Research Collection

Doctoral Thesis

The relevance of transhydrogenases and heterologous phosphagen kinases for microbial cofactor metabolism

Author(s): Canonaco, Fabrizio

Publication Date: 2003

Permanent Link: https://doi.org/10.3929/ethz-a-004525971

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.

ETH Library DISS. ETH.Nr. 15040

The relevance of transhydrogenases and heterologous phosphagen

kinases for microbial cofactor metabolism

A dissertation submitted to the

SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZURICH

for the degree of

Doctor of Natural Science

Presented by

Fabrizio Canonaco

Dipl. Natw. ETH

born April 22nd, 1975

in Locarno, Switzerland

Accepted on the recommendation of

PD Dr. U. Sauer

Prof. T. Wallimann

PD Dr. U. Schlattner

Prof. H. Hennecke

2003 Seite Leer Blank

2 ACKNOWLEDGEMENTS

and further back to the When I think back to the time point where I began my PhD,

in a how one recalls as have being important years of the study, it is surprising many people

I owe a it is not to name them all, but to some people way or another. In this section, possible PhD. tribute, because without them I would not have succeeded in finishing my

and it could not have been else. They The first big 'thank you' goes to my family,

to without me in I did not want do, always played a central role in my life, pushing anything their sacrifices and their encouragements I but supporting me in everything I chose. Without would not have been writing these pages.

for the sincere they offered mc To other people I owe my deepest gratitude, friendship for the stories behind the in the difficult moments I encountered. There is no place here

to me alive' in the last years, and essential contribution every one of them gave 'keeping

these nicknames, in neither I would be able to put them in words, so I simply list people by

as casual order. Jonne, Maud, Droopy, Pimpa, Zambo, Sonda, Claudia, Mitch, your loyalty

friends will never be forgotten.

in Uwe Sauer. In his My scientific and professional career had a very competent mentor

later as a PhD a student five and I continued group I began my scientific career as years ago, and to blame when are going bad student. As group leader, it would be easy people things

Uwe had words of encouragement when congratulate them as they succeed, but always

critics as small successes made motivation was low. He also had precise and appropriated my

For his of 'the boss' and the me blind towards new upcoming problems. original way being

I owe him contribution to my scientific maturation my deepest gratitude.

kinases was conducted in collaboration The part of the project involving phosphagen

the Cell Institute of the ETH Zürich. I'm with the group of Theo Wallimann, at Biology also co-referees for this deeply thankful to Uwe Schlattner and Theo Wallimann, which are had. I also thank work, for their intriguing suggestions and the stimulating discussions we in the examination of this Prof. Hauke Hennecke for having accepted the role of co-referee

doctoral thesis.

3 the contribution of students I had Parts of the results presented here were achieved with Pascal Baillod and Lorena the opportunity to train. I hope I could have helped Sylvia Heri,

Jaramillo, the way they helped me.

is not I therefore will miss and No job results in success if the work atmosphere optimal.

collaborative attitude of the Sauer Group people, where I always never forget the friendly and and Last to Marco, Eliane, Ursi, Lars, Jörg. felt part of a team. Special thanks Peter, Nicola, I the at the Institute of which had but not least, thanks to all the other people Biotechnology opportunity to know and work with.

4 TABLE OF CONTENTS

7 SUMMARY/ SOMMARIO

11 CHAPTER 1 General introduction

isomerase 55 CHAPTER 2 Metabolic flux response to phosphoglucose knock-out in Escherichia coli and impact of overexpression of the soluble transhydrogenase UdhA

69 CHAPTER 3 Dissecting the role of soluble and membrane-bound transhydrogenases UdhA and PntAB in Escherichia coli

95 CHAPTER 4 Functional expression of phosphagen kinase systems confers resistance to transient stresses in Saccharomyces cerevisiae by buffering the ATP pool

119 CHAPTER 5 Functional expression of arginine kinase improves recovery from pH stress of Escherichia coli

133 CONCLUSIONS AND OUTLOOK

135 ABBREVIATIONS LIST

139 CURRICULUM VITAE

5 Seite Leer / Blank leaf

6 "Obstacles are thosefrightful things you see

when you take your eyes offyour goal"

Henry Ford

SUMMARY

NADPH and how The scientific topic of this thesis is how micro-organisms generate that transient protection during they can be equipped with ATP buffering systems provide

nucleotides are essential cofactors for cellular metabolism, energy stress. ATP and pyridine utilized for linking catabolism to anabolism. While NADH is almost exclusively energy reactions. generation, NADPH provides reducing power for biosynthetic construction and metabolic flux In Chapter 2 and Chapter 3, we show through mutant have distinct roles in NADPH analysis that the two transhydrogenases of Escherichia coli

is an source of metabolism. The membrane-bound transhydrogenase PntAB important

the of NAPDH during exponential growth on glucose by catalyzing transhydrogenation

in the reverse NADH to NADPH. The soluble transhydrogenase UdhA, contrast, catalyzes

on but reaction from NAPDH to NADH. This function is not necessary during growth glucose

of which a of critical on substrates such as acetate, the catabolism generates higher proportion

NADPH.

constitute and Vertebrates contain so-called phosphagen kinase systems, which spatial demand. In to overcome of fluctuating energy temporal energy buffering systems periods

kinase in Saccharomcyes cerevisiae that Chapter 4 we install such phosphagen systems

show the functional of arginine normally does not contain such systems. We then expression

kinase Environmental conditions of fluctuating kinase, one particular phosphagen system.

S. cerevisiae in chemostat culture energy demand are installed by growing glucose-limited

intracellular ATP level was with interrupted feeding. In the absence of glucose feeding, the in but it was stabilized at a much higher level shown to drop significantly in the control strain, that buffering the arginine kinase-expressing strain. Our results demonstrate temporal energy

kinases that can be transferred to phylogenetically very is an intrinsic property of phosphagen

distant organism.

7 of kinase in the The 5th chapter affords then a short study on the expression arginine

that kinase significantly bacterium E. coli. Specifically, we could show arginine expression other that was to exert among improves the recovery after a transient pH stress, anticipated effects and ATP stress.

8 che si "GH ostacoli sono quelle cose spaventose scorgono quando si perde di vista il proprio obiettivo"

Henry Ford

SOMMARIO

l'NADPH nei L'argomento scientifico di questa tesi è come venga generato in con sistemi per FATP microorganismi e come questi possano venire equipaggiati tampone

lo stress L'ATP e i nucleotidi a grado di fornire una protezione transitoria durante energetico.

metabolismo cellulare e l'anabolismo al piridina sono cofattori essenziali del collegano di catabolismo. Mentre l'NADH è utilizzato quasi esclusivamente per la produzione energia,

biosintetiche. l'NADPH fornisce potere riduttivo per le reazioni

tramite la costruzione di mutanti e l'analisi Nel Capitolo 2 e nel Capitolo 3 mostriamo coli hanno ruoli dei flussi metabolici che le due transidrogenasi del batterio Escherichia

membranare PntAB è un distinti nel metabolismo dell'NADPH. La transidrogenasi catalizzando la importante sorgente di NADPH durante la crescita esponenziale su glucosio, necessaria durante la transidrogenazione dell'NADH in NADPH. Questa funzione non è

una in corne il cui catabolismo genera crescita su glucosio, ma è critica su substrati l'acetato,

proporzione più NADPH.

che costituisce un I vertebrati contengono il cosddetto sistema di fosfagene chinasi,

momenti di fluttuazione energetica. Nel sistema tampone spaziale e temporale per affrontare chinasi nel lievito Saccharomyces Capitolo 4, installiamo un taie sistema di fosfagene mostriamo funzionale délia cerevisiae che ne è normalmente privo. In seguito l'espressione

délia délie chinasi. Condizioni arginina chinasi, un particolare membro famiglia fosfagene crescendo S. ambientali che causano fluttuazioni nella domanda di energia vengono generate

con alimentazione discontinua. In cerevisiae in un medium di coltura limitato in glucosio

il livello intracellulare di ATP diminuisce in assenza di una alimentazione di glucosio,

ma risulta stabilizzato e mantiene un livello più maniera significativa nel ceppo di controllo, chinasi. I nostri risultati dimostrano quindi corne alto nel ceppo in grado di esprimere arginina intrinseca délie chinasi la funzione di tampone temporale sia una proprietà fosfagene

trasferibile ad organismi fîlogeneticamente molto distanti.

9 chinasi nel batterio Il 5° capitolo affronta un brève studio suU'espressione dell'arginina chinasi migliori in E. coli. Nello specifico, mostriamo come l'espressione dell'arginina

transiente da conosciuto per generare tra maniera significativa il recupero da uno stress pH, gli altri effetti uno stress da ATP.

10 CHAPTER 1

General introduction

11 NADH TWO FACTORS LINK CATABOLISM TO ANABOLISM: ATP AND

need to survive In principle, all living organisms have two goals. Firstly, they In living environmental challenges, and secondly they need to generate offspring. every

human all cellular organism, from the most primitive and simple microorganism to being, activity relies on the availability of energy. is adenosine The universal elementary energetic quantum for biological systems

issue in the universal nature of triphosphate (ATP), this being a fundamental and dogmatic

be to the numerous biochemistry (112). Its hydrolysis is highly exergonic and may coupled

them to In living endergonic cellular processes so as to drive completion. non-photosynthetic

substrate-level or by oxidative organisms, ATP can be generated either by phosphorylation phosphorylation.

substrates to smaller molecules by a series Energy is provided by breakdown of growth called catabolic Here, of enzymatic reactions, organized as a network of so pathways. nucleotides NADH and reductive charge is produced in form of the two reduced pyridine

the two are not metabolically NADPH (Figure 1). Although structurally very similar,

to the interchangeable. In fact, whereas the reducing power of NADH is exploited accomplish the free of otherwise endergonic ATP synthesis, NADPH is involved in utilizing energy

such as the metabolite oxidation for otherwise endergonic reductive biosynthesis, e.g.

between NADH and biosynthesis of fatty acids. This ability of the metabolism to differentiate towards their NADPH is guaranteed by the high degree of specificity of dehydrogenases

respective cofactors.

catabolic pathways Acetate

NADH

nadp+ Carbon source ("i^pj^toryjchaïnj isj^pjj

- Biomasse ANABOLIC PATHWAYS

NADPH with anabolism and catabolism. Figure 1. Link of the two pyridine nucleotides NADH and

free of metabolite In oxidative phosphorylation (also called respiration) the energy

substrates are oxidized by oxidation is exploited to synthesize ATP. In general, organic

12 as well as reductase activities. Thus, enzymes exhibiting specific dehydrogenase quinone which deliver the to electrons are transferred to mobile quinones, reducing equivalent

but also transfer terminal oxidoreductases. The latter enzymes are not only quinole oxidases,

free of this electrons transfer the reduction potential to specific electron acceptors. The energy

the membrane in which the respiratory is conserved by transporting H+ on the outside of

H+ across the membrane is complex is integrated. The generated electrochemical gradient ATP (H+-ATPase), then exploited by the membrane-integrated proton-translocating synthase which is able to generate ATP from ADP and free phosphate. therefore influenced by the The efficiency of oxidative phosphorylation is strongly

the first of During aerobic availability of the organic substrate oxidized in step respiration. with this reduction essential to provide the cell growth on glucose in Escherichia coli, power,

is reduced nicotinamide adenine dinucleotide (NADH). energy in form of ATP, supplied by

and the electrons are transferred This is oxidized to NAD+ by NADH dehydrogenases I and II

by terminal oxidases to oxygen (107,163).

on on the considered, but also We shall not forget, however, that depending organism substrates be utilized by the growth condition and substrate availability, many different may electron donor be utilized, e.g. dehydrogenases to oxidize quinones. In E. coli, other may if cells has phospholipids to formate as oxygen is not present, or glycerol-3-phosphate

breakdown (107).

nucleotides includes a series of The central role played by ATP and reduced pyridine

both and non- involvements of these molecules in cellular regulation, allosterically and involves a flow through ADP allosterically. Because ATP synthesis and utilization cyclic roles in fueling reactions AMP, it makes sense to find that all three adenylates play regulatory the Some are regulated primarily by as well as in biosynthetic pathways. enzymes

to the ratios or concentration of ATP, ADP or AMP; others respond [ATP]:[ADP] the the cellular can be expressed defining [ATP]:[AMP]. A summary index of energy status,

energy charge as

[ATP] + \[ADP] E_ [atp]+[adp]+[amp]

bonds in the Energy charge reflects the relative number of high-energy phosphate

on a number of enzymes. adenylate pool and the aggregate of the controls by adenylates large is one-half the The compound ADP, with a single anhydride bound phosphate, assigned

a reason, in charge value of ATP. The activity of adenylate kinase provides physiological

13 one-half the value of ATP. addition to the mathematical one, for assigning to ADP charge

is in E. the reaction This enzyme, the activity of which quite high coli, catalyzes

AMP + ATP <=± 2 ADP

1. ADP and ATP with a stoichiometry of 2 to thereby establishing an equilibrium between

from the actual intracellular concentrations of the three The energy charge is calculated

of a cell can adenylates, but it reflects their relative proportions. The energy charge vary of bacteria mathematically from 0 (all AMP) to 1 (all ATP); but in fact, the energy charge and it is invariant with under normal conditions will not lie outside the range 0.87 to 0.95,

decreases of a cell of a substrate for long periods growth rate. The energy charge deprived the cell dies. slowly; when it reaches a value of approximately 0.5, nucleotides in cells The necessity for the existence of two species of pyridine

of three facts. First, undergoing aerobic metabolism seems apparent as a consequence

constants near 1. dehydrogenase reactions involving pyridine nucleotides have equilibrium

are the reactants, in Secondly if in fueling reactions, oxidized pyridine nucleotides their function in reduced form. Thus, biosynthetic reactions and energy generation, they exert nucleotide must be largely for fueling reactions to proceed, the participating pyridine pool within the cell, most oxidized; for biosynthetic reactions, it must be largely reduced. Indeed, Useful of the NAD is in the oxidized form and most NADP in the reduced form. parameters

oxidoreduction state are

[NADH] ARC (anabolic reduction charge): ARC = [NADH]+[NAD+]

[NADPH] CRC (catabolic reduction charge): CRO [NADPH]+[NADP+]

and the ARC is The CRC in growing cells is maintained at a low level of 0.03 to 0.07,

approximately 10-fold higher.

