known method in the industry and fairly suspension transformed to permeability (LMH/psid). Millipore
uniform in its application. However, methods for testing a Express SHR 0.1 µm membrane challenge permeability (n = 6)
filter’s capability for mycoplasma removal are still evolving, was significantly higher than those of Manufacturer A (n = 6)
and there is currently no standard method for rating and Manufacturer B (n = 8) (ANOVA, α 0.05, P = 0.004).
mycoplasma clearance filters. As a result, testing should
be performed to verify the filter’s mycoplasma removal
Considerations for selecting a
properties, using the product under actual processing
mycoplasma clearance filter:
conditions.
Figure 1 illustrates mycoplasma reduction and permea- • Unit operation risk
• Process compatibility
• Mycoplasma clearance
• Filtration flux and capacity
• Filter validation

2
Consistent Filtration Figure 3. Millipore Express SHR media throughput
performance on disks and pleated devices
Characteristics of
Predictable, scalable throughput; 47 mm disk testing predicts
Cell Culture Media device performance within 20%..
In addition to reducing the risk of mycoplasma contamina-
Water Cell Culture Media
tion, a cell culture developer must also ensure that the
Scaling Factor 1000 Scaling Factor 1000
600
filtration train is robust and can handle variations in feed = 0.94 = 0.95
900 900
and batch media. To address this need, it is recommended

Throughput at 80% Flow Decay (L/m2)

Water Permeability at 23 °C (LMH/psi)
500
800 800
to characterize filter variability and to size the filtration
700 700
system appropriately. 400

Throughput (L/m2)
600 600
Towards this objective, Millipore initiated a case study
to mimic the way a cell culture developer would assess 300 500 500

filter lot variability with cell culture media. Throughput 400 400
200
testing (to at least 80% of initial flow) was performed, using 300 300

three samples each (n=3) of Millipore Express SHR 0.1 µm 100

200 200

membrane from three different, non-contiguous lots. 100 100

Throughput testing was run at 10 psi constant pressure, 0 0 0

47 mm 3 in. 47 mm 3 in. 0
using one batch of 5 g/L soy hydrolysate mixed with Disk Cartridge Disk Cartridge
n=7 n=4 n=4 n=2
10 g/L DMEM. Results, shown in Figure 2, demonstrate
that Millipore Express SHR 0.1 µm membranes have minimal
Water
lot-to-lot variation in filterability. Cell Culture Media Cell Culture Media
Scaling Factor Scaling Factor Throughput at
Figure
600 3 shows Millipore Express SHR
= 0.94
10000.1 µm through-
= 0.95
1000
80% Flow Decay (L/m2)
put performance on disks and pleated devices.
900 Throughput 900
Throughput at 80% Flow Decay (L/m2)
Water Permeability at 23 °C (LMH/psi)

500
testing performed with water and cell culture
800 media on 800

47 mm disks
400
and Opticap® XL3 devices showed
700 similar 700
Throughput (L/m2)

results, demonstrating consistent and predictable

600 filter 600
47 mm Disk 1
performance.
300 500 500 47 mm Disk 2
3 in. Cartridge
400 400
200
300 300
Figure 2. Lot-to-lot consistency of 200 200
Millipore
100Express SHR 0.1 µm membrane
100 100

1.50 0 0 0
47 mm 3 in. 47 mm 3 in. 0 500 1000 1500 2000 2500
Disk Cartridge Disk Cartridge
n=7 n=4 n=4 n=2 Time (sec)
1.25
Relative Vmax

1.00
CELL CULTURE MEDIA MIXING WITH
0.75
THE Mobius® MIX200 SYSTEM
When dissolving cell culture media, consistent mixing is
essential. Mobius single-use mixing systems are designed to
0.50
1 2 3 maintain uniformity as exhibited in this study. This technol-
Millipore Express SHR Lot ogy delivers economic and operational efficiency, saving
validation time and increasing operational flexibility. The
The study showed Millipore Express SHR 0.1 µm media provided
consistent performance with only minimal variation in filterability Mobius MIX200 system uses a bottom-mounted, magneti-
across three non-consecutive filter lots. cally driven impeller inside a 3-D conical single-use process
container made of Pureflex™ film. This high-purity, medical-
grade film was designed to provide strength and flexibility,
as well as inert contact and excellent gas barrier perfor-

3
mance. As a result Mobius mixing containers provide high Figure 4. Efficient Media Mixing with the Mobius
performance when engaged with the motor and electronics. MIX200 Disposable Mixing System
Mimicking the design of traditional stainless steel vessels, 0.020
the Mobius MIX200 system minimizes foaming potential,
particle generation and vessel contamination.

Resistance (psi/LMH)
0.015
Figure 4 shows the consistency of Mobius MIX200
system data when mixing serum-free media containing 5 g/L
0.010
soy hydrolysate. Samples taken from the top and bottom
ports of the Mobius MIX200 system and filtered through a
0.005 Disk Top
single lot of Millipore Express SHR membrane, show similar
Disk Bottom
results, demonstrating efficient mixing.
0.000
0 200 400 600 800 1,000
Summary Millipore Express SHR Throughput (L/m2)
This case study demonstrates the robustness and capability The study method was a throughput testing run in constant flow
of Millipore’s single-use Mobius MIX200 disposable mixing mode, mixing 200 liters of water with 5 g/L of soy hydrolysate.
The media was sampled from the top and bottom of the MIX200
technology and sterilizing-grade, mycoplasma clearance system and filtered through Millipore Express SHR 0.1 µm
filters with Millipore Express 0.1 µm membrane to meet membrane disks.
a cell culture developer’s needs for improved process
efficiency and reduced risk of microbial contamination.

To Place an Order or Receive

Technical Assistance
In the U.S. and Canada, call toll-free 1-800-MILLIPORE
(1-800-645-5476)
Outside of North America contact your local office. To find
the office nearest you, visit: www.millipore.com/offices

Internet: www.millipore.com

Technical Service: www.millipore.com/techservice

www.millipore.com/offices

Millipore, Mobius, Millipore Express, and Opticap are registered trademarks of

Millipore Corporation. The M mark, LucraTone, Pureflex, and Advancing Life Science
Together are trademarks of Millipore Corporation.
ATCC is a trademark of the American Type Culture Collection.
Lit. No. AN13621EN00 Rev. -  Printed in U.S.A. 02/09 DP-SBU-08-00652
© 2009 Millipore Corporation, Billerica, MA 01821 U.S.A. All rights reserved.