Journal 9

JBA-06934; No of Pages 12
Biotechnology Advances xxx (2015) xxxxxx
Contents lists available at ScienceDirect
Biotechnology Advances
journal homepage: www.elsevier.com/locate/biotechadv
Research review paper
Extraction and downstream processing of plant-derived

recombinant proteins
J.F. Buyel a,b,, R.M. Twyman c,1, R. Fischer a,b,2
a
b
c
Institute for Molecular Biotechnology, Worringerweg 1, RWTH Aachen University, 52074 Aachen, Germany
Fraunhofer Institute for Molecular Biology and Applied Ecology IME, Forckenbeckstrae 6, 52074 Aachen, Germany
TRM Ltd, PO Box 463, York, United Kingdom
a r t i c l e
i n f o
Article history:
Received 3 March 2015
Received in revised form 15 April 2015
Accepted 22 April 2015
Available online xxxx
Keywords:
Plant molecular farming
Downstream processing
Filtration
Clarication
Process chromatography
Protein purication
Process economics
a b s t r a c t
Plants offer the tantalizing prospect of low-cost automated manufacturing processes for biopharmaceutical proteins, but several challenges must be addressed before such goals are realized and the most signicant hurdles are
found during downstream processing (DSP). In contrast to the standardized microbial and mammalian cell platforms embraced by the biopharmaceutical industry, there are many different plant-based expression systems
vying for attention, and those with the greatest potential to provide inexpensive biopharmaceuticals are also
the ones with the most signicant drawbacks in terms of DSP. This is because the most scalable plant systems
are based on the expression of intracellular proteins in whole plants. The plant tissue must therefore be disrupted
to extract the product, challenging the initial DSP steps with an unusually high load of both particulate and soluble contaminants. DSP platform technologies can accelerate and simplify process development, including centrifugation, ltration, occulation, and integrated methods that combine solidliquid separation, purication
and concentration, such as aqueous two-phase separation systems. Protein tags can also facilitate these DSP
steps, but they are difcult to transfer to a commercial environment and more generic, exible and scalable strategies to separate target and host cell proteins are preferable, such as membrane technologies and heat/pH precipitation. In this context, claried plant extracts behave similarly to the feed stream from microbes or
mammalian cells and the corresponding purication methods can be applied, as long as they are adapted for
plant-specic soluble contaminants such as the superabundant protein RuBisCO. Plant-derived pharmaceutical
proteins cannot yet compete directly with established platforms but they are beginning to penetrate niche markets that allow the benecial properties of plants to be exploited, such as the ability to produce biobetters with
tailored glycans, the ability to scale up production rapidly for emergency responses and the ability to produce
commodity recombinant proteins on an agricultural scale.
2015 Elsevier Inc. All rights reserved.
Contents
1.
2.
3.
Introduction . . . . . . . . . . . . . . . . . . .
Downstream processing challenges specic to plants
2.1.
Product release and process containment . .
2.2.
Process-related impurities . . . . . . . . .
Extraction and clarication . . . . . . . . . . . .
3.1.
Tissue homogenization . . . . . . . . . . .
3.2.
Product extraction from biomass . . . . . .
3.3.
Removal of particulate contaminants . . . .
3.4.
Improved clarication strategies . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
0
0
0
0
0
0
0
0
0
Abbreviations: AEC, anion exchange chromatography; AFC, afnity chromatography; ATPS, aqueous two phase systems; CEC, cation exchange chromatography; CHO, Chinese hamster
ovary; DoE, design of experiments; DSP, downstream processing; EBA, expanded bed adsorption; ELP, elastin-like polypeptide; GMP, good manufacturing practice; HCP, host cell protein;
HIC,hydrophobic interaction chromatography; IEC, ion exchange chromatography; ITC, inverse transition cycling;MMC,mixed-mode chromatography; RSM, response surface methodology;
RuBisCO, ribulose-1,5-bisphosphate carboxylase/oxygenase; SEC, size exclusion chromatography; UF/DF, ultraltration and dialtration.
Corresponding author at: Fraunhofer Institute for Molecular Biology and Applied Ecology IME, Germany. Tel.: +49 241 6085 13162; fax: +49 241 6085 10000.
E-mail addresses: johannes.buyel@rwth-aachen.de (J.F. Buyel), richard@twymanrm.com (R.M. Twyman), scher@molbiotech.rwth-aachen.de (R. Fischer).
1
Tel.: +44 1430 872694.
2
Tel.: +49 241 6085 11020.
http://dx.doi.org/10.1016/j.biotechadv.2015.04.010
0734-9750/ 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Buyel JF, et al, Extraction and downstream processing of plant-derived recombinant proteins, Biotechnol Adv (2015),
J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx
4.
Integrated methods . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.1.
Combined extraction and solidliquid separation . . . . . . . . . . .
4.2.
Aqueous two-phase separation . . . . . . . . . . . . . . . . . . .
4.3.
Purication using genetic fusion tags . . . . . . . . . . . . . . . . .
4.4.
Expanded bed adsorption . . . . . . . . . . . . . . . . . . . . . .
5.
Product purication . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1.
Chromatography . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2.
Downstream process development . . . . . . . . . . . . . . . . .
5.3.
Membrane-based purication . . . . . . . . . . . . . . . . . . . .
5.4.
Minimal processing . . . . . . . . . . . . . . . . . . . . . . . .
6.
Process costs and competitiveness compared to other platforms . . . . . . . .
6.1.
Improved expression, reduced DSP costs . . . . . . . . . . . . . . .
6.2.
Lessons learned from other expression platforms . . . . . . . . . . .
6.3.
Vertical farming . . . . . . . . . . . . . . . . . . . . . . . . . .
6.4.
Standardization and competing expression platforms . . . . . . . . .
7.
Information box on DoE and process synthesis: Development of a DSP strategy
8.
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction
Biopharmaceutical proteins are almost universally manufactured
using animal cells or microbes cultivated in fermenters, and an entire industry has evolved based on the standardization, optimization and regulation of these platforms (Twyman et al., 2005). More recently,
biopharmaceutical products have been manufactured in plants, with
several candidates now in late-stage clinical development and one already approved for human use (Fischer et al., 2013; Paul et al., 2013).
Plants offer numerous advantages during the upstream production
phase including the low cost of infrastructure and production, the
built-in safety features reecting the inability of plants to support the
replication of human pathogens, and the unparalleled scalability of agricultural production (Sabalza et al., 2014). Another key feature of plants
is the diversity of upstream production systems, reecting the use of different plant species, tissues/cells, cultivation formats and expression
strategies, all of which can affect product yields and post-translational
modications such as glycosylation (Arcalis et al., 2013; Khan et al.,
2012; Makhzoum et al., 2014).
The biopharmaceutical industry has consolidated around a narrow
collection of platforms based on microbes and animal cells, and newcomers challenging these established markets face a high entry barrier
because the gold-standard systems benet from decades of incremental
strain and process improvement, plus a tailored regulatory framework
(Anonymous, 2001). In contrast, plants can be considered as a disruptive innovation with footholds already established in the market for
niche products (Paul et al., 2013). Here, there is a much lower entry barrier for new production platforms and the regulatory framework is still
developing (Fischer et al., 2012). This has given rise to an unusual industry landscape in which the platform can be matched to the product, with
the unique attributes of different plant-based production systems being
chosen to complement the properties of the recombinant protein, rather than all products being manufactured in a narrow range of standardized platforms. The diverse selection of plant-based production hosts
and representative product candidates have recently been reviewed
(Kuo et al., 2013; Merlin et al., 2014; Paul and Ma, 2011; Stoger et al.,
2014; Wilken and Nikolov, 2012).
One drawback of the diverse upstream production platforms based
on plants (Maschke et al., 2015; Menkhaus et al., 2004) is that their existence creates a challenge when it comes to downstream processing
(DSP), in contrast to the established biomanufacturing industry where
DSP is largely product-focused and platform independent. Economical
downstream processing relies on the use of standardized unit operations that have been developed to dovetail with upstream production
using fermenters. The adaption of downstream processing to plant-
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
based systems has therefore been easiest when plant cell suspension
cultures are used for production, because the upstream processes are
analogous to those used with microbes and animal cells (Schillberg
et al., 2013; Wilson and Roberts, 2012). However, the major advantages
of upstream production using plants have been realized by the development of whole-plant transient expression systems, which are rapid and
scalable (Bendandi et al., 2010; Lico et al., 2008; Pogue et al., 2010), and
transgenic plants, which have a slower development cycle but allow
agricultural-scale production (Menkhaus et al., 2004; Rybicki et al.,
2012). Both these systems are game-changers in terms of DSP because
plants have unique DSP-relevant attributes that are not shared with
fermenter-based systems. Such challenges will need to be addressed because plant-derived biopharmaceuticals are likely to emerge as market
leaders in the next few decades, particularly in developing countries
where they offer not only increased access to medicines but also to
the production technology itself via socially responsible licensing (Ma
et al., 2013).
2. Downstream processing challenges specic to plants
2.1. Product release and process containment
DSP can be dened as the recovery and purication of specic products from a complex source such as a biological matrix, and in general
terms the early steps are tailored for the production platform and the
later steps for the product. The biopharmaceutical industry has evolved
around a small number of platforms with similar characteristics so the
early stages of DSP have also become standardized and generally involve
centrifugation and/or ltration steps to remove cells and debris, leaving
the product in solution to be separated by more rened chromatography
and ltration processes. Such standardization is possible because biopharmaceutical products are usually secreted by cells into the cell culture
broth, allowing capture without cell disruption. This strategy can also be
applied to plants and efforts have been made to develop plant cell
suspension cultures that secrete proteins into the medium, and
even whole-plant systems in which the roots or leaves secrete proteins
into hydroponic medium or in which proteins are removed from the
plants by applying a vacuum. However, most of the plant-derived
biopharmaceuticals in clinical development are retained within the
plant cell and must be extracted by disrupting plant tissues during the
early stages of DSP (Wilken and Nikolov, 2012). Another signicant
issue is that biopharmaceutical manufacturing platforms based on
whole plants need to be compatible with good manufacturing practice
(GMP), but unlike fermenter systems it is difcult to constrain the entire
production chain within a clean-room environment (Fischer et al., 2012).
In some cases extraction is necessary because the product is targeted

to intracellular compartments that increase stability or allow particular
forms of post-translational modication (Twyman et al., 2013). But
even products that pass through the secretory pathway may not necessarily be secreted into the medium, because plant cells are surrounded
by a cell wall that retains large molecules such as proteins in the extracellular space beneath it, a compartment known as the apoplast. The
ability of proteins to pass through the cell wall is dependent on the
properties of the protein (size and charge) and the plant species, cell
type and cultivation process, although this can be facilitated by the
vacuum extraction process mentioned above. Additionally, the space
available in different subcellular compartments can limit protein accumulation and therefore the overall yields (Takaiwa, 2013).
In simple terms, the role of upstream production is to maximize the

yield of the product and the role of downstream purication is to
maximize its quality and purity. In conventional fermenter systems
these phases work in concert, but one of the major challenges with
plants (particularly whole-plant systems) is that the upstream
and downstream phases often work against each other. The yield of
biopharmaceuticals produced in plants can be increased by judicious intracellular targeting but this reduces the efciency of DSP by introducing many soluble and insoluble contaminants into the initial feed
stream. The abundance and nature of these contaminants therefore determines which separation processes are necessary during DSP. This
works against the development of standardized platform processes,
even though some general purication concepts have been suggested
such as extraction at low pH (b 5.0) to achieve the precipitation of
most HCPs (Azzoni et al., 2002; Buyel and Fischer, 2014d).
2.2. Process-related impurities

