Itaconate production by Ustilago maydis - darwin.bth.rwth...

"Itaconate production by Ustilago maydis;

the influence of genes and cultivation conditions"

Von der Fakultät für Mathematik, Informatik und

Naturwissenschaften der RWTH Aachen University

zur Erlangung des akademisches Grades einer

Doktorin der Naturwissenschaften

genehmigte Dissertation vorgelegt von

M.Sc.

Monika Maasem Panáková

aus

Šumperk - Tschechische Republik

Berichter: Universitätsprofessor Dr. em. Ulrich Klinner

Universitätsprofessor Dr. Michael Bölker

Universitätsprofessor Dr. Jan Schirawski

Tag der mündlichen Prüfung: 10.10.2013

Diese Dissertation ist auf den Internetseiten der Hochschulbibliothek online

verfügbar.

Summary

Due to the increasing CO2 emissions and the necessity for substitutions to the petrochemical

fuels, the interest in biofuels originating from alternative renewable resources is on the rise.

Itaconate represents a newly identified platform molecule, which can be used for the produc-

tion of various polymers and can also act as an intermediate for biofuel production. It can

be produced by microorganisms from hydrolyzed biomass, without these resources being

in competition with common sources of human diet. Ustilago maydis MB215 represents

an alternative itaconate producer; it exhibits an unicellular yeast-like form of growth and

possesses the ability to produce itaconate naturally. However, its itaconate yields do not

compare to those from the currently most favored producer, Aspergillus terreus. Therefore,

the aim of this work was to investigate and understand the itaconate synthesis pathway in

U . maydis and determine the factors influencing the production process in order to enhance

the efficiency of the itaconate production.

Several genes putatively involved in the itaconate synthesis pathway were determined and

deleted and/or overexpressed in order to elucidate the biosynthetic pathway. Deletion of

the UM06344 gene, presumably encoding a cis-aconitate decarboxylase, did not decrease

the itaconate production. Although overexpression of this gene might indirectly improve

the itaconate production, it is not directly involved. Nevertheless, three mutants with sig-

nificantly increased itaconate generation were obtained during the phenotypical screening

of the overexpression mutants. These mutants were subjected to further characterization

under selected cultivation conditions. On the other hand, disruption of the UM11778 gene,

encoding an enzyme similar to 3-methylitaconate-∆-isomerase (Mii), drastically decreased

itaconate production. The Mii enzyme can catalyze the isomerization between citraconate

and itaconate, and based on this finding and physiological characterization of itaconate

production by U . maydis, possible itaconate synthesis pathways were proposed.

Biotechnological production of chemicals is strongly influenced by the cultivation conditions.

In order to establish an optimal production process, factors having an impact on the cells and

consequently on the itaconate yields were determined and studied. The most influencing

factor proved to be the oxygen supply. Cultures suffering from oxygen limitation failed to

efficiently synthesize itaconate, whereas more aerated cultures were capable of significantly

enhanced itaconate production. The necessity for sufficient aeration originates from the

indispensability of cofactor regeneration during itaconic acid production. The respiratory

chain represents the first option after the cessation of the biomass formation, which is the

main route for the cofactor regeneration. However, during the oxygen limitation, these

cofactors are regenerated via by-product synthesis. The two most distinct by-products are

glycolipids and malate. A mutant with blocked glycolipid production exhibited increased

itaconate production with regard to the wt strain.

The initial glucose concentration also represents an important factor and therefore vari-

ous glucose amounts and their impact on the cells as well as on the itaconate yields were

studied. Even though calculations revealed that the maximum theoretical itaconate yield

increases with increasing initial glucose concentrations, the cultivation experiments showed

that the cells undergo a substantial osmotic stress while being cultivated in high-glucose-

concentration solutions. Understanding these relations led to the determination of the

optimal initial glucose concentration for the itaconate production in U . maydis, which under

the conditions tested is 65 g l−1.

Not only the depletion of the nitrogen (N) functions as a trigger for the itaconate synthesis but

its concentration influences the itaconate yield and production rate; in theory, more N means

high rates but low yields and vice versa. However, the biomass formation increases with the

increasing N concentration and soon the culture reaches oxygen limitation and the itaconate

production decreases. Taken all these facts into consideration, the optimal N concentration

for the itaconate production by U . maydis was determined to be 0.8 g l−1 NH4Cl.

In order to decrease the overall fermentation costs, alternative carbon sources for the it-

aconate production were tested. U . maydis is capable of utilization of xylose, arabinose and

sucrose as substrates for itaconate production. However, itaconate production from these

sugars was very low, and further optimization is required.

Based on the findings presented within this work, U . maydis proved to be a suitable alternat-

ive itaconate producer. Although its itaconate yields do not yet reach those from A. terreus,

there are several advantages U . maydis possesses in comparison to the Aspergillus and with

suggested further improvements the itaconate yields will improve.

Zusammenfassung

Aufgrund der Zunahme an CO2-Emissionen und der Notwendigkeit einen Ersatz für petro-

chemische Treibstoffe zu finden, zeigt sich ein steigendes Interesse an Biokraftstoffen aus

regenerativen Ausgangsstoffen.

Hierzu gehört Itaconat als ein neu identifiziertes Plattformmolekül, das sowohl für die Produk-

tion verschiedener Polymere als auch als Intermediär für die Biokraftstoff-Produktion ver-

wendet werden kann. Mikroorganismen können Itaconat aus hydrolysierter Biomasse bilden,

so dass die verwendeten Ressourcen nicht mit menschlichen Nahrungsquellen konkurrieren.

Bei Ustilago maydis MB215 handelt es sich um einen alternativen Itaconat-Produzenten, der

ein einzelliges, hefe-ähnliches Wachstum aufweist und die Fähigkeit besitzt Itaconat auf

natürliche Weise zu generieren. Die Ausbeute ist jedoch nicht mit der des zurzeit favorisier-

ten Itaconat-Produzenten Aspergillus terreus vergleichbar. Deshalb lag das Ziel dieser Arbeit

in der Untersuchung und dem Verständnis des Itaconat- Syntheseweges in U . maydis und

der Bestimmung der Einflussfaktoren im Produktionsprozess, um dessen Effizienz bei der

Itaconat-Produktion zu steigern.

Einige Gene, die mutmaßlich in den Syntheseweg von Itaconat eingebunden sind, kon-

nten bestimmt, deletiert und/oder überexpressioniert werden, um den tatsächlichen Bio-

syntheseweg aufzudecken. Das Entfernen des UM06344-Gens, das wahrscheinlich eine

cis-Aconitat-Decarboxylase kodiert, verringerte den Itaconat-Ausstoß nicht. Auch wenn

die Überexpression dieses Gens die Itaconat möglicherweise indirekt verbessert, ist es je-

doch nicht direkt beteiligt. Bei der phänotypischen Überprüfung der überexpressionier-

ten Mutanten wurden jedoch drei Mutanten gewonnen werden, die die Itaconat-Ausbeute

signifikant steigerten. Diese Mutanten wurden daraufhin unter ausgewählten Kultivier-

ungsbedingungen genauer charakterisiert. Die Spaltung des Gens UM11778, das ein 3-

Methylitaconat-∆-Isomerase (Mii) ähnliches Enzym kodiert, führte zu einer drastischen

Abnahme der Itaconat-Produktion. Das Mii Enzym kann die Isomerisierung zwischen Cit-

raconat und Itaconat katalysieren. Auf dieser Erkenntnis und der physiologischen Charakter-

isierung der Itaconat-Produktion von U . maydis aufbauend, werden mögliche Synthesewege

der Itaconat aufgezeigt.

Die biotechnologische Produktion der Chemikalien hängt stark von den genutzten Kultivier-

ungsbedingungen ab. Um einen optimalen Produktionsprozess zu gewährleisten, wurden

die Einflussfaktoren auf die Zellen und folglich auf die Itaconat-Ausbeute ermittelt und näher

untersucht. Als Faktor mit dem meisten Einfluss stellte sich hier die Sauerstoffzufuhr heraus.

Kulturen, die nur eine limitierte Sauerstoffzufuhr erhielten, konnten Itaconat nicht effizient

synthetisieren, wohingegen besser mit Sauerstoff versorgte Kulturen eine deutliche Produk-

tionssteigerung aufwiesen. Die Notwendigkeit ausreichender Sauerstoffzufuhr liegt in der

unverzichtbaren Kofaktor-Regeneration während der Itaconsäure-Produktion begründet.

Die Atmungskette ist die erste Regenerationsmöglichkeit nach Einstellung der Biomasse-

bildung, die den Hauptpfad der Kofaktor-Regeneration darstellt. Es ist anzumerken, dass

die Kofaktoren bei verringerter Sauerstoffzufuhr über Nebenprodukt-Synthese regeneriert

werden. Die beiden am häufigsten ausgeprägten Nebenprodukte sind Glykolipide und Malat.

Ein Mutant mit blockierter Glykolipid-Produktion zeigte eine gesteigerte Itaconat-Produktion

bezüglich des Wildtyp Stamms.

Die anfängliche Glukose-Konzentration stellt einen bedeutenden Faktor dar, so dass ver-

schiedene Glukosekonzentration eingestellt und die jeweiligen Auswirkungen auf die Zellen

als auch die sich jeweils ergebenden Itaconat-Ausbeuten untersucht wurden. Auch wenn die

Berechnungen ergeben, dass die maximale Itaconat-Ausbeute mit zunehmender initialer

Glukosekonzentration ansteigt, zeigen die durchgeführten Kultivierungsexperimente jedoch

dass die Zellen einem nicht zu vernachlässigbaren osmotischen Schock ausgesetzt werden,

wenn sie sich in hohen Glukosekonzentrationen befinden. Das Verständnis dieser Wirkungs-

beziehung führte zur Bestimmung der optimalen initialen Glukose-Konzentration für die

Itaconat-Produktion in U . maydis, die unter den getesteten Bedingungen einen Wert von 65

g l−1 ergab.

Die Itaconat-Ausbeute und Produktionsrate werden jedoch nicht nur durch die Erschöp-

fung des Stickstoffs (N), als Auslöser der Itaconat-Synthese, sondern auch durch die Stick-

stoffkonzentration beeinflusst. In der Theorie geht ein mehr an Stickstoff mit höheren Raten

aber geringerer Ausbeute einher und umgekehrt. Die Bildung der Biomasse nimmt hingegen

mit zunehmender Stickstoffkonzentration zu, so dass die Kultur bald darauf das eingestellte

Sauerstofflimit erreicht und sich die Itaconat-Produktion wieder verringert. Unter Berück-

sichtigung aller Fakten ergab sich eine optimale N-Konzentration von 0.8 g l−1 NH4Cl für die

Itaconat-Produktion von U . maydis.

Zur Reduktion der Fermentationskosten wurden alternative Kohlenstoffquellen für die Gewin-

nung der Itaconat getestet. U . maydis kann hierbei Xylose, Arabinose und Saccharose als

Substrate für die Itaconat-Produktion nutzen. Die Ausbeute mit diesen Zuckern war jedoch

sehr gering, so dass eine weitere Optimierung notwendig ist.

Basierend auf den Ergebnissen der vorliegenden Arbeit konnte U . maydis als passender

alternativer Itaconat-Produzent ermittelt werden. Obwohl die produzierte Menge an It-

aconat bisher nicht mit der durch A. terreus zu erreichenden Ausbeute mithalten kann, weist

U . maydis doch einige Vorteile auf, die im Vergleich mit Aspergillus und mit den vorgeschla-

genen Verbesserungen die Ausbeute noch weiter steigern werden können.

Acknowledgements

This thesis represents the final work of my Ph.D. study at the Institute of Applied Microbiology

at the RWTH Aachen University.

I would like to thank several individuals for their support and encouragement during the last

almost five years; Dr. Nicole Maaβen for her guidance throughout the practical part of the

work, Dr. Nick Wierckx for his much welcomed insights, ideas and patience during the writing

of the thesis. I would also like to thank him for the countless hours he spent proofreading

the manuscript. I would like to thank Prof. Dr. em. Ulrich Klinner, who has given me the

great opportunity to perform this challenging work. I would also like to express my gratitude

and respect to Prof. Dr. Michael Bölker for his much appreciated ideas, encouragement and

for being a member of the dissertation committee. I am also very grateful for his technical

support concerning cell strains, plasmids and primers. Moreover, I am very thankful for the

possibility he gave me to spend much valued time within his research team where I have

learnt methods I have used throughout this work. I would also like to thank Prof. Dr. Jan

Schirawski that he was so kind and agreed to be a member of the dissertation committee and

to write the report.

Last but not least, I would like to thank to Chris for his endless moral support, patience and

love, which have given me the strength to complete this thesis. A lot of thanks to Susan for

lifting up my spirits during dark times.

This work was performed as part of the Cluster of Excellence “Tailor-Made Fuels from Bio-

mass“, which is funded by the Excellence Initiative by the German federal and state govern-

ments to promote science and research at German universities.

Contents

1 Introduction 1

1.1 Biofuels for a sustainable society . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Alternative carbon sources for biofuel production . . . . . . . . . . . . . . . . . . 3

1.3 Itaconic acid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.4 Industrial bioproduction of itaconic acid . . . . . . . . . . . . . . . . . . . . . . . 6

1.5 Ustilago maydis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.6 Aim of this study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2 Materials and methods 11

2.1 Strains, plasmids and primers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.2 Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.3 Cultivation methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2.4 Analytical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2.4.1 Fermentation broth analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 18

2.4.2 Off-gas analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

2.5 Molecular biological methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

2.5.1 Genomic DNA isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

2.5.2 Plasmid DNA isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

2.5.3 Gel electrophoresis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

2.5.4 In vitro DNA methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

2.5.5 Transformation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

2.5.6 Southern blot . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

3 Results 25

3.1 Selection of the gene of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3.2 Disruption of UM06344 in U. maydis . . . . . . . . . . . . . . . . . . . . . . . . . 25

3.3 Overexpression of UM06344 in U. maydis . . . . . . . . . . . . . . . . . . . . . . . 27

3.3.1 Screening of transformants for increased itaconate production . . . . . . 28

3.3.2 Southern blot analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

3.4 Itaconate production kinetics under various cultivation conditions . . . . . . . 33

3.4.1 Influence of glucose concentration . . . . . . . . . . . . . . . . . . . . . . 34

3.4.2 Influence of culture aeration . . . . . . . . . . . . . . . . . . . . . . . . . . 38

3.4.3 Influence of nitrogen concentration . . . . . . . . . . . . . . . . . . . . . . 42

3.4.4 Selected optimal conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

3.5 Itaconate production during fermenter batch cultivation . . . . . . . . . . . . . 47

3.6 Biosurfactant production and its influence on the itaconate production . . . . 49

3.7 Alternative carbon sources for the itaconate production . . . . . . . . . . . . . . 51

3.8 Deletion of UM11778 in U. maydis . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

4 Discussion 57

4.1 Itaconate synthesis pathway in U. maydis . . . . . . . . . . . . . . . . . . . . . . 57

4.2 Factors influencing the itaconate production . . . . . . . . . . . . . . . . . . . . 63

4.2.1 Nitrogen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

4.2.2 Glucose . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

4.2.3 Oxygen supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

4.2.4 Growth-limiting factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

4.3 Alternative carbon sources for itaconate production . . . . . . . . . . . . . . . . 70

4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

4.5 Further recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

5 Supplementary data 75

List of Figures III

List of Tables V

Bibliography VII

CHAPTER 1

Introduction

1.1 Biofuels for a sustainable society

Increasing carbon dioxide emissions, a rising energy demand and limited availability of fossil

energy resources are leading to an ever growing necessity for a more sustainable society.

Especially transportation fuels, which are 98 % derived from petroleum, are highly depend-

ent on fossil resources. The use of these resources leads to high net CO2 emissions, and

the number of (often politically unstable) oil producing countries is small, leading to large

fluctuations in supply and price. Biofuels are considered a promising sustainable alternative

to petrochemical fuels. Their main advantage is that they are produced from biomass, which

means there is little net CO2 emission, and the biomass can be produced virtually everywhere

[1].

Up to now, research into biological transportation fuel substitutes has focused mainly on

ethanol. Saccharomyces cerevisiae is by far the most commonly used organism for ethanol

production, and this organism is used at a large scale for fuel production today. There are also

other ethanol fermenting organisms, such as Zymomonas mobilis [2]and Escherichia coli [3],

which have been engineered to produce ethanol at a high rate, titer and yield.

Alternatively, some molecules present in gasoline can also be produced by microorganisms,

both natural and engineered. The advantages of these molecules with regard to ethanol

are higher energy content, lower hygroscopicity and corrosivity. For instance, straight- and

branched-chain alkanes and alkenes can be produced through isoprenoid pathways, and

various alcohols and esters can be derived from fatty acid biosynthesis. Moreover, itaconic

acid, the newly identified key molecule for biofuel production [4], can be produced from

hydrolyzed biomass using engineered microorganisms (Figure 1.1).

1

Chapter 1: Introduction

Arabinose

Ribulose

R5P

Xylose

Xylulose X5P 6PG G6P

Glucose

G1P G´1P

Galactose

F6P

FBP

G3P

DHAP

M6P

Mannose

E4P

S7P

PEP

Pyruvate Formate Hydrogen

CO2

Acetaldehyd

Ethanol

CO2

Acetyl-CoA

CIT

ICT

OGA

SUC SUC-CoA

Glyoxylate

FUM

MAL

OAA

Propyonyl-CoA Propanol

Butyryl-CoA Butanol

Valeryl-CoA Isopentanol

Even-chain fatty acids

Alcohols

Odd-chain fatty acids

Even-chain fatty alcohol

Various esters

Odd-chain fatty alcohol

Even-chain alkane

Odd-chain alkane

IPP DMAPP

GPP

FPP

GGPP Geranylgeraniol

Diterpene

Farnesol

Sesquiterpene

Geraniol

Monoterpene

Isopentenol Isopentanol

Acetoacetyl-CoA

Cis-Ac

CO2

ITACONIC ACID

Figure 1.1: Overview of potential fuel precursors derived from various products of

hydrolyzed lignocellulose. Blue box: isoprenoid pathways and isoprenoid-

derived molecules. Yellow box: fatty acid pathways and fatty acid-derived

molecules. Green box: products of hydrolyzed lignocellulose. In green font

are shown short-chain alcohols. Abbreviations: 6PG, 6-phosphogluconate;

Cis-Ac, cis-aconitate; CIT, citrate; DHAP, dihydroxyacetone phosphate;

DMAPP, dimethylallyl pyrophosphate; E4P, erythrose-4-phosphate; F6P,

fructose-6-phosphate; FBP, fructose-1,6-bisphosphate; FPP, farnesyl pyro-

phosphate; FUM, fumarate; G’1P, galactose-1-phosphate; G1P, glucose-1-

phosphate; G3P, glyceraldehyde-3-phosphate; G6P, glucose-6-phosphate;

GGPP, geranylgeranyl pyrophosphate; GPP, geranyl pyrophosphate; ICT,

isocitrate; IPP, isopentenyl pyrophosphate; M6P, mannose-6-phosphate;

MAL, malate; OAA, oxaloacetate; OGA, 2-oxoglutarate; PEP, phosphoen-

olpyruvate; R5P, ribose-5-phosphate; S7P, sedoheptulose-7-phosphate;

SUC, succinate; SUC-CoA, succinyl coenzyme A; X5P, xylulose-5-phosphate

(figure adapted from [5]).

2

1.2 Alternative carbon sources for biofuel production

High-yield production of these molecules using microbial systems could function as a re-

placement of some or all of the supply currently derived from petroleum [5].

Currently, there are two main biomass substrates for biofuel production. Cane juice is favored

in Brazil, while starch-crops like corn are used in the USA [6]. Corn is used as human food,

and the production of sugar cane competes with food production for arable land. Using some

of human’s main diet sources for ethanol production has raised an intense discussion about

its effect on food production and costs. These concerns were the base and the driving force

in development of new ways of biofuels production based on biomass that does not compete

with human food production [5].

1.2 Alternative carbon sources for biofuel production

The next generation of biofuels will most likely be developed from abundantly present ligno-

cellulosic biomass due to the political pressure and physical limitations on tillable land [5].

Moreover lignocellulosic biomass such as straw or wood is not used as human nutrient.

Therefore there is no competition between human food and biofuel production.

Lignocellulose consists of cellulose and hemicellulose, which are tightly bound to lignin [7].

Cellulose is the major structural component of cell wall of plants. This polysaccharide is

organized in microfibrils stabilized by hydrogen bonds [8] and consists of D-glucose units

linked together via β-1,4 glycosidic bonds into linear chains. The length of these chains varies

with the origin and treatment of the biomass [9]. The cellulose chains bind together through

hydrogen bonds to form microfibrils, which are in turn interconnected by hemicelluloses

and polymers of various other molecules (sugars, pectin). This structure is then covered by

lignin. Hemicellulose consists mainly of pentoses (xylose, arabinose) and hexoses (glucose,

mannose, galactose). It has a lower molecular weight than cellulose, and its polymer chains

are considerably more branched [8]. The composition of hemicelluloses varies depending on

the biomass source [7]. Lignin is an amorphous polymer which functions mainly as physical

support of the plant cell wall, giving it its firmness. It also functions as a barrier against

microbial attack [8].

Pretreatment of lignocellulose is a necessary step, its main function being to make the cellu-

lose and hemicelluloses available for cellulolytic enzymes. Several kinds of pretreatment are

used. These can be divided into physical (e.g. milling, irradiation), physico-chemical and

chemical (e.g. several types of explosion, usage of oxidizing agents) or biological approaches

(treatment with enzymes of fungi and Actinomycetes) [10]. Most of the techniques require

high temperatures, long run time, extreme pH levels and are energy intensive. Moreover,

the target of each of these methods is different and therefore also the quantitative inter-

action between the pretreatment parameters must still be clarified [11]. Hydrolysis of the

3


pretreated lignocellulose is performed mainly by highly specific enzymes (endoglucanases,

exoglucanases, β-glucosidases) [12] or by using diluted acids (sulfuric, phosphoric or nitric

acid) [13]. The most important factors influencing hydrolysis include porosity, cellulose fiber

crystallinity and lignin and cellulose content [12].

The lignocellulose hydrolysates resulting from the pretreatment contain a highly diverse

mixture of sugars, lignin and their degradation products. The main components are usually

glucose and xylose, although arabinose, galacturonic acid and rhamnose can also be found in

small concentrations. Lignin and its residues also represent a large part of most hydrolysates

[5]. In order to reach economically efficient bioconversion of this feedstock to bulk and

commodity chemicals, a high percentage of the available monosaccharides must be used

effectively.

Glucose utilization is common among microorganisms. Even though exceptions can be found

in E. coli mutant [14] or Pichia stipitis wt [15], which can utilize also C5 sugars, the ability to

metabolize other sugars, such as xylose and arabinose is rather rare. This work shows, that

U . maydis possesses such ability of utilization the above mentioned sugars.

The interest in engineering of strains capable of utilizing other monosaccharides rises.

S. cerevisiae [16], [17], [18] and Z . mobilis [2] have been engineered with heterologous path-

ways for xylose and arabinose utilization. Although these steps were made in order to optimize

ethanol production, these advances are also applicable for the production of other molecules,

which could be used for biofuel production [5].

1.3 Itaconic acid

Itaconic acid (ITA, CAS number: 97-65-4) is a promising new platform compound for produc-

tion of alternative biofuels and chemicals. Based on its industrial potential, it was selected by

US Department of Energy as one of the top 12 building block chemicals to be produced from

biomass [19].

Itaconic acid is a white crystalline powder soluble in water. More than 80 000 tons are pro-

duced per year, mostly for the synthesis of polymers, but also for bioactive compounds in

the fields of medicine and agriculture [20]. Several starting molecules for various purposes

originating from itaconic acid have been already identified (Figure 1.2).