The levels of these charges undoubtedly play a vital role in coordinating fueling the reactions with biosynthesis. Bacteria may set the levels in two ways. Firstly, by regulating

an called relative pathway usage, as depicted below. Secondly, by utilizing enzyme

section. Here a vectorial transhydrogenase, which will be extensively described in a separate

in a reaction (proton transport) is coupled to a scalar reaction (oxidoreduction), resulting On of the change of the apparent equilibrium constant of the redox system. energization

calculated as mitochondrial membrane, the mass action ratio T for nicotinamide nucleotides, [nadph)[nad+]

~ [nadp+\-[nadh]

can reach values up to 500.

14 CELLULAR METABOLISM THE CENTRAL CARBON METABOLISM IS THE CORE OF

that are The complexity of the catabolic network is due to the different possibilities of metabolites. In fact, where main often available for a given conversion or synthesis

of the carbon flux be redirected to pathways are branched, a significant portion may is the central carbon metabolism alternative pathways. The core of substrate breakdown (PP) (Figure 2). Here, glucose is converted to trioses either by glycolysis, pentose phosphate The trioses formed either by glycolysis or by pathway, or Entner-Doudoroff (ED) pathway.

which is either accumulated as acetate or enters ED Pathway are then converted to AcCoA, the tricarboxylic acid (TCA) cycle.

their intermediates that serve also as starting The so cited pathways differ not only by but also different point for the biosynthesis of more complex cellular building blocks, provide

and substrate-level of types and amounts of pyridine nucleotides, provides phosphorylation

not and PP provides ADP to ATP. Glycolysis provides NADH but NADPH, pathway nucleotides, but NADPH but not NADH. The TCA cycle provides both reduced pyridine

does not reductive charge. The NADH to a larger extent, while the ED pathway generate any of fundamental for branching points which connect two pathways are therefore importance of ATP is control of NADPH and NADH production. Substrate-level phosphorylation and in overflow provided in glycolysis by pyruvate kinase, in TCA by succinate thiokinase,

metabolism by acetate kinase. of E. coli is by One of the main branching points in central carbon metabolism surveyed

the of the and first enzyme of the two enzymes: phosphoglucose isomerase, product pgi gene of the and glycolytic cascade, and glucose-6-phosphate dehydrogenase, the product zwfgene

These two are well conserved throughout the first enzyme of the PP pathway. enzymes different evolution and their cellular production rate varies very little during growth phases

reversible conversion of 6-phosphate (56). Phosphoglucose isomerase catalyses the glucose both in it is as a dimer or tetramer (107), (G6P) to fructose 6-phosphate (F6P), and present have different functions, but it periplasm and cytoplasm. It is unclear whether these two forms

and are not correlated with the is known that the dimer is not the precursor of tetramer, they

under anaerobic conditions (135). localization (135). The expression rate of pgi is doubled

wild on but Mutants lacking phosphoglucose isomerase grow one third as fast as type glucose these mutants on substrates that do not G6P, pgi rapidly on fructose (57). If grown provide

to be not essential still grow, since G6P appears (56).

15 ATP NADPH co2

^ GLUCOSE < > G6P

FADH

«W ^~*ATP NADH Y^^NADH Acetate-*—:,— AcCoA t ATP

Pyridine nucleotide transhydrogenase(s) NADPH UÄ\solub,e TCA cycle and glyoxylate shunt NADPH + NAD+ NADH + NADP+ f PntAB/inembrane-bound

The arrows indicate reaction Figure 2. Reaction network of E. coli central carbon metabolism. physiological 3-letter code. Abbreviations: G6P, 6-phosphate; F6P, directionality and enzymes are indicated by their glucose fructose 6-phosphate; 6PG, gluconate 6-phosphate; KDPG, 2-keto-3-deoxy 6PG; P5P, pentose 5-phosphate; PGA, 3- E4P, erythrose 4-phosphate; S7P, sedoheptulose 7-phosphate; GAP, glyceraldehyde 3-phosphate; OAA, phosphoglycerate; PEP, phosphoenolpyruvate; AcCoA, acetyl coenzyme A; OGA, oxoglutarate; soluble oxaloacetate; Eda, KDPG aldolase; Edd, 6-phosphogluconate dehydratase; UdhA, transhydrogenase, isomerase; PP PntAB, membrane-bound transhydrogenase; Zwf, G6P dehydrogenase; Pgi, phosphoglucose Entner-Doudoroff TCA, tricarboxylic acid. pathway, pentose phosphate pathway; ED pathway, pathway;

and is G6P dehydrogenase (Zwf) catalyses the irreversible conversion of G6P to 6PG, is detected in normally expressed constitutively. However, an increase in the expression rate with the increased need for NADPH the presence of superoxide agents (56), consistently show during oxidative stress. Mutants lacking G6P dehydrogenase do not any significant

slower on changes in growth rates on glucose and fructose, and grow only slightly gluconate

(55,93).

16 A second relevant branching point, is situated immediately after the step catalyzed by

of PP where 2 G6P dehydrogenase. Here 6PG may either flow to subsequent steps pathway,

ED where neither NADPH moles of NADPH are produced per mol of 6PG, or enter pathway,

ED of E. coli more than nor NADH are produced. Studies on the pathway began thirty years

found renewed interest. It is that the ED ago (178), but only recently this pathway thought

in E. coli and other enterobacteria to pathway had an essential evolutionary role allowing

in which acids are the main substrates (147). Until occupy the intestine of mammalian, sugar

determine which of these substrates is the most now, however, it was not possible to important for intestinal colonization (114). and the 2- The ED pathway consists of 6PG dehydratase (encoded by the edd gene) keto-3-deoxy 6PG (KDPG) aldolase (encoded by the eda gene). Furthermore, two transport

The former one is encoded and phosphorylation enzymes for gluconate are present. by gntU, The latter and exhibits low affinity, whereas the other, encoded by gntT, exhibits high affinity.

role with low cell The one is inducible by gluconate and has a main during growth density.

the of the which is expression of these genes is negatively controlled by product gntR gene, induced by gluconate. Other gluconate transporter and phosphorylation systems are known, unanswered Other roles for Edd but The question of why so many are needed is still (114). and and Eda have also been proposed, including respiration recovery after SOS-response

have been that their neutralization of some toxic compounds. Eda mutants generated, showing growth rate does not change, compared to the wild type, when grown on glucose or fructose,

Since no cofactors are needed in ED but it is reduced to one third when grown on gluconate.

and ED should be a "fast" if pathway, no cofactor recycling is necessary, pathway pathway, compared to the PP pathway (178).

the Regulation of the balance between NADH and NADPH by

TRANSHYDROGENASE SYSTEM

reduction Pyridine nucleotide transhydrogenase catalyzes the stereospecific transfer of

to the equivalents (hydride ion equivalents) between NADH and NADPH, according

following reaction

nH+0Ut + NADP++ NADH <=± NADPH + NAD++ nH+m

the two classes of Depending on the stereospecificity of hydride transfer, is found in the transhydrogenases have been identified. The AB transhydrogenase activity

17 bacteria as well cytosolic membrane of many heterotrophic and photosynthesizing (82,111),

in of and in mitochondria of as in parasitic protozoans (160,176), chloroplast plants (164), the animals. It couples the translocation of a proton (n=l) across the membrane to specific NADP+ and of the transfer of both the 4A hydrogen from the nicotinamide ring of NADH to

4B hydrogen from NADPH to NAD+ (34,84), and it is therefore referred as energy-linked or both proton-translocating (71). The BB transhydrogenases transfer the 4B hydrogen from without to NADPH and NADH through a ping-pong bi-bi-reaction mechanism, coupling The latter proton translocation (n=0) (36,164), therefore being referred as non-energy-linked.

E. coli Pseudomonas activity is found only in a few heterotrophic bacteria such as (13), spp.

(58), and Azotobacter vinelandii (164). membrane of animal Energy-linked transhydrogenases are located in the inner show mitochondria and in the cytoplasmic membrane of many bacteria (84), and high

E. structural, but important differences in polypeptide organization (111). The enzyme from

of the of animals but coli has an amino acid sequence similar to that transhydrogenase higher

acids is split into the PntA and PntB polypeptides (a and ß) having 510 and 460 amino residues, respectively, organized in an oc2ß2 unit (35). All proton-translocating which transhydrogenases have three major structural components, designated dl, dll, and dill,

which bind NAD+ and NADH, and probably assemble with a universal topology. Both di, side of dill, which binds NADP+ and NADPH, protrude from the matrix, on the cytosolic di membrane. The binding-change mechanism by which the hydride transfer occurs between

vitro and dill domain has recently been clarified by X-ray cristallography and separate in

with no bound the studies of domains, as illustrated in (9,10,83,84). Briefly, substrate,

which is switched to 'occluded' substrate dl:dll:dlll assumes an 'open' conformation, by Structural studies binding. So, di and dill come closer together and hydride transfer occurs.

and alanine have a common on di have suggested that transhydrogenase dehydrogenase may

of the ancestor, which was recruited by a membrane protein (the equivalent dll/dlll) during

evolution of transhydrogenase (85). Domain dll is membrane integrated, and is split in dlla

are twelve trans-membrane and dllb, part of the PntA and PntB peptides, respectively. There

The structure of dll domains that form a channel through which proton translocation occurs.

and the mechanism of proton-translocation have been studied extensively (10,71,72,83- of 85,111). An interesting feature has been proposed, by which the degree of protonation

residues in the dll domain regulates the activity of the energy-linked transhydrogenase (61).

Besides the possible involvement of the cAMP/CRP system (40), this is the only regulatory

issue known concerning transhydrogenase.

18 can either use or Like most proton pumps, the proton-translocating transhydrogenase The on the direction of the reaction catalyzed (155). generate a proton gradient, depending

absence of a hydride transfer from NADH to NADP+ (forward reaction) is slow in the proton of gradient, but is nevertheless coupled to proton extrusion (111). In the presence proton reaction is increased 5-10 gradient generated by another proton pump, the rate of the same by that of the times. The opposite (reverse) reaction is faster and its maximal velocity approaches proton gradient driven forward reaction (97).

of a few Non-energy-linked transhydrogenases are the intriguing property only each other but is organisms. This class of transhydrogenases show high sequence similarity to structurally unrelated to energy-linked transhydrogenases. They are soluble flavoproteins form containing flavin adenine dinucleotide (FAD) and are remarkable for the ability to large This polymers, which in E. coli are composed by seven or eight monomers (12,13). enzymes

and inhibited NADP+ The of a soluble are generally activated by NADPH by (13). presence of E. coli transhydrogenase in E. coli escaped identification till 1999, when a sequencing error the of the udhA was found to be homologous to genome was discovered, and the product gene first soluble transhydrogenase from P. fluorescens (13). E. coli is thereby the organism

and a membrane-bound nucleotide reported to possess both a soluble pyridine has been identified transhydrogenase. Later on (25,26), a non-energy-linked transhydrogenase

in the same tissue no transhydrogenase in potato tuber and pea leaves, whereas energy-linked

have been on the further catalytic activity was detected. So far, few investigations performed

and structural properties of soluble E. coli transhydrogenase (12,13).

has been a source Ever since its discovery, the physiological role of transhydrogenases essential function of speculation and matter of controversy. Earlier theories hypothesized an electrochemical Since, however, in NADPH generation or maintenance of the potential. exist for both transhydrogenase deletion is not lethal (179) and alternative cellular sources is excluded. On NADPH production and generation of the proton gradient, an essential role

Hoek and a the basis of a thermodynamic and kinetic approach, Rydström proposed protective

of demand for NADPH, function in maintaining the proton gradient or under conditions high In toxic or after anoxia E. such as during oxidative stress due to agents reoxygenation (75).

of in coli NADPH is not only used in reductive biosynthesis, but is also great importance

the mechanism to reducing glutathione, which is an essential component of response

oxidative stress.

those in Besides E. coli, which at that moment is a unique and particular case, organisms

the linked which transhydrogenases are found, have either the energy-linked or non-energy

19 to transhydrogenase. By considering thermodynamic argumentations, Voordouw attempted

the two He that provide a rationale for the distribution of transhydrogenases (164). suggested of NADPH where the energy-linked transhydrogenases are responsible for the formation The catabolism is NAD+-linked, such as in E. coli or in animal mitochondria. non-energy- linked transhydrogenase, in contrast, is supposed to act in reoxidizing NADPH in organism This however, cannot with partially or completely NADP+-linked catabolism. hypothesis,

in E. coli. explain the simultaneous presence of both transhydrogenase show The two transhydrogenase isoforms found in other micro-organisms strong

similar function for structural and catalytic similarities to the E. coli enzymes suggesting a

still no has been conducted in this these enzymes in all these organisms, although study that is direction. In animals and plants, transhydrogenase is present in a mitochondrial isoform level of more similar to the E. coli membrane-bound one, and shows tissue-dependent

tissues are transcription (2). Due to the reduced environmental variations to which e.g. human

the NADPH to NADH subjected, the presence of a soluble transhydrogenase catalyzing

reaction does not resemble to be necessary.

have no of Organisms without a transhydrogenase are thought to overproduction NADH NADPH(164). Imbalance in the reductive charge distribution between NADPH and

Some of them were is, however, a common problem in transhydrogenase-lacking organisms.

which able to overcome the problem by using a system of two or more enzymes, together may

of NADPH metabolism are give a transhydrogenase-like activity (25). Special cases yeasts,

and in particular S. cerevisiae where the low utilization of the PP pathway (51) causes an of increased production of NADH (109). In anaerobic processes aimed at the production

it is utilized to ethanol, NADH cannot be exploited by the respiratory chain, and therefore

synthesis glycerol. The yield of ethanol is strongly reduced and therefore also the rentability

exhibit both the of the process. In S. cerevisiae, which do not transhydrogenase activity (23),

E. coli were based membrane-bound as well as the soluble transhydrogenase from expressed,

from NADH to on the idea that the reductive charge would be transferred NADPH, thereby

both tend to improving the productivity. In both cases it was showed, however, that enzymes

In another UdhA was shown to catalyze the reaction on the opposite direction (1,108). study, Another compensate for the lethal Pgi knockout in S. cerevisiae (51). biotechnological

was found as a cofactor system application of the same soluble transhydrogenase regeneration

of in a cell-free system for the biological production hydromorphone (12). of the Further biotechnological applications of transhydrogenases rely on our knowledge the role of mechanisms related to cofactor metabolism. The investigation of

20 and the non- transhydrogenases in E. coli, the sole organism with both the energy-linked in this direction. To accomplish energy-linked transhydrogenases, may be a fundamental step and examine the this task it is essential to have the tools to modify cofactor metabolism effects on the whole metabolism.

variations Flux analysis is a powerful tool to track metabolic

include the Strategies for the dissection of the in vivo role of a certain gene or enzyme non-standard conditions. Classical generation of deletion mutants and of strains grown under

often not able to the intracellular approaches, such as physiological analysis, are depict in which often occurs at a metabolic level and are reflected consequences of these procedures, of metabolic fluxes, metabolic fluxes. Metabolic flux analysis is aimed to the determination

for the of metabolic responses and is therefore a potentially powerful tool investigation global

(131). of Flux analysis is commonly performed by stoichiometric flux balancing quantitative

is used to solve a of physiological data (158). This method uses linear algebra, which system intracellular fluxes. Since linear stoichiometric relationships, where the unknowns are

to solve the for a normally the system is underdetermined, it is often not possible system

single solution.

is the use of tracing. A A recent improvement in metabolic flux analysis isotope ' of the C- commonly used method involves the feeding with 13C-glucose, and analysis

the distribution of carbon fluxes at labeling pattern in proteinogenic amino acids. Here the of amino branching points of the central carbon metabolism is reflected in labeling pattern

matrix to reduce the number of constraints. acids and this may be applied to the stoichiometric

of isomers or The practical question of the detection of the relative abundance isotope be addressed in isotopomers, reflecting the labeling pattern of amino acids, may many ways.

at of metabolic products, In a first approach, fractional 13C-enrichments specific positions

carbon are determined using ID resulting from growth on specifically 13C-labeled substrates,

1 labeled and C NMR spectra of the purified compounds (41,101). Alternatively, uniformly in the natural abundance carbon substrates are fed. The information contained analyzed

which have not been metabolic products is then the fraction and position of carbon bonds correlation NMR broken during synthesis from the labeled substrate, analyzed by 2D [13C,!H] labeled substrates have be combined spectroscopy (132,148,150). Specifically and uniformly of deduced information Further in a method with allow to increase the amount (134).

