Industrial Purification Strategies for Monoclonal Antibodies

 

Narendra Kumar S¹*, Lingayya Hiremath¹, Praveen Kumar Gupta¹, Ajeet Kumar Srivastava¹, Poornima S⁴

¹Assistant Professor, Department of Biotechnology, R V College of Engineering, Bangalore

²PG-Student, M. Tech Biotechnology, R V College of Engineering, Bangalore-560059

 

ABSTRACT:

The global monoclonal antibody market in 2016 was around USD 85.4 billion and is expected to reach USD 140 billion by end of 2024 i.e., approximately 1.5 fold increase in market rate within 10 years. The therapeutics purification process monoclonal antibody through chromatography technique is expected to grow by 5% from 2011 to 2017, with a prediction of reaching to USD 9 billion by end of 2017. The global market interest is shifting from monoclonal antibody to fragmented antibody; nearly 60 fragmented antibodies have gone into clinical trials as of 2010. In this review, recovery of monoclonal antibody by centrifugation, depth filtration followed by purification by chromatography techniques i.e., capture step by affinity chromatography, intermediate step by anionic chromatography in flow through mode and polishing step by cationic chromatography is reviewed. The production of antigen binding fragment is carried out by enzymatic method followed by affinity chromatography, multimode chromatography and cationic chromatography. Monoclonal antibodies have been developed by these methods are used for targeting range of diseases. Many of such monoclonal antibodies have been approved and many such is still in development pipeline. This review gives an overview on the structure of antibody, production of monoclonal antibody and the downstream purification of monoclonal antibody. The downstream purification covers the basic techniques followed for most of the mAb purification done commercially. The downstream purification steps discussed here include centrifugation, tangential flow filtration, depth filtration and chromatographic techniques.

 

 

 

INTRODUCTION:

Monoclonal Antibody:

‘Mono’ means single, ‘clonal’ means the population that has originated from a single parent and ‘antibody’ is the Y shaped protein found in blood which acts against the antigen. Hence it is defined as the monospecific antibodies generated from a single parent cell. The concept of mAbs started with the development of hybridoma by Georges Kohler and Cesar Milstein in 1984, which produces antibody of monovalent affinity to a single epitope [1].

This started gaining importance in the field of immunology and immuno-oncology treatment. Monoclonal antibody developed with the hybridoma technique where the spleen of the mouse was fused with the B cells lead to the production of murine antibody.

 

Murine antibody developed an immunological side effect that was concerned in the oncological treatment. This made way to the development of humanized antibody in 1988 by Greg winter. This mainly eliminated the undesired reactions concerned with the murine backbone antibody. Antibody is also known as immunoglobulin. There are five classes of immunoglobulins based onvthe heavy chain namely A, D, G, M, E [2]. The immunoglobulins further contain light chain which can either be kappa or lambda subtype. The light chain is generally made of 211-217 amino acids. The light chain is encoded by gene located on chromosome 2 for kappa subtype and chromosome 22 for lambda subtype. Lambda as 4 subtypes λ1, λ2, λ3, λ4. In any healthy individual the ratio of kappa to lambda in serum is 3:1 [3,4].

 

Table1: Classes of immunoglobulin found in serum [2]

Type of Ig	IgM	IgG	IgA	IgE	IgD
Heavy chain	µ(mu)	γ(gamma)	α(alpha)	ε(epsilon)	δ(delta)
Light chain	Kappa or lambda
MW(kDa)	900	150	385	200	180
% In serum	6%	80%	13%	0.002%	1%
Antigen binding sites	10	2	4	2	2
Half life in serum (day)	5	23	6	2	3
Primary function	Primary response to antigen and fixes complements	Enhanced phagocytosis and neutralizes toxins and viruses	Found in secretion found to protect mucous membrane	Involved in allergic reactions	Involved in activation of B cell

 

 

 

STRUCTURE OF IMMUNOGLOBULIN

Immunoglobulin basic structure of immunoglobulin remains the same immaterial of the 5 types. The antibody resembles a Y shaped structure containing 2 identical heavy chain and 2 identical light chains. The heavy chain is made of 450 – 550 amino acids. In this section immunoglobulin G structure is discussed prominently since most of the commercially available monoclonal antibody that is available for therapeutic use generally has IgG backbone.  IgG is made of gamma heavy chains roughly made of 450 amino acids and having molecular weight of 150kDa where the heavy chain alone makes up to 50kDa each and since one antibody is made of two heavy chains therefore 100kDa and remaining is made of light chains 50kDa where each chain is 25kDa.

