CELL DAMAGE MECHANISMS AND STRESS RESPONSE IN ANIMAL CELL CULTURE
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of the Ohio State University
By
Claudia Berdugo, Ch. Eng. M. Sc.
Graduate Program in Chemical and Biomolecular Engineering
*****
The Ohio State University 2010
Dissertation Committee:
Dr. Jeffrey J. Chalmers, Adviser
Dr. Jessica Winter
Dr. Andre Palmer
ABSTRACT
Animal cell culture is a widely used technology for producing recombinant proteins. The
ability to make post translational modifications and secrete the active forms of the protein
into the culture medium represents major advantages over other processes. The growing
market demand for pharmaceuticals has created a need for increased production capacity;
however, achieving productivity gains in both the upstream stage and downstream processes can subject cells to aggressive environments such as those involving
hydrodynamic stresses. Although numerous studies have explored the consequences of
cell damage due to hydrodynamic stress, there has been a lack of understanding of the
mechanism of such damage at a cellular level. Cell damage can also influence biomedical
applications. Cells manipulated in instruments such as diagnosis and analysis devices can
experience hydrodynamic forces.
The level of cell damage is influenced by the hydrodynamic conditions in the bioprocess
or biomedical equipment as well as the cell line sensitivity. To evaluate and compare cell
sensitivity among different cell lines, a flow contraction device, previously designed by
our group was used. Cells were exposed to well defined and controlled hydrodynamic
forces and cell damage was estimated as a function of energy dissipation rate (EDR).
EDR is a scalar value that represents the rate of dissipation of kinetic energy per unit of
ii mass or volume. Using this methodology we found human cell lines highly sensitive to
hydrodynamic forces.
Hydrodynamic evaluations were performed in ten different bioreactor configurations
Impeller Sparger. The best configurations were chosen based on kLa response surface
model for testing in cell culture experiments. The configurations chosen were used to
evaluate the expression of stress proteins under moderate hydrodynamic stress in bioreactors as well as cell cycle profile and its relationship to recombinant protein production. The results suggest that for a clonal cell line evaluated G1 phase of the cell
cycle may be more conducive to producing the recombinant protein. In addition, a
relationship between hydrodynamic stress and expression of stress proteins was observed.
The type of stress protein and the level of expression seem to be dependent on cell type
and differences could even be observed between clones of the same cell line.
Cell damage was also evaluated in a fluorescent activated cell sorter (FACS) models
Vantage and Aria. Cells can be exposed to very high hydrodynamic forces when flowing
through channels and nozzle in the sorting process. Results indicate that not only are cells
damaged in a flow cytometer, but that this damage can vary from cell line to cell line as
well as from specific conditions/type of flow cytometer and flow conditions. In addition,
studies were conducted to evaluate cell growth behavior after stress as well as the effect
of sorting on cell cycle. Extended growth lag phase was observed in cells exposed to
hydrodynamic stress, and the sensitivity of any specific cell line can be a function of the
growth phase of the cell.
iii
Dedicated to my sister
iv
ACKNOWLEDGMENTS
I want to thank God for giving me the courage to accomplish this goal, and reminding me how vast the world is and how little we know.
I would like to thank my advisor, Professor Jeffrey J. Chalmers, for giving me the opportunity to join his research group, for his financial support and his insights in my research. I am also grateful for his support in getting me opportunities to go to the industry and getting involved in training and internship programs.
Special thanks to my committee members Dr. Jessica Winter and Dr. Andre Palmer for their comments, questions and directions during my qualified, candidacy, and defense exams. I want to thank also Dr. Andrea Doseff for her guidance during my candidacy exam and my research.
I would like to show my gratitude to GlaxoSmithKline. It was an honor for me to work at their research facilities at King of Prussia, PA. I learnt a great deal from that experience and I gained perspective into the special requirements of process development in the
v pharmaceutical industry, which helped me later in my research. In particular I would like
to gratefully acknowledge the enthusiastic supervision and friendship of
Dr. Oscar Lara Velasco, his inspiration and efforts to explain things clearly and simply
helped me to take advantage of that great opportunity. I wish to thank Dr. Ilse
Blumentals, a person with great charisma and wisdom, who believe in people and knows
how to bring the best of them. I also thank Dr. Prem Patel for his kindness and support in my internship. Thanks to everyone in the group of Process development.
I am grateful to all my colleagues in Dr. Chalmers group, lab mates in Dr. Yang’s group
and office mates from Dr. Bakshi’s group. I appreciate the company, talks, and
stimulating environment to learn and grow. I also appreciate the support of Paul Green,
Leigh Evrard, David Cade, Angela Benett, Susan Tesfai, Lynn Flanagan and Bill Cory, in
the department of Chemical Engineering, and the technical support in the FACS project
from Bryan, Nicole, Priya, Katrina, Serra, Arup and Adeline.
I am forever in debt to Ruben. I admire your gifted mind and I am thankful to have you as
my mentor. Your teaching and support helped me to defeat my fears when I started this journey, I wouldn’t have made it without your help. Thanks from the bottom of my heart.
It is a pleasure to thank so many friends that helped me in many ways to make it through
this work. Please forgive me if I forget to mention someone but, many precious moments
might escape at 2 am while I am writing this section. Nacho, thanks for being there every
time I needed a friend and thanks for going along with me on crazy ideas, even running
vi 10K under the rain. Oliver, a truly giving person with a big heart, thanks for taking care
of me. Sharing your roof has been like having a brother (and you are a good cook). Elba,
thanks for listening and for having always a word of support. Laura thanks for sharing
with me all the goodness of your heart. My runner mates: Nicole, Elsa and Jake, fun and
healthy moments we shared. Priya: thanks for the treasure of your friendship. Thanks to
all my dear friends for the emotional support, camaraderie, entertainment and caring they provided: Bryan, Katrina, Yadira, Daniel, Alejandra, Courtney, Leo, Adeline, Glenn,
Toño, Pamela, Miryam.
My friends and surrogate family in USA: Charlene, you are a blessing in my life, you made me feel I have family far from my land in a moment where I needed the most.
Sabrina and Shane, thanks for welcoming me in your house and giving me a home. So much love I felt in that family, I am really lucky for I am still there.
With immense joy I want to thank Jeremy. The peace and happiness that I found with you helped me to accomplish this goal. It is easier to bear the challenges because I know you are always at the finish line. Thanks for your constant help, support, and patience. You have drawn a smile in my heart and in my life.
