Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Autophagy inhibition and hypertonicity increased monoclonal antibody production in chinese hamster ovary… Nasseri, Sayyed Soroush 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_spring_nasseri_sayyedsoroush.pdf [ 4.74MB ]
Metadata
JSON: 24-1.0073489.json
JSON-LD: 24-1.0073489-ld.json
RDF/XML (Pretty): 24-1.0073489-rdf.xml
RDF/JSON: 24-1.0073489-rdf.json
Turtle: 24-1.0073489-turtle.txt
N-Triples: 24-1.0073489-rdf-ntriples.txt
Original Record: 24-1.0073489-source.json
Full Text
24-1.0073489-fulltext.txt
Citation
24-1.0073489.ris

Full Text

AUTOPHAGY INHIBITION AND HYPERTONICITY INCREASED MONOCLONAL ANTIBODY PRODUCTION IN CHINESE HAMSTER OVARY FED-BATCH CULTURES  by Sayyed Soroush Nasseri  B.Sc., Sharif University of Technology, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Chemical and Biological Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  December 2012  © Sayyed Soroush Nasseri, 2012  Abstract  Biopharmaceuticals have improved the treatment of many diseases such as cancers and strokes, with much reduced adverse effects. However, high prices pose a major challenge to their widespread application. Improvements to cell culture processes can ease this burden, and facilitate the access to high quality biological medications for increasing sections of world society. is thesis addresses two methods to increase the productivity of the Chinese Hamster Ovary (CHO) cells expressing monoclonal antibodies using either autophagy inhibition or hypertonicity. Autophagy is a cellular process whereby intracellular components are degraded in response to nutrient limitations and other stresses. It has previously been shown that 3-methyladenine (3MA) inhibition of autophagy can increase fed-batch tissue plasminogen activator production. is thesis investigated autophagy inhibition in fed-batch cultures of CHO cells producing monoclonal antibodies (MAb). Hypertonicity in mammalian cell cultures has also been shown to increase the MAb productivity of mammalian cell cultures. By investigating the timing and dose of 3-MA, more than two-fold increases in cell specific productivity and an up to two-fold increase in total MAb were obtained. Hypertonic conditions can similarly increase the productivity of fed-batch cultures by more than two-fold. e combination of these two approaches, however, did not result in any increase in productivity. Neither autophagy inhibition nor hypertonicity significantly changed the glycan profiles of the MAb product.  ii  Preface  I was the major contributor of this thesis alongside my supervisors, Dr. James Piret and Dr. Bhushan Gopaluni. I have performed most of the experimental design and work. Christopher Sherwood performed the amino acid analyses by HPLC analysis in Piret laboratory. Katrin Braasch (Michael Butler laboratory at e University of Manitoba) has performed the glycan analyses in Chapter 4, and provided the corresponding section in the Materials and Methods in Chapter 2.  iii  Table of Contents  Abstract ............................................................................................................................... ii	
   Preface ................................................................................................................................ iii	
   Table of Contents ................................................................................................................ iv	
   List of Tables .....................................................................................................................viii	
   List of Figures ....................................................................................................................... x	
   List of Abbreviations.......................................................................................................... xix	
   Acknowledgements ............................................................................................................ xxi	
   Dedication ....................................................................................................................... xxiii	
   Chapter 1: Introduction.......................................................................................................1	
   1.1	
   Biopharmaceutical Production Using Mammalian Cell Culture .............................. 1	
   1.2	
   Mammalian Cell Cultures in Industrial Settings...................................................... 2	
   1.3	
   Fed-Batch Processes ................................................................................................ 2	
   1.4	
   Programmed Cell Death ......................................................................................... 4	
   1.4.1	
   Apoptosis ......................................................................................................... 4	
   1.4.2	
   Autophagy........................................................................................................ 5	
   1.5	
   Effect of Autophagy Inhibition on Protein Production ............................................ 7	
   1.6	
   Osmolality in Cell Culture ...................................................................................... 8	
   1.7	
   Protein Glycosylation ............................................................................................ 10	
   1.8	
   Research Goals ...................................................................................................... 14	
   Chapter 2: Materials and Methods.....................................................................................15	
    iv  2.1	
   Cell Lines .............................................................................................................. 15	
   2.2	
   Cell Culture .......................................................................................................... 15	
   2.2.1	
   Maintenance Medium and Cultures ............................................................... 15	
   2.2.2	
   Batch and Fed-batch Cultures ........................................................................ 16	
   2.2.3	
   Cell Banking .................................................................................................. 20	
   2.3	
   Flow Cytometry .................................................................................................... 20	
   2.4	
   IgG ELISA ............................................................................................................ 21	
   2.5	
   Glycan Analysis ..................................................................................................... 22	
   2.6	
   Cell Counting ....................................................................................................... 22	
   2.7	
   Chemical Analysis ................................................................................................. 23	
   2.8	
   Osmometry ........................................................................................................... 23	
   2.9	
   HPLC Amino Acid Analysis .................................................................................. 23	
   2.10	
   Calculations ........................................................................................................ 24	
   Chapter 3: Establishing Fed-batch Cell Culture Protocol for MAb Producing CHO Cells .... ...........................................................................................................................................25	
   3.1	
   Fed-batch Culture Proof of Concept Study ........................................................... 25	
   3.2	
   Optimization of CHO-tPA Fed-Batch Protocol to CHO-MAb Cells Using DOE 28	
   3.2.1	
   First Screening Experiment ............................................................................ 30	
   3.2.2	
   Second Screening Experiment ........................................................................ 36	
   3.2.3	
   Augmenting the Second Screening Experiment .............................................. 42	
   3.2.4	
   Assessment of Possible Amino Acid Limitations ............................................. 46	
   3.3	
   Conclusions to Adapting CHO-tPA Fed-batch Protocol to MAb Cells ................. 50	
    v  Chapter 4: Autophagy Inhibition and Hypertonicity Effects on CHO-MAb Fed-batch Cultures ..............................................................................................................................52	
   4.1	
   Autophagy Inhibition Effect on the Protein Production in CHO-MAb Cultures .. 52	
   4.2	
   Autophagy Inhibition by 3-MA in CHOm56 Cells .............................................. 54	
   4.2.1	
   Dose Response of 3-MA in CHOm56 Fed-batch Cultures ............................ 54	
   4.2.2	
   Amino Acid Analysis for 3-MA Treated CHOm56 Fed-batch Cultures ......... 58	
   4.3	
   3-MA Treatment of CHOm44 Fed-batch Cultures Under Hypertonic Conditions .. .............................................................................................................................. 60	
   4.4	
   Increase of Protein Production in High Osmolalities............................................. 64	
   4.5	
   Autophagy Inhibition Under Different Osmolality Conditions ............................. 67	
   4.6	
   Effects of Autophagy Inhibition and Hypertonicity on the MAb Glycosylation Patterns .............................................................................................................................. 75	
   4.7	
   Conclusions to the Autophagy Inhibition and Hypertonicity Effects on CHO-MAb Fed-batch Cultures ............................................................................................................... 81	
   Chapter 5: Conclusions and view to future ........................................................................82	
   Bibliography .......................................................................................................................85	
   Appendices .........................................................................................................................99	
   Appendix A. Oxygen Limitation in Shake Flasks due to Increase in Volume .................. 99	
   Appendix B. Supplementary Data for Section 3.1 ........................................................ 101	
   Appendix C. Supplementary Data for Section 3.2.1 ..................................................... 103	
   Appendix D. Supplementary Data for Sections 3.2.2 and 3.2.3 ................................... 109	
   Appendix E. Supplementary Data for Section 4.1 ........................................................ 112	
    vi  Appendix F. Supplementary Data for Section 4.2.1 ..................................................... 113	
   Appendix G. Supplementary Data for Section 4.3 ........................................................ 115	
   Appendix H. Supplementary Data for Section 4.4 ....................................................... 116	
   Appendix I. Supplementary Data for Section 4.5 ......................................................... 118	
   I.1	
   Results from the First Replicate ....................................................................... 118	
   I.2	
   Results from the Second Replicate ................................................................... 122	
   Appendix J. Glycan Analysis Raw Data (Section 4.6) ................................................... 126	
   J.1	
   Glycan Analysis of Antibody Heavy Chains ..................................................... 128	
   J.2	
   Glycan Analysis of Antibody Light Chains ....................................................... 131	
    vii  List of Tables Table 2.1: Measured composition of CD CHO and Feed A media .......................................... 19	
   Table 3.1: List of proposed independent fed-batch process variables. Some items such as medium composition or feeding regime can be divided into more variables. .......................................... 29	
   Table 3.2: Design of experiment table for the first characterization experiment. ....................... 31	
   Table 3.3: IVC response analysis of variance table .................................................................... 35	
   Table 3.4: Total MAb response analysis of variance table ......................................................... 36	
   Table 3.5: Cell specific productivity response analysis of variance table .................................... 36	
   Table 3.6: DOE table for the second round of screening .......................................................... 37	
   Table 3.7: DOE analysis of the second screening experiment based on total MAb produced with the culture H removed. ............................................................................................................ 41	
   Table 3.8: DOE analysis of the second screening experiment based on final IVC with the culture H removed. .............................................................................................................................. 41	
   Table 3.9: e second round DOE table with augmented points shaded in blue and the culture H removed ............................................................................................................................... 42	
   Table 3.10: Parameter estimation for the total MAb polynomial model based on the significant parameters found in the half normal plot ................................................................................. 45	
   Table 4.1: Experimental design table for testing autophagy inhibition and hypertonicity effects on MAb production simultaneously ......................................................................................... 68	
   Table 4.2: P-values for the actual and normalized amounts of productivity and total MAb, using a one-tailed paired t-test compared to cultures without 3-MA .................................................. 73	
    viii  Table 4.3: P-values for the actual and normalized amounts of productivity and total MAb, using a one tailed paired t-test compared to cultures with low osmolality .......................................... 74	
   Table 4.4: Glycosylation analysis results for the heavy chain of the MAb produced by CHOMAb clones .............................................................................................................................. 79	
   Table 4.5: Glycosylation analysis results for the light chain of the MAb produced by CHO-MAb clones ....................................................................................................................................... 80	
   Table J.1: Experimental design table for testing autophagy inhibition and hypertonicity effects on MAb production simultaneously. e samples on which glycan analysis was performed are highlighted ............................................................................................................................. 126 Table J.2: e monosaccharide constituents of glycan structures and their graphical representations used in this thesis ........................................................................................... 126	
   Table J.3: e glycan structures identified and their graphical representations ........................ 127	
   Table J.4: Peak identification based on the GU values for the heavy chains of MAb samples and comparison of individual area percentage. .............................................................................. 130	
   Table J.5: Peak identification based on the GU values for the heavy chains of MAb samples and comparison of individual area percentage. .............................................................................. 133	
    ix  List of Figures Figure 3.1: Viable cell concentration of batch and fed-batch cultures of CHOm44 cell line ..... 26	
   Figure 3.2: Viability of batch and fed-batch cultures of CHOm44 cell line .............................. 26	
   Figure 3.3: Monoclonal antibody concentration of batch and fed-batch cultures of CHOm44 cell line ..................................................................................................................................... 27	
   Figure 3.4: Total amount of antibody produced in batch and fed-batch cultures, as well as the fed-batch volume change. e batch culture volume was assumed to be constant at 25 mL (Sampling and evaporation ignored)......................................................................................... 27	
   Figure 3.5: Viable cell concentrations of the cultures with the 3F feeding strategy .................... 32	
   Figure 3.6: Total amount of MAb produced in the cultures with the 3F feeding strategy ......... 32	
   Figure 3.7: Viable cell concentrations of the cultures with the 1F3F feeding strategy ............... 33	
   Figure 3.8: Total amount of MAb produced in the cultures with the 1F3F feeding strategy ..... 33	
   Figure 3.9: Glutamine concentration of the cultures with the with 3F feeding strategy; the initial zero level is based on the medium formulation ......................................................................... 34	
   Figure 3.10: Glutamine concentration in the cultures with the with 1F3F feeding strategy; the initial zero level is based on the medium formulation ............................................................... 34	
   Figure 3.11: Viable cell concentration of the cultures A, B, C, D and the centre point (culture I) ................................................................................................................................................. 38	
   Figure 3.12: Viable cell concentration of the cultures E, F, G, H and the centre point culture (culture I) ................................................................................................................................. 38	
   Figure 3.13: MAb concentration of the cultures A, B, C, D and the centre point culture (culture I) .............................................................................................................................................. 39	
    x  Figure 3.14: MAb concentration of the cultures E, F, G, H and the centre point culture (culture I) .............................................................................................................................................. 39	
   Figure 3.15: MAb weight of the cultures A, B, C, D plus the centre point culture (culture I) ... 40	
   Figure 3.16: MAb weight of the cultures E, F, G, H plus the centre point culture (culture I) ... 40	
   Figure 3.17: Viable cell concentration for the augmented points in the second screening the design ....................................................................................................................................... 42	
   Figure 3.18: MAb concentration for the augmented points in the second round of the design . 43	
   Figure 3.19: Total MAb produced for the augmented points in the second round of the design43	
   Figure 3.20: Half normal plot for the significance of factors based on the MAb response for the augmented design of experiments ............................................................................................. 44	
   Figure 3.21: Half normal plot for the significance of factors based on the IVC response for the augmented design of experiments ............................................................................................. 45	
   Figure 3.22: Selected amino acid results for the culture A, with feed increased on day 4 with 2x magnitude, no EN/CYD solution, no α-KG, with glutamine fed every day.............................. 47	
   Figure 3.23: Selected amino acid results for the culture B, with feed increased on day 4 with 2x magnitude, with both EN/CYD and α-KG and no glutamine feeding. .................................... 47	
   Figure 3.24: Selected amino acid results for the culture D, with feed increased on day 4 with 4x magnitude, with EN/CYD and glutamine feeding and no α-KG.............................................. 48	
   Figure 3.25: Selected amino acid results for the culture I (centre point), with feed increased on day 5 with 3x magnitude, with EN/CYD, α-KG and glutamine fed half the amount defined for other cultures. .......................................................................................................................... 48	
    xi  Figure 4.1: Viable cell and MAb concentrations of the CHOm44 fed-batch cultures treated with 3-MA ....................................................................................................................................... 52	
   Figure 4.2: After 3-MA addition average of cell specific productivity of the CHOm44 fed-batch culture with and without 3-MA addition ................................................................................. 53	
   Figure 4.3: Viable cell concentration of the cultures in the 3-MA dose response experiment for the CHOm56 cell line ............................................................................................................. 55	
   Figure 4.4: MAb concentrations for fed-batch cultures of CHOm56 treated with different 3-MA doses ........................................................................................................................................ 55	
   Figure 4.5: Average cell specific productivity of fed-batch cultures of CHOm56 after 3-MA addition ................................................................................................................................... 56	
   Figure 4.6: Comparison of the final MAb weight produced in 3-MA treated cultures treated of CHOm56 (C56) and CHOm44 (C44) clones, using EN/CYD 5X and 1X solutions respectively ................................................................................................................................................. 57	
   Figure 4.7: Selected amino acid concentrations of the CHOm56 fed-batch culture with no 3MA treatment .......................................................................................................................... 58	
   Figure 4.8: Selected amino acid concentrations of the CHOm56 fed-batch culture treated with 5 mM of 3-MA ........................................................................................................................... 59	