 

The light chain is disulphide bound to heavy chain by cysteine in the last position at the light chain and fifth cysteine position of the heavy chain. The two heavy chains are linked together at the hinge region by disulfide bond. The number of disulphide bond connecting the heavy chain varies according to the subtype of the IgG. 2 disulphide bonds connects the 2 chain in IgG1 and IgG4, 4 bonds for IgG2 and 11 bond for IgG3 [7].

 

The heavy chain is made of 4 fragments namely 3 constant chain fragments and 1 variable chain ^{fragments namely V}H ^{followed by C}H^{1, C}H^{2, C}H^{3 similarly light chain is made of 2 fragment one constant region and one heavy chain namely V}L ^{followed by C}L^{. The antigen binding site exists in V}H ^{and V}L ^{and this is known as Complementary Determining Regions (CDRs).}

 

^{There are 3 CDRs found on the one arm of the IgG namely CDR1, CDR2 and CDR3. The antigen binding to CDR is through non covalent interactions [5,6]. It is a reversible binding reaction where the strength of binding is determined by affinity constant. Here the antigen, antibody and complex is in molar concentrations to fit in the formula for calculating affinity constant.}

 

Antibody + Antigen  Antigen antibody complex

 

^{The affinity constant changes with antigen and antibody. This is dependent on factors such as pH, solvent and temperature. The affinity constant gives the value which tells us how well the antigen fits into the CDRs of an antibody; this is not connected with number of binding sites of an antibody. The interactions of antigen with an antibody are through non covalent interaction i.e. through Vander Waals interaction, hydrophobic interactions, hydrogen bonding and ionic interactions.}

 

^{Ka= [Antigen antibody complex] / [antigen][antibody]}

 

The therapeutic antibodies developed against many of diseases are generally having IgG backbone and only variations in the CDR region. Hence only the CDR sequences can be _patented.

 

Flow chart1: Monoclonal development process.

 

Downstream Processing of Monoclonal Antibodies:

Monoclonal antibodies are produced by mammalian cells along with various other proteins that required for the physiological mechanisms. These host cell proteins, DNA, lipids and cell debris should be removed from the desired protein of interest. This separation of desired protein from the other contaminants is the primary objective of downstream processing of monoclonal antibodies. The general steps involved in purification of a mAb are cell harvesting by centrifugation, clarification by depth filtration and purification by various chromatographic techniques.

 

Preliminary Recovery Step:

Cell harvesting starts with the removal cells from the media components where the desired protein of interest is present. This is done by simple centrifugation technique. In lab scale different types of centrifuges such as swing bucket, fixed angle centrifuges are used but in commercial scale of production centrifuges such as disc stack centrifuges are employed. This is followed by clarification by depth filtration and this is followed by purification by chromatographic techniques.

 

The preliminary steps for purification includes tangential flow filtration (TFF) where the cell culture fluid flows tangential to the porous membrane of pore size of 0.22µm and this is achieved by applying pressure. This results in removal of insoluble cell mass with the help of external energy. This results in increase cost of production which results in choosing alternate method i.e. centrifugation. There are other cons involved in choosing TFF which is easy fouling of membrane, pressured process results in cell shearing hence increased complexity of downstream steps.

 

CENTRIFUGATION:

Recovery of mAb is performed by centrifugation coupled with depth filtration. The basic operating principle is the separation based on centrifugal force. As the centrifugation starts the dense particles are pulled away from the central axis of rotation by centrifugal force acting away from the axis. The rate of settling is determined by the density of particles, viscosity of surrounding liquid and the centrifugal force acting. The greater the density difference the faster separation occurs.