4

1.3 Itaconic acid

Itaconic acid

3-Methyl THF

2-Methyl-1,4 BDO

Poly(itaconic acid)

Biofuels

Medical polymers

n

m n o

SBR Latex Technical polymers

Figure 1.2: Examples of possible itaconic acid application. Abbreviations: 3-MTHF -

3-methyltetrahydrofuran, MBD - 2-methyl-1,4-butanediol, SBR - styrene-

butadiene resin, [19], [21], [22].

There are chemical [23] and biological ways to produce itaconic acid. The biological way is

mostly studied in Aspergillus terreus, which produces itaconate via the Embden-Meyerhof-

-Parnas (EMP) pathway and the tricarboxylic acid (TCA) cycle. The enzymes involved in

the last part of the pathway are citrate synthase (EC: 2.3.3.1), pyruvate dehydrogenase (EC:

1.2.4.1), aconitase (EC: 4.2.1.3), malate dehydrogenase (EC: 1.1.1.37), pyruvate carboxylase

(EC: 6.4.1.1) and cis-aconitic acid decarboxylase (CAD; EC: 4.1.1.6). The first three enzymes

are localized in the mitochondrion, the last two enzymes in cytosol and malate dehydro-

genase can be found in both environments. The transport of the substrates between the

mitochondrion and the cytosol might be carried out via citrate/malate antiport (Figure 1.3)

[24].

The CAD activity was first identified by Bentley and Thiessen in A. terreus crude cell-free

extracts. This cytoplasmic enzyme [25] is implicated in the itaconate formation from the TCA

cycle intermediate cis-aconitate [26]. The sequence of the CAD gene of A. terreus has been

published quite recently [27]. Although this is the best characterized biochemical pathway

5


for the itaconate production, alternative routes may exist.

Figure 1.3: Proposed subcellular organization of the itaconate pathway in A. terreus.

Figure adapted from [24].

1.4 Industrial bioproduction of itaconic acid

Currently, almost all itaconic acid is produced using filamentous Aspergilli. Industrial itaconic

acid production by submerged fermentation of A. terreus was started by Pfizer Co. Inc. in

1955 in their plant in Brooklyn. This process was stopped in 1995. Nowadays, most of the

main itaconic acid suppliers come from China, some of them also from Japan, France and

the USA [20].

The optimal conditions for itaconate production by A. terreus are under phosphate limita-

tion with sugar concentrations between 100 and 150 g l−1 usually at 37 °C. Molasses is the

substrate of choice, but trace elements such as N, P, Fe, Mn, Mg, Cu and Zn significantly

influence the itaconate production, necessitating pre-purification of this substrate [28]. In

the initial growth phase the pH of the medium drops from 5 to less than 3. When phosphate

is depleted and sufficient biomass has been formed, the second (production) phase starts,

6

1.4 Industrial bioproduction of itaconic acid

where itaconate is produced, substantially free from other organic acids. In this phase, further

growth is prevented by keeping the phosphate level low.

It was previously shown that higher concentration of itaconate in culture broth inhibits its

production by A. terreus. Therefore the improvement of the production strain is focused on

engineering itaconate resistance [29]. However, depending on the cultivation conditions and

strain, A. terreus is able to produce 30 - 90 g l−1 itaconate in 3 - 25 days with a production rate

of 0.13 - 0.98 g l−1 h−1 and 0.34 - 0.62 g g−1 yield [30].

The process for the industrial purification of itaconic acid from the culture broth consists

of several steps; filtration in order to remove mycelia and other suspended solids, filtrate

concentration and itaconic acid crystallization (twice), decolorization of crystals with active

carbon, evaporation of the decolorized broth and recrystallization. Subsequently the itaconic

acid crystals are dried and packaged. The total itaconic acid recovery yield throughout the

whole process is approximately 80 % [20].

Filamentous fungi are known to be particularly difficult to handle in bioreactors [31]. Moreover,

certain fermentation conditions, such as substrate impurities, influence fermentations using

Aspergillus [32]. Also commercial fermentations must be cost-effective and therefore min-

imum of nutrients should be used. Furthermore utilization of both C5 and C6 sugars would

be a great advantage [19].

Therefore a search for alternative itaconate producers without disadvantages which are con-

nected with large scale cultivation of filamentous fungi is important. Many strains have

already been screened, both wild type and engineered (Table 1.1).

Among these, Ustilago is able to metabolize both C5 (arabinose, xylose) and C6 sugars (gluc-

ose, fructose) [this thesis, [33]]. This ability makes it an ideal microbial system for production

of biofuels and chemicals from hydrolyzed biomass.

Strain Itaconate concentration and time it was reached Reference

Pseudozyma antarctica 30 g l−1 [34]

Rhodotorula sp. 15 g l−1, 7 days [32]

Candida sp. 35 g l−1, 5 days [32]

Candida mutant 42 g l−1, 5 days [32]

Ustilago zeae * 15 g l−1, 5 days [35]

Ustilago maydis 53 g l−1, 5 days [32]

Table 1.1: Selection of alternative itaconate production hosts. * U . zeae is the former

designation for U . maydis.

7


1.5 Ustilago maydis

Ustilago maydis, a phytopathogenic basidiomycetous fungus, is the causative agent of smut

disease in corn (Figure 1.4A). U . maydis represents a microorganism with dimorphic life style.

The haploid form divides by budding, has a yeast-like appearance (sporidia) (Figure 1.4B)

and can be easily propagated on artificial media. The pathogenic dikaryotic filamentous

stage is formed by fusion of two sporidia that carry different alleles of a and b mating types

and needs its plant host to propagate [36].

Figure 1.4: (A) Smut disease on corn caused by dikaryotic form of U . maydis and (B)

the unicellular haploid form of U . maydis with buds (arrow), 630 ×.

U . maydis is a genetically well-characterized model organism and in 2006 its genome se-

quence has been published [37]. The genes are organized in highly compact manner and

many of them lack introns. The genome consists of about 6900 protein encoding genes and

many of them code for secreted proteins. The availability of genome sequence made identify-

ing of genes and metabolic pathways as well as developing U . maydis for biotechnological

purposes not only easier, but also quicker and more effective [38].

U . maydis is known to produce large concentrations of surface-active compounds. They can

be divided into two groups; mannosylerythritol lipids (MEL, ustilipids) and ustilagic acid.

MEL are found in the culture broth as extracellular oil droplets with higher density than water.

They consist of disaccharide units containing mannose and erythritol and a fatty acid of

8

1.6 Aim of this study

variable length. Ustilagic acid is a cellobiose lipid with antibiotic activity forming needle-like

crystals [38]. Both of these types of biosurfactants are produced under nitrogen (N) starvation

and their concentrations can reach up to 23 g l−1 [39]. The type of available carbon source is

the most important factor influencing the yield and ratio of both glycolipids [40].

The first large screening of Ustilago for metabolite products was performed in 1955 by Haskins

et al. [35]. 180 isolates underwent a screening for ustilipids and ustilagic acid production.

Several other metabolic products were identified, among them also the itaconate. In the

majority of the samples, more itaconate was produced in presence of CaCO3 than without.

This study represents the starting point of understanding Ustilago as a successful itaconate

producer.

The greatest advantage of industrial itaconic acid production using U . maydis is the fact, that

itaconate is secreted to the medium during cultivation by the yeast-like form of Ustilago and

therefore handling of the biomass in large scale fermenters even in high concentrations does

not bring the difficulties that are observed while fermenting filamentous fungi. Moreover

utilization of both C5 and C6 sugars by U . maydis is a fundamental industrial advantage.

Nevertheless as yet there is not much known about U . maydis and itaconate production. How-

ever, the preliminary study of Chandragiri and Sastry [41] and Rafi et al. [42] and the more

detailed study of Klement et al. [33] suggest that there is an increasing interest in this topic.

The goal of this work is to deliver more insights into this interesting but poorly studied subject.

1.6 Aim of this study

This study focuses on U . maydis as an alternative itaconate producer. The metabolic basis of

the itaconate production, as well as various aspects influencing the production process, was

investigated.

In order to find genes, which might be involved in the itaconate production pathway, mutants

with deletion and overexpression of genes possibly involved in the itaconate synthesis path-

way were created. These mutants were phenotypically characterized with regard to high-yield

itaconate production conditions. The influence of aeration of the culture, N concentration

as well as glucose concentration in the medium was studied. Moreover, various carbon

sources, which can be found in hydrolyzed biomass, were tested as potential substrates for

the itaconate production by U . maydis. Finally, the production of glycolipids as side products

of the itaconate synthesis was described.

Ultimately, the aim of this study is to optimize the itaconate production in U . maydis to such

extent, that it can be applied in sustainable production process.

9


10

CHAPTER 2

Materials and methods

2.1 Strains, plasmids and primers

Strain Description Origin

U . maydis MB215 wt DSM 17144, DSMZ Braun-

schweig, Germany *

U . maydis MB215 6_15 Overexpressed UM06344 This work



U . maydis MB215 ∆UM06344 ∆UM06344 *

U . maydis ∆cyp1∆emt1 ∆cyp1∆emt1, deleted cyp1 and

emt1 are responsible for glycol-

ipid production

[39] *

U . maydis MB215 ∆UM11778 ∆UM11778 *

E. coli DH5α ampR, fhuA2,

∆(argF − lacZ)U169,

phoA, glnV44, theta80,

delta(lacZ)M15, gyrA96, recA1,

relA1, endA1, thi−1, hsdR17

Strain no. 795, Cell collec-

tion of Institute of Applied

Microbiology, RWTH

Aachen University

S. cerevisiae DF5his1-1 his1−1, leu2−3, 2−112,

lys2−801, trp1−1, ura3−52

Strain was modified by H.

Ullrich from strain DH5

[43] *

Table 2.1: Strains used in this work. * Strains kindly provided by Prof. Michael Bölker.

11

Chapter 2: Materials and methods

Plasmid Description Origin

pMF1h ampR , hygR , oriC, lacZ , MCS in lacZ [44] *

pRS426 ori(f1), lacZ, T7 promoter, MCS (KpnI −SacI), T3

promoter, lacI , ori(pMB1), ampR , ori (2 micron),

URA3

*

pSMUT ampR , hygR [45] *

pMF2-3h hygR , Po2te f [44] *

UM06344 disruption vec-

tor

Integrated disruption cassette for UM06344, hygro-

mycin marker; derived from pRS426

This work

UM06344 overexpression

vector

Overexpressed UM06344; derived from pRS426 This work

Table 2.2: Plasmids used in this work. * Plasmids kindly provided by Prof. Michael

Bölker.

Primer Sequence 5’- 3’ Purpose

LF for * gtaacgccagggttttcccagtcacgac

gaatattcgcagcctgtcacgcggccca

acttg

Amplification of left flanking region

for UM06344 disruption vector

LF rev * attgtcacgccatggtggccatctaggc

cctggatacgaacgaaaacgagagatg

agc

Amplification of left flanking region

for UM06344 disruption vector

RF for * gtgcggccgcattaataggcctgagtg

gccactgggcagcagctagatcggcg

gacgc

Amplification of right flanking re-

gion for UM06344 disruption vec-

tor

RF rev * gcggattaacaatttcacacaggaaac

agcaatattgaaaagtcagatgggtaag

gagcg

Amplification of right flanking re-

gion for UM06344 disruption vec-

tor

Hyg pMF 2-3 h for ctgtcctgatttccaccctc Hygromycin probe for Southern

blot

Hyg pMF 2-3 h rev cagctatttacccgcaggac Hygromycin probe for Southern

blot

Continues on next page

12

2.1 Strains, plasmids and primers


pRS426+Hyg for * actcactatagggcgaattgggtacaata

ttggccgcactcgagtggcggcagatg

Amplification of hygromycin and

Po2te f for UM06344 overexpression

vector

Prom+CAD rev * tggattgggaaacgagtcgaggcatccgtg

gatgatgttgtctgtgtatg

Amplification of hygromycin and

Po2te f for UM06344 overexpression

vector

Prom+CAD for * catacacagacaacatcatccacggat

gcctcgactcgtttcccaatcca

Amplification of UM06344 gene for

UM06344 overexpression vector

CAD+pRS426 rev * ctcactaaagggaacaaaagctggagcaa

atattctactggtcgcgaacccag aggtca

Amplification of UM06344 gene for

UM06344 overexpression vector

Po2te f end for gctgtctcggcactatctttctttc Sequencing

UM06344 for agcagggctacaactcggta UM06344 probe for Southern blot

UM06344 rev tcttgatggtgacgctcttg UM06344 probe for Southern blot

UM06344 ORF rev taccgaattccgagttgcaagagcgtccg Amplification of UM06344 ORF

UM06344 ORF rev accggaattctccagatgcctcgactcg Amplification of UM06344 ORF

pUM UM06344 for ttaagttgggtaacgccagg Amplification of UM06344 overex-

pression cassette

pUM UM06344 rev ttccggctcctatgttgtgt Amplification of UM06344 overex-

pression cassette

Disruption cassette

for

aaagtgccatccatcctgga Amplification of integrated disrup-

tion cassette

Disruption cassette

rev

cgaaagcgtcgaagtggaag Amplification of integrated disrup-

tion cassette

Actin for cttcctgacggacaggtgat Amplification of part of actin gene

Actin rev tcctttcggatgtccaagtc Amplification of part of actin gene

pRS426-LF for acgacgttgtaaaacgacgg Sequencing

DownUM06344-LF

rev

agaggtctgcaatgcttggt Sequencing

DownUM06344-LF

for

agctctgcacagctctgttg Sequencing

Hyg-LF rev cgcgtgtcgattcacgtttt Sequencing

Hyg-RF r for ctgcaggcatgcaagcttca Sequencing

Hyg-RF for aacccagcatcaccaactgt Sequencing

Hyg-DownUM06344

rev

tgggcgagctcgaattcatc Sequencing

Continues on next page

13



pRS426-RF rev cgcaattaaccctcactaaaggg Sequencing

Primer 2 rev * cgataagccaggttaacctgtcttgttg

agacggatgagc

Screening

Table 2.3: List of used primers. * Primers were designed following the in vivo homo-

logous recombination [46].

2.2 Media

If not mentioned otherwise, all chemicals were supplied by Roth, Fluka or Merck.

YPD

10 g yeast extract

20 g pepton

20 g glucose

20 g agar for solid media

were dissolved in 1000 ml deionized water and sterilized in autoclave.

YPD-Hyg

YPD medium containing 400 µg ml−1 hygromycin (sterile filtered).

LB

10 g pepton

5 g NaCl

5 g yeast extract


were dissolved in 1000 ml deionized water and sterilized in autoclave.

LB-A

LB medium containing 100 mg l−1 ampicillin.

Modified Tabuchi medium (MTM; [47])

65, 110 or 250 g glucose, xylose, arabinose or sucrose

0.8, 1.6 or 4 g NH4Cl

0.5 g KH2PO4

0.2 g MgSO4

14

2.2 Media

0.14 g FeSO4 × 7H2O

1 g yeast extract

33 g CaCO3

was dissolved in such volume of water so that after sterilization in autoclave and subsequent

addition of appropriate volume of 50 % (w/w) sterile carbon source solution and 30 % (w/v)

sterile CaCO3 suspension total volume of 1 l medium was achieved.

SC-Ura

2.4 g (NH4)2SO4

1.7 g Difco Yeast Nitrogen Base w/o Amino Acids (BD)

12 g glucose

0.5 g Yeast Synthetic Drop-out Medium Supplement (Sigma)


were dissolved in 1000 ml deionized water and pH of 5.6 was adjusted using 10 M NaOH.

Consequently the mixture was sterilized in autoclave.

Regeneration agar

10 g yeast extract

20 g pepton

20 g sucrose

182.2 g sorbitol


were dissolved in 1000 ml deionized water and sterilized in autoclave. When medium cooled

down sufficiently, 400 µg ml−1 hygromycin (sterile filtered) was added. Agar plates were

poured according to Figure 2.1.

Figure 2.1: Outline of regeneration agar plate.

15


2.3 Cultivation methods

Cultivation on the solid media

All strains were transferred every two weeks on appropriate fresh plates and incubated for 24 -

48 h in 28 °C prior further use.

Cultivation in the liquid media

All strains were cultivated in 28 °C at 100 rpm on side-to-side motion agitator. If not men-

tioned otherwise, the cells were inoculated at starting OD O.5 AU (600 nm).

Precultures

If not mentioned otherwise, all precultures were cultivated in YPD and cells were washed

with 0.9 % NaCl solution before inoculation into main culture.

Precultures were carried out in 100 ml Erlenmeyer flasks containing 20 ml medium. The

incubation was performed at 28 °C and 100 rpm for 20 h.

Precultures for fermenter cultivation were carried out in 1 l Fernbach flasks containing 300

ml MTM supplemented with 30 g l−1 glucose, 1.6 g l−1 NH4Cl, w/o CaCO3. Cultivations were

carried out for 16 - 18 h in 28 °C at 100 rpm.

Cultivation for the screening tests

In order to screen overexpression mutants for increased itaconate production cells were

cultivated in 6-well plates (BD) containing 2.5 ml MTM (250 g l−1 glucose, 1.6 g l−1 NH4Cl).

Cells were cultivated in 28 °C at 100 rpm for 24, 48, 72 or 96 h, resp. Cultures were inoculated

with cells at OD 0.5 AU (600 nm).

To find out whether the disruption cassette was integrated into the right position in U . maydis

genome, DNA isolations were performed. For these purposes cells were cultivated in 10 ml

YPD in 100 ml Erlenmeyer flasks in 28 °C at 100 rpm for 24 h. Cultures were inoculated with

cells at OD 0.5 AU (600 nm).

Cultivation for the off-gas analysis

Cultivations in airtight DURAN bottles (Schott) were performed according to Robertz et al.

(1999) [48] with various filling volumes (10, 50 or 100 ml l−1, resp.) in MTM containing 65, 110

or 250 g l−1 glucose. In each case MTM was supplemented with 1.6 g l−1 NH4Cl. Bottles were

shaken in 28 °C at 100 rpm for about 144 h. Cells were inoculated at OD of 0.5 AU (600 nm).

Cultivation in the fermenter

Fermenter batch cultivations were performed in 2 l fermenter model Biostat A Plus (Sar-

torius).

16

2.3 Cultivation methods

Cells were collected from preculture by centrifugation (5 000 rpm, 5 min) and resuspended in

0.1 l MTM w/o CaCO3. The suspension was then transferred into the fermenter containing

1.4 l MTM w/o CaCO3.

Fermenter batch cultivation conditions were:

Working volume: 1.5 l

Temperature: 28 °C

Dissolved oxygen tension (DOT): 80 or 15 %

pH 6.

pH value was controlled automatically by adding 10 M NaOH. Sterile silicone antifoaming

emulsion was automatically added, when foam occurred.

An easyferm Plus K8 200 electrode (Hamilton) measured the pH level while an Oxyferm FDA

225 electrode (Hamilton) determined the DOT level.

If necessary, fermenters were cooled by Frigomix 1000 cooling system (Sartorius). Data were

collected using BioPAT MFCS/DA 2.1 software (Sartorius).

Cell growth quantification

Determination of optical density

The optical density was determined using Pharmacia LKB Ultrospec III (Pharmacia) at 600

nm through 1 cm thick cell layer.

In order to avoid the interference of the CaCO3 with the OD determination, 36% HCl was

used to dissolve the particles. In all cases 500 µl sample and 50 µl HCl (1/10 of the sample

volume) were used. However, the OD determination represents only an indication of the

biomass production, because the HCl used for the OD determination could precipitate some

of the cultivation products (i.e. organic acids or glycolipids).

The cells themselves were not influenced by the addition of HCl, which was demonstrated in

a separate experiment (data not shown). Briefly, the cells of U . maydis wt were cultivated for

144 h in YPD medium. Since there was no itaconate produced in this medium, the presence

of the buffer, CaCO3, which disturbs the OD determination, was not necessary. Each sample

was divided into two aliquots and only one was treated with HCl. Subsequently the OD was

measured in both aliquots.

Determination of the dry cell biomass from the fermenter batch cultivations

The samples of cell cultures of 5 ml were filtered through filter paper type MN 218 B (Macherey-

Nagel). These were washed with 5 ml deionized water, dried for 72 hours at 65 °C and

consequently weighted.

17


2.4 Analytical methods

2.4.1 Fermentation broth analysis

Samples were taken from a homogenous culture using a sterile pipet. A syringe was filled with

the content of the pipet and sterile filtered. The filtrate was used for organic acids, carbon

sources and TCA cycle intermediates determination.

Determination of the itaconate content

In order to verify the occurance of the itaconate in the culture broth, a fraction containing

the itaconate peak was isolated and its presence was confirmed by means of NMR (data not

shown; personal communication with Dr. Nicolas Lassauque).

Itaconate concentration determination was performed using Beckman Coulter System Gold

series HPLC (Beckman Coulter) consisting of an autosampler, binary pump, thermostatted

compartment for column, injection system and UV detector. The separation of cultivation

products was accomplished with LiChrospher 100 RP-18 column (5 µm particle size, Merck).

Chromatographic conditions were: mobile phase, 0.1 % phosphoric acid; flow-rate, 2 ml

min−1; injection volume, 30 µl; column temperature, 60 °C; UV detection, 210 nm; run time,

8 min, autosampler temperature: 4 °C. Data were recorded and processed using 32 Karat

Software (Beckman Coulter).

Determination of the carbon sources content

Glucose, xylose, arabinose and sucrose concentration was detected using HPLC based on

measurement of refraction index.

Determination was carried out using a set containing SDS 9414 Solvent Delivery System

(Schambeck SFD), thermostatted compartment for column type SFD 12560 (Schambeck SFD)

and refractometric detector RI 2000 (Schambeck SFD). Samples were injected to the system

manually using Exmire Microsyringe. Data were collected using PeakSimple Chromatography

Data System SRI Model 333 (SRI Instruments) and processed with PeakSimple 3.21 software

(Schambeck SFD).

Glucose, xylose and arabinose were separated using column filled with organic-acid resins

(300 × 0.8 mm) and fitting pre-column (CS-Chromatographie Service). Deionized water was

used as mobile phase.

Sucrose was separated using column filled with carbohydrate Ca2+ (300 × 0.8 mm) and fitting

pre-column (CS-Chromatographie Service). 0.016 M H2SO4 was used as mobile phase.

In both cases the chromatographic conditions were: column temperature, 60 °C; injection

volume, 20 µl; run time, 10 min (sucrose: 15 min); flow-rate, 0.8 ml min−1.

18

2.4 Analytical methods

Determination of TCA cycle intermediates

Fumarate, succinate, malate and pyruvate titers were determined using UltiMate 3000 HPLC

system (Dionex). The system consisted of a degasser, isocratic pump, autosampler, ther-

mostatted compartment for column and variable wavelength UV detector. The TCA cycle

intermediates separation was based on the ion exchange mehanism using Trentec 308R-

Gel.H column (300 × 0.8 mm, Trentec Anylysentechnik). The determination parameters

were: mobile phase, 5 mM H2SO4; flow-rate, 0.6 ml min−1; injection volume, 5 µl; column

temperature, 25 °C; UV detection, 210 nm; run time, 30 min, autosampler temperature: 4 °C.

Data were recorded and processed using Chromeleon Software (Dionex).

Determination of the ammonium content

The consumption or release of ammonium was determined by means of Kjeldahl method

[49]. Briefly, the sample was distilled together with NaOH and back titrated using Na2CO3

solution.

Surface tension determination

The surface tension of fermentation broth was measured using table-top tensiometer (Krüss),

using the du Noüy ring method.

Lipid-specific staining

The samples containing lipids underwent the lipid-specific staining according to Burdon

(1946) [50]. Solutions of Sudan black B and Safranin were used.

Cell photographs

Photographs of microscopical view of cells were obtained using Leica DM750 microscope

containing built-in camera type Leica ICC50 Digital Camera Module (Leica Microsystems).

Photographs were processed using either Debut Video Capture Software or LAS EZ software

(Leica Microsystems).

2.4.2 Off-gas analysis

The biochemical oxygen demand (BOD), CO2 production (CBP) and respiratory quotient

(RQ) was followed using OxiTop system [48].