I wish to thank my entire family for their love and support:
Inesita: Gracias por contagiarnos con tu entusiasmo por la vida. Claudia Imelda:
Gracias por estar pendiente de mi, por tus palabras y espiritualidad. Angela Maria: Ha sido hermoso conocerte y tenerte cerca, tu y yo sabemos cuánto REALMENTE cuesta
vii este reto y me siento orgullosa de como lo estás afrontando. Adri siempre muy cerca de
mi corazón por lo que compartimos creciendo juntas. Todos mis tíos y tías en Colombia,
mis primitos y toda mi familia. Sé que han estado pendientes de mi y les agradezco con gran amor desde mi corazón
Mami: Tu constante deseo de superación ha sido mi ejemplo en el logro de esta meta. Tu inmensa capacidad de amor es el símbolo de bondad que jamas he visto en nadie. Papi:
Gracias por todas las enseñanzas de amor a tus hijos y familia, por tu dedicación y esfuerzo. Ricardo: Hermanito este triunfo comenzo contigo, gracias por tu empeño en mostrarnos hasta donde podemos llegar. Mi admiración a mi brillante hermanito. Javi:
Gracias por tu gran corazón, te admiro muchísimo desde lejos por todo lo que has logrado. Tu carisma, dedicación y éxito son ejemplo para todos nosotros. Sandra:
Hermanita eres un hermoso regalo que la vida me dio. Agradezco inmensamente todo tu
apoyo. Sin ti no hubiera sido capaz de superar tantos tropiezos grandes y pequeños
durante estos años de lucha. Tus palabras siempre me iluminaron. Admiro tu tenacidad,
tu integridad y amor a tu familia. Dios te bendiga siempre. Mis cuñis: Felipe y Lina.
Gracias por el amor y cuidado a nuestra familia y sobrinitos, ustedes son dos hermanos
mas. Mis sobrinitos y sobrinita: Dany, Migue, Martín, Tomás y Manuela: Ustedes son la
alegría de mi corazón. A mi Hermosa familia: Los llevo en mi corazón y cada pequeño
logro en mi vida es reflejo de lo que sembramos con esfuerzo en nuestra familia.
A la memoria de Rociito, abuelita y Juan Camilo, vivos en el corazón de quienes los
amamos tanto.
viii
VITA
March 30, 1969….…….………… Born Duitama, Boyacá, Colombia 1995...... Chemical Engineering. Universidad Nacional de Colombia. Bogotá, Colombia. 2000……………………………... Master of Science in Biotechnology. Universidad Nacional Autónoma de México. México. 2000 2004……………………… Research Engineer. ECOPETROL. Colombia
2005 2006……………………… Departamental Fellowship. The Ohio State University 2006 2007……………………… Graduate Research Associate, The Ohio State University 2007 2008……………………… COOP GlaxoSmithKline, King of Prussia, PA
2008 Present…………………….. Graduate Research Associate, The Ohio State University.
PUBLICATIONS
Beltran, L.; Moreno, N.; Berdugo, C.; Zamora, A.; and Buitrago, G. (1998). “Estrategia para el diseno de un medio de cultivo para la fermentacion de B. thuringiensis”. Revista Colombiana de Biotecnologia. 1(1): 28 34
Berdugo, C.; Mena, J.; Acero, J.; and Mogollon, L. (2001). “Increasing the production of Desulfurizing Biocatalyst by means of Fed Batch Culture”. CT&F Ciencia, Tecnologia y Futuro. 1(2): 5 9
Berdugo, C.; Caballero, R. and Godoy, R. D. (2002) “Aqueous organic phase separation by membrane reactors in biodesulfurization reactions”. CT&F Ciencia, Tecnologia y Futuro. 2(3): 97 112
Acero, J.; Berdugo, C.; and Mogollon, L. (2003). “Biodesulfurization process evaluation with a Gordona rubropertinctus strain”. CT&F Ciencia, Tecnologia y Futuro. 2(4): 43 54 ix
Mollet, M.; Godoy Silva, R.; Berdugo, C. and Chalmers, J. J. (2007). Acute Hydrodynamic Forces and Apoptosis: A Complex Question. Biotechnology and Bioengineering. 98 (4): 772 788.
Mollet, M; Godoy Silva, R; Berdugo, C; and Chalmers, J. J. “Computer Simulations of the Energy Dissipation Rate in the Fluorescence Activated Cell Sorter”. Biotechnology and Bioengineering (2008). 100(2): 260 272
x
FIELDS OF STUDY
Major Field: Chemical Engineering
Minor Field: Biotechnology, Biochemical Engineering, Cell Culture.
xi
TABLE OF CONTENTS
Page Abstract…………………………………………………………………………. ii Acknowledgments…………………………………………………………….... v Vita……………………………………………………………………………... ix List of tables……………………………………………………………………. xv List of figures………………………………………………………………….... xvii
Chapters: 1. Introduction……………………………………………………………….. 1 1.1 Animal Cell culture……………………..……………………………..... 1 1.2 Objectives………………………………………………………………. 3 1.2.1 Goal…………………………………………………………………. 3 1.2.2 Methodology………………………………………………………… 3 1.3 Scopes of this study…………………………………………………….. 4 1.3.1 Cell damage in a fluorescent activated cell sorter (FACS) (chapter 3)…………………………………………………………...... 4 1.3.2 Cell damage in a fluorescence activated cell sorter (chapter 4)……... 6 1.3.3 Effect of hydrodynamic conditions on cell cycle, stress proteins and recombinant protein productivity (chapter 5)……………………..… 7 1.4 References……………………………………………………………… 9 2. Literature review: mixing, aeration and hydrodynamics in bioreactors….. 11 2.1 Abstract………………………………………………………………… 11 2.2 Introduction……………………………………………………………... 13 2.3 Aeration………………………………………………………………… 16 2.3.1 surface aeration……………………………………………………… 22 2.3.2 perfluorocarbons…………………………………………………….. 24 2.3.3 oxygen carriers………………………………………………………. 26 2.3.4 Membrane aeration………………………………………………….. 26 2.3.5 Sparging……………………………………………………………... 27 2.3.6 Sparger design………………………………………………………. 28 2.3.7 CO2 accumulation and removal……………………………………... 30 2.4 Mixing………………………………………………………………….. 34 2.4.1 Stirred tank reactors…………………………………………………. 36 2.4.2 Geometry……………………………………………………………. 37 2.4.3 Impeller……………………………………………………………… 37 2.4.4 Baffles……………………………………………………………….. 40 xii 2.4.5 Mixing times and energy dissipation rates………………………….. 40 2.5 The relationship of animal cells to hydrodynamic forces……………… 46 2.5.1 Molecular mechanism involved in the response of animal cells to mechanical forces……………….…………………………………… 48 2.5.2 Cell damage concept………………………………………………… 52 2.5.3 Hydrodynamical cell damage……………………………………….. 54 2.5.4 Quantification of cell damage……………………………………….. 57 2.5.5 Detrimental effects of sparging……………………………………... 63 2.6 Summary……………………………………………………………….. 69 2.7 References……………………………………………………………… 71 3. Cell damage in a fluorescent activated cell sorter……………..…………. 118 3.1 Abstract………………………………………………………………… 119 3.2 Introduction…………………………………………………………….. 120 3.3 Materials and methods…………………………………………………. 123 3.3.1 Cell culture……………..…………………………………………… 123 3.3.2 Single shear stress studies……………………….………………….. 124 3.3.3 Cell sorting…….………………………………………………….… 124 3.3.4 Cell damage analysis……………………………...………………… 125 3.3.5 Cell cycle analysis….…………………………………………….…. 126 3.3.6 Cell arrest in G2 phase…………………………………………..…. 127 3.3.7 Computational Fluid Dynamics (CFD) simulations……..…………. 127 3.4 Results………………………………………………………………….. 130 3.4.1 Single shear stress studies……….…………………………………... 130 3.4.2 Growth after hydrodynamic stress exposure………………………... 132 3.4.3 FACS sorting studies at different cell densities…………………….. 135 3.4.4 FBS protective effect in FACS sorting……………………………… 135 3.4.5 Effect of sorting on cell cycle……………………………………….. 137 3.4.6 FACS Flow Rate Measurements……………………………………. 139 3.4.7 CFD Simulations……………………………………………………. 140 3.5 Discussion………………………………………………………………. 141 3.6 References……………………………………………………………… 145 4. Effect of impeller sparger configurations on mass transfer capabilities and cell culture performance…………………………………. 169 4.1. Abstract…………………………………………………………………. 169 4.2. Introduction ……………………………………………………………. 170 4.3. Materials and methods………………………………………………….. 173 4.3.1 Methodology to evaluate kLa……………………………………….. 173 4.3.2 Experimental design ………………………………………………... 174 4.3.3 Volumetric oxygen mass transfer coefficient calculation kLa……… 175 4.3.4 Power number and Energy Dissipation rate calculation…………….. 177 4.3.5 Cell culture experiments……………………………………………. 178 4.4. Results and discussion…………………………………………………. 179 4.4.1 Estimation of kLa at 2L scale.……………………………………… 179 4.4.2 Data analysis for different configurations……….…………………… 180 4.4.3 Surface Response Analysis…………………………………………… 183