3. Extraction and clarication
There are some cases in which the intracellular accumulation of biopharmaceutical proteins is advantageous per se, especially where unprocessed or part-processed plant tissue (e.g. fruit juice or our paste)
is used for the oral administration of vaccines or prophylactic antibodies. However, the vast majority of plant-derived biopharmaceuticals
are intended to be extracted and puried in the conventional manner,
meaning that DSP must be adapted specically to deal with individual
expression systems. The idiosyncratic properties of different plants include the water content of the tissue which is much higher in leaves
than seeds (Virdi and Depicker, 2013), the presence and abundance of
proteases in the apoplast or other compartments (or even proteases
fully secreted into the medium), and plant-specic particulates released
during extraction (including intrinsic bers and oils but also extrinsic
and adventitious agents such as soil particles). Other factors include
plant-specic soluble contaminants such as polyphenols and organic
acids. These increase the viscosity of the feed stream and can modify
the product or damage processing equipment (Barros et al., 2011).
The feed stream may also contain abundant plant-specic host cell proteins (HCPs) that can interfere with chromatography (e.g. ribulose-1,5bisphosphate carboxylase/oxygenase, RuBisCO), plant specic metabolites that may contaminate the nal product (e.g. nicotine in the case
of tobacco) and plant viruses, which do not infect humans but are nevertheless important because they may be immunogenic or may mask
the presence of adventitious animal viruses introduced during processing. Finally, several plant-based expression platforms have been developed in which target proteins are expressed as fusions with peptides
that promote extraction, such as elastin-like polypeptides (ELPs) that
can be reversibly precipitated by temperature cycling and hydrophobic
proteins such as oleosins and hydrophobins which partition the recombinant proteins to the oil fraction (Conley et al., 2011).
3.1. Tissue homogenization

The most suitable method for the extraction of a biopharmaceutical
molecule depends greatly on the way it has been expressed (Fabian and
Ju, 2011; Georgiev et al., 2009). For products that have been secreted,
e.g. into cell culture supernatant or hydroponic medium, no specialized
extraction operations are required because the target is already present,
albeit highly diluted, in an accessible aqueous solution (Borisjuk et al.,
1999; Drake et al., 2009; Komarnytsky et al., 2000). In contrast, a dry
or wet milling step is often carried out prior to the extraction of target
proteins from seed-based expression systems such as maize, rice and
barley (Azzoni et al., 2002; Bai and Glatz, 2003a; Farinas et al., 2005;
Paraman et al., 2010). Thereafter, liquid extraction from milled seed or
unprocessed leafy biomass is similar and can involve blade-based homogenizers and/or presses (Bals and Dale, 2011; Buyel and Fischer,
2014e; Hassan et al., 2008, 2014; T.H. Kim et al., 2013). Nonmechanical methods such as enzymatic lysis (Blanco-Pascual et al.,
2014), sonication and freezethaw cycles (Hassan et al., 2008; Wang
and Zhang, 2012) can also be used, but these are much more difcult
to scale up.
3.2. Product extraction from biomass
A buffer is typically added during extraction to facilitate target protein recovery e.g. by stabilizing the pH at a desirable value, providing
sufcient salinity to maintain product solubility (Azzoni et al., 2002)
and providing antioxidants to prevent unwanted protein modications
(Ali et al., 2012; Azzoni et al., 2002; Buyel and Fischer, 2014d; Hassan
et al., 2008) (Table 1). For leafy biomass, pH values of 7.08.0 generally
Table 1
Buffer components and additives frequently used during the extraction of proteins from plant tissues or cells.
Buffer or additive
Function
Reference
Ascorbic acid
BME (-mercaptoethanol)
Carbonate
Citrate
DTT (dithiothreitol)
EDTA (ethylenediaminetetraacetic acid)
EMPIGEN-BB
Glycine
Phosphate
PVPP (polyvinylpolypyrrolidone)
SDS (sodium dodecylsulfate)
Sodium bisulte
Sodium sulte
TRIS
TX-100 (Triton X-100)
Urea
Antioxidant
Reducing agent
Buffer
Buffer
Reducing agent
Chelating agent
Detergent
Buffer
Buffer
Polyphenol-binding
Detergent
Antioxidant
Antioxidant
Buffer
Detergent
Chaotropic agent
Ali et al. (2012)

Love et al. (2012)
McLean et al. (2012)
Buyel and Fischer (2014c) and Soares et al. (2011)
Ahmad et al. (2012) and Fabian and Ju (2011)
Love et al. (2012)
Buyel et al. (2012)
Buyel et al. (2014b)
Buyel et al. (2014b) and Love et al. (2012)
Ali et al. (2012) and Barros et al. (2011)
Fabian and Ju (2011) and Wu et al. (2014)
Buyel and Fischer (2014f)
Love et al. (2012)
Nykiforuk et al. (2006) and Wang et al. (2013)
Ahmad et al. (2012) and Love et al. (2012)
Fabian and Ju (2011)
achieve the highest rates of protein recovery (Barros et al., 2013; Buyel
and Fischer, 2014d; Fu et al., 2010; Hassan et al., 2008) and buffer to biomass ratios of 150 have been used successfully (Fig. 1A). However, a
pH of 4.05.0 may be preferable if the target molecule is stable under
these conditions because many HCPs and even some phenols precipitate or become excluded from the extract under these conditions, thus
facilitating purication (Azzoni et al., 2002; Buyel and Fischer, 2014e;
Lai and Chen, 2012; Robic et al., 2010; Woodard et al., 2009). Other
pre-conditioning strategies such as blanching or heat precipitation at
~ 6570 C can also be used to reduce the concentration of HCPs by
N90% early in the process if the target proteins are thermostable
(Buyel et al., 2014a).
Aqueous buffers are the most common additives because they are
likely to be compatible with protein folding and activity (Berg et al.,
2002), but ionic liquids (Bi et al., 2011; Wang and Zhang, 2012) and supercritical uids such as CO2 can also be used (Ota et al., 2009) and can
increase product recovery by 23.5-fold. Organic solvents such as phenol or hexane are preferred for non-protein targets (Machado et al.,
2010; McPartland et al., 2012) and membrane proteins (Teng and
Wang, 2012), if the extracted proteins are to be analyzed by mass spectrometry (Wu et al., 2014), or if the biomass needs to be de-fatted before further processing (Hayden et al., 2012).
Several attempts have been made to replace the disruption of leafy
biomass with inltration and centrifugation methods, i.e. the biomass
is inltrated with extraction buffer which is then withdrawn from the
leaves by centrifugation (Kingsbury and McDonald, 2014; Turpen,
1999). This greatly improves the initial product purity and reduces the
particle burden because the leaf tissue remains intact and HCPs are
not released, but the method is only suitable for proteins targeted to
the apoplast and the recovery is generally less than 60%.
3.3. Removal of particulate contaminants
Tissue disruption is a component of most extraction processes,
resulting in a signicant particle burden and contributing to the high
DSP costs for plants that have frequently been reported (Buyel and
Fischer, 2014e; Menkhaus et al., 2004; Nikolov and Woodard, 2004;
Wilken and Nikolov, 2012). Even though the particle burden in hydroponic media and cell culture supernatants is lower, these feed streams
still need to be claried before target protein purication in order to
protect sophisticated downstream chromatography equipment and
sensitive lters. Combinations of centrifugation and/or ltration are
used for clarication (Roush and Lu, 2008). Centrifuges are reusable
and allow continuous operation, but cross-product contamination is

possible and the equipment can also release lubricants into the feed
stream. The up-front investment and maintenance costs are also higher
for centrifuges than for lters, and centrifuges are more difcult to scale
up. The maintenance costs for lters are minimal because most processes include single-use modules that do not need to be cleaned or recycled
(O'Brien et al., 2012; Pegel et al., 2011). Filters are compatible with solid
loads of up to 10% [w/v] (Buyel and Fischer, 2014g) and can rapidly be
adapted to different feed stocks (Buyel and Fischer, 2014e) because
nominal retention ratings in the 0.120.0 m range are available from
different vendors. Furthermore, depth lters with active chemical
groups can be used for the removal of certain soluble contaminants,
e.g. anion exchange surfaces for the removal of host cell DNA and RNA
(Yigzaw et al., 2006) or pigments (Naik et al., 2012).
3.4. Improved clarication strategies
The clarication of plant extracts can be improved by occulation
(Buyel and Fischer, 2014f), which is the aggregation of dispersed particles induced by charged polymers or electrolytes thus increasing the average particle diameter and facilitating solidliquid separation by
centrifugation or ltration (Buyel and Fischer, 2014b; Gregory and
Barany, 2011). Flocculation can be enhanced by selecting the optimal
pH, conductivity, polymer type and concentration, and such empirical
testing has been shown to more than double the capacity of depth lters
(Buyel and Fischer, 2014c). A polymer concentration of 0.55.0 g L1 is
often sufcient, and the addition of occulants to the feed stream is generally cost-effective because the cost of many occulants is less than
10 kg1. Flocculants can also be used selectively to precipitate and
thereby purify target proteins (Holler et al., 2007; Menkhaus et al.,
2002). However, hold steps of 1530 min may be required for efcient
occulation, which means the process becomes discontinuous or at
best a semi-continuous and the cost of dedicated holding tanks must
be anticipated during process development (Buyel and Fischer, 2014c;
Yasarla and Ramarao, 2012). The removal of polymers must be validated
but their safety has already been demonstrated in several processes
based on mammalian cells (Kang et al., 2013) and by the use of GMPapproved occulants.
The capacity of depth lters can be improved even further if
cellulose-based lter aids are used in combination with occulants
(Buyel et al., 2014b; Gottschalk, 2009). Filter aids at a concentration of
120 g L1 build up a loose cake on the rst lter layer and prevent viscous cell debris from causing a blockage. In combination with
Fig. 1. Extraction conditions and clarication costs for plant-derived biopharmaceuticals. A. A broad range of buffer:biomass ratios has been tested by different laboratories with more than
75% of the authors reporting a ratio of 10:1 or less. B. Consumables costs during DSP have been reduced by the simplication of ltration cascades and the introduction of occulants and
lter aids to enhance lter capacity. Error bars indicate standard deviations (n 3).
occulants, lter aids can increase the depth lter capacity by 35-fold, in
some cases achieving a capacity of N1000 L m2 (Fig. 1B) (Buyel et al.,
2014b). Like occulants, lter aids are safe, biodegradable and inexpensive (b 10 kg1) thus offering a sustainable solution to reduce DSP
costs for plant-derived biopharmaceuticals within a continuous operation mode.
separating into three phases have been developed by incorporating organic components such as butanol (Duman and Kaya, 2013; Narayan
et al., 2008).
4. Integrated methods
The partitioning behavior of target proteins can also be modied by

