Methods 116 (2017) 84–94

Contents lists available at ScienceDirect

Methods
journal homepage: www.elsevier.com/locate/ymeth

Affinity chromatography: A versatile technique for antibody purification

Sushrut Arora a, Vikas Saxena b, B. Vijayalakshmi Ayyar c,⇑
a
Department of Pediatrics, Baylor College of Medicine, Houston, TX 77030, USA
b
Center for Vascular and Inflammatory Diseases, School of Medicine, University of Maryland, Baltimore, MD 21201, USA
c
Department of Biochemistry and Molecular Biology, Baylor College of Medicine, Houston, TX 77030, USA

a r t i c l e i n f o a b s t r a c t

Article history: Antibodies continue to be extremely utilized entities in myriad applications including basic research,
Received 2 November 2016 imaging, targeted delivery, chromatography, diagnostics, and therapeutics. At production stage, antibod-
Received in revised form 16 December 2016 ies are generally present in complex matrices and most of their intended applications necessitate purifi-
Accepted 17 December 2016
cation. Antibody purification has always been a major bottleneck in downstream processing of
Available online 22 December 2016
antibodies, due to the need of high quality products and associated high costs. Over the years, extensive
research has focused on finding better purification methodologies to overcome this holdup. Among a
Keywords:
plethora of different techniques, affinity chromatography is one of the most selective, rapid and easy
Antibody
Antibody purification
method for antibody purification. This review aims to provide a detailed overview on affinity chromatog-
Affinity chromatography raphy and the components involved in purification. An array of support matrices along with various
Affinity matrices classes of affinity ligands detailing their underlying working principles, together with the advantages
Affinity ligands and limitations of each system in purifying different types of antibodies, accompanying recent develop-
ments and important practical methodological considerations to optimize purification procedure are
discussed.
Ó 2016 Elsevier Inc. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
2. Affinity chromatography: principle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3. Affinity chromatography: components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.1. Chromatography matrix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
3.2. Affinity ligands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.2.1. Biospecific ligands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.2.2. Pseudobiospecific ligands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.2.3. Synthetic ligands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.2.4. Affinity tags. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
4. Affinity chromatography: practical considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.1. Sample preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.2. Binding. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.3. Washing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
4.4. Elution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5. Conclusions and future perspectives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

⇑ Corresponding author.
E-mail address: vijiayyar@gmail.com (B.V. Ayyar).

http://dx.doi.org/10.1016/j.ymeth.2016.12.010
1046-2023/Ó 2016 Elsevier Inc. All rights reserved.
 S. Arora et al. / Methods 116 (2017) 84–94 85

1. Introduction specific antigenic determinants (epitopes). In the heavy and light

Antibodies are inducible defensive components of host immune
system programmed to identify and neutralize non-self-entities
such as pathogens. Antibodies are specific components of adaptive
immune system which not only directly engage in the elimination
of foreign particles but also trigger other non-specific effector func-
tions such as antibody-dependent cell-mediated cytotoxicity,
complement-dependent cytotoxicity, and antibody-dependent
cell-mediated phagocytosis to eradicate the threat to host’s
homeostasis.
An antibody is typified by the immunoglobulin G (IgG)
molecule, the most abundant subclass found in the serum of
mammals [1,2]. IgG finds extensive application in assay develop-
ment and antibody-based therapeutics [3,4] owing to their
relatively small size and stability during the isolation and purifi-
cation processes. An IgG is represented as a Y-shaped structure
consisting of four polypeptide chains, two heavy chains and
two light chains, connected by disulfide bonds (Fig. 1). These
heavy and light chains are named based on the difference in
their molecular weights. A light chain has a molecular weight
of
25 kDa whereas a heavy chain has a molecular weight of

50 kDa. Based on variability in their amino acids, both heavy
and light chains are further divided into constant (C) and vari-
able (V) regions. The constant region determines mechanisms
used to destroy antigen and segregates antibodies into five major
classes (IgM, IgG, IgA, IgD and IgE) depending on their type and
immune function. Variable region is further subdivided into
hypervariable (HV) and framework (FR) regions. The relatively
conserved amino acid sequences make-up the FR region. Con-
versely, there is great amount of variability in the sequence of
amino acids in HV regions (also known as complementarity
determining regions (CDRs)). Combination of different amino
acids in the CDRs confers to antibody its ability to recognize
chains, there are three HVs/CDRs which are connected to each
other by four FR regions. Functionally, antibodies are composed
of a fragment antigen binding (Fab), responsible for specific
binding to the antigen; and a fragment crystallizable (Fc), impor-
tant for effector mechanisms, entwining specific recognition of
antigen with non-specific effector mechanisms of the host
immune system (Fig. 1).
Antibodies are categorized into polyclonal, monoclonal and
recombinant forms. A polyclonal antibody is a mixture of antibod-
ies to a given antigen produced by different B cells. Such a mixture
contains antibodies with different affinities and specificities to dif-
ferent epitopes on the antigen. Monoclonal antibodies are pro-
duced from a single clone of a B cell, having singular epitope
specificity. The antibodies or antibody fragments generated
in vitro using molecular techniques are classified as recombinant
antibodies.
Antibodies are being exploited more and more for the obvious
benefits they offer. They are increasingly being utilized in various
scientific endeavors including but not limited to basic research,
chromatography, targeted delivery, imaging, diagnostics and ther-
apeutics [4–14]. Most of these applications require homogeneous
antibody preparations with high purity. Consequently, there is an
ever-increasing demand for robust, simple, efficient and cost-
effective purification methodologies for antibodies from complex
mixtures such as plasma, serum, ascites fluid, cell culture medium,
egg yolk, plant extracts, and bacterial and yeast cultures [15,16]. To
this end, a variety of chromatographic methodologies based on dif-
ferent biochemical properties (e.g. size, solubility, charge,
hydrophobicity and binding affinity) have been tested [15,16].
However, affinity chromatography based processes have been
reported to be the most efficient and widely employed techniques
owing to their exceptional specificity, ease of operation, yield and
relatively high throughput [16,17].