21 with mass (MS), techniques involve the combination of NMR spectroscopy spectrometry in order to additional use of 2H or 180 labeled substrates, which may be complemented by the further extend the collected information (149).

methods to Considerable efforts concentrated on the development of computational

of a vast obtain qualitatively reliable flux analysis, which involves the laborious integration

close as to the number of biochemical and physiological information in a model as possible

the determination of net metabolic fluxes from real system (171). Recently, a new method for will be used in this work to MS data and physiological data has been developed (52) and central carbon metabolism analyze the rate of formation of pyridine nucleotides from pathways and through the transhydrogenase reaction.

Cellular ATP buffering: the Phosphagen kinase system

is often a for Despite its key role in cellular energetics, the availability of ATP problem the of metabolism. For example, ATP production rate depends significantly on availability

conditions. On the other hand, the energetic needs substrates or on optimal environmental

of activation of ATP tasks, e.g. ion transport, or may suddenly increase because consuming where ATP capacity possibly stress responses. Under optimal growth conditions, synthesis

the cell to stock ATP for exceeds the cellular ATP needs, it would be very advantageous for

metabolism cannot face such moments where ATP consumption increases and augmented

well as the ATP to ADP energetic needs. Unfortunately, both ATP and ADP concentration, as

have to be constant under normal ratio exert essential regulatory functions and therefore kept since the liberated by physiological conditions (59,167). This is an important issue, energy local nucleotide concentration, i.e. on ATP hydrolysis is not fixed but strongly depends on the

how far the hydrolysis reaction is displaced from thermodynamical equilibrium (86).

was faced bacteria and primitive A more or less sudden drop of ATP availability by central unicellular eukaryotes by optimizing the efficiency of carbon source transport systems,

chain. as multicellular organisms evolved, carbon pathways as well as respiratory However,

cells and showed an their specialized tissues and cells like e.g. muscles, nervous spermatozoa

with a increased energetic need. The tasks accomplished by these new tissues are coupled which could not be entirely sudden and increased need of energy in form of ATP,

of compensated improving the kinetic or thermodynamic parameters energy producing

or subcellular pathways. Adaptations, like polyphosphate kinase in plants (157)

22 needs. the need for a fully new compartmentation could not fully cover these energetic Thus,

stock in another form. system arose, allowing the possibility to energy

transfer the of ATP to a A class of enzymes arose, able to phosphoryl group without loss of metabolically mostly inert molecule (a so called phosphagen) significant if to constitute a of such energy" compound energy and with the possibility large pool "high

amounts of are ATP production rate exceeds the need for energy. When, later, large energy work in the direction, thereby suddenly required, these enzymes will mainly opposite

refer to this as the reconstituting ATP from the phosphorylated phosphagen pool. We ability temporal energy buffering. chemical structure Eight phosphagens have been identified, all with the same basic

HN R

I (with Ras in Table 1) NH2

for where the guanidino group is the target phosphorylation (therefore phosphagen kinases are also called guanidino kinases) (43).

R Phosphagen

NH-(CH2)3-CHNH2COOH Arginine NH-CH2COOH Glycocyamine (Guanidinoacetate) N(CH3)-CH2COOH Creatine NH-(CH2)2-S03H Taurocyamine NH-(CH2)2-S02H Hypotaurocyamine NH-(CH2)2-HP04CH3 Opheline NH-(CH2)2-HP04-CH2CH(NH2)COOH Lombricine NH-(CH2)2-HP04-CH2CH(N(CH3)2)COOH Thalassemine

Table 1. The eight known phosphagens.

kinase is the The general idea is that the system formed by arginine and arginine (AK)

kinase This to be a common ancestor for the other guanidino systems (170). appears

as reasonable hypothesis by considering how AK evolved as first, probably simultaneously

the sole of a specific guanidino the appearance of metazoans (Figure 3). Moreover, presence the other the kinase is sufficient to make the system functional, whereas for phosphagens

for the and a simultaneous evolution of both a specific biosynthetic pathway phosphagen

in contrast to these phosphagens specific guanidino kinase was necessary. Therefore, arginine,

can be considered as metabolically 'inert' phosphagen.

23 lEchinodermsl

iTunicatesI

iHemichordatesI iDeuterostomi iCephalochordatesI

I Vertebrates Euk. protozoans (Trypanosoma cruzi, Paramecium caudatum) IMolluscs

4 HeHorizontal gene transfer iFlatworms

lEcdysozoa ISipunculidsl (arthropods, nematodes, pnapulsds) lLophotrocozoal

iCnidarians lAnnelids

I Ponferal (sponges) Echiuroidsl Metazoans iBrachiopods (marine worms) Phoronids

Metazoans. Abbreviations: AK, arginine kinase; Figure 3. The evolution of phosphagen kinase systems in kinase; ThK, thalassemine kinase; GK, CK, creatine kinase; HTK, hypotaurocyamine kinase; TK, taurocyamine glycocyamine kinase, OK, ophaline kinase; LK, lombricine kinase (from (43), modified).

and is involved in being Arginine is a proteinogenic amino acid biosynthetic pathway,

A second of (glycocyamine, thus a more "primitive" phosphagen. group phosphagens the of lombricine, hypotaurocyamine) appeared as the result of activity specific metabolic reaction and amidinotransferase (Figure 4). They do not participate to other

of a efficient therefore metabolically inert. A decisive step towards the establishment highly the evolution of guanidino phosphagen kinase system was accomplished by amidinotransferase reaction. Among these methyltransferases acting on the products of the

to which makes of the new phosphagens, creatine (Cr) seems possess unique properties tissues with and phosphocreatine / creatine kinase (PCr/CK) system the tool of choice for high sudden energy demands (43). in Surprisingly, there is evidence for the evolution of AK from a CK ancestor, e.g. The about the driving chordates and in the sea cucumber Stichopus japonicus (43). question

conclusions be drawn by force of such an involution are still unanswered, but some may

well as the kinetic and considering the physico-chemical properties of arginine and Cr, as and the of thermodynamical characteristics of the corresponding guanidino kinases degree

complexity of the system.

24 AMIDINO¬ METHYL- TRANSFERASE TRANSFERASE + + SAH t Guanidine Guamdme SAM--» Methylguamdine Arg -t- acceptor -Ornithine

H-R Hypotaurocyamine Taurocyamine

Guc!!i.uii;of'-;hylphosp!:.-":tv «Opheline

Arginine

Lombricine «Thalassemine

Guanidinoacetic acid -Creatine (Glycocyamine)

SAH, S-adenosyl- Figure 4. Biosynthesis of phosphagens. Abbreviations: SAM, S-adenosyl-L-methionine; homocysteine. For R, refer to Table 1.

of and PCr, Considering the mere physico-chemical properties phosphoarginine (PArg)

due to the of a it is notable that PCr is far more labile than other phosphagens, presence

This eliminates of the resonance states, methyl group vicinal to the guanidino group. many the thereby destabilizing the compound (125). As a direct consequence, apparent equilibrium is around times for constant for ATP formation under physiological conditions eight higher

on the ATP buffering CK than for AK (38,42,94), thereby causing decisive differences increase of behaviour of the two phosphagen kinase systems. A drop in pH causes a strong is much for CK, due to this equilibrium constant for both CK and AK, but the effect stronger the thermodynamic properties of PCr (63,151-153). for ATP is highly Moreover, as mentioned before, the free energy change hydrolysis

concentration. Under conditions and dependent on the variation in ATP optimal physiological values to -60 kJ/mol if ATP hydrolysis is kept far enough from equilibrium, free energy up and stabilizes the of a kinase increases may be attained (86). The presence phosphagen system

to kJ/mol, keeping high and maximal free energy change for ATP hydrolysis up -(75-80) by

PCr can constant ATP concentrations. Because of these intrinsic thermodynamic properties, be considered as the best phosphagen for energy buffering.

activities or In vertebrates, the amount of Cr needed is provided either by biosynthetic

in a two reaction, but can by dietary supplement (175). In mammals, biosynthesis occurs step

25 The first occurs in and it is catalyzed only barely cover the Cr need (Figure 5). step kidneys, and L-ornithine by L-arginine:glycine amidinotransferase, which produces guanidinoacetate is to the liver out of L-arginine and glycine (78,175). Successively, guanidinoacetate exported the where S-adenosyl-L-methionine:N-guanidinoacetate methyltransferase catalyzes

to the blood vessels, thus methylation of guanidinoacetate to Cr (79,175). Cr is then exported and brain. In lower reaching then the other CK-expressing tissues, such as muscles distribution of the vertebrates, the situation differs slightly, due to the different biosynthetic

Cr has been detected. Those organisms enzymes (156). In invertebrates, no biosynthesis

in thus forcing these species to express, however, CK, particularly spermatozoa (43),

accumulate Cr from the feed or from the environment (175).

Creatine fr0m dietary Supplement GASTROINTESTINAL TRACT

Citrulline m

» Creatine Creatine-* Phosphocreatine

UREA CYCLE GAMT Glycine

Guanidinoacetate y y Creatinine Phosphocreatinine Ornithine J Guanidinoacetate

MUSCLE KIDNEYS LIVER

thicker arrows Normal arrows indicate enzymatic reactions, Figure 5. Biosynthesis of creatine in mammals. reactions. Enzyme names are indicate transport issues, and dotted arrows indicate spontaneous non-enzymatic GAMT, SAM:N- enclosed in boxes. Abbreviations: AGAT, L-arginine:glycine amidinotransferase; guanidinoacetate methyltransferase; CK, creatine kinase.

Na+- and Cr Transport of Cr from and to the tissues is operated by Cl~-dependent

is one of the factors influencing the activity transporters (142). Intracellular Cr accumulation allosteric inhibitors of AGAT, and thus of these transporters (100). Cr and ornithine are strong and the efficiency of of Cr biosynthesis (140,175) The intracellular Cr turnover (indirectly) of Cr transport (directly) exert a strong influence on the biosynthesis (175).

the need of a (175). Cr is degraded spontaneously to creatinine, without specific enzyme cells. cells have to continuously This implies two important consequences for Cr-using First, accumulates in the intracellular import Cr to keep the Cr pool constant. Secondly, creatinine to Cr. effect because of the much lower solubility with respect space and exerts a toxic

to eliminate this Therefore, Cr-dependent organisms have to dispose of a system degradation-

product of the CK system.

26 for AK are with a much system. Organisms using arginine as a substrate dealing simpler it is the action Arginine is either synthesized in the urea cycle from ornithine or imported by

amino acid and a basic of different kind of transporters (39,177). Being a proteinogenic

central role in metabolism. Thus, the AK constituent of some cellular activities, arginine has a

concentration. reaction can not cause a too drastic change in arginine

of kinase Besides temporal buffering, a second essential property phosphagen systems in and cells (165- is the ability to function as an energy transport system especially large polar distributed in the cell, the diffusivity is a 167). For a molecule to be easily and efficiently of crucial issue. Being strongly dependent from the molecular size and mass, the diffusivity the phosphagens is doubled for PCr and PArg with respect to MgATP. In addition, diffusivity for fast of PCr is again 10% higher than for PArg, making PCr the best suited phosphagen

energy transport (45). the intracellular The largest difference between the AK and the CK system concern

two targeting and the tissue specificity of the enzymes. CK is known to exist as cytosolic in skeletal muscles and isoforms, M-CK and B-CK, which may form homodimers (MM-CK

heart, and BB-CK in neural tissues, brain, and other muscles) or heterodimers (in embryonic sarcomeric and uMtCK, tissues). In addition, two mitochondrial isoforms (sMtCK,

homodimers. The kinetic and ubiquitous) exist as homooctamers and thermodynamical, the structural properties of these isoenzymes have been extensively studied during past years

(for reviews see (33,105,128,133,166,174)).

in some crustaceans However, mitochondrial activity has also been found for AK

the of a (44,73,74), and in the tobacco hornworm Manduca sexta (30), and presence the crabs mitochondrial AK in Drosophila melanogaster has been suggested (106). In

be located in the Limulus polyphemus and Callinectes sapidus this activity appears to

this does not to a true intermembrane space of mitochondria (116). However, correspond mitochondria mitochondrial AK isoenzyme, since the AK isolated from cytosol and form

How of the AK is targeted to the appear to be the product of the same gene (74,106). part

mitochondria remains unclear.

introduction of a The availability of cytosolic and mitochondrial isoforms allows the

for or energy transport new concept: a phosphagen kinase system spatial energy buffering, where the (8,65) (Figure 7). MtCK is present in the intermembrane space of mitochondria, the octameric form functionally interacts with the adenine nucleotide translocator (ANT) on

inner membrane and with the oligomeric porin on the outer membrane (17,127,167,173,174).

or The cytosolic dimeric isoenzyme, in contrast, is present either as free cytosolic enzyme

27 The idea of a buffering functionally coupled to ATPases and glycolytic enzymes (167). spatial

ATP in mitochondria by oxidative by CK is based on the observation that the synthesized

it to PCr. The phosphorylation will be channelled directly to MtCK which converts porin

and serves as a substitute complex will then export PCr into the cytosol, where it accumulates and in the for of cytosolic CK to keep a high ATP/ADP ratio, globally locally optimal range tissues with ATPases (81,167,173). CK is prominently present not only in high energy

with the most fluctuations in workload. requirement, but more interestingly in tissues abrupt reduce the transient times needed On this basis, it was hypothesized that the CK system may

reaction has a much turnover rate to refill the ATP pool (174). In fact, the cytosolic CK higher

metabolite occurs with than ATP utilization or consumption rates. In addition, channeling

to sarcomeric Ca - MiCK coupled to porin and ANT (173), or cytosolic CK, e.g. coupled

in metabolic this reduction of transient times would pumps. Besides the increase efficiency,

metabolic such as also improve the transition between different work loads and states,

occurring during transient stress (165).