genetic fusion to hydrophobins, which are amphiphilic 715-kDa proteins naturally produced by lamentous fungi (Hakanpaa et al., 2006).
This method has been used to direct a green uorescent protein/
hydrophobin fusion to the detergent-rich phase of an ATPS system, increasing the concentration ~ 3-fold and achieving 6090% recovery
(Reuter et al., 2014).
The separation of target proteins and HCPs into different phases can
also be achieved by fusing the target proteins with oleosins, which naturally associate with oil bodies present in seeds (Bhatla et al., 2010). The
target protein therefore partitions to the oil fraction and is easily separated from cell debris and the majority of water-soluble proteins by centrifugation (Nykiforuk et al., 2006) or decantation (Kapchie et al., 2011).
This approach has been developed into a large-scale process for the production of insulin expressed in safower seeds (Markley et al., 2006).
Oleosins can also be used to purify non-fusion-proteins through specic
non-covalent interactions, e.g. a protein A-oleosin fusion produced in
safower has been used to purify a monoclonal antibody transiently
expressed in Nicotiana benthamiana with 7075% recovery (McLean
et al., 2012). Similarly, protein L or protein G fusions with a cellulosebinding domain can be used for the purication of antibodies on cellulose supports (Hussack et al., 2010).
Protein interactions resulting in oligomer formation can be triggered
by fusing target proteins to the maize storage protein zein or its synthetic variants. Zein contains a highly repetitive proline-rich motif with an
amphipathic helix that can associate with other molecules containing
the same motif, resulting in the formation of a storage organelle called
a protein body (Geli et al., 1994). The unique density of these compartments (~1.25 g mL1) allows them to be separated from cell debris and
other proteins by density gradient centrifugation or even direct centrifugation (Torrent et al., 2009). However, the scalability of this method
remains to be demonstrated. Furthermore, the reducing conditions required for disaggregation may be incompatible with protein activity
and may promote aggregation.
Another aggregation-dependent purication and concentration
method involves the genetic fusion of target proteins to ELPs, comprising repeats of a VPGXG motif where X is any amino acid except proline
(Urry, 1988). ELPs reversibly form insoluble aggregates due to hydrophobic interactions when a certain transition temperature Tt (~37 C)
is exceeded in the presence of salt (~1.5 M NaCl), a process known as inverse transition cycling (ITC) (Urry, 1997). The temperature or salt concentration required for phase transition can be reduced if the number of
VPGXG repeats is increased (Conley et al., 2009), allowing the purication method to be used with heat-sensitive proteins (Bischof and He,
2005). The insoluble aggregates can be separated from more than 95%
of soluble HCPs by centrifugation (Meyer and Chilkoti, 1999) or membrane ltration (Phan and Conrad, 2011) without chromatography.
ITC has been used to purify spider silk proteins (Weichert et al., 2014),
hemagglutinin (Phan et al., 2014) and lectins (Tian and Sun, 2011). A
more sophisticated single-step method has been described in which
ITC is combined with the removal of the ELP-tag from the target protein
by an ELP-protease fusion (Lan et al., 2011).
The use of hydrophobins, oleosins, zeins and ELP fusions for purication may not be suitable for process-scale biopharmaceutical
manufacturing processes due to intellectual property issues (Conley
et al., 2011) and regulatory concerns based on the potential immunogenicity of the fusion tag (Fischer et al., 2012). Even following protease
cleavage, residual amino acids may remain attached to the target protein thus altering its structure (Li, 2011).
Integrated methods for the processing of plant extracts include juice

extraction, aqueous two phase separation (ATPS), expanded bed adsorption (EBA) chromatography and various strategies based on the expression of fusion proteins (Bai and Glatz, 2003b; Gu, 2014). These
methods combine two or more of the following operations: extraction,
solidliquid separation, purication and concentration. The benets of
integrated methods include the need for fewer process steps, lower
up-front and consumables costs, faster processing with less opportunity
for target modication or degradation, and sometimes a continuous
mode of operation.
4.1. Combined extraction and solidliquid separation
Juice extraction can replace the homogenization and initial ltration
of leafy biomass. The biomass is fed into a juicer or screw press containing a conical rotating cylinder with a progressively narrowing bore, thus
pressing plant sap out of the biomass by constantly increasing the pressure (Yan et al., 2014). This method has been used to recover small molecules from plant biomass by applying a counter-current of organic
buffer (Rao et al., 2008). More recently, the method has been adapted
to extract antibodies and other proteins from leafy biomass without
using buffers (Buyel and Fischer, 2014g). Although this reduced the
recovery by 4065%, the protein concentrations increased 2.8-fold and
the process volume was reduced 5-fold which is desirable for DSP
(Lightfoot and Cockrem, 2013) and reduces costs (Bals and Dale,
2011). This method is also suitable for continuous operation and is compatible with occulation and the heat precipitation of HCPs (Buyel and
Fischer, 2014g).
4.2. Aqueous two-phase separation
ATPS occurs when two polymers or a polymer and a salt are added to
an aqueous solution above certain critical concentrations (Hatti-Kaul,
2001). Dispersed particles settle in the lower phase due to gravity,
whereas dissolved molecules such as proteins have unique partitioning
coefcients for the upper and lower phases allowing selective purication. ATPS methods integrate solidliquid separation and purication,
and they can be applied in either batch (Gu, 2014) or continuous
mode using liquidliquid extraction columns (Biazus et al., 2007) or
similar devices (Vazquez-Villegas et al., 2013). The potential application
of ATPS for the purication of plant-derived proteins has recently been
reviewed, and the author concluded that the advantages of integrated
separation, purication and concentration outweigh the high costs of
polymers, salts and waste disposal (Gu, 2014). The partitioning of target
proteins and HCPs can be inuenced by the choice and concentration of
polymers, the buffer pH and salinity (Gu, 2014; Hatti-Kaul, 2001;
Oelmeier et al., 2011). However, the presence of diverse proteins as
well as cell debris can affect phase separation and thus hinder the application of a priori model-based predictions. Instead, high throughput
screening combined with statistical experimental designs (design of experiments, DoE) is the method of choice for the optimization of ATPS
(Oelmeier et al., 2011) and this approach has already been applied to tobacco extracts (Balasubramaniam et al., 2003). Combinations of polyethylene glycol and sulfate or phosphate are often used for ATPS
(Balasubramaniam et al., 2003; Vazquez-Villegas et al., 2013) but citrate
(Soares et al., 2011) and even detergents can also be used to promote
phase formation (Joensuu et al., 2010). In addition to ATPS, systems
4.3. Purication using genetic fusion tags
4.4. Expanded bed adsorption

EBA chromatography is an integrated method that combines solid
liquid separation with target protein recovery (Schugerl and Hubbuch,
2005). A sample is loaded to the bottom of an EBA column while
the top plug is retracted from the packed or gravity-settled bed, causing
uidization of the resin, i.e. the beads are suspended within the geometric volume of the column due to the application of the feed stream
(Hubbuch et al., 2005). In contrast to packed-bed chromatography,
where the resin beads contact each other, the beads of an EBA resin
may be separated by several millimeters so that the system is much
more tolerant of high particle loads in the feed stream. This reduces
the need for expensive upstream clarication, e.g. a centrifuge operated
with a Q/ ratio of 4.56 107 m s1 was sufcient for the clarication
of canola seed extracts prior to EBA chromatography (Bai and
Glatz, 2003a). For packed-bed chromatography, Q/ ratios of
b5.0 108 m s1 are required to remove N80% of dispersed solids
(Lander et al., 2005), where Q is the volumetric feed ow [m3 s 1]
and (sigma factor) is the product of the centrifuge area [m2] and the
settling relative to gravity [] (g-factor of the centrifuge). Although
EBA chromatography can tolerate high volumetric particle loads,
the size and charge of these particles should be kept low to reduce the
likelihood of adsorption, which can result in a collapse of the uidized
bed particularly in the case of ion-exchange resins (Hubbuch et al.,
2006). EBA chromatography has been used to purify recombinant
-glucuronidase from canola seed extracts (Bai and Glatz, 2003b) and
patatin from potato (Straetkvern et al., 1999). However, a direct cost
comparison for the purication of a monoclonal antibody indicated
that a conventional packed-bed chromatography was more cost effective (Menkhaus and Glatz, 2005). An additional drawback of EBA is
the limited number of resins that is currently available from commercial
vendors.
5. Product purication
5.1. Chromatography
Although there is a limited choice of EBA resins, there are many different conventional chromatography resins using different GMP-ready
base matrices (e.g. Sepharose, cellulose and polymethacrylate). These
matrices are paired with ligands suitable for hydrophobic interaction
chromatography (HIC), mixed-mode chromatography (MMC), sizeexclusion chromatography (SEC, also called gel ltration), afnity chromatography (AFC) and ion-exchange chromatography (IEC), the latter
including anion exchange chromatography (AEC) and cation exchange
chromatography (CEC) (Anonymous, 2010). Differences in the selectivity of these ligands for target proteins and contaminating HCPs are used
to purify the target, typically using multiple operations with orthogonal
separation mechanisms (Nfor et al., 2011). The highest selectivity is
generally achieved by AFC in bind-and-elute mode so this is often
used as the initial capture step, e.g. protein A binding to monoclonal
antibodies (The CMC Biotech Working Group, 2009). The remaining
formats can also be operated in bind-and-elute mode, but further options during process development include weak partitioning mode to
remove tightly-bound impurities from a target molecule (Kelley et al.,
2008) and ow-through mode which is used to bind contaminants
such as HCPs while the target protein remains in the liquid phase
(Anonymous, 2013). Several salt-tolerant ion exchange ligands have
been developed that allow protein binding in buffers with conductivities 15 mS cm 1, e.g. HyperCel STAR AX (Pall, Port Washington,
NY, USA) and Sartobind STIC (Sartorius, Gttingen, Germany). These
resins allow the direct capture of target proteins from cell culture supernatants or tobacco leaf extracts (Buyel and Fischer, 2014d; Champagne
et al., 2013). Membrane chromatography can be used to replace conventional packed-bed columns because membrane cassettes have a
lower backpressure and therefore allow higher ow rates, which is
advantageous when processing plant extracts with target protein concentrations lower than 50 mg L1 (our unpublished data) (Menkhaus
and Roseland, 2008).
Continuous countercurrent chromatography has a signicant potential for the purication of plant-derived biopharmaceuticals because
continuous processing reduces the column sizes necessary for the
separation of plant extracts and therefore the associated time and
costs (Warikoo et al., 2012). This method has recently been used to isolate proteins with anti-cancer activity from bitter melon (Momordica
charantia) achieving 8694% purity (Li et al., 2012).
5.2. Downstream process development
The diverse combinations of plant expression systems and target
proteins that have been reported have resulted in an equally diverse
selection of purication processes that are difcult to transfer to
other products and that in many cases include operations that are not
particularly scalable, e.g. SEC (Abe et al., 2014) or density gradient
centrifugation for virus-like particles (Love et al., 2012). A lectin has
been puried from seeds in a ve-stage process including ammonium
sulfate precipitation, achieving 95% purity but only 20% recovery
(Devi et al., 2011), whereas a single-step purication of phosphinothricin
N-acetyltransferase from cotton seed achieved the same purity and 30%
recovery but involved several rounds of extract clarication (Wang et al.,
2013). Even afnity-based antibody purication has resulted in recovery
rates as low as 36% (Platis et al., 2008). Furthermore, custom-made afnity resins may be less stable than commercial products (Esteve-Turrillas
et al., 2011).
In contrast to process development based on trial-and-error, the
structured identication of HCPs and their subsequent characterization,
e.g. in terms of abundance, molecular mass, pI and hydrophobicity, can
facilitate the rational design of DSP operations and may allow the a
priori identication of separation conditions that are likely to achieve
purication (Aguilar et al., 2009; Buyel et al., 2013b; Lim et al., 2012;
Murad et al., 2011; Nfor et al., 2012) (Fig. 2). This approach may be
complemented by generic empirical and experience-based methods
(Buyel and Fischer, 2014d).
Although the principle of chromatography is similar for all biopharmaceutical feed streams, the challenge with plant extracts is to separate
an often scarce target protein from several highly-abundant HCPs, such
as RuBisCO (Buyel and Fischer, 2014d). This protein can be depleted by
precipitation with polyethylene glycol (Alam et al., 2013) (our unpublished data), protamine sulfate (Y.J. Kim et al., 2013) or phytate
(Boyhan and Daniell, 2011) as well as heat precipitation, as discussed
above (Buyel et al., 2014a). If the removal of RuBisCO is not possible
and it co-elutes with the target protein, separation may be facilitated
by the use of purication tags such as hexahistidine (Buyel et al.,
2012) or soybean agglutinin (Tremblay et al., 2011). These have been
used to purify proteins after transient expression in N. benthamiana
and a review on the most frequently used tags has recently been published (Pina et al., 2014). However, as mentioned above, genetic fusion
tags may increase the intellectual property and regulatory burden of
commercial biopharmaceuticals (Fischer et al., 2012). As an alternative,
directed protein engineering can be applied to achieve protein purication using established afnity chromatography methods. This has been
demonstrated for an IgA protein, which was engineered with a protein L
binding domain to facilitate afnity capture (Boes et al., 2011).
5.3. Membrane-based purication
Protein engineering can be labor-intensive and time-consuming and
is not always feasible, so more generic purication strategies are advantageous. Ultraltration and dialtration (UF/DF) can be used for protein
purication as a more scalable version of SEC if the molecular size of the
target protein and contaminating HCPs differ by a factor of ~ 2. UF/DF
was successfully applied to purify recombinant bovine lysozyme from
Fig. 2. Multi-dimensional plots showing the properties of tobacco HCPs. These plots can be used to identify windows of opportunity for the purication of target proteins (hatched regions).
A. In terms of molecular mass, a window of opportunity exists for HCPs N 100 kDa if the molecular mass, pI and e.g. elution salt concentration on sulfopropyl-Sepharose (SP) are considered.
B. This window is even larger if the actual abundance of HCPs is taken into account. C. If the monomeric sizes of HCPs are replaced by native oligomers, purication becomes more
challenging. HCPs may adopt an oligomeric state between monomers and their native biological assembly. D. If the abundance and oligomerization of HCPs are considered, promising purication conditions deviate markedly from those in the nave plot (A). Spot sizes in B and D are proportional to the abundance of the corresponding HCP in tobacco leaf extracts.
sugarcane extracts using 100-kDa and 3-kDa membranes (Barros et al.,

