Cover: Protein Chromatography, Second Edition by Giorgio Carta, Alois Jungbauer

Protein Chromatography

Process Development and Scale-Up

 

Giorgio Carta

Alois Jungbauer

 

Second Edition

 

 

 

 

 

 

Wiley Logo

All books published by Wiley-VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.

Library of Congress Card No.:
applied for

British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.

Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at <http://dnb.d-nb.de>.

© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany

All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law.

Print ISBN: 978-3-527-34666-0
ePDF ISBN: 978-3-527-82403-8
ePub ISBN: 978-3-527-82402-1
oBook ISBN: 978-3-527-82404-5

Preface

Chromatography is an essential tool for the production of purified proteins, in general, and biopharmaceuticals, in particular. The method allows handling the complex mixtures encountered in this industry using readily available stationary phases and equipment suitable for large-scale sanitary operation. Moreover, its practice as a process purification tool is recognized by regulatory agencies so that chromatography is an integral part of essentially all licensed biopharmaceutical processes. Regulatory agencies are increasingly requiring in-depth process understanding. As a result, chemists, engineers, and life scientists working in this field need to become familiar with the theory and practice of process chromatography.

In the past, the design and scale-up of chromatography units for biopharmaceutical purification was largely empirical. As a result, optimum designs often remained elusive. The last 10 years have, however, witnessed increasing emphasis on the use of models based on a mechanistic understanding of both physical and biomolecular properties. On the one hand, engineers, while possessing a strong foundation in transport phenomena and unit operations, often do not have adequate understanding of biomolecular properties. On the other hand, biochemists and biologists often have little understanding of the key scale-up relationships and models needed for optimum design. Our goal for this book and for the courses and workshops that we have taught over the years has been to help create “new phenotypes” – engineers who understand the biophysical properties of proteins and other biomacromolecules and can implement this understanding in the bioprocess setting and life scientists who understand transport phenomena and engineering models and who can apply these tools to the design of process units.

This book is based not only on our own experience in downstream processing but also on the interactions that we have had over the past 20 years with more than 1000 industrial scientists and engineers who have attended our practical protein chromatography courses and workshops. During this time, we have collected many practical examples and acquired a good feeling about the main difficulties that are encountered by both students and beginning practitioners in understanding protein chromatography from both theoretical and practical viewpoints. In order to meet this goal, following the first edition of this book published in 2010, we begin the second edition with a chapter on the biochemical and biophysical properties of proteins and their contaminants. We focus on the properties that are relevant for chromatographic separations such as size, surface charge and hydrophobicity, solution viscosity, and diffusivity and on how to preserve biological activity while addressing regulatory and environmental impact aspects. In Chapter 2, we provide a succinct, general introduction to chromatography identifying the key factors important for design and scale-up. This allows the reader who is not familiar with chromatography to put in proper context the various issues discussed in Chapters 3–10. Chapter 3 addresses the chemistry and structure of many different stationary phases, while Chapter 4 discusses laboratory and process columns and equipment. Both of these chapters are limited in scope to providing the reader with some familiarity with commercially available materials and equipment. No attempt was made to provide a comprehensive coverage, in large part because the field is rapidly expanding and new media and equipment are constantly being introduced. Mechanical design of equipment is also not considered in detail, as rarely, it is a task for separation scientists and engineers in the biopharmaceutical production setting. Chapters 5–9 are structured to acquaint the reader with the theory and models needed to design and scale-up chromatography units. Emphasis is placed on phenomenological models whose parameters can be determined through suitable experimental studies. Many specific numerical examples are provided to illustrate the application of these models to the analysis of laboratory data and to the prediction of column performance. A great deal of emphasis is placed on describing transport in the stationary phase, as adsorption kinetics is often limiting in industrial applications of biochromatography. Thus, Chapter 6 provides a detailed coverage of mass transfer effects and their relationship to the structure of the stationary phase. Chapter 7 explores the dynamic behavior of chromatography columns establishing a link between equilibrium properties, described in Chapter 5, and column behavior. Chapter 8 discusses how equilibrium and rate factors combine to determine column performance and how to model band broadening for practical conditions. Chapter 9 is focused on gradient elution chromatography. We chose to devote a separate chapter to this mode of operation, as, in our experience, it is often frequently less well understood despite its major importance in the practice of biochromatography. Finally, Chapter 10 is designed to bring everything together by providing guidance for the optimum design of process units. This chapter covers both the design of conventional batch chromatography units as well as the design of continuous or semicontinuous multicolumn systems for continuous biomanufacturing.