   Figure 4.9: Selected amino acid concentrations of the CHOm56 fed-batch culture treated with 10 mM of 3-MA ...................................................................................................................... 59	
   Figure 4.10: Viable cell concentrations for CHOm44 cultures treated with 3-MA on different dates ......................................................................................................................................... 61	
   Figure 4.11: MAb concentration profiles of CHOm44 cultures treated on different dates ........ 62	
    xii  Figure 4.12: Comparison of the average cell specific productivities of the cultures after 3-MA treatment (red bars) to the control cultures throughout the same time. .................................... 62	
   Figure 4.13: Comparison of the final MAb weight produced in cultures treated with 3-MA, using EN/CYD 5X and 1X solutions........................................................................................ 63	
   Figure 4.14: Viable cell concentrations of CHOm44 batch cultures under different osmolalities ................................................................................................................................................. 65	
   Figure 4.15: Viable cell concentrations of CHOm56 batch cultures under different osmolalities ................................................................................................................................................. 65	
   Figure 4.16: e average cell specific of CHOm44 and CHOm56 batch cultures under different osmolalities .............................................................................................................................. 66	
   Figure 4.17: e final MAb concentration of CHOm44 and CHOm56 batch cultures under different osmolalities ................................................................................................................ 66	
   Figure 4.18: Change of osmolality for cultures with EN/CYD 1X that undergo the lower osmolality regime ..................................................................................................................... 69	
   Figure 4.19: Change of osmolality for cultures with EN/CYD 5X that undergo the higher osmolality regime ..................................................................................................................... 69	
   Figure 4.20: Average of normalized cell specific productivity in the duration after addition of 3MA of different CHO-MAb clones with high and low osmolalities .......................................... 71	
   Figure 4.21: Average of normalized final weight of MAb produced in fed-batch cultures of different CHO-MAb clones with high and low osmolalities ..................................................... 71	
   Figure 4.22: Average of normalized cell specific productivity in the duration after addition of 3MA of different CHO-MAb clones with different 3-MA concentrations .................................. 72	
    xiii  Figure 4.23: Average of normalized final weight of MAb produced in fed-batch cultures of different CHO-MAb clones with different 3-MA concentrations ............................................. 72	
   Figure 4.24: Comparison of the relative percentage of the glycoforms found in the heavy chains of MAb produced by CHOm44 cells at different osmolality regimes and 3-MA concentrations... ................................................................................................................................................. 78	
   Figure 4.25: Comparison of the relative percentage of the glycoforms found in the heavy chains of MAb produced by CHOm56 cells at different osmolality regimes and 3-MA concentrations... ................................................................................................................................................. 78	
   Figure A.1: Viable cell concentrations of CHOm56 fed-batch cultures in different volumes cultured in 125 mL shake flasks ............................................................................................. 100 Figure A.2: Monoclonal antibody concentrations of CHOm56 fed-batch cultures in different volumes cultured in 125 mL shake flasks ................................................................................ 100	
   Figure B.1: Glucose concentration of batch and fed-batch cultures of CHOm44 cell line ...... 101	
   Figure B.2: Lactate concentration of batch and fed-batch cultures of CHOm44 cell line ....... 101	
   Figure B.3: Glutamine concentration of batch and fed-batch cultures of CHOm44 cell line .. 102	
   Figure C.1: Viability of the cultures with the 3F feeding strategy ........................................... 103	
   Figure C.2: Viability of the cultures with the 1F3F feeding strategy ....................................... 103	
   Figure C.3: e MAb concentrations produced in the cultures with the 3F feeding strategy .. 104	
   Figure C.4: e MAb concentrations produced in the cultures with the 1F3F feeding strategy ............................................................................................................................................... 104	
   Figure C.5: e glucose concentration profiles of the cultures with the 3F feeding strategy .... 105	
   Figure C.6: e glucose concentration profiles of the cultures with the 1F3F feeding strategy 105	
    xiv  Figure C.7: e lactate concentration profiles of the cultures with the 3F feeding strategy ..... 106	
   Figure C.8: e lactate concentration profiles of the cultures with the 1F3F feeding strategy . 106	
   Figure C.9: e integral of viable cells of the cultures with the 3F feeding strategy ................. 107	
   Figure C.10: e integral of viable cells of the cultures with the 1F3F feeding strategy .......... 107	
   Figure C.11: e cell specific productivity of the different fed-batch cultures (Table 3.2) ...... 108	
   Figure D.1: Viability of the cultures A, B, C, D and the centre point (culture I) .................... 109	
   Figure D.2: Viability of the cultures E, F, G, H and the centre point (culture I) .................... 109	
   Figure D.3: Viability of the cultures I2, I3, I4 and Hi (the augmented points in Table 3.9) ...... 110	
   Figure D.4: e integral of viable cells of the cultures A, B, C, D and the centre point (culture I) ............................................................................................................................................... 110	
   Figure D.5: e integral of viable cells of the cultures E, F, G, H and the centre point (culture I) ............................................................................................................................................... 111	
   Figure D.6: e integral of viable cells of the cultures I2, I3, I4 and Hi (the augmented points in Table 3.9) .............................................................................................................................. 111	
   Figure E.1: Viability of the CHOm44 fed-batch cultures with and without 3-MA................. 112	
   Figure E.2: e cell specific productivity of the CHOm44 fed-batch cultures with and without 3-MA ..................................................................................................................................... 112	
   Figure F.1: Cell specific productivity for CHOm56 fed-batch cultures of CHOm56 treated with 2.5 and 5 mM of 3-MA compared to the 3-MA-free control culture ...................................... 113	
   Figure F.2: Cell specific productivity of CHOm56 fed-batch cultures of CHOm56 treated with 7.5, 10 and 15 mM of 3-MA compared to the 3-MA-free control culture .............................. 113	
    xv  Figure F.3: Lysosomal content of CHOm56 fed-batch cultures of CHOm56 treated with 2.5 and 5 mM of 3-MA compared to the 3-MA-free control culture ............................................ 114	
   Figure F.4: Lysosomal content of CHOm56 fed-batch cultures of CHOm56 treated with 7.5, 10 and 15 mM of 3-MA compared to the 3-MA-free control culture ..................................... 114	
   Figure G.1: Viability of CHOm44 cultures treated with 3-MA on different dates .................. 115	
   Figure G.2: Cell specific productivity CHOm44 cultures treated with 3-MA on different dates ............................................................................................................................................... 115	
   Figure H.1: Viability of CHOm44 batch cultures under different osmolalities ....................... 116	
   Figure H.2: Viability of CHOm56 batch cultures under different osmolalities ....................... 116	
   Figure H.3: Cell specific productivity of CHOm44 batch cultures under different osmolalities ............................................................................................................................................... 117	
   Figure H.4: Cell specific productivity of CHOm56 batch cultures under different osmolalities ............................................................................................................................................... 117	
   Figure I.1: Viable cell concentrations for the CHOm44 cell cultures with different osmolality regimes and different 3-MA concentrations ............................................................................ 118	
   Figure I.2: MAb concentrations for the CHOm44 cell cultures with different osmolality regimes and different 3-MA concentrations......................................................................................... 118	
   Figure I.3: Viable cell concentrations for the CHOm56 cell cultures with different osmolality regimes and different 3-MA concentrations ............................................................................ 119	
   Figure I.4: MAb concentrations for the CHOm56 cell cultures with different osmolality regimes and different 3-MA concentrations......................................................................................... 119	
    xvi  Figure I.5: Total number of cells for CHOm44 cultures with different osmolality regimes and 3MA concentrations ................................................................................................................. 120	
   Figure I.6: Total number of cells for CHOm56 cultures with different osmolality regimes and 3MA concentrations ................................................................................................................. 120	
   Figure I.7: Average cell specific productivity in the duration after addition of 3-MA of different CHO-MAb clones with high and low osmolalities ................................................................. 121	
   Figure I.8: Final weight of MAb produced in fed-batch cultures of different CHO-MAb clones with high and low osmolalities ............................................................................................... 121	
   Figure I.9: Viable cell concentrations for the CHOm44 cell replicate cultures with different osmolality regimes and different 3-MA concentrations ........................................................... 122	
   Figure I.10: MAb concentrations for the CHOm44 replicate cell cultures with different osmolality regimes and different 3-MA concentrations ........................................................... 122	
   Figure I.11: Viable cell concentrations for the CHOm56 replicate cell cultures with different osmolality regimes and different 3-MA concentrations ........................................................... 123	
   Figure I.12: MAb concentrations for the CHOm56 cell cultures with different osmolality regimes and different 3-MA concentrations ............................................................................ 123	
   Figure I.13: Total cell number for the CHOm44 replicate set of cultures with different osmolality regimes and 3-MA concentrations ......................................................................... 124	
   Figure I.14: Total cell number for the CHOm56 replicate set of cultures with different osmolality regimes and 3-MA concentrations ......................................................................... 124	
   Figure I.15: Average cell specific productivity in the duration after addition of 3-MA for the replicate cultures of different CHO-MAb clones with high and low osmolalities .................... 125	
    xvii  Figure I.16: Final weight of MAb produced in replicate cultures of different CHO-MAb clones with high and low osmolalities ............................................................................................... 125	
   Figure J.1: Overlay of HPLC elution profiles and peak identification for the heavy chains of the MAb samples from A, B, D and E cultures. All peaks with more than 2% of area are identified. ............................................................................................................................................... 128	
   Figure J.2: Overlay of HPLC elution profiles and peak identification for the heavy chains of the MAb samples from G, H, J and K cultures. All peaks with more than 2% of area are identified. ............................................................................................................................................... 129	
   Figure J.3: Overlay of HPLC elution profiles for the light chains of the MAb samples from A, B, D and E cultures. ................................................................................................................... 131	
   Figure J.4: Overlay of HPLC elution profiles for the light chains of the MAb samples from G, H, J and K cultures. ............................................................................................................... 132	
    xviii  List of Abbreviations  3-MA  3-methyladenine  Asn  Asparagine  Asp  Aspartate  CNS  Central nervous system  CHO  Chinese hamster ovary  Cys  Cysteine  DMSO  Dimethyl sulfoxide  GFP  Green fluorescence protein  GlcNAc  N-acetylglucosamine  Gln  Glutamine  Glu  Glutamate  GS  Glutamine synthetase  IgG  Immunoglobulin G  IVC  Integral of viable cells  LC3  Microtubule-associated protein 1 light chain 3  MAb  Monoclonal antibody  mTOR  Mammalian target of rapamycin  PCD  Programmed cell death  PI3K  Phosphatidylinositide 3-kinase  RFP  Red fluorescence protein  xix  RFU  Relative fluorescence unit  Ser  Serine  r  reonine  Tyr  Tyrosine  tPA  Tissue plasminogen activator  xx  Acknowledgements  All my gratitude is due to the Almighty God; whose praise I cannot commend sufficiently, and whose generosity I cannot count entirely. He is praised for everything, difficult or simple; and in every time, turbulent it be or serene. e ultimate goal, without whom no endeavour is worth attempting, and the absolute aspiration, for whom no effort is in vain. I am forever grateful to my supervisors Dr. James Piret and Dr. Bhushan Gopaluni. Jamie has been a great mentor, whose steadfast support, unwavering patience and realistic optimism were great helps for this thesis to materialize and this project to conclude. Bhushan has been a wonderful supervisor, whose insightful inputs and enthusiastic encouragement have always been helpful. ank you both for providing me the opportunity to work here and be a part of your team. It was a pleasure to work with both of you, and an unforgettable experience to be carried to other stages of life. First among all of Piret Lab members, I would like to thank Dr. Mario Jardon, for paving the way for this project. Without your vast knowledge and kind support this project would have never come to pass. I would like to thank Dr. Malcolm Kennard, and appreciate his help and feedback from the beginning to the end of this project. My cordial thanks belong to Chris Sherwood, for running the lab and helping with everything, including some of my experiments. ank you Pascal, Kelsey, Navid, Xinbo, Venkata, René and Roger for being great friends and colleagues. anks to Steve, Vince and Victor at Michael Smith Laboratories, for their help in providing uninterrupted research experience. Also, I would like to thank Katrin Braasch and Dr.  xxi  Michael Butler for collaborating with this project. MAbNet and Pfizer are also acknowledged for providing funding and cell line for this research. ese years at the UBC would have never been this amazing without Mohammad-Sadegh, Mostafa, Mehdi, Alireza, Amin, Ali and each and every other friend who has had and will ever have a part in my heart. ank you all for your companionship, support and care. Last but not least, I extend my utmost respect and gratitude to my family. ank you my dear father for being such an excellent friend and life-long model; thank you my adored mother for your endless love and warm affection; and thank you my beloved brothers, Amir-Hossein and Ali, for all the great years passed together, and all the fondness they bestowed upon me.  xxii  Dedication  To my late grandfather, Haj Baba, whom I lost, while being this far away… To you, though you can never see this…  xxiii  Chapter 1: Introduction  1.1  Biopharmaceutical Production Using Mammalian Cell Culture Production of recombinant proteins using genetically engineered cells has gained a great  degree of importance in the medical and pharmaceutical industry (Butler 2005; Pavlou and Reichert 2004; Werner 2004; Wurm 2004). ese proteins are expanding the modern therapeutic arsenal for the treatment of heart disease, stroke, cancer and central nervous system diseases (Pavlou and Reichert 2004), generating a multi–billion dollar market (Werner 2004; Aggarwal 2011). Various host cell lines such as yeast, insect or mammalian cells have been used to develop cell lines capable of producing biopharmaceuticals. e host cell lines of choice for the most complex proteins are mammalian cells; especially since these cells perform complex post-translational protein modifications, required for protein function. For example, protein glycosylation (i.e. sugar molecules linked to proteins) can significantly affect protein half-life, immunogenicity and clinical efficacy. Prokaryotic hosts such as E. coli do not glycosylate proteins, and lower eukaryotic expression systems such as yeast and insect cells do not provide correct mammalian glycosylation. erefore mammalian cells and particularly Chinese Hamster Ovary (CHO) cells have emerged as the most widely used expression system for the production of complex therapeutic recombinant proteins (Gerngross 2004).  1  1.2  Mammalian Cell Cultures in Industrial Settings e most common industrial mammalian cell production uses fed-batch culture and serum-  free media in stainless steel stirred tanks or, increasingly, disposable bioreactors (Butler 2005; Chu and Robinson 2001; Hou et al. 2011). Stainless steel stirred tank bioreactors are used in largest-scale fed-batch mammalian cell culture systems, with working volumes up to 20000 L (S.-Y. Lee et al. 2012). In fed-batch processes concentrated nutrients are fed to maintain depleted nutrient levels. e feeding can be continuous or intermittent (Wlaschin and Wei-Shou Hu 2006), although intermittent feeding is more common to simplify operations. Previously, cell culture media were supplemented with fetal bovine serum, containing many growth factors. is is now being phased out and replaced with serum-free media to reduce animal product risks especially due to the outbreak of mad cow disease (Butler 2005). us present industrial practice is using animal-component free media with many advantages, including much reduced cost as well as lot-to-lot variations (a major problem of serum supplements).  1.3  Fed-Batch Processes Fed-batch processes provide the cells with necessary nutrients that are depleted over time and  for a variety of reasons are not added in excess to the initial medium (e.g. substrate toxicity or solubility limits). Feeding can extend the culture duration while maintaining cell productivity, thereby increasing the total amount of protein produced compared to a non-fed batch culture. Feeding can be continuous or intermittent, and either with pre-specified (Bibila et al. 1994) or adaptive amounts. Adaptive feeding can be based on the rate of nutrient consumption (W. 2  Zhou et al. 1997; Frahm et al. 2003) or maintaining a target nutrient concentration (Sauer et al. 2000), and operates with feedback control based on the monitoring of specific nutrient concentrations. Although using adaptive feeding may better suit a culture for a longer duration or higher production, it tends to be more complex and requires sophisticated equipment with more complicated operational needs. Feeding protocols have been tailored to keep nutrients at high or low levels (for example glucose or glutamine) in order to reduce the generation of undesired metabolites or by-products (such as lactic acid or ammonia). Generally, the performance of a cell culture is based on maximum cell concentrations, cell specific productivities and final product titres. In the case of CHO cells, final product titres have increased from 50-100 mg/L in the late 1980s to over 5 g/L (Wurm 2004). Maximum cell concentrations have increased from 1-2 to 10-15 million cells/mL (Wurm 2004), with reports of over 18 million cells/mL in fed-batch (Luo et al. 2012; J. Li et al. 2012). e reported cell specific productivities have increased from below 10 pg/cell.day in the 1980s to over 90 pg/cell.day (Wurm 2004). Culture times vary from ~7 days in batch culture up to ~20 days in fed-batch culture (Wurm 2004). Within the course of a fed-batch process, cultures stresses decrease growth rates, productivity and viability (the percentage of viable cells). A cellular stress can be defined as a fluctuation in external conditions of the cell, including physical, chemical or extracellular signals that have gone beyond a certain threshold. e response of the cell to the stresses will determine whether it can survive and continue to function (Kroemer et al. 2010). Cellular stresses can result from metabolite accumulation, osmolality increases, nutrient depletion, oxygen limitation,  3  shear stresses from bubble bursting, etc. ese stresses can cause cells to stop growing or producing proteins, or trigger cellular death mechanisms. Cellular death occurs through necrosis or apoptosis and has also been associated with autophagy; where apoptosis and autophagy are described as programmed cell death (PCD) pathways. Necrosis is premature cellular death that arises from sudden exposure to cellular stresses such as the shear force of an impeller, or extreme pH change (Mohan et al. 2009). Unlike necrosis, in which cells have no control over the process of death, programmed cell death pathways are mediated by well-characterized pathways.  