Each sample was set in 2 parallels, one for BOD (w/o NaOH) and the other for CBP (with

NaOH) determination. The flask for CBP determination contained filter paper soaked with

50% (w/v) NaOH solution for CO2 absorption. The other flask (for BOD determination) was

set without NaOH. While BOD was measured automatically by a sensor attached to the flask,

19


the CBP was calculated from BOD.

In order to find out whether the cells in the poorly aerated culture were cultivated under

oxygen limited conditions, calculations were performed based on the formula:

6 O2 + C6H12O6 → 6 CO2 + 6 H2O.

Results from the fermenter batch cultivation showed, that one half of utilized glucose was

used for biomass production (27 g/l utilized glucose/14 g/l biomass/first 24 h) and the other

half for energy generation. The amount of oxygen needed by the cells in the first 24 h of cul-

tivation to convert half of the utilized glucose into energy was calculated from the following

formula:

n (O2) = c(g l ucose)M(g l ucose) ∗ 0.5 ∗ 6

V q ;

where n is the amount of substance [mol], c is concentration [g/l], M is molecular weight

[g/mol], 0.5 represents the half of the utilized glucose used to generate energy, 6 is molar

coefficient of oxygen (from formula 1) and Vq is the quotient between 1 l to which all the

calculations were applied and the culture volume [l].

Consequently the amount of oxygen available to the cells at the beginning of the experiment

was calculated from the formula:

n (O2) = V22.4 ∗ 0.21;

where n is the amount of substance [mol], V is the volume of the gas phase present in the flask

[l], 22.4 is the volume of 1 mol of gas [l] and 0.21 represents the relative amount of oxygen in

air. The values used for the calculations (concentration of utilized glucose) were obtained

from the experiment dealing with the influence of the culture aeration on the itaconate

production. While in the poorly aerated cultures, the cells utilized 25 g glucose in the first 24

h, the cells in the highly aerated culture utilized 37 g glucose.

The RQ values were calculated based of the following formula:

RQ = |BODC BP |;

where the BOD and the CBP are first recalculated from g/l to mol.

20

2.5 Molecular biological methods


2.5.1 Genomic DNA isolation

The isolation of U . maydis genomic DNA was performed as described by Hoffman and Win-

ston (1987) [51] with some modifications. The final protocol was as follows:

• Cultivate the cells in 10 ml culture

• Collect the cells by centrifugation (5 000 rpm, 5 min) and resuspend the pellet in 0.5 ml

water

• Spin down the cell suspension (5 000 rpm, 1 min), discard the supernatant and resus-

pend the cells in remaining water

• Add 0.4 ml of buffer A (2 % Triton X-100, 1 % SDS, 100 mM NaCl, 10 mM Tris-HCl pH

8.0, 1 mM EDTA) and transfer the suspension to glass tube

• Add 0.4 ml phenol-chloroform-isoamyl alcohol (25:24:1) and 0.3 g acid-washed glass

beads (0.75 - 1 mm)

• Vortex for 2 × 5 min

• Add 0.2 ml TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) and spin down (12 000 rpm,

5 min)

• Transfer the aqueous layer to a fresh tube, add 1 ml 100 % ethanol and mix by inverting

• Spin it down (12 000 rpm, 2 min), decant the supernatant and add 0.4 ml TE buffer and

5 µl RNAse (10 mg ml−1, Roche)

• Incubate the mixture for 1 h in 37 °C

• Add 0.5 ml phenol-chloroform-isoamyl alcohol (25:24:1) and vortex strongly for 20 sec

• Spin it down (12 000 rpm, 10 min), transfer the aqueous layer to a fresh tube, add 0.5

ml chloroform-isoamyl alcohol (24:1) - repeat this step 3 ×

• Add 0.3 M ammonium acetate-ethanol solution - 2.5 fold sample volume, mix by

inverting and spin it down (12 000 rpm, 2 min)

• Discard the supernatant and dry the pellet

• Resuspend the pellet in 50 µl nuclease-free water and store it in 4 °C.

21


2.5.2 Plasmid DNA isolation

Plasmid DNA was isolated using either the QIAprep Spin Microprep Kit (Qiagen) or the In-

visorb Spin Plasmid Mini Two (stratec molecular).

2.5.3 Gel electrophoresis

The DNA fragments were separated using 1 % agarose gel in TAE buffer (4.84 g l−1 Tris, 0.372

g l−1 Na2EDTA, pH 8.0). The separation was performed in 75 V environment.

Consequently, the gel was stained with ethidium bromide solution and evaluated in UV light

(265 nm).

Fermentas GeneRuler 1 kb Ladder or Fermentas GeneRuler 100 bp Ladder (Fermentas) was

used as DNA fragment size standard.

For the extraction of the DNA fragments from gel, Biozym LE GP Agarose (Biozym) and Min-

iElute Gel Extraction Kit (Qiagen) were used. The gel concentration was chosen appropriately

to the DNA fragment size.

2.5.4 In vitro DNA methods

Purification of DNA

The restricted or amplified DNA was purified using High Pure PCR Product Purification Kit

(Roche) or MSB Spin PCRapace (Invitek).

DNA concentration determination

The concentration of the isolated genomic DNA was determined using UV-VIS Spectropho-

tometer UV-1650PC (Schimadzu) and UVProbe 2.31 software (Schimadzu) at 260 nm and

concentration was calculated using the following formula:

(dsDNA) = A260 nm ∗ 50 µl ml−1 ∗ dilution factor.

The concentration of plasmid DNA was estimated from gel of appropriate concentration

using DNA frangment size standard (Fermentas, see Gel electrophoresis).

Polymerase chain reaction (PCR)

All reactions were performed as described by Newton and Graham (1994) [52] using GoTaq

DNA Polymerase (Promega).

For the amplification of the ORFs and the DNA fragments for sequencing, Phusion High-

Fidelity DNA Polymerase (Biozym) was used.

All reactions were performed using FlexCycler (analytik Jena).

22


DNA sequencing

For the sequencing of the DNA fragments services of either 4base lab or Eurofins MWG

Operon were used.

2.5.5 Transformation

The transformation of the E. coli DH5α followed the heat shock method by Sambrook et. al

(1989) [53].

The yeast cells were transformed according to Gietz and Woods (2002) [54] using the lithium

acetate/ss DNA/PEG method. The in vivo homologous recombination was used for vector

construction [46].

The transformation of U . maydis cells was performed using protoplasts. The protocol fol-

lowed the method by Tsukuda et al. (1988) [55] using STC/PEG and sorbitol solution.

2.5.6 Southern blot

The Southern blot assays were performed as described by Ausubel et. al (1991) [56].

For the transfer of the DNA fragments pressure was applied evenly on the gel and the mem-

brane. For this purpose a stack of paper towels and weight was used.

The DIG Easy Hyb solution, Blocking reagent, Anti-Dioxigenin-AP Fab fragments, CDP Star

solution, DIG DNA Labeling mix for probe labeling, DNA Molecular Weight Marker VII, DIG-

labeled (all Roche) and Biodyne B 0.45 µm membrane (Pall Corporation) were used.

23


24

CHAPTER 3

Results

3.1 Selection of the gene of interest

In A. terreus, the cis-aconitate decarboxylase (CAD), an enzyme that converts cis-aconitate

into itaconate, plays a crucial role in the itaconate production [27]. A sequence of the gene

coding for CAD in A. terreus, was recently published by Kanamasa et al. [27]. The biosynthetic

pathway of the itaconate in U . maydis is not known, but the gene UM06344 encodes a protein

with 23% identity to the CAD gene of A. terreus (ATEG 09971). Although the similarity between

the A. terreus and U . maydis proteins is very low, UM06433 was investigated in more detail in

order to learn whether it plays a role in the itaconate production.

3.2 Disruption of UM06344 in U. maydis

In order to knockout the UM06344 gene a disruption cassette was constructed according to

Mézard et al. [46] and Jansen et al. [57]. This method is based on the homologous recom-

bination system of S. cerevisiae. The vector consisted of carrier plasmid pRS426, containing

the ampicillin selection marker and the ura3 gene, left (LF) and right (RF) flanking regions of

UM06344 and a hygromycin resistance marker (Figure 3.1).

The individual cassette parts were amplified using PCR and chimeric primers (LF for, LF

rev, RF for and RF rev; see Materials and methods, Table 2.3). In order to verify, whether

the constructs were correct, restriction analysis with SspI restriction enzyme (Roche) and

sequencing were performed. Four out of twelve tested constructs were correct.

Selected correct constructs were restricted with SspI and transformed into U . maydis proto-

plasts. About 5 µg DNA was transformed, resulting in about 150 colonies per plate after 120 -

25

Chapter 3: Results

168 h of cultivation. Some of the colonies were cultivated further in shaken flasks and those

with sufficient growth (about 80 transformants) were submitted to screening.

Figure 3.1: Map of UM06344 disruption vector (A), outline of the region around the

UM06344 gene in U . maydis genome (B) and the cassette integrated into

U . maydis genome (C). (A) HygR : hygromycin resistance marker, RF: right

flank, AmpR : ampicillin resistance marker, LF: left flank.

Genomic DNA from transformants was isolated and PCRs with three different sets of primers

were performed (disruption cassette for and rev, primer 2 rev and ORF RF rev; see Materi-

als and methods, Table 2.3). The aim was to find an UM06344 deletion mutant. Although

the cultivation experiments in YPD containing hygromycin suggested that the cassette was

successfully integrated into the genome, no bands of the correct size were found in the PCR

analyses. Thus, obtaining the mutants with deleted UM06344 gene was not successful.

A mutant with deleted UM06344 gene was kindly provided later by Prof. M. Bölker. A com-

parison of the itaconate production between U . maydis wt and ∆UM06344 mutant strain is

shown in Figure 3.2.

Substantially different results from cultivation experiments were obtained. If UM06344 in-

deed encodes a CAD enzyme, the deletion of this gene would likely lead to reduced itaconate

production. Figure 3.2A shows significantly increased itaconate titer from the ∆UM06344

mutant in comparison to the wt strain. However, the results from further cultivations were

inconsistent and did not confirm previous findings (Figure 3.2B). Nevertheless, none of the

cultures exhibited decreased itaconate production by the ∆UM06344 mutant with regard to

the wt, indicating that the UM06344 gene does not encode a CAD enzyme.

26

3.3 Overexpression of UM06344 in U. maydis

Figure 3.2: Production kinetics of the itaconate under various conditions. Cells were

cultivated in (A) 10 ml MTM containing 65 g l−1 glucose, 0.8 g l−1 NH4Cl

and CaCO3 (n = 3; the values are mean of three analytical determinations

± standard deviation) and (B) 50 ml MTM containing 65 g l−1 glucose,

1.6 g l−1 NH4Cl and CaCO3 (n = 1). Green line: U . maydis wt, black line:

U . maydis ∆UM06344 mutant. 5.1


Although the results from the experiments with the ∆UM06344 mutant were inconsistent,

certain cultures did show that this gene might be otherwise related to the itaconate produc-

tion. Therefore, in order to obtain additional information, an overexpression construct of this

gene was created.

The UM06344 overexpression vector construction was also based on the protocols of Mézard

et al. [46] and Jansen et al. [57]. The vector consisted of hygromycin marker, Po2tef promoter

and UM06344 ORF. Plasmid pRS426 containing an ampicillin resistance marker and ura3

gene was used as a carrier plasmid. The map of this construct is shown in Figure 3.3. Each part

of the construct was amplified via PCR using chimeric primers (pRS426+Hyg for, Prom+CAD

for and rev and CAD+pRS426 rev; see Materials and methods, Table 2.3). The constructs were

tested via restriction with NdeI (Fermentas). Ten of twelve tested constructs proved to be

correct, which was later confirmed by sequencing.

Prior the transformation, the vector was restricted using SspI and the mixture was heat in-

activated. About 5 µg of the DNA was transformed into the Ustilago maydis wt protoplasts.

Transformation resulted in 0 - 80 colonies per plate after 120 - 168 h of cultivation.

27

Chapter 3: Results

Figure 3.3: Outline of the UM06344 overexpression vector. HygR : hygromycin resist-

ance marker, AmpR : ampicillin resistance marker.

3.3.1 Screening of transformants for increased itaconate production

One hundred thirty eight transformants were submitted to screening for increased itaconate

production. The cells were cultivated in 6-well plates, where each well contained 2.5 ml MTM

with 250 g l−1 glucose, 1.6 g l−1 NH4Cl and CaCO3. The plates were cultivated in 28 °C at

100 rpm for 24-96 hours. The culture broth was filtered and the itaconate production was

analyzed by HPLC.

The UM06344 overexpression cassette contained four different DNA regions originating from

U . maydis genome. Three integration events could occur (Figure 3.5). (A) Random integra-

tion of single or multiple copies of the linearized cassette into the genome [58]. In this case

random gene-knockout takes place. (B) Integration of the cassette based on the homologous

recombination at a non-random site (in this case probably an exchange of the hsp70 gene).

As a result, the Po2tef promoter and the UM06344 ORF would get lost. (C) Integration based

on a single cross-over between any of the homologous cassette parts of the circular vector

and the genome. Although this last event is improbable since the vector was digested with

SspI before transformation, it is still possible that a trace of undigested vector remained.

In order to find a mutant where integration of the construct led to a gene knockout which

is beneficial for the itaconate production, 138 individual transformants were screened for

increased the itaconate production (Figure 3.4).

28


Figure 3.4: Itaconate production by UM06344 overexpression cassette transformants

compared to the U . maydis wt. The wt itaconate titers were used as 100 %.

In average, the mutants produced 1.2-fold more itaconate with regard to the wt. About half of

the mutants produced itaconate in a standard deviation range. However, the graph (Figure 3.4)

is clearly asymmetrical towards the improved production (42 mutants produced significantly

more itaconate and 27 significantly less itaconate with regard to the wt), indicating that even

though the UM06344 gene is unlikely to encode a CAD enzyme, its overexpression or random

integration of the construct does have a certain positive effect on the itaconate production.

Transformants with significantly higher itaconate titer were present. The effect of UM06344

overexpression or random integration of the construct is considered to be small. The increase

in the itaconate production was most likely due to a positional effect of a random insertion

of the overexpression construct into the genome (Figure 3.5A). The three mutants with the

highest itaconate production (U . maydis 6_15, 6_17 and 2_29) were submitted to further

characterization.

29

Chapter 3: Results

3.3.2 Southern blot analyses

As discussed previously, there are three different ways of integration of the UM06344 over-

expression cassette. To find out whether and how UM06344 overexpression construct was

integrated into the genome of the selected mutants, Southern blot analyses were performed.

Genomic DNA was digested with BamHI (Roche) and probes for the hygromycin resistance

marker and the UM06344 gene were used (Figure 3.5A,B).

In the case of the hygromycin probe, there was a band visible in the wt lane as well as in

the transformants lanes. Since this band was present in the wt sample, it had to be un-

specific. This may be explained by the position of the hygromycin probe in the cassette,

where the probe consisted partly of the hygromycin ORF and partly of the Hsp70 promoter

(Figure 3.5A,B). The presence of this promoter, originating from the U . maydis genome, de-

creased the specificity of the probe.

Each of the transformants exhibited a 4.9 kb band which corresponds to the original UM06344

gene (Figure 3.6). If the overexpression cassette would integrate into the genome based on

the single cross-over (Figure 3.5C), the bands on the Southern blot film would be as follows:

UM06344: 4946 and 10141 bp; Po2tef : 18403 bp; Hsp70 promotor: 12346 bp and in case of

Hsp70 terminator: 16077 bp. None of these bands are present on the film (Figure 3.6). This

indicates that the overexpression cassette did not integrate into the genomic DNA based

on single cross-over. Moreover, the transformant 6_15 exhibited a clearly brighter band

(Figure 3.6A1, B), which is most likely a result of multiple tandem integrations of the linear

construct. The integration numbers are summarized in Table 3.1.

The results from the Southern blot analyses revealed, that each of the selected mutants con-

tained at least one extra copy of the UM06344 gene in its genome.

Moreover, the UM06344 overexpression cassette integrated into the genome of each strain

based on the scenario in Figure 3.5A and therefore a random gene-knockout can be expected.

30


gDNAPo2

tef

C

UM06344 ORF

pBR322 originAmpR

2μ origin

ura3

HygR

Figure 3.5: Outline of the three possibilities of the integration of the UM06344 over-

expression cassette into the gDNA of U . maydis. (A) Random insertion in

the genome, (B) double cross-over between the Hsp70 promotor and the

terminator resulting in the loss of the Po2tef and UM06344 ORF parts and

(C) single cross-over of the circular plasmid with any of the homologous

regions in the cassette, here with the Po2tef promoter as an example. AmpR :

ampicillin resistance marker, HygR : hygromycin resistance marker.

31

Chapter 3: Results

Figure 3.6: Southern blot analysis of the genomic DNA of U . maydis wt and the trans-

formants with integrated UM06344 overexpression cassette (6_15, 6_17,

2_29). (A1, A2) UM06344 probe and (B) hygromycin probe was used. The

black arrows show the original UM06344 gene (A1, A2) or the unspecific

annealing position for hygromycin gene (B), the red arrows show the integ-

rated UM06344 gene (A1, A2) or gene coding for hygromycin originating

from the overexpression cassette (B). Abbreviations: M - marker (DNA

fragment sizes are shown in bp).

Transformant No. of UM06344 genes No. of hygromycin genes

U . maydis 6_15 3 + 1 original 3



Table 3.1: Summary of integration numbers of UM06344 overexpression cassette

parts.

32

3.4 Itaconate production kinetics under various cultivation conditions

3.4 Itaconate production kinetics under various

cultivation conditions

After the production mutants were selected, the aim of the further investigation focused on

selection of optimal cultivation conditions with regard to high-yield itaconate production.

A certain pattern in the distribution of the lines representing each determined parameter

was visible among all the strains and all studied conditions. Figure 3.7 shows an example of a

cultivation chart. The U . maydis wt cells were cultivated in 50 ml MTM containing 250 g l−1

glucose, 1.6 g l−1 NH4Cl and CaCO3 (basic cultivation conditions, Table 3.2). This chart rep-

resents a typical cultivation graph, which can be divided into two distinct phases: 1) A growth

phase, with a high rate of biomass growth and no itaconate production. 2) A production

phase, where the itaconate is produced while the biomass growth rate is greatly decreased.

The trigger for the transition from phase 1 to phase 2 was the ammonium depletion, which

usually occurred within the first 24 - 48 h.

Figure 3.7: A typical cultivation chart of U . maydis wt. Green line: glucose utilization,

blue line: OD, red line: ammonium consumption, yellow line: itaconate

production. Cells were cultivated under the basic cultivation conditions.

The MTM contained CaCO3. The values are mean of three analytical

determinations ± standard deviation. 5.2

33

Chapter 3: Results

Aside of the itaconate, a high concentration of malate is produced. This phenomenon is

known from the literature [47] and is strain-dependent.

In order to produce the itaconate with the highest possible titer and in the most efficient way,

several cultivation conditions which may influence the itaconate concentration were tested

in further experiments.

Briefly, three parameters were studied; glucose and ammonium concentration and culture

aeration. Two of these parameters were always kept constant and the third one varied (Table

3.2).

Condition 1 Condition 2 Condition 3

Setup 1 1.6 g l−1 NH4Cl 50 ml culture

65 g l−1 glucose

110 g l−1 glucose

250 g l−1 glucose

Setup 2 250 g l−1 glucose 1.6 g l−1 NH4Cl

10 ml culture

50 ml culture

100 ml culture

Setup 3 50 ml culture 250 g l−1 glucose

0.8 g l−1 NH4Cl

1.6 g l−1 NH4Cl

4 g l−1 NH4Cl

Table 3.2: Experimental design of optimization of the itaconate production paramet-

ers.

3.4.1 Influence of glucose concentration

Since the carbon source concentration is known to affect the biomass production and the

itaconate yield and titer in A. terreus [59], cultivation experiments with varying initial gluc-

ose concentrations were performed in order to find the optimal glucose concentration for

U . maydis.

Three different glucose concentrations were applied, 65, 110 and 250 g l−1. The cells were

cultivated in 50 ml MTM per 1 l flask in medium containing 1.6 g l−1 NH4Cl and CaCO3. As

can be seen in Figure 3.8, the high concentrations of available initial glucose seemed to have a

negative effect on the itaconate production. In the experiments with 65 and 250 g l−1 glucose,

the highest itaconate production rate was observed between 72 and 96 h while with 110 g l−1

glucose it was between 120 and 144 h of cultivation. In the case of 65 g l−1 glucose the average

itaconate production rate was 0.109 ± 0.051 g l−1 h−1, with 250 g l−1 glucose it was 0.027 ±0.017 g l−1 h−1 and with 110 g l−1 glucose 0.06 ± 0.016 g l−1 h−1.

34


Figure 3.8: The ammonium utilization (dashed lines) and the itaconate production

(full lines) in MTM containing 250 (A) and 65 (B) g l−1 glucose. Green

lines: U . maydis wt, red lines: U . maydis 6_15, blue lines: U . maydis 6_17

and yellow lines: U . maydis 2_29. The values are mean of three analytical


In the cultures with the high glucose concentration, the substrate was not yet depleted and the

itaconate titer was still increasing, suggesting that the itaconate production was not finished

after 144 h. Therefore, the final itaconate titer could be much higher. The mutant strains

exhausted the 65 g l−1 glucose already within 120 h. The end-point glucose concentrations as

well as the final OD and itaconate yield values are summarized in Table 3.3.

After 144 h, significantly more itaconate was produced from the medium containing 65 g l−1

glucose. The high glucose concentration seemed to have a negative effect on the itaconate

production rate and yield. The reason might be the osmotic stress caused by such high

glucose concentration. Indeed, the ammonium was depleted at a lower rate in the culture

with 250 g l−1 glucose, indicating that the growth rate was decreased under this condition. In

the experiments with only 110 g l−1 glucose the itaconate production did not improve much

compared to the 250 g l−1 culture.

An overview of the influence of the glucose concentration on the final itaconate titer is

summarized in Figure 3.9. While the itaconate titers obtained from the two higher glucose

concentrations (250 and 110 g l−1) were in average very similar, the final itaconate titers

obtained with the 65 g l−1 initial glucose concentration were significantly higher. Moreover,

all the three selected mutants produced substantially more itaconate with regard to the wt,

where the effect of the glucose concentration was the least visible.

35

Chapter 3: Results

Initial glucose

concentration

U . maydis

wt

U . maydis

6_15

U . maydis

6_17

U . maydis

2_29

OD (AU)

65 g l−1 17.0 ± 0.4 17.8 ± 1.4 15.3 ± 1.7 13.1 ± 1.5

110 g l−1 22.1 ± 0.7 25.9 ± 2.0 29.1 ± 0.6 19.5 ± 1.8

250 g l−1 17.9 ± 0.8 24.3 ± 1.2 25.5 ± 1.2 19.7 ± 1.4

Glucose (g l−1)

65 g l−1 2.0 ± 0.8 0 0 0

110 g l−1 64.0 ± 5.0 64.5 ± 15.6 67.8 ± 7.0 56.4 ± 11.5

250 g l−1 152.4 ± 6.1 145.8 ± 1.7 160.0 ± 3.5 169.7 ± 11.8

Yield (g g−1)

65 g l−1 0.05 0.121 0.074 0.161

110 g l−1 0.041 0.065 0.053 0.106

250 g l−1 0.032 0.017 0.025 0.051

Table 3.3: The summary of the end-point values of OD (AU), glucose concentration (g

l−1) and yield (itaconate/utilized glucose, g g−1). The values (mean of three

analytical determinations ± standard deviation) were obtained after 144 h

of cultivation.

Figure 3.9: An overview of the influence of the initial glucose concentration on the

itaconate titer. Initial glucose concentration: 250 g l−1 (red columns), 110

g l−1 (green columns) and 65 g l−1 (orange columns). The values (mean

of three analytical determinations ± standard deviation) were obtained

after 144 h of cultivation. 5.4

36


In order to verify the hypothesis concerning the influence of the osmotic stress caused by

the high glucose concentration, a cultivation experiment in air-tight flasks with medium

containing the high and low glucose concentrations was performed. In these flasks, the

biological oxygen demand (BOD) and the biological production of CO2 (CBP) were studied.