ATP consuming enzymes

ADP ATP

oxidative phosphorylation

creatine MtCK, mitochondrial creatine Figure 7. Energy transport by CK system. Abbreviations: CK, kinase; kinase (from (165), modified).

28 of A further relevant feature of the CK system is pH buffering (90,167). In case large in occur as result of ATP hydrolysis, energy utilization, intracellular acidification may muscle lactate particular during severe muscle exercise. Under anaerobic activity, production in loss of fibres involves further liberation of protons in the cytoplasm (90). This may result

disturbance as well as of the contraction force (32), as a consequence of the pH gradient

The of the PCr buffer influence of low pH on proteins and membranes (90). depletion by

ATP molecule. Thus, the cytosolic CK involves the utilization of one proton per generated

a replenishment of the ATP pool during intense energy utilization is accomplished by

for a reduction of local acidification (167). Although AK has not yet been investigated

with sites of possible role in proton buffering, it was shown that AK interacts high energy

its kinetic make of AK a less utilization, as CK does (124). Moreover, although parameters

AK exerts a efficient system than CK, it seems theoretically possible that proton buffering

function as well.

of As mentioned above, CK interacts with some proteins for a direct channeling

substrates and products of the reaction. The interaction of CK with AMP-activated protein

of metabolism in kinase (AMPK) may moreover exert a co-ordinated regulation energy

skeletal muscle, where AMPK is expressed at high levels. In response to metabolic stresses as involved in the intracellular ATP/AMP ratio falls, AMPK is activated, inhibiting enzymes

is exerted at energy-consuming tasks, such as biosynthetic pathways (69,70). Regulation post- sites transcriptional level by sequence-specific phosphorylation (37,172), and CK has three

shown how AMPK is able to in where such a phosphorylation may occur. It was moreover

are co-localized in skeletal vitro phosphorylate MM-CK, and that in vivo the two enzymes

and activated a muscle (121). AMPK is part of a protein cascade, being phosphorylated by

distinct kinase (AMPKK), although the regulation of the latter remains to be cleared (28,121). increased AMPK is activated by low ATP/AMP ratio (69,146) as well as it is inhibited by

This is in fact a efficient PCr/Cr ratio in a pH-dependent manner (121) (Figure 8). very

and attributes to AMPK an system by which the cells is able to regulate energy availability

energy-conserving role. which is well suited for By considering all these issues, CK appear to be a system very

These are well high-energy demanding tasks in very complex organisms. organisms very and protected form strong fluctuations of physiological conditions, such as temperature

intracellular pH. Under these conditions the weakness of PCr in term of thermodynamical and the properties is attenuated. The metabolic burden connected with the biosynthesis uptake

29 isoforms of is more than by the of Cr, as well as the many tissue specific CKs, compensated advantages of an efficient energy transport and buffering system.

kinase. arrows indicate enzymatic Figure 8. Regulation of CK activity by AMP-activated protein Gray indicates activation and e indicates inhibition). reactions, black arrows indicate regulatory issues (here, © AMP-activated kinase; Abbreviations: CK, creatine kinase; PCr, phosphocreatine; Cr, creatine; AMPK, protein AMPKK, AMPK-activated protein kinase.

of as exhibited by AK, on the contrary, appears to lack the vast palette functionalities,

well as a role in cannot CK, although less efficient energy transport function as pH buffering

makes of it a more be excluded. The lower dependence of PArg from physiological conditions

influence of the environment, like e.g. powerful tool for organisms or tissues exposed to the

It seems that unicellular were unable sponges or unicellular eukaryotes. however, organisms

therefore on horizontal gene to develop a phosphagen kinase system by themselves, relying and cruzi transfer from host-parasite relationships, as in case of arthropods Trypanosoma

of kinase in unicellular (115). A possible cause for the missing evolution phosphagen systems

from which AK evolved first. organisms may be the lack of the ancestor gene, involution of AK from CK. An These considerations may also explain the possible

of sudden in environmental conditions organism confronted e.g. with the appearance changes

from a more and more or the colonization of a new habitat would take advantage primitive

resistant system, although it is less efficient as is the case of the AK/PArg system.

like or The debate about the possible advantage primitive unicellular organisms yeasts

to be bacteria would take from the availability of an ATP buffering system appears intriguing

caudatum could from the coexistence as well. It was already shown how Paramecium profit

30 will the of both adenylate cyclase and AK in his cilia (110). In the next sections, we analyze kinase in possible biotechnological applications of the overexpression of phosphagen systems different organisms.

Previous studies on heterologous Phosphagen kinase expression

kinase was described The first attempt to heterologously express a phosphagen by

in E. coli as Koretsky and co-workers in 1989 (89). The B isoenzyme of CK was expressed

in vivo and levels of PCr were fusion protein. The enzyme was found to be active high

31 measured by P-NMR when the medium was supplemented with Cr. One year later, a study

of CK in the S. cerevisiae was on the overexpression of the B isoenzyme yeast published.

PCr levels as Cr was fed in the Here as well, in vivo activity was detected by measuring

concentration was observed In both medium, but no corresponding change in the ATP (19).

and PCr was studies, however, it remained unclear whether Cr transport actually occurred

Cr needs synthesized intracellularly. In fact, it is known that determination of intracellular wall extensive washing of cells before disruption in order to eliminate the Cr bound to the cell than the (U. Schlattner, unpublished results). Moreover, the Cr levels measured are higher resulted from solubility limit for Cr (144). Thus, the measured PCr could eventually have

extracellular CK activity leaking from lysed cells.

and to mitochondria. In another study, MtCK was expressed in S. cerevisiae targeted could be detected after Despite the correct localization of the enzyme, no PCr formation

medium supplementation with Cr (E. Furter 1994, unpublished results). Similarly, cytosolic

also in this case no PCr formation was observed CK was expressed in the same organism, but the absence of in correspondence to Cr feeding (27). The possible problem was reported to be

of this see intracellular Cr due to the lack of a transport system (for discussion topic, Chapter

M-CK was in E. coli, 4). The overexpression of the cytosolic enzymes B-CK and attempted Stolz 1995 / N. but PCr was not determined and growth parameters not analyzed (M.

Holmberg 1998, unpublished results).

in mammals and could show Other studies, focussing on the expression of CK plants

of ATP and surprising effects. In perfused mouse liver overexpressing cytosolic CK, no drop

pH occurred under hypoxic conditions as is generally seen in control liver, indicating a

In the same protective effect by CK in a tissue normally lacking CK activity (103). organ,

overexpression of MtCK generated a PCr pool, which could be efficiently exploited (104).

31 PCr The cytosolic B-CK expressed in tobacco plants resulted as well in the formation of a

be observed pool, although no physiological improvement could (50).

Applications of functional Phosphagen systems in biotechnology

the effect of In our work, we have used a new approach to investigate heterologous

expression of functional phosphagen systems, concentrating our efforts in providing sufficient

E. coli and S. amounts of phosphagen to the phosphagen kinase expressed and using

but also a detailed cerevisiae. These hosts allow not only a comparison with previous studies,

the knowledge of metabolism and biochemistry. Moreover, their low pathogenicity and

make of them an ideal to frequent utilization as model strain or in industrial processes system

analyze possible future applications of phosphagen kinases in biotechnology.

aware of We have limited our investigation to the AK/PArg and CK/PCr systems, being

the their structural and functional differences, as outlined before. As already described,

of CK/PCr system has been investigated quite extensively. The vast palette functionalities,

and the kinetic characteristics make of CK the providing a large free energy change for ATP, with most efficient phosphagen kinase to work with. Moreover, Cr is inert, only interacting

CK and not being involved in other biosynthetic pathways. However, the relative instability strict of PCr at low pH makes the CK/PCr system only suitable in organisms able to keep a

to install a mechanism or Cr- control over intracellular pH. In addition, the need Cr-uptake

biosynthesis together with co-expression of different CK isoenzymes contribute to a evaluate considerable burden for the cells. Therefore, it was of particular importance to

of possible negative effect connected to system complexity. Finally, the inhibitory effects should be creatinine have to be considered and, if necessary, a system to avoid accumulation

implemented.

Since the AK/PArg system is not relevant for humans, this system in only poorly

characterized. The overexpression study presented here is the first attempt of a functional

heterologous expression of AK. Therefore, it is difficult to concretely estimate the impact on

the chosen hosts. About the functions other than the temporal ATP buffering one can only

which makes of it a speculate at the moment. AK is also a simpler and monomeric enzyme,

less efficient actor in comparison to CK. Nevertheless, this lack of complexity probably

unicellular hosts. can be confers to the AK/PArg system an improved compatibility with Arg

provided directly as medium component or we can rely on endogeneous synthesis.

32 it is of fundamental By considering expression studies for biotechnological purposes, importance to consider possible drawbacks. In their natural cellular environment, phosphagen consider kinases have been reported to directly interact with many proteins. We shall possible

AMPK is to the SNF1, interactions in our expression system as well. analogous yeast protein

function It is reasonable to consider, both which appears to also exert the same (68-70). We AMPK and SNF1 utilizes the same recognition sequence to phosphorylate target proteins.

sites of analyzed the protein sequence of AK and we found two putative AMPK-dependent difficult phosphorylation (Serl30 and Thr32). Due to the lack of structural data, it is however

AMPK. we cannot to estimate whether these sites are effectively accessible for Similarly,

of a of estimate the impact of phosphorylation on activity, and therefore possible regulation

AK and CK by SNF1 in S. cerevisiae. In E. coli one can neglect such a possible interaction,

AMPK has been identified in this As second since no protein homologous to the organism.

inhibition of the Cr possible drawback, we should consider a possible biosynthesis pathway

is the of the whole in a single by Cr, as described above. The relevant point expression system

in a cell and not in different tissues as it is in mammals, therefore possibly resulting stronger

inhibitory effect.

A functional phosphagen kinase system is expected to provide cells with an energy

In normal batch cultures, buffer be used in case of sudden and intense energy consumption.

In order to find a for the however, this is a condition which rarely occurs. possible application

heterogeneously expressed phosphagen kinase systems it is thus important, to apply affect the conditions which cause serious depletion of the ATP pool or negatively capability will be of ATP formation. Environmental stress is known to generate such situations, as

described below.

Stress mechanisms in E. coli and S. cerevisiae

situations The intriguing ability of certain micro-organisms to overcome the stress in the last found in hostile and quickly changing environments has been extensively studied

between stress decades. This great interest originates mostly form the close relationship

fundamental for mankind. biotechnological response and many activities of importance Many

cheese and bread brewing applications, such as wine production (80,122), (20) making, of stressful conditions (4,123), and industrial processes involving the generation (46,136)

intended as a require robust, stress-resistant cells. But stress resistance not always is positive

is often related to the efficiency property of micro-organisms. Bacterial pathogenicity strongly

33 tract colonization of stress resistance mechanisms, e.g. as in the case of gastro-intestinal by

well on the enteric pathogenic bacteria (99,102). Moreover, food preservation relies as generation of combined stresses to prevent food spoilage (22).

both a stress E. coli can respond to a vast palette of stresses, by activating general

Some of these involve a more or less intense response and more specific mechanisms.

formation or utilization of cofactors. depletion of energy or a disturbance in the energy

acids or of Among these energetic stresses, the presence of weak organic strong inorganic encountered hostile acids is one of the most relevant. Acid stress is also the most frequent

at different levels. condition in enteric bacteria (7), where it may affect cellular activity Strong

inorganic acids lower drastically the extracellular pH, and this variation is reflected more or show maximal less strongly in the cytoplasm, depending on the organism. Since proteins

inhibition or denaturation activity at given pH values, the shift from this optimum may cause

the of But weak of surface or intracellular protein (15), depending on severity pH change.

deleterious effect on microbial and their organic acids are considered to have stronger growth in the mechanism of action is slightly different (53,143). Organic acids enter the cell internal protonated form, which is mostly lypophilic, and deprotonates in the cytoplasm. So, but acidification will occur with the same deleterious effects as for inorganic acids, eventually

with reduced changes in extracellular pH (11). Moreover, the efficacy of weak acid stress,

liberated as it is the case for acetate, relies on the possible toxicity of the so anion, e.g.

of sure interest propionate and benzoate (129). Focussing attention on weak organic acids is

media often occurs as the results of since presence of weak acids in culture by-products

accumulation (14).

As in all situations where survival is menaced, micro-organisms try to face the problem

result on one side from the stable in as efficiently as possible. Acid stress may change

environment. instead of conditions, especially in the case of organism colonizing a new Here, metabolism fighting the adverse conditions, micro-organisms adapted their structure or their which exhibit to the external conditions, evolving into acidophilic or alkalophilic organism,

acid stress occur as a their optimum at reduced or increased pH (14). On the other side, may

environmental variations. In those cases, the consequence of sudden and momentary It is development of resistance mechanism is preferred, which will act only if needed. bacteria and unicellular surprising, how different are the approaches developed by eukaryotes

evidence had been that to actively or passively overcome acid stress. Convincing provided

actors and acid stress resistance arises as the combination of several systems of regulators,

which often overlap (15).