1.4  Programmed Cell Death  1.4.1  Apoptosis  Apoptosis (known as type I PCD) is a non-reversible cellular suicide process in animal cells initiated in response to accumulation of internal or external non-lethal stimuli (Y. Lim et al. 2010). Internal apoptosis stimuli include hypoxia, lack of growth factors, DNA damage and reactive oxygen species. External signaling molecules can activate pro-apoptotic receptors on cell surfaces. Once initiated, apoptosis results in cell shrinkage, chromatin condensation and fragmentation, and plasma membrane blebbing. Apoptosis is a major death pathway in mammalian cell cultures, limiting the maximum cell density and leading to release of intracellular enzymes that may be detrimental to the product quality (Mohan et al. 2009). Consequently, anti-apoptotic engineering of cell lines to delay the onset of apoptosis has been an important and widely addressed research topic (Arden and Betenbaugh 2004). Either over4  expressing anti-apoptotic genes or knocking out pro-apoptotic genes have been used to downregulate apoptotic pathways (Y. Lim et al. 2010). Also addition of galactose in lieu of glucose or nucleosides such as adenosine are non-genetic approaches to mitigate the severity of apoptosis in mammalian cell cultures (Costa et al. 2010).  1.4.2  Autophagy  Autophagy (derived from Greek words for ‘self-eating’) is a eukaryotic cellular process whereby cellular components are non-specifically degraded by the lysosomal machinery of cells. Although it has been referred to as type II PCD, it is not a pure death mechanism. Autophagy occurs throughout lifecycle of a cell at housekeeping levels to degrade old or damaged organelles, but may be upregulated in response to cellular stresses such as starvation, lack of growth factors or accumulation of misfolded proteins (Kundu and ompson 2008), that may lead to cell survival or death. ere are three types of autophagy: macro-autophagy, micro-autophagy and chaperone mediated autophagy. Macro-autophagy is responsible for most of the intracellular protein degradation (Mizushima et al. 2002) and is called autophagy hereafter. In most cases, autophagy is induced through the inhibition of a protein called the mammalian target of rapamycin (mTOR). mTOR is a central metabolic sensor of the cell, involved in processing the information of (a) cell energy, amino acid and growth factor responses, (b) protein translation and ribosomal protein synthesis, (c) vesicular traffic, (d) cell size and (e) cell proliferation (Dreesen and Fussenegger 2010). When there is no limitation of nutrients, mTOR remains active and phosphorylates several proteins that direct cell growth. mTOR inactivation due to nutrient limitations reduces overall protein translation and up5  regulates proteins needed for survival rather than those needed for growth. Autophagy is then initiated as an alternate route to provide the cells with ATP and amino acids (Kundu and ompson 2008). However, cases of mTOR-independent autophagy are also reported, for example in presence of reactive oxygen species or endoplasmic reticulum stress (Kroemer et al. 2010; Maiuri et al. 2007). Autophagy starts with the formation of isolated double-membrane vesicles called autophagosomes. ese enclosed components are digested upon the subsequent fusion of these autophagosomes with lysosomes, resulting in the recycle of cellular components or the generations of energy (Kundu and ompson 2008). ere are more than 20 genes involved in autophagy (called ATG genes in yeast) (Kundu and ompson 2008), and many more that participate in related pathways, such as cell growth etc. (Klionsky et al. 2003). Vesicle formation is mediated by the activation of a mammalian class III phosphatidylinositide 3-kinase (PI3K). PI3K is a complex of proteins, namely Vsp34, p150, Atg14 and Beclin-1 (He and Klionsky 2009). Beclin-1 is activated during starvation by dissociation of a protein called Bcl-2 (Pattingre et al. 2005). Vesicle elongation takes place with the recruitment of two conjugation systems that both involve ubiquitin-like proteins. One of these conjugation systems is used to assay autophagy. is system involves the microtubuleassociated protein 1 light chain 3 (MAP1-LC3, commonly known as LC3). LC3-I (the soluble form of LC3) is cleaved by Atg4 and subsequently conjugates to phosphatidyl ethanolamine by Atg7 and Atg3. is lipidated form, LC3-II, attaches to the outer membrane of autophagosomes. Atg4 completes the formation of autophagosomes upon cleavage of phosphatidyl ethanolamine from LC3-II. 6  Lysosomes fuse with autophagosomes to form ‘autolysosomes’, mediated by LAMP-2 (lysosomal-associated membrane protein 2) and several other proteins. e inner membrane and contents of autophagosomes are digested by lysosomal enzymes (Maiuri et al. 2007). Degraded molecules, mostly amino acids are recycled to cytosol (He and Klionsky 2009). e ratio of LC3-II to LC3-I, measured by western blotting, is commonly used as a marker for autophagosomes (Klionsky et al. 2009; Rubinsztein et al. 2009). Tagging LC3 with GFP (green fluorescent protein), using genetic engineering techniques to create this fusion protein allows for the visualization and measurement of autophagosomes (Maiuri et al. 2007; Shvets et al. 2008). Since the GFP is sensitive to lysosomal hydrolases, the GFP-LC3 fusion protein is unable to show the end stages of autophagy, when lysosomes fuse in autophagosomes. To address this problem, tandem reporter LC3 proteins are developed with both GFP and RFP (red fluorescent protein). Upon fusion of lysosomes to autophagosomes, despite the digestion of GFP, RFP remains intact (Barth et al. 2010; S. Kimura et al. 2007). e fluorescent vesicles can be measured using fluorescence-activated cell sorting (FACS) or fluorescence microscopy (Maiuri et al. 2007; Shvets et al. 2008). e acidotropic stains such as LysoTracker can to some extent monitor autophagy in this stage, as these stains accumulate in the acidic autolysosomes. However, these stains detect other acidic organelles and so this method is non-specific (Klionsky et al. 2009).  1.5  Effect of Autophagy Inhibition on Protein Production Mammalian cell culture stresses can lead to autophagy in ways that are not usually found in  vivo such as glucose limitation, hypoxia and expression of aggregate prone proteins (He and 7  Klionsky 2009). Onset of autophagy can disrupt the protein production and processing machinery. us, study of autophagy and its effect on cells and their recombinant protein production in the bioprocessing context has been relevant and rewarding. Recently, (Jardon et al. 2012) observed that inhibition of autophagy in glutamine-limited cells by 3-methyladenine (3-MA) increased the amount of tissue plasminogen activator (tPA) protein production in CHO cells by almost 3-fold. 3-MA is a class III PI3K inhibitor that is widely used for autophagy inhibition in research (Seglen and Gordon 1982; Wu et al. 2010; Klionsky et al. 2009; Barth et al. 2010; Mohan et al. 2009). 3-MA inhibits autophagy by preventing the formation of autophagosomes. Although deemed non-toxic by some authors (Hwang and G. M. Lee 2008; Mohan et al. 2009),  3-MA can reduce cell viability and  proliferation rates, likely due to its non-specific inhibition of both class I and III PI3Ks (Wu et al. 2010). is effect of autophagy inhibition on protein production is part of the foundation for this thesis, investigating these phenomena in MAb producing CHO cells.  1.6  Osmolality in Cell Culture e osmolality of a solution is defined as the total moles of solute molecules per weight of  the solvent, and is usually given in units of milli-osmoles per kg or mOsm/kg. e osmolality of cell cultures is usually between 260 – 320 mOsm/kg, close to the 290 mOsm/kg osmolality of blood (Ozturk and Palsson 1991). In this thesis, osmolalities greater than 330 mOsm/kg are defined as hypertonic, and those in the range of 260 – 330 mOsm/kg are defined as isotonic conditions.  8  Hypertonic conditions can have many impacts on mammalian cells. At least 200 cell components including mRNAs and proteins are sensitive to osmolality (Burg et al. 2007). Acute elevation of osmolality can cause DNA damage and cell cycle arrest. Apoptosis, oxidative stress, inhibition of RNA transcription and translation, and mitochondrial depolarization can also occur with increased osmolality with the levels depending on the cell type. In addition, hypertonicity can affect other cellular responses related to ion transport, osmolyte accumulation and cytoskeletal organization. Hypertonicity in the medium causes water to flow out of the cells in response to the osmotic pressure. is can cause volume shrinkage of the cell and elevation of the concentration of cell components. A regulatory volume increase (RVI) response is initiated in cells facing hypertonic conditions (Kiehl et al. 2011), with a rapid intake of inorganic osmolytes by activated channels such that cells can re-establish an equilibrium with their surrounding and retain the cell size. However, this results in abnormally high intracellular ion concentrations and can perturb protein functions. After hours of hypertonic induction, cells can accumulate more compatible organic osmolytes and decrease the intracellular concentration towards those of isotonic conditions while maintaining cell volume. e accumulated organic osmolytes such as taurine, sorbitol, glycine betaine, glycerophosphocholine (GPC) and myo-inositol are less perturbing to protein functions (Burg et al. 2007). is accumulation results from upregulated production, increased intake or downregluated breakdown of organic osmolytes (Wehner et al. 2003). Despite induction of cellular stresses and reducing cell growth, hypertonic cell culture has also been shown to increase cell specific productivity in mammalian cell lines (Han et al. 2009; N. S. Kim and G. M. Lee 2002; T. K. Kim et al. 2000; Ozturk and Palsson 1991; Rönsch et al. 9  2003; Takagi et al. 2001). As the osmolality of cell culture media can easily be changed by addition of inorganic or organic osmolytes (such as NaCl or sorbitol), this can provide an easyto-implement method to increase protein production. However, the increase specific productivity may be counteracted by the reduction in cell growth, resulting in no gain in the overall production. Hypertonic conditions have also been reported to affect the glycosylation patterns of monoclonal antibodies produced by CHO cells (Konno et al. 2012; Hossler et al. 2009). Biphasic batch culture strategies or gradual increases of osmolality have been used to overcome the problem of cell growth suppression (M. S. Kim et al. 2002; C. F. Shen and Kamen 2012; Takagi et al. 2001). However, most of these studies have been performed under batch conditions and thus the study of hypertonicity under fed-batch cultures is worthy of further study.  1.7  Protein Glycosylation Most of the therapeutic proteins approved for human use undergo post-translational  modification, such as protein glycosylation, i.e. covalent linking of sugar structures to proteins. Glycosylation can influence the glycoprotein folding, stability, trafficking, immunogenicity, ligand recognition/binding and half-life. Erythropoietin, antibodies and blood factors are among the proteins shown to be dependent on particular glycoforms for their function (Walsh and Roy Jefferis 2006). e impact of glycosylation on clinical function has caused regulatory agencies to require that the glycan profiles of therapeutic proteins be kept within specified limits, set by a glycoform profile of the product that has tested as part of the approval process (R Jefferis 2009).  10  To sufficiently match these profiles is a major challenge faced by generic biopharmaceutical producers who seek to market patent-expired (biosimilar) therapeutics (Schellekens 2004). Although eukaryotic cells and some prokaryotes glycosylate proteins, the resulting glycoforms often have significant differences compared to those generated by human cells (R Jefferis 2009). For examples, high mannose glycoforms produced by yeast or high fucose and xylose glycoforms produced by plant cells result in shorter half-lives in vivo and may render the protein less potent or more immunogenic (Werner et al. 2007; Walsh and Roy Jefferis 2006). e ability of CHO cells to produce human-like glycan structures is one of the important reasons for their selection as industrial recombinant protein producers (Bennun et al. 2009). e glycans attached to proteins are in 2 categories: N-linked and O-linked. Glycan structures are attached through amide groups in asparagine in the case of N-linked glycosylation, or through hydroxyl groups of either serine or threonine in the case of O-linked glycosylation (Bennun et al. 2009). O-linked glycosylation is relatively rare in secreted proteins, including IgG (Allen et al. 1995). N-linked glycan structures are connected to an asparagine residue through the Nacetylgalactosamine (GlcNAc) residue of the glycan. ese asparagine residues are located in Asn-Xxx-(Ser, r) motifs, where Xxx can be any amino acid excluding proline (Taylor and Drickamer 2011). N-linked glycosylation starts in the rough ER, with the formation of a lipidlinked precursor oligosaccharide with 9 GlcNAc, 9 mannose and 3 glucose residues (Glc3Man9GlcNAc2). e precursor oligosaccharide is transferred en bloc to the polypeptide, and is trimmed and processed by several glycosidase and glycosyltransferase enzymes on its way to the Golgi apparatus. e glycan processing steps include: 11  a) Removal of glucose residues b) Trimming from 9 mannose residues to the minimum of 3 c) Addition of GlcNAc residues d) Addition of fucose residues e) Addition of galactose residues f) Addition of sialic acid residues Not all oligosaccharides are completely processed, and the structure of resulting glycans vary considerably (Taylor and Drickamer 2011). e oligosaccharides of human IgG-Fc is usually biantennary, complex and heterogeneous, with sparse (<10%) sialylation (Walsh and Roy Jefferis 2006). Antibodies produced by CHO cells include high amounts of non-galactosylated fucosylated glycan structures (G0F) as well as fucosylated structures with one (G1F) or two (G2F) galactose residues (Del Val et al. 2010). High-mannose structures are limited in IgG produced by CHO cells (Kornfeld et al. 1978) e heavy chains of IgGs have a glycosylation site in their Fc region, located at the CH2 domain at the asparagine 297 (Asn297) residue (Wright and Morrison 1997). Glycosylation at this conserved glycan site has been shown to be necessary for antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) (R Jefferis 2009). However, there is little or no direct contact between the IgG glycan structures and the receptors and complement components necessary for ADCC and CDC functions. Instead, the IgG glycan structure is buried inside the Fc structure and makes substantial contact with CH2 and CH3 domains. Absence of IgG glycan structures results in proximity of these domains leading to  12  ablated antibody functionality (Taylor and Drickamer 2011). Interestingly, afucosylated antibodies are also shown to have 20- to 100-fold higher ADCC (Walsh and Roy Jefferis 2006). Approximately 15% to 20% of human antibodies are glycosylated in the Fab region, located within the variable regions of light or heavy chains (Walsh and Roy Jefferis 2006). Light chain glycosylation has only been observed in the variable region (VL) (Fujimura et al. 2000). e glycan structure of the variable region can influence the affinity of the antibody for the antigen (Del Val et al. 2010). e variable region glycosylation can both increase (Wallick et al. 1988; Leibiger et al. 1999; Tachibana et al. 1997) or decrease (Fujimura et al. 2000; Co et al. 1993) the antigen affinity of antibodies. e magnitude of influence on affinity is dependent on the glycosylation site (Wright et al. 1991). Unlike Fc-linked glycans, higher galactosylation and sialylation are reported in Fab-linked glycans (Endo et al. 1995), perhaps due to more accessible sites on the Fab region for further glycan processing in the Golgi apparatus. Different culture methods have been shown to affect the glycosylation patterns of proteins (Butler 2005; Patel et al. 1992; Andersen et al. 2000; R. Kimura and Miller 1999). For example, high osmolality can increase the proportion of less processed high mannose glycoforms in MAb producing fed-batch CHO cultures. Fed-batch cultures with starting and final osmolalities of 300 and 395 mOsm/kg, respectively, produced IgG with 53% more Man5 structures (Pacis et al. 2011). High mannose glycan structures increase clearance rates by mannose receptors in humans (Wright and Morrison 1994). Other reports indicate higher ADCC in high-mannose IgGs due to their lack of fucose (Zhou et al. 2008). In general, changes in glycosylation patterns of a particular antibody due to culture conditions is important, especially when both antibody efficacy and regulatory approval can be influenced. 13  1.8  Research Goals As many biopharmaceutical patents are expiring, and generic production becomes more  important, there is an increasingly need to develop high performance protein production methods. e main hypothesis of this thesis was that the increased production by autophagy inhibition established for a tPA producing CHO cell line (CHO-tPA) could be applied to the more widely needed MAb production. e specific objectives are: 1-  To systematically adapt the CHO-tPA fed-batch protocol (Jardon et al. 2012) to MAb producing CHO cells (CHO-MAb).  2-  To explore the use autophagy inhibition for increased MAb production.  3-  To investigate the effects of hypertonicity influences on fed-batch productivity, including interactions with autophagy inhibition.  4-  To determine if there are significant influences on the MAb glycan structures produced under hypertonic and autophagy-inhibited conditions.  Chapter 2 describes the thesis Materials and Methods. Chapter 3 discusses the development of the MAb culture protocol based on the CHO-tPA fed-batch process. Chapter 4 investigates autophagy inhibition and hypertonicity effects on fed-batch process productivity and product quality. And finally, Chapter 5 presents the conclusions and recommendations for future work.  14  Chapter 2: Materials and Methods  2.1  Cell Lines Two clones of Chinese Hamster Ovary (CHO) cells (a gift from Pfizer Inc.) were used in  this project (Kennard et al. 2009). Both clones were derived from Lonza’s CHOK1SV CHO cell line and produced the human monoclonal anti-interleukin 1 β IgG1 antibody. e clones were originally named ChK2 437.90.44 and ChK2 437.89.56 and in this thesis they are referred to as CHOm44 and CHOm56, respectively.  2.2  Cell Culture Cell cultures were performed in 25, 125 and 500 mL sterile shake flasks (Corning and  Nalgene) in a 5% CO2 humidified incubator (Infors, Basel, Switzerland) at the temperature of 37ºC, with the speed of 140 rpm.  2.2.1  Maintenance Medium and Cultures  Cells were stored in 1 mL vials at 107 cell/mL and -150ºC (See 2.2.3). e maintenance medium for the cells was CD CHO (Gibco-Invitrogen, Grand Island, NY) supplemented with 6 mM glutamine (Gibco), 4.5 µg/mL bleocin (Calbiochem, La Jolla, CA) and 100 µg/mL hygromycin B (Gibco). ese antibiotics were intended to maintain selection pressure during thawing and maintenance of cells. ese vials were thawed rapidly in a 37ºC water bath. e 1 mL of cell suspension was transferred to a 125 mL flask containing 20 mL of pre-warmed maintenance medium, without 15  removing the DMSO. Newly thawed cells were passaged at least 3 times before being used in experiments. e normal maintenance cell culture routine was to passage cells when they reached 2×106 cells/mL, diluting them back to 0.2×106 cells/mL. Before starting a new set of experiments, the volume of maintenance cultures were increased and they were transferred to larger flasks if needed. Prior to using the cells as experimental inocula, cells were checked for their growth rate and viability, and were used while at their exponential growth phase. Before inoculation, the conditioned maintenance medium was removed by centrifugation, and cells were re-suspended in fresh maintenance medium.  