2013). The oligomeric state of HCPs must be characterized to select
membranes appropriate for the optimal UF/DF separation window
(Fig. 2). For feed streams with conductivities of 2050 mS cm 1 at
pH 8.0, we have found that RuBisCO is retained by a 100-kDa membrane
but not by a 300-kDa membrane, indicating an intermediate state of
oligomerization between monomers (~14.5 kDa and ~52.7 kDa for the
small and large subunits, respectively) and the biologically-active 8:8
oligomer (~537.6 kDa) (our unpublished data). Membranes have several additional benets, i.e. their high volumetric throughput which can
easily be scaled up to greater areas without the risk of bed instability
as observed for packed-bed chromatography (Orr et al., 2013), the
high extract concentration that can be achieved with a membrane cascade, resulting in lower costs (Lightfoot, 2005), and the availability of
cost-effective single-use membranes from several suppliers. One drawback of UF/DF is the limited availability of membranes with different
size ratings, especially in the 50500 kDa range that would allow the
precise and directed selection of membranes suitable for specic target
proteins.
5.4. Minimal processing
Purication can also be simplied if partially or minimally processed
plant material fullls the safety requirements for the intended application. Examples of such product candidates include oral vaccines/
prophylactic antibodies, and topical microbicides (Daniell et al., 2009;
Ramessar et al., 2008). Although such products have yet to be approved
as pharmaceuticals for human use, minimal processing is being
explored in the context of veterinary products that have a lower regulatory burden (Jacob et al., 2013; Kolotilin et al., 2014). The production of
technical enzymes can also benet from minimal processing strategies
(Biesgen et al., 2002).
6. Process costs and competitiveness compared to other platforms
6.1. Improved expression, reduced DSP costs
One of the major benets of plant-based production platforms is the
low cost of upstream production (Buyel and Fischer, 2012, 2014g; Tuse
et al., 2014; Walwyn et al., 2015). The latest generation of sophisticated
expression vectors has increased the yields achieved in leafy crops up to
5.0 g of recombinant protein per kg of biomass (Gleba et al., 2014;
Sainsbury and Lomonossoff, 2008) thus making plants even more competitive with mammalian systems. These vectors are often used for transient expression mediated by Agrobacterium tumefaciens, viral vectors
or a combination of the two systems (Gleba et al., 2005; Jones et al.,
2009). Virus-derived sequences, e.g. promoters and untranslated regions, are also frequently used to achieve high levels of recombinant
protein expression (Buyel et al., 2013a; Lacorte et al., 2010; Regnard
et al., 2010; Sainsbury and Lomonossoff, 2008). DSP costs of up to 80%
of total production costs are often quoted (Menkhaus et al., 2004;
Wilken and Nikolov, 2012), not including quality control and release
testing, and one might anticipate this proportion increasing to 90%
as yields continue to improve. However, these higher titers are more
likely to reduce DSP expenditure because the consumables costs are
driven by extract volume, and higher yields mean that fewer lters
and less chromatography resin are needed to achieve the purication of

the same amount of product (Buyel and Fischer, 2014g). These measures will probably maintain the DSP costs at the 80% level, but platform
standardization, the incorporation of recent improvements in clarication (occulation, lter aids and integrated methods) and the inclusion
of UF/DF for product purication will reduce these costs even further,
making plants more competitive than ever. For example, single-use
depth lters accounted, in our hands, for consumables costs of ~4000
per batch when processing 200 kg of tobacco leaf biomass. Incorporating occulation and lter aids during clarication increased the lter capacity (Buyel et al., 2014b) and reduced the associated costs to b200
per batch (our unpublished data).
6.2. Lessons learned from other expression platforms
Despite these positive developments, there is an urgent need to keep
pace with new developments in pharmaceutical manufacturing such
as complete continuous processing (Jungbauer, 2013), the implementation of single-use technologies and standardization (Shukla and
Gottschalk, 2013), optimal process design (Liu et al., 2004, 2013), process analytical technology (Glassey et al., 2011; Rathore et al., 2010;
Read et al., 2010a, 2010b), quality by design (Rathore, 2009) and optimal capacity usage for multi-product facilities (Iribarren et al., 2004).
In this context, developers working on plant-based production processes must not only learn from the history of cell culture processes, such as
the problems that arise from ever increasing product titers (Yang et al.,
2014), but also from solutions that have been developed in the food processing industry, which must also comply with strict safety regulations
(Abdel-Rahman et al., 2011; Manfreda and De Cesare, 2014; Randell and
Whitehead, 1997) and for which efcient and automated quality
control mechanisms have been implemented (Caldwell et al., 2009;
McMeekin et al., 2006).
antibodies produced in mammalian cells (The CMC Biotech Working

Group, 2009). Therefore, plant-based processes can drive process innovation in the future. Once standardization is achieved, more effective
platform DSP operations can be developed which will reduce the duration and cost of process development and also the upfront investment
and operational costs (Haimowitz and Warran, 2007; Tpfer et al.,
2000), especially when generic quality assurance assays can be introduced (our unpublished data).
Despite the cost-savings associated with standardization, expression
levels of 0.5 g kg 1 may be required for plant-based systems to
compete with antibodies derived from mammalian cells (Buyel and
Fischer, 2012, 2014g; Buyel et al., 2014b), where titers of 5.0 g L1
can readily be achieved (Kelley, 2009; Shukla and Thommes, 2010) reducing the cost of goods to 301500 g1 (Feder, 2012; Kelley, 2009;
Rathore et al., 2004). Therefore, plants cannot currently compete directly with mammalian cell cultures but they represent an economical alternative for proteins that cannot be expressed efciently in mammalian
cells, that are functionally inferior when produced in mammalian cells,
e.g. glucocerebrosidase (Pastores et al., 2014; Shaaltiel et al., 2007), or
that are needed more rapidly and/or in greater quantities than can be
achieved using mammalian cells. Therefore, it is likely that the current
generation of plant-derived pharmaceuticals will focus on specialized
niche products such as enzymes that function better when produced
in plants (Hayden et al., 2014), vaccines and other emergency medicinal
products required rapidly, such as antibodies and vaccines against Ebola
(Skarjinskaia et al., 2013), seasonal inuenza (Shoji et al., 2012), or
antibiotic-associated diarrhea (Nandi et al., 2005; Wolfson, 2013) and
commodity products such as antibodies produced on the potential ton
scale for the preparation of topical microbicides (Ramessar et al., 2008).
7. Information box on DoE and process synthesis: Development of a

DSP strategy
6.3. Vertical farming

A concept uniquely applicable to plants, which originated in the
1960s but has re-emerged more recently (Despommier, 2009;
Marchetti, 1979) is the fully-automated indoor cultivation of plants on
stacked trays, a process known as vertical farming. The advantages of
this approach include reduced environmental pollution from soil, fertilizers, improved logistics and better-controlled growth conditions which
are relevant to molecular farming (Fischer et al., 2012). Along with the
rst commercial vertical farming units (Despommier, 2013), a fully automatic growth, inltration, incubation and harvest facility for transient
protein expression in N. benthamiana has recently been designed and
constructed (Wirz et al., 2012). In the future, such facilities should also
integrate automated product extraction and subsequent DSP operations. For this purpose, the membrane technologies mentioned above
can be used as an initial and generic purication and concentration
tool, removing the majority of plant HCPs (our unpublished data).
6.4. Standardization and competing expression platforms
The standardization of plant-based biopharmaceutical manufacturing including DSP should also be a major goal in plant biotechnology
to close the current gap between the highly-industrialized production
of recombinant proteins in mammalian cell cultures and the small number of pilot-scale processes using intact plants. This standardization can
be achieved by the consolidation of current expression systems, paying
special attention to the most promising, e.g. tobacco and cereals (Melnik
and Stoger, 2013), to a stage where they can introduce specic glycosylation patterns (Strasser, 2013) or produce low levels of alkaloids (Buyel
et al., 2015). There is currently no standard plant-based process and
this should allow the rapid incorporation of improved methods such
as membrane chromatography and ATPS into manufacturing processes,
in contrast to the clear path that has been laid out for monoclonal
A major challenge in the development of manufacturing processes