Despite the structural similarity with the first, beside many updates, the second edition provides a lot of new material. The second edition provides an expanded analysis of surface heterogeneity of molecular properties including consideration of domain-specific chemistry in multidomain proteins as well as new information about the mechanical stability of proteins. Regulatory aspects, including references to ICH guidelines, quality by design (QbD) concepts, viral clearance considerations, and biosimilars are included as well as references to the process mass intensity. Many new examples dealing with data for actual monoclonal antibodies (mAbs) are provided to replace or complement examples based on model proteins that were used throughout the first edition. The new mAb examples provide a more realistic view of current and future industrial practice as these molecules are much larger and much more complex than commonly used model proteins. Data for such molecules were not widely available for the first edition but are available now. Their inclusion, we think, provides much greater practical value demonstrating the application of engineering principles and chromatography theories in a much more realistic context. Many updates were made to stationary phases and their interactions with antibodies, including an expanded coverage of new protein A resins that have become available during the past 10 years. Greater insights into how proteins interact with chromatographic surfaces, protein unfolding and on-column aggregation, and much more detailed views of the physical structures of established and new chromatographic matrices are also included. Improved chromatographic workstations and new chromatography column hardware, including single-use columns and systems, have also become widely available since the first edition and an overview of examples of these technologies is included in Chapter 4. Advances made in the measurement and theoretical description of protein adsorption isotherms have also been made and are covered in Chapter 5. In this context, high-throughput screening methods have become very popular and provide the means of obtaining adsorption data that, combined with mechanistic descriptions of transport and column dynamics, allow predictions of column behaviors with previously unmet accuracy. Advances in understanding and modeling competitive binding in multicomponent protein mixtures are now covered with reference to actual mAb systems, including detailed examples on the separation of mAb charge variants and mAb monomer–dimer mixtures at high loadings. In this context, Chapter 6 now includes an expanded analysis of multicomponent batch adsorption, Chapter 7 includes expanded analyses of single and multicomponent breakthrough dynamics for mAbs, and Chapter 8 includes a coverage of the description of combined dynamics and kinetics effect for mAbs in chromatographic columns. The treatment of gradient elution in Chapter 9 has been expanded by providing a more general version of the Yamamoto model, by introducing the concept and calculation of the isoresolution line, which provides the means to understand the optimization of residence time and gradient duration, by illustrating the concept of buffer design for pH-gradient elution, and by discussing the effects of overloading on gradient elution chromatography of proteins as well as its description based on hybrid modeling approaches. Finally, column design, covered in Chapter 10, has been expanded by introducing the Pareto concept of balancing productivity and capacity utilization/buffer consumption and discussing the impact of multicolumn and continuous processing on this balance. Design charts and application examples are now provided for the analysis of periodic countercurrent systems for protein capture.

Similar to the first edition, we feel that the book's main value is not in “de novo” process development – rather, it focuses mainly on how to optimally design and scale up columns for a process whose steps have already been defined. Nevertheless, understanding these concepts will also aid the scientist involved in early process development identify process steps that are scalable and can be efficiently translated from the laboratory to the manufacturing suite. We are convinced that proper application of the theory covered in this book combined with adequate experiments is instrumental for a successful application of chromatography on a large scale. The book also provides extensive references to original literature, textbooks, and books in chromatography, for those seeking greater detail. We have strived to make the notation consistent across the whole book and to check the correctness of the mathematical equations. Notwithstanding these efforts, we strongly suspect that some inconsistencies might still exist. We would be very grateful if you could inform us of any such issue so that they can be remedied.