2.2.2  Batch and Fed-batch Cultures  e initial fed-batch protocol for CHO-MAb fed-batch cultures was based on the protocol for CHO-tPA (Jardon et al. 2012) fed-batch processes. CHO-tPA batch and fed-batch cultures were performed in CD CHO medium (Gibco) supplemented with 25 ng/mL insulin-like growth factor (IGF), 4 mM glutamine and 4x anti-clumping agent (Jardon et al. 2012). Most of the CHO-tPA cultures were started at the initial viable cell concentration of 0.3×106 cells/mL. However, the initial cell concentration was not fixed in the CHO-tPA protocol, and there were many cultures started at higher concentrations such as 1×106 cells/mL. Fed-batch cultures were fed daily with CHO CD Efficient Feed A (Feed A) (Gibco). Table 2.1 shows part of the composition of CD CHO and Feed A solutions. In addition, two mixtures of other amino acid feed solutions were added to the cultures with the following concentrations defined as 1X: (a) Amino acid mixture I (EN) consisted of 10 mM L-glutamic acid (Sigma, St. Louis, MO) and 75 mM L-asparagine (Sigma) in 0.1 M NaOH; (b) Amino acid 16  mixture II (CYD) contained 10 mM L-cystine disodium salt hydrate (MP Biomedicals, Illkirch, France), 15 mM L-tyrosine disodium salt (Sigma) and 10 mM aspartic acid (Invitrogen) dissolved in 0.1 M HCl. In the platform process, the two amino acids mixtures were prepared at 5X concentration and then diluted in cell culture grade water before addition to the cultures, although in some experiments, the 5X concentration was used directly. e EN and CYD solutions were prepared separately, but always added at the same time to the cultures. us, the addition of EN and CYD is referred to as the addition of EN/CYD, although they were never mixed before addition to the culture since it caused precipitate formation. e volume of feeding was based on addition of certain volumes of feeding solutions, calculated as a percentage of the initial culture volume. is platform process protocol started with the daily addition of 4% Feed A, 2% EN 1X and 2% CYD 1X (percentage of the initial volume) (Jardon et al. 2012). e volume of feeding was increased by 3-folds when the cells reached the 5×106 cells/mL concentration. e time in which the feeding volume was tripled was not rigidly defined in the CHO-tPA protocol, and it was subject to change. e standard initial volume of cultures was 25 mL in 125 mL flasks. As mentioned before, the CHO-MAb fed-batch protocol was developed taking the CHOtPA protocol as the baseline. However, some changes were made from the beginning. CHOMAb cells usually don’t form aggregates, and they don’t need any growth factor for survival. us, neither anti-clumping agent nor IGF were added to the CD CHO medium used for CHO-MAb cells. Also, the amount of sampling was standardized to have consistent dilution rates. e standard sampling routine was daily removal of 6% of the initial culture volume, i.e.  17  1.5 mL for a culture started with 25 mL volume. is slightly modified version the CHO-tPA was taken as the ‘platform process’. In the case of using 5X concentrated version of EN/CYD was used, the volume added daily was 0.4% EN 5X and 0.4% CYD 5X. e volume of feeding was tripled in the same manner of platform process, when cell density passed 5×106 cells/mL. Sampling was done similar to that of the batch cultures. e volume changes throughout the cultures run with the platform process. e oxygen transport might become limited with the increase in volume. is issue is addressed in Appendix A, and it was shown that in the range of volumes we used, there is no indication of oxygen transport limitation. 3-methyladenine (3-MA) (Sigma) was used to chemically inhibit autophagy, and was made into a 500 mM stock dissolved in 0.5 M HCl. pH adjustments were done after 3-MA addition with a solution of 7.5% w/v NaHCO3 (Invitrogen).  18  Table 2.1: Measured composition of CD CHO and Feed A media  Component  CD CHO (mM)  Feed A (mM)  Feed A/ CD CHO Ratio  Alanine  –  –  –  Asparagine  5.82  21.84  3.75  Aspartic Acid  1.2  5.51  4.59  Cysteine  0.4  0.95  2.38  Glucose  32  160  5.00  Glutamic Acid  1.61  7.17  4.45  Glutamine  –  –  –  Glycine  –  –  –  Histidine  1.01  4.59  4.54  Hydroxyproline  1.21  5.58  4.61  Isoleucine  2.34  10.63  4.54  Leucine  3.58  16.04  4.48  Lysine  2.6  10.94  4.21  Methionine  0.75  3.35  4.47  Phenylalanine  1.15  4.86  4.23  Proline  4.56  20.69  4.54  Serine  2.95  15.63  5.30  reonine + Arginine  2.34  10.83  4.63  Tyrosine  0.88  0.1  0.11  Valine  2.66  11.94  4.49  19  2.2.3  Cell Banking  Original cell vials received from Pfizer, with the volume of 1 mL at 10×106 cells/mL, were thawed, grown and expanded in the maintenance medium described above. e cryopreservation medium consisted of equal amounts of the culture conditioned maintenance medium and a 2x medium consisting of 85% CD CHO with 6 mM glutamine and 15% DMSO (Sigma). Cells were centrifuged and re-suspended in the cryopreservation medium at the concentration of 10×106 cells/mL. en, cells were quickly loaded in 1 mL cryovials, and put in special freezing containers filled with 100% isopropanol. Cryovials were put at -80ºC for at least 4 hours and then transferred into suitable boxes for cryostorage at -150ºC.  2.3  Flow Cytometry Cells were stained with several cell permeable dyes (Molecular Probes, Eugene, OR) in order  to examine and quantify a number of physiological parameters by means of flow cytometry: LysoTracker Green DND-26 for staining lysosomes, DiIC1(5) to stain mitochondria that preserved a high membrane potential, and propidium iodide to stain dead cells. To stain the cells for flow cytometry, 1.0×105 cells were spun and re-suspended in a solution of 10 mM glucose, 0.09% trypsin in phosphate buffered saline (PBS), incubated with 50 nM DiIC1(5) and 75 nM LTG for 15 min at 37ºC. Afterwards, the stained cells were spun, and cell pellets were resuspended in a solution of 10 mM glucose in 0.5 µg/mL propidium iodide in PBS, and placed on ice until analysis. A FACScalibur machine (BD Biosciences, Franklin Lakes, NJ) was used for flow cytometry, using 488 and 633 nm lasers, and data analysis was performed using FlowJo (Tree Star, Ashland, OR). 20  2.4  IgG ELISA Monoclonal antibody concentration was measured using a sandwich IgG ELISA assay. 96  well MaxiSorp plates (Nunc, Penfield, NY) were coated with goat anti-human IgG FCγ antibody fragment (Jackson Immuno Research, West Grove, PA) dissolved in 0.1 M sodium carbonate buffer at pH 9.5 with 100 µL/well, and were incubated at either for an hour at 37ºC or 4ºC overnight. Afterwards, plates were washed with a wash buffer consisting of PBS with 0.015% Tween 20 (Sigma) three times, using Nunc-Immuno™ Wash 12 (Nunc). Wells were blocked using 200 µL/well of a blocking buffer consisted of PBS, 1% bovine serum albumin (Sigma) and incubated for 1 to 1.5 hours at room temperature. awed culture supernatant were diluted in a dilution buffer consisting of PBS, 0.5% bovine serum albumin and 0.01% Tween 20. To generate a standard curve, IgG standards were also prepared from human IgG (Sigma) using serial dilutions in the dilution buffer, in the range of 1000-3.9 ng/mL. After washing out the blocking buffer three times using the wash buffer, standard and diluted samples were loaded in the plated with 100 µL/well, and incubated either for 2 hours at room temperature or for an hour at 37ºC. After washing three times using wash buffer, goat anti-human IgG (H+L)-alkaline phosphatase conjugated antibody was diluted 4000-fold in the dilution buffer and was added to each well at 100 µL/well. After an hour of incubation at room temperature, plates were washed three times. Afterwards, 1 mg/mL para-nitrophenylphosphate (Sigma) solution was prepared using a substrate buffer consisting of 0.1 M diethanolamine-HCl (Sigma), 5 mM MgCl2 (Sigma) in dH2O at pH 9.8. e para-nitrophenylphosphate solution was loaded 100 µL/well and incubated in dark for 5-7 minutes for the colour formation reaction to proceed. e alkaline phosphatase enzyme catalyzes the reaction of para-nitrophenylphosphate with water that results 21  in formation of yellow-coloured para-nitrophenol. e reaction was stopped with addition of 50 µL/well of 3M NaOH (Sigma). en the absorbance of each well was read, within 30 minutes of stopping reaction, using a plate reader (Vmax Kinetic microplate reader, Molecular Devices, Sunnyvale, CA or Tecan M200, Männedorf, Switzerland) at 405 nm, using a reference of 492 nm.  2.5  Glycan Analysis Glycan analysis was performed at the University of Manitoba, using in-gel method (Royle et  al. 2006). By running the IgG1 on a reduced gel the glycan structures of the heavy and light chain could be analyzed separately. PNGase F was used to cleave the glycans from the protein before they were fluorescently labeled with 2-aminobenzamide (Sigma). e glycan samples were then cleaned up using a GlycoClean S cartridge (Prozyme, Hayward, CA) and analyzed by Normal Phase-HPLC (Guile et al., 1996) using a Waters HPLC system and Breeze. Using a dextran ladder standard the glucose units (GU) values were determined for all peaks in the profiles making up more than 2% of the total integration area. e peaks of each profile were than assigned to structures based on their GU value and additional information from an exoglycosidase digest using Glycobase (Dublin-Oxford Glycobiology site, NIBRT) as a guide.  2.6  Cell Counting Cell concentrations were counted by trypan blue (Sigma) dye exclusion method using a  Cedex automatic cell counter system (Innovatis, Bielefeld, Germany). In order to remove the possible cell aggregates, samples were diluted by 0.25% trypsin-EDTA. 22  2.7  Chemical Analysis Glucose, lactate, glutamate and glutamine content of the samples were measured with a 7100  MBS Bioanalyzer (YSI Life Sciences, Yellow Springs, OH), using enzymatic electrochemical reactions.  2.8  Osmometry Osmolality of the samples was measured using a freezing point depression osmometer.  Samples were diluted 1:2 with dH2O, if needed.  2.9  HPLC Amino Acid Analysis e method used here measures the free amino acid content of the cell culture supernatants.  Protein content of the supernatants was removed the cells with Nanosep centrifugal ultra-filters (Pall Life Sciences, Ann Arbor, MI) with 3 kDa molecular weight cut-off, spinning them for 20 minutes at 10000g. e prepared samples were tagged with AccQFluor reagent (Waters, Milford, MA) and prepared as described in the AccQTag amino acid analysis method (Waters). e samples were run in Waters’ HPLC system, using 3.9×150mm 4µm silica base bonded with C18 HPLC columns (Waters). e peaks were resolved and the area under each was compared to the area under the standard, and the concentrations are calculated. e standard was a mixture of Amino Acid Standard H (Pierce Biotechnology, Rockford, IL) supplemented with asparagine, glutamine, hydroxyproline and tryptophan.  23  2.10 Calculations Integral of viable cells (IVC) was calculated by the following equation, in which X is the viable cell concentration, V is the volume of the culture, and t is the time: IVC=  t XVdt 0  (Equation 2.1)  Total MAb was calculated using the following equation, in which P is the product concentration and V is the culture volume: IVC=PV  (Equation 2.2)  Cell specific productivity was calculated by the following equation, in which P is the product concentration and IVC is the integral of viable cells. qp =  1 d PV XV dt  = Δ  (PV) Δ IVC  (Equation 2.3)  One-tailed paired t-test was done using Microsoft Excel ‘TTEST’ function. All other statistical tests, ANOVA calculations and designs of experiments are done using JMP software (SAS Institute, Cary, NC). Indices related to glycosylation analysis were calculated using the following formulae, where G, S, F and A refer to galactose, sialic acid, fucose and N-acetylglucosamine (GlcNAc) with the corresponding number of molecules attached to the glycan. Galactosylation Index  = Sialylation Index  =  G2+G1+G0  S0 × 0 + S1 × 1 +(S2 × 2)  Fucosylation Index = Antennarity Index=    (G2+0.5  ×  G1)  S0+S1+S2 F0 × 0 + F1 × 1 F0+F1  A0 × 0 + A1 × 1 + A2 × 2 + A3 × 3 +(A4 × 4) A0+A1+A2+A3+A4  (Equation 2.4) (Equation 2.5) (Equation 2.6) (Equation 2.7)    24  Chapter 3: Establishing Fed-batch Cell Culture Protocol for MAb Producing CHO Cells  Different cell lines have varying responses to identical culture milieux due to different nutrient requirements and environmental sensitivities. In particular, different expression systems and recombinant proteins can significantly affect the metabolism of cell lines. us, a cell culture protocol developed for one CHO cell lines may not be optimized or even compatible for another CHO cell line. A fed-batch culture protocol was originally developed in our lab for a CHO cell line expressing tPA (Jardon et al. 2012). is protocol was used as a platform process in a way analogous to industrial practice where previously successful process methods are used with new cell lines (Heath 2010). is allowed the more rapid development and optimization of a new fed-batch protocol for the CHO-MAb cell lines. To systematically adapt the platform protocols, several aspects were analyzed by design of experiments (DOE) studies.  3.1  Fed-batch Culture Proof of Concept Study A CHOm44 fed-batch culture using the platform protocol was first compared to a batch  culture, in order to assess the applicability of the protocol. e fed-batch culture was fed with daily 4% of Feed A (of the initial volume), and 2% of EN and CYD 1X solutions, with tripling of the feed volumes on Day 7. As seen in Figure 3.1, the fed-batch cultivation of CHO-MAb cells significantly extended the culture run time and increased the maximum viable cell  25  concentration by approximately 2-fold. e cells were viable for 10 days longer in the fed-batch culture (Figure 3.2), delaying the sharp drop in viability until day 15. 16  Viable Cell Conc. (106 c/mL)  14  Batch  12  Fed-Batch  10 8 6 4 2 0  0  2  4  6  8  10 Time (day)  12  14  16  18  20  18  20  Figure 3.1: Viable cell concentration of batch and fed-batch cultures of CHOm44 cell line  100 90 Viability (%)  80 70 60 50  Batch  40  Fed-Batch  30 20 10 0  0  2  4  6  8  10  12  14  16  Time (day) Figure 3.2: Viability of batch and fed-batch cultures of CHOm44 cell line  26  450 400  Batch  MAb Conc. (mg/L)  350  Fed-Batch  300 250 200 150 100 50 0  0  2  4  6  8  10  12  14  16  18  20  Time (day)  Figure 3.3: Monoclonal antibody concentration of batch and fed-batch cultures of CHOm44 cell line  90 80  35  Batch MAb  30  Fed-Batch MAb  70  25  Fed-Batch Volume  60 50  20  40  15  30  10  20  5 0  Volume (mL)  Total MAb (mg)  40  10 0  2  4  6  8  10  12  14  16  18  20  0  Time (day)  Figure 3.4: Total amount of antibody produced in batch and fed-batch cultures, as well as the fed-batch volume change. The batch culture volume was assumed to be constant at 25 mL (Sampling and evaporation ignored).  27  Both cultures were continued until they reached a viability of less than 10%, to assess the maximum length of the cultures. In practice however, cell cultures are harvested well before reaching this point, to prevent the degradation of the product by the enzymes released after the lysis of dead cells, and to reduce the release of host cell proteins that are a burden on downstream processing. e fed-batch culture produced considerably higher MAb concentration and total MAb (Figure 3.3 and Figure 3.4). Since the total MAb in the fed-batch culture remained essentially constant in the last days of the culture, it could be that the decrease in MAb concentration was due to the dilution. However, it may be that the decrease in the MAb concentration in the batch culture was due to the MAb degradation in the presence of cell lysate. Perhaps in the fed-batch culture more sustained protein production compensated for any degradation. Overall, the CHO-tPA platform fed-batch protocol did considerably increase production by the CHO-MAb cell line. e supplementary data for this experiment can be found in Appendix B.  3.2  Optimization of CHO-tPA Fed-Batch Protocol to CHO-MAb Cells Using DOE By performing a series of screening experimental designs, the platform protocol for the  CHO-MAb cells was investigated in an effort to increase production. e CHO-tPA fed-batch protocol involves multiple process variables or factors that are dependent or independent, controlled or uncontrolled. e controllable independent process variables (Table 3.1) were modulated in several fractional factorial DOEs (designs of experiments). e number of factors was limited by our capacity for multiple parallel experiments and our ability to control some of the factors. erefore a subset of factors was chosen from the list of independent process 28  variables (Table 3.1), and tested in a trial-and-error approach to determine their significance and need for further consideration. Table 3.1: List of proposed independent fed-batch process variables. Some items such as medium composition or feeding regime can be divided into more variables.  Independent Fed-batch Process Variables Initial Cell Concentration Medium Composition Feed Composition Feeding Regime Temperature O2 Pressure and Dissolved Oxygen CO2 Pressure pH  Setting the values and ranges to test for each factor is a crucial and often challenging task, and even a few improper choices can render an entire set of experiments useless. ese ranges are set based on previous experience, but often with a considerable degree of risk. e fed-batch platform protocol was taken as the baseline for all subsequent experiments. e selected factors were varied slightly from the original levels in the platform protocol. Using statistical tests we could determine whether these changes had any meaningful effect on the final MAb titres and the specific productivities of the CHO-MAb cells.  29  3.2.1  First Screening Experiment  In the first round of screening experiments three factors expected to be influential were chosen. First, the feeding regime was tested to determine if a simpler approach would be effective, i.e. the stepwise tripling of the feed volume above 5×106 cells/mL (1F3F) versus using the flat tripled amount from the first day (3F). Second, the initial cell concentration (X0) was varied from 0.3×106 to 1.0×106 cells/mL. ird, the initial glutamine (Gln0) varied from 0–4 mM and was selected since glutamine influenced the metabolism and protein production in the CHO-tPA cells (Jardon et al. 2012). Since the first factor is a categorical factor that cannot be changed continuously, the common fractional factorial designs with centre points weren’t used. Instead, a 2×2×3 factorial design was used, as shown in Table 3.2. e CHO-MAb clone used in this study was CHOm56 and culture I corresponds to the platform fed-batch protocol. e supplementary data for this section can be found in Appendix C. In Figure 3.5 and Figure 3.7 the viable cell concentrations of the experimental runs are shown separately for the sake of clarity. e cultures with no glutamine (A, D, G and J) had slower growth than the rest with lower peak viable cell concentrations; signs of adaptation to the conditions followed the long lag time. However those cultures started with high initial cell concentration (D and J) seemed to have a shorter lag time than what was observed in the cultures with low initial cell concentration (A and G).  30  Table 3.2: Design of experiment table for the first characterization experiment.  Culture Feeding Initial Cell Conc. (106 cell/mL) Initial Glutamine Conc. (mM) A 3F 0.3 0 B 3F 0.3 2 C 3F 0.3 4 D 3F 1 0 E 3F 1 2 F 3F 1 4 G 1F3F 0.3 0 H 1F3F 0.3 2 I 1F3F 0.3 4 J 1F3F 1 0 K 1F3F 1 2 L 1F3F 1 4 Glutamine was not depleted in any of the cultures (Figure 3.9 and Figure 3.10), leveling out at around 0.5 mM, even in the cultures that had no initial glutamine. is suggested that these cells actually produced glutamine (up to ~0.5-0.75 mM), and that was may be due to glutamine synthetase (GS) activity in this Lonza CHOK1SV cell line. GS is an enzyme that converts a glutamate and an ammonium ion to glutamine. Also it might indicate the release of intracellular glutamine upon cell death and lysis (Piret et al. 1991). e HPLC results in the end of this chapter confirmed these results, to show that the nonzero glutamine results were not the result of assay artifacts (See 3.2.4). e integral of viable cell concentrations over time (IVC) was used to represent the overall cell concentration in the culture. As there was a difference in the volumes of the cultures due to the different feeding strategies (addition of higher volumes in the 3F than the 1F3F feeding), the total weight of MAb produced was used instead of the MAb concentration. Also, the average cell specific productivity (qp), was calculated as an important characteristic of the cells. 31  Viable Cell Conc. (106 cell/mL)  14 12 10  A  B  C  D  E  F  8 6 4 2 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  11  12  13  14  Time (day) Figure 3.5: Viable cell concentrations of the cultures with the 3F feeding strategy  45 40  Total MAb (mg)  35 30  A  B  C  D  E  F  25 20 15 10 5 0  0  1  2  3  4  5  6  7  8  9  10  Time (day) Figure 3.6: Total amount of MAb produced in the cultures with the 3F feeding strategy  32  14  Viable Cell Conc. (106 cell/mL)  12 10  G  H  I  J  K  L  8 6 4 2 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  11  12  13  14  Time (day) Figure 3.7: Viable cell concentrations of the cultures with the 1F3F feeding strategy  45 40 35 Total MAb (mg)  30  G  H  I  J  K  L  25 20 15 10 5 0  0  1  2  3  4  5  6  7  8  9  10  Time (day) Figure 3.8: Total amount of MAb produced in the cultures with the 1F3F feeding strategy  33  Glutamine Conc. (mM)  4.5 4  A  B  3.5  C  D  3  E  F  2.5 2 1.5 1 0.5 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time (day) Figure 3.9: Glutamine concentration of the cultures with the with 3F feeding strategy; the initial zero level is based on the medium formulation  4.5  Glutamine Conc. (mM)  4 3.5 3  G  H  I  J  K  L  2.5 2 1.5 1 0.5 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time (day) Figure 3.10: Glutamine concentration in the cultures with the with 1F3F feeding strategy; the initial zero level is based on the medium formulation  34  e analysis of variance showed that although X0 and Gln0 were found to be significant factors for the IVC response (Table 3.3), there was no significant factor found for the MAb response (Table 3.4) and the qp response (Table 3.5). However, Gln0 would remain not significant if we excluded the glutamine-free results (A, D, G and J). Overall, only two of the factors were found significant for the IVC response, but none for the MAb and qp responses. Since total amount of MAb is the more important protein outcome, and there may be some other factors that may influence it, we chose to consider another set of factors in a second round of screening experiments. For the next experiments, as we needed to limit the factors of interest for further testing, we kept Gln0 at the platform value (4 mM). Also, the initial cell concentration wasn’t defined rigidly in the platform process. For many experiments, the initial cell concentration of 0.3×106 cells/mL was used. However, if we needed to shorten experimental run time, we could also use 1.0×106 cells/mL as the initial cell concentration. Table 3.3: IVC response analysis of variance table  Factors Feeding X0 Gln0 Feeding×X0 Feeding×Gln0 X0×Gln0 Gln0×Gln0  F Ratio 1.5220 101.2912 259.3281 0.9486 6.4086 9.0521 77.0376  Prob > F 0.2849 0.0005* <0.0001* 0.3852 0.0646 0.0396* 0.0009*  35  Table 3.4: Total MAb response analysis of variance table  Factors Feeding X0 Gln0 Feeding×X0 Feeding×Gln0 X0×Gln0 Gln0×Gln0  F Ratio 0.0275 2.1347 3.3460 2.4350 0.1609 0.2990 1.0737  Prob > F 0.8763 0.2178 0.1414 0.1937 0.7088 0.6136 0.3587  Table 3.5: Cell specific productivity response analysis of variance table  Factors Feeding X0 Gln0 Feeding×X0 Feeding×Gln0 X0×Gln0 Gln0×Gln0 3.2.2  F Ratio 3.4201 0.0517 0.2465 2.8876 0.0599 0.2195 3.4201  Prob > F 0.3156 0.8576 0.7066 0.3386 0.8472 0.7211 0.3156  Second Screening Experiment  In the second round of screening experiments, another set