for plant-derived biopharmaceuticals is the rational selection of optimal
combinations of production host, expression strategy (e.g. promoter and
subcellular localization), and effective DSP. Statistical experimental designs (DoE) are useful in this context as we and others have repeatedly
demonstrated (Ahmad et al., 2012; Buyel and Fischer, 2012, 2014a,
2014e; Buyel et al., 2013a, 2014a, 2014b; Dhaneshwar et al., 2014;
Farinas et al., 2005; Fu et al., 2010; Gecchele et al., 2014; Wang and
Zhang, 2012). The core idea of DoE is to alter more than one parameter
at a time in a pre-dened series of experiments in order to reduce the
number of experiments (and associated costs) required to identify a
process optimum (Czitrom, 1999), to obtain information about process
parameters that interact with each other, allowing the identication of
global rather than local optima (Montgomery, 2007), and to provide
predictive empirical models for production processes that are too complex for a mechanistic description (Myers et al., 2009). Importantly,
these models also support the process documentation that is requested
by the regulatory authorities (FDA, 2009). The sequence of a DoE approach is an initial full or fractional factorial screening design to identify
the most relevant process parameters (factors, e.g. pH, temperature) affecting the outcome (response, e.g. purity, activity) of an experiment
(Anderson and Kraber, 1999; Montgomery, 2007) followed by one or
several rounds of renement, e.g. by response surface methodology
(RSM). This is used to precisely determine the absolute quantitative effects of the relevant factors and to identify the most desirable windows
of operation (Myers et al., 2009). Desirability can be dened as process
robustness, i.e. the insensitivity of the response towards changes in the
factors (a plateau in the RSM), or extreme responses such as minimal
turbidity after clarication or maximum yield after chromatography. Finally, the results of individual DoE runs can be combined by so-called
expert systems that predict combinations of process steps yielding an
overall optimal result (Iribarren et al., 2004).
8. Conclusions
Plant-based expression systems are now entering the commercial
arena for biopharmaceutical protein manufacturing and are beginning
to compete directly with conventional fermenter-based platforms
using microbes or mammalian cells. Some plants are starting to emerge
from the vast pool of potential species as standard expression hosts. This
declining complexity, together with the standardization of the initial
DSP steps, has already resulted in relevant cost savings during product
recovery operations. Novel technical solutions (e.g. miniaturized laboratory equipment and continuous DSP operations) and statistical designs
(e.g. DoE software) can be used to exploit and optimize integrated purication operations such as ATPS, further reducing the production costs
for plant-derived biopharmaceuticals. In this context, plant biotechnology can take advantage of the lessons learned from fermenter-based
systems and process engineers can begin to implement rational DSP designs focusing on mechanistic separation models.
Acknowledgments
This work was funded in part by the European Research Council
Advanced Grant Future-Pharma, proposal number 269110, the FhG
Internal Programs under Grant No. Attract 125-600164 and the
Fraunhofer-Zukunftsstiftung (Fraunhofer Future Foundation). The
authors have no conict of interest to declare.
References
Abdel-Rahman A, Anyangwe N, Carlacci L, Casper S, Danam RP, Enongene E, et al. The
safety and regulation of natural products used as foods and food ingredients. Toxicol
Sci 2011;123:33348.
Abe M, Yuki Y, Kurokawa S, Mejima M, Kuroda M, Park EJ, et al. A rice-based soluble form
of a murine TNF-specic llama variable domain of heavy-chain antibody suppresses
collagen-induced arthritis in mice. J Biotechnol 2014;175:4552.
Aguilar O, Glatz CE, Rito-Palomares M. Characterization of green-tissue protein extract
from alfalfa (Medicago sativa) exploiting a 3-D technique. J Sep Sci 2009;32:322331.
Ahmad MN, Liew SL, Yarmo MA, Said M. Optimization of protease extraction from horse
mango (Mangifera foetida Lour) kernels by a response surface methodology. Biosci
Biotechnol Biochem 2012;76:143844.
Alam I, Sharmin SA, Kim KH, Kim YG, Lee JJ, Lee BH. An improved plant leaf protein extraction method for high resolution two-dimensional polyacrylamide gel electrophoresis and comparative proteomics. Biotech Histochem 2013;88:6175.
Ali M, Homann T, Kreisel J, Khalil M, Puhlmann R, Kruse HP, et al. Characterization and
modeling of the interactions between coffee storage proteins and phenolic compounds. J Agric Food Chem 2012;60:116018.
Anderson MJ, Kraber SL. Keys to successful designed experiments. ASQ The global voice
of quality, 6; 1999. p. 6.
Anonymous. Q7A good manufacturing practice guidance for active pharmaceutical
ingredients. In: Services USDoHaH, editor. Rockville, MD, USA: Food and Drug
Administration; 2001. p. 58.
Anonymous. Strategies for protein purication. Uppsala: GE Healthcare; 2010.
Anonymous. Multimodal chromatography. Uppsala: Ge Healthcare; 2013.
Arcalis E, Stadlmann J, Rademacher T, Marcel S, Sack M, Altmann F, et al. Plant species and
organ inuence the structure and subcellular localization of recombinant glycoproteins. Plant Mol Biol 2013;83:10517.
Azzoni AR, Kusnadi AR, Miranda EA, Nikolov ZL. Recombinant aprotinin produced in
transgenic corn seed: extraction and purication studies. Biotechnol Bioeng 2002;
80:26876.
Bai Y, Glatz CE. Bioprocess considerations for expanded-bed chromatography of crude canola extract: sample preparation and adsorbent reuse. Biotechnol Bioeng 2003a;81:
77582.
Bai Y, Glatz CE. Capture of a recombinant protein from unclaried canola extract using
streamline expanded bed anion exchange. Biotechnol Bioeng 2003b;81:85564.
Balasubramaniam D, Wilkinson C, Van Cott K, Zhang CM. Tobacco protein separation by
aqueous two-phase extraction. J Chromatogr A 2003;989:11929.
Bals B, Dale BE. Economic comparison of multiple techniques for recovering leaf protein in
biomass processing. Biotechnol Bioeng 2011;108:5307.
Barros GOF, Woodard SL, Nikolov ZL. Phenolics removal from Transgenic Lemna minor extracts expressing mAb and impact on mAb production cost. Biotechnol Prog 2011;27:
4108.
Barros GOF, Ballen MAT, Woodard SL, Wilken LR, White SG, Damaj MB, et al. Recovery of
bovine lysozyme from transgenic sugarcane stalks: extraction, membrane ltration,
and purication. Bioprocess Biosyst Eng 2013;36:140716.
Bendandi M, Marillonnet S, Kandzia R, Thieme F, Nickstadt A, Herz S, et al. Rapid, highyield production in plants of individualized idiotype vaccines for non-Hodgkin's lymphoma. Ann Oncol 2010;21:24207.
Berg JM, Tymoczko JL, Stryer L. Tertiary Structure: water-soluble proteins fold into compact structures with nonpolar cores. Biochemistry-Us. 5 ed. New York: W H Freeman;
2002.
Bhatla SC, Kaushik V, Yadav MK. Use of oil bodies and oleosins in recombinant protein
production and other biotechnological applications. Biotechnol Adv 2010;28:
293300.
Bi W, Tian M, Row KH. Ultrasonication-assisted extraction and preconcentration of
medicinal products from herb by ionic liquids. Talanta 2011;85:7016.
Biazus JPM, Santana JCC, Souza RR, Jordao E, Tambourgi EB. Continuous extraction
of alpha- and beta-amylases from Zea mays malt in a PEG4000/CaCl2 ATPS.
J Chromatogr B 2007;858:22733.
Biesgen C, Hillebrand H, Herbers K. Technical enzymes produced in transgenic plants.
Phytochem Rev 2002;1:7985.
Bischof JC, He X. Thermal stability of proteins. Ann N Y Acad Sci 2005;1066:1233.
Blanco-Pascual N, Aleman A, Gomez-Guillen MC, Montero MP. Enzyme-assisted extraction of kappa/iota-hybrid carrageenan from Mastocarpus stellatus for obtaining bioactive ingredients and their application for edible active lm development. Food Funct
2014;5:31929.
Boes A, Spiegel H, Delbruck H, Fischer R, Schillberg S, Sack M. Afnity purication of a
framework 1 engineered mouse/human chimeric IgA2 antibody From tobacco.
Biotechnol Bioeng 2011;108:280414.
Borisjuk NV, Borisjuk LG, Logendra S, Petersen F, Gleba Y, Raskin I. Production of recombinant proteins in plant root exudates. Nat Biotechnol 1999;17:4669.
Boyhan D, Daniell H. Low-cost production of proinsulin in tobacco and lettuce
chloroplasts for injectable or oral delivery of functional insulin and C-peptide. Plant
Biotechnol J 2011;9:58598.
Buyel JF, Fischer R. Predictive models for transient protein expression in tobacco
(Nicotiana tabacum L.) can optimize process time, yield, and downstream costs.
Buyel JF, Fischer R. Characterization of complex systems using the design of experiments
approach: transient protein expression in tobacco as a case study. J Vis Exp 2014a;1:
e51216.
Buyel JF, Fischer R. Downstream processing of biopharmaceutical proteins produced in
plants: the pros and cons of occulants. Bioengineered 2014b;5:13842.
Buyel JF, Fischer R. Flocculation increases the efcacy of depth ltration during the downstream processing of recombinant pharmaceutical proteins produced in tobacco.
Plant Biotechnol J 2014c;12:24052.
Buyel JF, Fischer R. Generic chromatography-based purication strategies accelerate the
development of downstream processes for biopharmaceutical proteins produced in
plants. Biotechnol J 2014d;9:56677.
Buyel JF, Fischer R. Scale-down models to optimize a lter train for the downstream purication of recombinant pharmaceutical proteins produced in tobacco leaves.
Biotechnol J 2014e;9:41525.
Buyel JF, Fischer R. Synthetic polymers are more effective than natural occulants for the
clarication of tobacco leaf extracts. J Biotechnol 2014f;195:3742.
Buyel JF, Fischer R. A juice extractor can simplify the downstream processing of plantderived biopharmaceutical proteins compared to blade-based homogenizers. Process
Biochem 2014g;50:85966.
Buyel JF, Bautista JA, Fischer R, Yusibov VM. Extraction, purication and characterization
of the plant-produced HPV16 subunit vaccine candidate E7 GGG. J Chromatogr B
2012;880:1926.
Buyel JF, Kaever T, Buyel JJ, Fischer R. Predictive models for the accumulation of a uorescent marker protein in tobacco leaves according to the promoter/5'UTR combination.
Biotechnol Bioeng 2013a;110:47182.
Buyel JF, Woo JA, Cramer SM, Fischer R. The use of quantitative structure-activity relationship models to develop optimized processes for the removal of tobacco host cell proteins during biopharmaceutical production. J Chromatogr A 2013b;1322:1828.
Buyel JF, Gruchow HM, Boes A, Fischer R. Rational design of a host cell protein heat precipitation step simplies the subsequent purication of recombinant proteins from
tobacco. Biochem Eng J 2014a;88:16270.
Buyel JF, Opdensteinen P, Fischer R. Cellulose-based lter aids increase the capacity of
depth lters during the downstream processing of plant-derived biopharmaceutical
proteins. Biotechnol J 2014b;10:58491.
Buyel JF, Buyel JJ, Haase C, Fischer R. The impact of Pseudomonas syringae type III effectors on transient protein expression in tobacco. Plant Biol 2015;17:48492.
Caldwell D, Davis S, Moreno Masey R, Gray J. Automation in food processing. In: Nof
SY, editor. Springer Handbook of Automation. Berlin Heidelberg: Springer; 2009.
p. 104159.
Champagne J, Balluet G, Gantier R, Toueille M. Salt tolerant anion exchange chromatography for direct capture of an acidic protein from CHO cell culture. Protein Expr Purif
2013;89:11723.
Conley AJ, Joensuu JJ, Jevnikar AM, Menassa R, Brandle JE. Optimization of elastin-like
polypeptide fusions for expression and purication of recombinant proteins in plants.
Conley AJ, Joensuu JJ, Richman A, Menassa R. Protein body-inducing fusions for high-level
production and purication of recombinant proteins in plants. Plant Biotechnol J
2011;9:41933.
Czitrom V. One-factor-at-a-time versus designed experiments. Am Stat 1999;53:6.
Daniell H, Singh ND, Mason H, Streateld SJ. Plant-made vaccine antigens and
biopharmaceuticals. Trends Plant Sci 2009;14:66979.
Despommier D. The rise of vertical farms. Sci Am 2009;301:807.
Despommier D. Farming up the city: the rise of urban vertical farms. Trends Biotechnol
2013;31:3889.
Devi SK, Singh SS, Singh SJ, Rully H, Singh LR. Purication and characterization of a
magnesium ion requiring N-Acetyl-D-glucosamine specic lectin from seeds of
Quercus ilex L. Biosci Biotechnol Biochem 2011;75:17527.
10
Dhaneshwar AD, Chaurasiya RS, Hebbar HU. Process optimization for reverse micellar extraction of stem bromelain with a focus on back extraction. Biotechnol Prog 2014;30:
84555.
Drake PMW, Barbi T, Sexton A, McGowan E, Stadlmann J, Navarre C, et al. Development of
rhizosecretion as a production system for recombinant proteins from hydroponic
cultivated tobacco. FASEB J 2009;23:35819.
Duman YA, Kaya E. Three-Phase partitioning as a rapid and easy method for the purication and recovery of catalase from sweet potato tubers (Solanum tuberosum). Appl
Biochem Biotechnol 2013;170:111926.
Esteve-Turrillas FA, Mercader JV, Agullo C, Abad-Somovilla A, Abad-Fuentes A. Development of immunoafnity columns for pyraclostrobin extraction from fruit juices and
analysis by liquid chromatography with UV detection. J Chromatogr A 2011;1218:
49029.
Fabian C, Ju YH. A review on rice bran protein: its properties and extraction methods. Crit
Rev Food Sci Nutr 2011;51:81627.
Farinas CS, Leite A, Miranda EA. Aqueous extraction of recombinant human proinsulin
from transgenic maize endosperm. Biotechnol Prog 2005;21:146671.
FDA. Guidance for industry Q8(R2) pharmaceutical development. In: Services USDoHaH,
editor. Rockville, MD, USA: U.S. Department of Health and Human Services; 2009.
[Revision 2 ed.].
Feder J. Panel discussion. In: Feder J, editor. Large-scale mammalian cell culture.
Amsterdam: Elsevier Science; 2012. p. 170.
Fischer R, Schillberg S, Hellwig S, Twyman RM, Drossard J. GMP issues for recombinant
plant-derived pharmaceutical proteins. Biotechnol Adv 2012;30:4349.
Fischer R, Schillberg S, Buyel JF, Twyman RM. Commercial aspects of pharmaceutical protein production in plants. Curr Pharm Des 2013;19:54717.
Fu H, Machado PA, Hahm TS, Kratochvil RJ, Wei CI, Lo YM. Recovery of nicotine-free proteins from tobacco leaves using phosphate buffer system under controlled conditions.
Bioresour Technol 2010;101:203442.
Gecchele E, Schillberg S, Merlin M, Pezzotti M, Avesani L. A downstream process allowing
the efcient isolation of a recombinant amphiphilic protein from tobacco leaves.
J Chromatogr B 2014;960:3442.
Geli MI, Torrent M, Ludevid D. Two structural domains mediate two sequential events in
[gamma]-zein targeting: protein endoplasmic reticulum retention and protein body
formation. Plant Cell 1994;6:191122.
Georgiev MI, Weber J, Maciuk A. Bioprocessing of plant cell cultures for mass production
of targeted compounds. Appl Microbiol Biotechnol 2009;83:80923.
Glassey J, Gernaey KV, Clemens C, Schulz TW, Oliveira R, Striedner G, et al. Process analytical technology (PAT) for biopharmaceuticals. Biotechnol J 2011;6:36977.
Gleba Y, Klimyuk V, Marillonnet S. Magnifectiona new platform for expressing recombinant vaccines in plants. Vaccine 2005;23:20428.
Gleba YY, Tuse D, Giritch A. Plant viral vectors for delivery by agrobacterium. Curr Top
Microbiol Immunol 2014;375:15592.
Gottschalk U. Process scale purication of antibodies. Hoboken, New Jersey: John Wiley &
Sons; 2009.
Gregory J, Barany S. Adsorption and occulation by polymers and polymer mixtures. Adv
Colloid Interface Sci 2011;169:112.
Gu Z. Recovery of recombinant proteins from plants using aqueous two-phase
partitioning systems: an outline. In: Labrou NE, editor. Protein downstream processing. Berlin: Springer Science + Business Media; 2014. p. 7787.
Haimowitz J, Warran J. Economic value of standardization. Ottawa: Standards Council of
Canada; 2007. p. 51.
Hakanpaa J, Szilvay GR, Kaljunen H, Maksimainen M, Linder M, Rouvinen J. Two crystal
structures of Trichoderma reesei hydrophobin HFBI the structure of a protein amphiphile with and without detergent interaction. Protein Sci 2006;15:212940.
Hassan S, van Dolleweerd CJ, Ioakeimidis F, Keshavarz-Moore E, Ma JK. Considerations for
extraction of monoclonal antibodies targeted to different subcellular compartments
in transgenic tobacco plants. Plant Biotechnol J 2008;6:73348.
Hassan S, Keshavarz-Moore E, Ma J, Thomas C. Breakage of transgenic tobacco roots for
monoclonal antibody release in an ultra-scale down shearing device. Biotechnol
Bioeng 2014;111:196201.
Hatti-Kaul R. Aqueous two-phase systems. A general overview. Mol Biotechnol 2001;19:
26977.
Hayden CA, Egelkrout EM, Moscoso AM, Enrique C, Keener TK, Jimenez-Flores R, et al. Production of highly concentrated, heat-stable hepatitis B surface antigen in maize. Plant
Biotechnol J 2012;10:97984.
Hayden CA, Smith EM, Turner DD, Keener TK, Wong JC, Walker JH, et al. Supercritical uid
extraction provides an enhancement to the immune response for orally-delivered
hepatitis B surface antigen. Vaccine 2014;32:12406.
Holler C, Vaughan D, Zhang CM. Polyethyleneimine precipitation versus anion exchange
chromatography in fractionating recombinant beta-glucuronidase from transgenic
tobacco extract. J Chromatogr A 2007;1142:98105.
Hubbuch J, Thommes J, Kula MR. Biochemical engineering aspects of expanded bed adsorption. Technology transfer in biotechnology: from lab to industry to production,
92; 2005. p. 10123.
Hubbuch JJ, Brixius PJ, Lin DQ, Mollerup I, Kula MR. The inuence of homogenisation conditions on biomass-adsorbent interactions during ion-exchange expanded bed adsorption. Biotechnol Bioeng 2006;94:54353.
Hussack G, Grohs BM, Almquist KC, McLean MD, Ghosh R, Hall JC. Purication of Plantderived antibodies through direct immobilization of afnity ligands on cellulose.
J Agric Food Chem 2010;58:34519.
Iribarren OA, Montagna JM, Vecchietti AR, Andrews B, Asenjo JA, Pinto JM. Optimal process synthesis for the production of multiple recombinant proteins. Biotechnol Prog
2004;20:103243.
Jacob SS, Cherian S, Sumithra TG, Raina OK, Sankar M. Edible vaccines against veterinary
parasitic diseasescurrent status and future prospects. Vaccine 2013;31:187985.
Joensuu JJ, Conley AJ, Lienemann M, Brandle JE, Linder MB, Menassa R. Hydrophobin fusions for high-level transient protein expression and purication in Nicotiana
benthamiana. Plant Physiol 2010;152:62233.
Jones HD, Doherty A, Sparks CA. Transient transformation of plants. Methods Mol Biol
2009;513:13152.
Jungbauer A. Continuous downstream processing of biopharmaceuticals. Trends
Biotechnol 2013;31:47992.
Kang Y, Hamzik J, Felo M, Qi B, Lee J, Ng S, et al. Development of a novel and efcient cell
culture occulation process using a stimulus responsive polymer to streamline antibody purication processes. Biotechnol Bioeng 2013;110:292837.
Kapchie VN, Hauck CC, Wang H, Murphy PA. Process improvement for Semipuried
Oleosomes on a pilot-plant scale. J Food Sci 2011;76:C85360.
Kelley B. Industrialization of mAb production technology: the bioprocessing industry at a
crossroads. MAbs 2009;1:44352.
Kelley BD, Tobler SA, Brown P, Coffman JL, Godavarti R, Iskra T, et al. Weak partitioning
chromatography for anion exchange purication of monoclonal antibodies.
Khan I, Twyman RM, Arcalis E, Stoger E. Using storage organelles for the accumulation
and encapsulation of recombinant proteins. Biotechnol J 2012;7:1099108.
Kim TH, Yoo CG, Lamsal BP. Front-end recovery of protein from lignocellulosic biomass
and its effects on chemical pretreatment and enzymatic saccharication. Bioprocess
Biosyst Eng 2013a;36:68794.
Kim YJ, Lee HM, Wang YM, Wu J, Kim SG, Kang KY, et al. Depletion of abundant plant
RuBisCO protein using the protamine sulfate precipitation method. Proteomics
2013b;13:21769.
Kingsbury NJ, McDonald KA. Quantitative evaluation of E1 endoglucanase recovery from
tobacco leaves using the vacuum inltrationcentrifugation method. Biomed Res Int
2014;2014:483596.
Kolotilin I, Topp E, Cox E, Devriendt B, Conrad U, Joensuu J, et al. Plant-based solutions for
veterinary immunotherapeutics and prophylactics. Vet Res 2014;45.
Komarnytsky S, Borisjuk NV, Borisjuk LG, Alam MZ, Raskin I. Production of recombinant
proteins in tobacco guttation uid. Plant Physiol 2000;124:92733.
Kuo YC, Tan CC, Ku JT, Hsu WC, Su SC, Lu CA, et al. Improving pharmaceutical protein production in Oryza sativa. Int J Mol Sci 2013;14:871939.
Lacorte C, Ribeiro SG, Lohuis D, Goldbach R, Prins M. Potatovirus X and Tobacco mosaic
virus-based vectors compatible with the Gateway cloning system. J Virol Methods
2010;164:713.
Lai HF, Chen Q. Bioprocessing of plant-derived virus-like particles of Norwalk virus capsid
protein under current Good Manufacture Practice regulations. Plant Cell Rep 2012;
31:57384.
Lan DM, Huang GR, Shao HW, Zhang LC, Ma LX, Chen SW, et al. An improved
nonchromatographic method for the purication of recombinant proteins using
elastin-like polypeptide-tagged proteases. Anal Biochem 2011;415:2002.
Lander R, Daniels C, Meacle F. Efcient, scalable clarication of diverse bioprocess streams.
Bioprocess Int 2005;3:3240.
Li YF. Self-cleaving fusion tags for recombinant protein production. Biotechnol Lett 2011;
33:86981.
Li YN, Yin LH, Zheng LL, Xu LN, Xu YW, Zhao YY, et al. Application of high-speed countercurrent chromatography coupled with a reverse micelle solvent system to separate
three proteins from Momordica charantia. J Chromatogr B 2012;895:7782.
Lico C, Chen Q, Santi L. Viral vectors for production of recombinant proteins in plants.
J Cell Physiol 2008;216:36677.
Lightfoot EN. Can membrane cascades replace chromatography? Adapting binary ideal
cascade theory of systems of two solutes in a single solvent. Sep Sci Technol 2005;
40:73956.
Lightfoot EN, Cockrem MCM. Complex tness diagrams: downstream processing of biologicals. Sep Sci Technol 2013;48:17537.
Lim S, Chisholm K, Cofn RH, Peters RD, A-Mughrabi KI, Wang-Pruski G, et al. Protein
proling in potato (Solanum tuberosum L.) leaf tissues by differential centrifugation.
J Proteome Res 2012;11:2594601.
Liu JH, Fan LT, Seib P, Friedler F, Bertok B. Downstream process synthesis for biochemical
production of butanol, ethanol, and acetone from grains: generation of optimal and
near-optimal owsheets with conventional operating units. Biotechnol Prog 2004;
20:151827.
Liu S, Simaria AS, Farid SS, Papageorgiou LG. Designing cost-effective biopharmaceutical
facilities using mixed-integer optimization. Biotechnol Prog 2013;29:147283.
Love AJ, Chapman SN, Matic S, Noris E, Lomonossoff GP, Taliansky M. In planta production
of a candidate vaccine against bovine papillomavirus type 1. Planta 2012;236:
130513.
Ma JKC, Christou P, Chikwamba R, Haydon H, Paul M, Ferrer MP, et al. Realising the value
of plant molecular pharming to benet the poor in developing countries and emerging economies. Plant Biotechnol J 2013;11:102933.
Machado PA, Fu H, Kratochvil RJ, Yuan YH, Hahm TS, Sabliov CM, et al. Recovery of
solanesol from tobacco as a value-added byproduct for alternative applications.
Makhzoum A, Benyammi R, Moustafa K, Tremouillaux-Guiller J. Recent advances on host
plants and expression cassettes' structure and function in plant molecular pharming.
Biodrugs 2014;28:14559.
Manfreda G, De Cesare A. The challenge of dening risk-based metrics to improve food
safety: inputs from the BASELINE project. Int J Food Microbiol 2014;184:27.
Marchetti C. 1012: a check on the earth-carrying capacity for man. Energy 1979;4:
110717.
Markley N, Nykiforuk C, Boothe J, Moloney M. Producing proteins using transgenic
oilbody-oleosin technology. Biopharm Int 2006;19:34-+.
Maschke RW, Geipel K, Bley T. Modeling of plant in vitro cultures: overview and estimation of biotechnological processes. Biotechnol Bioeng 2015;112:112.