34 intracellular acidification as a To prevent entry of extracellular protons and

from a series of defence mechanisms consequence, micro-organisms take advantage passive cell wall itself. It (Figure 9). In E. coli, the first significant barrier to the environment is the

influx is due to its structure and to porins seems that the efficacy of cell wall against proton it (15,95), but their exact contribution to pH stress resistance is still unclear (15). However,

to reflect external was hypothesized that, in the case ofE. coli, porins are responsible partially of by pH changes in the periplasmic space (22). The buffering capacity periplasm, guaranteed for small influxes. As the large amount of peptidoglycans (14) is able to compensate proton barrier and the periplasm becomes acidified, the cytoplasmic membrane is the second strong of the real defence against intracellular acidification. The low proton permeability well. The cytoplasmic barrier is of fundamental importance for energy transduction as high

relies on its protection level of cytoplasmic membrane from acidification of cytoplasm the flux composition, which provides a net positive surface charge, thereby reducing protons

In E. to acidic (22,66), as well as on the control of ion channels (96). coli, adaptation slightly

membrane to increase the protective pH will cause a change in the cytoplasmic composition

effect of cell membrane (21,31). of extracellular However, all these passive barriers can compensate only small changes acidification weak pH (1-2 pH units) and are even of little help against the cytoplasm by reduced if to organic acids (22). Moreover, buffering capacity of cytoplasm is much compared

membrane influences the periplasm (14), so proton flux through the cell strongly cytoplasmic of acidification is pH, and acidification of cell interior occurs. Active counteraction cytoplasm which extrudes accomplished in first place by proton pumping. One of the actors is ATPase,

molecule of ATP. and co-workers one proton in the periplasm by consuming one Kobayashi under mild acid observed how in enteric bacteria ATPase synthesis was increased up to 3-fold

how acid stress stress (pH 5) (88). It has moreover always been recognized, response strongly of depends from the availability of potassium in the medium (92). In fact, there are two types

concentration. One is an H /K - potassium transporter which affects the intracellular proton in which E. coli cell antiporter, extrudes protons, imports protons and represents the main way net regulates the potassium availability (18,47). However, during recovery from pH stress, no and increase of intracellular potassium was observed (22). In a recent study, Zakharyan

as an H /K - Trchounian suggested the presence of the Kup potassium transporter, acting

this mechanism works for antiporter (180). It was however not clear whether only proton extrude both import and maintenance the proton gradient or if it can also efficiently protons

35 has also been and potassium ions to the periplasm. Moreover, the NhA H+/Na+-antiporter related to proton extrusion form acidified cytoplasm (14,15).

used GABA, Figure 9. Stress resistance mechanism in E. coli. Abbreviations Glu, glutamate, Arg, arginine, Na+/K+ GadC, y-aminobutyric acid, Kup, H+/K+ symporter, TrkA, H+/K+ antiporter, NhaA, antiporter, RpoS, stationary- glutamate/GABA antiporter, GadAB, glutamate decarboxylase, AdiA, arginine decarboxylase, Color code present at neutral phase dependent sigma factor as, HdeAB, periplasmic chaperones grey, proteins at acidic transcription Dashed as well as at acidic pH, white, proteins expressed only pH by RpoS-dependent mechanism arrows represent activities with unclear

acid The effectiveness of proton extrusion in the periplasm for strong or long-term acidification stress has often been matter of debate. In fact, it seems that, although periplasmic

cannot be accumulated a certain limit in the is as deleterious as for cytoplasm, protons beyond

if not the for the cell to extrude protons to periplasm. Porins appear to be the major only way

limitation in the So, low the extracellular space, thereby being a probable process.

which are true intracellular pH also induces a series of acid resistance systems (ARS)

acidification or eliminate defensive systems that reduce the negative effects of protons

are all induced and are therefore without need for a transporter. These ARS by RpoS present

36 which in the stationary phase. A first ARS is the glucose-repressed and oxidative system, may

and the role of those are not clear at the time, involve many proteins. The identity proteins

effect of on DNA and (98). Recently, a although it seems that they reduce the protons proteins

HdeAB. These are part of this system was identified as the periplasmic chaperones proteins

active as synthesized in the cytoplasm and exported to the periplasm, where they are The heterodimers, binding unfolded proteins and preventing their aggregation (60). analysis two-dimensional of protein expression under weak and strong acid stress by gel involved in electrophoresis showed the specific overexpression of many proteins sugar the exact metabolism. It is however not clear whether these proteins are induced by RpoS, or

Another the cytoplasmic role they may have in stress resistance (11). system, involving but in regulatory protein Fur, has been found to play a role in stress resistance, principally however still to be sensing pH changes. The exact role of this resistance system have in LB medium explained (54,67). The glucose-repressed ARS is present during growth only,

have not been identified, but and the complex components responsible for the induction yet

tells this is not in minimal glutamate could be one of those (5). As the name system present

medium complemented with glucose (7).

and the acid resistance. The second ARS is a glutamate-dependent one provides highest effective if the It allows cells to survive several hours at an external pH of 2, and it is most pH

in the medium is lower than 3 (5). The system relies on the availability of glutamate

and converted to (endogenous glutamate seems not to suffice) which is imported y-

of a aminobutyric acid (GABA), by the glutamate decarboxylase GadAB upon consumption

to the in exchange of proton (29). The generated GABA molecule is then exported periplasm

In the more acidic another glutamate molecule by the glutamate/GABA antiporter GadC. alkalinization of periplasm, the GABA molecule is protonated, contributing to the periplasm.

of extreme The contribution to the reduction of periplasm acidification may be importance

of these relies since porins limit the extrusion of periplasmic protons. So, the efficacy systems and in the reduction of acidity by elimination of one proton in both cytoplasm periplasm (15).

in LB acid also induces the regardless RpoS induces gadBC at every pH, but medium, gadA

from growth phase (5,29). Another important function exerted by the glutamate-dependent under anaerobic conditions, by ARS, seems to be the possibility to enhance energy production

coupling electrogenic antiport and amino acid decarboxylation. In fact, by consuming

membrane can be created cytoplasmatic protons, a proton gradient across the cytoplasmatic with this and exploited to generate ATP. The uniqueness of the glutamate decarboxylase lower regard and in comparison to other decarboxylases is its ability to work at cytoplasmatic

37 limited anaerobiosis, pH (130,141). Since in E. coli the ATP production is seriously during

the this system could be an efficient way to overcome problem.

is an one. A third ARS, offering a more modest level of protection arginine-dependent and converted to Analogously to the glutamate-dependent ARS, here arginine is imported The thus agmatine, by the arginine decarboxylase AdiA upon consumption of a proton. of another generated agmatine molecule is then exported to the periplasm in exchange The glutamate molecule by an unknown glutamate/agmatine antiporter (29). system protects under the cells for short-term stresses at pH 2.5, appearing to be active at its maximum pH 3, The and is not only induced by RpoS, but also activated by the CysB regulatory protein (5). arginine-dependent ARS requires anaerobiosis to be active at full strength, and this may speak

lack of as for the for a possible role in ATP generation under oxygen, glutamate-dependent

ARS (7). in Strengthening the ability of E. coli to fight acid stress, this organism and general AH to show acid habituation (AH). The many enteric bacteria have the ability (or adaptation) then of the cell at a mild acidic These cells show response involves a pre-incubation pH.

in this case seems higher resistance against stress at lower pH. The mechanism of resistance

caused low The AH to involve principally an improved repair of DNA damage by pH. system

of the anion released some seems also to allow the cell to reduce the deleterious effect by weak organic acids, such as benzoic acid (7).

and to overcome acid stress The presence of all these overlapping complex system

to E. which varies only confers the ability to exert a strong control over cytoplasmic pH coli,

and is at a constant value of 0.1 pH unit per every unit of extracellular pH change (22), kept

of 7.6 for external pH values between 4.5 and 9.0 (11). However, in order to fight strong

minimal such as the stresses, the environmental conditions have to match some requirements, weak acids availability of amino acids or other complex components (see above). Moreover,

this to be due more to the strongly affect growth, even at low concentrations, and appears it is released anion in the cytoplasm than because of acidification. In a recent study, reported

maximum rate if 8 mM acetate, 2 mM benzoate, or 5 E. coli grow with at least halved growth

mM propionate are present in growth medium (126).

the investment of a It is generally recognized, that recovery from acid stress requires

for metabolic these, large amount of resources which would else be utilized purposes. Among

the dissipation of proton motive-force as well as the increased ATP-demanding proton

from stress extrusion are considered to play a relevant role during recover (6,14).

38 and a food S. cerevisiae is both a biotechnologically useful organism frequent spoilage

include not bakery and agent. Industrial applications involving S. cerevisiae only brewing, material for the wine production, but also the production of the yeast itself, as starting numbers successive applications (4,123). During these processes, cells are subjected to a large of stress factors, including heat, change in pH, toxic chemicals, ions, oxidizing compounds,

S. cerevisiae shows little and momentary lack of nutrients. If grown under optimal conditions,

stress resistance mechanisms if sign of stress response, but immediately switches on

stress conditions change (62). S. cerevisiae has the amazing capacity to easily overcome

This to survive is conditions which results deleterious for many other organisms. ability

In the onset of stress conditions that conferred by a multifaceted stress response (62). general,

the cell to and enter a stationary are strong or long enough to be sensed, causes stop growth

to and a is started where the cell try phase where emergency defences are activated, process

last from minutes to several hours in yeast adapt to the new conditions. This last step may

cells are now cultures, after which the normal exponential growth is recovered. These adapted

is able to activate a stress response as to this new stress factor (118). Although yeast general As it is well, the key role is played by specialized mechanism in response to specific stresses.

these stress affect the formation or the case for E. coli, also in S. cerevisiae some of responses

acid stress and the utilization of energy cofactors, thereby impairing growth. Among those,

more carbon starvation are for sure among the stresses encountered frequently during industrial processes (4). is The distinction between the stress caused by strong acids and weak acids necessary

of (Figure 10). The low pK of the former causes intracellular acidification as a consequence

the lowered extracellular pH, while the latter have a higher pK, and only acidify cytoplasm by

entering in the undissociated form, liberating the proton only in the intracellular space (120).

would have a on In both cases, however, the lowered cytoplasmic pH strong negative impact

the cellular mechanisms if the cell would not fight it by alkalizing the intracellular space (24).

combined action of S. cerevisiae alkalize the cytoplasm by proton transport, both through the

Na+/H+- and C17HC03"-antiporter, and by the work of the plasma membrane ATPase Pmal,

the total membrane which is the most abundant membrane protein, constituting over 20% of

in form of protein (76,137). Besides exploiting the proton gradient to generate energy ATP, membrane. The the ATPase is also responsible maintain the correct ion balance across the exterior fulfilment of this activity is achieved by extruding protons to the cell upon

the ATPase was consumption of ATP (137). In case of extensive intracellular acidification,

the intracellular this shown to be activated (76,119,161,162). The maintenance of pH by

39 to 60% of the total mechanism can be energetically expensive (48,137) (138), consuming up restrict cellular ATP (139). The resulting significant ATP depletion may growth

and this occurs (120,162,168). Therefore, the activity of Pmal has to be tightly regulated, post-transcriptionally by the action of the stress-induced protein Hsp30. This regulator protein inhibits the of Pmal, thereby is expressed during late heat or acid stress response and activity

role. The knockout of a quicker having a probable energy conservation Hsp30 produces the alkalization of cytoplasm, but also a reduced growth during recovery phase (119).

.Cmklp ' ATP MECHANICAL Pdrl2 xcoo-«- Turgor pressure— - STRESS

OXIDATIVE xcoo weak acid STRESS dit») on S—*-XCOOH + Ff Protein denaturation

pH STRESS

Enzyme inhibition (e.g. Pfk, Hexokinases)

HY I strong acid GROWTH ^ ATPl REDUCTION

Figure 10. Acid stress response of S. cerevisiae.

acidification If the stressful condition is caused by weak organic acids, intracellular may

the anionic acid form is liberated in the be a less serious matter (16,120). In this case,

the acid nature. Benzoate is cytoplasm, inhibiting growth in different ways, depending on

a reduced ATP (76,91). known to inhibit glycolytic enzymes generating thereby production and formation of Sorbate in presence of oxygen cause mitochondrial disorganization resulting

such as acetate or simply high levels of superoxide free radicals (117). Other acids, propionate Other than in other inhibit metabolic reactions or cause an increased turgor pressure (120).

metabolize weak acids, S. cerevisiae responds to stress yeasts, many of which can efficiently extracellular During caused by weak acids, by transporting the anionic form to the space (3).

and a weak organic acids, the ABC transporter Pdrl2 is activated both on a transcriptional membrane. The post-transcriptional level, approaching the abundance of Pmal in the plasma extrusion of weak acid role of this transporter is to catalyze the ATP-dependent organic

40 causes severe ATP (2 anions (113,118). Hence, the response to weak acid stress depletion

a system to ATP molecules per molecule weak acid entering cytoplasm), needing regulation

and avoid an excessive energy consumption consequent growth impairment

(119,145,159,162,168). Repression of Pdrl2 is achieved through post-transcriptional The modification by the action of the Ca2+/calmodulin-dependent protein kinase Cmklp (77).

involves the inhibition of Cmkl, and therefore an adaptive response to weak organic acids increased activity of Pdrl2 (120). Pdrl2 is regulated also on a transcriptional level, although Pdrl2 becomes the exact mechanism is not known. The weak acid anion extruded by

lower extracellular so the cell has to avoid the futile protonated as a consequence of the pH, acid This is cycle of anion expulsion and successive re-enter as undissociated (118,169).

of the cell a mechanism which is probably achieved by a reduction of the porosity wall, by

currently still unknown (120). also Besides increased ATP consumption, energetic stress may be achieved by reducing

on the ATP formation. The availability of energy in form of ATP strictly depends availability

source for S. cerevisiae, and it is of the carbon source. Glucose is the preferred carbon

and are crucial therefore not surprising that pathways involved in glucose sensing signalling is available, ATP regulators of cellular physiology (64) (Figure 11). If enough glucose the Ras/cAMP and the availability is high and AMP levels are low. As a consequence, system

over the AMP-dependent protein kinase (PKA) are activated, thereby allowing to progress and AMP levels also nutrients decision point in the Gi-phase of the cell cycle. The same ATP cascade to inhibit the AMP-activated protein kinase SNFl and the signalling factor leading concentration the ATP the elicitation of the general stress response. As the glucose drops, the activation of concentration increases and the AMP concentration increases. Therefore,

ATP tasks. At the same time the signalling SNFl occurs, which inhibits all the consuming activated. cascade relative to the activation of the general stress response is Moreover, growth

the RAS/cAMP is blocked by the reduced presence of PKA and of system (49,87,154).

