Host cell protein information

1. ©2013
Landes
Bioscience.
Do
not
distribute
Bioengineered 4:5, 288–291; September/October 2013; © 2013 Landes Bioscience
Commentary
288 Bioengineered Volume 4 Issue 5
Commentary to: Tait AS, Hogwood CE, Smales
CM, Bracewell DG. Host cell protein dynamics
in the supernatant of a mAb producing CHO
cell line. Biotechnol Bioeng 2012; 109:971-82;
PMID:22124969; http://dx.doi.org/10.1002/
bit.24383 and Hogwood CE, Tait AS, Koloteva-
Levine N, Bracewell DG, Smales CM. The dynam-
ics of the CHO host cell protein profile during
clarification and protein A capture in a platform
antibody purification process. Biotechnol Bioeng
2013; 110:240-51; PMID:22806637; http://dx.doi.
org/10.1002/bit.24607.
Keywords: host cell protein, Chinese
hamster ovary (CHO), mammalian cell
culture, downstream processing, protein
A chromatography, monoclonal antibody,
proteomics
Submitted: 12/06/12
Revised: 12/18/12
Accepted: 12/21/12
http://dx.doi.org/10.4161/bioe.23382
*Correspondence to: C Mark Smales;
Email: c.m.smales@kent.ac.uk
During the production of recom-
binant protein products, such as
monoclonal antibodies, manufacturers
must demonstrate clearance of host cell
impurities and contaminants to appro-
priate levels prior to use in the clinic.
These include host cell DNA and RNA,
product related contaminants such as
aggregates, and importantly host cell pro-
teins (HCPs). Despite the importance of
HCP removal, the identity and dynamics
of these proteins during cell culture and
downstream processing (DSP) are largely
unknown. Improvements in technologies
such as SELDI-TOF mass spectrometry
alongside the gold standard technique of
ELISA has allowed semi-quantification
of the total HCPs present. However, only
recently have techniques been utilized
in order to identify those HCPs pres-
ent and align this with the development
of approaches to monitor the dynam-
ics of HCPs during both fermentation
and downstream processing. In order to
enable knowledge based decisions with
regards to improving HCP clearance it
is vital to identify potential problematic
HCPs on a cell line and product specific
basis. Understanding the HCP dynamics
will in the future help provide a platform
to rationally manipulate and engineer
and/or select suitable recombinant CHO
cell lines and downstream processing
steps to limit problematic HCPs.
Introduction
At present therapeutic recombinant mono-
clonal antibodies (mAb) and Fc-fusion
proteins dominate the biopharmaceutical
Host cell protein dynamics in recombinant CHO cells
Impacts from harvest to purification and beyond
Catherine EM Hogwood,1
Daniel G Bracewell2
and C Mark Smales1,
*
1
Centre for Molecular Processing and School of Biosciences; University of Kent; Canterbury, Kent, UK; 2
Advanced Centre for Biochemical Engineering;
Department of Biochemical Engineering; University College London; Torrington Place, London, UK
market, accounting for 35% of all biother-
apeutic proteins.1
Indeed, industry is now
capable of producing in excess of 5 g/L of
recombinant protein product in mamma-
lian cell cultures.1
The production of com-
plex recombinant proteins such as mAbs,
requires a system that possesses the cellular
machinery capable of processing, folding,
assembling and post-translationally modi-
fying the product to generate the authen-
tic required target protein, and Chinese
hamster ovary cells (CHO) are routinely
used for the expression of such proteins.2
The recombinant product is secreted from
the cell into the surrounding media and
hence it is necessary to recover this from
the harvested cell culture fluid (HCCF)
via a series of downstream processing
(DSP) steps. These steps are designed to
purify the product, removing host cell
DNA/RNA, lipids, host cell proteins
(collectively referred to as process related
impurities) and product related contami-
nants.3
The requirements placed upon
this process include removal of HCPs in
the final product to < 1–100 ppm.4,5
The
process contaminants are of concern in
the biopharmaceutical sector as adverse
clinical effects have been reported.6,7
Of
concern is not only that CHO HCPs in
the final product could illicit an immune
response in the patient but also that due
to the similarity between many CHO
and human proteins cross-reactivity may
result in autoimmunity.6
These concerns
underpin the importance of understand-
ing HCP identity, the processes by which
they appear in the HCCF and dynamics
during recombinant protein production
and subsequent DSP steps.