McLean MD, Chen RJ, Yu DQ, Mah KZ, Teat J, Wang HF, et al. Purication of the therapeutic
antibody trastuzumab from genetically modied plants using safower Protein
A-oleosin oilbody technology. Transgenic Res 2012;21:1291301.
McMeekin TA, Baranyi J, Bowman J, Dalgaard P, Kirk M, Ross T, et al. Information systems
in food safety management. Int J Food Microbiol 2006;112:18194.
McPartland TJ, Patil RA, Malone MF, Roberts SC. Liquidliquid extraction for recovery of
paclitaxel from plant cell culture: solvent evaluation and use of extractants for
partitioning and selectivity. Biotechnol Prog 2012;28:9907.
Melnik S, Stoger E. Green factories for biopharmaceuticals. Curr Med Chem 2013;20:
103846.
Menkhaus TJ, Glatz CE. Antibody capture from corn endosperm extracts by packed bed
and expanded bed adsorption. Biotechnol Prog 2005;21:47385.
Menkhaus TJ, Roseland J. Recovery of proteins from corn and soybean extracts by membrane adsorption. Biotechnol Prog 2008;24:107584.
Menkhaus TJ, Eriksson SU, Whitson PB, Glatz CE. Host selection as a downstream strategy:
polyelectrolyte precipitation of beta-glucuronidase from plant extracts. Biotechnol
Bioeng 2002;77:14854.
Menkhaus TJ, Bai Y, Zhang C, Nikolov ZL, Glatz CE. Considerations for the recovery of recombinant proteins from plants. Biotechnol Prog 2004;20:100114.
Merlin M, Gecchele E, Capaldi S, Pezzotti M, Avesani L. Comparative evaluation of recombinant protein production in different biofactories: the green perspective. Biomed
Res Int 2014.
Meyer DE, Chilkoti A. Purication of recombinant proteins by fusion with thermallyresponsive polypeptides. Nat Biotechnol 1999;17:11125.
Montgomery DC. Design and analysis of experiments. John Wiley & Sons Incorporated;
2007.
Murad AM, Souza GHMF, Garcia JS, Rech EL. Detection and expression analysis of recombinant proteins in plant-derived complex mixtures using nanoUPLC-MSE. J Sep Sci
2011;34:261830.
Myers RH, Montgomery DC, Anderson-Cook CM. Response surface methodology: process
and product optimization using designed experiments. Wiley; 2009.
Naik AD, Menegatti S, Reese HR, Gurgel PV, Carbonell RG. Process for purication
of monoclonal antibody expressed in transgenic Lemna plant extract using
dextran-coated charcoal and hexamer peptide afnity resin. J Chromatogr A 2012;
1260:616.
Nandi S, Yalda D, Lu S, Nikolov Z, Misaki R, Fujiyama K, et al. Process development and
economic evaluation of recombinant human lactoferrin expressed in rice grain.
Transgenic Res 2005;14:23749.
Narayan AV, Madhusudhan MC, Raghavarao KSMS. Extraction and purication of Ipomoea
peroxidase employing Three-phase partitioning. Appl Biochem Biotechnol 2008;151:
26372.
Nfor BK, Zuluaga DS, Verheijen PJT, Verhaert PDEM, van der Wielen LAM, Ottens M.
Model-based rational strategy for chromatographic resin selection. Biotechnol Prog
2011;27:162943.
Nfor BK, Ahamed T, Pinkse MW, van der Wielen LA, Verhaert PD, van Dedem GW, et al.
Multi-dimensional fractionation and characterization of crude protein mixtures: toward establishment of a database of protein purication process development parameters. Biotechnol Bioeng 2012;109:307083.
Nikolov ZL, Woodard SL. Downstream processing of recombinant proteins from transgenic feedstock. Curr Opin Biotechnol 2004;15:47986.
Nykiforuk CL, Boothe JG, Murray EW, Keon RG, Goren HJ, Markley NA, et al. Transgenic expression and recovery of biologically active recombinant human insulin from
Arabidopsis thaliana seeds. Plant Biotechnol J 2006;4:7785.
O'Brien TP, Brown LA, Battersby DG, Rudolph AS, Raman LP. Large-scale, single-use depth
ltration systems for mammalian cell culture clarication. Bioprocess Int 2012;10:
507.
Oelmeier SA, Dismer F, Hubbuch J. Application of an aqueous two-phase systems highthroughput screening method to evaluate mAb HCP separation. Biotechnol Bioeng
2011;108:6981.
Orr V, Zhong LY, Moo-Young M, Chou CP. Recent advances in bioprocessing application of
membrane chromatography. Biotechnol Adv 2013;31:45065.
Ota M, Watanabe H, Kato Y, Watanabe M, Sato Y, Smith RL, et al. Carotenoid production
from Chlorococcum littorale in photoautotrophic cultures with downstream supercritical uid processing. J Sep Sci 2009;32:232735.
Paraman I, Fox SR, Aspelund MT, Glatz CE, Johnson LA. Recovering corn germ
enriched in recombinant protein by wet-fractionation. Bioresour Technol 2010;
101:23944.
Pastores GM, Petakov M, Giraldo P, Rosenbaum H, Szer J, Deegan PB, et al. A Phase 3, multicenter, open-label, switchover trial to assess the safety and efcacy of taliglucerase
alfa, a plant cell-expressed recombinant human glucocerebrosidase, in adult and pediatric patients with Gaucher disease previously treated with imiglucerase. Blood
Cells Mol Dis 2014;53:25360.
Paul M, Ma JK. Plant-made pharmaceuticals: leading products and production platforms.
Biotechnol Appl Biochem 2011;58:5867.
Paul MJ, Teh AYH, Twyman RM, Ma JKC. Target product selection where can molecular
pharming make the difference? Curr Pharm Des 2013;19:547885.
Pegel A, Reiser S, Steurenthaler M, Klein S. Evaluating disposable depth ltration platforms for mAb harvest clarication. Bioprocess Int 2011;9:526.
Phan HT, Conrad U. Membrane-based inverse transition cycling: an improved means for
purifying plant-derived recombinant protein-elastin-like polypeptide fusions. Int J
Mol Sci 2011;12:280821.
Phan HT, Hause B, Hause G, Arcalis E, Stoger E, Maresch D, et al. Inuence of elastin-like
polypeptide and hydrophobin on recombinant hemagglutinin accumulations in
transgenic tobacco plants. PLoS One 2014;9:e99347.
Pina AS, Lowe CR, Roque ACA. Challenges and opportunities in the purication of recombinant tagged proteins. Biotechnol Adv 2014;32:36681.
11
Platis D, Drossard J, Fischer R, Ma JKC, Labrou NE. New downstream processing strategy
for the purication of monoclonal antibodies from transgenic tobacco plants.
J Chromatogr A 2008;1211:809.
Pogue GP, Vojdani F, Palmer KE, Hiatt E, Hume S, Phelps J, et al. Production of
pharmaceutical-grade recombinant aprotinin and a monoclonal antibody product
using plant-based transient expression systems. Plant Biotechnol J 2010;8:63854.
Ramessar K, Rademacher T, Sack M, Stadlmann J, Platis D, Stiegler G, et al. Cost-effective
production of a vaginal protein microbicide to prevent HIV transmission. Proc Natl
Acad Sci U S A 2008;105:372732.
Randell AW, Whitehead AJ. Codex Alimentarius: food quality and safety standards for international trade. Rev Sci Tech 1997;16:31321.
Rao RN, Talluri MVNK, Krishna TSVNM, Ravindranath K. Continuous counter current extraction, isolation and determination of solanesol in Nicotiana tobacum L. by nonaqueous reversed phase high performance liquid chromatography. J Pharm Biomed
Anal 2008;46:3105.
Rathore AS. Roadmap for implementation of quality by design (QbD) for biotechnology
products. Trends Biotechnol 2009;27:54653.
Rathore AS, Latham P, Levine H, Curling J, Kaltenbrunner O. Costing issues in the production of biopharmaceuticals manufacturing costs are crucial to overall prot margins. Biopharm Int 2004;17:4654.
Rathore AS, Bhambure R, Ghare V. Process analytical technology (PAT) for biopharmaceutical products. Anal Bioanal Chem 2010;398:13754.
Read EK, Park JT, Shah RB, Riley BS, Brorson KA, Rathore AS. Process analytical technology
(PAT) for biopharmaceutical products: Part I. Concepts and applications. Biotechnol
Bioeng 2010a;105:27684.
Read EK, Shah RB, Riley BS, Park JT, Brorson KA, Rathore AS. Process analytical technology
(PAT) for biopharmaceutical products: Part II. Concepts and applications. Biotechnol
Bioeng 2010b;105:28595.
Regnard GL, Halley-Stott RP, Tanzer FL, Hitzeroth II, Rybicki EP. High level protein expression in plants through the use of a novel autonomously replicating geminivirus shuttle vector. Plant Biotechnol J 2010;8:3846.
Reuter LJ, Bailey MJ, Joensuu JJ, Ritala A. Scale-up of hydrophobin-assisted recombinant
protein production in tobacco BY-2 suspension cells. Plant Biotechnol J 2014;12:
40210.
Robic G, Farinas CS, Rech EL, Miranda EA. Transgenic soybean seed as protein expression
system: aqueous extraction of recombinant beta-glucuronidase. Appl Biochem
Biotechnol 2010;160:115767.
Roush DJ, Lu YF. Advances in primary recovery: centrifugation and membrane technology. Biotechnol Prog 2008;24:48895.
Rybicki EP, Chikwamba R, Koch M, Rhodes JI, Groenewald JH. Plant-made therapeutics: an
emerging platform in South Africa. Biotechnol Adv 2012;30:44959.
Sabalza M, Christou P, Capell T. Recombinant plant-derived pharmaceutical proteins:
current technical and economic bottlenecks. Biotechnol Lett 2014;36:236779.
Sainsbury F, Lomonossoff GP. Extremely high-level and rapid transient protein production
in plants without the use of viral replication. Plant Physiol 2008;148:12128.
Schillberg S, Raven N, Fischer R, Twyman RM, Schiermeyer A. Molecular farming of pharmaceutical proteins using plant suspension cell and tissue cultures. Curr Pharm Des
2013;19:553142.
Schugerl K, Hubbuch J. Integrated bioprocesses. Curr Opin Microbiol 2005;8:294300.
Shaaltiel Y, Bartfeld D, Hashmueli S, Baum G, Brill-Almon E, Galili G, et al. Production
of glucocerebrosidase with terminal mannose glycans for enzyme replacement
therapy of Gaucher's disease using a plant cell system. Plant Biotechnol J 2007;5:
57990.
Shoji Y, Farrance CE, Bautista J, Bi H, Musiychuk K, Horsey A, et al. A plant-based system
for rapid production of inuenza vaccine antigens. Inuenza Other Respi Viruses
2012;6:20410.
Shukla AA, Gottschalk U. Single-use disposable technologies for biopharmaceutical
manufacturing. Trends Biotechnol 2013;31:14754.
Shukla AA, Thommes J. Recent advances in large-scale production of monoclonal antibodies and related proteins. Trends Biotechnol 2010;28:25361.
Skarjinskaia M, Ruby K, Araujo A, Taylor K, Gopalasamy-Raju V, Musiychuk K, et al. Hairy
roots as a vaccine production and delivery system. Adv Biochem Eng Biotechnol
2013;134:11534.
Soares PAG, Nascimento CO, Porto TS, Correia MTS, Porto ALF, Carneiro-da-Cunha MG. Purication of a lectin from Canavalia ensiformis using PEG-citrate aqueous two-phase
system. J Chromatogr B 2011;879:45760.
Stoger E, Fischer R, Moloney M, Ma JKC. Plant molecular pharming for the treatment of
chronic and infectious diseases. Annu Rev Plant Biol 2014;65:74368.
Straetkvern KO, Schwarz JG, Wiesenborn DP, Zarakos E, Lihme A. Expanded bed adsorption for recovery of patatin from crude potato juice. Bioseparation 1999;7:33345.
Strasser R. Engineering of human-type O-glycosylation in Nicotiana benthamiana plants.
Bioengineered 2013;4:1916.
Takaiwa F. Increasing the production yield of recombinant protein in transgenic seeds by
expanding the deposition space within the intracellular compartment. Bioengineered
2013;4:1369.
Teng Z, Wang Q. Extraction, identication and characterization of the water-insoluble
proteins from tobacco biomass. J Sci Food Agric 2012;92:136874.
The CMC Biotech Working Group. A-Mab: a case study in bioprocess development. In:
The CMC biotech working group, editor. 2.1 ed. Emeryville, CA, USA: CASSS An International Separation Science Society; 2009. p. 278.
Tian L, Sun SSM. A cost-effective ELP-intein coupling system for recombinant protein purication from plant production platform. PLoS One 2011;6:e24183.
Tpfer A, Blum U, Eickhoff G, Junginger I, Blind K, Grupp H, et al. Economic benets of
standardization. Berlin: DIN German Institute for Standardization e. V; 2000. p. 39.
Torrent M, Llompart B, Lasserre-Ramassamy S, Llop-Tous I, Bastida M, Marzabal P, et al.
Eukaryotic protein production in designed storage organelles. BMC Biol 2009;7:5.
12
Tremblay R, Diao H, Huner N, Jevnikar AM, Ma SW. The development, characterization,