Inhibition of ATP General r consuming task stress response AMPK (SNFl) A© Msn2/Msn4 ÜSTREs | © ' ' Progiess over 'START' decision Glucose ©" w Glucose AMP|—PKA— t in of cell (extracellular) (intracellular)"""""*"* ATP| poin G,-phase cycle ©Y ©^ 4 ' Ras Adenylate kinase cAMP

in cerevisiae. Black arrows indicate Figure 11. Stress response to carbon starvation Saccharomyces indicate indicates activation and e indicates inhibition), grey arrows interactions or regulatory issues (here, e used: AMP-activated protein metabolic reactions, and dotted arrows indicate transport. Abbreviations AMPK, cAMP, AMP. kinase; PKA, AMP-dependent protein kinase; STREs, stress-response elements; cyclic

41 with kinase, as observed in As seen before, SNFl could also interact phosphagen

kinase if stress is too humans (87,121). That would lead to the inhibition of the phosphagen

certain level. Thus a short-term transient glucose severe, and ATP levels drop under a

of a kinase limitation would be the ideal condition to study the possible impact phosphagen under carbon starvation stress.

deletion mutants of of the In Chapter 2 and Chapter 3, we have analyzed defined enzymes

central carbon metabolisms of E. coli in combination with the transhydrogenases. Through the role for both physiological observation and metabolic flux analysis, we have investigated and the natural the energy-linked and the non-energy-linked transhydrogenase, depict in E. coli. A regulation of these relevance of the presence of two transhydrogenases possible metabolism is of a central role in central carbon two enzymes within the framework

discussed. of both the In Chapter 4 and Chapter 5, we show the effect of heterologous expression and E. coli under conditions AK/PArg and the CK/PCr ATP buffering system in S. cerevisiae and starvation stress. of reduced ATP availability, such as transient acidic

42 REFERENCES

Hahn- 1. Anderlund, M., Nissen, T.L., Nielsen, J., Villadsen, J., Rydström, J., Escherichia coli Hägerdal, B., and Kielland-Brandt, M.C. (1999). Expression of the nucleotide in pntA and pntB genes, encoding nicotinamide transhydrogenase, anaerobic Saccharomyces cerevisiae and its effect on product formation during glucose fermentation. Appl Environ Microbiol 65, 2333-2340. and 2. Arkblad, E.L., Egorov, M., Shakhparonov, M., Romanova, L., Polzikov, M., nucleotide Rydström, J. (2002). Expression of proton-pumping nicotinamide Biochem B transhydrogenase in mouse, human brain and C. elegans. Comp Physiol Biochem Mol Biol 133, 13-21. cells of 3. Arneborg, N., Jespersen, L., Jakobsen, M. (2000). Individual short-term Saccharomyces cerevisiae and Zigosaccharomyces balii exhibit different Arch Microbiol 125-128. intracellular pH responses to acetic acid. 174, of industrial baker's 4. Attfield, P.V. (1997). Stress tolerance: the key to effective strains yeast. Nat Biotechnol 15, 1351-1357. the acid 5. Audia, J.P., Webb, C.C., and Foster, J.W. (2001). Breaking through enteric bacteria. Int J Med barrier: an orchestrated response to proton stress by Microbiol 291, 97-106. the 6. Axe, D.D., and Bailey, J.E. (1995). Transport of lactate and acetate through 8-19. energized cytoplasmic membrane of Escherichia coli. Biotechnol Bioeng 47, in 7. Bearson, S., Bearson, B., and Foster, J.W. (1997). Acid stress responses enterobacteria. FEMS Microbiol Lett 147, 173-180.

muscle: the 8. Bessman, S.P., and Geiger, P.J. (1981). Transport of energy in phosphorylcreatine shuttle. Science 211, 448-452.

9. Bizouarn, T., Fjellström, O., Meuller, J., Axelsson, M., Bergkvist, A., Johansson, nicotinamide C, Goran Karlsson, B., and Rydström, J. (2000). Proton translocating deduced nucleotide transhydrogenase from Escherichia coli. Mechanism of action 211-228. from its structural and catalytic properties. Biochim Biophys Acta 1457, The 10. Bizouarn, T., Meuller, J., Axelsson, M., and Rydström, J. (2000). transmembrane domain and the proton channel in proton-pumping transhydrogenases. Biochim Biophys Acta 1459, 284-290. Acid- and base-induced 11. Blankenhorn, D., Phillips, J., and Slonczewski, J.L. (1999). proteins during aerobic and anaerobic growth of Escherichia coli revealed by two- dimensional gel electrophoresis. JBacteriol 181, 2209-2216. N.C. 12. Boonstra, B., Bjorklund, L., French, C.E., Wainwright, I., and Bruce, vinelandii and construction of (2000). Cloning of the sth gene from Azotobacter Lett chimeric soluble pyridine nucleotide transhydrogenases. FEMS Microbiol 191, 87-93.

The udhA 13. Boonstra, B., French, CE., Wainwright, I., and Bruce, N.C. (1999). J encodes a soluble nucleotide gene of Escherichia coli pyridine transhydrogenase. Bacteriol 181, 1030-1034.

in bacteria. Microbiol Rev 49, 359- 14. Booth, I.R. (1985). Regulation of cytoplasmic pH 378.

43 to acid stress. 15. Booth, LR., Cash, P., and O'Byrne, C. (2002). Sensing and adapting

Antonie van Leeuwenhoek 81, 33-42.

of the 16. Bracey, D., Holyoak, CD., and Coote, P.J. (1998). Comparison inhibitory is effect of sorbic acid and amphotericin B on Saccharomyces cerevisiae: growth Microbiol 1056-1066. inhibition dependent on reduced intracellular pH? JAppl 85, membranes. 17. Brdiczka, D. (1991). Contact sites between mitochondrial envelope Biochim Acta 1071, Structure and function in energy- and protein-transfer. Biophys 291-312.

in Escherichia 18. Brey, R.N., and Rosen, B.P. (1979). Cation/proton antiport systems coli. Properties of the calcium/proton antiporter. J Biol Chem 254, 1957-1963. 31P NMR measurements of the 19. Brindle, K., P. Braddock, and S. Fulton. (1990). creatine kinase. ADP concentration in yeast cells genetically modified to express Biochemistry 29, 3295-3302.

shock treatment on 20. Broadbent, J.R., and Lin, C (1999). Effect of heat shock or cold the resistance of Lactococcus lactis to freezing and lyophilization. Cryobiology 39, 88-102.

Acid habituation 21. Brown, J.L., Ross, T., McMeekin, T.A., and Nichols, P.D. (1997). in low of Escherichia coli and the potential role of cyclopropane fatty acids pH tolerance. Int JFood Microbiol 31, 163-173.

and Gould 22. Brown, M.H., and Booth, LR. (1991). Foodpreservatives, Russell W.A. 23. Bruinenberg, P.M., Jonker, R., van Dijken, J.P., Scheffers, (1985). chemostat cultures of Utilization of formate as an additional source by glucose-limited for Candida utlilis CBS 621 and Saccharomyces cerevisiae CBS 8066: evidence absence of transhydrogenase activity in yeast. Arch Microbiol 142, 302-306. Am J 24. Busa, W.B., Nuccitelli, R. (1984). Metabolic regulation via intracellular pH. Physiol 206, 409-438.

NADP turnover in the 25. Bykova, N.V., and Meiler, I.M. (2001). Involvement of matrix Plant 448- oxidation of NAD-linked substrates by pea leaf mitochondria. Physiol 111, 456. and Moller, 26. Bykova, N.V., Rasmusson, A.G., Igamberdiev, A.U., Gardeström, P., I.M. (1999). Two separate transhydrogenase activities are present in plant mitochondria. Biochem Biophys Res Commun 265, 106-111. of the 27. Çakar, Z.P., U. Sauer, and J. Bailey. (1999). Metabolic engineering yeast: perils of auxotrophic hosts. Biotech lett 21, 611-616.

28. Carling, D., Zammit, V.A., and Hardie, D.G. (1987). A common bicyclic protein cholesterol kinase cascade inactivates the regulatory enzymes of fatty acid and biosynthesis. FEBS Lett 223, 217-222. and J.W. 29. Castanie-Cornet, M.P., Penfound, T.A., Smith, D., Elliott, J.F., Foster, 3525-3535. (1999). Control of acid resistance in Escherichia coli. J Bacteriol 181, of the tobacco 30. Chamberlin, M. (1997). Mitochondrial arginine kinase in the midgut hornworm (Manduca sexto). J Exp Biol 200, 2789-2796. of 31. Chang, Y.Y., Eichel, J., and Cronan, J.E., Jr. (2000). Metabolic instability Escherichia coli cyclopropane fatty acid synthase is due to RpoH-dependent proteolysis. J Bacteriol 182, 4288-4294.

44 contraction of rabbit 32. Chase, P.B., and Kushmerick, M.J. (1988). Effects of pH on fast and slow skeletal muscle fibers. Biophys J 53, 935-946.

muscle. Mol Cell Biochem 33. Clark, J.F. (1994). The creatine kinase system in smooth 133/134,221-232. and of reconstitutively 34. Clarke, D.M., and P.D. Bragg. (1985). Purification properties Eur J Biochem active nicotinamide nucleotide transhydrogenase of Escherichia coli. 149,517-523. Nucleotide 35. Clarke, D.M., Loo, T.W., GiUam, S., and Bragg, P.D. (1986). sequence nucleotide of of the pntA and pntB genes encoding the pyridine transhydrogenase Escherichia coli. Eur JBiochem 158, 647-653.

characteristics of the 36. Cohen, P.T., and Kaplan, N.O. (1970). Kinetic pyridine Chem 4666- nucleotide transhydrogenase from Pseudomonas aeruginosa. JBiol 245, 4672.

D.G. Similar substrate 37. Dale, S., Wilson, W.A., Edelman, A.M., and Hardie, (1995). HMG- recognition motifs for mammalian AMP-activated protein kinase, higher plant CoA reductase kinase-A, yeast SNFl, and mammalian calmodulin-dependent protein kinase I. FEBS Lett 361, 191-195.

A of the function 38. Davuluri, S.P., Hird, F.J.R., McClean, R.M. (1981). re-appraisal Biochem and synthesis of phosphoarginine and phosphocreatine in muscle. Comp Physiol 69B, 329-336. cationic amino acids in animal 39. Deves, R., and Boyd, CA. (1998). Transporters for 487-545. cells: discovery, structure, and function. Physiol Rev 78,

40. Dills, S.E., and Dobrogosz, W.J. (1977). Cyclic adenosine 3',5'-monophosphate 854-865. regulation of membrane energetics in Escherichia coli. JBacteriol 131, and of 41. Eggeling, L., Sahm, H, de Graaf, AA. (1996). Quantification directing 1- metabolite flux: Application to amino acid overproduction. Adv Biochem Eng 54, 30. and functional 42. Ellington, W.R. (1989). Phosphocreatine represents a thermodynamic J Biol 177-194. improvement over other muscle phosphagens. Exp 143, of 43. Ellington, W.R. (2001). Evolution and physiological roles phosphagen systems. Annu Rev Physiol 63, 289-325. activities of 44. Ellington, W.R., and Hines, A.C. (1991). Mitochondrial phosphagen invertebrates. Biol Bull 180, 505-507. kinases are not widely ditributed in the of octameric 45. Ellington, W.R., Roux, K., and Pineda, A.O., Jr. (1998). Origin creatine kinases. FEBSLett 425, 75-78. E., 46. Enfors, S.O., Jahic, M., Rozkov, A., Xu, B., Hecker, M., Jürgen, B., Kruger, Boone, Schweder, T., Hamer, G., O'Beirne, D., Noisommit-Rizzi, N., Reuss, M., Fuchs, L., L., Hewitt, C, McFarlane, C, Nienow, A., Kovacs, T., Tragardh, C, S., Revstedt, J., Friberg, P.C., Hjertager, B., Blomsten, G., Skogman, H., Hjort, Vrabel, P., and Hoeks, F., Lin, H.Y., Neubauer, P., van der Lans, R., Luyben, K., scale bioreactors. J Manelius, A. (2001). Physiological responses to mixing in large Biotechnol 85, 175-185. loci in Escherichia coli K- 47. Epstein, W., and Kim, B.S. (1971). Potassium transport 12. J Bacteriol 108, 639-644.

45 of the amount of 48. Eraso, P., Cid, A., and Serrano, R. (1987). Tight control yeast and plasma membrane ATPase during changes in growth conditions gene dosage. FEBS Lett 224, 193-197. stress-induced and 49. Estruch, F. (2000). Stress-controlled transcription factors, genes Rev 469-486. stress tolerance in budding yeast. FEMS Microbiol 24, and Kallio, 50. Farres, J., Holmberg, N., Schlattner, U., Bailey, J.E., Wallimann, T., first towards P.T. (2002). Expressing creatine kinase in transgenic tobacco—a step in Res 11, 49-59. introducing an energy buffering system plants. Transgenic and Sauer U. 51. Fiaux, J., Çakar, P., Sonderegger, M., Wüthrich, K, Szyperski, T., cerevisiae and Pichia (2003). Metabolic-flux profiling of the yeasts Saccharomyces stipitis. Eukaryotic cell in press of Escherichia coli 52. Fischer, E., and Sauer, U. (2003). Metabolic flux profiling Eur JBiochem 21fS mutants in central carbon metabolism using GC-MS. (in press)

of acid 53. Foster, J.W. (1999). When protons attack: microbial strategies adaptation. Curr Opin Microbiol 2, 170-174. ferric 54. Foster, J.W., and Hall, H.K. (1992). Effect of Salmonella typhimurium uptake J Bacteriol 174, regulator (fur) mutations on iron- and pH-regulated protein synthesis. 4317-4323.

55. Fraenkel, D.G. (1968). Selection of Escherichia coli mutants lacking glucose-6- JBacteriol 95, phopshate dehydrogenase or gluconate-6-phosphate dehydrogenase. 1267-1271.

and Salmonella: cellular and 56. Fraenkel, D.G. (1996)Glycolysis in Escherichia coli 2 molecular biology (Neidhardt, F. C, ed) Vol. 1, 2nd Ed., pp. 189-198, vols., American Society for Microbiology, Washington, D.C. and metabolism in an 57. Fraenkel, D.G., and Levisohn, S.R. (1967). Glucose gluconate Bacteriol 1571- Escherichia coli mutant lacking phosphoglucose isomerase. J 93, 1578.