2. ©2013
Landes
Bioscience.
Do
not
distribute
www.landesbioscience.com Bioengineered 289
Commentary Commentary
A (LDH-A) and protein disulfide isom-
erase (PDI), most likely released due to
cellular lysis or breakdown either during
fermentation or primary recovery.9
How can such information on the
generation of the exact nature, abun-
dance, identity and relevance of these
HCPs contaminants be used in furthering
our process understanding with respect
to HCPs? From the cost perspective, a
small-scale evaluation revealed that ~80%
of mAb production costs were down-
stream process related.19
Further, despite
improvements in product titer upstream,
downstream process improvements to
cope with this increased load have been
slow. Hence information on HCP pres-
ence and removal is important in order to
make knowledge-based process decisions
within the context of the whole process.
Although DSP of monoclonal antibod-
ies for example relies heavily on centrifu-
gation and filtration steps, few reports
focus on the impact of these on the HCP
profile which is subsequently in the feed
stream introduced into the chromatogra-
phy workflow. Published data once again
highlights the importance of process
understanding regarding HCPs in this
respect, as different clarification decisions
influences the HCP profile and its abun-
dance.12
Various HCPs also responded in
a different manner to this initial DSP step
depending on the clarification step cho-
sen.12
There are also likely to be cell line
specific differences in the response to DSP
with respect to HCPs, which should also
be taken into consideration when recom-
binant CHO cell lines are selected.12
Recombinant protein purification
relies heavily upon chromatography steps,
with protein A capture chromatogra-
phy being the “gold standard” for mAb
purification, demonstrating ~98% prod-
uct purity prior to subsequent polishing
steps.3,20
However, despite this excellent
performance HCPs remain. This is likely
to be due to recent issues raised in the lit-
erature, such as protein-protein interac-
tions between product and HCPs during
protein A chromatography, an issue that
largely remains a black-box and likely to
be highly dependent on the cell line (i.e.,
the HCPs in the feedstream) and product
in question.21
Interactions between HCPs,
product and chromatography resins are
of additional methods to quantify and/
or identify HCPs; such as the use of fou-
rier transform mid infrared spectroscopy
(FT-MIR) and 2D-LC/MS,14,15
these
approaches can be further complemented
to aid in identification of greater numbers
of HCPs and to follow their fate during
DSP. An increased need for rapid and
accurate HCP detection and quantifi-
cation during the recombinant protein
production workflow may be met by
such approaches and robotic systems as
described by Rey et al.16
Ultra-scale down
mimics of process scale unit operations
may also allow the rapid assessment of
the effects of processing on the HCP pro-
file,9,12,17
reducing costs in acquiring pro-
cess understanding and providing suitable
evidence for scale-up. Finally, the recent
publication of the CHO-K1 genome,18
the
principle mammalian host cell used for
recombinant protein production, should
lead to the emergence of more informat-
ics based approaches, such as described by
Gutierrez et al.6
in which immunoinfor-
matic tools were used to evaluate immu-
nogenic potential of CHO HCPs.