xiii 4.4.3.1 Effect of sparger position on mass transfer coefficient………………. 183 4.4.3.2 Effect of antifoam on mass transfer coefficient……………………… 184 4.4.3.3 Analysis of reactor configuration that include perforated tube………. 184 4.4.3.4 Analysis of reactor configuration that include sintered sparger……… 184 4.4.3.5 Analysis of reactor configuration that include PBT down pumping and rushton impeller in the bottom………………………………………………. 185 4.4.4 EDR calculation in the configurations evaluated…………………….. 185 4.4.5 Cell Culture…………………………………………………………… 186 4.5. Conclusions………………………………………………………………. 187 4.6. References……………………………………………………………… 189 5. Effect of hydrodynamic conditions on cell cycle stress proteins and recombinant protein productivity………………………………………. 209 5.1 Abstract…………………………………………………………………. 209 5.2 Introduction…………………………………………………………….. 210 5.2.1 Cell Damage Mechanisms………………………………………………. 210 5.2.2 Mammalian Stress Response……………………………………………. 212 5.2.2.1 Structure and function of Stress Proteins………………………………. 213 5.2.2.2 Induction of response………………………..………………………… 215 5.2.2.3 Stress proteins and hydrodynamic stress…………………………….. 216 5.3 Materials and methods………………………………………………….. 218 5.3.1 Cell line……………………………………………………………… 218 5.3.2 Static Cell Cultures……………………………………………….…. 219 5.3.3 Bioreactors……………………………………………………………. 220 5.3.4 Continuous stress……………………………………………………... 221 5.3.5 Stress Proteins analysis……………………………………………….. 222 5.3.5.1 Cell fixation………………………………………………………….. 222 5.3.5.2 Staining and analysis………………………………………………… 222 5.3.6 Analytical methods………………………………………………….. 223 5.3.7 Cell cycle analysis…………………………………………………… 224 5.3.8 Parameter calculation……….………………………………………. 224 5.3.8.1 Growth rate…………………………………………………………. 224 5.3.8.2 Integral of viable cell concentration (IVC)…………………………. 225 5.3.8.3 Specific Productivity……………………………………………….. 225 5.4 Results and discussion…………………………………………………... 226 5.4.1 Effect of configuration on cell cycle profile…………………….…... 226 5.4.2 Effect of configuration on expression of stress proteins..…………… 229 5.4.3 Expression of stress proteins in Reactors Versus TFlask……………. 232 5.4.4 Expression of Stress proteins in a highly sensitive cell line………… 235 5.5 Conclusions……………………………………………………………… 237 5.6 References……………………………………………………………… 241 6. Conclusions and recommendations…………………….…………………. 266 Bibliography……………………………………………………………………. 271
xiv LIST OF TABLES
Table Page
2.1 Specific oxygen uptake rates (q ) reported for animal cell lines……… O2 93
2.2 Examples of empirically derived equations for kLa estimation found in literature………………………………………………………………... 94 2.3 Power number for selected impellers for turbulent regime…………….. 95 2.4 Relative Oxygen carrying capacities (k) of water, perfluorochemicals and gas vesicles (Modified from Sundararajan and Ju, 2006)………….. 96 2.5 Benefits and drawbacks of PFCs (modified from Lowe et al., 1998)…. 97 2.6 Correlations derived by Cui et al. (1996) for power drawn in systems with multiple Rushton impellers………………………………………... 98 2.7 Methodologies reported in literature for measurement of mechanical properties of animal cells and/or determination effect of hydrodynamical forces on animal cells………………………………… 99 2.8 Some parameters reported in literature to correlate the effect of hydrodynamical forces on cells………………………………………… 100 3.1 Median and range of energy dissipation rate that cells are exposed to in the Torture Chamber at different volumetric flow rates……...………… 149 3.2 Grid size and type of injection in the calculation of EDR……………… 150 3.3 Highest level of EDR that a particle would experience in a BD FACSVantage for different operating conditions in a 70 and 100 m nozzle ……………………………………………...... 151 3.4 Effect of sorting on cell cycle profile and fraction of cells in G2 phase.. 152 3.5 Highest level of EDR that a particle would experience in a BD FACS Aria for different conditions in a 70 m nozzle……………………….. 153 4.1 Experimental design to evaluate effect of sparger position. Side: Sparger is located at side of the impeller near to the wall of the tank. Edge: Sparger is located below the edge of the impeller. Centered: Sparger is located under the impeller in the center. High (h): Distance of the sparger with respect to the lower impeller, negative ( ) indicates negative value in the axes with respect to the impeller located in the center (0, 0, 0). Clearance (C): distance between impellers……………. 192 4.2 Configurations impeller sparger evaluated ……………………………. 193 xv 4.3 O2 Experimental table to test K L a capabilities of the 2L bioreactors for configurations that include perforated tube. (00) indicates center points, ( + or similar pattern ) indicate corner points…………………………... 193
4.4 O2 Experimental table to test K L a capabilities of the 2L bioreactors for configurations that include sintered sparger. (00) indicates center points, ( +, or similar pattern) indicate corner points…………………. 194
4.5 O2 Experimental table to test K L a capabilities of the 2L bioreactors for configurations that include sintered sparger of 50 m size pore. (00) indicates center points, ( +, or similar pattern) indicate corner points…. 194
4.6 O2 Data treatment to calculate K L a , Configuration C2 ………………… 195 4.7 Summary of Fit…………………………………………………………. 196 4.8 Parameter Estimates……………………………………………………. 196
4.9 O2 Prediction correlations to calculate K L a for the configurations evaluated……………………………………………………………….. 197
4.10 O2 K L a range reached with configurations including perforated tube and a combination of impellers……………………………………………… 198
4.11 O2 K L a range reached with configurations including sintered sparger and a combination of impellers……………………………………………… 198
4.12 O2 K L a range reached with configurations including sintered sparger and a combination of impellers……………………………………………… 198 4.13 Input data and power number and constants calculations using
O2 measured K L a for configuration C2 (Dual impeller and perforated tube)…………………………………………………………………….. 199 4.14 Input data and power number and constants calculations using