Finally, we would like to thank our students who, over the years, have helped us develop and teach the laboratory and discussion sessions of our short courses, which have formed the basis for much of this book and which could not have been held without their tremendous efforts and enthusiastic support. In particular, for these efforts, we would like to thank Antonio Ubiera, Timothy Pabst, Emily Schirmer, Jamie Harrington, Melani Stone, Jeremy Siebenmann-Lucas, Theresa Bankston, Yinying Tao, Robert Deitcher, Ernie Perez-Almodovar, Tarl Vetter, Yige Wu, Joe Basconi, Jing Guo, Shaojie Zhang, Mimi Zhu, Jason Reck, Arch Creasy, Yiran Wang, Preston Fuks, Andreas Alberti, Lucas Kimerer, and Joey Roberts at the University of Virginia and Tina Paril, Kerstin Buhlert, Rene Überbacher, Anne Tscheliesnig, Alfred Zoechling, Christine Machold, Anna Christler, Dominik Sauer, Goncalo Silva, Katrin Reiter, Nico Lingg, Nicole Walch, Nikolaus Hammerschmidt, Peter Satzer, Stephan Hinterberger, Susanne Schweiger, Tobias Schneider, Walpurga Kreper, and our colleague Rainer Hahn at BOKU. We also thank all the people who have attended our courses and who have provided very valuable feedback and have shared with us much of their practical experiences.