HCPs from Harvest to Purification
Taking a top-down approach with regard
to investigating HCPs in the manufactur-
ing workflow requires the determination
of those HCPs present during fermenta-
tion of recombinant CHO cells. Various
upstream decisions can impact upon the
HCP profile; including cell culture media,
culture feeding strategies and bioreactor
control.8
However, cell culture duration is
reported as having the greatest impact on
the observed HCP profile as determined
using 2D-PAGE, SELDI-TOF and ELISA
analysis.8,9
Comparing harvest days (vary-
ing from day 10 to day 14) for HCP
content revealed a dynamic HCP profile
affected by (1) the cell line and its recom-
binant protein product, as some proteins
may be cell line or product specific, (2) cell
viability, in which an increase in HCPs is
observed when viability declines8,9
and (3)
the viable and non-viable cell populations,
as viable cells exhibited increased sensitiv-
ity to shear experienced during centrifuga-
tion.9,17
Thus, it is not surprising that the
HCP profile largely consists of intracellu-
lar proteins, such as lactate dehydrogenase
The HCP Monitoring
and Assessment Toolbox
The current toolbox available to measure/
monitor total HCP concentration includes
enzyme-linked immunosorbent assays
(ELISA), of which few kits are commer-
cially available. These kits are produced
by injecting animal models with an HCP
mixture to raise antibodies. The HCP
mixture is commonly the null cell line
(containing an empty vector) at a cellular
harvest level where the “general” HCP pop-
ulation is well represented between both
producer and null.8,9
There are a number
of potential drawbacks to this technique,
for example if the protein is not present in
the mixture or does not illicit an immune
response in the animal model then it will
not be detected in the sample. This raises
the question of how well any one ELISA
covers the HCP profile, however ELISA
is widely used in the biopharmaceutical
industry to determine HCP levels and is
the current gold-standard methodology.7
The biotechnology industry use both com-
mercially available HCP ELISA kits and
customised in-house designed assays.7
Two-dimensional polyacrylamide gel
electrophoresis (2D-PAGE) is an approach
previously applied to both bacterial and
mammalian cell lines to determine HCP
dynamics.8,10,11
2D-PAGE as a qualitative
technique allows a profile to be generated,
from which process conditions for exam-
ple can be compared and changes quan-
tified. This approach is often combined
with ELISA technology in order to further
quantify HCPs. Technical limitations of
2D-PAGE include that only proteins of
high abundance in a protein mixture will
be visualized. Further, when this tech-
nique is applied to product producing
cell lines (such as mAbs) the product can
“swamp” the profile either masking pro-
tein spots or making it difficult to visual-
ize low abundant contaminating proteins
on the same gel, in which case the null
cell line is usually investigated.8,12
In addi-
tion to this global proteomics approach,
SELDI-TOF mass spectrometry has
enabled changes in the HCP profile to be
rapidly determined.9,13
Large volumes of supernatant mate-
rial are not required for SELDI-TOF and
2D-PAGE analysis. With the emergence

3. ©2013
Landes
Bioscience.
Do
not
distribute
290 Bioengineered Volume 4 Issue 5
engineering to reduce problematic HCPs
may lessen the need for additional purifi-
cation steps to remove these from the final
biopharmaceutical product.
Another potential problem that exists
at present and poses an additional level
of complexity is other trace level impuri-
ties. Other trace level impurities, below
current detection level techniques, are
most likely present either in the HCCF
and/or the purified protein product (e.g.,
proteases) that can potentially impact
upon protein integrity.27
Indeed, degra-
dation of an Fc-fusion protein led to the
identification of a homolog of the prote-
ase cathepsin D in fractions post-protein
A purification.28
It also appears any link
between the concentration of HCP and
protease activity is tenuous as it has been
reported that at high levels of HCPs, pro-
tease related mAb fragmentation appeared
not to occur.29
These contaminants can
play a significant role in the degradation
of the product, in particular mAbs, lead-
ing to functional titer loss. Cytokines are
another contaminant previously shown to
be present in the supernatant of cultured
CHO cells. Latent transforming growth
factor-β1 has recently been shown to be
secreted by CHO cells and to be func-
tional in human cells.30
Although DSP
removed this cytokine, any trace could
cause profound effects in a patient. This
again highlights the importance of under-
standing potential contaminants and at
which steps they are removed during DSP.
Thus, identifying, characterizing and
understanding HCPs and their process
interactions during recombinant protein
production is essential in developing or
tailoring rational approaches to remove
them, safeguarding the final biopharma-
ceutical product for the patient.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were
disclosed
careful selection of membranes for flow-
through polishing. Evaluation of various
commercially available anion exchange
membranes reported differences in HCP
removal and this appears to be further
influenced by process conditions, dem-
onstrating the importance of process
understanding even at the later stages of
purification.26
All of the literature data points toward
the fact that during process development
HCPs should be considered on a cell line
and recombinant protein product basis,
particularly with regard to the influence
of culture conditions, cell culture dura-
tion, primary recovery, clarification steps
and those HCPs carrying through chro-
matography procedures. This informa-
tion can then be used to feed-back further
upstream to aid either cell line selection
strategies (i.e., select a cell line with a
reduced number of problematic HCPs)
and/or to influence DSP decisions.