O2 measured K L a for configuration C5 (PBT impeller and sintered sparger)…………………………………………………………………. 199 5.1 Stress agents reported that induce stress proteins expression ……...… 245 5.2 State of art on relationship between protein expression and cell cycle phases …………………………………………………………………... 246
xvi LIST OF FIGURES
Figure Page 1.1 Aspects considered in the methodology to evaluate the effect of hydrodynamic forces on mammalian cell cultures…….………….… 10 2.1 Geometrical configuration, flow patterns and total non gassed power drawn into the liquid as a function of impeller spacing for a mixing system with multiple Rushton turbines (Adapted from Hudcova et al., 1989)……………………………………………………….……... 101 2.2 Impeller configurations commonly employed in animal cell culture.... 102 2.3 Regions of highest shear rate and highest energy dissipation rate behind the blades of a Rushton turbine……………………….……… 103 2.4 Effect of the gas flow rate on the power drop under gassed conditions for two impeller geometries in Newtonian, water like fluids. (Adapted from Galindo and Nienow, 1993)……………………….…. 104 2.5 Effect of the number and size of the baffles on the power drawn by an impeller in a cylindrical stirred tank reactor (Adapted from Oldshue, 1983)…………………………………………………….…. 105 2.6 Change in the tank diameter and impeller diameter as the volume of the vessel increases from 0.5 to 10,000 liters keeping constant the geometrical ratios H/T = 1 and T/D = 3………………………….…... 106 2.7 Lines of constant, maximum EDR in a vessel as a function of impeller rotational speed and diameter for Rushton turbine in water. H/T = 1 and T/D = 3……………………………………………….…. 107 2.8 Average EDR for the whole vessel as a function of impeller diameter and RPM using a Rushton turbine in water. H/T = 1 and T/D = 3………………………………………………………….…………… 108 2.9 Calculated maximum and average energy dissipation rate as a function of RPM for an Applikon bioreactor containing four baffles. The single points correspond to experimental measurements without baffles (Adapted from Mollet et al., 2004)……………………….….. 109 2.10 Molecular signaling and response cascade in endothelial and smooth muscle cells (A) before and (B) after stimulation by hydrodynamic forces. …………………………………………………………….….. 110
xvii 2.11 Effect of agitation rate on cell concentration, viability and aggregate diameter of murine NSC in batch suspension in a 125 mL spinner flask. Data points are average of duplicate runs. (a) Cell 1 1 concentration: ( ) 60 rev·min ; ( ) 100 rev·min . Viability: ( ) 60 rev·min 1; ( ) 100 rev·min 1.(b) Average aggregate diameter:( 1 1 ) 60 rev·min ; ( ) 100 rev·min . Standard deviation:( ) 60 rev·min 1; ( ) 100 rev·min 1. (Adapted from Sen et al. 2001)….…. 112 2.12 Effect of stirring speed on cell concentration after 7 days of culture of Vero cells on Cytodex microcarriers on 250 mL spinner vessels (Data from Hirtenstein and Clark, 1980)………………………….…. 113 2.13 Experimental curves for the percentage of damage experienced by cells in a custom design microfluidic device for single abuse experiments. Adapted from Ma et al. (2002), Mollet et al. (In Press) and Mollet et al., (submitted)…………………………………….…... 114 2.14 Summary of the reported energy dissipation rate at which cells are affected as well as the reported levels of energy dissipation rate in various bioprocess environments. Adapted from Ma et al. (2002) and Mollet et al. (2004).……………………………….………………….. 115 2.15 A three dimensional plot of the number of cells associated with each bubble as a function of cell concentration (cell·mL 1) and Pluronic F 68 concentration. The dots indicated experimental data and the surface is the plot of a multiple variable regression. (Adapted from Ma et al., 2004)…………………………………………………….… 117 3.1 Photograph (a), top view (b), and a perspective view (c) of the flow contraction device …………………………………………..……….. 154 3.2 Single Pass set up: Cells are centrifuged and resuspended into fresh media. They pass once through the Torture chamber at different flow rates ………………………………………………….…………….…. 155 3.3 Simplified sketch flow cytometer ……………...………………….…. 156 3.4 Exit sheath flow rate as a function of flow rate scale and sheath pressure for the 70 m nozzle (a), and sample flow rate as a function of flow rate scale and sheath pressure for a 70 m nozzle (b)……….. 157 3.5 Cell damage estimated from single pass experiments in the shear stress device ……………………………………..……………….….. 158 3.6 Growth kinetic of THP1 cells after stress exposure in shear stress device (TC), cells were exposed at 90 mL/min ……...…………….… 158 3.7 Growth kinetic of THP1 cells after sorting in FACS Aria, cells were grown in media with 10% FBS ………..…………………………..… 159
xviii 3.8 Growth kinetic of THP1 cells after sorting in FACS Aria, cells were grown in media with 30% FBS ……………………………………… 159 3.9 Growth kinetic of THP1 cells after sorting in FACS Aria, cells were grown in media with 10% FBS, 30%FBS and conditioned media…. 160 3.10 Cells suspensions at different cell concentrations were sorted under the same conditions. Cell damage was calculated based on the amount of LDH in the supernatant after sorting ………………….… 160 3.11 THP1 cells sorted in media with 0% FBS. Cells were counted with hematocytometer before and after sorting. Aria reports number of events …………………..…………………………….……………… 161 3.12 THP1 cells sorted in media with 10% FBS. Cells were counted with hematocytometer before and after sorting. Aria reports number of events ……………………………………………………………….. 161 3.13 Statistical analysis of THP1 cells sorted in media with 10% FBS versus cells sorted in media with 0%FBS …..……………………… 162 3.14 Cell cycle profile before and after stress exposure in flow cytometer.. 163 3.15 Cell cycle profile of cells arrested in G2 phase before and after stress exposure in flow cytometer…………………………………………... 164 3.16 Statistical analysis of fraction of THP1 cells in G2 phase before and after sorting………………………………………………………….. 165 3.17 Photograph of BD FACS Aria Flow cell components………………. 166 3.18 View of the nozzle geometry and mesh used for the simulation. Geometry and mesh were built in Gambit……………………………. 167 3.19 Fluent output of the simulations of particles flowing through the nozzle The color coded figures correspond to the levels of EDR…… 168 4.1 Type of impellers evaluated: Rushton (a), Pitch Balde Turbine (b)………………………………………………………...…………… 200 4.2 Geometrical configuration to evaluate effect of impeller and sparger
O2 location on K L a . h: distance between lower impeller and sparger, C: distance between impellers, sparger location: side, center, edge…. 200
4.3 O2 Sample calculation of K L a from collected data. DO profile……….. 201
4.4 O2 Sample calculation of K L a from collected data. Obtaining slope….. 201
4.5 O2 Actual vs. Predicted K L a after fitting a response surface to data
O2 collected during the K L a assay for configuration C2………………. 202
xix 4.6 O2 Effect of location of sparger on K L a Sparger was located in three different positions: At side of the impeller, near to the wall of the tank, edge of the impeller and centered under the impeller……..……. 202