The results are shown in Figure 3.10. Since there were no differences among the four strains,

the chart of the 2_29 mutant was used as an example.

Figure 3.10: Influence of various glucose concentration in MTM on BOD and CBP in

the U . maydis 2_29 strain. The cells were cultivated in 50 ml MTM in 1 l

flask containing 1.6 g l−1 NH4Cl and CaCO3. Blue line: BOD (250 g l−1

glucose), red line: BOD (65 g l−1), orange line: CBP (250 g l−1) and green

line: CBP (65 g l−1 glucose). The values in are mean of three analytical

determinations.

The most apparent differences in the BOD and the CBP were present between 8 and 36 h of

cultivation. Then the values reached stationary phase and the previous differences caused by

the glucose concentration vanished.

The presence of the high glucose concentration and consequently also the high osmotic

pressure led to slower O2 consumption and CO2 production compared to the low glucose

concentration. This indicates decreased growth rate and therefore overall lower metabolism

level (e.g. N utilization in Figure 3.8A).

37

Chapter 3: Results

The respiratory quotient (RQ) was among all the four strains the same - slightly above 1 (data

not shown). The reason, why the RQ is above 1 could be the extra CO2 released from the

reaction between CaCO3 present in the culture and the acids produced by the cells.

3.4.2 Influence of culture aeration

The aeration of the culture, as one of the putative factors influencing the itaconate production,

was studied.

The cells of U . maydis wt as well as the 6_15, 6_17 and 2_29 mutants were cultivated in 1 l

flasks with 10 ml (highly aerated), 50 ml or 100 ml (poorly aerated) MTM containing 250 g l−1

glucose, 1.6 g l−1 NH4Cl and CaCO3. The flasks were agitated at 100 rpm.

The findings revealed that the aeration of the culture has a strong influence on the itaconate

titer. Substantially more itaconate (in average 10-fold more) was produced in the 10 ml culture

(Figure 3.11. Moreover, the mutants produced significantly more itaconate in comparison to

the wt regardless of the culture aeration. Interestingly, the itaconate production rate of strain

2_29 was approximately 6-fold higher than the other strains under low aeration, indicating

that this mutant is capable of relatively efficient itaconate production even at low aeration

levels.

Figure 3.11: The kinetics of ammonium utilization (dashed lines) and the itaconate

production (full lines) in 100 ml culture (A) and 10 ml culture (B). Green

lines: U . maydis wt, red lines: U . maydis 6_15, blue lines: U . maydis

6_17 and yellow lines: U . maydis 2_29. The values are mean of three

analytical determinations ± standard deviation. 5.5

38


Interestingly, while in the 10 ml culture the N exhaustion was the trigger for the itaconate pro-

duction as expected, in the 100 ml culture the itaconate production started even though the

N was not depleted. The severe oxygen limitation might in this case have induced itaconate

production, although the exact mechanism is not clear.

In the poorly aerated culture, the final biomass concentration and the N consumption rate

were greatly reduced, indicating that the cultures were severely oxygen limited. As a result,

the cells in this culture consumed 4-fold less glucose in comparison to the highly aerated

culture. The end-point OD values as well as the final glucose concentration and the itaconate

yield values are summarized in Table 3.4.

Culture

volume

U . maydis

wt

U . maydis

6_15

U . maydis

6_17

U . maydis

2_29

OD (AU)

10 ml 45.3 ± 2.4 33.7 ± 2.9 42.3 ± 1.8 27.2 ± 1.3

50 ml 17.9 ± 0.8 24.3 ± 1.2 25.5 ± 1.2 19.7 ± 1.4

100 ml 10.1 ± 0.9 10.7 ± 0.7 11.9 ± 1.2 11.3 ± 0.7

Glucose (g l−1)

10 ml 102.7 ± 16.2 100.7 ± 4.9 88.7 ± 10.2 131.2 ± 18.4

50 ml 166.3 ± 8.6 151.4 ± 9.2 149.8 ± 4.7 165.6 ± 5.0

100 ml 179.1 ± 2.9 195.8 ± 6.9 194.0 ± 3.0 217.6 ± 4.0

Yield (g g−1)

10 ml 0.022 0.076 0.095 0.189

50 ml 0.002 0.003 0.005 0.012

100 ml 0.008 0.013 0.022 0.107




of cultivation.

As shown in the Figure 3.12, the aeration of the culture plays an important role in the op-

timization of the high-yield itaconate production conditions. The final itaconate titers were

increased manifold in the highly aerated cultures in comparison to the poorly aerated ones

(100 and 50 ml). There were no big differences in the itaconate titers obtained from the latter

two cultures by any of the three mutants. The only difference was observed at the wt strain.

Moreover, the selected mutants were able to produce significantly more itaconate with regard

to the wt in the 10 ml cultures (6_15: 3.6-fold more, 6_17: 4.8-fold more and 2_29: 7-fold more).

39

Chapter 3: Results

Figure 3.12: An overview of the influence of the culture aeration on the itaconate titer.

Culture volumes: 100 ml (red columns), 50 ml (green columns) and 10 ml

(orange columns). The values (mean of three analytical determinations

± standard deviation) were obtained after 144 h of cultivation. 5.6

In order to verify the reason for the low cell performance in the poorly aerated culture in

comparison to the highly aerated one, cultivation experiments in 1 l air-tight flasks were

performed. The BOD and the CBP were determined in MTM with 250 g l−1 glucose, 1.6 g l−1

NH4Cl and CaCO3 and flasks were filled with either 10 or 100 ml of medium.

Each strain used in this experiment demonstrated the same pattern in O2 consumption and

CO2 production. Therefore, results from strain 2_29 are given as an example (Figure 3.13).

The highly aerated cultures used oxygen at an almost linear rate, which was 3-fold higher than

the initial rate in the poorly aerated culture. In the case of 100 ml culture the BOD reached

stationary phase after the first 30 h of cultivation, indicating that oxygen was depleted at this

point in the closed flasks.

In order to find out whether the cells in the poorly aerated cultures were cultivated under

oxygen limited conditions, calculations were performed (for the calculations see Material

and methods, section 2.4.2).

40


Figure 3.13: Overview of the results from the off-gas analysis. The BOD values are

located in the positive and the CBP values in the negative areas of the

charts. The U . maydis 2_29 mutant was used as an example. The results

were obtained from the cultivation in MTM containing 250 g l−1 glucose,

1.6 g l−1 NH4Cl and CaCO3. Blue line: BOD in 10 ml, green line: BOD

in 100 ml, orange line: CBP in 100 ml and red line: CBP in 10 ml. The

values are mean of three analytical determinations.

Since the experiments were performed in closed flasks, only a fixed amount of oxygen is

available to the cells. In the 100 ml culture (900 ml air) 8 mmol oxygen was available at the

beginning of the experiment. Assuming that half of the glucose consumed during growth is

used to generate energy according to formula above, the cells would need 42 mmol oxygen.

Therefore, the cells likely reached the oxygen limitation after few hours of cultivation.

In the case of the 10 ml culture (990 ml air), there was 9 mmol oxygen available. Although the

biomass concentration was higher under these conditions, the 10-fold lower culture volume

means that the cells would need only 6.3 mmol oxygen. These cells therefore were cultivated

under non-limiting oxygen conditions.

These calculations clarified why the cells performed so poorly in the 100 ml culture, and also

indicate that the 10 ml culture was not limited by the total available oxygen amount.

41

Chapter 3: Results

3.4.3 Influence of nitrogen concentration

Previous experiments showed that in some cases the itaconate production is triggered while

N is still present in the medium. Therefore cultivations in MTM containing excess as well as

limiting concentration of N were performed in order to test whether other growth limitations

can also trigger the itaconate production.

U . maydis wt and mutants 6_15, 6_17 and 2_29 were cultivated for 96 h in 50 ml MTM per 1 l

flask in medium containing 250 g l−1 glucose, CaCO3 and either 0.8, 1.6 or 4 g l−1 NH4Cl.

The N/P molar ratio in MTM containing 0.8 g l−1 NH4Cl is 4.54 and with 1.6 g l−1 NH4Cl

8.65. In MTM with 4 g l−1 NH4Cl it is already 20.8. Assuming that the cellular composition

of U . maydis is similar to yeast, (N/P ratio of 14.7; [60]), the presence of 0.8 and 1.6 g l−1

NH4Cl in medium will lead to N-limitation, while 4 g l−1 NH4Cl will cause a P-limitation. In

all calculations it was assumed that 0.1 g l−1 N was added with the yeast extract.

In the case of 4 g l−1 NH4Cl, the N was not depleted and yet the itaconate production started

(Figure 3.14A). The sudden decrease in the N utilization likely means that the phosphate was

exhausted in the first 24 h, indicating that under these conditions the limiting factor and

therefore the itaconate production trigger was the phosphorus.

Figure 3.14: The kinetics of ammonium utilization (dashed lines) and the itaconate

production (full lines) in MTM containing 4 g l−1 NH4Cl (A) and 0.8 g

l−1 NH4Cl (B). Green lines: U . maydis wt, red lines: U . maydis 6_15, blue

lines: U . maydis 6_17 and yellow lines: U . maydis 2_29. The values are

mean of three analytical determinations ± standard deviation. 5.7

42


Nevertheless the limiting concentration of N led to higher titers of the itaconate with regard

to the medium with limiting concentration of phosphorus (Figure 3.14B). This suggests that

an N-limitation is a better trigger for the high-yield itaconate production than P-limitation,

although there were no substantial differences in the itaconate titers with 4 and 1.6 g l−1

NH4Cl among the strains (Figure 3.15). The only exception represented the strain 2_29, which

produced also significantly more itaconate with 4 g l−1 NH4Cl (44-fold with regard to the

wt). This indicates that this mutant is capable of relatively efficient itaconate production

regardless of the available N concentration.

Figure 3.15: An overview of the influence of the concentration of the N supply on the

itaconate titer. NH4Cl concentrations: 4 g l−1 (red columns), 1.6 g l−1

(green columns) and 0.8 g l−1 (orange columns). The values (mean of

three analytical determinations ± standard deviation) were obtained

after 96 h of cultivation

The excess of N led to 1.6-fold more biomass even though the glucose consumption was

comparable among the conditions. The end-point glucose concentration, OD values and

itaconate yields are listed in Table 3.5.

As shown in the Figure 3.15, all the selected mutants produced significantly more itaconate

with regard to the wt under the highly N limited conditions (6_15: 6.6-fold more, 6_17: 10.5-

fold more and 2_29: 9-fold more with 0.8 g l−1 NH4Cl).

43

Chapter 3: Results

Initial NH4Cl

concentration

U . maydis

wt

U . maydis

6_15

U . maydis

6_17

U . maydis

2_29

OD (AU)

0.8 g l−1 19 ± 1 19 ± 2 21 ± 1 18 ± 2

1.6 g l−1 17 ± 1 20 ± 1 22 ± 1 14 ± 1

4 g l−1 31 ± 2 30 ± 4 31 ± 6 30 ± 3

Glucose (g l−1)

0.8 g l−1 192 ± 4 186 ± 6 185 ± 5 186 ± 5

1.6 g l−1 179 ± 10 162 ± 8 157 ± 4 171 ± 5

4 g l−1 192 ± 10 199 ± 6 209 ± 7 214 ± 11

Yield (g g−1)

0.8 g l−1 0.0074 0.0449 0.0696 0.0604

1.6 g l−1 0.0012 0.0024 0.0033 0.0109

4 g l−1 0.0008 0.004 0.0115 0.0553




of cultivation.

3.4.4 Selected optimal conditions

Based on the results from the previous experiments, optimal cultivation conditions for high-

yield itaconate production from U . maydis were selected and tested.

The cells were cultivated in MTM containing CaCO3 and the selected conditions were: high

culture aeration (10 ml culture per 1 l flask), low initial concentration of carbon source (65 g

l−1 glucose) and low N concentration (0.8 g l−1 NH4Cl). Four strains of U . maydis were com-

pared; wt and the three selected mutants 6_15, 6_17 and 2_29. The kinetics of the itaconate

production and yield, ammonium utilization, malate production, cell growth, and glucose

utilization were studied (Figure 3.16).

Itaconate production started, when the N, which was present in the medium in growth-

limiting concentration, was depleted (Figure 3.16A), which is in correlation with the previous

experiments. All the four strains exhausted the N within the first 24 h of cultivation, but the

itaconate production in the wt strain started only after 72 h.

The highest rate in the itaconate production was observed between 24 and 48 h by the strain

2_29 (0.29 ± 0.029 g l−1 h−1). Also the most itaconate was produced by this strain, 18.8 g

l−1 after 144 h, which is 9.8-fold more itaconate than wt produced (1.9 g l−1 after 144 h).

Generally, all three selected mutants produced significantly more itaconate than the wt.

44


Figure 3.16: The kinetics of the ammonium utilization and the itaconate production

(A, the dashed lines - ammonium concentration, the full lines - itaconate

production), malate production (B), cell growth (C) and glucose util-

ization (D) under selected optimal cultivation conditions. Green lines:

U . maydis wt, red lines: U . maydis 6_15, blue lines: U . maydis 6_17,

yellow lines: U . maydis 2_29. The values are mean of three analytical


Although the cells were cultivated under selected optimal conditions, the itaconate titer ob-

tained in this experiment by the strain 2_29 did not improve when compared to the previous

experiments. There was a comparable itaconate concentration reached in the cultivation of

the 10 ml culture in the experiments with various aeration levels (22.6 g l−1 itaconate after

144 h, strain 2_29, Figure 3.16B). This may suggest that the most influencing condition for the

itaconate production from those studied is the aeration of the culture.

Aside of the itaconate, malate was produced in high titers. Any side product which accom-

panies the itaconate production process potentially decreases the efficiency of the itaconate

production by the cells. Figure 3.16B shows production kinetics of the malate. In contrast

to the itaconate, the malate was produced directly from the beginning of the cultivation. It

45

Chapter 3: Results

appears that the malate yield decreases when the itaconate yield is increased. The ratio values

between the itaconate and the malate are summarized in Table 3.6. While the highest ratio

was obtained from the 2_29 strain, the lowest one belonged to the wt strain.

U . maydis wt U . maydis 6_15 U . maydis 6_17 U . maydis 2_29

Yield (g g−1) 0.03 0.24 0.22 0.3

Itaconate (g l−1) 2 ± 0.1 15 ± 0.2 14 ± 1 19 ± 0.2

Itaconate/malate 0.3 4.1 4.0 8.9

Table 3.6: Summary of values of the itaconate yield (itaconate/utilized glucose), titer

and itaconate/malate ratio. The values were calculated from samples ob-

tained after 144 h of cultivation. The itaconate titer values are mean of

three analytical determinations ± standard deviation.

A new pattern in the cell growth was observed during this experiment (Figure 3.16C). This

rapid increase in the OD in the first 24 h of cultivation and a decline in the OD after 48 - 72 h

was not present in any of the previous experiments. Nevertheless, a decrease in the OD was

visible among all strains after about 96 h.

On the contrary, the glucose was utilized throughout the whole experiment and with no signi-

ficant differences among the strains (Figure 3.16D), as was already observed in the previous

experiments. After 144 h, there was between 1.5 g l−1 (2_29 strain) and 5 g l−1 (wt) glucose left

in the culture.

A strong positive influence of the integration of the UM06344 overexpression cassette on the

itaconate titer was observed. Comparison between the results obtained from this experiment

revealed that the condition optimization had significantly higher effect on the mutants than

on the wt strain. The difference in the itaconate titer obtained from the mutants based on

the cultivation conditions was however huge. While the strain 2_29 produced only about 4 g

l−1 itaconate from 0.8 g l−1 NH4Cl in the N concentration optimization experiment, about 11

g l−1 itaconate from 65 g l−1 glucose in the glucose concentration optimization experiment

and about 22 g l−1 itaconate in the 10 ml culture in the experiment focusing on the culture

aeration, the same strain produced almost 19 g l−1 itaconate under the selected conditions.

Therefore, not only an improvement in the itaconate production was reached by the culture

conditions optimization but moreover, further optimization should be conducted with main

focus on the high-yield strain.

46

3.5 Itaconate production during fermenter batch cultivation

3.5 Itaconate production during fermenter batch

cultivation

The results from the shaken flasks experiments revealed, that the most important factor for

the itaconate production is the aeration of the culture. Since the conditions in the fermenter

are highly controlled, the oxygen limitation can be excluded. Moreover, cultivation in the

fermenter is closer to the industrial production. Therefore, fermenter batch cultivations were

performed.

U . maydis 2_29 strain was cultivated alongside the wt strain in 1.5 l MTM containing 208 or

200 g l−1 glucose and 1.6 g l−1 NH4Cl. The dissolved oxygen tension (DOT) was 80 %. The pH

level was kept at 6 by automatic addition of NaOH, and thus no CaCO3 was present, enabling

the accurate determination of dry biomass weight.

Even though the U . maydis wt cells in the fermenter were cultivated in the same medium as

those in the shaken flask experiments, the discrepancy between the itaconate titer obtained

from the fermenter batch cultivation and the shaken flask experiments was large. The highest

itaconate titer produced by wt strain was obtained in the experiment dealing with the culture

aeration (10 ml culture, 3.2 g l−1 itaconate after 144 h). This is manifold less in comparison to

the fermenter batch cultivation. The reason could be the culture aeration, since it appears to

be the most influencing factor for the itaconate production. Apparently, the oxygen transfer

rate in the 10 ml culture is still much lower with regard to the transfer rate in the fermenter

and therefore only such low itaconate titer was obtained.

Big differences in the itaconate titers between the wt and the 2_29 strain were observed

(Figure 3.17A). Although the wt strain produced 37 g l−1 itaconate after 116 h, a decrease in

the production rate is obvious. While at the beginning the itaconate production rate was 0.8 g

l−1 h−1, at the end of the cultivation it dropped to 0.095 g l−1 h−1. The 2_29 mutant produced

29 g l−1 itaconate after 160 h, but with a rather stable production rate of about 0.21 g l−1 h−1.

Nevertheless, the wt strain reached a higher final itaconate titer and yield than the 2_29 strain.

There can be several reasons for that, e.g. the higher amount of biomass in the 2_29 strain cul-

ture (Figure 3.17B). Such a high quantity of biomass requires certain amount of maintenance

energy, which would negatively affect the itaconate production. Another reason could be that

the 2_29 strain was selected under conditions, which were at least to a certain extend oxygen

limited. If the mutation in strain 2_29 specifically affected the itaconate production under

low oxygen conditions, the improved production may be lost in a highly aerated fermenter.

In order to obtain further information about the influence of the culture aeration on the it-

aconate production under highly controlled conditions, fermenter batch cultivation with low

DOT (15 %) was performed. Even though this cultivation was only a preliminary experiment

and needs to be repeated in order to confirm the findings, some very interesting results were

47

Chapter 3: Results

obtained.

The cells of U . maydis wt and the 2_29 strain were cultivated under the same conditions as

above (1.5 l MTM containing 270 g l−1 glucose, 1.6 g l−1 NH4Cl, pH 6 and no CaCO3). As

mentioned previously, the only difference was the DOT, which was kept at about 15 %.

The difference in the itaconate production between the wt and the 2_29 strain was vast

(Figure 3.17C). While the 2_29 strain produced almost 20 g l−1 itaconate after 115 h, the

itaconate titer obtained from wt cells after the same cultivation time period was only 0.3 g l−1.

It appears, that the wt cells did not even start the itaconate production.

0

50

100

150

200

250

0

5

10

15

20

25

30

35

40

0 20 40 60 80 100 120 140 160

Glu

cose

(g

/l)

Dry

bio

mas

s w

eigh

t (g

/l)

Time (h)

0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0 20 40 60 80 100 120 140 160

0

5

10

15

20

25

30

35

40

Am

mo

niu

m (

g/l

)

Time (h)

ITA

(g

/l)

0

5

10

15

20

25

0

1

2

3

4

0 24 48 72 96 120

ITA

(g

/l)

Mal

ic a

cid

(g

/l)

Time (h)

0

50

100

150

200

250

300

0

5

10

15

20

25

30

35

0 24 48 72 96 120

Glu

cose

(g

/l)

Dry

bio

mas

s w

eigh

t (g

/l)

Time (h)

A B

D C

Figure 3.17: Fermenter batch cultivations with 80 % (A, B; n = 2) and 15 % DOT (C,

D; n = 1). (A) Ammonium consumption (dashed lines) and itaconate

production (full lines), (B) dry biomass weight (dashed lines) and glucose

utilization (full lines), (C) malate (dashed lines) and itaconate produc-

tion (full lines) and (D) dry biomass weight (dashed lines) and glucose

utilization (full lines). Green lines: U . maydis wt, blue lines: U . maydis

2_29. 5.10

48

3.6 Biosurfactant production and its influence on the itaconate production

A relatively high concentration of malate was produced alongside the itaconate. Even though

the titers were comparable among the both strains (about 3 g l−1), when compared to the

itaconate concentrations, the differences became apparent. The ratio itaconate/malate in

the case of the wt strain is 0.14, for the mutant 5.5. Furthermore, in contrary to the previous

findings, the malate production started later, parallel to the itaconate production.

While there were differences in the glucose utilization among the strains especially in the

second half of the cultivation, the biomass production from both the strains was comparable

(Figure 3.17D).

The results from these two fermenter batch cultivations confirmed the necessity of the proper

culture aeration for high-yield itaconate production as well as the fact, that also in the case of

fermenter batch experiments, the condition optimization should be performed with regard

to the mutant strain.

3.6 Biosurfactant production and its influence on the

itaconate production

The production of surface active compounds by the cells of U . maydis is well known [38].

The determination of the surface tension of the culture broth from cultivation experiments

revealed that all three selected mutants produced glycolipids under the conditions men-

tioned above (data not shown). Production of oil droplets was also observed by microscopy

(Figure 3.18). Under optimal conditions, up to about 23 g l−1 of these surface active com-

pounds can be produced [39], indicating that the synthesis of glycolipids as side products can

represent a major obstacle for the itaconate production, because it is energy and resources

demanding.

In order to investigate the extent of influence of the glycolipid synthesis on the itaconate pro-

duction, cultivation experiments with the wt and the U . maydis ∆cyp1 ∆emt1 strain (kindly

provided by Prof. Michael Bölker, Phillips University Marburg; see Materials and methods,

Table 2.1) were conducted. The cyp1 and emt1 are genes responsible for the production of

ustilagic acid and MEL respectively in U . maydis [39].

The cells were cultivated in 10 ml MTM containing CaCO3, 65 g l−1 glucose and 0.8 g l−1

NH4Cl. The kinetics of the itaconate production, ammonium utilization, cell growth, and

glucose utilization were studied (Figure 3.19).

The itaconate production was significantly increased in the ∆cyp1 ∆emt1 strain with regard

to the wt (about 2-fold, Figure ??A). Moreover, the production started after the N depletion,

which occurred within the first 24 h, only in the case of ∆cyp1 ∆emt1 strain.

49

Chapter 3: Results

Figure 3.18: Microscopic view of U . maydis 6_17 mutant without staining (A, 630 x)

and after the lipid-specific staining with Sudan black B and Safranin (B,

1000 x). Some lipid droplets are marked with arrows.

Figure 3.19: The kinetics of the ammonium utilization and the itaconate production

(A, dashed lines - ammonium concentration, full lines - itaconate produc-

tion) and cell growth and glucose utilization (B, dashed lines - biomass

production, full lines - glucose utilization) under selected optimal cul-

tivation conditions. Green lines: U . maydis wt, purple lines: U . maydis

∆cyp1 ∆emt1. The values are mean of three analytical determinations ±standard deviation. 5.11

The ∆cyp1 ∆emt1 strain grew much better in comparison to the wt, probably because this

mutant did not need to invest the gained energy into the glycolipid production (Figure ??B).