January 1, 2020

Giorgio Carta
Charlottesville, Virginia, USA

Alois Jungbauer
Vienna, Austria

Nomenclature

a coefficient in dimensionless van Deemter equation (Eqs. (2.4), (2.5), and (8.51)) or isotherm parameter
A coefficient in van Deemter equation (Eq. (8.50)), m
Aexternal surface area outside particles per unit column volume, m2/m3
Ai combined equilibrium parameter for retention in IEC (Eq. (9.24)), RPC (Eq. (9.39)), or HIC (Eq. (9.44)), variable units
Ainternal surface area inside particles per unit column volume, m2/m3
As asymmetry factor (Eq. (8.12))
b coefficient in dimensionless van Deemter equation (Eqs. (2.4), (2.5), and (8.51))
B coefficient in van Deemter equation (Eq. (8.50)), m2/s
B0 hydraulic permeability (=ηLuP), m2
c solute concentration in pore liquid, kg/m3, or coefficient in dimensionless van Deemter equation (Eqs. (2.4), (2.5), and (8.51)).
images solute concentration in pore liquid averaged over the particle volume, kg/m3
C solute concentration in mobile phase, kg/m3, or coefficient in van Deemter equation (Eq. (8.50)), s
Cf peak compression factor in linear gradient elution (Eqs. (9.15), (9.29), and (9.42))
CF solute concentration in feed, kg/m3
CM mobile-phase modulator (or modifier) concentration in IEC and HIC, mol/m3
C0 initial solute concentration, kg/m3
Cs mobile-phase solute concentration at particle surface, kg/m3
C* mobile-phase solute concentration in equilibrium with the stationary phase, kg/m3
CV number of bed volumes of mobile phase passed through the column
dc column diameter, m
dp particle diameter, m
dpore pore diameter, m
D0 molecular diffusivity (or diffusion coefficient) in the mobile phase, m2/s
DL axial dispersion coefficient (see Eqs. (7.1) and (8.47)), m2/s
De effective pore diffusivity (Eq. (6.9)), m2/s
De, b effective diffusivity in the mobile phase (see Eq. (8.47)), m2/s
Ds effective adsorbed phase or surface diffusivity (see Eq. (6.14)), m2/s
images convection-enhanced effective intraparticle diffusivity (Eq. (6.20)), m2/s
DBC dynamic binding capacity or amount of protein held in column at a specified percentage of breakthrough (see Eqs. (2.15), (8.60), and (10.5)), kg/m3
ED eddy diffusivity (Eq. (8.47)), m2/s
EBC equilibrium binding capacity or amount of protein held in column at equilibrium with feed, kg/m3
F fractional approach to equilibrium (images)
Fp ratio of intraparticle and column superficial velocities (see Eqs. (6.15) and (6.18))
h reduced HETP (=H/dp, see Eqs. (2.6) and (8.51))
H height equivalent to a theoretical plate (HETP, see Eqs. (2.3) and (8.4)), m
H(t) unit step function
I ionic strength (Eq. (1.16)), mol/m3
J mass transfer flux, kg/(m2 s) or mol/(m2 s)
J(n, 1) J-function (item B in Table 8.1, Eqs. (8.39) and (8.40))
k rate coefficient for LDF model with adsorbed phase concentration driving force (Eq. (8.30)), s−1
ka second-order adsorption rate constant (see Eqs. (6.22) and (8.63)), m3/(kg s)
kb Boltzmann constant (=1.38 × 10−23 J/K)
kc rate coefficient for LDF model with mobile-phase concentration driving force (Eq. (8.31)), s−1
k retention factor (images or φm for the linear isotherm case, see Eq. (2.10))
images average retention factor (Eq. (2.13))
kf film or boundary layer mass transfer coefficient (see Eqs. (6.1), (6.3)–(6.5)), m/s
K linear isotherm slope or Henry constant (see Eq. (5.5)), m3/kg
KD distribution coefficient in size exclusion chromatography (Eqs. (1.11) and (3.15))
Ke equilibrium constant for ion exchange (Eq. (5.16))
KL Langmuir isotherm equilibrium constant (Eq. (5.4)), m3/kg
length of packed column in SMB separator (see Eq. (10.53)), m
L length of packed chromatographic column or zone length in SMB separator, m
L0 uncompressed column length (see Eq. (10.17)), m
Lcri critical, compressed column length, m
m distribution coefficient (images)
M mobile-phase modulator (or modifier) or amount injected, kg
Mj flow rate ratio in zone j of SMB separator (Eq. (10.50))
Mr molecular mass
n number of transfer units (see Table 8.2)
N number of plates (see Eqs. (2.3) and (8.5))
p switch time for SMB separator (see Eq. (10.53)), s
P productivity (Eq. (10.1)), kg/(m3 s)
ΔP column pressure drop, Pa
q adsorbed solute concentration, kg/m3
qF adsorbed solute concentration in equilibrium with feed, kg/m3
qm monolayer adsorption capacity (e.g. see Eq. (5.4)), kg/m3
qmax maximum adsorption capacity in SD or SMA models (see Eqs. (5.22) and (5.24)), kg/m3
q0 concentration of charged ligands in the stationary phase (see Eq. (5.18)), mol/m3
images adsorbed concentration averaged over particle volume (Eq. (6.29)), kg/m3
images total concentration in stationary phase including amounts adsorbed and held in the pores (Eq. (6.24)), kg/m3
images total concentration in stationary phase averaged over particle volume (Eq. (6.29)), kg/m3
images adsorbed concentration in equilibrium with mobile phase, kg/m3
Q volumetric flow rate, m3/s
r particle radial coordinate, m
rh hydrodynamic radius (Eq. (1.28)), m
rm molecule radius, m
rp particle radius, m
rpore pore radius, m
images volume-averaged particle radius (Eq. (3.16)), m
R separation factor isotherm parameter (Eq. (5.17)); R = 1 for a linear isotherm, R = 0 for a rectangular isotherm
Rs chromatographic resolution (Eqs. (2.9), (10.27), and (10.35))
S sensitivity coefficient for retention in RPC and HIC (Eqs. (3.6), (3.8) and (9.39), (9.44)), or column cross-sectional area, m2
tb breakthrough time (see Figure 8.14), s
tCycle total cycle time (see Figure 10.4), s
tF duration of feed injection, s
tG parameter in EMG function (Eq. (8.14)), s, or duration of gradient, s
tmax time elapsed from injection at peak maximum, s
tR retention time (Eq. (7.22)), s
ts time required for separation (Eq. (2.14)), s
tsh time at which shock front emerges from column, s
T temperature, K
u superficial mobile-phase velocity (Eq. (4.1)), m/s
us adsorbent superficial velocity in SMB separator (see Figure 10.13), m/s
uj superficial mobile-phase velocity in zone j of TMB-equivalent to SMB separator (see Figure 10.13), m/s
images superficial mobile-phase velocity in zone j of actual SMB separator (Eq. (10.54)), m/s
v interstitial velocity of mobile phase (=u/ε, Eq. (4.2)), m/s
v reduced velocity (=vdp/D0, Eqs. (2.7) and (8.52))
vc chromatographic velocity for simple waves (Eq. (7.28)), m/s
vsh shock velocity (Eq. (7.30)), m/s
V liquid-phase volume, m3
Vb fluid volume passed through column at breakthrough, m3
Vc column volume, m3
VF feed volume loaded to column, m3
V0 column extraparticle void volume (εVc), m3
Vp volume of adsorbent particles, m3
VR retention volume, m3
w solubility in solution, kg/m3
w0 solubility in pure water, kg/m3
W baseline width of pulse response peak (see Figure 8.1), s or m3
x column axial coordinate, m
X dimensionless solute concentration in mobile phase (Eq. (7.12))
Y dimensionless solute concentration in stationary phase (Eq. (7.12))
z protein effective charge (Eq. (5.15))