4.7 O2 Surface response results on effect of antifoam on K L a ……………. 203
4.8 O2 Effect of configuration impeller/sparger on K L a . Perforated tube and combination of impellers………………………………………… 203
4.9 O2 Effect of configuration impeller/sparger on K L a . Sintered sparger and combination of impellers………………………………………… 204
4.10 O2 Effect of configuration impeller/sparger on K L a . Perforated tube and combination of impellers including dual impeller (PBT down – pumping and rushton impeller)……………………………………….. 204 4.11 Summary of the reported energy dissipation rate at which cells are affected as well as the reported levels of energy dissipation rate in various bioprocess environments. Adapted from Ma et al. (2002) and Mollet et al. (2004)…………………………………………………… 205 4.12 Cell growth in bioreactors with best configurations…………………. 207
4.13 CO2 profile in cell culture evaluation………………………………… 207 4.14 Recombinant protein profile in cell culture evaluation……………… 208 5.1 Summary of the reported energy dissipation rate at which cells are affected as well as the reported levels of energy dissipation rate in various bioprocess environments. Adapted and improved from Ma et al. (2002) and Mollet et al. (2004)…………….……………….…..... 247 5.2 Diagram of the experimental setup for continuous, chronic exposure of suspended animal cells to high levels of hydrodynamic forces. Adapted with permission from Godoy silva et al (2009)……………. 249 5.3 Growth kinetic (a) and viability (b) for four cell cultures performed in bioreactor with two different impeller/sparger configurations. One configuration operated with perforated tube sparger and dual impeller (PBT in the top and Rushton bottom), the second configuration operated with sintered sparger (50 m) and PBT impeller. Two strains of the CHO clonal cell line were used strain A and strain B were seeded in each bioreactor as is indicated in the label………….. 250
xx 5.4 Concentration of glucose (a) and lactate (b) for four cell cultures performed in bioreactor with two different impeller/sparger configurations. One configuration operated with perforated tube sparger and dual impeller (PBT in the top and Rushton bottom), the second configuration operated with sintered sparger (50 m) and PBT impeller. Two strains of the CHO clonal cell line were used strain A and strain B were seeded in each bioreactor as is indicated in the label………………………………………………………………. 251 5.5 Cell cycle diagram (A) Typical cell cycle histogram obtained in flow cytometer (B). Content of DNA increases along the x axis. First peak correspond to cells in G1 phase, small area in the center with lines correspond to S phase, peak further right correspond to cells in G2 phase…………………………………………………………………. 252 5.6 Cell cycle profile (A), Relationship between cell cycle and titer: subpopulation in G1 phase (B), subpopulation in G2 phase (C), and subpopulation in S phase (D)………………………………………… 253 5.7 Cell cycle profile comparison between reactors with different hydrodynamic conditions. Two impeller/sparger configurations were evaluated: A. Open tube sparger and dual impeller Rushton bottom – PBT top. B. Sintered sparger and PBT impeller……………………… 254 5.8 Relationship between cell cycle and titer subpopulation in G1 phase for two different reactor configurations. 1. Open tube sparger and dual impeller Rushton bottom – PBT top. 2. Sintered sparger and PBT impeller………………………………………………………….. 255 5.9 Typical plots for stress proteins analysis in flow cytometry. Dot plot (a) indicates gated populations for live cells, dead cells, debri and aggregates. Histogram to estimate mean fluorescence intensity of samples as indicative of level of expression of stress proteins (b)…… 256 5.10 Analysis of relationship between population expressing HSP70 and specific productivity …………………….………………………….. 257 5.11 Analysis of relationship between population expressing HSP90 and specific productivity…………………………………………………. 258 5.12 Profile of stress proteins expression. Two different strains of a clonal CHO cell line where seeded in two bioreactors with the configuration sintered sparger and PBT Impeller. Strain A (a), and Strain B (b)…... 259 5.13 Profile of stress proteins expression. Two different strains of a clonal CHO cell line where seeded in two bioreactors with the configuration open tube sparger and dual impeller (PBT top, Rushton bottom). Strain A (a), and Strain B (b)………………………………………… 260
xxi 5.14 Growth kinetic of cell line CHO 6E6. (a), and viability, (b), for cell culture in agitated bioreactor (2 L working volume) and TFlasks, to compare profile of stress protein expression…………………………. 261 5.15 Concentration of glucose, (a), and lactate, (b), as a function of time of two batchs of cells growing in TFlasks Vs. cells growing in agitated bioreactor (2L working volume). Cell line CHO 6E6……………….. 262 5.16 Comparison on expression of stress proteins between reactor and TFlasks. Cell line CHO 6E6…………………………………………. 263 5.17 THP1 cell suspension was subjected to recycle through the TC at 50 mL·min 1, cells passed 10 times through TC. Samples were taken 1, 6, 8 10 hr in recovery to evaluate HSP70 and HSP90 expression. Purple line corresponds to unstained control, green line corresponds
to control (non stressed cells), red line correspond to sample cells stressed at 50 mL/min……………………………………………….. 264 5.18 THP1 cell suspension was subjected to recycle through the TC at 90 mL·min 1, corresponding to 1.1×108 W·m 3, cells passed 10 times through TC. After stress, cells were incubated at 37 °C in a 5% CO2 atmosphere. Samples were taken 2 and 7 hr in recovery to evaluate HSP70 and HSP90 expression. Purple line corresponds to unstained control, green line corresponds to control (non stressed cells), red line correspond to sample cells stressed at 90 mL/min…………………… 265
xxii
CHAPTER 1
INTRODUCTION
1.1. ANIMAL CELL CULTURE
Cultures of animal cells are able to reproduce clinically active copies of proteins used by the human body. These proteins have been used in a recent generation of drugs for the replacement or reinforcement of deficient or defective proteins or component characteristics of different diseases. (Cartwright, 1994). Although recombinant proteins can also be obtained from bacteria, animal cell culture has emerged as an attractive technology for producing recombinant proteins. These cells have the machinery to make post-translational modifications as well as regulation mechanisms to assure proper folding and also secrete these active forms into the suspending media. These unique properties allow the production of proteins in their functional configuration, which can
represent significant savings in the overall process. Other technologies that use bacteria
typically require significant downstream processes that include lysis of cells in order to
release the recombinant protein. They also require additional steps like solubilization and
refolding in order to obtain the bioactive conformation of the protein.