 

Relative Centrifugal force = Mω²r

 

Where, M is mass of particle, mass of particle, ω is the angular velocity, r is the distance of particle form the axis.

Angular velocity = {2п(RPM)}/ 60

 

Where, RPM is the revolutions per minute.

 

Since the rotors differ according to the manufacturers, relative centrifugal force represents the centrifugal force. The relation between RCF and RPM is given by

 

RCF = 1.118*10^-5r (RPM)²

 

Where, r is the radius of rotor from the centre of axis of rotation.

 

In case of commercial scale centrifugation a disc stack centrifuge is used. Few parameters are considered very important during the start of this process such as the cell density, viability. In industrial scale terms such as clarification efficiency is considered important. Clarification efficiency in case of centrifugation is determined by relative amount of debris in the centrate and in feed.  Clarification efficiency is determined by centrifugation feed rate, centrifugal force applied, operating pressure, bowl geometry and the discharge frequency.

 

This step separates the cell, cell debris, genomic DNA, cell membrane and its anchoring proteins from the external medium containing proteins.

 

Depth Filtration:

Depth filtration is also known as deep bed filtration where separation occurs by size exclusion and physical adsorption on to the membrane. The mechanism of filtration is retaining particles through the medium rather than absolute filtration where the particles retain onto the surface of the filter. The membrane is generally made of cellulose fibers, diatomaceous earth, etc which is positively charged and this tends to retain negatively charged particles such as DNA and endotoxins.

 

The thickness of this membrane is 2-4mm. In lab scale single use filters with low filter area are employed. The pore size varies from 0.1-4µm with two or more distinct layers for lab scale purpose [8]. The filtration pore size, flux, filtration area is optimized at lab scale and scaled up for industrial scale. The clarification efficiency depends on the particle size distribution, area available for filtration. In industrial set up 2 depth filtration systems are arranged consecutively to increase the clarification efficiency. This step separated off the fragmented DNA, endotoxins, negatively charged proteins.

 

 

 

Table2: Few of the commercially available depth filters

Manufacturer	Material of construction	Name of depth filter	Retention range (μm)	Number of filter layer
Pall seitz HP series	Cellulose, diatomaceous earth and resins	PDK7 PDK6 PDK5 PDK4 PDE2	20-4 20-3 20-1 15-0.4 3.5-0.2	2
Cuno Zeta plus	Cellulose, diatomaceous earth, perlite	10SP02A 30SP02A 60SP02A 60ZA05A 90ZA08A	7-1 5-0.8 5-0.65 0.8-0.6 0.2-1	2
Millipore Millistak+	Cellulose, diatomaceous earth	D0HC C0HC	9-0.6 2-0.2	2
Sartorius Sartoclear P	Cellulose, diatomaceous earth, binding matrix	PB1 PB2	11-4 8-1	2

 

Few of the mAb production involves primary step of flocculation or precipitation. This is followed in only few mAb production processes.

 

Chromatographic Techniques:

Chromatography is a technique used for separation of desired product from other components in the medium. Chromatography concept was given in 1901 by Mikhail Tsvet [9] and was popularized by Richard Kuhn, Edgar Lederer in 1930 [10]. In the year 1938 chromatographic separation of particles based on charge was put forth by Harold and Taylor [11]. Invention of two dimensional chromatography by Martin, Gordon and Consden in 1944 lead to development of separation of proteins[12]. This paved way for the modern chromatographic method for purification of mAbs. There are various chromatographic techniques but generally affinity followed by ion exchange chromatography is used for mAb purification.

 

Affinity Chromatography:

It is the simplest form of chromatography where the separation of mAbs occurs based on the affinity of the IgG with protein A antigen. The column resin is made of gel matrix usually agarose matrix on which protein A obtained from Staphylococcus aureus. This protein A had a signal sequence region, IgG binding region and a cell wall anchoring region. The IgG binding region contains 5 homologous IgG binding domains namely E, D, A, B and C. Now-a-days recombinant for of protein A wherein it facilitates an easy and mild elution condition at pH of 4.5 rather than the native pH 3.3 elution.