Fig. 1. Structure of a typical immunoglobulin, IgG. IgG are large molecules of approximately 150–155 kDa containing 2 pairs of heavy and light chains composed of different
domains. The heavy chain consists of a variable domain (VH) and three constant domains (CH1, CH2, and CH3). The two heavy chains are connected by disulfide bonds (SS) in
the hinge region. The light chain has one variable domain (VL) and only one constant domain (CL).
86 S. Arora et al. / Methods 116 (2017) 84–94
2. Affinity chromatography: principle
Affinity chromatography is the most selective and versatile
form of liquid chromatography [16,18,19] relying on the reversible
and specific binding between a protein and its cognate ligand, e.g.
binding of a hormone with its receptor, interaction of an enzyme
with its substrate, and binding of an antibody with its target anti-
gen. This specificity of the interaction is used in affinity chro-
matography by immobilizing one of the interacting agents, called
‘‘affinity ligand” on to the column as a stationary phase support.
Target protein is eluted from a complex mixture by selective
absorption on the affinity ligand. Complex mixture containing tar-
get protein (e.g. plasma or bacterial culture) is applied in a mobile
phase that has the right composition and pH for binding of the tar-
get protein with affinity ligand, while other components pass
through. The retained target is then eluted from the column.
Method of choice for the target elution varies according to the
affinity of target and ligand interaction. Following elution, column
is allowed to regenerate prior to the application of next sample.
3. Affinity chromatography: components
Affinity chromatography employs different types of binding
agents (ligands) and solid phase supports (matrices). By optimizing
various properties, such as specificity, selectivity, reproducibility,
conjugation chemistries, and cost-effectiveness of these compo-
nents, this method can be applied for the large-scale purification
of antibodies to achieve desired yield and purity of the product.
Over the years, substantial efforts have been made by scientific fra-
ternity on finding or generating quality matrices and affinity
ligands to aid in the development of effective methodologies for
affinity chromatography, with an aim of increased purity and yield
of the desired antibody in an economic and efficient way.
3.1. Chromatography matrix
A prerequisite for affinity chromatography is a support system
for the ligands and targets. During early phase of chromatography
method development natural compounds were used both as sup-
port and binding agents, mainly in the purification of enzyme.
Examples of such efforts include purification of amylase by insol-
uble starch [20–24], alginic acid by polygalacturonase [25], pepsin
by crystal form of edestin [26] and elastase by elastin [27] as sup-
port and binding agent. In later years, the range of biomolecules of
interest expanded, however, availability of ligand and support sys-
tem from natural sources remained limited. This led to the quest
for supporting materials from alternate sources. Desired properties
expected of these supporting materials included exhibition of min-
imum non-specific absorption, insolubility in the buffer system
used, adaptability to the appropriate flow characteristics, availabil-
ity of adequate surface area, and easy activation, for the coupling of
Table 1
Different matrices for use in affinity chromatography.
Matrix Class Examples Commercial Names
Natural Agarose Sepharose
Dextrans Sephadex and Superdex
Cellulose beads Cellulose beads
Synthetic Acrylamide Bio-Gel
Polystyrene Polystyrene
Polymethacralate Separon HEMA
Inorganic Porous silica Porous silica
Glass fiber Glass fiber
TiO2 TiO2
ligand. Additionally, they should be uniformly distributable in a
column, have macroporosity and hydrophilicity along with the sta-
bility to withstand various chemical and mechanical stresses. To
obtain an ideal matrix material different support materials and
conjugation chemistries were explored. In 1936, Landsteiner and
van der Scheer used diazo coupling method to couple haptens to
chicken erythrocyte stroma to isolate antibodies that could bind
to the immobilized haptens [28]. Another group, in 1951, used acti-
vated cellulose for the immobilization of bovine serum albumin
(BSA) for the isolation of anti-BSA antibodies from the serum of
BSA immunized rabbit [29]. Subsequently, several groups used
binding agents coupled to solid supports such as polyaminostyrene
[30] and glass beads [31], for antigen immobilization for the purifi-
cation of enzyme or antibodies. However, two major breakthrough
discoveries changed the way chromatography matrix developed.
Firstly, beaded agarose was developed in 1964 [32] which was fol-
lowed by the development of cyanogen bromide (CNBr) immobi-
lization method in 1967 [33]. These advances allowed for the
development of convenient means for covalently coupling peptides
and proteins to the polysaccharide-based materials. CNBr based
coupling method found several usages in affinity chromatography,
however, due to its toxic nature it became a health hazard for
users. Therefore, several other coupling reagents, such as carbodi-
imide, tresyl chloride, glutaraldehyde and phosphoryl chloride
were developed (refer [34]). However, their activation remained
a challenge. Therefore, pre-activated matrices were developed
and made available commercially [16]. These pre-activated matri-
ces were developed using chemistries involving primary amines,
sulfhydryls, aldehydes, hydroxyls, and carboxylic acids for cou-
pling ligands to matrices covalently. Based on the origin of com-
pounds, matrices for use in affinity chromatography were divided
into three classes; natural, synthetic and inorganic (Table 1).
Affinity separation based on the magnetic beads has gained
increased attention in the last two decades. Biocompatible mag-
netic beads were in use since 1970s in chemistry, medicine, and
biotechnology [35]. A detailed review describing their use in pro-
tein and peptide purification had been published earlier [36]. Mag-
netic beads coupled with appropriate ligand-based purification
employ simple steps of mixing and incubating beads with the tar-
get sample containing antibody of interest. Magnetic beads having
antibodies bound to them are then captured by the separator, fol-
lowed by washing, and elution of the bound antibodies [36,37].
Magnetic beads-based procedure is simple, fast, involves less sheer
pressure, and is applicable to samples with very high protein con-
centrations. Therefore, magnetic beads are frequently used in high-
throughput automation [38]. To prepare magnetic support, mag-
netic particles are entrapped by copolymerization with the poly-
mer, onto which the biomolecules can be immobilized [39].
Different kind of supports such as polystyrene, agarose, cellulose,
silica, and porous glass can be entrapped onto the magnetic parti-
cles of different sizes. Due to their low cost and robustness, mag-
netic beads are used in a variety of applications and are available
commercially in different formats.
The aforementioned chromatographic matrices can be applied
to columns, however, their relatively high cost and a high pressure
drop along the column is not appreciated in certain applications.
References Therefore, an alternative method for the bioseparation of large pro-
[33] teins was used in the form of development of affinity membranes.
[181] Affinity membranes offer improved mass-transfer efficiencies, ease
[182] of handling, macroporosity, larger surface area and higher dynamic
[183,184] binding capacity [40]. Due to compactness of bed, affinity mem-
[185,186] branes can be used in various sizes and formats with better acces-
[187,188]
sibility of ligands for affinity interaction, making them easier to
[189,190] scale-up. Application of affinity membranes is similar to other
[191,192]
matrices described above, wherein membranes are activated for
[193]
the coupling of affinity ligands, and used for capturing antibodies
 S. Arora et al. / Methods 116 (2017) 84–94 87

Table 2
Binding characteristics of different immunoglobulin-binding proteins.

Species Antibody Class Protein A Protein G Protein A/G Protein L§

Human
Total IgG +++ +++ +++ +++
IgG1 +++ +++ +++ +++
IgG2 +++ +++ +++ +++
IgG3 + +++ +++ +++
IgG4 +++ +++ +++ +++
IgM +  + +++
IgD    +++
IgE ++  ++ +++
IgA +  + +++
IgA1 +  + +++
IgA2 +  + +++
Fab + + + +++
scFv +  + +++
Mouse
Total IgG +++ +++ +++ +++
IgG1 + ++ ++ +++
IgG2a +++ +++ +++ +++
IgG2b +++ +++ +++ +++
IgG3 +++ +++ +++ +++
IgM    +++
Rat
Total IgG + ++ ++ +++
IgG1 + ++ ++ +++
IgG2a  +++ +++ +++
IgG2b  + + +++
IgG2b +++ +++ +++ +++
Cow
Total IgG + +++ +++ 
IgG1 + +++ +++ 
IgG2 +++ +++ +++ 
Goat
Total IgG + +++ +++ 
IgG1 + +++ +++ 
IgG2 +++ +++ +++ 
Sheep
Total IgG + +++ +++ 
IgG1 + +++ +++ 
IgG2 +++ +++ +++ 
Horse
⁄
Total IgG + +++ +++
⁄
IgG(ab) +  +
⁄
IgG(c) +  +
⁄
IgG(T)  +++ +++
Rabbit
Total IgG +++ +++ +++ +
Guinea pig
Total IgG +++ + +++ +
Pig
Total IgG +++ + +++ +++
Dog
⁄
Total IgG +++ + +++
Hamster
Total IgG ++ ++ ++ +++
Donkey
⁄
Total IgG ++ +++ +++
Cat
⁄
Total IgG +++ + +++
Monkey
⁄
Total IgG +++ +++ +++
Chicken
Total IgY    ++