The deletion of the glycolipid production represents a great advantage for the itaconate

50

3.7 Alternative carbon sources for the itaconate production

production. The cells can use their energy for the itaconate production rather than for

biosurfactant synthesis. Moreover, the absence of the glycolipids in the culture broth may sig-

nificantly simplify the downstream processing, i.e. the membrane filtration or centrifugation

and thus reduce the overall production costs.

3.7 Alternative carbon sources for the itaconate

production

The ability to utilize the products of hydrolyzed biomass would be a great advantage in order

to establish U . maydis as a successful industrial itaconate producer. Therefore utilization of

L-arabinose, xylose and sucrose as well as itaconate production from them by various strains

of U . maydis was studied.

The cells of U . maydis wt and the 6_15, 6_17 and 2_29 mutants were cultivated in 50 ml

medium in 1 l flasks. The MTM contained 1.6 g l−1 NH4Cl, CaCO3 and 240, 260 or 240 g l−1

L-arabinose, xylose or sucrose.

The itaconate titers obtained from these alternative carbon sources were significantly lower

in comparison to the glucose (Table 3.7). One reason could be the cultivation conditions,

which may have not been optimal. Fermenter batch cultivation experiments performed by

other members of our research group revealed that the use of mixed carbon sources leads to

efficient production of itaconate (Dr. Nicole Maaßen, personal communication).

The cell growth on the different carbon sources was comparable. With regard to glucose, the

cells were able to produce even more biomass from these alternative carbon sources. The

end-point OD values obtained after 144 h are summarized in Table 3.7. While about 60 g l−1

arabinose and 28 g l−1 xylose was utilized by the cells over the 144-h cultivation, the cells used

about 241 g l−1 sucrose.

The U . maydis cells were able to utilize L-arabinose, xylose and sucrose, the two first being

products of biomass hydrolysis. Although the culture conditions clearly need to be optimized,

the use of these alternative carbon sources is possible.

51

Chapter 3: Results

Carbon

source

U . maydis

wt

U . maydis

6_15

U . maydis

6_17

U . maydis

2_29

OD (AU)

Arabinose 15 ± 2 26 ± 1 27 ± 2 23 ± 2

Xylose nd 18 ± 2 18 ± 2 15 ± 0.4

Sucrose 21 ± 2 24 ± 1 24 ± 1 17 ± 1

ITA (g l−1)

Arabinose 0.016 ± 0.002 0.017 ± 0.002 0.013 ± 0.0003 0.015 ± 0.001

Xylose 0.023 ± 0.0007 0.103 ± 0.019 0.176 ± 0.025 0.054 ± 0.012

Sucrose 0.091 ± 0.01 0.148 ± 0.021 0.212 ± 0.015 0.123 ± 0.013

Table 3.7: The summary of the end-point values of OD (AU) and itaconate (g l−1) from

various carbon sources. The values (mean of three analytical determina-

tions ± standard deviation) were obtained after 144 h of cultivation; ITA -

itaconate, nd - not determined.

3.8 Deletion of UM11778 in U. maydis

Results obtained from previous experiments revealed that the UM06344 gene does not func-

tion as a cis-aconitate decarboxylase, as was originally assumed.

Therefore, another gene putatively involved in the itaconate synthesis pathway, UM11778,

was selected by Prof. Michael Bölker and an U . maydis UM11778 knockout was constructed

by his work group and kindly provided for further analyses.

The basic local alignment search tool analysis (BLAST; [61]) of the protein encoded by

this gene revealed that it belongs to the prpF superfamily and it has 33% identity to a 3-

methylitaconate-∆-isomerase (Mii) from Eubacterium barkeri. This enzyme, among other

functions, catalyzes the isomerization of itaconate to citraconate [62].

The strains of U . maydis wt and the U . maydis ∆UM11778 were cultivated in 50 ml medium

in 1 l flasks. The MTM contained 65 g l−1 glucose, 1.6 g l−1 NH4Cl and CaCO3.

Fumarate, itaconate, glucose, succinate, malate and pyruvate were detected in the culture

broth. The production kinetics of these substances are shown in Figure 3.20.

There were great production differences between the∆UM11778 strain and the wt strain. The

deletion of the UM11778 gene led to a significant decrease in the itaconate production (0.014

g l−1 after 192 h), indicating that this gene plays a crucial role in the itaconate biosynthesis.

Still, a clearly defined peak of the itaconate was visible in the chromatogram (Figure 3.21).

The itaconate titer obtained from the wt strain was about 330-fold higher than the itaconate

titer from the ∆UM11778 strain.

52


Figure 3.20: The production kinetics of various substances putatively involved in the

itaconate synthesis pathway. Green lines - U . maydis wt, brown lines -

U . maydis ∆UM11778; n = 1. 5.12

The Figure 3.20 shows, that all the TCA cycle intermediates were produced in significantly

lower concentration by the ∆UM11778 strain with regard to the wt. The pyruvate concentra-

tion, conversely, was higher in the ∆UM11778 strain after 24 hours.

Apart from the peaks of the identified compounds, a peak of unknown substance eluting at

53

Chapter 3: Results

13.4 min was present in the culture broth of ∆UM11778 strain during the whole experiment

(Figure 3.22). A similarity with the terreic acid obtained by Bonnarme et al. [25] is possible.

The cultivation experiments with the ∆UM11778 strain revealed, that this gene is indeed

strongly involved in the itaconate synthesis pathway of U . maydis. This finding represent a

significant step forward in understanding the process of the itaconate synthesis.

Figure 3.21: HPLC profile of culture broth of U . maydis ∆UM11778 strain obtained

after 192 h of cultivation. A clearly defined itaconate peak is visible at

14.4 min.

54


Figure 3.22: HPLC profile of supernatant from 2-day-old culture of U . maydis

∆UM11778 strain. A peak of unidentified compound is visible at 13.4

min.

55

Chapter 3: Results

56

CHAPTER 4

Discussion

The aim of this work was to characterize the itaconate production in U . maydis, understand

the itaconate synthesis pathway and describe conditions and aspects which may influence

the itaconate yields.

U . maydis strain MB215 is presented as an alternative itaconate producer and compared

to the currently most used one, A. terreus. The great advantage of U . maydis with regard

to A. terreus is its unicellular growth. However, the production characteristics are for the

time being much lower than those from A. terreus. Therefore, various aspects putatively

influencing the itaconate production were studied within this work in order to improve them.

Genes which might be involved in the synthesis pathway of itaconate were either deleted

and/or overexpressed and the consequent cellular phenotype was examined. Various initial

carbon and nitrogen source concentrations as well as oxygen supply were investigated in

order to understand their impact on the itaconate yields. Moreover, generation of various

side products and their influence on the itaconate yields was also analyzed.

4.1 Itaconate synthesis pathway in U. maydis

In A. terreus the itaconate is synthesized via Embden-Meyerhof-Parnas (EMP) pathway fol-

lowed by the initial steps of the tricarboxylic acid (TCA) cycle. The last step of the pathway

involves the conversion of cis-aconitate to itaconate (Figure 4.1, cis-aconitate route). This

reaction is catalyzed by cis-aconitate decarboxylase (EC: 4.1.1.6). Bonnarme et al. [25] proved

by employing 14C-labelled TCA cycle intermediates, that A. terreus uses this pathway to syn-

thesize the itaconate. The detected radioactivity levels corroborate the cis-aconitate route

but exclude the citraconate route (Figure 4.1). These experiments also excluded the pentose

57

Chapter 4: Discussion

phosphate pathway, leaving the EMP route the only one used for processing the glucose

during itaconate synthesis in A. terreus. These findings were supported by detection of most

of the TCA cycle intermediates in the culture broth.

The itaconate synthesis pathway in U . maydis remains unclear. There are three possible

routes for itaconate production in U . maydis.

The first route includes the condensation of two acetyl-CoA molecules and further production

of 1,2,3-tricarboxypropanoate from succinate [63]. This pathway was excluded for A. terreus

by Bonnarme et al. [25] based on the disproportion of the itaconate yields. Although this

route is considered highly unlikely due to the inefficiency of the pathway and the fact that

there has not been any further literature dealing with this route published, it can not be

completely excluded for the itaconate production of U . maydis. The theoretical itaconate

yield of this pathway is 56 Cmol %. This value was calculated based on the stoichiometry of

the following reaction:

1.5 glucose → 3 CO2 + 3 acetyl-CoA → 4 CO2 + 1 itaconate.

The second possible pathway (Figure 4.1, cis-aconitate route) is similar to the pathway in

A. terreus, via the conversion of cis-aconitate to itaconate, which is catalyzed by cis-aconitate

decarboxylase (CAD). Based on a blast analysis between the protein sequence of the CAD

from A. terreus [27] and the U . maydis genome, the UM06344 gene was selected as a putative

CAD in U . maydis. Both of these genes contain similar conserved domains.

If the UM06344 gene does indeed encode a CAD enzyme, the deletion of this gene should have

led to decreased itaconate production, which was not the case. Moreover, the itaconate titers

from the deletion mutant were in some cases even higher than the those from the wt strain.

Thus, although a certain influence of this gene on the itaconate production was observed,

the findings suggest that this gene is not the cis-aconitic acid decarboxylase responsible for

itaconate production.

The overexpression of UM06344 or random integration of the overexpression construct led

on average to 1.2-fold higher itaconate titers with regard to the wt, although a wide range of

titers was observed. This suggests that the UM06344 overexpression construct influences the

itaconate production only to a small extent.

Three consequences of the overexpression of the UM06344 gene or random integration of the

overexpression construct simultaneously affect the final cell phenotype;

• An increase in the itaconate production due to the overexpression of the UM06344 gene

itself, as originally expected. However, since the deletion of UM06344 gene did not result in

decreased itaconate production, this gene is not directly involved in the itaconate production.

58


Glucose

Acetyl-CoA Pyruvate

CO2

Oxaloacetate

CO2

R-citramalate

Itaconate (Citraconate route)

Cis-aconitate

Itaconate (Cis-aconitate route)

CO2

Cis-aconitic acid decarboxylase

Citraconate

Mii (UM11778)

R-citramalate synthase

R-citramalate dehydratase

EMP pathway

Figure 4.1: Putative itaconate synthesis routes in U . maydis and A. terreus (figure was

adapted from [25]); Mii - 3-methylitaconate-∆-isomerase, EMP - Embden-

Meyerhof-Parnas pathway.

59


• As no terminator for the UM06344 ORF was present in the overexpression construct, the

termination was not efficient and therefore the gene expression did not reach its putative

maximum. However, if the overexpression cassette integrated into the U . maydis genome

in front of a terminator, the expression of the UM06344 gene would be greatly enhanced.

Such arrangement would contribute to the variation of the itaconate titers obtained from the

transformant screening.

• A gene knock-out as a result of the overexpression construct integration into the U . maydis

genome. If a gene involved in the itaconate production pathway would be disrupted, the

itaconate production would be either decreased or obstructed. However, a disruption of

another gene involved in a competing route might as well lead to an increase of the metabolic

flux through the itaconate synthesis pathway.

Therefore, it is difficult to distinguish between the outcome resulting from the UM06344

overexpression and the random integration of this construct and the consequent mutation.

The latter two points can have either negative or positive effect due to the integration locus.

Moreover, the second point strongly influences the first one.

A further BLAST analysis of the protein encoded by the UM06344 gene revealed a 96%

identity to 2-methylcitrate dehydratase (PrpD, EC: 4.2.1.79) from Sporisorium reilianum.

Furthermore, an identity to the 2-methylcitrate dehydratase higher that 60 % was found

among wide range of microorganisms, e.g. Puccinia graminis, several strains of Burkholderia,

Salmonella enterica and E. coli. The PrpD of E. coli showed the highest affinity to 2-methylcit-

rate and citrate and no affinity to cis- or trans-aconitate [64]. Thus, overexpression of

UM06344 or random integration of the overexpression construct could increase the rate

of the dehydration of the citrate to form the cis-aconitate, and therefore the concentration

of the substrate for the next step in the itaconate synthesis pathway, the conversion of the

cis-aconitate into itaconate [65]. This may explain the slight overall improvement of itaconate

production resulting from introduction of the UM06344 overexpression construct. However,

this applies only if U . maydis employs the cis-aconitate route (Figure 4.1) for the itaconate

production.

Nowakowska-Waszczuk proposed yet another possible itaconate production route for A.

terreus, which involves the production of citramalate from pyruvate and acetyl-CoA (Fig-

ure 4.1, citraconate route; [66]). This assumption originates from the experiment with isolated

mitochondria from cell extracts from A. terreus. These preparations were unable to oxidize

the TCA cycle intermediates and therefore Nowakowska-Waszczuk concluded, that the it-

aconate synthesis pathway most likely does not involve the TCA cycle, but rather occurs

via dehydration of citramalate. However, this pathway for itaconate production in A. terreus

was excluded by the findings by Bonnarme et al. [25] based on the 13C and 14C labeling

experiments. Nevertheless, for U . maydis this route should still be taken into consideration.

Velarde et al. [62] described an enzyme catalyzing the isomerization between 3-methylitacona-

60


te and 2,3-dimethylmaleate in Eubacterium barkerii. This enzyme, 3-methylitaconate-∆-

isomerase (Mii, EC 5.3.3.6), also catalyzes the reversible conversion between itaconate and

citraconate. The BLAST analysis between the protein sequence of Mii and the U . maydis

genome resulted in three putative 3-methylitaconate-∆-isomerases (UM11778, UM06058 and

UM02807). Although the sequence similarities are very low (33 - 35 %), a conserved domain

search classifies them into the same PrpF superfamily as the Mii enzyme. These three genes

were deleted in the U . maydis MB215 wt strain by Prof. Michael Bölker and kindly provided

for phenotypic characterization.

Since the deletion of the UM11778 gene almost completely disrupted the itaconate produc-

tion, the respective enzyme plays a crucial role in the itaconate synthesis pathway. This

finding could indicate that the citraconate route is in U . maydis active. However, a clearly

defined peak was visible in the HPLC profile of the culture broth from this strain (Figure 3.21).

Although the deletion of the other two genes, UM06058 and UM02807, did not result in

any detectable decrease in itaconate production, the residual itaconate production in the

UM11778 knockout could be explained by the activity of the enzymes encoded by either one

of these two genes or even the UM06344 mentioned above.

As described previously, Mii catalyzes the isomerization between itaconate and citracon-

ate. In theory, citraconate can be produced by dehydration of R-citramalate by the enzyme

R-citramalate dehydratase (EC: 4.2.1.35). This R-citramalate in turn can be formed from pyr-

uvate and acetyl-CoA, catalyzed by R-citramalate synthase (EC: 2.3.1.182). This citraconate

route is summarized in Figure 4.1.

A BLAST analysis of a well characterized R-citramalate dehydratase from Methanocaldococcus

jannaschii revealed a putative R-citramalate dehydratase in the U . maydis genome (UM06010).

However, this gene is annotated as 3-isopropylmalate dehydratase (EC: 4.2.1.33). The max-

imal identity between these two proteins is only 35 %. Rubin et al. [67] characterized the

UM06010 gene as a LEU1 encoding the isopropylmalate isomerase. Therefore, this gene can

be excluded from the genes participating in the citraconate pathway.

The presence of the R-citramalate synthase gene in the genome of U . maydis was not con-

firmed since the BLAST analysis of this protein from Sulfobacillus acidophilus resulted in two

hypothetical proteins; UM04115 and UM04156. The BLAST analysis of the protein encoded

by the first gene resulted in 97% identity with a LYS20-homocitrate synthase (EC: 2.3.3.14) of

Sporisorium reilianum and 93% identity with the same enzyme from Ustilago hordei. Since

this enzyme is involved in the lysine biosynthesis, the role of the UM04115 gene in the itacon-

ate synthesis pathway is not likely. The BLAST analysis of the protein encoded by the second

gene showed a 89% and 88% identity to the LEU4-2-isopropylmalate synthase (EC: 2.3.3.13)

from S. reilianum and U . hordei, resp. The LEU4-2-isopropylmalate synthase is a part of the

biosynthesis route of leucine and similarly to the LYS20-homocitrate synthase, it is also not

likely to be involved in the itaconate synthesis route.

61


Since the itaconate is not essential for the viability of U . maydis, it is considered a secondary

metabolite. The secondary metabolites are usually produced under specific cultivation con-

ditions and support the cells in adjusting to the environmental fluctuations or in competition

with other microorganisms [68]. Secondary metabolism genes are often arranged together in

clusters and coregulated [69].

A gene cluster for another secondary metabolite of U . maydis, the ustilagic acid, has been

identified and described [68]. The expression of this gene cluster is induced under N starva-

tion, leading to extensive production of ustilagic acid. Since itaconate synthesis is triggered

among other by N starvation, it might be possible, that a similar gene cluster responsible for

itaconate synthesis and involving the UM11778 gene is present in the genome of U . maydis.

The R-citramalate dehydratase and the R-citramalate synthase genes might be located in

such cluster. However, further investigations are necessary.

To summarize the findings and the assumptions about the itaconate synthesis pathway, it is

very difficult to identify which particular pathway U . maydis uses for the itaconate synthesis.

The UM11778 gene has a significant influence on the itaconate production, but putative

genes encoding the remaining enzymes in the citraconate route could not be identified.

Thus, the presence of the remaining enzymes, R-citramalate dehydratase and R-citramalate

synthase, in U . maydis is so far only speculative. Although the citraconate pathway may

exist, due to the insufficient identification of the genes involved in it, further investigation is

needed.

The cis-aconitate pathway can also be neither dismissed nor confirmed. Although several

of its intermediates were detected in the culture broth of U . maydis cells, the activity and

consequently the involvement of the key enzyme of this pathway, the putative U . maydis

CAD encoded by the UM06344 gene, remains unclear.

The theoretical itaconate yields from both of these pathways are 83 Cmol %.The calculation

is based on the theoretical itaconate yield and the stoichiometry of the following reaction,

which summarizes both of the itaconate synthesis pathways (Figure 4.1):

1 glucose → 1 CO2 + 1 itaconate.

Since the highest itaconate yield obtained during the cultivation experiments was only 38

Cmol %, both these routes can be taken into consideration.

However, without further investigation, no unequivocal conclusion concerning the itaconate

synthesis pathway in U . maydis can be drawn.

62

4.2 Factors influencing the itaconate production


The biotechnological production process of itaconate is greatly influenced by the cultivation

conditions. Therefore, their optimization may lead to an increase in the itaconate yields.

Hence, U . maydis wt and the three selected high-yield itaconate mutants of U . maydis (strains

6_15, 6_17 and 2_29) were phenotypically characterized under various cultivation conditions.

The most promising parameters for optimization in order to obtain an economically feasible

production process were chosen to be the initial glucose and ammonium concentration

and the oxygen supply. The carbon source usually represents the highest expense within

the fermentation process, while as mentioned above, N depletion can act as an itaconate

synthesis trigger. However, the most important factor influencing the itaconate synthesis

might be the oxygen supply as it is known for example for the citric acid production [70]. The

influence of these factors is interconnected.

Theoretical maximum yields and production rates of itaconate depend on the initial glucose

and ammonium concentrations as shown in Figure 4.2. Numbers were estimated based on

the experiment dealing with various initial glucose concentrations (Figure 3.8). The theoret-

ical maximum itaconate yields were calculated according to the following formula:

ycomp = ymax∗ (g l ucosei ni t − g l ucosebi om)

(g l ucosei ni t )= ymax∗ (g l ucosei ni t − ( f ∗ammoni umi ni t ))

g l ucosei ni t,

where ycomp (Cmol %) is the theoretical maximum yield (%) compensated for biomass pro-

duction, ymax is theoretical maximum yield (83 %) based on the stoichiometry of the reaction

from glucose to itaconate and CO2, glucosei ni t is the initial glucose concentration (g l−1),

glucosebi om is the glucose concentration used for biomass production (g l−1), f is a factor

related to the C/N ratio of the biomass estimated from Figure 3.8 (utilized glucose (g) per

utilized ammonium (g) in the first 24 h of cultivation) and ammoniumi ni t is the initial am-

monium concentration (g l−1).

The itaconate production rates were calculated from formula below:

rp,max = rp,spec ∗ NH4+

i ni t ,

where rp,max is the theoretical itaconate production rate (g l−1 h−1) at a specific nitrogen

concentration, rp,spec is the specific itaconate production rate estimated from Figure 3.8 (g

ITA l−1 h−1 g−1 NH4+) and NH4

+i ni t is the initial ammonium concentration (g l−1). Great

disproportion can be observed in the production rates between the strains and therefore two

63


different rp,spec were used. The itaconate production rate of U . maydis wt (Figure 4.2, black

full line) is almost 2-fold lower than that of the 2_29 mutant (Figure 4.2, black dashed line).

Figure 4.2: Theoretical yields and production rates of itaconate vs. ammonium con-

centration by U . maydis wt (black full lines) and 2_29 mutant (black

dashed lines) strains. Yields are based on the initial glucose concentration;

green - 65 g l−1, red - 110 g l−1, yellow - 250 g l−1 glucose. Arrows indicate

the NH4+ concentrations tested (0.25, 0.52 and 1.35 g l−1).

4.2.1 Nitrogen

The concentration of the N source represents an important factor influencing the itaconate

production. N is essential for the cells, amino acids and consequently proteins, as well as

polyamines, nucleic acids and vitamins being the products of its utilization. Moreover, nitro-

gen source depletion represents a common trigger for production of various biotechnological

products (citric acid by A. niger [71], [72] or lipids by oleaginous yeast [73]).

There are several substrates used in order to supplement the cells with N. The most common

ones are nitrate, nitrite, ammonium and urea. While not all microorganisms are capable of

nitrate utilization, U . maydis possess this ability, although the presence of ammonium ions

drastically decreases the activity of the nitrate reductase [74].

The cells in the cultivation experiments throughout this work were supplemented with NH4Cl.

However, NH4NO3 is used in most cases for the itaconate production by A. terreus [75], [76],

64


[29], [26] due to its less significant effect on the culture pH [47].

The optimal N initial concentration should be selected based on the specific preferences

for the production process. In theory, lower initial N concentrations lead to high yields but

very low production rates, while high N concentrations lead to low yields and high produc-

tion rates (Figure 4.2). However, with increasing initial N concentration, the culture quickly

approaches oxygen limitation due to the high biomass production. Therefore, an optimal

combination of glucose and N concentrations should be selected. As an example, low glucose

concentration would not be suitable in combination with high concentration of ammonium,

because all the glucose would be utilized for biomass production.

In the cultivation experiments, the highest itaconate yields were achieved with low N con-

centration, which is in contradiction to the theoretical expectations (Figure 4.2). Under the

conditions used, higher N concentration likely led to a high biomass formation and con-

sequently to oxygen limitation, which prevented the cells from efficient itaconate production

(Figure 3.11).

4.2.2 Glucose

As shown in Figure 4.2, high glucose concentrations are more favorable, since higher itaconate

yields can be achieved. However, high glucose concentrations can have negative effects of the

cells. Moreover, the highest expense within the itaconate fermentation process represents

the carbon source [77]. Therefore, optimization of the glucose concentration is necessary.

High glucose concentrations (about 250 g l−1) significantly decreased the level of cell meta-

bolism, e.g. the ammonium depletion occurred at a lower rate, growth rate decreased and

hence also the itaconate production was reduced. Most likely the reason for this is the high

osmotic pressure caused by the presence of the glucose and the high energy demand re-

quired to sustain the cell integrity in such environment. The hyperosmotic environment

itself represents a stress and nutrient shortage just adds to that [78]. Such finding is not

completely unexpected, because the cellular adaptations (e.g. up- or down-regulations of

metabolic pathways, switching on and off sensing systems) to an environment containing

nutrients in growth-limiting concentrations are closely related to responses to stresses such

as a hyperosmotic environment. As an example, Diano et al. [79] reported a slower growth

of A. niger when a high carbon source concentration was used (100 g l−1 glucose or 90 g

l−1 xylose). During the cultivation experiments performed within the present work, large

differences in the metabolic rate based on the various initial glucose concentrations (65 or

250 g l−1, respectively) were observed (Figure 3.10).