and demonstration of a novel strategy for purication of recombinant proteins
expressed in plants. Transgenic Res 2011;20:135766.
Turpen TH. Tobacco mosaic virus and the virescence of biotechnology. Philos Trans R Soc
Lond B Biol Sci 1999;354:66573.
Tuse D, Tu T, McDonald KA. Manufacturing economics of plant-made biologics: case studies in therapeutic and industrial enzymes. Biomed Res Int 2014;2014:256135.
Twyman RM, Schillberg S, Fischer R. Transgenic plants in the biopharmaceutical market.
Expert Opin Emerg Drugs 2005;10:185218.
Twyman RM, Schillberg S, Fischer R. Optimizing the yield of recombinant pharmaceutical
proteins in plants. Curr Pharm Des 2013;19:548694.
Urry D. Entropic elastic processes in protein mechanisms. I. Elastic structure due to an inverse temperature transition and elasticity due to internal chain dynamics. J Protein
Chem 1988;7:134.
Urry DW. Physical chemistry of biological free energy transduction as demonstrated by
elastic protein-based polymers. J Phys Chem B 1997;101:1100728.
Vazquez-Villegas P, Espitia-Saloma E, Rito-Palomares M, Aguilar O. Low-abundant protein
extraction from complex protein sample using a novel continuous aqueous twophase systems device. J Sep Sci 2013;36:3919.
Virdi V, Depicker A. Role of plant expression systems in antibody production for passive
immunization. Int J Dev Biol 2013;57:58793.
Walwyn DR, Huddy SM, Rybicki EP. Techno-economic analysis of horseradish peroxidase
production using a transient expression system in Nicotiana benthamiana. Appl
Biochem Biotechnol 2015;175:84154.
Wang XQ, Zhang XW. Optimal extraction and hydrolysis of Chlorella pyrenoidosa proteins.
Wang CX, Lee TC, Crowley KS, Bell E. Purication of phosphinothricin acetyltransferase
using Reactive brown 10 afnity in a single chromatography step. Protein Expr
Purif 2013;90:12934.
Warikoo V, Godawat R, Brower K, Jain S, Cummings D, Simons E, et al. Integrated continuous production of recombinant therapeutic proteins. Biotechnol Bioeng 2012;109:
301829.
Weichert N, Hauptmann V, Menzel M, Schallau K, Gunkel P, Hertel TC, et al.