+ = weak binding, ++ = medium binding, +++ = strong binding,  = no binding, * = information not available.
§
Binding occurs only if the appropriate kappa light chains are present. Reproduced with permission from [15].
88 S. Arora et al. / Methods 116 (2017) 84–94
[41]. Affinity membranes find widespread use in high-throughput
applications and are available commercially as SwellGel disks.
3.2. Affinity ligands
Affinity ligands are the principle component of a successful
affinity chromatographic set-up [19]. Typical qualities of a suitable
ligand for affinity chromatography include affinity to the target,
specificity, immobilization feasibility, stability in harsh washing
and elution conditions, and retention of target binding capacity
after attachment to the matrix. A plethora of research has been
conducted to develop and mend these ligands for improved protein
purification. However, only a few of these ligands have found con-
sistent use for antibody purification. These ligands are classified in
three broad groups: biospecific ligands, pseudobiospecific ligands
and synthetic ligands. Additionally, recombinant affinity tags serve
as alternative ligands for purification of recombinant antibodies.
3.2.1. Biospecific ligands
Biospecific ligands are naturally derived molecules and usually
have high binding affinities to antibodies. This class includes anti-
body binding proteins of biological origins, such as bacterially
derived proteins, antigens, lectins and anti-antibodies. Biospecific
ligands offer high specificity and selectiveness [42].
3.2.1.1. Bacterially derived proteins. Bacterially derived proteins
possess high affinity towards antibodies and, thus, are dominant
tool for antibody purification. Among these, protein A, G and L
are the most commonly used ligands for purification of full-
length antibodies. Protein A was isolated from the cell wall of Sta-
phylococcus aureus and consists of five IgG-binding domains (desig-
nated E, D, A, B and C) [36]. Protein A binds to the Fc region of IgG
at the junction between its CH2 and CH3 domains [38] and the
heavy chain variable region between CDR2 and CDR3 of VH3 family
[43]. Protein G was isolated from C and G groups of Streptococcus
spp. and binds strongly to the Fc region of IgG and with low affinity
to the CH1 domain of Fab region [41,44]. In addition to IgG, protein
G also binds to albumin, a2 macroglobulin and kinogen [45] which
leads to reduced purification efficiencies. Both protein A and G
bind to antibodies from different species, with varied affinities
based on their subclasses and source (Table 2). While protein A
binds only to certain species and antibody subclasses; protein G
is broad-spectrum and possess low binding capacity and is less
stable during harsh elution steps. Over the years recombinantly
engineered forms of protein A and G, overcoming their drawbacks,
were developed. For example, a genetically engineered protein A
and G fusion protein was developed, combining the IgG binding
domains of both protein A and G. This fusion protein binds to all
the human IgG subclasses (Table 2) [46–48].
Protein L, isolated from Peptostreptococcus magnus, is another
bacterial protein that is useful for antibody purification. Protein L
has nanomolar affinity towards kappa (k) light chains of variable
region of antibody. Though, it binds only to j1, j3, and j4 subclass
of the antibody light chains, not recognizing j2 and k subgroups
and thus, has the ability to bind to the different antibody isotypes
(IgG, IgM, IgY, IgD and IgE) (Table 2) [17].
3.2.1.2. Lectins. Lectins are carbohydrate-binding proteins that bind
specifically and reversibly to the glycosylation sites on the antibod-
ies rendering them useful for antibody purification. Some of the
common lectins used for antibody purification include con-
canavalin A (Con A), wheat germ agglutinin (WGA), mannan-
binding protein (MBPs), and jacalin. Con A, isolated from Canavalia
ensiformis, binds to a-D-mannose and a-D-glucose [49,50], wheat
germ agglutinin (WGA) binds to the sialic acid and molecules con-
taining N-acetyl-D-glucosamine residue [50], MBPs, isolated from
mammalian serum, binds to mannose and N-acetyl-D-
glucosamine residue [51], jacalin, isolated from jackfruit Artocarpus
integrifolia, binds to a-D-galactosyl groups [52]. Lectins are suitable
for IgD, IgA and IgM purifications, which are otherwise difficult to
purify using bacterially derived ligands.
3.2.1.3. Antigens and anti-antibodies as ligands. Antigen-specific
antibodies can be purified using whole antigen or specific peptides
as ligands. Alternatively, ligands mimicking structure or sequence
of the antigens can also be utilized. This approach leads to the
selection of antibodies with varying affinities, though requiring
further optimization of elution conditions based on the affinity of
antibodies to be purified [53]. Similarly, anti-antibodies also offer
high-specificity options for antibody purification. The constant
domains of antibodies are used as potential targets for this process.
However, antibodies have stability issues under harsh pH and high
salt concentrations [54], generally required for elution, leading to
leeching of antibodies from column. Single-domain antibodies
(sdAb) are especially useful for antibody purification as they have
small size, high affinity and stability and a three-dimensional
structure that enable binding to novel epitopes [55]. Additionally,
sdAb are easily produced in large scale using microbial systems
and can be tweaked for specificity and affinity depending on the
type of target. However, sdAb are monomeric, thus, limited in their
binding capacity (owing to their monovalency) during the purifica-
tion process.
Biospecific ligands, owing to their specificity and selectivity,
have been used extensively for purification of antibodies; however,
they also have various disadvantages associated with them, mak-
ing them less attractive. Their manufacturing and processing is
costly and laborious [56]. Other drawbacks associated with
biospecific ligands include low stability, lot-to-lot variability, high
risk of contamination and toxicity, ligand leakage, difficulty with
sterilization, low binding capacities, limited life cycles and low
scale-up potential [56,57]. Additionally, they are difficult to immo-
bilize in proper orientation owing to their large size. Consequently,
alternative ligands have regularly been researched to overcome
these limitations.
3.2.2. Pseudobiospecific ligands
Pseudobiospecific ligands, a group alternative ligands, exploit
intrinsic properties of the immunoglobulins as they are developed
on the basis of multiple non-covalent forces involved in the inter-
action of immunoglobulins with affinity ligands. These ligands are
promising candidates as they are cheaper, robust, less toxic, struc-
turally well defined, and highly stable as they meet the standards
of good manufacturing protocols (GMPs) involving harsh sanita-
tion or sterilization conditions [58,59]. Their affinity is relatively
lower than the biospecific ligands, but it is sufficient enough to
ensure specificity and selectivity towards target antibody [60].
Some of the most common pseudobiospecific ligands used in the
purification of antibodies include thiophilic [61,62], hydroxyap-
atite [63], L-histidine [64–66], chelating metal ions [67,68] and
mixed mode ligands [69,70]. The development of pseudobiospeci-
fic ligands for the purification of antibodies begun in early nineties
and since then several products, such as Thiosorb, T-gel, HA-
Ultrogel, and MEP HyperCel were commercialized and used
extensively.
3.2.2.1. Thiophilic adsorption chromatography (TAC). TAC is based on
the specific propensity of the immunoglobulins to bind with the
thioether-substituted organic sulfone compounds, in a reaction
termed as thiophilic interaction [71]. TAC is based on the utiliza-
tion of a synthetic pseudo-affinity matrix, termed thiophilic gel
(T-gel) [72]. T-gels are the most commonly used adsorbent, and
adsorption depends just on the presence of a kosmotropic salt.
 S. Arora et al. / Methods 116 (2017) 84–94
T-gels are synthesized by a reaction between divinylsulfone and 2-
mercaptoethanol and the chemical formula for an immobilized
ligand can be depicted as agarose-CH2-CH2-SO2-CH2-CH2-SCH2-
CH2OH [73], indicative of the presence of a linear ligand with
two sulfur atoms. Sulfur atoms provide specificity for selective
binding of the immunoglobulins under high salt concentrations.
Due to their small size and high charge density, phosphates and
sulfates are the most commonly used salts in the buffer system
applied in TAC. Composition of the buffer system is determined
empirically, depending on the free energy of hydrogen bonding
or on the basis of Hofmeister series, so as to avoid protein precip-
itation in the presence of high salt concentrations [71]. Bound
immunoglobulins can be eluted by lowering salt concentration,
however, salt concentration for the elution of different antibody
formats require optimization to achieve highest recovery of the
antibody in its purest form. Boschetti has reviewed a range of thio-
philic ligands elsewhere [62].
The simplicity of control over the adsorption behavior as a func-
tion of salt concentration, chemical stability of the matrix, high
protein binding capacity, and affinity toward several classes of
antibodies has made TAC a method of choice in several applica-
tions [17]. Thiophilic chromatography has found application in
the purification of IgG from cows [74], mouse hybridoma [74,75],
IgG from human and rabbit [76]; IgY from egg yolk [77]; F(ab0 )2
[78]; and Fab fragment [61], and scFvs [79] from E. coli.
Besides salt-dependent thiophilic ligands, development of the
salt-independent mercaptoheterocyclic ligands to capture IgG with
higher adsorption capacity were reported [73]. For instance, Qian
et al. [80] synthesized thiophilic magnetic polymer beads which
were found to have strong specificity against IgG in a salt-
independent manner. Magnetic beads were functionalized by using
a heterocyclic ligand of 2-mercaptonicotinic acid (MNic). Beads