1 There are a variety of pharmaceutical products obtained through the use of cell cultures.
Cells Hamster Ovary (CHO cells) was the cell line used for the production of tPA (tissue plasminogen activator), a protein used in the treatment of cardiac disease. Since its
approval in 1986, this cell line has been used to obtain other proteins such as β-interferon used in multiple sclerosis, Factor IX used in hemophilia B and erythropoietin used in anemia among others. Over the last decade, a large number of monoclonal antibodies, produced in CHO cells have became approved, or are in various stages of approval. This success of antibodies is primarily the result of high specificity of the antibodies that can be preserved in this CHO derived products as well as the “humanized” characteristics of these antibodies that prevent human immune response.
The advantages of cell culture technology are reflected in the growing market of recombinant proteins, whose sales were estimated at $54.5 billion in 2007 and are expected to increase up to $75.8 billion by 2012 (Research and markets, 2008). Of course, the impact of the development of animal cell cultures should not be assessed purely as an issue of economics. The key benefit of the technology is the potential for developing treatments to previously incurable diseases and providing an improved quality of life to those who receive treatments thus derived. Due to the popular demand for monoclonal antibody and recombinant protein production, and continued economic pressures, the pharmaceutical industry continues working to increase production capacity, however, such increasing demands for productivity not only in the upstream stage but also in the downstream processes can subject cells to aggressive environments including
2 hydrodynamic stress. More broadly, the effect of hydrodynamic forces on animal cells
used for research purposes is also of interest.
1.2. OBJECTIVES
1.2.1. Goal
This work’s intent is to improve the understanding of the effect of hydrodynamic stress on mammalian cells. Cell damage due to hydrodynamic stress occurs in biomedical devices such as Flow cytometer as well as industrial equipment. We focused on the quantification of cell damage of both industrially relevant cell lines and human cell lines used for research purposes.
1.2.2. Methodology
Two cell lines were chosen to evaluate the stress response in this work. Chinese Hamster
Ovary (CHO) and THP1. CHO is a cell line widely used in the pharmaceutical industry for the production of recombinant proteins. CHO cells were exposed to different hydrodynamics conditions in order to evaluate the stress response as well as the effect on viability and productivity of the cell culture. THP1 is a human cell line, which is routinely studied in screening assays to assess cytotoxicity induced by potential agents.
THP1, being from human origin, also allows for many other types of analysis and represents a model of study of human cells used in flow cytometry analysis. In addition,
3 other human leukemia cell lines as well as primary cells were used in screening of cell
sensitivity.
A summary of aspects considered in the methodology is presented in Figure 1.1. The
methodology will be discussed in detail in every chapter, briefly, the analysis include:
characterization of cell cultures in static cultures, cell sensitivity screening using single pass assays, cell damage in flow cytometer (THP1 cells), cell damage in bioreactors
(CHO), stress response, and hydrodynamic characterization of bioreactors and flow
cytometer.
1.3. SCOPES OF THIS STUDY
1.3.1. Cell damage in a fluorescent activated cell sorter (FACS). (Chapter 3)
The first part of the work developed to study cell damage in FACS was published in the paper:
Mollet, M., Godoy-Silva, R., Berdugo, C, Chalmers, J.J. (2008). Computer Simulations
of the Energy Dissipation Rate in a Fluorescence-Activated Cell Sorter: Implications to
Cells. Biotechnology and Bioengineering, Vol. 100, No. 2, June 1, 2008.
The work reported in this paper focused on the BectonDickenson flow cytometer with a
model name of FACS Vantage. Chinese hamster ovary cells (CHO) and a human
leukemia cell line (THP1) were used as model to evaluate integrity in cell sorting.
4 Claudia Berdugo’s contribution to this work includes:
- Experiments of cell sorting in FACS Vantage with THP1 cells
- Data and results included in Figure 4.7 as well as data from cell sorting include in
Tables 4.3 and 4.4
- Editing of the document, bibliographical review, analysis and discussion.
The second part of the work is presented in this chapter and focus in hydrodynamics studies and cell damage on FACS Aria. This instrument is equipped with three lasers
(488nm, 633nm and 405 nm), 13 fluorescent channels and 2 scatter channels. FACS Aria offers expanded detection of fluorescence with respect to the Vantage since the last one does not have 405 nm laser. Another important difference between Vantage and Aria is a fundamentally different design in the fluidics: the point of interrogation of cells by the laser is “upstream” of the point of the drop creation (unlike the FACS Vantage in which the cells in drops are interrogated), and the actual nozzle where the “sheath fluid: and fluid containing the cells converges is significantly different. Both instruments are widely used in Biomedical research, we focused on the quantification of cell damage, based upon practitioners reported damage.
The effect of hydrodynamic stress induced by sorting these cells in these two instruments was quantified both experimentally as well as theoretically using computational fluid dynamics packages. The results indicate that not only are cells damaged in a flow cytometer, but that this damage can vary from cell line to cell line as well as specific 5 conditions/type of flow cytometer and flow conditions in rapid contractions (the torture
chamber). In addition, the sensitivity of any specific cell line can be a function of the
growth phase of the cell.
1.3.2. Hydrodynamic studies in bioreactors. (Chapter 4)
The work presented in this chapter was developed as part of an internship at
GlaxoSmithKline.