Transglutamination allows production and characterization of native-sized ELPylated
spider silk proteins from transgenic plants. Plant Biotechnol J 2014;12:26575.
Wilken LR, Nikolov ZL. Recovery and purication of plant-made recombinant proteins.
Biotechnol Adv 2012;30:41933.
Wilson SA, Roberts SC. Recent advances towards development and commercialization of
plant cell culture processes for the synthesis of biomolecules. Plant Biotechnol J 2012;
10:24968.
Wirz H, Sauer-Budge AF, Briggs J, Sharpe A, Shu SD, Sharon A. Automated production of
plant-based vaccines and pharmaceuticals. J Lab Autom 2012;17:44957.
Wolfson W. Grow your own: Protalix BioTherapeutics produces drugs in carrot cells.
Chem Biol 2013;20:96970.
Woodard SL, Wilken LR, Barros GOF, White SG, Nikolov ZL. Evaluation of monoclonal antibody and phenolic extraction from transgenic Lemna for purication process development. Biotechnol Bioeng 2009;104:56271.
Wu XL, Xiong EH, Wang W, Scali M, Cresti M. Universal sample preparation method integrating trichloroacetic acid/acetone precipitation with phenol extraction for crop
proteomic analysis. Nat Protoc 2014;9:36274.
Yan Q, Miazek K, Grande PM, Domnguez de Mara P, Leitner W, Modigell M. Mechanical
pretreatment in a screw press affecting chemical pulping of lignocellulosic biomass.
Energy Fuel 2014;28:69817.
Yang Y, Farid SS, Thornhill NF. Data mining for rapid prediction of facility t and
debottlenecking of biomanufacturing facilities. J Biotechnol 2014;179:1725.
Yasarla LR, Ramarao BV. Dynamics of occulation of lignocellulosic hydrolyzates by polymers. Ind Eng Chem Res 2012;51:684761.
Yigzaw Y, Piper R, Tran M, Shukla AA. Exploitation of the adsorptive properties of depth
lters for host cell protein removal during monoclonal antibody purication.
Biotechnol Prog 2006;22:28896.

Journal 9

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal 9

Uploaded by

Copyright:

Available Formats

JBA-06934; No of Pages 12

Biotechnology Advances xxx (2015) xxxxxx

Contents lists available at ScienceDirect

Research review paper

Extraction and downstream processing of plant-derived

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

In some cases extraction is necessary because the product is targeted

In simple terms, the role of upstream production is to maximize the

2.2. Process-related impurities

3.1. Tissue homogenization

Ali et al. (2012)

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

and allow continuous operation, but cross-product contamination is

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

The partitioning behavior of target proteins can also be modied by

Integrated methods for the processing of plant extracts include juice

4.3. Purication using genetic fusion tags

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

4.4. Expanded bed adsorption

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

sugarcane extracts using 100-kDa and 3-kDa membranes (Barros et al.,

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

and less chromatography resin are needed to achieve the purication of

antibodies produced in mammalian cells (The CMC Biotech Working

7. Information box on DoE and process synthesis: Development of a

6.3. Vertical farming

A major challenge in the development of manufacturing processes

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

J.F. Buyel et al. / Biotechnology Advances xxx (2015) xxxxxx

Tremblay R, Diao H, Huner N, Jevnikar AM, Ma SW. The development, characterization,

Weichert N, Hauptmann V, Menzel M, Schallau K, Gunkel P, Hertel TC, et al.

You might also like