Future Prospects:
An Engineered Approach?
It is envisaged that ultimately informa-
tion on HCPs and their dynamics will
be used to make knowledge based pro-
cess decisions. This may take the form of
(1) cell screens of multiple recombinant
CHO cell lines in order to select the most
suitable cell line for your protein to be
produced i.e., exhibiting reduced levels of
problematic HCPs; (2) cell line engineer-
ing to generate a more robust cell line to
decrease HCP release during fermenta-
tion and clarification steps or to knock-
down a problematic HCP (although
targets would have to be selected with
extreme caution); (3) media formula-
tion and feeding strategies to improve
cell strength and (4) selecting optimal
downstream processing steps tailored
to allow removal of problematic HCPs.
Knowledge-based decisions such as cell
therefore being extensively investigated at
present. In the case of mAbs and protein
A chromatography, a proportion of HCPs
co-elute via interaction with the prod-
uct itself rather than with the chromato-
graphic resin.21-23
However, non-specific
interactions between HCPs and protein A
resins have also been reported.12,23
As mAb purification follows a rather
fixed template, research aiming to
improve purification focus mainly on the
replacement or improvement of the cap-
ture step, thus providing further infor-
mation on HCP dynamics. For example,
combining multiple interactions within
one resin (mixed-mode chromatography)
has recently been evaluated for their suit-
ability as a primary capture step in the
purification of mAb from CHO super-
natant.24
While yields were compliant
with a capture step, HCP removal varied
among the resins tested and demonstrated
that populations of HCPs also differed.
Therefore the physiochemical characteris-
tics of HCPs in the feed stream play a role
in dictating their chromatographic behav-
ior.24
Taking an alternative approach to
the standard capture chromatography,
Borlido et al.25
recently demonstrated that
magnetic separation techniques could
provide an alternative purification strat-
egy that is fast, highly specific and poten-
tially more cost effective than the gold
standard protein A. Boronic acid mag-
netic particles (as an alternative ligand to
protein A) were successfully used for mAb
capture from CHO HCCF. This method
provided a higher yield under optimised
conditions than that of protein A and
showed promise for HCP removal (remov-
ing only 12% less than protein A).25
The
conditions under which DSP steps are run
may also be adapted, such as alternative
washes during capture chromatography to
disrupt product-HCP interactions.22
The improvement of polishing steps
with regard to HCP removal include the
References
1. Marichal-Gallardo PA, Alvarez MM. State-of-the-art
in downstream processing of monoclonal antibodies:
process trends in design and validation. Biotechnol
Prog 2012; 28:899-916; PMID:22641473; http://
dx.doi.org/10.1002/btpr.1567.
2. Zhu J. Mammalian cell protein expression for bio-
pharmaceutical production. Biotechnol Adv 2012;
30:1158-70; PMID:21968146; http://dx.doi.
org/10.1016/j.biotechadv.2011.08.022.
3. Guiochon G, Beaver LA. Separation science is the
key to successful biopharmaceuticals. J Chromatogr
A 2011; 1218:8836-58; PMID:21982447; http://
dx.doi.org/10.1016/j.chroma.2011.09.008.
4. Champion K, Madden H, Dougherty J, Shacter E.
Defining your product profile and maintaining con-
trol over it, part 2. Bioprocess Int 2005; 5:52-7.
5. Chon JH, Zarbis-Papastoitsis G. Advances in the pro-
duction and downstream processing of antibodies.
N Biotechnol 2011; 28:458-63; PMID:21515428;
http://dx.doi.org/10.1016/j.nbt.2011.03.015.
