Modeling based process development of fed-batch bioprocesses [Elektronische Ressource] : L-Valine production by Corynebacterium glutamicum / vorgelegt von Michael Alexander Brik Ternbach

Informations

• Publié par: rheinisch-westfalischen_technischen_hochschule_-rwth-_aachen
• Publié le: 01 janvier 2005
• Nombre de lectures: 9
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

Modeling Based Process Development of
Fed-Batch Bioprocesses:
L-Valine Production by
Corynebacterium glutamicum
Von der Fakulta¨t fu¨r Maschinenwesen der Rheinisch-Westfa¨lischen Technischen
Hochschule Aachen zur Erlangung des akademischen Grades eines Doktors der
Ingenieurwissenschaften genemigte Dissertation
vorgelegt von
Michael Alexander Brik Ternbach
aus
Maassluis, Niederlande
Berichter: Universita¨tsprofessor Dr.-Ing. Jochen Bu¨chs
Universita¨tsprofessor Dr.rer.nat. Christian Wandrey, Universit¨at Bonn
Tag der mu¨ndlichen Pru¨fung: den 13. April 2005
Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online verfu¨gbar.’Therefore, mathematical modeling does not make sense
without defining, before making the model, what its use is
and what problem it is intended to help to solve.’
James E. Bailey (1998)
’The ability to ask the right question is more than half the
battle of finding the answer.’
Thomas J. Watson
’You will only see it when you understand it.’
(Translated freely from the Dutch: ’Je gaat het pas zien als je het
door hebt.’)
Johan CruijffContents
1. Introduction 5
2. Theory 7
2.1. Biotechnological Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.1. Bioreactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.2. Modes of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2. Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.1. Mechanistic and Black-Box Models . . . . . . . . . . . . . . . . . . 9
2.2.2. Macroscopic Modeling . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2.3. Structured and Unstructured Models . . . . . . . . . . . . . . . . . 10
2.3. A Model of a Fed-Batch Process . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.1. Mass Balance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.3.2. Kinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2.4. Fitting Models to Experimental Data. . . . . . . . . . . . . . . . . . . . . 17
2.4.1. Parameter Estimation . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.4.2. Parameter Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.4.3. Accuracy of the Fitted Model . . . . . . . . . . . . . . . . . . . . . 19
2.4.4. Jacobian in dynamic systems . . . . . . . . . . . . . . . . . . . . . 21
2.5. Experimental Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.5.1. Experimental Design for Model Discrimination . . . . . . . . . . . 26
2.5.2. Experimental Design for Parameter Estimation . . . . . . . . . . . 36
2.5.3. Model Discrimination and Parameter Estimation . . . . . . . . . . 39
2.6. Optimization of Bioprocesses . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.7. L-valine Production in Corynebacterium glutamicum . . . . . . . . . . . . 41
3. Dynamic Modeling Framework 45
3.1. The Modeling Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.2. Parameter Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.3. Designing Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.3.1. Model Discriminating Design . . . . . . . . . . . . . . . . . . . . . 48
3.3.2. Design for Parameter Estimation . . . . . . . . . . . . . . . . . . . 49
3.3.3. Design an Optimized Production Process . . . . . . . . . . . . . . 50
1Contents
3.4. Application: L-Valine Production Process Development . . . . . . . . . . 50
3.4.1. Designed Experiments . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.4.2. Preparation of Measured Data . . . . . . . . . . . . . . . . . . . . 54
3.4.3. Calculation of other Characteristic Values . . . . . . . . . . . . . . 56
4. Simulative Comparison of Model Discriminating Design Criteria 61
4.1. Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.2. First Case: Fermentation . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.2.1. Prior Situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.2.2. Designed Experiments . . . . . . . . . . . . . . . . . . . . . . . . . 65
4.2.3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.3. Second Case: Catalytic Conversion . . . . . . . . . . . . . . . . . . . . . . 74
4.3.1. Prior Situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
4.3.2. Experimental Designs . . . . . . . . . . . . . . . . . . . . . . . . . 76