 

The protein A offers a better separation of IgG from other protein because of its specificity and affinity only to IgG- Fc hence on running a protein A chromatography would result in lowering volume of the load, a very good purity at very less time. The protein A column is made of protein tagged to the matrix hence the column needs to be stored in cold when not in used. There is a huge disadvantage of protein leaking out into the elute of chromatogram.

 

The generally used buffer is phosphate buffer system which is cost effective and operates in the required pH range of IgG. The steps involved in protein A chromatography is equilibration of the column with low conductivity buffer, load application, washing to remove partly bound fragments or proteins with low pH buffer and elution of the desired protein at very low pH of about 3-2.5 pH.

 

The elution is at low pH hence storage of protein in the elution buffer might be harmful for the protein and may affect the function of protein because of which the pH of the eluted sample is pH adjusted to slightly higher pH [13-17].

 

There are other affinity protein resins available such as protein L and protein G. Protein L resins are employed for separation of Fab fragment with light chain being kappa since the protein L has affinity with kappa chain of the IgG.

 

Ion Exchange Chromatography:

The operating principle in this chromatography is the relative adsorption of the charged species to the oppositely charged matrix. The ion exchange resins can be either positively charged i.e., anion exchanger or if the matrix is negatively charged it is the cation exchanger. For any type of ion exchange chromatography buffer selection is important and there are range of buffers available for the same. Buffers available for cation and anioin exchange chromatography are listed below.

 

 

 

 

 

_{Table3: Buffers available for cation exchange chromatography}

pH range	Buffer substance	Concentration (mM)	Counter ion	pKa (25°C)
1.4- 2.4	Maleic acid	20	Na+	1.92
2.6- 3.6	Methyl malonic acid	20	Na+ or Li+	3.07
2.6- 3.6	Citric acid	20	Na+	3.13
3.3- 4.3	Lactic acid	50	Na+	3.86
3.3- 4.3	Formic acid	50	Na+ or Li+	3.75
3.7- 4.7	Succinic acid	50	Na+	4.21
5.1- 6.1	Succinic acid	50	Na+	5.64
4.3- 5.3	Acetic acid	50	Na+or Li+	4.75
5.2- 6.2	Methyl malonic acid	50	Na+ or Li+	5.56
5.6- 6.6	MES	50	Na+or Li+	6.27
6.7- 7.7	Phosphate	50	Na+	7.20
7.0- 8.0	HEPES	50	Na+ or Li+	7.56
7.8- 8.8	BICINE	50	Na+	8.33

 

 

 

Table4: Buffers available for anion exchange chromatography

pH range	Buffer substance	Concentration (mM)	Counter ion	pKa (25°C)
4.3- 5.3	N- Methylpiperazine	20	Cl-	4.75
4.8- 5.8	Piperazine	20	Cl- or HCOO-	5.33
5.5- 6.5	L-Histidine	20	Cl-	6.04
6.0- 7.0	Bis- Tris	20	Cl-	6.48
6.2- 7.2	Bis- Tris propane	20	Cl-	6.65
8.6- 9.6	Bis- Tris propane	20	Cl	9.10
7.3- 8.3	Triethanolamine	20	Cl- orCH3COO-	7.76
7.6- 8.6	Tris	20	Cl-	8.07
8.0- 9.0	N- Methyl- diethanolamine	20	SO42-	8.52
8.0- 9.0	N- Methyl- diethanolamine	50	Cl- or CH3COO-	8.52
8.4- 9.4	Diethanolamine	20(at pH 8.4)	Cl-	8.88
8.4- 9.4	Diethanolamine	50 (at pH 8.8)	Cl-	8.88
8.4- 9.4	Propane 1,3- Diamino	20	Cl-	8.88
9.0- 10.0	Ethanolamine	20	Cl-	9.50
9.2- 10.2	Piperazine	20	Cl-	9.73
10.0-11.0	Propane 1,3- Diamino	20	Cl-	10.55
10.6-11.6	Piperidine	20	Cl-	11.12

 

 

 

The general operating procedure for any ion exchange chromatography is equilibrating the resin, load application, washing with slightly high conductivity and final elution with very high conductivity generally with 1N NaCl of conductivity 86mS/cm.