6. Gutiérrez AH, Moise L, De Groot AS. Of [Hamsters]
and men: A new perspective on host cell pro-
teins. Hum Vaccin Immunother 2012; 8:1172-4;
PMID:23124469; http://dx.doi.org/10.4161/
hv.22378.
7. Wang X, Hunter AK, Mozier NM. Host cell pro-
teins in biologics development: Identification, quan-
titation and risk assessment. Biotechnol Bioeng
2009; 103:446-58; PMID:19388135; http://dx.doi.
org/10.1002/bit.22304.

4. ©2013
Landes
Bioscience.
Do
not
distribute
www.landesbioscience.com Bioengineered 291
24. Pezzini J, Joucla G, Gantier R, Toueille M, Lomenech
AM, Le Sénéchal C, et al. Antibody capture by
mixed-mode chromatography: a comprehensive study
from determination of optimal purification condi-
tions to identification of contaminating host cell
proteins. J Chromatogr A 2011; 1218:8197-208;
PMID:21982448; http://dx.doi.org/10.1016/j.chro-
ma.2011.09.036.
25. Borlido L, Azevedo AM, Sousa AG, Oliveira PH,
Roque AC, Aires-Barros MR. Fishing human mono-
clonal antibodies from a CHO cell supernatant
with boronic acid magnetic particles. J Chromatogr
B Analyt Technol Biomed Life Sci 2012; 903:163-
70; PMID:22857861; http://dx.doi.org/10.1016/j.
jchromb.2012.07.014.
26. Weaver J, Husson SM, Murphy L, Wickramasinghe
SR. Anion exchange membrane adsorbers for flow-
through polishing steps: Part II. Virus, host cell
protein, DNA clearance, and antibody recovery.
Biotechnol Bioeng 2012; 110: PMID:22951992.
27. Gao SX, Zhang Y, Stansberry-Perkins K, Buko A, Bai
S, Nguyen V, et al. Fragmentation of a highly purified
monoclonal antibody attributed to residual CHO cell
protease activity. Biotechnol Bioeng 2011; 108:977-
82; PMID:21404269; http://dx.doi.org/10.1002/
bit.22982.
28. Robert F, Bierau H, Rossi M, Agugiaro D, Soranzo
T, Broly H, et al. Degradation of an Fc-fusion recom-
binant protein by host cell proteases: Identification
of a CHO cathepsin D protease. Biotechnol Bioeng
2009; 104:1132-41; PMID:19655395; http://dx.doi.
org/10.1002/bit.22494.
29. Cordoba AJ, Shyong BJ, Breen D, Harris RJ. Non-
enzymatic hinge region fragmentation of antibodies
in solution. J Chromatogr B Analyt Technol Biomed
Life Sci 2005; 818:115-21; PMID:15734150; http://
dx.doi.org/10.1016/j.jchromb.2004.12.033.
30. Beatson R, Sproviero D, Maher J, Wilkie S, Taylor-
Papadimitriou J, Burchell JM. Transforming
growth factor-β1 is constitutively secreted by
Chinese hamster ovary cells and is functional in
human cells. Biotechnol Bioeng 2011; 108:2759-
64; PMID:21618471; http://dx.doi.org/10.1002/
bit.23217.
16. Rey G, Wendeler MW. Full automation and valida-
tion of a flexible ELISA platform for host cell protein
and protein A impurity detection in biopharma-
ceuticals. J Pharm Biomed Anal 2012; 70:580-
6; PMID:22727805; http://dx.doi.org/10.1016/j.
jpba.2012.05.027.
17. Tait AS, Aucamp JP, Bugeon A, Hoare M. Ultra
scale-down prediction using microwell technology
of the industrial scale clarification characteristics by
centrifugation of mammalian cell broths. Biotechnol
Bioeng 2009; 104:321-31; PMID:19507199; http://
dx.doi.org/10.1002/bit.22393.
18. Hammond S, Kaplarevic M, Borth N, Betenbaugh
MJ, Lee KH. Chinese hamster genome database:
an online resource for the CHO community at
www.CHOgenome.org. Biotechnol Bioeng 2012;
109:1353-6.