were prepared by microsuspension polymerization with divinyl-
benzene (DVB) and vinyl acetate (VAc). By using these beads
authors were able to purify antibodies, with >94% purity and
>99% bioactivity, from human serum in batch mode and under
physiological conditions, with the assistance of magnetic decanta-
tion [80].
3.2.2.2. Hydroxyapatite chromatography (HAC). Hydroxyapatite
(HA), (Ca5(PO4)3OH)2, is a naturally occurring compound, and its
synthetic forms are available both as calcium and phosphate com-
pounds and as an alternative fluorapatite (FA), (Ca5(PO4)3F)2. HA is
used as porous ceramic microsphere to generate mixed-mode
resin. HA mediates its interaction through (i) P-site, comprising
of a triplet of phosphate groups wherein each phosphate with a
pair of negatively charged oxygen residues interacts with amino
groups of proteins, similar to cation exchangers; (ii) C-site, a dou-
blet of positively charged calcium residues that mediates calcium
chelation of protein through carboxyl clusters, and calcium coordi-
nation on DNA through phosphoryl groups in a metal-affinity type
of mechanism. The P-site and C-site bindings are cooperative;
exhibiting positive cooperation with antibodies and negative coop-
eration with other solutes. Traditional elution strategies involve
use of phosphate [81–85] or sodium chloride gradients [81–83];
however, recently sulfate and borate gradients were also employed
[83]. Application of HAC in the one-step purification of IgG was
reported to be equally effective as protein A and protein G
[81,82,86–88]. HAC method was also adopted to support IgA
[63,89,90] and IgM [89,91–93] capture, with capacities ranging
from 10–20 mg/mL and yields with 90% purity. HAC’s application
in removal of antibody aggregate during large-scale production
for industrial setup has been highlighted [83,94,95]. In these
applications, polyethylene glycol (PEG) is recommended. PEG is
proposed to enhance aggregate removal as it imposes secondary
size selectivity on HA [95]. PEG addition is suggested to be most
89
effective when used along with chloride gradients as it helps in
better removal of host cell proteins, leached protein A, DNA, endo-
toxin, and virus [95].
3.2.2.3. Hydrophobic charge-induction chromatography (HCIC). The
HCIC method is based on the usage of dual-mode ionizable
hydrophobic ligands in a pH-dependent manner [96]. As protein
binding occurs in almost neutral conditions, this method represent
a near physiological condition for antibody purification. Both thio-
philic adsorption chromatography and hydroxyapatite chromatog-
raphy, require high salt concentrations; whereas, HCIC only
requires variation in pH as the modification of pH of the mobile
phase induces a net ionic charge on the ligand and the adsorbed
antibody, leading to the electrostatic repulsion-induced desorp-
tion. The 4-mercapto-ethyl-pyridine (MEP) attached via a
hydrophobic spacer arm to the hydrophilic matrix is a dual-mode
ligand of choice for selective adsorption of antibodies. Presence
of pyridine ring on MEP provides an added advantage in the sepa-
ration of immunoglobulins [97,98]. When sample is applied with-
out adjustment of ionic strength or pH, the IgG dynamic binding
capacity ranges from 25 to 35 mg/mL at pH 7–8 (physiological
pH). MEP, which is non-charged structure in neutral conditions
(pKa 4.8), binds efficiently with antibodies; and under acidic pH
(4.0–4.5) of the mobile phase, a net positive charge is imparted
on both the ligand and bound target molecules, resulting in the
desorption of the antibody owing to the repulsion forces between
the sorbent and positively charged antibody [99]. Due to simplicity
of the method, purification and concentration effects are achieved
in a single step [100]. Chemical properties of MEP render it inert,
thus, not interacting with impurities of the sample; enhancing
the washing-off of impurities in the flow-through [101]. Taken
together, HCIC is an efficient method of purifying antibodies from
samples obtained from diverse sources, such as animal sera, ascites
fluid, and cell culture supernatant [84,102–104].
3.2.2.4. Immobilized metal affinity chromatography (IMAC). In IMAC,
a metal ion functions as an affinity ligand which is bound to a
chelating agent attached to a support. The adsorption of proteins
is based on the interaction between electron donor groups located
on the protein surface and immobilized metal ions [105]. Amino
acid moieties such as histidine, cysteine, serine, glutamate, aspar-
tate, and tryptophan present on the outer surface of a biomolecule,
contain electron donor atoms either in their amino terminus or in
their side chains and, therefore, have higher affinity towards
transitional metal ions such as Co2+, Cu2+, Fe2+, Ni2+, and Zn2+
[106–111]. The ligands for IMAC are designed to exploit this very
property of biomolecules in general, and antibodies in particular,
wherein ligands are coupled to a covalently linked chelating com-
pound and spacer group attached to the matrix [112]. Chelating
compounds vary from bidentate (a-amino phenylalanine tetra-
zole) [113], tridentate (iminodiacetic acid, IDA), tetradentate
(nitrilotriacetic acid, NTA), to pentadentate (tris (carboxymethyl)
ethylene diamine) compounds, based on the number of occupied
coordination bonds.
Several reports indicated use of IMAC for the purification of
antibodies from different species and subclasses [114–123]. Anti-
bodies purification up to a purity level exceeding 90% in a single
step can be achieved using IMAC [124,67]. This efficiency emanates
from the cluster of histidine residues present at the junction of the
second and third constant domains of the heavy chain of an anti-
body [67,115,125]. These histidine moieties endow the antibodies
with high affinity towards immobilized metal ions. According to an
estimate, copper-immobilized IDA can bound up to 14–16 mg/mL
of polyclonal or monoclonal antibodies [68]. IMAC has been uti-
lized in the purification of IgG from different sources e.g. human
[115], humanized murine IgG [125], and goat IgG [126].
90 S. Arora et al. / Methods 116 (2017) 84–94
IMAC has been the method of choice for the purification of
recombinant proteins, especially under denaturing conditions,
e.g. inclusion bodies in E. coli expression system, which are nuclear
or cytoplasmic aggregates of protein and require high concentra-
tions of chaotropic salts for solubilization. The recombinant anti-
bodies could be synthesized so as to have a histidine tag at
amino- or carboxy-terminal of the protein to enhance millimolar
affinity towards metal ions. Following chromatographic step, the
fusion tag can be efficiently cleaved and removed by proteases
[127]. This approach has extensively been utilized in the purifica-
tion of Fab fragments [128] and scFvs [5,129,130]. Depending on
the antibody properties, following purification, proper folding can
be achieved through a combination of different approaches [131].
Guo et al. used an on-column refolding system to obtain functional
protein from inclusion bodies by fusing single chain Fv (scFv) anti-
body fragment with interferon-c inducible protein 10 (IP10), a che-
mokine inducing chemotaxis of activated T- and NK-cells [132].
Overall, IMAC is superior to protein A or G methods owing to
the robustness of system in terms of IMAC matrices, lower cost,
and mild elution of proteins with salts [133]. However, proteins
without affinity to metal ions are not applicable to IMAC. Addition-
ally, buffers containing chelating agents, such as EDTA, ammonium
salts, can strip metal ions from resin, posing limitations to IMAC
mediated purifications.
3.2.2.5. Boronate affinity chromatography (BAC). The BAC method
utilizes boronic acid or its derivative as ligands [134]. Boronic acids
form a stable cyclic ester in alkaline aqueous solution with a com-
pound containing cis-diol groups. Dissociation of boronic-esters is
achieved by changing medium to an acidic pH. BAC utilizes pres-
ence of glycans in the Fc region of the antibody for purification
[135,136]. Recently, Liu et al. [137] prepared a restricted access
boronate affinity porous monolith, as a mimic of protein A for
the specific capture of antibodies. This biomimetic was based on
using steric hindrance of the porous monolith and using boronic
acid as specific ligand. BAC holds promise owing to its low cost,
high stability, and fast elution kinetics based on changing pH
condition.
3.2.3. Synthetic ligands
Advancement in high-throughput lead generation and screen-
ing by high-resolution methods such as X-ray crystallography
and nuclear magnetic resonance (NMR) has helped in the develop-
ment of compounds that can overcome weaknesses associated
with natural ligands, such as fragility, and extreme costs in produc-
tion and procedure optimizations. These new generation com-
pounds are low in molecular weight and more environment
friendly due to increased options of their recycling. These com-
pounds are generated either by using a template, based on ratio-
nale design considering its structure and functional aspect or
using combinatorial approach to identify a template from a library;
or a combination of both the approaches by generating a library
based on a template [138]. The product generated by this approach
could be further modified to obtain desired specificity and affinity
and, thus, minimizing efforts in process optimization. The products
generated could be divided into two broad categories based on the
starting material viz. peptidyl and non-peptidyl ligands.
3.2.3.1. Peptidyl ligand. Amalgamate of different combinations of
amino acids in form of a small peptide used in the purification of
its cognate protein are termed as peptidyl ligands. Yang et al.
[139] screened a hexamer peptide beads library composed of his-
tidine at amino-terminus and positively charged amino acids at
carboxy-terminus flanking the aromatic amino acids. This
approach led to the identification of a peptide, His-Trp-Arg-Gly-
Trp-Val, which can purify IgG subclasses from human, cow, mouse,
goat, and rabbit. Another screening of a synthetic tetrapeptide
library targeted against anti-granulocyte macrophage-colony stim-