Greek Symbols

α selectivity (images)
β gradient slope (Eq. (9.6)) (mol/m3)/s or (mol/m3)/m3 or safety margin for SMB separator (see Eq. (10.51))
δ stagnant film or boundary layer thickness (Eq. (6.2)), m
images SMB separator parameter (Eq. (10.68))
δ(t) Dirac delta function
Δ peak width at half-peak height (see Figure 8.1), s or m3
ε extraparticle (or external) void fraction (Eq. (4.3))
εp intraparticle (or internal) void fraction (see Figure 2.8)
ε0 extraparticle void fraction of uncompressed bed
εt total column void fraction (Eq. (2.1))
images power input per unit mass in an agitated tank for use in Eqs. (6.6) and (6.7), m2/s3
φ ratio of stationary and mobile-phase volumes in the column (= (1 − ε)/ε)
γ normalized gradient slope (=βL/v = βV0/Q, see Eq. (9.11)), mM
images shear rate, s−1
η mobile-phase viscosity, Pa s
ηE elution recovery yield (see Eq. (10.2))
[η] intrinsic viscosity (Eq. (1.24)), m3/kg
ϕ volume fraction of organic modifier in RPC
λD Debye length (Eq. (3.9)), m
λcri critical bed compression factor (=(L0 − Lcri)/L0)
λm ratio of protein and pore radii (=rm/rpore)
μ0 zeroth moment of pulse response peak (Eq. (8.1)), (kg s)/m3 or kg
μ1 first moment of pulse response peak (Eq. (8.2)), s or m3
ρ density of mobile phase, kg/m3
σ steric hindrance parameter in SMA model (Eq. (5.23)) or standard deviation of pulse response peak (Eq. (8.3)), s or m3
σG parameter in EMG function (Eq. (8.14)), s
τ dimensionless time (=εvt/L, Eq. (7.13)) or shear stress (Eq. (1.22)), Pa
τa time constant for affinity binding, s
τG parameter in EMG function (Eq. (8.14)), s
τp tortuosity factor for intraparticle diffusion (see Eq. (6.9))
τ1 dimensionless time (images), see Eqs. (7.17), (8.32), and (8.62))
ψp hindrance parameter for pore diffusion (Eqs. (6.10) and (6.11))
ζ dimensionless column length (=x/L, Eq. (7.13))

Dimensionless Transport Parameters

Bi Biot number (=rpkf/De)
PeL Peclet number based on column length (=vL/DL)
Pep intraparticle Peclet number (see Eq. (6.21))
Re Reynolds number (=ρudp/η)
Sc Schmidt number (=η/ρD0)
Sh Sherwood number (=kfdp/D0)
St Stanton number (=(1 − ε)kL/us)
nfilm number of transfer units for film mass transfer (=3φkfL/vrp, see Table 8.2)
npore number of transfer units for pore diffusion (images, see Table 8.2)
nsolid number of transfer units for solid diffusion (images, see Table 8.2)