N.C 58. French, CE., Boonstra, B., Button, K.A., and Bruce, (1997). Cloning, the soluble nucleotide transhydrogenase of sequence, and properties of pyridine Pseudomonasfluorescens. JBacteriol 179, 2761-2765. V.K., 59. From, A.H., Zimmer, S.D., Michurski, S.P., Mohanakrishnan, P., Ulstad, oxidative Thoma, W.J., and Ugurbil, K. (1990). Regulation of the phosphorylation rate in the intact cell. Biochemistry 29, 3731-3743. that 60. Gajiwala, K.S., and Burley, S.K. (2000). HDEA, a periplasmic protein supports acid resistance in pathogenic enteric bacteria. J Mol Biol 295, 605-612. the mitochondrial 61. Galante, Y.M., Lee, Y., and Hatefi, Y. (1980). Effect of pH on Biol Chem energy-linked and non-energy-linked transhydrogenation reactions. J 255, 9641-9646.

of 62. Gasch, A.P., and Werner-Washburne, M. (2002). The genomics yeast responses 181-192. to environmental stress and starvation. Funct Integr Genomics 2,

of K to 63. Golding, E.M., Teague, W.E., Jr., and Dobson, G.P. (1995). Adjustment and ATP varying pH and pMg for the creatine kinase, adenylate kinase hydrolysis Biol 198 Pt equilibria permitting quantitative bioenergetic assessment. JExp ( 8), 1775-1782.

46 Ruis, H., and 64. Gorner, W., Durchschlag, E., Wolf, J., Brown, E.L., Ammerer, G., nuclear localization Schuller, C (2002). Acute glucose starvation activates the signal factor. Embo 135-144. of a stress-specific yeast transcription J21,

there's more than one 65. Greenhaff, P.L. (2001). The creatine-phosphocreatine system: J 657. song in its repertoire. Physiol 537, L.L. 66. Haest, C.W., de Gier, J., den Kamp, J.O., Bartels, P., and van Deenen, derived with (1972). Chages in permeability of Staphylococcus aureus and liposomes varying lipid composition. Biochim Biophys Acta 255, 720-733. in the acid tolerance of 67. Hall, H.K., and Foster, J.W. (1996). The role of fur response from its role in Salmonella typhimurium is physiologically and genetically separable iron acquisition. J Bacteriol 178, 5683-5691. kinase in the 68. Hardie, D.G. (1999). Roles of the AMP-activated/STvFi protein family Biochem Soc 13-27. response to cellular stress. Symp 64, fuel 69. Hardie, D.G., and Carling, D. (1997). The AMP-activated protein-kinase: gaugfe of the mammalian cell? Eur JBiochem 246, 259-273. kinase: the 70. Hardie, D.G., and Hawley, S.A. (2001). AMP-activated protein energy charge hypothesis revisited. BioEssays 23, 1112-1119. nicotinamide 71. Hatefi, Y., and M. Yamaguchi. (1992)The energy-transducing L., nucleotide transhydrogenase in Molecular mechanisms in bioenergetics (Ernster, Science ed), pp. 265-281, Elsevier nucleotide a 72. Hatefi, Y., and M. Yamaguchi. (1996). Nicotinamide transhydrogenase: FASEB J10, model for utilization of substrate binding energy for proton translocation. 444-452. and 73. Hird, F.J., and McLean, R.M. (1983). Synthesis of phosphocreatine Biochem B 76, phosphoarginine by mitochondria from various sources. Comp Physiol 41-46. mitochondrial 74. Hird, F.J., and Robin, Y. (1985). Studies on phosphagen synthesis by preparations. Comp Biochem Physiol B 80, 517-520. of nicotinamide nucleotide 75. Hoek, J.B., and Rydström, J. (1988). Physiological roles transhydrogenase. Biochem J 254, 1-10. Crimmins, K., Brown, 76. Holyoak, CD., Stratford, M., McMullin, Z., Cole, M.B., and A.J., and Coote, P.J. (1996). Activity of the plasma membrane H(+)-ATPase and of Saccharomyces optimal glycolytic flux are required for rapid adaptation growth sorbic acid. Environ cerevisiae in the presence of the weak-acid preservative Appl Microbiol 62, 3158-3164. Bauer, B., 77. Holyoak, CD., Thompson, S., Ortiz Calderon, C, Hatzixanthis, K., of Cmkl Kuchler, K., Piper, P.W., and Coote, P.J. (2000). Loss Ca(2+)-calmodulin- weak acid resistance, dependent protein kinase in yeast results in constitutive organic of the Pdrl2 ATP-binding cassette associated with a post-transcriptional activation transporter. Mol Microbiol 37', 595-605. acid of human 78. Humm, A., Huber, R., and Mann, K. (1994). The amino sequences 101-107. and pig L-arginine:glycine amidinotransferase. FEBS Lett 339, and of human 79. Isbrandt, D., and von Figura, K. (1995). Cloning sequence analysis Acta 264-267. guanidinoacetate N-methyltransferase cDNA. Biochim Biophys 1264,

47 inverse correlation 80. Ivorra, C, Perez-Ortin, J.E., and del Olmo, M. (1999). An A molecular between stress resistance and stuck fermentations in wine yeasts. study. Biotechnol Bioeng 64, 698-708. J Muscle Res Cell 81. Iyengar, M.R. (1984). Creatine kinase as an intracellular regulator. Motil 5, 527-534. M.N. L.A. 82. Jackson, J.B., N.P.J. Cotton, R. Williams, T. Bizouarn, Hutton, in Sazanoi, and CM. Thomas. (1993). Proton-translocating transhydrogenase bacteria. Biochem Soc Transact 21, 1010-1013.

and mechanism of 83. Jackson, J.B., Peake, S.J., and White, S.A. (1999). Structure proton-translocating transhydrogenase. FEBS Lett 464, 1-8. Bizouarn, T., 84. Jackson, J.B., Quirk, P.G., Cotton, N.P., Venning, J.D., Gupta, S., transfer in Peake, S.J., and Thomas, CM. (1998). Interdomain hydride proton- translocating transhydrogenase. Biochim Biophys Acta 1365, 79-86. J.D. The 85. Jackson, J.B., White, S.A., Quirk, P.G., and Venning, (2002). alternating site, binding change mechanism for proton translocation by transhydrogenase. Biochemistry 41, 4173-4185.

Basic Res Cardiol 88, 86. Kammermeier, H. (1993). Meaning of energetic parameters. 380-384.

and 87. Kemp, B.E., Mitchelhill, K.I., Stapleton, D., Michell, B.J., Chen, Z.P., AMP-activated Witters, L.A. (1999). Dealing with energy demand: the protein kinase. Trends Biochem Sei 24, 22-25. pH 88. Kobayashi, H., Suzuki, T., and Unemoto, T. (1986). Streptococcal cytoplasmic ATPase. J is regulated by changes in amount and activity of a proton-translocating Biol Chem 261, 627-630. of creatine kinase is active 89. Koretsky, A.P., and Traxler, B.A. (1989). The B isozyme coli: in vivo detection 31P NMR. FEBS Lett 243, as a fusion protein in Escherichia by 8-12.

in 90. Korzeniewski, B., and Zoladz, J.A. (2002). Influence of rapid changes cytosolic in skeletal muscle: theoretical studies. Biochem J pH on oxidative phosphorylation 365, 249-258.

Studies on the 91. Krebs, H.A., Wiggins, D., Stubbs, M., Sols, A., Bedoya, F. (1983). mechanism of the antifungal action of benzoate. Biochem J 214, 657-663. in the 92. Kroll, R.G., and Booth, LR. (1981). The role of potassium transport generation coli. Biochem 691-698. of a pH gradient in Escherichia J198, metabolism in 6- 93. Kupor, S.R., and D.G. Fraenkel. (1972). Glucose 1904-1910. phosphogluconolactonase mutants ofEscherichia coli. JBiol Chem 247,

free on the of the 94. Lawson, J.W., and Veech, R.L. (1979). Effects of pH and Mg2+ Keq transfer creatine kinase reaction and other phosphate hydrolyses and phosphate reactions. J Biol Chem 254, 6528-6537. Induction of the PhoE porin 95. Lazim, Z., Humphrey, T.J., and Rowbury, R.J. (1996). acid in Escherichia coli. Lett Appl by NaCl as the basis for salt-induced sensitivity Microbiol 23, 269-272. and Booth, LR. 96. Levina, N., Totemeyer, S., Stokes, N.R., Louis, P., Jones, M.A., activation of (1999). Protection of Escherichia coli cells against extreme turgor by

48 for MscS MscS and MscL mechanosensitive channels: identification of genes required activity. Embo J18, 1730-1737. of the membrane-bound 97. Liang, A., and R.L. Houghton. (1980). Structural aspects 185-188. Escherichia coli pyridine nucleotide transhydrogenase. FEBSLett 109,

98. Lin, J., Lee, LS., Frey, J., Slonczewski, J.L., and Foster, J.W. (1995). Comparative and analysis of extreme acid survival in Salmonella typhimurium, Shigellaflexneri, Escherichia coli. J Bacteriol 177, 4097-4104.

and J.W. 99. Lin, J., Smith, M.P., Chapin, K.C., Baik, H.S., Bennett, G.N., Foster, Escherichia coli. (1996). Mechanisms of acid resistance in enterohemorrhagic Appl Environ Microbiol 62, 3094-3100.

and S.C 100. Loike, J.D., Zalutsky, D.L., Kaback, E., Miranda, A.F., Silverstein, rat and human muscle (1988). Extracellular creatine regulates creatine transport in cells. Proc Natl Acad Sei USAK5, 807-811.

H. 101. Marx, A., de Graaf, AA, Wiechert W, Eggeling, L, Sahm, (1996). Determination of the fluxes in the central carbon metabolism of Corynebacterium with metabolite glutamicum by nuclear magnetic resonance spectroscopy combined balancing. Biotechnol Bioeng 49, 111-129.

102. Merrell, D.S., and Camilli, A. (2002). Acid tolerance of gastrointestinal pathogens. Curr Opin Microbiol 5, 51-55.

103. Miller, K., Halow, J., and Koretsky, A.P. (1993). Phosphocreatine protects kinase from and ischemia. Am J transgenic mouse liver expressing creatine hypoxia Physiol 265, C1544-1551. of 104. Miller, K., Sharer, K., Suhan, J., and Koretsky, A.P. (1997). Expression functional mitochondrial creatine kinase in liver of transgenic mice. Am J Physiol 272, C1193-1202.

and M. 105. Mühlebach, S.M., M. Gross, T. Wirz, T. Wallimann, J.-C Perriard, of the creatine kinase Wyss. (1994). Sequence homology and structure predictions isoenzymes. Mol Cell Biochem 133/134, 245-262. mitochondrial 106. Munneke, L.R., and Collier, G.E. (1988). Cytoplasmic and arginine Genet 131-141. kinases in Drosophila: evidence for a single gene. Biochem 26, and molecular 107. Neidhart, F.C (1996). Escherichia coli and Salmonella: cellular D.C. biology. 2nd Ed. 2 vols., American Society for Microbiology, Washington, M.C 108. Nissen, T.L., Anderlund, M., Nielsen, J., Villadsen, J., and Kielland-Brandt, in cerevisiae (2001). Expression of a cytoplasmic transhydrogenase Saccharomyces Yeast results in formation of 2-oxoglutarate due to depletion of the NADPH pool. 18, 19-32.

and Villadsen, J. 109. Nissen, T.L., Hamann, C.W., Kielland-Brandt, M.C., Nielsen, J., cerevisiae (2000). Anaerobic and aerobic batch cultivations of Saccharomyces mutants impaired in glycerol synthesis. Yeast 16, 463-474. in the 110. Noguchi, M., Sawada, T., and Akazawa, T. (2001). ATP-regenerating system cilia of Paramecium caudatum. J Exp Biol 204, 1063-1071. Molecular 111. Olausson, T., O. Fjellström, J. Meuller, and J. Rydström. (1995).

- Biochim biology of nicotinamide nucleotide transhydrogenase a unique proton pump. Biophys Acta 1231, 1-19.

49 Natl Acad Sei USA 112. Pace, N.R. (2001). The universal nature of biochemistry. Proc 98, 805-808.

of a 113. Panaretou, B., and Piper, P.W. (1992). The plasma membrane yeast acquires ATPase novel heat-shock protein (hsp30) and displays a decline in proton-pumping and the to Eur JBiochem levels in response to both heat shock entry stationary phase. 206, 635-640. Entner-Doudoroff 114. Peekhaus, N., and T. Conway. (1998). What's for dinner?: metabolism in Escherichia coli. J Bacteriol 180, 3495-3502. Torres, 115. Pereira, CA., Alonso, G.D., Paveto, M.C, Iribarren, A., Cabanas, M.L., kinase characterization H.N., and Flawia, M.M. (2000). Trypanosoma cruzi arginine Biol Chem and cloning. A novel energetic pathway in protozoan parasites. J 275, 1495-1501. transmission electron 116. Pineda, A.O., Jr., and Ellington, W.R:. (1998). Immunogold J Zool microscopic localization of arginine kinase in arthropod mitochondria. Exp 281, 73-39.

reveal a action of 117. Piper, P. (1999). Yeast superoxide dismutase mutants pro-oxidant weak organic acid food preservatives. Free Radical Biol Med 21, 1219-1227. R., 118. Piper, P., Mahe, Y., Thompson, S., Pandjaitan, R., Holyoak, C, Egner, ABC is Muhlbauer, M., Coote, P., and Kuchler, K. (1998). The pdrl2 transporter in Embo J17, required for the development of weak organic acid resistance yeast. 4257-4265.

and M. 119. Piper, P.W., Ortiz-Calderon, C, Holyoak, C, Coote, P., Cole, (1997). Hsp30, the integral plasma membrane heat shock protein of Saccharomyces of membrane Cell cerevisiae, is a stress-inducible regulator plasma H(+)-ATPase. Stress Chaperones 2, 12-24. M. Weak 120. Piper, P.W., Ortiz-Calderon, C, Hatzixanthis, K., Mollapour, (2001). with resistance to acid acid adaptation: the stress response that confers yeasts organic food preservatives. Microbiology 147, 2635-2642. T.A. and 121. Ponticos, M., Q. Long Lu, J.E. Morgan, D. Grahame Hardie, Partridge, kinase a D. Carling. (1998). Dual regulation of the AMP-activated protein provides EMBO J. novel mechanism for the control of creatine kinase in skeletal muscle. 17, 1688-1699.

millennium: novel 122. Pretorius, LS. (2000). Tailoring wine yeast for the new approaches to the ancient art of winemaking. Yeast 16, 675-729.

baker's room 123. Randez-Gil, F., Sanz, P., and Prieto, J.A. (1999). Engineering yeast: for improvement. Trends Biotechnol 17, 237-244. C Interaction in 124. Reddy, S.R., Houmeida, A., Benyamin, Y., and Roustan, (1992). actin. Eur JBiochem 206, vitro of scallop muscle arginine kinase with filamentous 251-257.