ulating factor (GMCSF) mAb from mouse ascetic fluid resulted in
the identification of Ala-Pro-Ala-Arg peptide [140]. Similarly, Fas-
sina et al. [141] screened a synthetic multimeric peptide library
for purification of IgG from human serum.
Besides screening peptide library, a group of ligands were syn-
thesized as biomimetic ligands to extract properties of natural
ligands. Examples of combinatorial approach in rational design of
ligand came from two different studies [142,143] which synthe-
sized ligand mimicking protein sequence of the IgG-Fcc receptor,
implicated in the binding of proteins to IgG-Fc region for the purifi-
cation of mouse and human IgG. Sugita et al. [142] generated an
octameric peptide library using spot-synthesized peptide array
and identified two peptide sequences by amino acid substitution
assay. Similarly, Gong et al. [143] fine-tuned Fc-III peptide, identi-
fied from a phage display screening with a cyclic peptide library
[144], and synthesized Fc-II-4C peptide containing two disulfide
bonds, thus, displaying higher stability and affinity against IgG
from various species. Several other groups had made attempts to
identify small ligands mimicking protein A [145–149].
3.2.3.2. Non-peptidyl ligands. Non-peptidyl ligands, also known as
non-peptidic ligands (NPL), were initially prepared following
exploration of potential binding site on the target protein. These
ligands were developed to overcome weakness of synthetic-
peptide ligands, which had vulnerable peptide bonds that are fis-
sile and subject to degradation. The basic element of NPL were
developed based on chemistries involving peptoids, protease-
resistant peptides synthesized by placing replacement at the nitro-
gen atom rather than at the a-carbon, or by incorporating ethyl
moiety between the two carbon atoms, thus, assigning novel sec-
ondary structures to polypeptides [150,151]. Conversely, non-
peptidic low molecular weight compounds were chemically syn-
thesized with high affinity and specificity for IgG. These de novo
synthesized and optimized molecules had very high affinity, dura-
bility and capacity, with very low cost for compound production.
Examples of NPL include dichloropyridine (AvidAL) [152], thio-
philic gel (divinylsulfone activated agarose) and its derivatives
coupled with hetero-aromatic ligands [62,153], and histidyl-
Sepharose matrix [66,154]. Several other small ligands were
designed based on the properties of textile dyes as biomimetic
ligands, by following NPL approach [155]. Among them, chlorotri-
azine based dyes led to the synthesis of de novo designs of several
such compounds, e.g. ligand 22/8 synthesized from artificial pro-
tein A (ApA) [156,157] and cyanuric chloride [158]. These
triazine-based compounds were further modified and made avail-
able commercially as MAbsorbent A1P and A2P. Another group of
NPL compounds derived from dyes, based on their ability to bind
proteins selectively and in a reversible manner, include, Cibacron
Blue FG-3A and mono- or di-chloro-triazine (developed by cou-
pling azo, anthraquinone, and phatalocyanin choromophores con-
jugated with a chlorotriazine ring) [156,157]. The derivation of
these synthetic ligands opened up numerous possibilities of using
combinatorial strategies to develop low cost and stable ligands for
the purification of antibodies.
3.2.4. Affinity tags
Affinity tags are short polypeptide sequences, proteins/protein
domains or enzymes, appended as fusion partners with the target
proteins. These tags have high specificity to different ligands and,
thus, they find considerable use in purification of recombinant
antibody formats owing to the associated high yields and purity
obtained in just a few steps [159,160]. The cDNA of the desired
tag is incorporated either onto the N- or C-terminus of the anti-
body expressing gene, so that the tag peptide/protein is expressed
 S. Arora et al. / Methods 116 (2017) 84–94
Table 3
Major contaminants in different sources of antibodies.
Source Major contaminants
Serum Serum proteins, albumin, transferrin, a2-
macroglobulin, haptoglobulin, ceruloplasmin
Ascitic fluid Lipids, albumin, transferrin, lipoproteins,
endogenous IgG, other host proteins
Hybridoma Serum proteins, phenol red, water, albumin,
transferrin, a2-macroglobulin, haptoglobulin,
ceruloplasmin, bovine IgG, viruses
Bacterial expression of Proteins from the host bacterial cells and
recombinant antibodies growth media
Reproduced with permission from [15].
in the same open reading frame as the antibody, resulting in the
expression of antibodies fused to the tag [160]. Specific ligands
are then used to purify these antibodies fused to the tag. If
required, the fused tag can be removed to generate purified anti-
bodies [160].
Various affinity tags are reported with their own sets of
strengths and weaknesses [16,161–164]. An ideal affinity tag will
have the advantage of not disturbing antibody structure and func-
tion. Additionally, it should be easily removable from the final pro-
duct, if needed. However, affinity tags might alter affinity and
specificity, lead to antibody aggregations, cause incorrect folding,
are prone to protease degradation and can be difficult to cleave-
off [165–167]. Polyhistidine tag is the most commonly used tag
for antibody purification. Other tags utilized for antibody purifica-
tion include FLAG, c-myc, hemagglutinin (HA) and avidin tags. At
times, more than one tag, i.e. combinatorial tagging, offering extra
benefits such as enhanced protein solubility and improved purifi-
cation, can be used [16,160,163].
4. Affinity chromatography: practical considerations
Affinity purification of antibodies typically involves 4 steps; i)
preparing the sample to load on the affinity column; ii) loading
and incubation of the sample with the ligand to promote binding;
iii) washing away the non-bound components from the column;
and iv) eluting (recovering) the bound antibodies from the immo-
bilized ligand to collect purified antibodies. Some practical consid-
erations associated with these steps are briefly discussed below.
For detailed protocols please refer [15].
4.1. Sample preparation
Preparation of starting material is the first step in an affinity
purification procedure. In case of working with mAbs, it is impor-
tant to identify their isotype to select ligand for efficient purifica-
tion. Both ELISA- and lateral flow-based tests are available for
this purpose. Sample characterization to verify if the antibodies
are expressed extracellularly or intracellularly should be carried
out to determine the type of extraction and clarification proce-
dures to be used. Simple extraction procedures involve osmotic
shock, French press or sonication to facilitate the release of anti-
bodies extracellularly. If the antibodies are expressed as aggre-
gates, disruption by harsh conditions, using denaturing agents
such as urea and guanidine hydrochloride, can be used.
Sources of antibodies (ascetic fluid, hybridoma supernatant and
microbial culture) contain a bulk of other proteins (Table 3) which
may interfere with the purification procedure. These contaminants
can be removed by sample clarification techniques using centrifu-
gation or filtration. Precipitation, using caprylic acid or ammonium
sulfate, is another effective way of clarifying the starting material
along with providing enrichment by isolation or removal of one
91
or more soluble components of a homogeneous mixture by form-
ing a precipitate. Precipitation procedure can be used to reduce
the bulk of the material and can either be used to precipitate anti-
bodies (positive) or other proteins (negative); end product being
antibody-enriched sample.
4.2. Binding
Once the sample is prepared, it is added to the column to bind
with the ligand. Based on the affinity of antibody to the ligand, spe-
cial considerations, such as slow flow rates and longer incubation
of samples with the ligands, are often required when purifying
low affinity antibodies. Each ligand has its own specific optimum
binding properties such as temperature, pH, etc. It is necessary to
carefully consider these conditions for binding buffer preparation
to promote maximum binding between the antibody and the
ligand. Binding can also be promoted by altering salt concentration
if hydrophobic or ionic interactions are involved. In case of proteins
containing disulfide bonds, addition of reducing agents such as b-
mercaptoethanol or dithiothreitol (DTT) can be helpful.
4.3. Washing
Once the antibody is allowed to interact with ligand, the col-
umn is treated with adequate washing buffer to remove all the
non-specific components (unbound or weakly-bound) sticking to
the matrix or ligand. Non-specific binding is generally contributed
by weak ionic interactions or hydrophobic interactions, which are
usually of low binding strength compared to specific interactions.
Binding of non-specific components can be easily dealt by addition
of salt (NaCl, MgCl2), by altering the pH and/or addition of deter-
gents (Tween-20, Triton X-100) to the buffers. Sometimes addition
of blocking agents such as bovine serum albumin or low quantities
of mimicking agents can be helpful.
4.4. Elution
Elution can be specific or non-specific depending on the
method. Specific elution is carried out by adding a mimicking agent
which can either compete for binding with the ligand or the anti-
body, thus, displacing antibody from the ligand causing specific
elution. Specific elution is a gentle, but slow method. This type of
elution requires optimization of the concentration of competitor
a priori to ensure complete displacement of the antibody from
ligand. For example, application of imidazole for elution in anti-
body purification using Ni-NTA IMAC.
Elution performed by changing solvent conditions (pH, ionic
strength and polarity) is referred to as non-specific elution. Non-
specific elution is faster compared to specific elution, however,
due to being harsh, it can cause ligand leaching; limiting the num-
ber of times the column can be used. Elution under low pH is the
most commonly used method among non-specific elution. How-