of factors was chosen for adapting and optimizing the platform process protocol to the CHO-MAb cell line. As in the first round of screening experiments, the clone used was CHOm56. e addition of EN/CYD amino acid solutions was tested as a factor. EN/CYD supplementation of Feed A was used to maintain the amino acids depleted in the CHO-tPA cell cultures. Two of the defining factors of the platform feeding protocol, the magnitude (Feed Increase Magnitude) and the timing (Feed Increase Time) of the increase in feed volume were selected as possible significant factors. Furthermore, there was evidence that CHO-tPA cell feeding of a mixture of glutamine and α-ketoglutarate (α-KG) increased the production of tPA (Jardon 2012). ese factors were tested for improving the CHO-MAb cell culture protocol in a 36  5-2  Resolution-III 2III fractional factorial design with one centre point. e centre point (culture I) had each factor set in the middle of range tested. e design was based on the platform process, although none of the cultures directly represent the platform. Details of the design can be seen in Table 3.6. e supplementary data for this section can be found in Appendix D. Figure 3.11 and Figure 3.12 show the trends of viable cell concentrations, with the centre point (I) repeated in both figures to ease comparisons. In both figures, the centre point culture had a higher viable cell concentration almost at all times. In terms of MAb concentration, B had a higher final product concentration than the centre point (Figure 3.13). However due to the very different final volumes of the cultures, the total MAb produced provides a more valid comparison. In Figure 3.15 and Figure 3.16 the protein production profiles can be seen in terms of total weight of the protein. Nevertheless the centre point had a higher production than the rest of the cultures except culture B that yielded a similar result. is indicated a curvature in both the MAb and the cell growth responses of the cultures; such that the optimum point may be found within this experimental design space. Table 3.6: DOE table for the second round of screening  Culture  Feed Increase Time  Feed Increase Magnitude  EN/CYD  A B C D E F G H I  4 4 4 4 6 6 6 6 5  2 2 4 4 2 2 4 4 3  0 1 0 1 0 1 0 1 0.5  α-KG Feed (mM) 0 2 2 0 2 0 0 2 1  Glutamine Feed (mM) 0.5 0 0 0.5 0.5 0 0 0.5 0.25 37  14  Viable Cell Conc. (106 cell/mL)  12 10  A  B  D  I  C  8 6 4 2 0  0  1  2  3  4  5  6  7 8 Time (day)  9  10  11  12  13  14  15  14  15  Figure 3.11: Viable cell concentration of the cultures A, B, C, D and the centre point (culture I)  14 Viable Cell Conc. (106 cell/mL)  12 10  E  F  H  I  G  8 6 4 2 0  0  1  2  3  4  5  6  7  8 9 Time (day)  10  11  12  13  Figure 3.12: Viable cell concentration of the cultures E, F, G, H and the centre point culture (culture I)  38  1600  MAb Conc. (mg/L)  1400 1200  A  B  D  I  C  1000 800 600 400 200 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  14  15  Figure 3.13: MAb concentration of the cultures A, B, C, D and the centre point culture (culture I)  1200  MAb Conc. (mg/L)  1000  E  F  H  I  G  800 600 400 200 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  14  15  Figure 3.14: MAb concentration of the cultures E, F, G, H and the centre point culture (culture I)  39  60  Total MAb Weight (mg/L)  50  A  B  D  I  C  40 30 20 10 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  14  15  14  15  Figure 3.15: MAb weight of the cultures A, B, C, D plus the centre point culture (culture I)  60  Total MAb weight (mg)  50  E  F  H  I  G  40 30 20 10 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  Figure 3.16: MAb weight of the cultures E, F, G, H plus the centre point culture (culture I)  40  For the response final total MAb produced (Table 3.7) and final IVC (Table 3.8), none of the factors were significant as none of them has a P-value (Prob > F) less than 0.05, the conventional α cut-off value (Maldonado and West 2011). As can be seen in Figure 3.12, Figure 3.14, and Figure 3.16 the culture H never came out of the lag phase and had a very low cell growth and production. is result seemed abnormal, and may have resulted by a feeding error during the initial days of the experiment. Removing the culture H from analysis made the fractional factorial design incapable of capturing the significant effects, since the minimum number of experiments needed for a Resolution-III design, neglecting the centre points, is 8 (Montgomery 2009). Perhaps augmenting the current design with more centre points and redoing the culture H could result in more conclusive results. Table 3.7: DOE analysis of the second screening experiment based on total MAb produced with the culture H removed.  Factors Feed Increase Time Feed Increase Magnitude EN/CYD α-KG Glutamine Feed  F Ratio 0.0936 0.3369 1.0421 0.1316 0.1513  Prob > F 0.7886 0.6203 0.4147 0.7515 0.7348  Table 3.8: DOE analysis of the second screening experiment based on final IVC with the culture H removed.  Factors Feed Increase Time Feed Increase Magnitude EN/CYD α-KG Glutamine Feed  F Ratio 0.1836 0.0634 0.1050 0.0044 0.5204  Prob > F 0.7100 0.8248 0.7767 0.9531 0.5456  41  3.2.3  Augmenting the Second Screening Experiment  e second screening experiment was augmented with three more centre points, plus the previously failed culture H (Table 3.9). e supplementary data for this section can be found in Appendix D. Table 3.9: The second round DOE table with augmented points shaded in blue and the culture H removed  Culture  Feed Increase Time  Feed Increase Magnitude  EN/CYD  A B C D E F G –H– I I2 I3 I4 H1  4 4 4 4 6 6 6 –6– 5 5 5 5 6  2 2 4 4 2 2 4 –4– 3 3 3 3 4  0 1 0 1 0 1 0 –1– 0.5 0.5 0.5 0.5 1  α-KG Feed (mM) 0 2 2 0 2 0 0 –2– 1 1 1 1 2  Glutamine Feed (mM) 0.5 0 0 0.5 0.5 0 0 –0.5– 0.25 0.25 0.25 0.25 0.5  Viable Cell Conc. (106 cell/mL)  10 8  I(2)  I(3)  I(4)  H(1)  6 4 2 0  0  1  2  3  4  5  6  7  8 9 Time (day)  10  11  12  13  14  15  Figure 3.17: Viable cell concentration for the augmented points in the second screening the design  42  1200  MAb Conc. (mg/L)  1000  I(2)  I(3)  I(4)  H(1)  800 600 400 200 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  14  15  14  15  Figure 3.18: MAb concentration for the augmented points in the second round of the design  50 45  Total MAb (mg)  40 35  I(2)  I(3)  I(4)  H(1)  30 25 20 15 10 5 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  Figure 3.19: Total MAb produced for the augmented points in the second round of the design  43  As it can be seen in Figure 3.17, Figure 3.18 and Figure 3.19 the three added centre point cultures had very similar profiles of growth and production. e repeated culture H1, unlike the previous unsuccessful culture, had no apparent lag and produced a high amount of MAb. Deviation from the line of normal error distribution is one of the ways to show the significance of factors. For the IVC response, none of the factors were significant, and they didn’t deviate enough from the line in the half normal plot in Figure 3.21. In contrast, the half normal plot for the total MAb produced (Figure 3.20) shows that α-KG and EN/CYD were significant factors, as they deviate from the line of normal error distribution. Furthermore, there were two second-order terms found to be significant. In this Resolution-III design, each second-order term is aliased with other second order terms. us the significance of a second-order term, doesn’t add much to our knowledge.  Figure 3.20: Half normal plot for the significance of factors based on the MAb response for the augmented design of experiments  44  Figure 3.21: Half normal plot for the significance of factors based on the IVC response for the augmented design of experiments  e coefficients of the polynomial model based on the significant effects show (Table 3.10) the positive relationship between the total MAb and addition of both EN/CYD and α-KG. Table 3.10: Parameter estimation for the total MAb polynomial model based on the significant parameters found in the half normal plot  Term Estimate Std Error t-Ratio P-value Intercept 45.15 1.92 23.44 <.0001* EN/CYD 10.99 1.36 8.07 <.0001* α-KG 5.35 1.36 3.93 0.0057* EN/CYD×EN/CYD -20.15 2.36 -8.54 <.0001* EN/CYD×α-KG 7.14 1.36 5.24 0.0012* e aim of the optimization study was to identify modifications to the platform process protocol to further increase the productivity of the CHO-MAb cell lines. e results showed that the platform process was sufficiently optimized for the new CHO-MAb cells lines and the 45  modifications tested did not have significant effects on MAb levels. e positive estimate of the EN/CYD and α-KG factors also showed that the inclusion of these elements in the feeding protocol increased the production of protein. Although α-KG was found to increase the total MAb, finally we decided to not change the protocol. is was mainly because introduction of αKG would have introduced more complexity while confusing the comparison of two important productivity increase methods, as will be discussed in Chapter 4.  3.2.4  Assessment of Possible Amino Acid Limitations  e growth and production of a cell line can be affected by nutrient limitation. Some nutrients are used for providing energy, and some are used for producing biomolecules. Amino acids are precursors of proteins, and limitation in some amino acids, besides energetic restrictions, can decrease protein translation. is could have been a possible reason why we didn’t find many significant effects in the pre-augmented experimental design, despite our anticipation that some of the factors such as EN/CYD were found to be significant. To confirm that amino acid limitations were not an issue, we tested the amino acid content of selected samples of the pre-augmentation DOE cultures with HPLC. Since the capacity for HPLC analysis was limited by cost, only a subset of cultures were chosen. Cultures A, B, D and I (Table 3.9) were tested with HPLC, as described in Chapter 2 (Section 2.9). For the sake of clarity only a subset of the results, for the more important amino acids will be shown here.  46  7  Run A  Amino Acid Conc. (mM)  6 5  Asp (D) Glu (E)  4  Ser (S) Asn (N)  3  Gln (Q) 2  Tyr (Y) Cys (C)  1 0  5  7  9  11  Day Figure 3.22: Selected amino acid results for the culture A, with feed increased on day 4 with 2x magnitude, no EN/CYD solution, no α-KG, with glutamine fed every day.  18  Run	
  B	
    Amino Acid Conc. (mM)  16 14  Asp (D)  12  Glu (E)  10  Ser (S)  8  Asn (N)  6  Gln (Q) Tyr (Y)  4  Cys (C)  2 0  4  5  6  7  8 Day  9  10  11  12  Figure 3.23: Selected amino acid results for the culture B, with feed increased on day 4 with 2x magnitude, with both EN/CYD and α-KG and no glutamine feeding.  47  20 18 Amino Acid Conc. (mM)  16 14  Asp (D)  12  Glu (E)  10  Run D  Ser (S) Asn (N)  8  Gln (Q)  6  Tyr (Y)  4  Cys (C)  2 0  6  9 Day  12  Figure 3.24: Selected amino acid results for the culture D, with feed increased on day 4 with 4x magnitude, with EN/CYD and glutamine feeding and no α-KG.  12  Run I  Amino Acid Conc. (mM)  10  Asp (D)  8  Glu (E) Ser (S)  6  Asn (N) Gln (Q)  4  Tyr (Y) Cys (C)  2 0  5  7  Day  9  11  Figure 3.25: Selected amino acid results for the culture I (centre point), with feed increased on day 5 with 3x magnitude, with EN/CYD, α-KG and glutamine fed half the amount defined for other cultures.  48  Cysteine was shown to be the only amino acid that was completely consumed in the culture A (Figure 3.22), the culture that is not fed with EN/CYD solution. Since cysteine is poorly soluble in water, it cannot be provided at high concentrations in cell culture media. e limitation in cysteine level despite addition of Feed A solution that contains cysteine, showed the necessity of the addition of extra cysteine through EN/CYD solution. e glutamate concentration in the culture A increased constantly, despite no EN/CYD addition. is shows that the glutamate in Feed A solution was sufficient for the cells, thereby it could have been removed from the EN/CYD feed. However, since other experiments related to other parts of the thesis were being carried out using EN/CYD, it was kept intact to keep the consistency between different experiments. e results in Figure 3.23 showed that even without glutamine feeding and despite continuous dilution in the culture B, glutamine level was kept constant. is confirmed the previous results, showing that CHO-MAb cell lines were not dependent on glutamine in their production phase. is lack of dependence doesn’t necessarily mean glutamine feeding will not result in any benefit in the MAb production; but the MAb results showed that glutamine feeding did not significantly increase the MAb concentration unlike in the case of CHO-tPA cells (Jardon et al. 2012). As expected by the using of higher volumes of Feed A and EN/CYD solutions, Figure 3.24 shows relatively higher concentrations of amino acids, although not twice as high. Considering the lower total cell number for the culture D and lower MAb weight produced (Figure 3.15), this showed that higher availability of amino acids can lead to higher cell specific consumption, that has not resulted in neither higher growth nor protein production. 49  e amino acid concentrations of the centre point culture (culture I) in Figure 3.25 showed no depletion of any of amino acids. Also there was no excessive build up of any amino acid except for glutamate for similar reasons as described above. e main incentive for testing these samples was the lack of significance found for factors such as EN/CYD in the DOE before augmentation. e amino acid analysis was done preaugmentation, to check whether any amino acid limitation developed. Overall, there was no unexpected amino acid depletion that can cause any of the cultures to act differently. e nonsignificance issue was resolved with augmenting the design with three more centre points and repeating one of the cultures.  3.3  Conclusions to Adapting CHO-tPA Fed-batch Protocol to MAb Cells An initial goal of this study was to develop an optimized fed-batch protocol for a MAb  expressing CHO cell line. To reach this goal, a fed-batch cell culture protocol established for the CHO-tPA cell line was used and applied to the CHO-MAb cell line. In order to optimize the platform protocol for the CHO-MAb cell line, several factors that described the CHO-tPA feeding protocol were examined in a series of DOE screening experiments. e changes in many factors were not statistically significant, resulted in insignificant improvement to overall MAb production and even in some cases detrimental to viable cell and MAb concentration. A newly introduced α-KG addition was found to be beneficial. But since it needed several more optimization steps and the magnitude of effect was not very high, it was not included in the CHO-MAb protocol. e amino acid content of some of the cultures was tested to check for possible limitations in the course of culture. It was shown that using the platform protocol 50  provides enough amino acids throughout the durations of the fed-batch cultures, and no serious limitations were observed. Overall, the CHO-tPA platform fed-batch protocol was shown to provide satisfactory cell growth and protein production in CHO-MAb cells. Changes to the platform protocol resulted in either non-significant improvements or negative impact on cell’s growth and production. As a result, the platform protocol was adopted in its entirety for use with CHO-MAb cells.  51  Chapter 4: Autophagy Inhibition and Hypertonicity Effects on CHO-MAb Fed-batch Cultures  4.1  Autophagy Inhibition Effect on the Protein Production in CHO-MAb Cultures Inhibition of autophagy was shown to be beneficial in the production of tPA in a CHO-tPA  system (Jardon et al. 2012). e applicability of this finding to the CHO-MAb system was tested initially using the CHOm44 cell line, in a fed-batch culture with EN/CYD 1X feed. 5 mM of 3MA was used to inhibit autophagy at the time of addition, in accordance with the first reports on the use of 3-MA for inhibiting autophagy (Seglen and Gordon 1982). Time of 3-MA addition was set to one day after the peak of the cell concentration, when the cells were more likely to be under cellular stresses due to high cell density and limited feed. is is especially important as 3MA can itself induce autophagy if introduced in nutrient rich conditions (Wu et al. 2010)  Cell Conc. (FB + 3-MA)  12  1200  3-MA  1000  Cell Conc. (FB) 10  MAb Conc. (FB + 3-MA)  8  800  MAb Conc. (FB) 600  6 400  4  200  2 0  MAb Conc. (mg/L)  Viable Cell Conc. (106 cell/mL)  14  0  2  4  6  8 10 Time (day)  12  14  16  18  0  Figure 4.1: Viable cell and MAb concentrations of the CHOm44 fed-batch cultures treated with 3-MA  52  Consistent with the results from the CHO-tPA cells (Jardon et al. 2012), addition of 3-MA showed an increase of more than two-fold in the product concentration, despite some loss of the viable cell concentration (Figure 4.1). 10 9  Cell Specific Productivity (pg/cell.day)  8 7  FB FB + 3-MA  6 5 4 3 2 1 0  After 3-MA addition  Figure 4.2: After 3-MA addition average of cell specific productivity of the CHOm44 fed-batch culture with and without 3-MA addition  e cell specific productivity of the cultures is also improved significantly, as shown in Figure 4.2. e average cell specific productivity of the culture with 3-MA was increased more than two-fold, and the average cell specific productivity of the cultures in the duration after 3-MA treatment was increased almost five-fold (Figure 4.2). e trends of the cell specific productivity of the cultures over time, as well as the viability of the culture can be found in Appendix E. e results showed that inhibition of autophagy with 3-MA could improve MAb production in a similar fashion to that of previously reported case of tPA production. However, the results 53  were only in the isotonic settings (with EN/CYD 1X feed). By the time of running this experiment, we hadn’t considered the effect of osmolality on the productivity of the cell cultures. Also, we didn’t know if there exist any different response to 3-MA in different clones of the CHO-MAb cell line.  4.2  Autophagy Inhibition by 3-MA in CHOm56 Cells  4.2.1  Dose Response of 3-MA in CHOm56 Fed-batch Cultures  In the previous study, no optimization was made on the 3-MA treatment dose in fed-batch culture and was based on the levels first reported for the use of 3-MA. us, experiments were performed to assess the response of the CHO-MAb cells to different levels of 3-MA. CHOm56, another clone that expressed the same MAb, was used in this study. CHOm56 was reported to have a higher productivity than CHOm44 (Kennard et al. 2009), thus in theory was a more favourable choice for maximizing MAb production. EN/CYD 5X was used in order to limit the volume increase in the cell culture flasks. As using EN/CYD 5X provided the same amount of nutrients in a lower volume, it eliminated the need for changing to bigger flasks, which in turn caused space limitation in the incubator.  54  No 3-MA 7.5 mM 3-MA  Viable Cell Conc. (106 cell/mL)  12  2.5 mM 3-MA 10 mM 3-MA  5 mM 3-MA 15 mM 3-MA  10 8 6 4 2 0  0  1  2  3  4  5  6  7 8 Time (day)  9  10  11  12  13  14  Figure 4.3: Viable cell concentration of the cultures in the 3-MA dose response experiment for the CHOm56 cell line  Although deemed non-toxic by some authors (Hwang and G. M. Lee 2008), the dosedependent toxic effect of 3-MA was evident in Figure 4.3, where higher concentration of 3-MA resulted in higher loss of viable cells.  MAb Conc. (mg/L)  1800  No 3-MA  1600  2.5 mM 3-MA  1400  5 mM 3-MA  1200  7.5 mM 3-MA  1000  10 mM 3-MA 15 mM 3-MA  800 600 400 200 0  0  1  2  3  4  5  6  7 8 Time (day)  9  10  11  12  13  14  Figure 4.4: MAb concentrations for fed-batch cultures of CHOm56 treated with different 3-MA doses  55  Cell Specific Productivity (pg/cell.day)  35 30 25  Control  20 15 10 5 0  0  2.5  5  7.5  10  15  3-MA Concentration (mM)  Figure 4.5: Average cell specific productivity of fed-batch cultures of CHOm56 after 3-MA addition  All cultures except the culture treated with 5 mM of 3-MA, produced less than the control culture (Figure 4.4). Even in the culture with 5 mM 3-MA, the magnitude of increase in production is much lower than the magnitude achieved with CHOm44 cells using EN/CYD 1X feed (11% vs. ~200%). Figure 4.5 shows that the average cell specific productivities also didn’t change considerably with the exception of 5 mM 3-MA addition. e productivity of the culture with 5 mM of 3-MA increased by almost 65%, which was much lower than the almost five-fold increase on the productivity in the case with CHOm44 (Figure 4.2). However, the final product concentrations of the experiment with CHOm56 (Figure 4.4) were much higher (up to threefold) than the final MAb concentrations reached with the CHOm44 cell line (Figure 4.1). Similarly, the productivity of the control culture in the CHOm56 culture (Figure 4.5) was almost ten times higher than the CHOm44 control culture and two times higher than the 3-MA treated culture. 56  In Figure 4.6 we can see that the final weight of MAb produced in the CHOm56 cultures were notably higher than that of the CHOm44 cultures. e total MAb produced was used to knock out the effect of different dilution rates in the comparison. In order to verify that amino acid limitations didn’t cause these results, we checked the amino acid concentration of some samples from this experimental set. e results are presented in the following section 4.2.2. e lysosomal content of the cells during the culture time was measured to assess its correlation with inhibition of autophagy (Figure F.3 and Figure F.4). However, no meaningful correlation was observed. e lysosomal content data alongside other supplementary materials for this section can be found in Appendix F. 90 80  MAb Weight (mg)  70 60 50 40 30 20 10 0  (C56) 0  (C56) 2.5  (C56) 5  (C56) 7.5  (C56) 10  (C56) 15  (C44) 0  (C44) 5  3-MA Concentration (mM)  Figure 4.6: Comparison of the final MAb weight produced in 3-MA treated cultures treated of CHOm56 (C56) and CHOm44 (C44) clones, using EN/CYD 5X and 1X solutions respectively  57  4.2.2  Amino Acid Analysis for 3-MA Treated CHOm56 Fed-batch Cultures  Inhibition of autophagy with 3-MA can potentially change the metabolism of the cell. In order to check the possibility that the response of cultures treated with 3-MA was not due to depletion of any amino acids, amino acid profiles of several samples from cultures without and with 5 mM, 10 mM 3-MA were analyzed by HPLC. e samples were selected form the time of 3-MA treatment onward. 