19. Vermasvuori R, Hurme M. Economic compari-
son of diagnostic antibody production in perfusion
stirred tank and in hollow fiber bioreactor processes.
Biotechnol Prog 2011; 27:1588-98; PMID:21954092;
http://dx.doi.org/10.1002/btpr.676.
20. Low D, O’Leary R, Pujar NS. Future of antibody
purification. J Chromatogr B Analyt Technol Biomed
Life Sci 2007; 848:48-63; PMID:17134947; http://
dx.doi.org/10.1016/j.jchromb.2006.10.033.
21. Nogal B, Chhiba K, Emery JC. Select host cell pro-
teins coelute with monoclonal antibodies in protein
A chromatography. Biotechnol Prog 2012; 28:454-
8; PMID:22275211; http://dx.doi.org/10.1002/
btpr.1514.
22. Shukla AA, Hinckley P. Host cell protein clearance
during protein A chromatography: development of
an improved column wash step. Biotechnol Prog
2008; 24:1115-21; PMID:19194921; http://dx.doi.
org/10.1002/btpr.50.
23. Tarrant RD, Velez-Suberbie ML, Tait AS, Smales
CM, Bracewell DG. Host cell protein adsorption
characteristics during protein A chromatography.
BiotechnolProg2012;28:1037-44;PMID:22736545;
http://dx.doi.org/10.1002/btpr.1581.
8. Jin M, Szapiel N, Zhang J, Hickey J, Ghose S.
Profiling of host cell proteins by two-dimensional dif-
ference gel electrophoresis (2D-DIGE): Implications
for downstream process development. Biotechnol
Bioeng 2010; 105:306-16; PMID:19739084; http://
dx.doi.org/10.1002/bit.22532.
9. Tait AS, Hogwood CE, Smales CM, Bracewell DG.
Host cell protein dynamics in the supernatant of a
mAb producing CHO cell line. Biotechnol Bioeng
2012; 109:971-82; PMID:22124969; http://dx.doi.
org/10.1002/bit.24383.
10. Aldor IS, Krawitz DC, Forrest W, Chen C, Nishihara
JC, Joly JC, et al. Proteomic profiling of recombi-
nant Escherichia coli in high-cell-density fermenta-
tions for improved production of an antibody frag-
ment biopharmaceutical. Appl Environ Microbiol
2005; 71:1717-28; PMID:15811994; http://dx.doi.
org/10.1128/AEM.71.4.1717-1728.2005.
11. Grzeskowiak JK, Tscheliessnig A, Toh PC,
Chusainow J, Lee YY, Wong N, et al. 2-D DIGE to
expedite downstream process development for human
monoclonal antibody purification. Protein Expr Purif
2009; 66:58-65; PMID:19367714; http://dx.doi.
org/10.1016/j.pep.2009.01.007.
12. Hogwood CE, Tait AS, Koloteva-Levine N,
Bracewell DG, Smales CM. The dynamics of the
CHO host cell protein profile during clarification
and protein A capture in a platform antibody puri-
fication process. Biotechnol Bioeng 2013; 110:240-
51; PMID:22806637; http://dx.doi.org/10.1002/
bit.24607.
13. Kumar N, Maurya P, Gammell P, Dowling P,
Clynes M, Meleady P. Proteomic profiling of
secreted proteins from CHO cells using Surface-
Enhanced Laser desorption ionization time-of-flight
mass spectrometry. Biotechnol Prog 2008; 24:273-
8; PMID:18163642; http://dx.doi.org/10.1021/
bp070244o.
14. Capito F, Skudas R, Kolmar H, Stanislawski B. Host
cell protein quantification by Fourier transform mid
infrared spectroscopy (FT-MIR). Biotechnol Bioeng
2013; 110:252-9; PMID:22811255; http://dx.doi.
org/10.1002/bit.24611.
15. Schenauer MR, Flynn GC, Goetze AM. Identification
and quantification of host cell protein impurities
in biotherapeutics using mass spectrometry. Anal
Biochem 2012; 428:150-7; PMID:22640604; http://
dx.doi.org/10.1016/j.ab.2012.05.018.