ever, like other proteins, antibodies are also very sensitive to pH
and using very low pH can result in their aggregation. Elution at
low temperature or by using additives such as salts (NaCl, MgCl2
or LiCl), guanidine hydrochloride, urea or sometimes organic sol-
vents can be considered to enhance elution.
5. Conclusions and future perspectives
Downstream processing is the final stage of antibody produc-
tion that not only determines the quality of the end product but
also massively impacts the commercial costs. There are many steps
such as purification, quality assessment, virus testing and efficacy
testing that an antibody has to go through before it makes way
92 S. Arora et al. / Methods 116 (2017) 84–94
to the market. Antibody purification serves as a limiting factor in
the antibody production [168] and has been the bottleneck when
it comes to cost reduction [169,170], resulting in increasing the
overall costs of the antibody-related products.
Advances in commercial antibody market and the growing need
for antibody-based therapeutic applications has augmented the
pressure on industries to reduce the cost and time of deliverables
to market. Last five years have witnessed a tremendous surge in
the FDA approved antibody-based therapeutics [4], making anti-
bodies the most promising and fastest growing niches within the
biopharmaceuticals sector [171]. But forlornly, there is no such
growth evident in the development of downstream processing
methodologies, which is lagging way behind, even after the mark-
edly higher requirements for robust and scalable large scale purifi-
cation processes [172]. The downstream processing alone accounts
for 50–80% of the total manufacturing costs of the therapeutics
antibodies [53,173]. It is noteworthy that development of
antibody-based therapeutics requires purified antibodies at
numerous additional stages and not just at the final stage of their
production [174], which expectedly leads to remarkably high sell-
ing costs of these therapeutics.
Over the years, protein A-based purification of antibodies has
been the stalwart technology for industrial antibody purification
[174,175], as majority of commercial antibodies have been either
monoclonal or polyclonal [176]. It is unmistakable apparent that
it will require a big leap to move from tested and established chro-
matography methods that have been the mainstay of antibody
purification for many decades. Consequently, a lot of recent work
has been directed at improving protein A for stability and binding
capacity [177,178]. However, the entry of recombinant antibodies
in the market is surely going to dent this dominance, though not
in near future. Additionally, there are augmented efforts concen-
trated at increasing the throughput of antibody purification, pro-
viding additional benefits of robustness, automation, and time-
and cost-effectiveness [172,174,179,180]. However, these trials
are still in their infancy and have a lot of challenges to address
[172,179] before they find extensive implementation.
Conflict of interest
The authors declare no conflict of interest.
References
[1] S.E. Salmon, B.A. Smith, J. Clin. Invest. 49 (1970) 1114–1121.
[2] B.L. Larson, H.L. Heary Jr., J.E. Devery, J. Dairy Sci. 63 (1980) 665–671.
[3] I. Correia, mAbs 2 (2010) 221–232.
[4] B.V. Ayyar, S. Arora, R. O’Kennedy, Trends Pharmacol. Sci. 37 (2016) 1009–
1028.
[5] B.V. Ayyar, S. Hearty, R. O’Kennedy, Anal. Biochem. 407 (2010) 165–171.
[6] K. Welbeck, P. Leonard, N. Gilmartin, B. Byrne, C. Viguier, S. Arora, R.
O’Kennedy, J. Immunol. Methods 364 (2011) 14–20.
[7] V. Saxena, C.K. Lai, T.C. Chao, K.S. Jeng, M.M. Lai, J. Virol. 86 (2012) 4139–4150.
[8] J.A.G.L. van Buggenum, J.P. Gerlach, S. Eising, L. Schoonen, R.A.P.M. van Eijl, S.
E.J. Tanis, M. Hogeweg, N.C. Hubner, J.C. van Hest, K.M. Bonger, K.W. Mulder,
Sci. Rep. 6 (2016) 22675.
[9] S.N. Rylova, L. Del Pozzo, C. Klingeberg, R. Tönnesmann, A.L. Illert, P.T. Meyer,
H.R. Maecke, J.P. Holland, J. Nucl. Med. 57 (2016) 96–102.
[10] F. Bootz, B. Ziffels, D. Neri, Inflamm. Bowel Dis. 22 (2016) 2098–2105.
[11] H.J. Kim, M.H. Jeong, H.J. Park, W.C. Kim, J.E. Kim, Food Chem. 196 (2016)
1144–1149.
[12] L. Xue, T. Hickling, R. Song, J. Nowak, B. Rup, Clin. Exp. Immunol. 183 (2016)
102–113.
[13] B.V. Ayyar, K.R. Aoki, M.Z. Atassi, Infect. Immun. 83 (2015) 1465–1476.
[14] P. Delfani, L. Dexlin Mellby, M. Nordström, A. Holmér, M. Ohlsson, C.A.K.
Borrebaeck, C. Wingren, PLoS One 11 (2016) e0159138.
[15] S. Arora, B.V. Ayyar, R. O’Kennedy, in: N.E. Labrou (Ed.), Protein Downstream
Processing: Design, Development and Application of High and Low-
Resolution Methods, Humana Press, Totowa, NJ, 2014, pp. 497–516.
[16] B.V. Ayyar, S. Arora, C. Murphy, R. O’Kennedy, Methods 56 (2012) 116–129.
[17] A.C.A. Roque, C.S.O. Silva, M.Â. Taipa, J. Chromatogr. A 1160 (2007) 44–55.
[18] R.R. Walters, Anal. Chem. 57 (1985) 1099A–1101A. 1102A-1106A passim.
[19] D.S. Hage, J.A. Anguizola, C. Bi, R. Li, R. Matsuda, E. Papastavros, E. Pfaunmiller,
J. Vargas, X. Zheng, J. Pharm. Biomed. Anal. 69 (2012) 93–105.
[20] E. Starkenstein, Biochem. Z. 24 (1910) 210–218.
[21] L. Ambard, Bull. Soc. Chim. Biol. 3 (1921) 51–65.
[22] O. Holmbergh, Biochem. Z. 258 (1933) 134–140.
[23] Y. Tokuoka, J. Agric. Chem. Soc. Jpn. 13 (1937) 586–594.
[24] D. Hockenhull, D. Herbert, Biochem. J. 39 (1945) 102–106.
[25] H. Lineweaver, R. Jang, E. Jansen, Arch. Biochem. 20 (1949) 137–152.
[26] J.H. Northrop, J. Gen. Physiol. 17 (1934) 359–363.
[27] N.H. Grant, K.C. Robbins, Arch. Biochem. Biophys. 66 (1957) 396–403.
[28] K. Landsteiner, J. Van der Scheer, J. Exp. Med. 63 (1936) 325–339.
[29] D.H. Campbell, E. Luescher, L.S. Lerman, Proc. Natl. Acad. Sci. U.S.A. 37 (1951)
575–578.
[30] G. Manecke, K.-E. Gillert, Naturwissenschaften 42 (1955) 212–213.
[31] G.B. Sutherland, D.H. Campbell, J. Immunol. 80 (1958) 294–298.
[32] S. Hjerten, Biochim. Biophys. Acta 79 (1964) 393–398.
[33] R. Axen, J. Porath, S. Ernback, Nature 214 (1967) 1302–1304.
[34] J.H. Luong, W.H. Scouten, Curr. Protoc. Protein Sci. (2008). Chapter 9, Unit 9.3.
[35] P.L. Kronick, G.L. Campbell, K. Joseph, Science 200 (1978) 1074–1076.
[36] I. Safarik, M. Safarikova, Biomagn. Res. Technol. 2 (2004) 7.
[37] D. Wilkinson, Scientist 14 (2000) 25–28.
[38] R.S. Sista, A.E. Eckhardt, V. Srinivasan, M.G. Pollack, S. Palanki, V.K. Pamula,
Lab. Chip 8 (2008) 2188–2196.
[39] M. Koneracka, P. Kopčanský, M. Timko, C. Ramchand, A. De Sequeira, M.
Trevan, J. Mol. Catal. B Enzym. 18 (2002) 13–18.
[40] M. Teeters, S. Conrardy, B. Thomas, T. Root, E. Lightfoot, J. Chromatogr. A
(2003) 165–173.
[41] H. Sun, B. Ge, S. Liu, H. Chen, J. Sep. Sci. 31 (2008) 1201–1206.
[42] Z. Liu, P.V. Gurgel, R.G. Carbonell, Biotechnol. Prog. 29 (2013) 91–98.
[43] C. Boi, J. Chromatogr. B 848 (2007) 19–27.
[44] S.P. Chambers, Drug Discov. Today 7 (2002) 759–765.
[45] P.J. Haney, C. Draveling, W. Durski, K. Romanowich, M. Walid Qoronfleh,
Protein Expr. Purif. 28 (2003) 270–279.
[46] J.J. Marchalonis, J.L. Atwell, J.W. Goding, Immunology 34 (1978) 97–103.
[47] S. Kabir, Immunol. Invest. 31 (2002) 263–278.
[48] O. Corey, H. David, B. Raychelle, B. Min, W. Chunling, K. Elizabeth, Bioaffinity
chromatography, in: D.S. Hage, J. Cazes (Eds.), Handbook of Affinity
Chromatography, 2nd ed., CRC Press, 2005, pp. 101–126.
[49] K. Kwack, J. Biochem. Mol. Biol. 33 (2000) 188–192.
[50] T. Ishihara, T. Kadoya, H. Yoshida, T. Tamada, S. Yamamoto, J. Chromatogr. A
1093 (2005) 126–138.
[51] A.C.A. Roque, C.R. Lowe, M.Â. Taipa, Biotechnol. Prog. 20 (2004) 639–654.
[52] K. Hagiwara, D. Collet-Cassart, K. Kunihiko, J.-P. Vaerman, Mol. Immunol. 25
(1988) 69–83.
[53] B. Chackerian, L. Briglio, P.S. Albert, D.R. Lowy, J.T. Schiller, J. Virol. 78 (2004)
4037–4047.
[54] A. Wörn, A. Plückthun, J. Mol. Biol. 305 (2001) 989–1010.
[55] A.L. Horenstein, I. Durelli, F. Malavasi, Purification of clinical-grade
monoclonal antibodies by chromatographic methods, in: C.M. Smales, D.C.
James (Eds.), Therapeutic Proteins, Humana Press, 2005, pp. 191–208.
[56] S. Sheng, F. Kong, Pharm. Biol. 50 (2012) 1038–1044.
[57] J.N. Arnold, R.A. Dwek, P.M. Rudd, R.B. Sim, Immunol. Lett. 106 (2006) 103–
110.
[58] J. Champagne, C. Delattre, C. Shanthi, B. Satheesh, L. Duverneuil, M.A.
Vijayalakshmi, Chromatographia 65 (2007) 639–648.
[59] R.R. Prasanna, A.S. Kamalanathan, M.A. Vijayalakshmi, J. Mol. Recognit. 28
(2015) 129–141.
[60] C.R. Lowe, J.C. Pearson, Methods Enzymol. 104 (1984) 97–113.
[61] M. Fiedler, A. Skerra, Protein Expr. Purif. 17 (1999) 421–427.
[62] E. Boschetti, J. Biochem, Biophys. Methods 49 (2001) 361–389.
[63] E. Luellau, U. von Stockar, S. Vogt, R. Freitag, J. Chromatogr. A 796 (1998) 165–
175.
[64] S.M. Bueno, K. Haupt, M.A. Vijayalakshmi, J. Chromatogr. B Biomed. Appl. 667
(1995) 57–67.
[65] K. Haupt, S.M. Bueno, M.A. Vijayalakshmi, J. Chromatogr. B Biomed. Appl. 674
(1995) 13–21.
[66] Q. Luo, H. Zou, Q. Zhang, X. Xiao, J. Ni, Biotechnol. Bioeng. 80 (2002) 481–489.
[67] D. Todorova-Balvay, O. Pitiot, M. Bourhim, T. Srikrishnan, M. Vijayalakshmi, J.
Chromatogr. B Analyt. Technol. Biomed. Life Sci. 808 (2004) 57–62.
[68] R.R. Prasanna, M.A. Vijayalakshmi, J. Chromatogr. A 1217 (2010) 3660–3667.
[69] V. Brenac Brochier, A. Schapman, P. Santambien, L. Britsch, J. Chromatogr. A
1177 (2008) 226–233.
[70] S.S. Ranjini, D. Bimal, A.P. Dhivya, M.A. Vijayalakshmi, J. Chromatogr. B Analyt.
Technol. Biomed. Life Sci. 878 (2010) 1031–1037.
[71] J. Porath, M. Belew, Trends Biotechnol. 5 (1987) 225–229.
[72] J. Porath, F. Maisano, M. Belew, FEBS Lett. 185 (1985) 306–310.
[73] H. Lakhiari, D. Muller, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 818
(2005) 53–59.
[74] P. Konecny, R.J. Brown, W.H. Scouten, J. Chromatogr. A 673 (1994) 45–53.
[75] U.B. Finger, W. Brummer, E. Knieps, J. Thommes, M.R. Kula, J. Chromatogr. B
Biomed. Appl. 675 (1996) 197–204.
[76] A. Lihme, P.M. Heegaard, Anal. Biochem. 192 (1991) 64–69.
[77] P. Hansen, J.A. Scoble, B. Hanson, N.J. Hoogenraad, J. Immunol. Methods 215
(1998) 1–7.
[78] G.K. Yurov, G.L. Neugodova, O.A. Verkhovsky, B.S. Naroditsky, J. Immunol.
Methods 177 (1994) 29–33.
 S. Arora et al. / Methods 116 (2017) 84–94