12  No 3-MA  Amino Acid Conc. (mM)  10  Asp (D)  8  Glu (E) 6  Ser (S) Asn (N)  4  Gln (Q) Tyr (Y)  2 0  Cys (C) 6  7  8  9  10  11  12  13  14  15  Day  Figure 4.7: Selected amino acid concentrations of the CHOm56 fed-batch culture with no 3-MA treatment  58  12  5 mM 3-MA  Amino Acid Conc. (mM)  10 Asp (D)  8  Glu (E) 6  Ser (S) Asn (N)  4  Gln (Q) Tyr (Y)  2 0  Cys (C)  6  7  8  9  10  11  12  13  14  15  Day  Figure 4.8: Selected amino acid concentrations of the CHOm56 fed-batch culture treated with 5 mM of 3-MA  12  10 mM 3-MA  Amino Acid Conc. (mM)  10  Asp (D)  8  Glu (E) 6  Ser (S) Asn (N)  4  Gln (Q) Tyr (Y)  2 0  Cys (C) 6  7  8  9  10  11  12  13  14  15  Day  Figure 4.9: Selected amino acid concentrations of the CHOm56 fed-batch culture treated with 10 mM of 3MA  59  Amino acids were not depleted in the cultures with no 3-MA (Figure 4.7), 5 mM of 3-MA (Figure 4.8) and 10 mM of 3-MA (Figure 4.9). Most amino acids exhibited an almost steady state concentration, which shows the efficacy of the selected platform fed-batch protocol for the CHO-MAb cultures. Glutamine levels in all cultures increased slightly due to the glutamine synthetase activity, previously observed in other cultures too. ose results showed that it was unlikely that differences in the performance of different cultures are due to limitations in amino acids. Previously there had been cases of glutamate build up in cultures due to introducing extra glutamate in the EN/CYD solution besides what existed in Feed A (see 3.2.4). However, the amino acid profiles showed that the same feeding strategy used here helped to maintain glutamate level in the last stages of the cultures, thereby undermining the need for removing glutamate from EN/CYD solution. For the results of this set of experiments to be justifiable, we needed to address two main questions: Whether different clones responded remarkably different to the 3-MA treatment, and whether using concentrated EN/CYD solution had a role in obtaining much higher MAb titre.  4.3  3-MA Treatment of CHOm44 Fed-batch Cultures Under Hypertonic Conditions In order to verify the role of different clones in the production response to 3-MA a more  limited 3-MA dose response experiment was done using the CHOm44 cell line. Unlike the first 3-MA treatment experiment with CHOm44, EN/CYD 5X solution was used. As 5 mM of 3MA was shown to be the best concentration for increasing both qp and MAb concentration, it was used as the sole concentration of 3-MA in this experiment. 3-MA addition was carried out at two different points in time (3 and 5 days), to check if timing had an effect on the fact that 360  MA didn’t always cause the final product to increase, as it was seen in the first trial (Figure 4.1). Since the cultures were started by initial cell concentration of 2×106 cell/mL instead of 0.3×106 cells/mL, the duration of cultures was shortened to 13 days instead of the usual 14 days. e cultures treated with 3-MA exhibit the usual decline in the number of viable cells after the treatment (Figure 4.10), and after some time, the timing of 3-MA addition seems to not to have an important effect on the number of viable cells. As shown in Figure 4.11, the MAb concentration of the cultures treated with 3-MA are not very different from the control culture (the culture with no 3-MA). Interestingly, the MAb concentration of the control culture is higher until the last days of the experiment, contrary to the increase that was observed by 3-MA when EN/CYD 1X feed was used (Figure 4.1).  20  No 3-MA  Viable Cell Conc. (106 cell/mL)  18 16  5 mM Day 3  14  5 mM Day 5  12 10 8 6 4 2 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time (day)  Figure 4.10: Viable cell concentrations for CHOm44 cultures treated with 3-MA on different dates  61  900 800  No 3-MA  MAb Conc. (mg/L)  700  5 mM Day 3  600  5 mM Day 5  500 400 300 200 100 0  0  1  2  3  4  5  6  7 8 Time (day)  9  10  11  12  13  14  Figure 4.11: MAb concentration profiles of CHOm44 cultures treated on different dates  Cell Specific Productivity (pg/cell.day)  10  No 3-MA  5 mM 3-MA  9 8 7 6 5 4 3 2 1 0  Day 3 (EN/CYD 5X)  Day 5 (EN/CYD 5X)  EN/CYD 1X  Figure 4.12: Comparison of the average cell specific productivities of the cultures after 3-MA treatment (red bars) to the control cultures throughout the same time.  62  EN/CYD 5X  35  EN/CYD 5X  EN/CYD 5X 30  MAb Weight (mg)  25  EN/CYD 1X  20 15 EN/CYD 1X  10 5 0  C56(No 3-MA)  C56(5mM 3-MA Day 3)  C56(5mM 3-MA Day 5)  C44(No 3-MA)  C44(5mM 3-MA)  Figure 4.13: Comparison of the final MAb weight produced in cultures treated with 3-MA, using EN/CYD 5X and 1X solutions  It is noteworthy that the productivity of the cultures after addition of 3-MA was almost the same as the productivity of the culture using EN/CYD 1X feed (Figure 4.12), but the control cultures using EN/CYD 5X and 1X were remarkably different. e rest of the data related to this experiment can be found in Appendix G. e only difference between the control cultures here and in part 4.1 was the use of EN/CYD 5X versus 1X. As mentioned before, we started to feed the cultures with EN/CYD 5X solution to reduce the volume increase in fed-batch cultures. e change to EN/CYD 5X provided 40% less increase in volume, enabling us to remain in smaller flask, thus occupying less space in the incubator. e use of EN/CYD 5X solution for feeding also provides us with less dilution of the final product. However, as seen in Figure 4.13, the difference in the MAb 63  concentrations was not caused by mere dilution effect, because the total MAb produced in the cultures fed with EN/CYD 5X were higher, knowing that the cultures had the same initial volumes. A possible explanation of the increase of the MAb production in cultures fed with EN/CYD 5X, and the ineffectiveness of 3-MA to increase production in them, was the rise of osmolality in these cultures. Since the EN/CYD solution contained HCl and NaOH, to aid in dissolving the amino acids, addition of a more concentrated version of EN/CYD solution means introducing a non-consumable source of hypertonicity that cannot be removed by cells.  4.4  Increase of Protein Production in High Osmolalities Here the effect of osmolality on MAb productivity was investigated. Both CHOm44 and  CHOm56 clone were cultured under batch condition starting with a viable cell concentration of 1.0×106 cell/mL, to speed up the experiment. As can be seen in Figure 4.14, higher osmolalities resulted in decrease of viable cell concentration in CHOm44 cells. e response of the CHOm56 cells was slightly different, where the cells under 400 mOsm/kg didn’t show a significant difference from the control (325 mOsm/kg) (Figure 4.15). e cell specific productivity of the CHOm44 cultures increased with the elevation of osmolality, almost up to 2-fold at 500 mOsm/kg (Figure 4.16). Similarly to the nonresponsiveness of CHOm56 to hypertonicity at 400 mOsm/kg, the productivity of CHOm56 was also similar to that of isotonic settings. e productivity also increased up to 1.7-fold for CHOm56 at 500 mOsm/kg. Notwithstanding the increase in productivity, for both clones, the MAb concentration remained the same at 400 mOsm/kg, even decreasing at 500 mOsm/kg 64  (Figure 4.17), which was due to much decreased viable cell concentration. e profiles of the cell specific productivities of this experiment, as well as other related data, can be found in Appendix H.  Viable Cell Conc. (106 cell/mL)  12  C44-Osm(325) C44-Osm(400)  10  C44-Osm(500)  8 6 4 2 0  0  1  2  Time (day)  3  4  5  Figure 4.14: Viable cell concentrations of CHOm44 batch cultures under different osmolalities  Viable Cell Conc. (106 cell/mL)  12  C56-Osm(325)  10  C56-Osm(400)  8  C56-Osm(500)  6 4 2 0  0  1  2  Time (day)  3  4  5  Figure 4.15: Viable cell concentrations of CHOm56 batch cultures under different osmolalities  65  Cell Specific productivty (pg/cell.day)  18 16  325 mOsm/kg 400 mOsm/kg 500 mOsm/kg  14 12 10 8 6 4 2 0  CHOm44  CHOm56  Figure 4.16: The average cell specific of CHOm44 and CHOm56 batch cultures under different osmolalities  300  325 mOsm/kg 400 mOsm/kg 500 mOsm/kg  MAb Conc. (mg/L)  250  200  150  100  50  0  CHOm44  CHOm56  Figure 4.17: The final MAb concentration of CHOm44 and CHOm56 batch cultures under different osmolalities  66  If osmolality elevation happens over the course of time (for example in a fed-batch culture), with the cells having time to adapt to that, the results might be different. In general, the results are comparable with the previous reports (discussed in the literature review); hypertonicity increases the productivity, but the overall effect on production might be undermined by diminished cell growth (Han et al. 2010).  4.5  Autophagy Inhibition Under Different Osmolality Conditions e previous section (4.4) showed that hypertonicity could increase the cell specific  productivity of the CHO-MAb cells in batch cultures. But, the overall amount of product actually decreases due to growth inhibition. However, we have already seen high MAb production in the cultures that were fed with EN/CYD 5X, possibly due to higher osmolality. is could be due to the gradual increase in protein production. Also, we have observed that the inhibition of autophagy with 3-MA wasn’t very effective in increasing the MAb concentrations in the cultures fed with EN/CYD 5X, whereas a 2-fold increase was seen in the cultures fed with EN/CYD 1X cultures. However, since the clones that were used in these experiments were different, and there was a possibility that the observed different response to 3-MA was due to clonal variations. To address all these concerns, we performed a set experiments, assessing the effect of autophagy inhibition by 3-MA in combination with different osmotic regimes in fedbatch cultures, using both CHOm44 and CHOm56 clones. Feeding with EN/CYD 1X or 5X, provides two different regimes of gradual osmolality change. ree different levels of 3-MA were also tested to check for possible impacts in interaction with autophagy inhibition. Clonal difference in response to both hypertonicity and autophagy inhibition was also assessed, which is 67  important especially due to their different inherent productivity. e experimental design is shown in the Table 4.1. e experiments were run in duplicate, in order to confirm the results statistically. Table 4.1: Experimental design table for testing autophagy inhibition and hypertonicity effects on MAb production simultaneously  Culture  Cell Line  A B C D E F G H I J K L  CHOm44 CHOm44 CHOm44 CHOm44 CHOm44 CHOm44 CHOm56 CHOm56 CHOm56 CHOm56 CHOm56 CHOm56  Osmolality Regime [EN/CYD Conc.] High [5X] High [5X] High [5X] Low [1X] Low [1X] Low [1X] High [5X] High [5X] High [5X] Low [1X] Low [1X] Low [1X]  3MA Conc. (mM) 0 5 10 0 5 10 0 5 10 0 5 10  Figure 4.18 and Figure 4.19 show the osmolality change in the cultures fed with EN/CYD 1X and 5X, respectively. e cultures with 3-MA exhibit higher osmolalities, since 3-MA was dissolved in HCl, and its addition involves raising the pH with NaHCO3, which is similar to addition of NaCl to the solution. In Figure 4.18, the change of osmolalities remained relatively small for EN/CYD 1X; in the most extreme case reducing from 320 mOsm/kg to 268 mOsm/kg. us, it can fit in our definition of isotonic settings. In contrast to that, in the cultures fed with EN/CYD 5X (Figure 4.19), the osmolalities of the cultures increase from 320 mOsm/kg to a range between 422 to 474 mOsm/kg. e rise in osmolality is gradual, with the rate of change in osmolality increasing slightly after the day the feeding volume increased. 68  500 C44-Osm(Lo)-3MA(0) C44-Osm(Lo)-3MA(5) C44-Osm(Lo)-3MA(10) C56-Osm(Lo)-3MA(0) C56-Osm(Lo)-3MA(5) C56-Osm(Lo)-3MA(10)  Osmolality (mOsm/kg)  450 400 350 300 250 200  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day)  Figure 4.18: Change of osmolality for cultures with EN/CYD 1X that undergo the lower osmolality regime  500  Osmolality (mOsm/kg)  450 400 350  C44-Osm(Hi)-3MA(0) C44-Osm(Hi)-3MA(5)  300  C44-Osm(Hi)-3MA(10) C56-Osm(Hi)-3MA(0)  250  C56-Osm(Hi)-3MA(5) C56-Osm(Hi)-3MA(10)  200  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure 4.19: Change of osmolality for cultures with EN/CYD 5X that undergo the higher osmolality regime  69  e cell culture and protein production data for both replicates can be found in Appendix I. In this study, changes in the average cell specific productivity and protein production were determined and for ease of comparison were normalized to their respective control cultures, either based on No 3-MA or isotonic (Low Osmolality) conditions. e data average and standard deviation were calculated from the combining normalized data of both replicates. e control cultures, by which the cultures are normalized, are noted with the blue colour in the following figures. e significance of the data was calculated based on one-tailed paired t-test. Figure 4.20 shows the normalized average of the cell specific productivities, from both replicates of the experimental design in Table 4.1. It clearly shows that addition of 3-MA at concentration of 5 and 10 mM to isotonic CHOm44 cultures has increased the cell specific productivity by 2 and 2.4 folds, and in CHOm56 cultures by 1.25 and 1.6 folds. At high osmolality, the only notable increase of qp was at 5 mM of 3-MA: 22% for CHOm44 and 9% for CHOm56. e average of increase in the normalized total amount of MAb is much smaller for both isoand hypertonic conditions. Both hypertonicity and 3-MA addition reduce the cell growth, and this counteracts the increase in cell specific productivity. Addition of 5 and 10 mM of 3-MA resulted in an average of 56% and 37% of increase in the final MAb in CHOm44 isotonic cultures. e similar case for CHOm56 cell cultures resulted in 7% and 10% increase in total MAb by addition of 5 and 10 mM of 3-MA. Addition of 3-MA to hypertonic cultures resulted in decrease on the final product.  70  3.5 Normalized Cell Specific Productivity  No 3-MA 5 mM 3-MA  3  10 mM 3-MA 2.5 2 1.5 1 0.5 0  C44-Osm(Lo)  C44-Osm(Hi)  C56-Osm(Lo)  C56-Osm(Hi)  Figure 4.20: Average of normalized cell specific productivity in the duration after addition of 3-MA of different CHO-MAb clones with high and low osmolalities  2 No 3-MA  1.8  5 mM 3-MA  Normalized Total MAb  1.6  10 mM 3-MA  1.4 1.2 1 0.8 0.6 0.4 0.2 0  C44-Osm(Lo)  C44-Osm(Hi)  C56-Osm(Lo)  C56-Osm(Hi)  Figure 4.21: Average of normalized final weight of MAb produced in fed-batch cultures of different CHOMAb clones with high and low osmolalities  71  3  Normalized Cell Specific Productivity  Low Osmolality 2.5  High Osmolality  2 1.5 1 0.5 0  C44-(0 mM)  C44-(5 mM)  C44-(10 mM)  C56-(0 mM)  C56-(5 mM)  C56-(10 mM)  Cell Line-(3-MA Conc.) Figure 4.22: Average of normalized cell specific productivity in the duration after addition of 3-MA of different CHO-MAb clones with different 3-MA concentrations  2.5 Low Osmolality  Normalized Total MAb  2  High Osmolality  1.5  1  0.5  0  C44-(0 mM)  C44-(5 mM)  C44-(10 mM)  C56-(0 mM)  C56-(5 mM)  C56-(10 mM)  Cell Line-(3-MA Conc.) Figure 4.23: Average of normalized final weight of MAb produced in fed-batch cultures of different CHOMAb clones with different 3-MA concentrations  72  In Figure 4.22, the cell specific productivities of hypertonic cultures are normalized to their counterparts with low osmolality regime, each at different 3-MA concentration. is clearly shows the effect of hypertonicity on the cell specific productivity, where the most notable increase of the cell specific productivity at high osmolality was achieved in CHOm44 cell line with no 3-MA treatment, increasing more than 2-fold. CHOm56 cells with no and 5 mM 3MA shows almost 25% and 11% increase in the cell specific productivity under hypertonic conditions. e rest of cultures showed reduction in cell specific productivity. e case for total MAb was also similar to the cell specific productivity. Under the hypertonic regime, the MAb production of 3-MA-free CHOm44 culture increased by average of 63% (Figure 4.23). CHOm56 cells with no and 5 mM of 3-MA had 21% and 7% increase under the hypertonic regime. Hypertonic regime didn’t result in any improvement in the MAb production of the rest of the cultures. Table 4.2: P-values for the actual and normalized amounts of productivity and total MAb, using a one-tailed paired t-test compared to cultures without 3-MA  Cell Line (Osmolality) C44 (Low) C44 (High) C56 (Low) C56 (High)  3-MA Conc. 5 10 5 10 5 10 5 10  P-value for Actual Values Total Productivity MAb 0.059* 0.057* 0.077* 0.098* 0.150 0.246 0.482 0.123 0.001** 0.088 0.169 0.340 0.086* 0.278 0.161 0.411  P-value for Normalized Values Total Productivity MAb 0.106 0.125 0.121 0.154 0.159 0.246 0.498 0.116 0.035** 0.099* 0.184 0.333 0.097* 0.282 0.154 0.074*  Based on the traditional α=0.05, only a few of the P-values derived from the paired t-test were significant (marked with **) (Table 4.2 and Table 4.3). However, based on the less strict 73  α=0.1 some other factors were found significant (marked with *). Nevertheless, it should be noted that P-values are not the sole determinant of the significance of the factors, and decisions of significance can encompass other elements such as previous knowledge (J. J. Lee 2011; Pieracci et al. 2010; Maldonado and West 2011). It is important to note that “absence of evidence is not evidence of absence” (Altman and Bland 1995), and the lack of significance in the normalized data may well be due to small sample size (Ranstam 2012; Harvey and Lang 2010). Also, defining a cut-off value for the fold increase and interpreting data on that basis, can drastically change the significances calculated (Dalman et al. 2012). us it seems logical that, inferring from our previous knowledge and the current results, hypertonicity in 3-MA-free conditions can indeed increase the total MAb despite the P-value being higher than 0.05. Table 4.3: P-values for the actual and normalized amounts of productivity and total MAb, using a one tailed paired t-test compared to cultures with low osmolality  Cell Line (3-MA Conc.) C44 (0 mM) C44 (5 mM) C44 (10 mM) C56 (0 mM) C56 (5 mM) C56 (10 mM)  P-value for Actual Values Osmolality Total Productivity MAb Low 0.059* 0.167 High 0.050* 0.387 Low 0.011** 0.105 High 0.001** 0.163 Low 0.267 0.096* High 0.063* 0.204  P-value for Normalized Values Total Productivity MAb 0.167 0.200 0.387 0.415 0.105 0.075* 0.163 0.169 0.096* 0.101 0.204 0.192  Finally these results clearly showed that there are clonal differences in the change of cell specific productivity and final production with 3-MA and hypertonicity or a combination of both. ere seems to be a production capacity that couldn’t be overcome by any of these methods. However, it seems that both 3-MA addition and hypertonicity can overcome some of 74  the mechanisms that has caused cells to decrease in productivity. e change in the magnitude of increase was greatest for CHOm44 cells (bigger error bars compared to that of CHOm56), indicating that 3-MA and hypertonicity may have a bigger effect on cells with lower productivities.  4.6  Effects of Autophagy Inhibition and Hypertonicity on the MAb Glycosylation  Patterns Hypertonicity and autophagy inhibition by 3-MA were shown to increase MAb production to a great extent in CHOm44 cells and to some extent in CHOm56 cells. However, these changes in the fed-batch process might change other important properties of the final product such as glycan profiles, which may render it unsuitable for therapeutic usage. For example, hypertonicity can affect the glycosylation patterns of recombinant proteins by mammalian cells (R. Kimura and Miller 1999; Pacis et al. 2011). Also, 3-MA was shown to modulate the tPA glycan profiles (Jardon et al. 2012). Since glycosylation may affect protein’s half-life and therapeutic effectiveness (Elliott et al. 2003; Sinclair and Elliott 2005) and it is one of the important tools for assessing biosimilarity (Walsh 2010), they should be among the studies needed for assessments of MAbs by regulatory agencies (Walsh and Roy Jefferis 2006). e glycan profiles of several selected samples were analyzed and the results. Since the analyses were not done in our laboratory, and the number of samples was limited, only a subset was chosen, from cultures with 0 and 5 mM of 3-MA under different osmolality regimes from the second replicate (See Appendix I.2). e extended glycosylation results and legends to glycan structures can be found in the Appendix J. 75  e relative proportion of oligosaccharides of the heavy chains of selected samples from CHOm44 and CHOm56 cultures are shown in Figure 4.24 and Figure 4.25, respectively; where G and F refer to galactose and fucose with the corresponding number of molecules attached to the core heptasaccharide glycan structure G0 (a list of glycan structures can be found in Table J.3). Although there were some variations in the glycan profiles of the MAb products, no single trend of change can be observed. Most of the heavy chain glycan profiles were relatively stable under 3-MA treatment or varying osmolality, especially in the MAb produced in CHOm56 cultures. e glycosylation profiles of both heavy and light chains are also described in terms of different indices in Table 4.4 and Table 4.5, respectively. Fucosylation, sialylation, galactosylation and antennarity indices are the proportion of the occurrence of fucose, sialic acid, galactose and GlcNAc residues per glycan structure. e data from both tables suggest that varying conditions had little to affect the glycosylation indices. e results are compatible with the properties of the heavy chain antibody glycosylation described in literature; high amounts of non-galactosylated fucosylated glycan structures (G0F) followed by fucosylated structures with one (G1F) or two (G2F) galactose residues (Del Val et al. 2010), with limited high-mannose structures (Kornfeld et al. 1978) and scarcity of sialylation (Walsh and Roy Jefferis 2006). e MAb produced by CHO-MAb cells was glycosylated on the light chain. e glycosylation sites in Fv regions of antibodies (which light chain glycans exclusively occur in that part) are found in only ~20% human IgGs. Also, there is no strong relationship between their Fv glycan structure and their clearance rate (L. Huang et al. 2006), and unlike the glycosylation in heavy chains, they do not affect the MAb configuration (Del Val 76  et al. 2010). Some reports suggest that glycosylation of the variable regions of either heavy or light chains may affect the binding of the antibody to antigens (R Jefferis 2005), although the observations were not conclusive (Wallick et al. 1988; Co et al. 1993). e light chains are more accessible for glycotransferase enzymes, thus carry more processed sialylated glycans, as can be seen in Table 4.5. e glycan structures have important effects on antibody effector and binding functions. us, the relative robustness of the glycan profiles under both 3-MA treatment and varying osmolality regimes is supporting evidence that neither of the two production increase approaches have compromised the consistence and biosimilarity of their MAb product to that of a normal fed-batch process.  