A significantly higher itaconate yield was obtained in the low-glucose-concentration environ-

ment (65 g l−1). Since the osmotic pressure was much lower, also the level of cell metabolism

65


was substantially increased. This finding is in contrast to the theoretical yield estimations

(Figure 4.2), where a higher glucose concentration would be more favorable for itaconate

production. In the presence of the high glucose concentration, the osmotic pressure in

addition to the oxygen limitation cause the decrease in the itaconate production. Until this

issue has been resolved, lower glucose concentrations should be applied for itaconate pro-

duction with U . maydis. However, rather low titers can be reached with these concentrations.

Such difficulty can be overcome by alternative fermentation strategies such as fed-batch or

continuous cultures.

While the glucose concentration in the production medium for itaconate by A. terreus is usu-

ally between 130 and 160 g l−1 [80], [20], [81], [29], [75], [27], for U . maydis and the itaconate

production, the determined optimal sugar concentration was about 2-fold lower (65 g l−1).

This may suggest, that either A. terreus possesses better mechanisms for coping with osmotic

stress caused by the presence of high glucose concentrations than U . maydis, or that the

oxygen limitation itself prevents U . maydis from reaching high itaconate yields.

4.2.3 Oxygen supply

The importance of sufficient oxygen supply for production of organic acids generally [79] and

itaconate in particular is indisputable. The role of the oxygen in the process of the itacon-

ate synthesis is very well shown in the experiment with A. terreus performed by Gyamerah

[82], where a 5 min interruption of the oxygen supply during fermentation led to complete

cessation of the itaconate production. The process was only slowly renewed after 24 h. The

same results were obtained by Guevarra et al. [83] with Ustilago cynodontis. The findings

from the experiments concerning the culture aeration within this work with U . maydis are

in agreement with the literature. Cells of U . maydis cultivated in higher aerated shake flasks

(10 ml culture/1 l flask) produced up to 6.5-fold more itaconate than those in poorly aerated

ones (100 ml culture/1 l flask). A DOT of 20% in the cultivation experiments with A. niger

is considered to be a critical value, under which the culture is oxygen-limited [79]. Such

situation can be recognized by decreased oxygen uptake rate and consequently also lowered

CO2 production rate. As a result, the glucose utilization rate diminished as well as the biomass

yield. A very similar situation was observed during the cultivation of the U . maydis cells in the

poorly aerated shaken flask cultures within this work. Further experiments and calculations

revealed that the cells cultivated in the poorly aerated cultures depleted the total available

oxygen within the first few hours and during the rest of the several-day experiment the cells

suffered from lack of oxygen. On the other hand, the cells cultivated in the higher aerated

media had more than enough available oxygen and did not experience the stress of oxygen

depletion.

66


Nevertheless, a limitation in the oxygen transfer rate was most likely present, because Kle-

ment et al. [33] obtained approximately 20 g l−1 itaconate with the wt strain of U . maydis.

This is substantially more than the itaconate titers obtained with the U . maydis wt strain

within this work (0.2 g l−1 itaconate) under otherwise comparable conditions. This 20 g l−1

itaconate was reached although the ratio of flask volume to culture volume was relatively

low (12.5). However, these cultures were agitated at a speed of 300 rpm, which significantly

increased the oxygen transfer rate, while all the cultivation experiments within this work were

performed at a speed of only 100 rpm. Therefore, it is likely that even the ‘high’ aeration

shaken flasks used in this study were to some extent limited in the oxygen transfer rate.

During glycolysis and the consequent itaconate or citrate synthesis routes (the synthesis and

the production conditions are very similar for both of these organic acids), NADH and ATP

are produced in excess (Figure 4.3; [70]). A high internal concentration of ATP may have

an inhibitory effect on the enzymes involved in glycolysis [28], and an excess of NADH can

cause metabolic cessation [84]. Since ATP and NADH are no longer used for cell growth

during itaconate production (in the case of U . maydis after the N or P exhaustion), the it-

aconate production leads to excess of these cofactors, which must be regenerated (Figure 4.3).

Figure 4.3: Putative possibilities of cofactor regeneration in U . maydis.

During the citric acid production process in A. niger one of the two major cofactor regenera-

tion routes includes an alternative (ubiquinol) oxidase (AOX; [28]). The main ability of this

terminal oxidase is to bypass the proton-pumping cytochrome pathway. Consequently, the

final ATP yield is lower with regard to the full length pathway, while the NADH regeneration

remains preserved. This is advantageous mainly when ATP is already in excess (during the

organic acids production).

67


U . maydis possess a gene encoding such alternative oxidase (UM02774) [85]. Thus, under

sufficiently aerobic conditions, the cofactors are regenerated via the respiratory chain, mak-

ing a sufficient oxygen supply in U . maydis cultures an important factor in the process of

itaconate production. Therefore this oxidase activity represents a very important factor in the

maintenance of the cofactor balance, and can be considered as the physiological basis for the

strong relation between oxygen supply and itaconate production.

Under anaerobic or oxygen limited conditions, the cofactors can be regenerated via various

routes. For S. cerevisiae it is the production of ethanol and glycerol [84], for E. coli ethanol

and a mixture of organic acids (e.g. lactate, acetate, formate [86]) and for A. niger mannitol or

other polyols [79]. While S. cerevisiae does not use malate production in order to regenerate

the NADH [84], for example Corynebacterium glutamicum does [87]. The main NADH regen-

erating reaction is catalyzed by a cytoplasmic malate dehydrogenase (EC: 1.1.1.37). However,

such regeneration is efficient only when malate is produced in cytosol from oxaloacetate [24]

since the malate production via the TCA cycle leads to further cofactor imbalance.

U . maydis possesses two genes encoding probable malate dehydrogenase (UM00403 and

UM04919). Since the shaken flask experiments were conducted under certain oxygen limita-

tion and large concentrations of malate were detected in the cultures, it is quite likely that

also U . maydis is capable of the cofactor regeneration via malate production [88]. Moreover,

this finding suggests, that the cis-aconitate route for the itaconate synthesis (Figure 4.1) might

be active in U . maydis, since malate is one of the key intermediates in the proposed pathway

[24]. Therefore, malate production can be used as an indicator for cofactor imbalance or

oxygen limitation during the production of itaconate by U . maydis.

Alternatively, lipid production represents yet another possibility to regenerate ATP and NADH

[89]. In comparison to U . maydis, A. terreus is not capable of lipid synthesis. The acetyl-CoA

used for fatty acid synthesis is transported from mitochondria, where it is produced, to the

cytosol in the form of citrate. Consequently, the citrate is cleaved back to oxaloacetate and

acetyl-CoA (citrate/malate cycle). This reaction is catalyzed by ATP-citrate lyase (EC: 2.3.3.8),

an enzyme present in oleaginous microorganisms. NADH aside of ATP is being recycled

in a cytosolic pyruvate regeneration cycle consisting of pyruvate, oxaloacetate and malate,

which essentially produces NADPH at a cost of NADH and ATP [89]. Acetyl-CoA from the

citrate/malate cycle together with the NADPH enables the consequent lipid biosynthesis.

The BLAST analysis between the protein sequence of ATP-citrate lyase from Aspergillus oryzae,

a typical oleaginous fungus, and the U . maydis genome resulted in a gene encoding a hypo-

thetical protein (UM01005). Consequent protein BLAST analysis of this putative U . maydis

ATP-citrate lyase revealed up to 59% identity to lyases of various microorganisms. Based

on all the previously listed findings, it is very likely that U . maydis regenerates the cofactors

also via the lipid production. However, the first choice for the cofactor regeneration during

the itaconate production is the respiratory chain, followed by both the malate and the lipid

68


synthesis.

The production of surface active compounds by fungi and namely by Ustilago is known for a

long time and very well described [90], [39]. The cultivation experiments performed within

this work with U . maydis MB215 wt and ∆cyp1∆emt1 (U . maydis MB215 ∆cyp1 crossed with

U . maydis FB1 ∆emt1) revealed that disabling the glycolipid production has a positive effect

on the itaconate production. When the genes responsible for the glycolipid production in

U . maydis (cyp1 and emt1, [39]) were deleted, the itaconate titers as well as the biomass

production significantly increased in comparison to the parental strain. The inability of the

U . maydis cells to produce the lipids comes as an advantage in the process of the itaconate

production not only from these biosurfactants is not only favorable from an energetic point

of view, but also for the downstream processing.

As described previously the aeration of the culture is the most important factor influencing

the itaconate production. However, U . maydis mutant 2_29 is capable of significantly in-

creased itaconate production with regard to the other mutants as well as to the wt, even in the

poorly aerated cultures, whether it was the shaken flask culture experiments or the fermenter

batch cultivations with only 15 % DOT. Interestingly, fermenter batch cultivation with 80 %

DOT resulted in substantially higher itaconate titers from U . maydis wt strain in comparison

to the 2_29 mutant. Since this mutant was selected under oxygen limited conditions, it might

be capable of higher itaconate production selectively under oxygen limitation. However,

these fermentation experiments were preliminary and need to be repeated in order to exclude

any errors. The lower malate production as well as the ability of the 2_29 strain to sustain

its metabolism even under oxygen-limited conditions may imply that this strain possesses

alternative ways for NADH and ATP regeneration. This feature of the 2_29 strain might be

very important considering the putative application of this strain in the industrial itaconate

production, because it represents a decrease of the production costs due to the lower energy

demand required for aeration of the culture.

One of the most profound advantages of U . maydis with regard to A. terreus is its unicellu-

lar yeast-like form of growth. Cultivation experiments performed by Park et al. [81] with

A. terreus showed that since Aspergillus forms hyphae, it is very difficult to supply the culture

with sufficient oxygen. The filamentous growth of Aspergillus increases the viscosity of the

culture, which greatly influences the oxygen transfer into the medium and subsequently the

respiration rate of the cells. Moreover, a too high impeller tip speed in stirred batch cultures

leads to a shear stress and hyphal breaks from mechanical agitation. Consequently, the condi-

tions switch from aerobic to oxygen-limited, resulting in production of large concentrations

of various polyols, causing problems in industrial process [79]. Furthermore, A. terreus forms

pellets during its growth, and cells in the center of such pellet are exposed to conditions of

substantial oxygen limitation.

Thus, the hyphal growth pattern of Aspergillus leads to severe limitations in oxygen transfer,

69


a factor which is critical for itaconate production. Therefore, alternative ways to increase the

oxygen supply, such as decreased biomass concentrations, enriched oxygen supply [91] or

alternative reactor designs [92] need to be employed. However, all of these measures either

decrease the efficiency of itaconate production or add to the total operational cost.

Herein lies the main advantage of U . maydis as an alternative itaconate production host.

Since its cells exhibit a yeast-like unicellular growth pattern, the above mentioned difficulties

can be avoided, leading to an overall more efficient and reliable process.

4.2.4 Growth-limiting factors

For most microorganisms, the production of chemicals is coupled to growth. The most com-

mon example is S. cerevisiae and ethanol production. Therefore, in order to obtain high titers

of desired product, large concentrations of biomass, essentially an undesired by-product, are

produced as well [93]. Such phenomenon is not observed in U . maydis cultivations. After

exhaustion of N source, which occurs usually within the first 24 hours, the growth rate sig-

nificantly decreases and itaconate production starts (Figure 3.7). This represents rather an

exception among microorganisms, but also a great advantage, because after the first 24 hours

of cultivation, the utilization of the carbon source can be directed mostly to the production of

chemicals, without being used for unnecessary further biomass production.

In the cases of A. terreus [94], [59] and Pseudozyma antarctica [34], it is also N and P which

can trigger itaconate production. Interestingly, P is the more efficient trigger for the itaconate

production in A. terreus. In the cultivation experiments with A. terreus performed by Riscald-

ati et al. [95] the critical P concentration is 10 mg l−1. Consequently, the primary mycelia

growth is terminated and itaconate excretion starts as well as the secondary growth [59].

Secondary growth can be also observed in the U . maydis cultures [33]. After N depletion from

the medium, such growth is possible most likely due to the utilization of N stored within the

cells. However, secondary growth and consequent extra N utilization may lead to a reduced

concentration of proteins necessary for the itaconate production. This can be the reason for

the use of a P-limitation in the Aspergillus process.

4.3 Alternative carbon sources for itaconate production

Since the highest itaconate titers can be reached by utilizing glucose, it is currently the most

suitable carbon source for itaconate production. However, utilizing solely glucose is not

economically admissible due to its high price, and in bulk scale the use of edible substrates is

undesirable from a societal perspective. In order to find a suitable substituent, several other

70

4.4 Conclusions

carbon sources were investigated as putative substrates for the itaconate production.

One of the best substrates would be corn starch [20]. However, despite its low price, high

purity and stability in mass supply, its usage is rather limited. The reason is its gelatinization

during heat sterilization. This obstacle can be overcome by starch hydrolysis, although this

process is accompanied with several further difficulties [76], [20]. Other substrates commonly

used for the itaconate synthesis are sucrose, xylose [96] or glycerol [97]. Newly, also citric

acid has been tested as a putative substrate for the itaconate synthesis [92]. Citric acid with

its price being only 1/10 of the price of itaconate could prove itself to be an economically

interesting carbon source. However, its price still exceeds the price of glucose. Nonetheless,

the most frequently used substrates are beet and sugarcane molasses [32].

The most interesting substrates for the purposes of this work are those, which can be found

in hydrolyzed biomass. Not only are these carbon sources cheap, but most importantly, they

do not compete with common human diet sources. Aside of glucose, it is mainly xylose and

to a certain extent also arabinose, which are present in the hydrolysates in the highest con-

centrations [5]. However, the utilization of a single carbon source at a time (xylose, sucrose or

arabinose) by U . maydis resulted in significantly lower itaconate titers compared to glucose.

Begum et al. [98] showed, that combining two substrates in ratio 1:1 significantly increases the

citric acid titers by A. niger. Fermenter batch cultivations with mixed substrates conducted

within our institute with U . maydis cells resulted in similar findings. Solely used substrates

gave substantially lower itaconate titers, while mixtures of carbon sources led to much higher

itaconate concentrations (Maaßen et al., unpublished results).

Therefore, in order to employ these alternative carbon sources in successful high-yield it-

aconate production, they should be used in mixtures. This finding is also favorable from an

economical aspect, because separation and further purification of the substrates from the

hydrolyzed biomass is therefore not necessary.

4.4 Conclusions

This work describes U . maydis as an alternative itaconate producer to the current indus-

trial workhorse A. terreus. The itaconate titers from U . maydis do not yet reach those from

A. terreus. However, genes putatively involved in the itaconate synthesis pathway were se-

lected and respective U . maydis mutants created. The UM06344 gene, initially considered

as a putative cis-aconitate decarboxylase, is not essential for the itaconate production. On

the other hand, the UM11778 gene, putatively encoding 3-methylitaconate-∆-isomerase,

is crucial for itaconate production. This enzyme is thought to catalyze the last step in the

putative itaconate synthesis pathway, the conversion of citraconate to itaconate, and its

deletion greatly reduced the itaconate production. However, until the enzyme activity of

71


the enzyme encoded by UM11778 can be determined, a clear conclusion concerning the

itaconate synthesis pathway can not yet be drawn.

Nevertheless, high-yield itaconate mutants were obtained and the itaconate production

from these mutants was significantly improved with regard to the parental strain. These

mutants were selected under oxygen limited conditions, which is reflected by the ability of

mutant 2_29 to cope with low oxygen environments and the production. Even though some

of these improvements need to be further investigated and explained, the findings described

above brought certain insight into the itaconate production pathway. However, the route still

remains just putative.

Substantial improvements to the cultivation medium were employed. Contrary to theoretical

estimations, the most favorable glucose and NH4Cl concentrations were the lowest ones

tested. This is most likely due to oxygen limiting conditions in the shake flask experiments

used in this study.

Therefore, the influence of the various levels of oxygen supply on the itaconate production

as well as on the overall cell metabolism was investigated. As expected, culture aeration

represents a crucial factor in the final itaconate titers and the cell viability. From the three

aeration levels tested, the highest level proved to be the best. However, even the highly

aerated cultures in the shaken flask experiment are likely oxygen limited to a certain extend

and therefore further increase in the culture aeration could be beneficial.

Various carbon sources were tested as putative substrates for itaconate production, most

of them to be found in hydrolyzed biomass. Not only is U . maydis capable of utilization

of these sugars, moreover, it possesses the ability to form itaconate from these substrates.

However, further optimization is clearly required since the itaconate titers reached in these

experiments were very low.

Based on the findings presented within this work, U . maydis appears to be a suitable itacon-

ate producer due to its morphological qualities and the ability to produce itaconate naturally.

Nevertheless, the efficiency levels of the itaconate production of U . maydis do not meet those

of A. terreus. However, the previously stated results show, that improvement of the production

conditions and the host itself is possible and therefore, the U . maydis strain should ultimately

reach the efficiency levels of A. terreus.

4.5 Further recommendations

Although substantial improvements in itaconate production were achieved within this work,

there are still many aspects of both the process and the organism itself, which should be

further investigated and eventually employed.

Mutants with significantly improved itaconate production were obtained and several at-

72

4.5 Further recommendations

tempts with various methods (LILIS PCR - Ligated Linker Supported PCR or the recovery of

the integrated plasmid using PCR) were conducted in order to determine the integration loci

of the overexpression cassette without success. Clarification of the genetic background of

improved production in these mutants will be beneficial for consequent strain enhancement

and therefore new approaches, such as the whole genome sequencing of pooled mutants,

should be employed.

Although the phenotypic characterization of the mutants provided certain insight into the

itaconate synthesis pathway, it remains unclear via which biosynthetic route itaconate is

produced and whether there is no other pathway involved. The labeling of putative precurs-

ors of itaconate as described by Bonnarme et al. [25] would enable a definite conclusion in

this respect. Furthermore, several enzymatic steps such as R-citramalate and citraconate

production are thus far only theoretical, and therefore the identification of the genes and

enzymes responsible for these steps would be necessary.

For further U . maydis strain improvement, cell enhancements performed within this work

should be combined in one strain. The 2_29 mutant strain would be a good choice for a

parental strain, since its itaconate production is substantially increased and malate synthesis

reduced. In addition, the genes responsible for the glycolipid production in U . maydis (cyp1

and emt1) should be deleted in order to facilitate the downstream process and increase

carbon efficiency. If no lipids were produced, more glucose can be utilized for itaconate

production. Since the UM11778 gene plays a crucial role in the itaconate synthesis pathway,

its overexpression would likely also be highly beneficial as would also be the identification of

the other genes involved in the citraconate pathway. Further improvement of the itaconate

titers could also be obtained by identification of the itaconate transporter through the cell

wall and consequent overexpression of the gene encoding this protein.

As findings within this work revealed, the aeration of the culture represents a key factor

in the cell metabolism and consequently also in the itaconate yields. Nevertheless, even

in the highly aerated cultures within the shaken flask experiments, the oxygen supply was

to a certain extent limited. Therefore, further increase of the aeration of the cultures may

increase the itaconate yields. There are multiple ways to enhance the culture aeration; in

the fermenter, adding oxygen either directly into the culture or to the air stream is possible

as well as simple stirring with high speed. Since U . maydis exhibits an unicellular form of

growth, the negative effect of the shear stress does not need to be taken into consideration.

To ensure optimal oxygen supply in shaken flask experiment, increasing the agitation rate

should be sufficient.

Aside of the depletion of the N being the trigger for the itaconate synthesis, P may represent

another interesting growth limiting factor. However, experiments with low P and constant N

concentration would be necessary in order to verify this hypothesis as well as to pose a valid

comparison between N and P as the triggers for the itaconate production for U . maydis.

73


In order to establish a sustainable production process, alternative carbon sources should

be employed. As described above, sugars from hydrolyzed biomass should be utilized in

mixtures in order to reach high itaconate yields. However, both the optimal composition and

percentage of such mixture is yet to be determined.

Identification of the peak of the unknown substance visible in the HPLC profile of the

∆UM11778 strain might reveal further information about the itaconate production pathway

in U . maydis.

As mentioned above, current itaconate yields from U . maydis do not reach the same level as

those from A. terreus. However, applying these recommendations and further understanding

of the itaconate synthesis pathway will most certainly lead to the optimal production process

and consequently also to the desired itaconate titers and yields.

74

CHAPTER 5

Supplementary data

75

Chapter 5: Supplementary data

0h

24h

48h

72h

96h

120

h14

4h

Fig.

Aw

t0.

00.

02±

0.0

0.17

±0.

040.

2±

0.01

1.40

±0.

231.

75±

0.19

1.92

±0.

13

∆U

M06

344

0.0

1.23

±0.

076.

08±

0.25

9.85

±0.

3712

.38±

0.97

13.8

6±

1.03

13.8

3±

0.70

Fig.

Bw

tn

d0.

009

0.09

50.

345

0.40

20.

465

4.2

∆U

M06

344

nd

0.00

80.

20.

358

0.37

70.

483

5.05

Tab

le5.

1:C

omp

aris

onof

ITA

pro

du

ctio

n(g

l−1

)by

U.m

ayd

isw

tan

d∆

UM

0634

4st

rain

s;n

d-

not

det

erm

ined

(Fig

ure

3.2)

.

0h

24h

48h

72h

96h

120

h14

4h

ITA

(gl−

1)

0.0

0.01

3±

0.00

10.

02±

0.00

10.

05±

0.02

0.08

±0.

040.

13±

0.04

0.18

±0.

07

NH

4+

(gl−

1)

0.52

±0.

00.

06±

0.03

0.0

0.0

0.0

0.0

0.0

OD

(AU

)0.

4±

0.01

9.6±

0.44

16.9

±0.

818

.5±

1.05

16.5

±0.

9715

.5±

0.45

17.9

±0.

78

Glu

cose

(gl−

1)

251±

4.0

206±

5.1

196±

7.3

183±

4.7

170±

4.4

161±

6.1

152±

6.1

Tab

le5.

2:A

typ

ical

cult

ivat

ion

char

tofU

.may

dis

wt(

Figu

re3.

7).

76

Stra

in0

h24

h48

h72

h96

h12

0h

144

h

Fig.

A

ITA

(gl−

1)

wt

0.0

0.01

6±

0.00

20.

19±

0.04

60.

5±

0.11

1.17

±0.

212.

25±

0.38

3.1±

0.38

6_15

0.0

0.01

4±

0.00

10.

14±

0.03

0.35

±0.

090.

75±

0.22

1.22

±0.

361.

74±

0.55

6_17

0.0

0.02

6±

0.00

80.

56±

0.11

1.08

±0.

103

1.4±

0.06

1.76

±0.

032.

3±

0.13

2_29

0.0

0.01

2±

0.00

20.

76±

0.1

1.85

±0.

43.

08±

0.3

3.2±

0.3

4.18

±0.

34

NH

4+

(gl−

1)

wt

0.52

±0.

00.

14±

0.02

0.0

0.0

0.0

0.0

0.0

6_15

0.52

±0.

00.

1±

0.03

0.0

0.0

0.0

0.0

0.0

6_17

0.52

±0.

00.

11±

0.04

0.0

0.0

0.0

0.0

0.0

2_29

0.52

±0.

00.

08±

0.01

0.0

0.0

0.0

0.0

0.0

Fig.

B

ITA

(gl−

1)

wt

0.0

0.01

2±

0.00

20.

082±

0.01

40.

53±

0.17

2.0±

0.35

2.8±

0.29

3.15

±0.

13

6_15

0.0

0.01

±0.

0006

0.14

±0.

021

1.5±

0.00

64.

6±

0.09

7.9±

0.17

7.9±

0.29

6_17

0.0

0.01

1±

0.00

030.

15±

0.03

61.

0±

0.07

82.

75±

0.3

4.54

±0.

544.

8±

0.53

2_29

0.0

0.02

8±

0.00

31.

25±

0.29

3.9±

0.64

8.0±

0.8

10.0

4±

0.82

10.5

±0.

52

NH

4+

(gl−

1)

wt

0.52

±0.

00.

00.

00.

00.

00.

00.

0

6_15

0.52

±0.

00.