[79] K. Trisler, L.L. Looger, V. Sharma, M. Baker, D.E. Benson, S. Trauger, P.G.
Schultz, V.V. Smider, J. Biol. Chem. 282 (2007) 26344–26353.
[80] H. Qian, C. Li, Z. Lin, Y. Zhang, Colloids Surf. B Biointerfaces 75 (2010) 342–
348.
[81] R. Giovannini, R. Freitag, Biotechnol. Bioeng. 73 (2001) 522–529.
[82] S. Schubert, R. Freitag, J. Chromatogr. A 1142 (2007) 106–113.
[83] P. Gagnon, Nat. Biotechnol. 25 (2009) 287–293.
[84] L. Guerrier, I. Flayeux, E. Boschetti, J. Chromatogr. B. Biomed. Sci. Appl. 755
(2001) 37–46.
[85] A. Jungbauer, R. Hahn, K. Deinhofer, P. Luo, Biotechnol. Bioeng. 87 (2004)
364–375.
[86] L.H. Stanker, M. Vanderlaan, H. Juarez-Salinas, J. Immunol. Methods 76 (1985)
157–169.
[87] A. Jungbauer, C. Tauer, M. Reiter, M. Purtscher, E. Wenisch, F. Steindl, A.
Buchacher, H. Katinger, J. Chromatogr. 476 (1989) 257–268.
[88] C. Poiesi, A. Tamanini, S. Ghielmi, A. Albertini, J. Chromatogr. 465 (1989) 101–
111.
[89] K. Aoyama, J. Chiba, J. Immunol. Methods 162 (1993) 201–210.
[90] H. Leibl, R. Tomasits, H.M. Wolf, M.M. Eibl, J.W. Mannhalter, J. Chromatogr. B
Biomed. Appl. 678 (1996) 173–180.
[91] G. Coppola, J. Underwood, G. Cartwright, M.T. Hearn, J. Chromatogr. 476
(1989) 269–290.
[92] A.J. Henniker, K.F. Bradstock, Biomed. Chromatogr. 7 (1993) 121–125.
[93] E. Fishbein, D. Alarcon-Segovia, J. Immunol. Methods 33 (1980) 93–99.
[94] P. Gagnon, K. Beam, Curr. Pharm. Biotechnol. 10 (2009) 440–446.
[95] P. Gagnon, J. Immunol. Methods 336 (2008) 222–228.
[96] S.C. Burton, D.R. Harding, J. Chromatogr. A 814 (1998) 71–81.
[97] S. Oscarsson, J. Porath, J. Chromatogr. 499 (1990) 235–247.
[98] K.L. Knudsen, M.B. Hansen, L.R. Henriksen, B.K. Andersen, A. Lihme, Anal.
Biochem. 201 (1992) 170–177.
[99] G. Zhao, Y. Sun, J. Chromatogr. A 1165 (2007) 109–115.
[100] W. Schwart, D. Judd, M. Wysocki, L. Guerrier, E. Birck-Wilson, E. Boschetti, J.
Chromatogr. A 908 (2001) 251–263.
[101] E. Boschetti, Trends Biotechnol. 20 (2002) 333–337.
[102] R.Z. Wang, D.Q. Lin, H.F. Tong, H.L. Lu, S.J. Yao, J. Chromatogr. B Analyt.
Technol. Biomed. Life Sci. 936 (2013) 33–41.
[103] L. Guerrier, P. Girot, W. Schwartz, E. Boschetti, Bioseparation 9 (2000) 211–
221.
[104] A. Schwarz, F. Kohen, M. Wilchek, J. Chromatogr. B Biomed. Appl. 664 (1995)
83–88.
[105] J. Porath, J. Carlsson, I. Olsson, G. Belfrage, Nature 258 (1975) 598–599.
[106] P.G. Taylor, C. Martinez-Torres, E.L. Romano, M. Layrisse, Am. J. Clin. Nutr. 43
(1986) 68–71.
[107] J.H. Swain, L.B. Tabatabai, M.B. Reddy, J. Nutr. 132 (2002) 245–251.
[108] S. Storcksdieck genannt Bonsmann, R.F. Hurrell, J. Food Sci. 72 (2007) S019–
S029.
[109] J. Porath, Protein Expr. Purif. 3 (1992) 263–281.
[110] E. Sulkowski, BioEssays 10 (1989) 170–175.
[111] F.H. Arnold, Biotechnology N.Y. 9 (1991) 151–156.
[112] F. Helfferich, Nature 189 (1961) 1001–1002.
[113] G. Lei, L. Liu, X. Xiong, Y. Wei, X. Zheng, J. Sep. Sci. 31 (2008) 3002–3008.
[114] S. Vançan, E.A. Miranda, S.M.A. Bueno, Process Biochem. 37 (2002) 573–579.
[115] J. Porath, B. Olin, Biochemistry 22 (1983) 1621–1630.
[116] G. Serpa, E.F.P. Augusto, W.M.S.C. Tamashiro, M.B. Ribeiro, E.A. Miranda, S.M.
A. Bueno, J. Chromatogr. B 816 (2005) 259–268.
[117] D. Das, T.M. Allen, M.R. Suresh, Protein Expr. Purif. 39 (2005) 199–208.
[118] T. Zimmerman, C. Petit Frère, M. Satzger, M. Raba, M. Weisbach, K. Döhn, A.
Popp, M. Donzeau, J. Immunol. Methods 314 (2006) 67–73.
[119] S. Martins, A. Karmali, J. Andrade, M.L. Serralheiro, Mol. Biotechnol. 33 (2006)
103–113.
[120] D. Platis, J. Drossard, R. Fischer, J.K.C. Ma, N.E. Labrou, J. Chromatogr. A 1211
(2008) 80–89.
[121] M.B. Ribeiro, M. Vijayalakshmi, D. Todorova-Balvay, S.M.A. Bueno, J.
Chromatogr. B 861 (2008) 64–73.
[122] S. Jain, M.N. Gupta, Biotechnol. Appl. Biochem. 39 (2004) 319–322.
[123] K. Ramessar, T. Rademacher, M. Sack, J. Stadlmann, D. Platis, G. Stiegler, N.
Labrou, F. Altmann, J. Ma, E. Stoger, T. Capell, P. Christou, Proc. Natl. Acad. Sci.
U.S.A. 105 (2008) 3727–3732.
[124] R.C. Cheung, J.H. Wong, T.B. Ng, Appl. Microbiol. Biotechnol. 96 (2012) 1411–
1420.
[125] J.E. Hale, D.E. Beidler, Anal. Biochem. 222 (1994) 29–33.
[126] V. Boden, J.J. Winzerling, M. Vijayalakshmi, J. Porath, J. Immunol. Methods
181 (1995) 225–232.
[127] P.A. Walker, L.E. Leong, P.W. Ng, S.H. Tan, S. Waller, D. Murphy, A.G. Porter,
Biotechnology N.Y. 12 (1994) 601–605.
[128] Y. Zhao, L. Gutshall, H. Jiang, A. Baker, E. Beil, G. Obmolova, J. Carton, S.
Taudte, B. Amegadzie, Protein Expr. Purif. 67 (2009) 182–189.
[129] B. McDonnell, S. Hearty, W.J.J. Finlay, R. O’Kennedy, Anal. Biochem. 410
(2011) 1–6.
[130] S. Sakamoto, F. Taura, R. Tsuchihashi, W. Putalun, J. Kinjo, H. Tanaka, S.
Morimoto, Hybridoma 29 (2010) 481–488.
[131] R. Vincentelli, S. Canaan, V. Campanacci, C. Valencia, D. Maurin, F. Frassinetti,
L. Scappucini-Calvo, Y. Bourne, C. Cambillau, C. Bignon, Protein Sci. 13 (2004)
2782–2792.
[132] J.Q. Guo, Q.M. Li, J.Y. Zhou, G.P. Zhang, Y.Y. Yang, G.X. Xing, D. Zhao, S.Y. You,
C.Y. Zhang, Protein Expr. Purif. 45 (2006) 168–174.
93
[133] G. Serpa, E.F. Augusto, W.M. Tamashiro, M.B. Ribeiro, E.A. Miranda, S.M.
Bueno, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 816 (2005) 259–
268.
[134] P. Muhammad, D. Li, Z. Liu, Encyclopedia of Analytical Chemistry, John Wiley
& Sons Ltd, Chichester, 2006, pp. 1–18.
[135] L. Borlido, A.M. Azevedo, A.C. Roque, M.R. Aires-Barros, J. Chromatogr. A 1218
(2011) 7821–7827.
[136] A.M. Azevedo, A.G. Gomes, L. Borlido, I.F. Santos, D.M. Prazeres, M.R. Aires-
Barros, J. Mol. Recognit. 23 (2010) 569–576.
[137] Y. Liu, Y. Lu, Z. Liu, Chem. Sci. 3 (2012) 1467–1471.
[138] Y.D. Clonis, J. Chromatogr. A 1101 (2006) 1–24.
[139] H. Yang, P.V. Gurgel, R.G. Carbonell, J. Pept. Res. 66 (2005) 120–137.
[140] S.A. Camperi, N.B. Iannucci, G.J. Albanesi, M. Oggero Eberhardt, M.
Etcheverrigaray, A. Messeguer, F. Albericio, O. Cascone, Biotechnol. Lett. 25
(2003) 1545–1548.
[141] G. Fassina, A. Verdoliva, G. Palombo, M. Ruvo, G. Cassani, J. Mol. Recognit. 11
(1998) 128–133.
[142] T. Sugita, M. Katayama, M. Okochi, R. Kato, T. Ichihara, H. Honda, Biochem.
Eng. J. 79 (2013) 33–40.
[143] Y. Gong, L. Zhang, J. Li, S. Feng, H. Deng, Bioconjug. Chem. 27 (2016) 1569–
1573.
[144] W.L. DeLano, M.H. Ultsch, A.M. de Vos, J.A. Wells, Science 287 (2000) 1279–
1283.
[145] B. D’Agostino, P. Bellofiore, T. De Martino, C. Punzo, V. Rivieccio, A. Verdoliva,
J. Immunol. Methods 333 (2008) 126–138.
[146] H. Yang, P.V. Gurgel, R.G. Carbonell, J. Chromatogr. A 1216 (2009) 910–918.
[147] K. Sakamoto, Y. Ito, T. Hatanaka, P.B. Soni, T. Mori, K. Sugimura, J. Biol. Chem.
284 (2009) 9986–9993.
[148] L.N. Lund, P.-E. Gustavsson, R. Michael, J. Lindgren, L. Nørskov-Lauritsen, M.
Lund, G. Houen, A. Staby, P.M. St. Hilaire, J. Chromatogr. A 1225 (2012) 158–
167.
[149] S. Menegatti, A.D. Naik, P.V. Gurgel, R.G. Carbonell, J. Sep. Sci. 35 (2012)
3139–3148.
[150] T. Clackson, J.A. Wells, Trends Biotechnol. 12 (1994) 173–184.
[151] H.U. Saragovi, M.I. Greene, R.A. Chrusciel, M. Kahn, Biotechnology N.Y. 10
(1992) 773–778.
[152] T.T. Ngo, N. Khatter, J. Chromatogr. A 597 (1992) 101–109.
[153] F. Anspach, D. Petsch, W. Deckwer, Bioseparation 6 (1996) 165–184.
[154] A. El-kak, M.A. Vijayalakshmi, J. Chromatogr. B Biomed. Sci. Appl. 570 (1991)
29–41.
[155] C.R. Lowe, S.J. Burton, N.P. Burton, W.K. Alderton, J.M. Pitts, J.A. Thomas,
Trends Biotechnol. 10 (1992) 442–448.
[156] R. Li, V. Dowd, D.J. Stewart, S.J. Burton, C.R. Lowe, Nat. Biotechnol. 16 (1998)
190–195.
[157] S.F. Teng, K. Sproule, A. Husain, C.R. Lowe, J. Chromatogr. B Biomed. Sci. Appl.
740 (2000) 1–15.
[158] S.F. Teng, K. Sproule, A. Hussain, C.R. Lowe, J. Mol. Recognit. 12 (1999) 67–75.
[159] D.W. Wood, Curr. Opin. Struct. Biol. 26 (2014) 54–61.
[160] X. Zhao, G. Li, S. Liang, J. Anal. Methods Chem. 2013 (2013) 581093.
[161] U.D. Palanisamy, A. Hussain, S. Iqbal, K. Sproule, C.R. Lowe, J. Mol. Recognit.
12 (1999) 57–66.
[162] A.C.A. Roque, M.Â. Taipa, C.R. Lowe, J. Chromatogr. A 1064 (2005) 157–167.
[163] G. Gupta, C.R. Lowe, J. Mol. Recognit. 17 (2004) 218–235.
[164] D. Dong, H. Liu, Q. Xiao, R. Li, J. Chromatogr. B 870 (2008) 51–54.
[165] M.L. Isaksen, K. Fitzgerald, in: R. Kontermann, S. Dübel (Eds.), Antibody
Engineering, Springer, Berlin Heidelberg, Berlin, Heidelberg, 2001, pp. 282–
291.
[166] A. Goel, D. Colcher, J.-S. Koo, B.J.M. Booth, G. Pavlinkova, S.K. Batra, Biochim.
Biophys. Acta 1523 (2000) 13–20.
[167] H. Schmeisser, P. Kontsek, D. Esposito, W. Gillette, G. Schreiber, K.C. Zoon, J.
Interferon Cytokine Res. 26 (2006) 866–876.
[168] F. Li, N. Vijayasankaran, A. Shen, R. Kiss, A. Amanullah, mAbs 2 (2010) 466–
479.
[169] P. Gronemeyer, R. Ditz, J. Strube, Bioengineering 1 (2014) 188–212.
[170] G.E. Khoury, B. Khogeer, C. Chen, K.T. Ng, S.I. Jacob, C.R. Lowe, Pharm.
Bioprocess. 3 (2015) 139–152.
[171] D.M. Ecker, S.D. Jones, H.L. Levine, mAbs 7 (2015) 9–14.
[172] J. Welsh, Pharm. Bioprocess. 3 (2015) 1–3.
[173] C.R. Lowe, Curr. Opin. Chem. Biol. 5 (2001) 248–256.
[174] P.M. Schmidt, M. Abdo, R.E. Butcher, M.-Y. Yap, P.D. Scotney, M.L. Ramunno,
G. Martin-Roussety, C. Owczarek, M.P. Hardy, C.-G. Chen, L.J. Fabri, J.
Chromatogr. A 1455 (2016) 9–19.
[175] G.R. Bolton, K.K. Mehta, Biotechnol. Prog. 32 (2016) 1193–1202.
[176] M. Hornsby, M. Paduch, S. Miersch, A. Sääf, T. Matsuguchi, B. Lee, K.
Wypisniak, A. Doak, D. King, S. Usatyuk, K. Perry, V. Lu, W. Thomas, J. Luke, J.
Goodman, R.J. Hoey, D. Lai, C. Griffin, Z. Li, F.J. Vizeacoumar, D. Dong, E.
Campbell, S. Anderson, N. Zhong, S. Gräslund, S. Koide, J. Moffat, S. Sidhu, A.
Kossiakoff, J. Wells, Mol. Cell. Proteomics 14 (2015) 2833–2847.
[177] R. Godawat, K. Brower, S. Jain, K. Konstantinov, F. Riske, V. Warikoo,
Biotechnol. J. 7 (2012) 1496–1508.
[178] J. Pollock, G. Bolton, J. Coffman, S.V. Ho, D.G. Bracewell, S.S. Farid, J.
Chromatogr. A 1284 (2013) 17–27.
[179] V. Girard, N.-J. Hilbold, C.K.S. Ng, L. Pegon, W. Chahim, F. Rousset, V.
Monchois, J. Biotechnol. 213 (2015) 65–73.
[180] J. Spooner, T. Wilkinson, B.P. Kemp, Pharm. Bioprocess. 3 (2015) 411–424.
[181] J. Turková, Journal of Chromatography Library, Elsevier, 1993.
94 S. Arora et al. / Methods 116 (2017) 84–94