Large-scale cell cultures require efficient mass transfer systems for economically viable production. Mass transfer capabilities are determined by the bioreactor’s agitation and aeration configuration. The aeration and agitation configuration also dictates the level of shear stress that cells will experience during culture. A balance should be accomplished to obtain the best mass transfer capabilities and low shear stress during bioreactor design.
In this work, the performance of ten different impeller-sparger configurations was evaluated. A full factorial experimental design with two center points was used to characterize the mass transfer capabilities for every configuration. The best impeller-
O2 sparger configuration was chosen based on the K L a response surface model for testing
in cell culture experiments.
The evaluation included different types of sparger and impeller as well as relative
location of impeller along the shaft and clearance between impellers. Results indicated 6 that the volumetric oxygen mass transfer coefficient is affected by the relative sparger
location and is less affected by clearance between impellers.
Once the best impeller sparger configuration was identified a cell culture experiment was carried out to assess its effects on culture performance. The proposed configuration supported high cell density cultures through improved gas dispersion, acceptable shear
rates and low foam formation.
1.3.3. Effect of hydrodynamic conditions on cell cycle, stress proteins and
recombinant protein productivity. (Chapter 5).
It has been proposed that the dependence of protein expression on cell cycle phase can be used quantitatively in animal cell bioreactor optimisation. Protein synthesis has been associated on particular cell cycle phases for different recombinant genes, however there is not a consensus in the literature regarding to the cell cycle phase where the cells are the most productive.
In this work, a potential dependence of recombinant proteins productivity on cell cycle was studied. In addition, we have evaluated the expression of stress proteins under different hydrodynamic conditions and we analyzed the relationship between the expression of stress proteins and recombinant protein productivity. Also, the expression of stress proteins in highly sensitive cell lines will be discussed. Stress proteins investigated have been studied in the context of heat and nutritional stress but it has not
7 been determined the expression of stress proteins in response to hydrodynamic
conditions.
The results strongly suggest that secretion occurs primarily during the G1 cell cycle phase. Some differences were observed depending on the feeding strategy and the
hydrodynamic conditions. On the other hand, the expression of stress proteins follows a
characteristic profile that is related with the state of the culture.
8
1.4. REFERENCES
Cartwright, 1994. Animal Cells as bioreactors. Cambridge University Press.
Godoy-silva, R., Berdugo, C. and Chalmers, J. (Peer review). Literature Review: Mixing, Aeration and hydrodynamics in bioreactors. Wiley Encyclopedia of Industrial Biotechnology.
Ma, N.; Koelling, K. and Chalmers, J. J. (2002). The fabrication and use of a transient contractional flow device to quantify the sensitivity of mammalian and insect cells to hydrodynamic forces. Biotechnology and Bioengineering. 80:428-437.
Mollet, M., Godoy-Silva, R., Berdugo, C, Chalmers, J.J. (2008). Computer simulations of the energy dissipation rate in a fluorescence activated cell sorter: implications to cells. Biotechnology and Bioengineering. 100(2): 260-272.
Mollet, M.; Godoy-Silva, R.; Berdugo, C. and Chalmers, J. J. (2007). Acute Hydrodynamic Forces and Apoptosis: A Complex Question. Biotechnology and Bioengineering. 98 (4): 772-788.
Research and markets, 2008. Biologics Pipelines: Who has the next promising recombinant proteins? http://www.researchandmarkets.com/reports/603343
9
THP1 CHOCHO cell line CHO-K1CHO-6E6 (Cell(ATCC) line ATCC) (GlaxoSmithKline) (Cell(ATCC) line ATCC)
Static Cultures Flow Cytometer Reactor Studies Static Cultures Reactor Studies
CellCell culture: cycle, Single Pass LDH Different Single Pass configurations to StressCell cycle, Proteins: evaluate Mass HSPStress 70, proteins HSP 90 Transfer RecombinantHSP70, HSP90 Protein LDH Stress proteins capabilities LDH kLa, EDR Molecular markers Molecular markers Continuous Stress Stress proteins Hydrodynamics Stress proteins Stains Stains characterization CellCell culture: cycle, Stress Proteins: Molecular markers Cell cycle, Molecular markers HSP 70, HSP 90 Stains Stress proteins Stains RecombinantHSP70, HSP90 Protein Recombinant protein
Figure 1.1. Aspects considered in the methodology to evaluate the effect of hydrodynamic forces on mammalian cell cultures.
10
CHAPTER 2
LITERATURE REVIEW: MIXING, AERATION AND HYDRODYNAMICS IN
BIOREACTORS
The content of this chapter is currently under the process of peer review as a chapter contribution for the Wiley Encyclopedia of Industrial Biotechnology. The chapter was written in conjunction with Dr. Ruben Godoy, a former PhD student at Dr. Chalmers’ group.
2.1. ABSTRACT
Animal cell culture has been used extensively for the production of a wide variety of substances with biological and therapeutic activity. In spite of being a mature area of development, no universally accepted method of design or scale up of animal cell culture has been developed.
When designing or scaling up a new process, mixing and aeration are among the most important operations to be considered since they achieve some of the most fundamental
11 objectives carried out in a bioreactor such us keeping certain level of homogeneity in the physicochemical parameters of the culture and providing oxygen to the cells. On one
side, oxygen is a crucial nutrient involved in growth and energy production in animal cell
culture; however, the low solubility of oxygen in water based media forces its continuous
supply to the culture, usually through direct sparging. Foam and cell bubble interactions
resulting from gas sparging may be problematic or even catastrophic for the culture but
they are usually controlled through the use of antifoaming substances and protective
surfactants.
Mixing, on the other hand, is absolutely necessary in animal cell culture to keep cells
suspended and improving mass and heat transfer. While some degree of mixing is
obtained through air sparging, particularly in the case of air lift bioreactors, most bioreactors rely on some kind of mechanical device (impeller) to draw kinetic energy into
the broth. Even though the choice of type and geometric characteristics of the impeller
are highly subjective, the optimum stirring speed is commonly evaluated experimentally
in bench scale bioreactors; even so, a large proportion of under mixed, large scale animal
cell culture bioreactors are currently in operation in many facilities as a result of a
wrongfully perceived “shear sensitivity” of animal cells. Although it has been
conclusively shown that certain culture conditions can lead to cell damage and death,
such damage is very dependent on the particular cell line, the culture characteristics such
as surface dependent vs. free suspension growth and the presence of any gas medium
interface where bubbles can rupture, especially when protective surfactants are not present. While experimental evaluation is advisable, for common industrial cell lines and
12 culture media “shear” sensitivity is not an issue and therefore scaling up conditions
should be focused on improving mass transfer limitations.