2±

0.0

0.0

0.0

0.0

0.0

0.0

6_17

0.52

±0.

00.

2±

0.01

0.0

0.0

0.0

0.0

0.0

2_29

0.52

±0.

00.

1±

0.0

0.0

0.0

0.0

0.0

0.0

Table 5.3: Influence of initial glucose amount on ITA production and ammonium

utilization by U . maydis wt and mutants (Figure 3.8).

77


Initial glucose amount (g l−1) Strain ITA (g l−1)

250

wt 3.1 ± 0.38

6_15 1.74 ± 0.55

6_17 2.27 ± 0.13

2_29 4.18 ± 0.34

110

wt 1.9 ± 0.32

6_15 2.9 ± 0.49

6_17 2.23 ± 0.32

2_29 5.67 ± 0.95

65

wt 3.15 ± 0.13

6_15 7.86 ± 0.29

6_17 4.8 ± 0.53

2_29 10.45 ± 0.52

Table 5.4: An overview of the influence of the initial glucose amount on ITA titers by

U . maydis wt and mutants (Figure 3.9).

78

Stra

in0

h24

h48

h72

h96

h12

0h

144

h

Fig.

A

ITA

(gl−

1)

wt

0.0

0.01

2±

0.00

040.

03±

0.00

730.

13±

0.03

80.

25±

0.08

0.4±

0.08

0.6±

0.15

6_15

0.0

0.01

1±

0.00

080.

016±

0.00

50.

08±

0.02

0.25

±0.

050.

47±

0.04

0.71

±0.

007

6_17

0.0

0.01

3±

0.00

20.

033±

0.01

80.

18±

0.06

70.

45±

0.22

0.83

±0.

441.

28±

0.49

2_29

0.0

0.01

1±

0.00

070.

25±

0.07

71.

52±

0.15

2.52

±0.

363.

07±

0.43

3.52

±0.

53

NH

4+

(gl−

1)

wt

0.52

±0.

00.

27±

0.01

0.2±

0.01

0.09

±0.

030.

02±

0.01

0.0

0.0

6_15

0.52

±0.

00.

32±

0.02

0.21

±0.

020.

08±

0.04

0.05

±0.

030.

03±

0.01

0.01

±0.

0

6_17

0.52

±0.

00.

31±

0.02

0.2±

0.05

0.09

±0.

050.

05±

0.02

0.03

±0.

010.

01±

0.0

2_29

0.52

±0.

00.

32±

0.02

0.17

±0.

010.

13±

0.05

0.09

±0.

040.

05±

0.02

0.05

±0.

02

Fig.

B

ITA

(gl−

1)

wt

0.0

0.04

6±

0.00

40.

22±

0.11

1.19

±0.

322.

13±

0.36

2.56

±0.

083.

23±

0.07

6_15

0.0

0.05

5±

0.00

52.

16±

0.82

6.18

±1.

779.

52±

1.68

10.8

1±

1.33

11.4

8±

1.37

6_17

0.0

0.05

8±

0.00

72.

42±

0.53

7.69

±1.

6411

.56±

1.45

14.0

±0.

5415

.5±

1.0

2_29

0.0

0.06

1±

0.00

013.

12±

0.35

10.3

8±

1.82

16.2

6±

1.42

20.4

1±

0.85

22.6

±1.

43

NH

4+

(gl−

1)

wt

0.52

±0.

00.

07±

0.03

0.0

0.0

0.0

0.0

0.0

6_15

0.52

±0.

00.

00.

00.

00.

00.

00.

0

6_17

0.52

±0.

00.

00.

00.

00.

00.

00.

0

2_29

0.52

±0.

00.

7±

0.02

0.0

0.0

0.0

0.0

0.0

Table 5.5: Influence of culture aeration on ITA production and ammonium utilization

by U . maydis wt and mutants (Figure 3.11). 79


Culture volume (ml) Strain ITA (g l−1)

100

wt 0.6 ± 0.15

6_15 0.71 ± 0.007

6_17 1.28 ± 0.5

2_29 3.52 ± 0.53

50

wt 3.1 ± 0.38

6_15 1.74 ± 0.55

6_17 2.3 ± 0.13

2_29 4.18 ± 0.34

10

wt 3.23 ± 0.07

6_15 11.5 ± 1.4

6_17 15.5 ± 1.0

2_29 22.6 ± 1.4

Table 5.6: An overview of the influence of the culture aeration on ITA titers by


80

Stra

in0

h24

h48

h72

h96

h

Fig.

A

ITA

(gl−

1)

wt

0.0

0.01

4±

0.00

30.

006±

0.00

10.

016±

0.00

40.

046±

0.02

3

6_15

0.0

0.00

8±

0.00

020.

019±

0.01

20.

12±

0.05

0.21

±0.

08

6_17

0.0

0.00

8±

0.00

040.

0103

±0.

009

0.12

4±

0.09

0.48

±0.

2

2_29

0.0

0.00

5±

0.00

010.

028±

0.01

90.

73±

0.1

2.04

±0.

32

NH

4+

(gl−

1)

wt

1.35

±0.

00.

83±

0.09

0.81

±0.

050.

76±

0.11

0.6±

0.11

6_15

1.35

±0.

00.

91±

0.12

0.86

±0.

130.

71±

0.07

0.83

±0.

08

6_17

1.35

±0.

00.

89±

0.05

0.86

±0.

120.

79±

0.1

0.67

±0.

12

2_29

1.35

±0.

00.

89±

0.02

0.83

±0.

110.

71±

0.08

0.61

±0.

11

Fig.

B

ITA

(gl−

1)

wt

0.0

0.01

7±

0.00

10.

027±

0.00

40.

23±

0.04

0.44

±0.

05

6_15

0.0

0.03

8±

0.02

60.

57±

0.32

1.66

±0.

442.

9±

0.35

6_17

0.0

0.02

9±

0.01

30.

095±

0.06

52.

2±

0.6

4.6±

0.16

2_29

0.0

0.02

9±

0.01

70.

65±

0.21

2.0±

0.44

3.95

±0.

16

NH

4+

(gl−

1)

wt

0.25

0.0

0.0

0.0

0.0

6_15

0.25

0.0

0.0

0.0

0.0

6_17

0.25

0.0

0.0

0.0

0.0

2_29

0.25

0.0

0.0

0.0

0.0

Table 5.7: Influence of initial amount of nitrogen supply on ITA production and

ammonium utilization by U . maydis wt and mutants (Figure 3.14).

81


Initial NH4+ amount (g l−1) Strain ITA (g l−1)

4

wt 0.046 ± 0.023

6_15 0.21 ± 0.12

6_17 0.48 ± 0.2

2_29 2.04 ± 0.32

1.6

wt 0.18 ± 0.07

6_15 0.29 ± 0.07

6_17 0.54 ± 0.04

2_29 1.04 ± 0.12

8

wt 0.44 ± 0.05

6_15 2.9 ± 0.35

6_17 4.6 ± 0.16

2_29 4.0 ± 0.16

Table 5.8: An overview of the influence of the initial nitrogen supply on ITA titers by


82

Stra

in0

h24

h48

h72

h96

h12

0h

144

h

Fig.

A

ITA

(gl−

1)

wt

0.0

0.02

±0.

00.

2±

0.04

0.2±

0.01

1.4±

0.2

1.8±

0.2

1.9±

0.13

6_15

0.0

1.5±

0.3

5.8±

0.9

10.7

±0.

113

.6±

0.2

14.2

±1.

014

.8±

0.2

6_17

0.0

1.4±

0.3

5.3±

0.1

9.6±

1.1

13.0

±1.

313

.7±

1.0

14.0

±1.

0

2_29

0.0

1.1±

0.03

8.0±

0.7

14.4

±1.

018

.5±

0.14

18.7

±0.

318

.8±

0.2

NH

4+

(gl−

1)

wt

0.25

±0.

00.

00.

00.

00.

00.

00.

0

6_15

0.25

±0.

00.

00.

00.

00.

00.

00.

0

6_17

0.25

±0.

00.

00.

00.

00.

00.

00.

0

2_29

0.25

±0.

00.

00.

00.

00.

00.

00.

0

Fig.

BM

alat

e(g

l−1

)

wt

nd

1.7±

0.6

3.8±

0.8

5.5±

1.1

6.3±

0.5

6.4±

0.6

6.4±

0.6

6_15

nd

1.0±

0.2

2.4±

0.3

2.8±

0.07

3.2±

0.5

3.6±

0.5

3.6±

0.2

6_17

nd

1.1±

0.1

2.2±

0.15

3.4±

0.6

4.1±

0.3

3.9±

0.5

3.5±

0.6

2_29

nd

0.28

±0.

031.

2±

0.11

1.7±

0.13

2.3±

0.09

2.2±

0.3

2.1±

0.3

Fig.

CO

D(A

U)

wt

0.5

18.3

±1.

422

.4±

0.8

21.4

±1.

121

.6±

0.4

20.7

±1.

417

.5±

1.0

6_15

0.5

20.6

±0.

421

.6±

1.1

18.1

±1.

617

.5±

0.7

16.1

±1.

115

.6±

0.4

6_17

0.5

19.0

±0.

619

.7±

1.0

18.2

±1.

813

.2±

1.6

9.8±

1.8

7.4±

1.7

2_29

0.5

16.0

±1.

119

.2±

1.1

19.1

±1.

39.

5±

1.1

7.8±

0.3

7.1±

0.6

Fig.

DG

luco

se(g

l−1

)

wt

65.0

39.7

±3.

724

.4±

3.5

15.6

±4.

318.

11±

3.2

4.4±

0.3

4.8±

0.4

6_15

65.0

38.8

±2.

325

.8±

1.5

15.2

±2.

28.

4±

3.5

4.4±

1.8

2.7±

1.5

6_17

65.0

38.0

±1.

924

.0±

4.9

12.6

±1.

82.

7±

0.9

2.0±

0.5

2.1±

0.4

2_29

65.0

39.1

±3.

625

.0±

2.7

12.7

±3.

23.

6±

2.1

1.6±

0.4

1.6±

0.2

Table 5.9: Cultivation of U . maydis wt and mutants under selected optimal condi-

tions; nd - not determined (Figure 3.16).

83


DO

Tan

d

stra

in0

h4

h20

h26

h50

h76

h92

h98

h11

6h

80%

,wt

ITA

(gl−

1)

0.0

0.0

0.19

0.77

20.6

32.2

34.6

35.2

36.9

NH

4+

(gl−

1)

0.51

0.48

0.00

90.

00.

00.

00.

00.

00.

0

DW

(gl−

1)

11.9

12.9

27.1

28.6

28.9

30.3

31.5

30.9

31.1

Glu

cose

(gl−

1)

199.

619

7.1

163.

615

0.8

104.

174

.753

.044

.024

.3

DO

Tan

d

stra

in2

h7

h21

h27

h32

h56

h62

h85

h11

0h

128

h13

4h

138

h16

0h

80%

,2_2

9

ITA

(gl−

1)

0.0

0.0

0.03

0.65

2.63

9.4

12.8

619

.56

23.9

525

.71

25.9

126

.52

28.9

9

NH

4+

(gl−

1)

0.61

0.57

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

DW

(gl−

1)

21.7

21.6

35.3

37.4

nd

34.8

34.5

35.3

32.0

30.7

30.4

3231

.6

Glu

cose

(gl−

1)

209.

020

2.7

182.

316

2.6

147.

611

9.7

110.

078

.354

.038

.235

.432

.925

.2

DO

TSt

rain

1h

20h

28h

51h

76h

94h

100

h11

6h

15%

wt

ITA

(gl−

1)

0.02

0.00

60.

038

0.23

0.24

0.26

0.36

0.34

Mal

ate

(gl−

1)

0.0

0.07

2.0

2.4

2.5

2.2

2.6

2.5

DW

(gl−

1)

12.1

23.7

25.2

26.6

24.8

25.0

23.8

25.0

Glu

cose

(gl−

1)

267.

123

0.0

215.

117

0.0

150.

013

9.0

147.

413

7.7

2_29

ITA

(gl−

1)

0.02

0.02

1.04

8.9

16.1

17.2

18.2

19.3

Mal

ate

(gl−

1)

0.0

0.0

0.5

2.2

3.0

3.1

nd

3.5

DW

(gl−

1)

16.1

27.4

31.7

31.2

28.1

29.8

30.1

29.5

Glu

cose

(gl−

1)

277.

724

7.7

225.

716

0.2

115.

197

.396

.289

.8

Tab

le5.

10:F

erm

ente

rb

atc

hcu

ltiv

ati

ons

ofU

.may

dis

wt

an

d2_

29st

rain

s;D

W-

dry

bio

ma

ssw

eigh

t,n

d-

not

det

erm

ined

(Fig

-

ure

3.17

).

84

Stra

in0

h24

h48

h72

h96

h12

0h

144

h

Fig.

A

ITA

(gl−

1)

wt

0.0

0.02

±0.

00.

17±

0.04

0.2±

0.01

1.4±

0.23

1.75

±0.

191.

92±

0.13

∆cy

p1∆

emt1

0.0

0.06

±0.

011.

36±

0.12

3.0±

0.65

3.96

±0.

764.

19±

0.99

4.1±

0.53

NH

4+

(gl−

1)

wt

0.25

±0.

00.

00.

00.

00.

00.

00.

0

∆cy

p1∆

emt1

0.25

±0.

00.

00.

00.

00.

00.

00.

0

Fig.

B

OD

(AU

)w

t0.

518

.3±

1.4

22.4

±0.

821

.4±

1.1

21.6

±0.

420

.7±

1.4

17.5

±1.

0

∆cy

p1∆

emt1

0.5

22.4

±0.

936

.3±

1.3

41.1

±1.

044

.6±

2.4

41.4

±2.

540

.2±

2.0

Glu

cose

(gl−

1)

wt

65.0

39.7

±3.

724

.4±

3.5

15.6

±4.

38.

1±

3.2

4.4±

0.3

4.8±

0.4

∆cy

p1∆

emt1

65.0

36.9

±0.

520

.0±

2.0

6.3±

1.0

1.0±

0.4

1.1±

0.5

1.1±

0.5

Table 5.11: Overview of the influence of the biosurfactant production on ITA titers

using U . maydis wt and ∆cyp1 ∆emt1 strains (Figure ??).

I


Strain 24 h 48 h 72 h 96 h 120 h 192 h

ITA (g l−1)wt 0.009 0.1 0.35 0.4 0.47 4.2

∆UM11778 0.0 0.004 0.004 0.01 0.012 0.014

Fumarate (g l−1)wt 0.01 0.06 0.14 0.35 0.44 0.56

∆UM11778 0.02 0.02 0.05 0.12 0.22 0.32

Glucose (g l−1)wt 60.6 55.3 50.1 37.1 28.5 0.0

∆UM11778 61.0 54.9 43.0 32.1 21.9 0.0

Succinate (g l−1)wt 0.25 1.0 2.8 3.0 2.8 3.2

∆UM11778 0.0 0.0 0.0 0.0 0.0 0.7

Malate (g l−1)wt 0.83 2.28 4.6 6.2 6.7 9.2

∆UM11778 0.0 0.0 0.12 0.95 2.0 5.0

Pyruvate (g l−1)wt 0.1 0.15 0.3 0.1 0.08 0.0

∆UM11778 0.6 0.23 0.16 0.14 0.16 0.0

Table 5.12: The production kinetics of various substances putatively involved in ITA

synthesis pathway. U . maydis wt and ∆UM11778 strains were used (Fig-

ure 3.20).

II

List of Figures

1.1 Overview of potential fuel precursors derived from various products of hydro-

lyzed lignocellulose. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2 Examples of possible itaconic acid application. . . . . . . . . . . . . . . . . . . . 5

1.3 Proposed subcellular organization of the itaconate pathway in A. terreus. . . . . 6

1.4 Smut disease on corn caused by dikaryotic form of U . maydis and the unicellu-

lar haploid form of U . maydis with buds. . . . . . . . . . . . . . . . . . . . . . . . 8

2.1 Outline of regeneration agar plate . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3.1 Integrated disruption cassette. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3.2 Comparison of production of the itaconate by U . maydis wt and ∆UM06344

strains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.3 Outline of the UM06344 overexpression vector. . . . . . . . . . . . . . . . . . . . 28

3.4 Itaconate production by all the UM06344 overexpression transformants. . . . . 29

3.5 Outline of the integrated UM06344 overexpression vector. . . . . . . . . . . . . . 31

3.6 Southern blot analyses of U . maydis wt and UM06344 overexpression trans-

formants. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

3.7 A typical cultivation chart of U . maydis wt . . . . . . . . . . . . . . . . . . . . . . 33

3.8 Influence of glucose concentration on the itaconate production and ammonium

utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

3.9 An overview of the influence of the initial glucose concentration on the itacon-

ate titer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

3.10 Influence of various glucose concentration in MTM on BOD and CBP in the

U . maydis 2_29 strain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3.11 Influence of the culture aeration on the itaconate titer . . . . . . . . . . . . . . . 38

3.12 An overview of the influence of the culture aeration on the itaconate titer . . . . 40

3.13 Overview of the results from the off-gas analysis. . . . . . . . . . . . . . . . . . . 41

3.14 Influence of the initial NH4Cl concentration on the itaconate titer . . . . . . . . 42

III

List of Figures

3.15 An overview of the influence of the initial NH4Cl concentration on the itaconate

titer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

3.16 Kinetics of the itaconate production under selected optimal conditions . . . . . 45

3.17 Fermenter batch cultivation charts . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

3.18 Microscopic view of biosurfactants produced by U . maydis cells . . . . . . . . . 50

3.19 The itaconate production by U . maydis wt and ∆cyp1 ∆emt1 strains . . . . . . . 50

3.20 The production kinetics of various substances putatively involved in the itacon-

ate synthesis pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

3.21 HPLC profile of supernatant from U . maydis ∆UM11778 strain with clearly

visible itaconate peak . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

3.22 HPLC profile of supernatant from U . maydis ∆UM11778 strain containing a

large peak of undefined compound . . . . . . . . . . . . . . . . . . . . . . . . . . 55

4.1 Putative itaconate synthesis routes in U . maydis and A. terreus . . . . . . . . . . 59

4.2 Theoretical yields and production rates of itaconate vs. ammonium concentra-

tion by U . maydis wt and 2_29 mutant strains . . . . . . . . . . . . . . . . . . . . 64

4.3 Putative possibilities of cofactor regeneration in U . maydis. . . . . . . . . . . . . 67

IV

List of Tables

1.1 Selection of alternative itaconate production hosts . . . . . . . . . . . . . . . . . 7

2.1 List of used strains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

2.2 List of used plasmids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

2.3 List of used primers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3.1 Summary of integration numbers of UM06344 overexpression cassette parts . . 32

3.2 Experimental design of optimization of the itaconate production parameters . 34

3.3 The summary of the end-point values of OD, glucose concentration and yield

based on various initial glucose concentration . . . . . . . . . . . . . . . . . . . . 36


based on various culture aeration . . . . . . . . . . . . . . . . . . . . . . . . . . . 39


based on various initial NH4Cl concentration . . . . . . . . . . . . . . . . . . . . 44

3.6 Selection of final values obtained from the optimal cultivation conditions ex-

periment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46


based on various carbon sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

5.1 Comparison of ITA production by U . maydis wt and ∆UM06344 strains . . . . . 76

5.2 A typical cultivation chart of U . maydis wt . . . . . . . . . . . . . . . . . . . . . . 76

5.3 Influence of initial glucose amount on ITA production and ammonium utilization 77

5.4 An overview of the influence of the initial glucose amount on ITA titers by

U . maydis wt and mutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

5.5 Influence of culture aeration on ITA production and ammonium utilization . . 79

5.6 An overview of the influence of the culture aeration on ITA titer . . . . . . . . . 80

5.7 Influence of initial amount of nitrogen supply on ITA production and am-

monium utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

V

List of Tables

5.8 An overview of the influence of the initial nitrogen supply on ITA titer . . . . . . 82

5.9 Cultivation of U . maydis wt and mutants under selected optimal conditions . . 83

5.10 Fermenter batch cultivations of U . maydis wt and 2_29 strains . . . . . . . . . . 84

5.11 Overview of the influence of the biosurfactant production on ITA titers . . . . . I

5.12 The production kinetics of various substances putatively involved in ITA syn-

thesis pathway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . II

VI

Bibliography

[1] Leonardo D Gomez, Clare G Steele-King, and Simon J McQueen-Mason. Sustainable

liquid biofuels from biomass: the writing’s on the walls. The New Phytologist, 178(3):473–

85, 2008. 1

[2] Ali Mohagheghi, Kent Evans, Yat-Chen Chou, and Min Zhang. Cofermentation of glucose,

xylose, and arabinose by genomic DNA-integrated xylose/arabinose fermenting strain

of Zymomonas mobilis AX101. Applied biochemistry and biotechnology, 98-100:885–98,

2002. 1, 4

[3] L R Jarboe, T B Grabar, and L P Yomano K T Shanmugan. Development of Ethanologenic

Bacteria. Advances in biochemical engineering/biotechnology, 108:237–261, 2007. 1

[4] S Pischinger. Tailor-Made Fuels from Biomass. Technical Report Exc 236, RWTH Aachen,

Aachen, 2008. 1

[5] J L Fortman, Swapnil Chhabra, Aindrila Mukhopadhyay, Howard Chou, Taek Soon Lee,

Eric Steen, and Jay D Keasling. Biofuel alternatives to ethanol: pumping the microbial

well. Trends in Biotechnology, 26(7):375–81, 2008. 2, 3, 4, 71

[6] Mehdi Dashtban, Heidi Schraft, and Wensheng Qin. Fungal Bioconversion of Ligno-

cellulosic Residues ; Opportunities & Perspectives. International Journal of Biological

Sciences, 5(6):578–595, 2009. 3

[7] Badal C Saha. Hemicellulose bioconversion. Journal of industrial microbiology &

biotechnology, 30(5):279–91, 2003. 3

[8] Gary Brodeur, Elizabeth Yau, Kimberly Badal, John Collier, K B Ramachandran, and

Subramanian Ramakrishnan. Chemical and physicochemical pretreatment of lignocel-

lulosic biomass: a review. Enzyme research, 2011:1 – 17, 2011. 3

[9] Dieter Klemm, Brigitte Heublein, Hans-Peter Fink, and Andreas Bohn. Cellulose: fascin-

ating biopolymer and sustainable raw material. Angewandte Chemie (International ed.

in English), 44(22):3358–93, 2005. 3

VII

Bibliography

[10] Mohammad J Taherzadeh and Keikhosro Karimi. Pretreatment of lignocellulosic wastes

to improve ethanol and biogas production: a review. International journal of molecular

sciences, 9(9):1621–51, 2008. 3

[11] Mads Pedersen, Katja S Johansen, and Anne S Meyer. Low temperature lignocellulose

pretreatment: effects and interactions of pretreatment pH are critical for maximizing

enzymatic monosaccharide yields from wheat straw. Biotechnology for biofuels, 4(1):1–

10, 2011. 3

[12] Ye Sun and Jiayang Cheng. Hydrolysis of lignocellulosic materials for ethanol production:

a review. Bioresource technology, 83(1):1–11, 2002. 4

[13] Y Y Lee, Iyer Prashant, and R W Torget. Dilute-Acid Hydrolysis of Lignocellulosic Biomass.

In Advances In Biochemical Engineering/Biotechnology, volume 65, pages 94 – 115. 1999.

4

[14] S Song and C Park. Organization and regulation of the D-xylose operons in Escherichia

coli K-12: XylR acts as a transcriptional activator. Journal of bacteriology, 179(22):7025–

32, 1997. 4

[15] a L Does and L F Bisson. Characterization of Xylose Uptake in the Yeasts Pichia heedii

and Pichia stipitis. Applied and environmental microbiology, 55(1):159–64, 1989. 4

[16] B Hahn-Hägerdal, C F Wahlbom, M Gárdonyi, W H van Zyl, R R Cordero Otero, and

L J Jönsson. Metabolic engineering of Saccharomyces cerevisiae for xylose utilization.