[182] J. Štamberg, Sep. Purif. Methods 17 (1988) 155–183.
[183] Z. Jedliński, J. Paprotny, J. Polym. Sci. A Polym. Chem. 5 (1967) 2957–2960.
[184] A. Verdoliva, F. Pannone, M. Rossi, S. Catello, V. Manfredi, J. Immunol.
Methods 271 (2002) 77–88.
[185] E. Klein, E. Eichholz, D.H. Yeager, J. Membr. Sci. 90 (1994) 69–80.
[186] C. Staak, F. Salchow, P.H. Clausen, E. Luge, J. Immunol. Methods 194 (1996)
141.
[187] O. Wichterle, D. Lim, Nature 185 (1960) 117–118.
[188] M.W.H. Roberts, C.M. Ongkudon, G.M. Forde, M.K. Danquah, J. Sep. Sci. 32
(2009) 2485–2494.
[189] K.K. Unger, Journal of Chromatography Library, Elsevier, 1979.
[190] F. Xi, J. Wu, J. Chromatogr. A 1057 (2004) 41–47.
[191] G.C. Serafica, G. Belfort, J. Pimbley, Biotechnol. Bioeng. 43 (1994) 21–36.
[192] M. Schuster, E. Wasserbauer, A. Neubauer, A. Jungbauer, Bioseparation 9
(2000) 259–268.
[193] Y. Li, H.G. Spencer, J. Biotechnol. 26 (1992) 203–211.