77  25  Relative area of glycans (%)  C44-Osm(Lo)-3MA(0) 20  C44-Osm(Lo)-3MA(5) C44-Osm(Hi)-3MA(0) C44-Osm(Hi)-3MA(5)  15  10  5  0  G0  G0F  G1F  Man5  Figure 4.24: Comparison of the relative percentage of the glycoforms found in the heavy chains of MAb produced by CHOm44 cells at different osmolality regimes and 3-MA concentrations  25  Relative area of glycans (%)  C56-Osm(Lo)-3MA(0) 20  C56-Osm(Lo)-3MA(5) C56-Osm(Hi)-3MA(0) C56-Osm(Hi)-3MA(5)  15  10  5  0  G0  G0F  G1F  Man5  Figure 4.25: Comparison of the relative percentage of the glycoforms found in the heavy chains of MAb produced by CHOm56 cells at different osmolality regimes and 3-MA concentrations  78  Table 4.4: Glycosylation analysis results for the heavy chain of the MAb produced by CHO-MAb clones  Clone  Osmolality  3-MA Conc. Fucosylation Sialylation Antennarity Galactosylation (mM) Index Index Index Index 0  0.39  –  1.59  0.24  5  0.39  –  1.59  0.26  0  0.42  –  1.61  0.31  5  0.40  –  1.56  0.25  0  0.41  –  1.58  0.28  5  0.40  –  1.52  0.22  0  0.41  –  1.53  0.25  5  0.41  –  1.53  0.25  High CHOm44 Low  High CHOm56 Low  79  Table 4.5: Glycosylation analysis results for the light chain of the MAb produced by CHO-MAb clones  Clone  Osmolality  3-MA Conc. Fucosylation Sialylation Antennarity Galactosylation (mM) Index Index Index Index 0  0.37  –  1.51  0.26  5  0.40  0.05  1.57  0.48  0  0.42  0.03  1.47  0.37  5  0.42  0.05  1.54  0.44  0  0.40  0.05  1.57  0.46  5  0.34  0.05  1.46  0.37  0  0.43  0.03  1.46  0.38  5  0.32  0.05  1.26  0.22  High CHOm44 Low  High CHOm56 Low  80  4.7  Conclusions to the Autophagy Inhibition and Hypertonicity Effects on CHO-MAb  Fed-batch Cultures Autophagy inhibition by 3-MA was previously shown to increase tPA production in CHO cells by almost 3-fold (Jardon et al. 2012). e impact of autophagy inhibition was tested in two different clones of a MAb producing CHO cell line, and compared with the established effect of hypertonicity in increasing productivity. It was shown that both autophagy inhibition and hypertonicity could be effective in increasing productivity of the CHO-MAb cells. However, the magnitude of these effects was dependent on different clones, suggesting the productivity of the high producer clone (CHOm56) couldn’t increase with the extent similar to that of the low producer clone (CHOm44). Combining the two strategies did not result in increasing the cell specific productivity or total MAb further. e glycan structures of the proteins were checked to assess the potential changes that might happen due to change of the culture milieu resulted from autophagy inhibition or hypertonicity. It was found that most of the changes in the glycan structure pattern were not pronounced, and regardless of the conditions, the cells were able to produce relatively consistent glycoforms. Robustness of glycosylation patterns is a favourable argument to demonstrate the product quality doesn’t change using autophagy inhibition or hypertonicity, and reduces the regulatory concerns.  81  Chapter 5: Conclusions and view to future  Biopharmaceuticals are the new prospects of treating diseases like cancer without many adverse effects in their traditional treatments. However, their high price poses a major challenge to their widespread application. Improvements to the manufacturing process can ease this burden, and help to provide easy access to high quality biological medications at the disposal of larger sections of society. is thesis addresses two methods of increasing the productivity of the CHO cells expressing monoclonal antibodies. Firstly, we established a high-performance fed-batch protocol for the CHO-MAb cell line, based on the previously established platform protocol for the CHO-tPA cell line. rough a series of screening design of experiments, several factors that described the platform process were modulated. Deviations from the platform process of many of the speculated factors either didn’t significantly improve the process or had a negative impact. Moreover, the amino acid profiles of the cultures cultivated using the developed CHO-MAb fed-batch protocol were determined. Fortunately, the established protocol was capable of providing all amino acids without any serious depletion. Secondly, we showed that autophagy inhibition, which significantly increased the protein production in CHO-tPA cultures, also increased the productivity of CHO-MAb cells. However, the extent of increase was found extremely dependent on particular clones. Autophagy inhibition was able to increase the productivity of CHOm44 clone (the low producer) more than 2-fold, in isotonic conditions. e magnitude of increase was lower in CHOm56 clone (the high producer). 82  irdly, hypertonicity was also able to increase the productivity of CHOm44 cell more than 2-fold. Similar to autophagy inhibition, the magnitude of increase in productivity due to hypertonicity was lower in CHOm56 cells. Despite the capability of both autophagy inhibition and hypertonicity in increasing productivity, merging both strategies was not beneficial in improving productivity. Finally, we assessed the impact of autophagy inhibition and hypertonicity on the glycosylation patterns of the MAb product. It was found that the glycan profiles were resistant to radical changes under both autophagy inhibition and hypertonic regimes. Since any change in the cell culture milieu can reflect itself in the quality of the products, this is a favourable argument in demonstrating that the MAb produced under these conditions is biosimilar to what is produced in the control fed-batch cultures. is thesis extends the findings of the impact of autophagy modulation on significantly increasing CHO-tPA cells recombinant protein productivity. We have shown that not all the cell lines and clones respond the same to the modulation of autophagy. e CHO-MAb clone with higher productivity was less responsive to the modulation of autophagy and osmolality. is might suggested a connection between the specific productivity of the cell lines and their response to autophagy inhibition, especially as CHO-tPA cells (that were very responsive to autophagy inhibition) also had a relatively low productivity. us, the relationship between the productivity of the cell lines and their response to autophagy can be interesting subject to look at in the future. Since the extent of increase in productivity using both autophagy inhibition and hypertonicity wasn’t very different, and the glycan profiles were similar in different conditions, 83  the applicability of 3-MA in industrial fed-batch culture may be questioned; especially as it involves introduction of a chemical agent that may cause upstream, downstream and regulatory complications. us, comparing the effect of hypertonicity and autophagy inhibition on CHOtPA or other cell lines would be interesting to assess, since it may provide insights on the usability of 3-MA inhibition of autophagy in fed-batch cultures. Furthermore, imitating the effect of chemicals that inhibit autophagy using cell line engineering approaches should be considered as a long-term approach.  84  Bibliography Aggarwal S. 2011. “What’s Fueling the Biotech Engine—2010 to 2011.” Nature Biotechnology, 29 (12): 1083–9. Allen AC, Harper SJ and Feehally J. 1995. “Galactosylation of N- and O-linked Carbohydrate Moieties of IgA1 and IgG in IgA Nephropathy.” Clinical and Experimental Immunology, 100 (3): 470–4. Altman DG and Bland JM. 1995. “Absence Of Evidence Is Not Evidence Of Absence.” British Medical Journal, 311 (7003): 485. Andersen DC, Bridges T, Gawlitzek M and Hoy C. 2000. “Multiple Cell Culture Factors Can Affect the Glycosylation of Asn-184 in Plasminogen Activator.” Biotechnology and Bioengineering, 70 (1): 25–31. Arden N and Betenbaugh MJ. 2004. “Life and Death in Mammalian Cell Culture: Strategies for Apoptosis Inhibition.” Trends in Biotechnology, 22 (4): 174–80. Barth S, Glick D and Macleod KF. 2010. “Autophagy: Assays and Artifacts.” Journal of Pathology, 221 (2): 117–24. Bennun SV, Krambeck FJ and Betenbaugh MJ. 2009. “Glycoengineering and Modeling of Protein N-Glycosylation.”. In M. Al-Rubeai, ed. Cell Line Development. Dordrecht: Springer Netherlands, 217–231. 85  Bibila TA, Ranucci CS, Glazomitsky K, Buckland BC and Aunins JG. 1994. “Monoclonal Antibody Process Development Using Medium Concentrates.” Biotechnology Progress, 10 (1): 87–96. Burg MB, Ferraris JD and Dmitrieva NI. 2007. “Cellular Response to Hyperosmotic Stresses.” Physiological Reviews, 87: 1441–1474. Butler M. 2005. “Animal Cell Cultures: Recent Achievements and Perspectives in the Production of Biopharmaceuticals.” Applied Microbiology and Biotechnology, 68 (3): 283– 91. Chu L and Robinson DK. 2001. “Industrial Choices for Protein Production by Large-scale Cell Culture.” Current Opinion in Biotechnology, 12 (2): 180–7. Co MS, Scheinberg DA, Avdalovic NM, McGraw K, Vasquez M, Caron PC and Queen C. 1993. “Genetically Engineered Deglycosylation Of e Variable Domain Increases e Affinity Of An Anti-CD33 Monoclonal Antibody.” Molecular Immunology, 30 (15): 1361– 1366. Costa AR, Rodrigues ME, Henriques M, Azeredo J and Oliveira R. 2010. “Guidelines to Cell Engineering for Monoclonal Antibody Production.” European Journal of Pharmaceutics and Biopharmaceutics, 74 (2): 127–38.  86  Dalman MR, Deeter A, Nimishakavi G and Duan Z-H. 2012. “Fold Change and P-value Cutoffs Significantly Alter Microarray Interpretations.” BMC Bioinformatics, 13 Suppl 2 (Suppl 2): S11. Dreesen IAJ and Fussenegger M. 2010. “Ectopic Expression of Human mTOR Increases Viability, Robustness, Cell Size, Proliferation, and Antibody Production of Chinese Hamster Ovary Cells.” Biotechnology and Bioengineering, 108 (4): 853–866. Elliott S, Lorenzini T, Asher S, Aoki K, Brankow D, Buck L, Busse L, Chang D, et al. 2003. “Enhancement of erapeutic Protein in Vivo Activities rough Glycoengineering.” Nature Biotechnology, 21 (4): 414–21. Endo T, Wright A, Morrison SL and Kobata A. 1995. “Glycosylation Of e Variable Region G-site Specific Maturation Of e Sugar Chains.” Molecular Immunology, 32 (13): 931– 940. Frahm B, Lane P, Märkl H and Pörtner R. 2003. “Improvement of a Mammalian Cell Culture Process by Adaptive, Model-based Dialysis Fed-batch Cultivation and Suppression of Apoptosis.” Bioprocess and Biosystems Engineering, 26 (1): 1–10. Fujimura Y, Tachibana H, Eto N and Yamada K. 2000. “Antigen Binding of an Ovomucoidspecific Antibody Is Affected by a Carbohydrate Chain Located on the Light Chain Variable Region.” Bioscience, Biotechnology, and Biochemistry, 64 (11): 2298–2305.  87  Gerngross TU. 2004. “Advances in the Production of Human erapeutic Proteins in Yeasts and Filamentous Fungi.” Nature Biotechnology, 22 (11): 1409–14. Han YK, Kim Y-G, Kim JY and Lee GM. 2010. “Hyperosmotic Stress Induces Autophagy and Apoptosis in Recombinant Chinese Hamster Ovary Cell Culture.” Biotechnology and Bioengineering, 105 (6): 1187–92. Han YK, Koo TY and Lee GM. 2009. “Enhanced Interferon-b Production by CHO Cells rough Elevated Osmolality and Reduced Culture Temperature.” Biotechnology Progress, 25: 1440–1447. Harvey BJ and Lang TA. 2010. “Hypothesis Testing, Study Power, and Sample Size.” Chest, 138 (3): 734–7. He C and Klionsky DJ. 2009. “Regulation Mechanisms and Signaling Pathways of Autophagy.” Annual Review of Genetics, 43: 67–93. Heath C. 2010. “Advancing Our Cell Culture Platform: Incorporating Lessons Learned and New Technologies.”. In S. J. Shire et al., eds. Current Trends in Monoclonal Antibody Development and Manufacturing. New York, NY: Springer New York, 63–72. Hossler P, Khattak SF and Li ZJ. 2009. “Optimal and Consistent Protein Glycosylation in Mammalian Cell Culture.” Glycobiology, 19 (9): 936–49.  88  Hou JJC, Codamo J, Pilbrough W, Hughes B, Gray PP and Munro TP. 2011. “New Frontiers in Cell Line Development: Challenges for Biosimilars.” Journal of Chemical Technology & Biotechnology, 86 (7): 895–904. Huang L, Biolsi S, Bales KR and Kuchibhotla U. 2006. “Impact of Variable Domain Glycosylation on Antibody Clearance: An LC/MS Characterization.” Analytical Biochemistry, 349 (2): 197–207. Hwang SO and Lee GM. 2008. “Nutrient Deprivation Induces Autophagy As Well As Apoptosis in Chinese Hamster Ovary Cell Culture.” Biotechnology, 99 (3): 678–685. Jardon MA, Sattha B, Braasch K, Leung AO, Côté HCF, Butler M, Gorski SM and Piret JM. 2012. “Inhibition of Glutamine-dependent Autophagy Increases t-PA Production in CHO Cell Fed-batch Processes.” Biotechnology and Bioengineering, 109 (5): 1228–38. Jardon MA. 2012. Modulating Autophagy and Glutamine Metabolism in CHO Cells to Increase Fed-batch Process Performance. e University of British Columbia. Jefferis R. 2009. “Glycosylation of Antibody erapeutics: Optimisation for Purpose.”. In L. Faye & V. Gomord, eds. Methods in Molecular Biology, Recombinant Proteins From Plants. Totowa, NJ: Humana Press, 223–238. Jefferis R. 2005. “Glycosylation of Recombinant Antibody erapeutics.” Biotechnology Progress, 21 (1): 11–16.  89  Kennard ML, Goosney DL, Monteith D, Zhang L, Moffat M, Fischer D and Mott J. 2009. “e Generation of Stable, High MAb Expressing CHO Cell Lines Based on the Artificial Chromosome Expression (ACE) Technology.” Biotechnology and Bioengineering, 104 (3): 540–53. Kiehl TR, Shen D, Khattak SF, Jian Li Z and Sharfstein ST. 2011. “Observations of Cell Size Dynamics Under Osmotic Stress.” Cytometry. Part A  : the Journal of the International Society for Analytical Cytology, 79 (7): 560–9. Kim MS, Kim NS, Sung YH and Lee GM. 2002. “Biphasic Culture Strategy Based On Hyperosmotic Pressure For Improved Humanized Antibody Production In Chinese Hamster Ovary Cell Culture Biphasic Culture Strategy Based On Hyperosmotic Pressure For Improved Humanized Antibody Production In Chinese Hamste.” In Vitro Cellular & Developmental Biology – Animal, 38 (6): 314–319. Kim NS and Lee GM. 2002. “Response of Recombinant Chinese Hamster Ovary Cells to Hyperosmotic Pressure: Effect of Bcl-2 Overexpression.” Journal of Biotechnology, 95 (3): 237–48. Kim TK, Ryu JS, Chung JY, Kim MS and Lee GM. 2000. “Osmoprotective Effect of Glycine Betaine on rombopoietin Production in Hyperosmotic Chinese Hamster Ovary Cell Culture: Clonal Variations.” Biotechnology Progress, 16 (5): 775–81.  90  Kimura R and Miller WM. 1999. “Glycosylation of CHO-derived Recombinant tPA Produced Under Elevated pCO2.” Biotechnology Progress, 15 (1): 146. Kimura S, Noda T and Yoshimori T. 2007. “Dissection of the Autophagosome Maturation Process by a Novel Reporter Protein, Tandem Fluorescent-Tagged LC3.” Autophagy, 3 (5): 452–460. Klionsky DJ, Abeliovich H, Agostinis P, Agrawal DK, Aliev G, Askew DS, Baba M, Baehrecke EH, et al. 2009. “Guidelines for the Use and Interpretation of Assays for Monitoring Autophagy in Higher Eukaryotes.” Autophagy, 4 (2): 151–175. Klionsky DJ, Cregg JM, Dunn WA, Emr SD, Sakai Y, Sandoval IV, Sibirny A, Subramani S, et al. 2003. “A Unified Nomenclature for Yeast Autophagy-Related Genes.” Developmental Cell, 5 (October): 539–545. Konno Y, Kobayashi Y, Takahashi K, Takahashi E, Sakae S, Wakitani M, Yamano K, Suzawa T, et al. 2012. “Fucose Content of Monoclonal Antibodies Can Be Controlled by Culture Medium Osmolality for High Antibody-dependent Cellular Cytotoxicity.” Cytotechnology, 64 (3): 249–65. Kornfeld S, Li E and Tabas I. 1978. “e Synthesis of Complex-type Oligosaccharides.” e Journal of Biological Chemistry, 253 (21): 7771–7778. Kroemer G, Mariño G and Levine B. 2010. “Autophagy and the Integrated Stress Response.” Molecular Cell, 40 (2): 280–93. 91  Kundu M and ompson CB. 2008. “Autophagy: Basic Principles and Relevance to Disease.” Annual Review of Pathology, 3: 427–55. Lee JJ. 2011. “Demystify Statistical Significance--time to Move on from the p Value to Bayesian Analysis.” Journal of the National Cancer Institute, 103 (1): 2–3. Lee S-Y, Kwon Y-B, Cho J-M, Park K-H, Chang S-J and Kim D-I. 2012. “Effect of Process Change from Perfusion to Fed-batch on Product Comparability for Biosimilar Monoclonal Antibody.” Process Biochemistry, 47 (9): 1411–1418. Leibiger H, Wüstner D, Stigler R-D and Marx U. 1999. “Variable Domain-linked Oligosaccharides of a Human Monoclonal IgG: Structure and Influence on Antigen Binding.” Biochemistry Journal, 338: 529–538. Li J, Wong CL, Vijayasankaran N, Hudson T and Amanullah A. 2012. “Feeding Lactate for CHO Cell Culture Processes: Impact on Culture Metabolism and Performance.” Biotechnology and Bioengineering, 109 (5): 1173–86. Lim Y, Wong NSC, Lee YY, Ku SCY, Wong DCF and Yap MGS. 2010. “Engineering Mammalian Cells in Bioprocessing – Current Achievements and Future Perspectives.” Biotechnology and Applied Biochemistry, 189 (55): 175–189. Luo J, Vijayasankaran N, Autsen J, Santuray R, Hudson T, Amanullah A and Li F. 2012. “Comparative Metabolite Analysis to Understand Lactate Metabolism Shift in Chinese Hamster Ovary Cell Culture Process.” Biotechnology and Bioengineering, 109 (1): 146–56. 92  Maiuri MC, Zalckvar E, Kimchi A and Kroemer G. 2007. “Self-eating and Self-killing: Crosstalk Between Autophagy and Apoptosis.” Nature Reviews. Molecular Cell Biology, 8 (9): 741–52. Maldonado F and West CP. 2011. “Interpreting the p Value.” Chest, 139 (3): 723–4; author reply 724. Mizushima N, Ohsumi Y and Yoshimori T. 2002. “Autophagosome Formation in Mammalian Cells.” Cell Structure and Function, 429 (27): 421–429. Mohan C, Kim Y and Lee GM. 2009. “Apoptosis and Autophagy Cell Engineering.”. In M. AlRubeai, ed. Cell Line Development. Dordrecht: Springer Netherlands, 195–216. Montgomery DC. 2009. Design and Analysis of Experiments 7th ed., Hoboken, NJ: Wiley. Ozturk S and Palsson B. 1991. “Effect of Medium Osmolarity on Hybridoma Growth, Metabolism, and Antibody Production.” Biotechnology and Bioengineering, 37: 989–993. Pacis E, Yu M, Autsen J, Bayer R and Li F. 2011. “Effects of Cell Culture Conditions on Antibody N-linked Glycosylation-what Affects High Mannose 5 Glycoform.” Biotechnology and Bioengineering, 108 (10): 2348–2358. Patel TP, Parekh RB, Moellering BJ and Prior CP. 1992. “Different Culture Methods Lead to Differences in Glycosylation of a Murine IgG Monoclonal Antibody.” Biochemical Journal, 285: 839–845. 93  Pattingre S, Tassa A, Qu X, Garuti R, Liang XH, Mizushima N, Packer M, Schneider MD, et al. 2005. “Bcl-2 Antiapoptotic Proteins Inhibit Beclin 1-dependent Autophagy.” Cell, 122 (6): 927–39. Pavlou AK and Reichert JM. 2004. “Recombinant Protein erapeutics--success Rates, Market Trends and Values to 2010.” Nature Biotechnology, 22 (12): 1513–9. Pieracci J, Perry L and Conley L. 2010. “Using Partition Designs to Enhance Purification Process Understanding.” Biotechnology and Bioengineering, 107 (5): 814–24. Piret JM, Devens DA and Cooney CL. 1991. “Nutrient and Metabolite Gradients in Mammalian Cell Hollow Fiber Bioreactors.” e Canadian Journal of Chemical Engineering, 69 (April): 421–428. Ranstam J. 2012. “Why the P-value Culture Is Bad and Confidence Intervals a Better Alternative.” Osteoarthritis and Cartilage / OARS, Osteoarthritis Research Society, 20 (8): 805–8. Royle L, Radcliffe CM, Dwek RA and Rudd PM. 2006. “Detailed Structural Analysis of Nglycans Released from Glycoproteins in SDS-PAGE Gel Bands Using HPLC Combined with Exoglycosidase Array Digestions.”. In I. Brockhausen, ed. Methods in molecular biology, vol. 347: Glycobiology Protocols. Totowa, NJ: Humana Press, 125–43. Rubinsztein DC, Cuervo AM, Ravikumar B, Sarkar S, Korolchuk V, Kaushik S and Klionsky DJ. 2009. “In Search of an “autophagomometer”.” Autophagy, 5 (5): 585–589. 94  Rönsch H, Krämer R and Morbach S. 2003. “Impact of Osmotic Stress on Volume Regulation, Cytoplasmic Solute Composition and Lysine Production in Corynebacterium Glutamicum MH20-22B.” Journal of Biotechnology, 104 (1-3): 87–97. Sauer PW, Burky JE, Wesson MC, Sternard HD and Qu L. 2000. “A High-yielding, Generic Fed-batch Cell Culture Process for Production of Recombinant Antibodies.” Biotechnology and Bioengineering, 67 (5): 585–97. Schellekens H. 2004. “Biosimilar erapeutic Agents: Issues with Bioequivalence and Immunogenicity.” European Journal of Clinical Investigation, 34 (12): 797–9. Seglen PO and Gordon PB. 1982. “3-Methyladenine: Specific Inhibitor of Autophagic/lysosomal Protein Degradation in Isolated Rat Hepatocytes.” Proceedings of the National Academy of Sciences of the United States of America, 79 (6): 1889–92. Shen CF and Kamen A. 2012. “Hyperosmotic Pressure on HEK 293 Cells During the Growth Phase, but Not the Production Phase, Improves Adenovirus Production.” Journal of Biotechnology, 157 (1): 228–36. Shvets E, Fass E and Elazar Z. 2008. “Utilizing Flow Cytometry to Monitor Autophagy in Living Mammalian Cells.” Autophagy, 4 (5): 621–628. Sinclair AM and Elliott S. 2005. “Glycoengineering: e Effect of Glycosylation on the Properties of erapeutic Proteins.” Journal of Pharmaceutical Sciences, 94 (8): 1626–35.  95  Tachibana H, Kim J and Shirahata S. 1997. “Building High Affinity Human Antibodies by Altering the Glycosylation on the Light Chain Variable Region in N-acetylglucosaminesupplemented Hybridoma Cultures.” Cytotechnology, 23 (1-3): 151–9. Takagi M, Moriyama T and Yoshida T. 2001. “Effects of Shifts Up and Down in Osmotic Pressure on Production of Tissue Plasminogen Activator by Chinese Hamster Ovary Cells in Suspension.” Journal of Bioscience and Bioengineering, 91 (5): 509–514. Taylor ME and Drickamer K. 2011. Introduction to Glycobiology 3rd ed., Oxford; New York: Oxford University Press. Del Val IJ, Kontoravdi C and Nagy JM. 2010. “Towards the Implementation of Quality by Design to the Production of erapeutic Monoclonal Antibodies with Desired Glycosylation Patterns.” Biotechnology Progress, 26 (6): 1505–27. Wallick SC, Kabat EA and Morrison SL. 1988. “Glycosylation of a VH Residue of a Monoclonal Antibody Against Alpha (1→6) Dextran Increases Its Affinity for Antigen.” e Journal of Experimental Medicine, 168 (3): 1099–109. Walsh G and Jefferis R. 2006. “Post-translational Modifications in the Context of erapeutic Proteins.” Nature Biotechnology, 24 (10): 1241–52. Walsh G. 2010. “Biopharmaceutical Benchmarks 2010.” Nature Biotechnology, 28 (9): 917–24.  