 

CONCLUSION:

Downstream techniques have to be upgraded regularly to keep up with the new technology, regulatory requirements. They are pointed towards increasing titer yield, increased recovery at each step; with no contamination or any resin issues (leakage, storage problems). There are lots of publications available in terms of mAb purification but future trends point out towards fragmented antibody production and purification which needs to be worked upon.

 

New resins are available in market for each type of chromatography with better binding capacity and operating flow rates. Hence the selection of best resin is also one important perspective which needs focus on. In this paper brief overview on only two types of chromatography i.e. affinity and ion exchange chromatography have been discussed further there are many types available which are not used generally but used only for specific mAb production processes. There are non chromatographic steps available for the same purpose but only time can decide if they prove to be beneficial and cost effective compared to the conventional chromatographic methods.

 

REFERENCE:

1.     Cambrosio A, Keating P. Between fact and technique: the beginnings of hybridoma technology. Journal of the History of Biology 1992; 25 (2): 175–230.

2.     Schroeder, H. W., and Cavacini, L. (2010). Structure and Function of Immunoglobulins. The Journal of Allergy and Clinical Immunology, 125: S41–S52.

3.     Janeway CA Jr, Travers P, Walport M, et al. Immunobiology: The Immune System in Health and Disease. 5th edition. New York: Garland Science; 2001.

4.     Owen, Judith A, Jenni Punt, Sharon A. Stranford, Patricia P. Jones, and Janis Kuby. Kuby Immunology. New York: W.H. Freeman, 2013

5.     Davies DR, Cohen GH. Interactions of protein antigens with antibodies. Proceedings of the National Academy of Sciences of the United States of America. 1996;93(1):7-12.

6.     Gabrielli, Elena et al. “Antibody Complementarity-Determining Regions (CDRs): A Bridge between Adaptive and Innate Immunity.” PloS one, 2009.

7.     Liu H, May K. Disulfide bond structures of IgG molecules: Structural variations, chemical modifications and possible impacts to stability and biological function. mAbs.2012;4(1):17-23.

8.     Liu HF, Ma J, Winter C, Bayer R. Recovery and purification process development for monoclonal antibody production. mAbs. 2010;2(5):480-499. 9. Martin J, Synge RL, “A new form of chromatography employing two liquid phases," J Biochem, 1941; 35:1358 - 1368.

9.     Csaba Horváth, “High-Performance Liquid Chromatography: Advances  and Perspectives”, NY academic press, 1980, chapter 2.

10.   Taylor JB, Swaisgood HE, “A unified partition coefficient theory for chromatography, immobilized  enzyme  kinetics,  and  affinity chromatography” Biotechnol. Bioeng, 1981;23(6):1349–1363.

11.   Kirk PL, Duggan EL, “Biochemical Analysis”, Anal. Chem., 1954;26 (1):163–176.

12.   Richman DD, Cleveland PH, et.al., The binding of staphylococcal protein A by the sera of different animal species, J. Immunol., 1982;128:2300.

13.   Amersham Pharmacia Biotech, Handbook, Antibody Purification, 2000.

14.   Uhlen M, Guss B, et.al., Complete Sequence of the Staphylococcal Gene Encoding Protein A, J. Biol. Chem. 1984;259:1695.

15.   Abrahmsen L, Moks T, et.al., Analysis of signals for secretion in the staphylococcal protein A gene, EMBO J. 1985;4:3901.

16.   Moks T, Abrahmsen L, et.al., Staphylococcal protein A consists of five IgG-binding domainsEur. J. Biochem. 1986;156:637.

 

 

Accepted on 11.08.2017      © RJPT All right reserved

Research J. Pharm. and Tech 2017; 10(10):3561-3566.