2.2. INTRODUCTION
Animal cells have an innate ability for proper folding and post translational processing of proteins that makes them preferable as host for producing biological components of
therapeutic and diagnostic interest; as a result, the culture of mammalian cell have been
used extensively for the production of biologics, including virus vaccines, monoclonal
antibodies (MAbs), hormones, enzymes, growth and blood factors. Although the
industrial exploitation of animal cell cultures started over five decades ago with the production of Salk polio virus vaccine in primary monkey kidney cells (Griffiths, 2000), it has been during the last 21 year period when there has been a rapid increase in the number of FDA approved products produced in mammalian cell culture, starting in 1986 with the production of recombinant tissue plasminogen activator (tPA) with genetically engineered CHO cells and the simultaneous introduction of the first Mab, ORTOCLONE
OKT3 (muromonab CD3) as a treatment for solid transplant rejection. Since then, 29
MAbs and many other therapeutic agents have been approved for marketing in the USA for the diagnosis or treatment of diverse diseases, a majority of which are produced in animal cell cultures (Ozturk, 2006; Thomson Centerwatch, 2007). By 2005, one estimate of the world market for antibodies produced in animal cell culture was 14 billion dollars and this number was expected to grow to more than 16 billion by 2006 (Research and
Markets, 2007). While the specific, and or volumetric productivity continues to improve,
13 i.e. greater than 5 g/L of product is now considered the norm, this rapid growth in the
number and demand of biologics has driven the industry to expand the size and number
of production facilities and lead to statements that shortfalls in manufacturing capacity exist (Molowa and Mazanet, 2003). All of this positive growth puts significant pressure to continually improve the size and productivity of commercial animal cell culture systems.
Despite the fact that reports exist showing that animal cell culture has been conducted for over a hundred years (Harrison, 1907; Carrel, 1912), no universally accepted method of design or scale up of animal cell culture has been developed. Diverse culture methodologies including static cultures (T flask), roller bottles, cultures on microcarriers and freely suspended cells in batch, fed batch or perfusion systems, among others, have evolved. While freely suspended, fed batch processes have emerged as the predominate ones, much of this evolution has been guided by the particularities of the process, the cell line and, most importantly, the in house expertise of the researchers specific to each organization. Although there are reports of large scale (1000 to 2000 liter) airlift bioreactors for protein/antibody production (Birch et al., 1985; Varley and Birch, 1999;
Hesse et al., 2003), at least 70% of the licensed processes for recombinant proteins, antibodies and vaccines using microcarriers or freely suspended cultures use traditional stirred tank bioreactors with reported capacities up to 20,000L (Chu and Robinson, 2001;
Butler, 2005; Meier, 2005). Several reasons account for this preference including the vast empirical knowledge accumulated for the design, scale up and operation of this type of reactor in the chemical and biochemical industries over the last century, the versatility of
14 this reactor type allowing its adaptation to several different processes with minor or no
modifications and, finally, the relative simplicity. As a result, most guidelines for bioprocesses design and scale up are based on stirred tank reactors and as such will be
the primary focus for the remainder of this review.
All animal cell processes are aerobic; oxygen is a crucial nutrient involved in growth and
energy production in such cells. Unfortunately, because of its low solubility in water based media, dissolved oxygen is consumed quickly, requiring a continuous supply in
order to keep cells alive. The easiest way of providing oxygen to a stirred cultured is
through surface aeration; however, surface aeration is not sufficient for cultures beyond
~100 L. Other aeration methods (i.e. perfluorocarbons, membrane aeration, oxygen
carriers) have been used successfully at scales up to 500 L but their use is limited by high
cost and downstream concerns (perfluorocarbons) or by limited design data and
difficulties in maintenance (membrane aeration). These limitations have placed air and
oxygen enriched air sparging as the most common and simplest method for continuously providing oxygen in bioreactors.
Agitation, on the other hand, is essential for satisfactory mass transfer and homogeneity
in a bioreactor with or without sparging; this homogeneity is crucial for process control.
Although sparging provides some degree of agitation by itself, most animal cell cultures
rely on some sort of mechanical device to impart enough kinetic energy to the fluid so a
certain degree of homogeneity is reached.
15 Both agitation and sparging are essential for the success of industrial cultures of animal
cells; yet, both are associated with cell damage as a result of cell bubble, liquid cell
and/or solid cell interactions. Significant progress has been made over the last two
decades in understanding of cell damage mechanisms and in this chapter we will attempt
to present an overview of the current practices and protocols for aeration and mixing in bioreactor design and operation as well as their connection to hydrodynamical damage in
animal cell processes.
2.3. AERATION
Oxygen is a key substrate in animal cell cultures; unfortunately, its sparing solubility in
water based media (7.8 mg·l 1 when bubbling air in water at 760 mm Hg and 25°C)
requires a continuous supply throughout the culture for most larger scale situations. At
high cell concentrations, the rate of oxygen consumption may exceed the rate of oxygen
supply; in those conditions, the dissolved oxygen (DO) concentration falls down to a point (the critical oxygen concentration) where it becomes limiting. Reported critical
oxygen concentration for animal cells ranges typically between 1 and 10% of air
saturation (Gotoh et al., 2001; Zeng and Bi, 2006). Below such range, typically the
oxygen consumption rate as well as the mithocondrial activity decrease, while the
specific glucose and glutamine consumption rates increase (to compensate for ATP production) with the corresponding increase in the lactate and ammonium formation rates
(Miller et al., 1987; Ozturk and Palsson, 1990; Lin and Miller, 1992). Together with a reduced mithocondrial activity, there is a reduction in the specific secretion rate of
16 proteins and volumetric accumulation of product which translates into higher cell viability, delayed death and lower chromosomal damage but also into low product yield
(Ozturk and Palsson, 1990; Packer and Fuehr, 1977).
On the other hand, elevated oxygen concentration (~100% of air saturation) may stimulate the generation of reactive oxygen species (ROS) that can alter cellular macromolecules such as DNA, proteins and lipids, impairing cell growth or even causing death. (Zeng and Bi, 2006).
As a result, cell culture should be constrained within an optimal range of dissolved oxygen that must be determined experimentally, taking into account that optimal DO values for cell growth may be different than optimal values for protein production. Oh et al. (1989) reported satisfactory cultivation of mammalian cells from approximately 5 to
100% of air saturation. Jan et al. (1997) studied the effect of dissolved oxygen from 10%
to 150% of air saturation on the growth of a murine hybridoma cells at steady state in
serum free continuous culture; while no significant effect on cell viability, growth rate or
specific antibody production rate was found, an increase in the amount of glucose utilized
at higher oxygen concentrations was detected. It was speculated that this increased
glucose consumption rate was associated with an increased activity of antioxidant
enzymes needed to reduce the cytotoxic effect of ROS. Using the same system, Kunkel et
al. (1998) described an increase in the level of galactosylation of the mAb chains with
increased dissolved oxygen.
17 For common animal cells lines under favorable oxygen supply, reported specific oxygen