96  Wehner F, Olsen H, Tinel H, Kinne-Saffran E and Kinne RKH. 2003. “Cell Volume Regulation: Osmolytes, Osmolyte Transport, and Signal Transduction.” Reviews of Physiology, Biochemistry and Pharmacology, 148: 1–80. Werner RG, Kopp K and Schlueter M. 2007. “Glycosylation of erapeutic Proteins in Different Production Systems.” Acta Paediatrica (Oslo, Norway  : 1992). Supplement, 96 (455): 17–22. Werner RG. 2004. “Economic Aspects of Commercial Manufacture of Biopharmaceuticals.” Journal of Biotechnology, 113 (1-3): 171–82. Wlaschin KF and Hu Wei-Shou. 2006. “Fedbatch Culture and Dynamic Nutrient Feeding.” Advances in Biochemical Engineering / Biotechnology, 101 (June): 43–74. Wright A and Morrison SL. 1994. “Effect of Altered CH2-associated Carbohydrate Structure on the Functional Properties and in Vivo Fate of Chimeric Mouse-human Immunoglobulin G1.” e Journal of Experimental Medicine, 180 (3): 1087–96. Wright A and Morrison SL. 1997. “Effect of Glycosylation on Antibody Function: Implications for Genetic Engineering.” Trends in Biotechnology, 15 (1): 26–32. Wright A, Tao M, Kabat EA and Morrison SL. 1991. “Antibody Variable Region Glycosylation: Position Effects on Antigen Binding and Carbohydrate Structure.” e EMBO Journal, 10 (10): 2717–23.  97  Wu Y-T, Tan H-L, Shui G, Bauvy C, Huang Q, Wenk MR, Ong C-N, Codogno P, et al. 2010. “Dual Role of 3-methyladenine in Modulation of Autophagy via Different Temporal Patterns of Inhibition on Class I and III Phosphoinositide 3-kinase.” e Journal of Biological Chemistry, 285 (14): 10850–61. Wurm FM. 2004. “Production of Recombinant Protein erapeutics in Cultivated Mammalian Cells.” Nature Biotechnology, 22 (11): 1393–8. Zhou Q, Shankara S, Roy A, Qiu H, Estes S, McVie-Wylie A, Culm-Merdek K, Park A, et al. 2008. “Development of A Simple and Rapid Method for Producing Non-Fucosylated Oligomannose Containing Antibodies With Increased Effector Function.” Biotechnology and Bioengineering, 99 (3): 652–665. Zhou W, Rehm J, Europa A and Hu Wei-Shu. 1997. “Alteration of Mammalian Cell Metabolism by Dynamic Nutrient Feeding.” Cytotechnology, 24 (2): 99–108.  98  Appendices  Appendix A. Oxygen Limitation in Shake Flasks due to Increase in Volume Due to addition of feed solutions, the volume of a fed-batch culture can increase during the course of operation. In our fed-batch protocols the amount of feed solutions added and the samples taken are almost equal in the beginning of cultures (1 mL Feed A + 0.5 mL EN 1X + 0.5 mL CYD 1X and 1.5 mL sampling). Of course, this is dependent on whether we used EN/CYD 1X or 5X, where when using 5X, the volume initially decreased. However, after increasing the feed solution volume 3-fold, the volume increased rapidly. As the cells were cultured in shake flasks without feedback control of the O2 level, surface area to volume ratio changes in the flasks could result in oxygen limitation. We tested the performance of cell cultures using two different volumes one at the maximum volume recommended by the flask manufacturer (35 mL in a 125 mL flask), and the other at half of that volume (17.5 mL). e cultures were fed according to the CHO-tPA fed-batch protocol, but without addition of EN/CYD, and the sampling volume was adjusted to keep the volume constant (i.e. 1 mL sampled and1 mL Feed A added). As seen in Figure A.1, the viable cell concentrations of both cultures are practically the same, especially considering the errors associated with the measurements. Figure A.2 suggests that the MAb concentration of the culture with 35 mL was higher. is shows that having higher volumes in the range recommended by the manufacturer did not result in oxygen limitations that reduced growth or MAb production. ese results were used as a basis to omit oxygen limitation from the culture variables considered.  99  14  Viable Cell Conc. (106 cell/mL)  12 10 17.5 mL  8  35 mL  6 4 2 0  0  1  2  3  4  5  6 7 Time (day)  8  9  10  11  12  Figure A.1: Viable cell concentrations of CHOm56 fed-batch cultures in different volumes cultured in 125 mL shake flasks  1400 1200  MAb Conc. (g/mL)  1000 800 600 17.5 mL  400  35 mL 200 0  0  1  2  3  4  5  6 7 Time (day)  8  9  10  11  12  Figure A.2: Monoclonal antibody concentrations of CHOm56 fed-batch cultures in different volumes cultured in 125 mL shake flasks  100  Appendix B. Supplementary Data for Section 3.1  40 Batch  Glucose Conc. (mM)  35  Fed-Batch  30 25 20 15 10 5 0  0  2  4  6  8  10  12  14  16  18  20  18  20  Time (day)  Figure B.1: Glucose concentration of batch and fed-batch cultures of CHOm44 cell line  20 Batch  18 Lactate Conc. (mM)  16  Fed-Batch  14 12 10 8 6 4 2 0  0  2  4  6  8  10  12  14  16  Time (day)  Figure B.2: Lactate concentration of batch and fed-batch cultures of CHOm44 cell line  101  5 Batch  4.5 Glutamine Conc. (mM)  4  Fed-Batch  3.5 3 2.5 2 1.5 1 0.5 0  0  2  4  6  8  10  12  14  16  18  20  Time (day)  Figure B.3: Glutamine concentration of batch and fed-batch cultures of CHOm44 cell line  102  Appendix C. Supplementary Data for Section 3.2.1 e alphabetical codenames and the feeding strategies are introduced in Table 3.2.  100 90  Viability (%)  80 70 60  A  B  C  D  E  F  50 40  0  1  2  3  4  5  6 7 8 Time (day)  9  10  11  12  13  9  10  11  12  13  14  Figure C.1: Viability of the cultures with the 3F feeding strategy  100 90  Viability (%)  80 70  G  H  I  J  K  L  60 50 40  0  1  2  3  4  5  6 7 8 Time (day)  14  Figure C.2: Viability of the cultures with the 1F3F feeding strategy  103  700 600  MAb Conc. (mg/L)  500  A  B  C  D  E  F  400 300 200 100 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  12  13  14  Time (day) Figure C.3: The MAb concentrations produced in the cultures with the 3F feeding strategy  700  MAb Conc. (mg/L)  600 500  G  H  I  J  K  L  400 300 200 100 0  0  1  2  3  4  5  6  7  8  9  10  11  Time (day) Figure C.4: The MAb concentrations produced in the cultures with the 1F3F feeding strategy  104  70 60  Glucose Conc. (mM)  50 40 30 20 10 0  0  1  A  B  C  D  E  F  2  3  4  5  6  7  8  9  10  11  12  13  14  12  13  14  Time (day) Figure C.5: The glucose concentration profiles of the cultures with the 3F feeding strategy  70  Glucose Conc. (mM)  60 50  G  H  I  J  K  L  40 30 20 10 0  0  1  2  3  4  5  6  7  8  9  10  11  Time (day) Figure C.6: The glucose concentration profiles of the cultures with the 1F3F feeding strategy  105  18 16  A  B  C  D  E  F  14  Lactate Conc. (mM)  12 10 8 6 4 2 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time (day) Figure C.7: The lactate concentration profiles of the cultures with the 3F feeding strategy  18 16  G  H  I  J  K  L  14  Lactate Conc. (mM)  12 10 8 6 4 2 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time (day) Figure C.8: The lactate concentration profiles of the cultures with the 1F3F feeding strategy  106  Integral of Viable Cells (109 cell.day)  7 6 5  A  B  C  D  E  F  4 3 2 1 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  11  12  13  14  Time (day) Figure C.9: The integral of viable cells of the cultures with the 3F feeding strategy  7 6  IVC (109 cell.day)  5  G  H  I  J  K  L  4 3 2 1 0  0  1  2  3  4  5  6  7  8  9  10  Time (day) Figure C.10: The integral of viable cells of the cultures with the 1F3F feeding strategy  107  Cell Specific Productivity (pg/cell/day)  12 10  9.54 8.37  8.08  8.30  8 6.45  6.18  6  5.02  5.44  5.22  J  K  5.64  4.54  3.73  4 2 0  A  B  C  D  E  F  G  H  I  L  Figure C.11: The cell specific productivity of the different fed-batch cultures (Table 3.2)  108  Appendix D. Supplementary Data for Sections 3.2.2 and 3.2.3 e alphabetic codenames of culture are introduced in Table 3.9. 100 90  Viability (%)  80 70  A  B  D  I  C  60 50 40 30 20 10 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  14  15  12  13  14  15  Figure D.1: Viability of the cultures A, B, C, D and the centre point (culture I)  100 90 E H  80 Viability (%)  70  F I  G  60 50 40 30 20 10 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  Figure D.2: Viability of the cultures E, F, G, H and the centre point (culture I)  109  100 90 80  Viability (%)  70 60  I(2)  I(3)  I(4)  H(1)  50 40 30 20 10 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  14  15  13  14  15  Figure D.3: Viability of the cultures I2, I3, I4 and Hi (the augmented points in Table 3.9)  4  A  B  D  I  C  Integral of Viable Cells (109 cell.day)  3.5 3 2.5 2 1.5 1 0.5 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  Figure D.4: The integral of viable cells of the cultures A, B, C, D and the centre point (culture I)  110  4  E  F  H  I  G  Integral of Viable Cells (109 cell.day)  3.5 3  2.5 2  1.5 1  0.5 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  14  15  14  15  Figure D.5: The integral of viable cells of the cultures E, F, G, H and the centre point (culture I)  Integral of Viable Cells (109 cell.day)  3.5 3 2.5  I(2)  I(3)  I(4)  H(1)  2 1.5 1 0.5 0  0  1  2  3  4  5  6  7 8 9 Time (day)  10  11  12  13  Figure D.6: The integral of viable cells of the cultures I2, I3, I4 and Hi (the augmented points in Table 3.9)  111  Appendix E. Supplementary Data for Section 4.1 100 90 80  Viability (%)  70 60 50 40 30  FB  20  FB + 3-MA  10 0  0  2  4  6  8 Time (day)  10  12  14  16  14  16  Figure E.1: Viability of the CHOm44 fed-batch cultures with and without 3-MA  Cell Specific Producitivty (pg/cell.day)  12 10 8 6 4 FB  FB + 3-MA  2 0  0  2  4  6  8  10  12  Time (day) Figure E.2: The cell specific productivity of the CHOm44 fed-batch cultures with and without 3-MA  112  Appendix F. Supplementary Data for Section 4.2.1  Cell Specific Productivity (pg/cell.day)  80 70  No 3-MA  60  2.5 mM 3-MA  50  5 mM 3-MA  40 30 20 10 0  -10  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time (day)  Cell Specific Productivity (pg/cell.day)  Figure F.1: Cell specific productivity for CHOm56 fed-batch cultures of CHOm56 treated with 2.5 and 5 mM of 3-MA compared to the 3-MA-free control culture  60 No 3-MA  50  7.5 mM 3-MA 10 mM 3-MA  40  15 mM 3-MA  30 20 10 0 -10  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  -20 Time (day) Figure F.2: Cell specific productivity of CHOm56 fed-batch cultures of CHOm56 treated with 7.5, 10 and 15 mM of 3-MA compared to the 3-MA-free control culture  113  700  No 3-MA  Lysosomal Content (RFU)  600  2.5 mM 3-MA  500  5 mM 3-MA  400 300 200 100 0  0  1  2  3  4  5  6  7 8 Time (day)  9  10  11  12  13  14  Figure F.3: Lysosomal content of CHOm56 fed-batch cultures of CHOm56 treated with 2.5 and 5 mM of 3MA compared to the 3-MA-free control culture  600 No 3-MA Lysosomal Content (RFU)  500  7.5 mM 3-MA 10 mM 3-MA  400  15 mM 3-MA  300 200 100 0  0  1  2  3  4  5  6  7 8 Time (day)  9  10  11  12  13  14  Figure F.4: Lysosomal content of CHOm56 fed-batch cultures of CHOm56 treated with 7.5, 10 and 15 mM of 3-MA compared to the 3-MA-free control culture  114  Appendix G. Supplementary Data for Section 4.3 100 90  Viabilty (%)  80 70 60 50  No 3-MA  40 5 mM Day 3  30 20  5 mM Day 5  10 0  0  1  2  3  4  5  6 7 Time (day)  8  9  10  11  12  13  14  12  13  Cell Specific Productivity (pg/cell.day)  Figure G.1: Viability of CHOm44 cultures treated with 3-MA on different dates  25  No 3-MA  20  5 mM Day 3  15  5 mM Day 5  10 5 0  0  1  2  3  4  5  6 7 Time (day)  8  9  10  11  Figure G.2: Cell specific productivity CHOm44 cultures treated with 3-MA on different dates  115  Appendix H. Supplementary Data for Section 4.4  100 90 80  Viability (%)  70 60  C44-Osm(325)  50  C44-Osm(400)  40  C44-Osm(500)  30 20 10 0  0  1  2  3  4  5  4  5  Time (day) Figure H.1: Viability of CHOm44 batch cultures under different osmolalities  100 90 80  Viability (%)  70 60  C56-Osm(325)  50  C56-Osm(400)  40  C56-Osm(500)  30 20 10 0  0  1  2  3 Time (day)  Figure H.2: Viability of CHOm56 batch cultures under different osmolalities  116  Cell Specific Productivity (pg/cell.day)  20  C44-Osm(325)  18  C44-Osm(400)  16  C44-Osm(500)  14 12 10 8 6 4 2 0  0  1  2  3 Time (day)  4  5  Figure H.3: Cell specific productivity of CHOm44 batch cultures under different osmolalities  Cell Specific Productivity (pg/cell.day)  20  C56-Osm(325)  18  C56-Osm(400)  16  C56-Osm(500)  14 12 10 8 6 4 2 0  0  1  2  3 Time (day)  4  5  Figure H.4: Cell specific productivity of CHOm56 batch cultures under different osmolalities  117  Appendix I. Supplementary Data for Section 4.5 I.1  Results from the First Replicate  Viable Cell Conc. (106 cell/mL)  25  C44-Osm(Lo)-3MA(0) C44-Osm(Hi)-3MA(0)  C44-Osm(Lo)-3MA(5) C44-Osm(Hi)-3MA(5)  C44-Osm(Lo)-3MA(10) C44-Osm(Hi)-3MA(10)  20 15 10 5 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.1: Viable cell concentrations for the CHOm44 cell cultures with different osmolality regimes and different 3-MA concentrations  1800 C44-Osm(Lo)-3MA(0)  1600  C44-Osm(Lo)-3MA(5)  MAb Conc. (mg/L)  1400  C44-Osm(Lo)-3MA(10)  1200  C44-Osm(Hi)-3MA(0)  1000  C44-Osm(Hi)-3MA(5)  800  C44-Osm(Hi)-3MA(10)  600 400 200 0  0  1  2  3  4  5  6  7 8 Time(Day)  9  10  11  12  13  14  Figure I.2: MAb concentrations for the CHOm44 cell cultures with different osmolality regimes and different 3-MA concentrations  118  Viable Cell Conc. (106 cell/mL)  25  20  C56-Osm(Lo)-3MA(0)  C56-Osm(Lo)-3MA(5)  C56-Osm(Lo)-3MA(10)  C56-Osm(Hi)-3MA(0)  C56-Osm(Hi)-3MA(5)  C56-Osm(Hi)-3MA(10)  15  10  5  0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.3: Viable cell concentrations for the CHOm56 cell cultures with different osmolality regimes and different 3-MA concentrations  1800 C56-Osm(Lo)-3MA(0)  1600  C56-Osm(Lo)-3MA(5)  MAb Conc. (mg/L)  1400  C56-Osm(Lo)-3MA(10)  1200  C56-Osm(Hi)-3MA(0)  1000  C56-Osm(Hi)-3MA(5)  800  C56-Osm(Hi)-3MA(10)  600 400 200 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.4: MAb concentrations for the CHOm56 cell cultures with different osmolality regimes and different 3-MA concentrations  119  1000 C44-Osm(Hi)-3MA(0)  900  C44-Osm(Hi)-3MA(5)  800  Viable Cell Numer (106 cell)  C44-Osm(Hi)-3MA(10) C44-Osm(Lo)-3MA(0)  700  C44-Osm(Lo)-3MA(5)  600  C44-Osm(Lo)-3MA(10)  500 400 300 200 100 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.5: Total number of cells for CHOm44 cultures with different osmolality regimes and 3-MA concentrations  1000 C56-Osm(Lo)-3MA(0)  Viable Cell Numer (106 cel)  900  C56-Osm(Lo)-3MA(5)  800  C56-Osm(Lo)-3MA(10)  700  C56-Osm(Hi)-3MA(0)  600  C56-Osm(Hi)-3MA(5)  500  C56-Osm(Hi)-3MA(10)  400 300 200 100 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.6: Total number of cells for CHOm56 cultures with different osmolality regimes and 3-MA concentrations  120  12 Cell Specific Productivity (pg/cell.day)  No 3-MA 5 mM 3-MA  10  10 mM 3-MA  8 6 4 2 0  C44-Osm(Lo)  C44-Osm(Hi)  C56-Osm(Lo)  C56-Osm(Hi)  Figure I.7: Average cell specific productivity in the duration after addition of 3-MA of different CHO-MAb clones with high and low osmolalities  60 No 3-MA 50  5 mM 3-MA  Total MAb (mg)  10 mM 3-MA 40 30 20 10 0  C44-Osm(Lo)  C44-Osm(Hi)  C56-Osm(Lo)  C56-Osm(Hi)  Figure I.8: Final weight of MAb produced in fed-batch cultures of different CHO-MAb clones with high and low osmolalities  121  I.2  Results from the Second Replicate  Viable Cell Conc. (106 cell/mL)  25  20  C44-Osm(Lo)-3MA(0)-R  C44-Osm(Lo)-3MA(5)-R  C44-Osm(Lo)-3MA(10)-R  C44-Osm(Hi)-3MA(0)-R  C44-Osm(Hi)-3MA(5)-R  C44-Osm(Hi)-3MA(10)-R  15  10  5  0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.9: Viable cell concentrations for the CHOm44 cell replicate cultures with different osmolality regimes and different 3-MA concentrations  1800  C44-Osm(Lo)-3MA(0)-R  1600  C44-Osm(Lo)-3MA(5)-R  MAb Conc. (mg/L)  1400  C44-Osm(Lo)-3MA(10)-R  1200  C44-Osm(Hi)-3MA(0)-R  1000  C44-Osm(Hi)-3MA(5)-R  800  C44-Osm(Hi)-3MA(10)-R  600 400 200 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.10: MAb concentrations for the CHOm44 replicate cell cultures with different osmolality regimes and different 3-MA concentrations  122  25  C56-Osm(Lo)-3MA(0)-R  Viable Cell Conc. (106 cell/mL)  C56-Osm(Lo)-3MA(5)-R C56-Osm(Lo)-3MA(10)-R  20  C56-Osm(Hi)-3MA(0)-R C56-Osm(Hi)-3MA(5)-R  15  C56-Osm(Hi)-3MA(10)-R  10  5  0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.11: Viable cell concentrations for the CHOm56 replicate cell cultures with different osmolality regimes and different 3-MA concentrations  1800  C56-Osm(Lo)-3MA(0)-R  MAb Conc. (mg/L)  1600  C56-Osm(Lo)-3MA(5)-R  1400  C56-Osm(Lo)-3MA(10)-R  1200  C56-Osm(Hi)-3MA(0)-R  1000  C56-Osm(Hi)-3MA(5)-R C56-Osm(Hi)-3MA(10)-R  800 600 400 200 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.12: MAb concentrations for the CHOm56 cell cultures with different osmolality regimes and different 3-MA concentrations  123  1000 C44-Osm(Lo)-3MA(0)-R  Viable Cell Number (106 cell)  900  C44-Osm(Lo)-3MA(5)-R  800  C44-Osm(Lo)-3MA(10)-R  700  C44-Osm(Hi)-3MA(0)-R  600  C44-Osm(Hi)-3MA(5)-R  500  C44-Osm(Hi)-3MA(10)-R  400 300 200 100 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.13: Total cell number for the CHOm44 replicate set of cultures with different osmolality regimes and 3-MA concentrations  1000  C56-Osm(Lo)-3MA(0)-R  Viable Cell Number (106 cell)  900  C56-Osm(Lo)-3MA(5)-R  800  C56-Osm(Lo)-3MA(10)-R  700  C56-Osm(Hi)-3MA(0)-R  600  C56-Osm(Hi)-3MA(5)-R  500  C56-Osm(Hi)-3MA(10)-R  400 300 200 100 0  0  1  2  3  4  5  6  7  8  9  10  11  12  13  14  Time(Day) Figure I.14: Total cell number for the CHOm56 replicate set of cultures with different osmolality regimes and 3-MA concentrations  124  Cell Specific Productivity (pg/cell.day)  12 No 3-MA 10  5 mM 3-MA 10 mM 3-MA  8 6 4 2 0  C44-Osm(Lo)-R  C44-Osm(Hi)-R  C56-Osm(Lo)-R  C56-Osm(Hi)-R  Figure I.15: Average cell specific productivity in the duration after addition of 3-MA for the replicate cultures of different CHO-MAb clones with high and low osmolalities  60  No 3MA 5 mM 3-MA  50  10 mM 3-MA  Total MAb (mg)  40 30 20 10 0  C44-Osm(Lo)-R  C44-Osm(Hi)-R  C56-Osm(Lo)-R  C56-Osm(Hi)-R  Figure I.16: Final weight of MAb produced in replicate cultures of different CHO-MAb clones with high and low osmolalities  125  Appendix J. Glycan Analysis Raw Data (Section 4.6) A subset of samples from the second replicate of experiments described in Section 4.6 (data in Appendix J.2) was selected for glycan analysis. e selected samples are taken on day 12 from the highlighted cultures in Table J.1. Table J.1: Experimental design table for testing autophagy inhibition and hypertonicity effects on MAb production simultaneously. The samples on which glycan analysis is performed are highlighted  Culture  Cell Line  A B C D E F G H I J K L  CHOm44 CHOm44 CHOm44 CHOm44 CHOm44 CHOm44 CHOm56 CHOm56 CHOm56 CHOm56 CHOm56 CHOm56  Osmolality Regime [EN/CYD Conc.] High [5X] High [5X] High [5X] Low [1X] Low [1X] Low [1X] High [5X] High [5X] High [5X] Low [1X] Low [1X] Low [1X]  3MA Conc. (mM) 0 5 10 0 5 10 0 5 10 0 5 10  Table J.2: The monosaccharide constituents of glycan structures and their graphical representations used in this thesis  Monosaccharide  Graphical Representation  Mannose GlcNAc Fucose Galactose Sialic Acid  126  Table J.3: The glycan structures identified and their graphical representations  Structure M1B  Graphical Representation  M2 A1 A2 (G0) F6A1 F6A2 (G0F) A1G1  A2B  M5 (Man5) F6A1G1 F6A2[6]G1 (G1F) F6A2[3]G1 (G1F) F(6)A2G2 (G2F) A2G2S1 127  J.1  Glycan Analysis of Antibody Heavy Chains  Figure J.1: Overlay of HPLC elution profiles and peak identification for the heavy chains of the MAb samples from A, B, D and E cultures. All peaks with more than 2% of area are identified.  128  Figure J.2: Overlay of HPLC elution profiles and peak identification for the heavy chains of the MAb samples from G, H, J and K cultures. All peaks with more than 2% of area are identified.  129  Table J.4: Peak identification based on the GU values for the heavy chains of MAb samples and comparison of individual area percentage.  Peak Number 1 2 3 4 5 6  GU Value Structure 4.94 A1 5.43 F(6)A1/A2 5.85 F(6)A2/A1G1/A2B 6.2 M5 6.6 F(6)A2[6]G1 6.74 F(6)A2[3]G1  Run A 5.12 30.25 61.18 3.45  Run B 2.38 25.42 66.34 2.07 4.19  Run D 23.11 65.39 2.72 6.09 2.69  Run E 3.63 29.91 58.05 2.96 5.45  Run G 3.47 25.84 59.86 2.78 5.37 2.33  Run H 6.79 36.13 46.11 4.01 4.87 2.09  Run J 4.23 31.9 50.51 4.75 6.43 2.19  Run K 4.35 31.86 50.44 4.74 6.42 2.19  130  J.2  Glycan Analysis of Antibody Light Chains  Figure J.3: Overlay of HPLC elution profiles for the light chains of the MAb samples from A, B, D and E cultures.  131  Figure J.4: Overlay of HPLC elution profiles for the light chains of the MAb samples from G, H, J and K cultures.  132  Table J.5: Peak identification based on the GU values for the heavy chains of MAb samples and comparison of individual area percentage.  Peak Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26  GU Value Structure 3.00 3.10 M1B 3.23 3.31 3.38 NA 3.48 M2 3.59 3.69 3.79 NA 3.85 4.17 NA 4.62 4.95 A1 5.20 5.43 F6A1/A2 5.64 5.86 F6A2/A1G1/A2B 6.19 M5 6.32 F6A1G1(?) 6.47 6.61 F(6)A2[6]G1 6.73 F(6)A2[3]G1 6.81 7.10 7.49 F(6)A2G2 7.98 A2G2S1  Run A  Run B 4.31  Run D 5.34  Run E  Run G 6.30  Run H 7.98  Run J 8.20  10.39  6.52 2.13 3.65 3.78 2.89 2.35  Run K 3.57  3.17 8.92 8.78  10.71  7.35  5.75 2.37 2.16 7.22 3  3.09 3.04 3.98 2.39  4.55  4.12  17.62  17.92 2.16 31.04 3.09  2.09 5.44  51.81 2.04  2.58 2.28  2.98 2.71 2.67 2.04 3.09 2.68  10.65 3.48 2.81  9.23  2.71 4.68  2.34  5.07 4.82 6.48 3.56 3.12  8.44  7.75  5.02  6.88  8.09  22.09  12.53  14.55 2.66 21.01 3.40  20.38  22.61 3.7 3.16  18.56 2.61 22.52 3.95 2.45  2.67 2.38  3.06 2.47  3.13 2.45  4.46 2.87 2.41  5.57 3.51 3.22  3.80 3.44 3.25  2.40 4.22  30.14 2.82  20.70 3.90 2.93  2.48 9.60 2.50 4.07 8.78 3.22 3.62 4.38 2.09 7.47 2.13 18.69 2.50 11.25 3.94 2.44  2.92 2.19 3.97 3.49 2.61  4.23 3.05 2.09  5.05 2.23 133  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073489/manifest

Comment

Related Items