Advances in biochemical engineering/biotechnology, 73:53–84, 2001. 4

[17] Marko Kuyper, Maurice J Toirkens, Jasper a Diderich, Aaron a Winkler, Johannes P van

Dijken, and Jack T Pronk. Evolutionary engineering of mixed-sugar utilization by a

xylose-fermenting Saccharomyces cerevisiae strain. FEMS yeast research, 5(10):925–34,

2005. 4

[18] Antonius J A Van Maris, Aaron A Winkler, Marko Kuyper, Wim T A M De Laat, Johannes

P Van Dijken, Jack T Pronk, A Fleminglaan, and A X Delft. Development of Efficient Xylose

Fermentation in Saccharomyces cerevisiae : Xylose Isomerase as a Key Component. Adv

Biochem Engin/Biotechnol, 108(April):179–204, 2007. 4

[19] T Werpy and G Petersen. Top Value Added Chemicals from Biomass Volume I, Results

of Screening for Potential Candidates from Sugars and Synthesis Gas. Technical report,

U.S. Department of Energy, 2004. 4, 5, 7

[20] Mitsuyasu Okabe, Dwiarti Lies, Shin Kanamasa, and Enoch Y Park. Biotechnological pro-

duction of itaconic acid and its biosynthesis in Aspergillus terreus. Applied microbiology

and biotechnology, 84(4):597–606, 2009. 4, 6, 7, 66, 71

VIII

Bibliography

[21] Tania Betancourt, Juan Pardo, Ken Soo, and Nicholas A. Peppas. Characterization of

pH-Responsive Hydrogels of Poly(Itaconic acid-g-Ethylene Glycol) Prepared by UV-

Initiated Free Radical Polymerization as Biomaterials for Oral Delivery of Bioactive

Agents. Journal of Biomedical Materials Research, 93(1):175–188, 2010. 5

[22] H Warson and C. H. Finch. Applications of Synthetic Resin Latices: Latices in diverse

applications. John Wilea and Sons Ltd., England, volume 3 edition, 2001. 5

[23] Dushyant Shekhawat, James E Jackson, and Dennis J Miller. Process model and eco-

nomic analysis of itaconic acid production from dimethyl succinate and formaldehyde.

Bioresource technology, 97(2):342–7, 2006. 5

[24] W. M. Jaklitsch, C. P. Kubicek, and M. C. Scrutton. The subcellular organization of itacon-

ate biosynthesis in Aspergillus terreus. Journal of General Microbiology, 137(3):533–539,

1991. 5, 6, 68

[25] P Bonnarme, B Gillet, a M Sepulchre, C Role, J C Beloeil, and C Ducrocq. Itaconate

biosynthesis in Aspergillus terreus. Journal of bacteriology, 177(12):3573–8, 1995. 5, 54,

57, 58, 59, 60, 73

[26] Ronald Bentley and Clara P. Thiessen. Biosynthesis of itaconic acid in Aspergillus terreus,

II. Early stages in glucose dissimilation and the role of citrate. Methods, pages 689–701,

1956. 5, 65

[27] Shin Kanamasa, Lies Dwiarti, Mitsuyasu Okabe, and Enoch Y Park. Cloning and func-

tional characterization of the cis-aconitic acid decarboxylase (CAD) gene from Asper-

gillus terreus. Applied microbiology and biotechnology, 80(2):223–9, 2008. 5, 25, 58,

66

[28] Maria Papagianni. Advances in citric acid fermentation by Aspergillus niger: biochemical

aspects, membrane transport and modeling. Biotechnology advances, 25(3):244–63, 2007.

6, 67

[29] Kazutoyo Yahiro, Tetsushi Takahama, Yong S Park, and Mitsuyasu Okabe. Breeding of

Aspergillus terreus Mutant TN-484 for Itaconic Acid Production with High Yield. Journal

of Fermentation and Bioengineering, 79(5):506–508, 1995. 7, 65, 66

[30] Anja Kuenz, Yvonne Gallenmüller, Thomas Willke, and Klaus-Dieter Vorlop. Microbial

production of itaconic acid: developing a stable platform for high product concentra-

tions. Applied microbiology and biotechnology, 96(5):1209–16, 2012. 7

[31] P a Gibbs, R J Seviour, and F Schmid. Growth of filamentous fungi in submerged culture:

problems and possible solutions. Critical reviews in biotechnology, 20(1):17–48, 2000. 7

IX

Bibliography

[32] Th. Willke and K.-D. Vorlop. Biotechnological production of itaconic acid. Applied

Microbiology and Biotechnology, 56(3-4):289–295, 2001. 7, 71

[33] Tobias Klement, Sofia Milker, Gernot Jäger, Philipp M Grande, Pablo Domínguez de

María, and Jochen Büchs. Biomass pretreatment affects Ustilago maydis in producing

itaconic acid. Microbial cell factories, 11(1):43, 2012. 7, 9, 67, 70

[34] William E. Levinson, Cletus P. Kurtzman, and Tsung Min Kuo. Production of itaconic

acid by Pseudozyma antarctica NRRL Y-7808 under nitrogen-limited growth conditions.

Enzyme and Microbial Technology, 39(4):824–827, 2006. 7, 70

[35] R Haskins, J A Thorn, and B Boothroyd. Biochemistry of ustilaginales. XI. Metabolic

products of Ustilago zeae in submerged culture. Canadian Journal of Microbiology,

1:749–756, 1955. 7, 9

[36] Thomas Brefort, Gunther Doehlemann, Artemio Mendoza-Mendoza, Stefanie Reiss-

mann, Armin Djamei, and Regine Kahmann. Ustilago maydis as a Pathogen. Annual

review of phytopathology, 47:423–45, 2009. 8

[37] Jörg Kämper, Regine Kahmann, Michael Bölker, Li-Jun Ma, Thomas Brefort, Barry J

Saville, Flora Banuett, James W Kronstad, Scott E Gold, Olaf Müller, Michael H Perlin,

Han a B Wösten, Ronald de Vries, José Ruiz-Herrera, Cristina G Reynaga-Peña, Karen

Snetselaar, Michael McCann, José Pérez-Martín, Michael Feldbrügge, Christoph W Basse,

Gero Steinberg, Jose I Ibeas, William Holloman, Plinio Guzman, Mark Farman, Jason E

Stajich, Rafael Sentandreu, Juan M González-Prieto, John C Kennell, Lazaro Molina, Jan

Schirawski, Artemio Mendoza-Mendoza, Doris Greilinger, Karin Münch, Nicole Rössel,

Mario Scherer, Miroslav Vranes, Oliver Ladendorf, Volker Vincon, Uta Fuchs, Björn

Sandrock, Shaowu Meng, Eric C H Ho, Matt J Cahill, Kylie J Boyce, Jana Klose, Steven J

Klosterman, Heine J Deelstra, Lucila Ortiz-Castellanos, Weixi Li, Patricia Sanchez-Alonso,

Peter H Schreier, Isolde Häuser-Hahn, Martin Vaupel, Edda Koopmann, Gabi Friedrich,

Hartmut Voss, Thomas Schlüter, Jonathan Margolis, Darren Platt, Candace Swimmer,

Andreas Gnirke, Feng Chen, Valentina Vysotskaia, Gertrud Mannhaupt, Ulrich Güldener,

Martin Münsterkötter, Dirk Haase, Matthias Oesterheld, Hans-Werner Mewes, Evan W

Mauceli, David DeCaprio, Claire M Wade, Jonathan Butler, Sarah Young, David B Jaffe,

Sarah Calvo, Chad Nusbaum, James Galagan, and Bruce W Birren. Insights from the

genome of the biotrophic fungal plant pathogen Ustilago maydis. Nature, 444(7115):97–

101, 2006. 8

[38] Michael Bölker, Christoph W Basse, and Jan Schirawski. Ustilago maydis secondary

metabolism-from genomics to biochemistry. Fungal genetics and biology : FG & B, 45

Suppl 1:S88–93, 2008. 8, 9, 49

X

Bibliography

[39] Sandra Hewald, Katharina Josephs, and Michael Boelker. Genetic Analysis of Bios-

urfactant Production in Ustilago maydis. Applied and Environmental Microbiology,

71(6):3033–3040, 2005. 9, 11, 49, 69

[40] S Spoeckner, V Wray, M Nimtz, and S Lang. Glycolipids of the smut fungus Ustilago may-

dis from cultivation on renewable resources. Applied Microbiology and Biotechnology,

51(1):33–39, 1999. 9

[41] Ramesh Chandragiri and R.C. Sastry. SYNTHESIS OF ITACONIC ACID USING USTILAGO

MAYDIS. Canadian Journal on Chemical Engineering & Technology, 2(7):128–135, 2011.

9

[42] Mokula Rafi, M G Hanumanthu, S Rizwana, Kadiyala Venkateswarlu, and Dowlath-

abad Muralidhara Rao. Effect of different physico-chemical parameters on fermentative

production of itaconic acid by Ustilago maydis. Journal of Microbiology and Biotechno-

logy Research, 2(5):794–800, 2012. 9

[43] D Finley, E Ozkaynak, and a Varshavsky. The yeast polyubiquitin gene is essential for

resistance to high temperatures, starvation, and other stresses. Cell, 48(6):1035–46, 1987.

11

[44] A Brachmann, J König, C Julius, and M Feldbrügge. A reverse genetic approach for gen-

erating gene replacement mutants in Ustilago maydis. Molecular genetics and genomics

: MGG, 272(2):216–26, 2004. 12

[45] M Bölker, H U Böhnert, K H Braun, J Görl, and R Kahmann. Tagging pathogenicity genes

in Ustilago maydis by restriction enzyme-mediated integration (REMI). Molecular &

general genetics : MGG, 248(5):547–52, 1995. 12

[46] C Mézard, D Pompon, and a Nicolas. Recombination between similar but not identical

DNA sequences during yeast transformation occurs within short stretches of identity.

Cell, 70(4):659–70, 1992. 14, 23, 25, 27

[47] D Guevarra and Takeshi Tabuchi. Accumulation of Itaconic, 2-Hydroxyparaconic, Itatar-

taric, and Malic Acids by Strains of the Genus Ustilago. Agric. Biol. Chem., 54(9):2353–

2358, 1990. 14, 34, 65

[48] M Robertz, T Muckenheim, S Eckl, and L Webb. Kostengünstige Labormethode zur

Bestimmung der mikrobiellen Bodenatmung nach DIN 19737. Wasser & Boden, 51(5):48–

53, 1999. 16, 19

[49] D Herbert, P J Phipps, and R E Strange. Chemical analysis of microbial cells. In Norris J

R and Ribbons D W, editors, Methods in Microbiology. Academic Press London and New

York, 1971. 19

XI

Bibliography

[50] K L Burdon. Fatty material in bacteria and fungi revealed by staining dried, fixed slide

preparations. Journal of bacteriology, 52(6):665–78, 1946. 19

[51] C S Hoffman and F Winston. A ten-minute DNA preparation from yeast efficiently

releases autonomous plasmids for transformation of Escherichia coli. Gene, 57(2-3):267–

72, 1987. 21

[52] C R Newton and A Graham. PCR. In PCR. Oxford: BIOS Scientific Publishers Limited,

1994. 22

[53] J Sambrook, E F Fritsch, and T Maniatis. Molecular cloning. A laboratory manual. New

York: Cold Spring Habour Laboratory Press., 1989. 23

[54] R D Gietz and A Woods. Transformation of yeast by lithium acetate/single-stranded

carrier DNA/polyethylen glycol method. Methods in enzymology, 350:87–96, 2002. 23

[55] T Tsukuda, S Carleton, S Fotheringham, and W K Holloman. Isolation and characteriz-

ation of an autonomously replicating sequence from Ustilago maydis. Molecular and

cellular biology, 8(9):3703–9, 1988. 23

[56] F M Ausubel, R Brent, R E Kingston, D D Moore, J A Smith, and K Struhl. Current

protocols in molecular biology. Current protocols in molecular biology., 1991. 23

[57] Gregor Jansen, Cunle Wu, Babette Schade, David Y Thomas, and Malcolm Whiteway.

Drag&Drop cloning in yeast. Gene, 344:43–51, 2005. 25, 27

[58] J U N Wang, David W Holden, and Sally A Leongtt. Gene transfer system for the phyto-

pathogenic fungus Ustilago maydis. Genetics, 85(February):865–869, 1988. 28

[59] Gregor Tevz, Mojca Bencina, and Matic Legisa. Enhancing itaconic acid production by

Aspergillus terreus. Applied microbiology and biotechnology, 87(5):1657–64, 2010. 34, 70

[60] Walker. Yeast nutrition. In Yeast Physiology and Biotechnology, chapter Yeast nutr, pages

51–75. John Wiley & Sons, 1998. 42

[61] Stephen F Altschup, Warren Gish, Webb Miller, Eugene W. Myers, and David J. Lipman.

Basic Local Alignment Search Tool. Journal of Molecular Biology, 215:403–410, 1990. 52

[62] Milko Velarde, Sofia Macieira, Markus Hilberg, Gerd Bröker, Shang-Min Tu, Bernard T

Golding, Antonio J Pierik, Wolfgang Buckel, and Albrecht Messerschmidt. Crystal struc-

ture and putative mechanism of 3-methylitaconate-delta-isomerase from Eubacterium

barkeri. Journal of molecular biology, 391(3):609–20, 2009. 52, 60

[63] I R Shimi and M S Nour El Dein. Biosynthesis of itaconic acid by aspergillus terreus.

Archiv für Mikrobiologie, 44:181–8, 1962. 58

XII

Bibliography

[64] Lindsay Blank, Jeffrey Green, and John R Guest. AcnC of Escherichia coli is a 2-

methylcitrate dehydratase (PrpD) that can use citrate and isocitrate as substrates. Mi-

crobiology (Reading, England), 148(Pt 1):133–46, 2002. 60

[65] Mariet van der Werf and Martinus Caspers. Production of itaconic acid, 2009. 60

[66] A. Nowakowska-Waszczuk. Utilization of Some Tricarboxylic-acid-cycle Intermediates

by Mitochondria and Growing Mycelium of Aspergillus terreus. Journal of General

Microbiology, 79(1):19–29, 1973. 60

[67] Brian P Rubin, Dangsheng Li, and William K Holloman. The LEU1 gene of Ustilago

maydis. Gene, 140:131–135, 1994. 61

[68] Beate Teichmann, Lidan Liu, Kay Oliver Schink, and Michael Bölker. Activation of

the ustilagic acid biosynthesis gene cluster in Ustilago maydis by the C2H2 zinc finger

transcription factor Rua1. Applied and Environmental Microbiology, 76(8):2633–40, 2010.

62

[69] Dirk Hoffmeister and Nancy P Keller. Natural products of filamentous fungi: enzymes,

genes, and their regulation. Natural product reports, 24(2):393–416, 2007. 62

[70] Michael Mattey and Bjorn Kristiansen. Citric acid biotechnology. Taylor and Francis,

1999. 63, 67

[71] Levente Karaffa and Christian P Kubicek. Aspergillus niger citric acid accumulation: do

we understand this well working black box? Applied microbiology and biotechnology,

61(3):189–96, 2003. 64

[72] Jaehoon Choe and Young J E Yoo. Effect of Ammonium Ion Concentration and Applica-

tion to Fed-Batch Culture for Overproduction of Citric Acid. Journal of Fermentation

and Bioengineering, 72(2):106–109, 1991. 64

[73] Christopher T. Evans and Colin Ratledge. Effect of Nitrogen Source on Lipid Accu-

mulation in Oleaginous Yeasts. Journal of General Microbiology, 130:1693–1704, 1984.

64

[74] C M Lewis and J R Fincham. Regulation of nitrate reductase in the basidiomycete

Ustilago maydis. Journal of bacteriology, 103(1):55–61, 1970. 64

[75] Lies Dwiarti, Ken Yamane, Hitoshi Yamatani, Prihardi Kahar, and Mitsuyasu Okabe.

Purification and characterization of cis-aconitic acid decarboxylase from Aspergillus

terreus TN484-M1. Journal of bioscience and bioengineering, 94(1):29–33, 2002. 64, 66

XIII

Bibliography

[76] C S K Reddy and R P Singh. Enhanced production of itaconic acid from corn starch and

market refuse fruits by genetically manipulated Aspergillus terreus SKR10. Bioresource

technology, 85(1):69–71, 2002. 64, 71

[77] Max Roehr and Christian P Kubicek. Production of primary metabolism. In H. J. Rehm

and G. Reed, editors, Biotechnology, chapter Further or, pages 363–379. VCH, Weinheim,

second edition, 1996. 65

[78] Stefan Hohmann. Osmotic Stress Signaling and Osmoadaptation in Yeasts. Microbiology

and Molecular Biology Reviews, 66(2):300–372, 2002. 65

[79] A Diano, S Jensen-Bekker, J Dynesen, and J Nielsen. Polyol Synthesis in Aspergillus niger

: Influence of Oxygen Availability , Carbon and Nitrogen Sources on the Metabolism.

Biotechnology and Bioengineering, 94(5):899–908, 2006. 65, 66, 68, 69

[80] Mitsuyasu Okabe, Noriyasu Ohta, and Yong Soo Park. Itaconic acid production in an air-

lift bioreactor using a modified draft tube. Journal of Fermentation and Bioengineering,

76(2):117–122, 1993. 66

[81] Yong Soo Park, Noriyasu Ohta, and Mitsuyasu Okabe. EFFECT OF DISSOLVED OXYGEN

CONCENTRATION AND IMPELLER TIP SPEED ON ITACONIC ACID PRODUCTION BY

ASPERGILLUS TERREUS. Biotechnology Letters, 15(6):583–586, 1993. 66, 69

[82] M H Gyamerah. Oxygen requirement and energy relations of itaconic acid fermentation

by Aspergillus terreus NRRL 1960. Applied Microbiology and Biotechnology, 44:20–26,

1995. 66

[83] Elvira D Guevarra and Takeshi Tabuchi. Production of 2-Hydroxyparaconic and Itatar-

taric Acids by Ustilago cynodontis and Simple Recovery Process of the Acids. Agric. Biol.

Chem., 54(9):2359 – 2365, 1990. 66

[84] Vishist K Jain, Benoit Divol, Bernard a Prior, and Florian F Bauer. Effect of alternative

NAD+-regenerating pathways on the formation of primary and secondary aroma com-

pounds in a Saccharomyces cerevisiae glycerol-defective mutant. Applied microbiology

and biotechnology, 93(1):131–41, 2012. 67, 68

[85] Oscar Juárez, Guadalupe Guerra, Federico Martínez, and Juan Pablo Pardo. The

mitochondrial respiratory chain of Ustilago maydis. Biochimica et biophysica acta,

1658(3):244–51, 2004. 68

[86] Qingzhao Wang, Mark S Ou, Y Kim, L O Ingram, and K T Shanmugam. Metabolic flux

control at the pyruvate node in an anaerobic Escherichia coli strain with an active

pyruvate dehydrogenase. Applied and environmental microbiology, 76(7):2107–14, 2010.

68

XIV

Bibliography

[87] Bastian Blombach, Tanja Riester, Stefan Wieschalka, Christian Ziert, Jung-Won Youn,

Volker F Wendisch, and Bernhard J Eikmanns. Corynebacterium glutamicum tailored for

efficient isobutanol production. Applied and environmental microbiology, 77(10):3300–

10, 2011. 68

[88] Oscar Juárez, Guadalupe Guerra, Isabel Velázquez, Oscar Flores-Herrera, R E Rivera-

Pérez, and Juan P Pardo. The physiologic role of alternative oxidase in Ustilago maydis.

The FEBS journal, 273(20):4603–15, 2006. 68

[89] J P Wynn, a a Hamid, Y Li, and C Ratledge. Biochemical events leading to the diversion of

carbon into storage lipids in the oleaginous fungi Mucor circinelloides and Mortierella

alpina. Microbiology (Reading, England), 147(Pt 10):2857–64, 2001. 68

[90] M A Z Coelho, P F F Amaral, and I Belo. Yarrowia lipolytica : an industrial workhorse. In

A Méndez-Vilas, editor, Current Research, Technology and Education Topics in Applied

Microbiology and Microbial Biotechnology, pages 930–944. 2010. 69

[91] Anca-Irina Galaction, Dan Cascaval, Corneliu Oniscu, and Marius Turnea. Enhance-

ment of oxygen mass transfer in stirred bioreactors using oxygen-vectors. 1. Simulated

fermentation broths. Bioprocess and biosystems engineering, 26(4):231–8, 2004. 70

[92] Eyal Bressler and Sergei Braun. Conversion of citric acid to itaconic acid in a novel liquid

membrane bioreactor. Journal of Chemical Technology and Biotechnology, 72(August

1999):66–72, 2000. 70, 71

[93] Léonie G M Boender, Erik a F de Hulster, Antonius J a van Maris, Pascale a S Daran-

Lapujade, and Jack T Pronk. Quantitative physiology of Saccharomyces cerevisiae at

near-zero specific growth rates. Applied and environmental microbiology, 75(17):5607–

14, 2009. 70

[94] Kirsten Welter. Biotechnische Produktion von Itaconsäure aus nachwachsenden Rohstof-

fen mit immobilisierten Zellen. Dissertation, Technische Universität Carolo-Wilhelmina

zu Braunschweig, 2000. 70

[95] E Riscaldati, M Moresi, F Federici, and M Petruccioli. Effect of pH and stirring rate on

itaconate production by Aspergillus terreus. Journal of Biotechnology, 83(3):219–30,

2000. 70

[96] H Kautola, N Vassilev, and Y Y Linko. Continuous itaconic acid production by immobil-

ized biocatalysts. Journal of biotechnology, 13(4):315–23, 1990. 71

[97] M I Juy and M E Lucca. Study of itaconic acid production by Aspergillus terrus MJL05

strain with different variable. Revista Colombiana de biotecnología, 12(2):187–193, 2010.

71

XV

Bibliography

[98] A Begum, N Choudhury, and M Islam. Citric Acid Fermentation by Gamma Ray Induced

Mutants of Aspergillus niger in Different Carbohydrate Media. Journal of Fermentation

and Bioengineering, 70(4):286–288, 1990. 71

XVI

Curriculum Vitae

Personal detailsName: Monika Maasem Panáková

Date of birth: 7.1.1982

Place of birth: Šumperk, Czech Republic

Nationality: Czech

Education9/1988 - 8/1990 5th Elementary School, Šumperk, Czech Republic

9/1990 - 8/1996 3r d Elementary School with Extended Foreign Language Education,

Šumperk, Czech Republic

9/1996 - 6/2001 Grammar School in Šumperk, Šumperk, Czech Republic; A-levels with

final "excellent" grade

9/2001 - 8/2004 University of Pardubice, Czech Republic, Faculty of Chemical Techno-

logy; Bachelor’s degree in Clinical Biology and Chemistry with final

"excellent" grade

9/2004 - 6/2006 University of Pardubice, Czech Republic, Faculty of Chemical Techno-

logy; Master’s degree in Analysis of Biological Materials with final "ex-

cellent" grade; Master thesis "The Influence of Low Molecular Weight

Hyaluronic Acid to Human Chondrocytes”

4/2008 - 10/2013 RWTH Aachen University, Aachen, Germany, Institute of Applied Mi-

crobiology; Ph.D. thesis "Itaconate production by Ustilago maydis; the

influence of genes and cultivation conditions"

Professional experiences2/2005 - 3/2006 Research assistant at Contipro Group, Dolní Dobrouc, Czech Republic

7/2006 - 8/2006 Research assistant at Department of Biochemistry, Faculty of Medicine

of Charles University in Prague, Hradec Králové, Czech Republic

10/2006 - 8/2007 Research assistant at Institute of Molecular Biophysics, CNRS Orleans,

France

Declaration of independent work

I declare that I have independently written the work presented here, and I have not used any

help other than from the stated sources and resources.

Aachen, January 29, 2013

Itaconate production by Ustilago maydis - darwin.bth.rwth...

Documents

Transcript of Itaconate production by Ustilago maydis - darwin.bth.rwth...