4.3.3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4.4. Discussion and Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5. L-Valine Production Process Development 87
5.1. Material and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.1.1. Organism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.1.2. Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.1.3. Fermentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
5.1.4. Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
5.2. Modeling Based Process Development . . . . . . . . . . . . . . . . . . . . 101
5.2.1. Preliminary Experiments . . . . . . . . . . . . . . . . . . . . . . . 101
5.2.2. Initialization: Batch Data . . . . . . . . . . . . . . . . . . . . . . . 102
5.2.3. Model Discriminating Experiment I . . . . . . . . . . . . . . . . . 104
5.2.4. Intuitiv Discrimination between L-Isoleucine and Pantothenic Acid 110
5.2.5. Model Discriminating Experiment II . . . . . . . . . . . . . . . . . 113
5.2.6. Optimized Process . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5.3. Biological Information from Modeling Based Process Development . . . . 121
5.3.1. Modeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
5.3.2. By-products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
5.4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Appendici 131
A. Nomenclature 133
2Contents
B. Ratio between OD and Dry Weight. 139600
C. The Influence of the Oxygen Supply. 141
D. Dilution for Measurement of Amino Acids by HPLC. 151
E. Models used in the First Model Discriminating Experiment. 153
F. Models used in the Second Model Discriminating Experiment. 161
G. Alternative Off-line Optimized Production Experiments 173
G.1. Total Volumetric Productivity . . . . . . . . . . . . . . . . . . . . . . . . . 173
G.2. Overall Yield of Product on Substrate . . . . . . . . . . . . . . . . . . . . 174
Summary 177
Zusammenfassung 179
Acknowledgements 181
Bibliography 183
3Contents
41. Introduction
The OECD (the Organisation of Economic Co-operation and Development) defines
biotechnology as ”the application of scientific and engineering principles to the process-
ing of materials by biological agents.” Biotechnology is actually already thousands of
years old, when we realize that for instance already ancient Egyptians used yeast for
making bear and bread. A major difference between modern biotechnology and those
early applications lies in the possibility to precisely identify and change the genes which
govern the desired traits.
Nowadays, modern biotechnolgy is applied in many different fields. Biotechnology
plays an important role in pharmaceutical industry, the so called ’red’ biotechnology
(Moore, 2001), for for instance diagnostic kits, production of vaccines, antibodies and
othermedicines. Infoodandagriculture, biotechnolgyisusedforinstancetobreedhigh-
yielding crops and reduce the need for insecticides or herbicides (’green’ biotechnology,
(Enriquez,2001)). Alsoinseveralotherindustrialprocesses,enzymesormicro-organisms
are used for instance for the production of intermediary productsfor chemical industries
(’white’ biotechnology, (Frazzetto, 2003)).
Since several decades, modern biotechnology is a major technology which increased
substantiallyespeciallyinthelastdecade. IntheErnst&YoungReport2003, morethan
$41 billion revenues of biotechnolgical companies were mentioned for 2002. Moreover,
these figures do not include large pharmaceutical or large agribusiness companies for
which biotechnology forms a part of the business.
Thecompetitiveness ofbiotechnological industriesdependsstronglyonknowledgeand
innovations (Pownall,2000). Inmanycases, suchas theintroduction ofnewpharmaceu-
tical products, it is especially important to keep the time to market as short as possible.
So it is important to have fast process development procedures.
On the other hand, it can be very important to keep production costs as low as
possible and guarantee a good product quality. This can benefit from optimization
of the production organism or the process conditions. With modern biotechnological
techniques, new organisms or catalysts are developed fast. Due to the strong interaction
between the used organism and the optimal process conditions, this even increases the
need for proper and fast process development.
It iswell accepted that processmodelscan bevery usefulforprocessoptimization and
-control. One of the main reason why this is not used very extensively in industry, is
probably the fact that the development of the model itself is often rather difficult and
time-consumi