Sur la de creatine chez les 125. Roche, J., van Thoai, N., and Robin, Y. (1957). présence Biochim Acta 24 invertébrés et as signification biologique. Biophys Inhibition of 126. Roe, A.J., O'Byrne, C, McLaggan, D., and Booth, LR. (2002). and Escherichia coli growth by acetic acid: a problem with methionine biosynthesis homocysteine toxicity. Microbiology 148, 2215-2222.

50 127. Rojo, M., R. Hovius, R.A. Demel, K. Nicolay, and T. Wallimann. (1991). Mitochondrial creatine kinase mediates contact formation between mitochondrial membranes. J Biol Chem 266, 20290-20295.

128. Saks, V.A., Khuchua, Z.A., Vasilyeva, E.V., Yu Belikova, O., and Kuznetsov, muscle cells. A.V. (1994). Metabolic compartmentation and substrate channeling in Mol Cell Biochem 133/134, 155-192.

of food 129. Salmond, C.V., Kroll, R.G., and Booth, LR. (1984). The effect preservatives coli. J Gen Microbiol 130 Pt 2845-2850. on pH homeostasis in Escherichia ( 11), and Kok, 130. Sanders, J.W., Leenhouts, K., Burghoorn, J., Brands, J.R., Venema, G., lactis and J. (1998). A chloride-inducible acid resistance mechanism in Lactococcus its regulation. Mol Microbiol 27, 299-310. microbial 131. Sauer, U. (2001). Evolutionary engineering of industrially important phenotypes. Adv Biochem Eng Biotechnol 73, 129-169. and 132. Sauer, U., Hatzimanikatis, V., Bailey, J.E., Hochuli, M., Szyperski, T., Bacillus subtilis. Nat Wüthrich, K. (1997). Metabolic fluxes in riboflavin-producing Biotechnol 15, 448-452. and 133. Schlattner, U., Forstner, M., Eder, M., Stochowiak, O., Fritz-Wolf, K., mitochondrial Wallimann, T. (1998). Functional aspects of the X-ray structure of Mol Cell Biochem 125-140. creatine kinase: a molecular physiology approach. 184,

134. Schmidt, K., Norregaard, L.C, Pedersen, B., Meissner, A., Duus, J.O., Nielsen, metabolic fluxes from J.O., and Villadsen, J. (1999). Quantification of intracellular fractional enrichment and 13C-13C coupling constraints on the isotopomer distribution in labeled biomass components. Metab Eng 1, 166-179.

from Eschrichia coli 135. Schreyer, R., and A. Bock. (1980). Phosphoglucose isomerase conditions. K 10: purification, properties and formation under aerobic and anaerobic Arch. Microbiol. 127, 289-298. and 136. Schweder, T., Kruger, E., Xu, B., Jürgen, B., Blomsten, G., Enfors, S.O., stress in Hecker, M. (1999). Monitoring of genes that respond to process-related large-scale bioprocesses. Biotechnol Bioeng 65, 151-159.

as a novel of 137. Serrano, R. (1984). Plasma membrane ATPase of fungi and plants type 87-126. proton pump. Curr Top Cell Regul 23, ATPase in 138. Serrano, R. (1988). Structure and function of proton translocating plasma membranes of plants and fungi. Biochim Biophys Acta 947, 1-28. membrane in The 139. Serrano, R. (1991)Transport across yeast vacuolar and plasma Harbor molecular biology of the yeast Saccharomyces, pp. 523-586, Cold Spring Laboratories, Cold Spring Harbor, NY. ornithine. A 140. Sipila, I. (1980). Inhibition of arginine-glycine amidinotransferase by of possible mechanism for the muscular and chorioretinal atrophies in gyrate atrophy 79-84. the choroid and retina with hyperornithinemia. Biochim Biophys Acta 613, and 141. Small, P.L., and Waterman, S.R. (1998). Acid stress, anaerobiosis gadCB: 214-216. lessons from Lactococcus lactis and Escherichia coli. Trends Microbiol 6,

142. Snow, R.J., and Murphy, R.M. (2001). Creatine and the creatine transporter: a review. Mol Cell Biochem 224, 169-181.

571. 143. Sorrells, K.M., Enigl, D.C, and Hatfield, J.R. (1989). JFood Prot 52,

51 B.M. The Merck 144. Stecher, P.G., Finkel, M.J., Siegmund, O.H., Szafranski, (1960). Index ofchemicals and drugs. 7th Ed., Merck & Co. Inc., Rahway, NJ.

action on 145. Stratford, M., Anslow, P.A. (1996). Comparison of trhe inhibitory and medium-chain Saccharomyces cerevisiae of weak-acid preservatives, uncouplers, fatty acids. FEMS Microbiol Lett 142, 53-58. Characterisation of 5- 146. Sullivan, J.E., Carey, F., Carling, D., and Beri, R.K. (1994). and the AMP-activated protein kinase in human liver using specific peptide substrates Res Commun 200, effects of 5'-AMP analogues on enzyme activity. Biochem Biophys 1551-1556.

Escherichia coli F-18 and E. coli 147. Sweeney, N.J., D.C. Laux, and P.S. Cohen. (1996). intestine. K-12 eda mutants do not colonize the streptomycin-treated mouse large Infect Immun 64, 3504-3511. of 148. Szyperski, T. (1995). Biosynthetically directed fractional 13C-labeling proteinogenic amino acids. An efficient analytical tool to investigate intermediary metabolism. Eur J Biochem 232, 433-448.

flux in biotechnology 149. Szyperski, T. (1998). 13C-NMR, MS and metabolic balancing research. Q Rev Biophys 31, 41-106. and 150. Szyperski, T., J.E. bailey, and K. Wüthrich. (1996). Detecting dissecting two-dimensional NMR metabolic fluxes using biosynthetic fractional 13C labeling and spectroscopy. TIBTECH14, 453-459.

on the creatine 151. Teague, W.E., Jr., and Dobson, G.P. (1992). Effect of temperature kinase equilibrium. JBiol Chem 267, 14084-14093. of the kinase 152. Teague, W.E., Jr., and Dobson, G.P. (1999). Thermodynamics arginine reaction. J Biol Chem 274, 22459-22463. of K for 153. Teague, W.E., Jr., Golding, E.M., and Dobson, G.P. (1996). Adjustment the creatine kinase, adenylate kinase and ATP hydrolysis equilibria to varying Pt 509-512. temperature and ionic strength. J Exp Biol 199 ( 2), and cell control in 154. Thevelein, J.M. (1992). The RAS-adenylate cyclase pathway cycle Saccharomyces cerevisiae. Antonie Van Leeuwenhoek 62, 109-130. of van K. On the reversibility 155. van de Stadt, R.J., Niuewenhuis, F.J., Dam, (1971). 173-176. the energy-linked transyhdrogenase. Biochim Biophys Acta 234, Distribution of creatine, 156. Van Pilsum, J.F., Stephens, G.C, and Taylor, D. (1972). in the animal guanidinoacetate and the enzymes for their biosynthesis kingdom. Implications for phylogeny. Biochem J126, 325-345. D., 157. van Voorthuysen, T., Regierer, B., Springer, F., Dijkema, C, Vreugdenhil, in and Kossmann, J. (2000). Introduction of polyphosphate as a novel phosphate pool J the chloroplast of transgenic potato plants modifies carbohydrate partitioning. Biotechnol 11, 65-80.

flux balance models 158. Varma, A., and Palsson, B.O. (1994). Stoichiometric quantitatively predict growth and metabolic by-product secretion in wild-type Escherichia coli W3110. Appl Environ Microbiol 60, 3724-3731. J.P. Effect of 159. Verduyn, C, Postma, E., Scheffers, W.A., and Van Dijken, (1992). on the benzoic acid on metabolic fluxes in yeasts: a continuous-culture study regulation of respiration and alcoholic fermentation. Yeast 8, 501-517.

52 and 160. Vermeulen, A.N., Kok, J.J., van den Boogaart, P., Dijkema, R., Claessens, J.A. (1993). Eimeria refractile body proteins contain two potentially functional characteristics: transhydrogenase and carbohydrate transport. FEMS Microbiol Lett 110, 223-229. The 161. Viegas, CA., Almeida, P.F., Cavaco, M., and Sa-Correia, I. (1998). H(+)- ATPase in the plasma membrane of Saccharomyces cerevisiae is activated during in growth latency in octanoic acid-supplemented medium accompanying the decrease intracellular pH and cell viability. Appl Environ Microbiol 64, 779-783. ATPase of 162. Viegas, CA., and Sa-Correia, I. (1991). Activation of plasma membrane Saccharomyces cerevisiae by octanoic acid. J Gen Microbiol 137, 645-651. New 163. Voet, D., and Voet, J.G. (1995). Biochemistry. 2nd Ed., John Wiley and Sons, York.

two different 164. Voordouw, G., van der Vies, S.M., Themmen, A.P.N. (1983). Why are Eur J types of pyridine nucleotide transhydrogenase found in living organisms. Biochem 131, 527-533.

165. Wallimann, T. (1994). Dissecting the role of creatine kinase. Curr Biology 1, 42-46. and 166. Wallimann, T., and W. Hemmer. (1994). Creatine kinase in non-muscle tissues cells. Mol Cell Biochem 133/134, 193-220.

167. Wallimann, T., M. Wyss, D. Brdiczka, K. Nicolay, H.M. Eppenberger. (1992). Intracellular compartmentation structure and mention of creatine kinase isoenzymes in circuit' for tissues with high and fluctuating energy demands: the 'phosphocreatine 21-40. cellular energy homeostasis. Biochem J281, in their 168. Warth, A.D. (1988). Effect of benzoic acid on growth yield of yeasts differing resistance to preservatives. Appl Environ Microbiol 54, 2091-2095.

169. Warth, A.D. (1989). Transport of benzoic and propanoic acids by Zigosaccharomyces balii.JGen Microbiol 135, 1383-1390. Zool 170. Watts, D.C (1975). Evolution of phosphagen kinase in the chordate line. Symp Soc London 36, 105-127.

171. Wiechert, W. (2002). Modeling and simulation: tools for metabolic engineering. J Biotechnol 94,37-63.

172. Woods, A., Cheung, P.C., Smith, F.C, Davison, M.D., Scott, J., Beri, R.K., and Carling, D. (1996). Characterization of AMP-activated protein kinase beta and in vitro. JBiol Chem 271, gamma subunits. Assembly of the heterotrimeric complex 10282-10290.

173. Wyss, M., and T. Wallimann. (1992). Metabolite channeling in aerobic energy metabolism. J Theor Biol 158, 129-132.

174. Wyss, M., J. Smeitnik, R.A. Wevers, and T. Wallimann. (1992). Mitochondrial metabolism. Biochim Acta creatine kinase: a key enzyme of aerobic energy Biophys 1102,119-166.

175. Wyss, M., and Kaddurah-Daouk, R. (2000). Creatine and creatinine metabolism. Physiol Rev 80, 1107-1213.

176. Yi, Y., and Samuelson, J. (1994). Primary structure of the Entamoeba histolytica of a vacuolar membrane gene (Ehvmal) encoding the catalytic peptide putative proton-transporting ATPase (V-ATPase). Mol Biochem Parasitol 66, 165-169.

53 Jr. The amino 177. Young, G.B., D.L. Jack, D.W. Smith, and M.H. Saier (1999). Acta 1415, 306-322. acid/auxin:proton symport permease family. Biochim. Biophys.

metabolism in a 178. Zablotny, R., and D.G. Fraenkel. (1967). Glucose and gluconate J Bacteriol 93, mutant of Escherichia coli lacking gluconate-6-phosphate dehydrase. 1579-1581.

and characterization 179. Zahl, K.J., C Rose, and R.L. Hanson. (1978). Isolation partial nucleotide Arch of a mutant ofEscherichia coli lacking pyridine transhydrogenase. Biochem Biophys 190, 598-602. in Escherichia coli is 180. Zakharyan, E., and Trchounian, A. (2001). K+ influx by Kup Microbiol Lett 61-64. accompanied by a decrease in H+ efflux. FEMS 204,

54 CHAPTER 2

Metabolic flux response to phosphoglucose isomerase knock-out in Escherichia coli and impact of overexpression

of the soluble transhydrogenase UdhA

FABRIZIO CANONACO, TRACY A. HESS, SYLVIA HERI, TAOTAO

WANG, THOMAS SZYPERSKI, AND UWE SAUER

FEMS Microbiology Letters 204, 247-252 (2001)

55 Abstract

Blocking glycolytic breakdown of glucose by inactivation of phosphoglucose isomerase in Escherichia coli led to a greatly reduced maximum specific growth

ratios rate. Examination of the operational catabolic pathways and their flux using

[U- CJ-glucose labeling experiments and METAFoR analysis provide evidence for

catabolism in the pentose phosphate (PP) pathway as the primary route ofglucose the knock-out mutant. The resulting extensive flux through the PP pathway disturbs

the apparently the reducing power balance, since overexpression of recently identified soluble transhydrogenase UdhA improves significantly the growth rate of

the the Pgi mutant. The presented results provide first evidence that UdhA restores cellular redox balance by catalyzing electron transferfrom NADPH to NADH.

56 INTRODUCTION

central carbon metabolism By providing energy, building blocks, and co-factors, constitutes the biochemical backbone of all cells. Nevertheless, our understanding of systemic properties of the central metabolic network lacks behind the accumulated wealth of detailed is the biochemical and genetic knowledge on its individual components. One reason redundancy of this complex network, in the sense that more than one reaction or pathway

firm conclusions on metabolic catalyzes a given conversion of intermediates. Thus,

not be drawn on the basis of consequences of genetic manipulations can often physiological of metabolic characterization only (7). When based on isotopic tracer data, however, methods flux analysis can potentially distinguish between alternative pathways (15,17,19).

cells on mixtures of One approach is based on tracing intact carbon fragments in grown uniformly [U-13Cô] and unlabeled glucose (16,18). The resulting 13C labeling patterns of of metabolic intermediates are analyzed by nuclear magnetic resonance (NMR) spectroscopy

alternative that amino acids that are synthesized from these intermediates. Because pathways

intact that from lead to common intermediates or products yield different fragments originate

in the 13C fine structures that a single glucose source molecule, specific multiplet patterns

Probabilistic relate the reflect the in vivo usage of reactions are generated. equations determined intensities of the multiplet components to the relative abundance of intact carbon fragments (16). Thus, this method, also referred to as metabolic flux ratio (METAFoR) analysis, identifies the active metabolic pathways and the ratios of their fluxes during a labeling experiment ( 11,13,14,16,18).

the metabolic of Here we use METAFoR analysis by NMR to investigate consequences phosphoglucose isomerase (Pgi) inactivation in Escherichia coli. Since Pgi is located at the first juncture of different pathways for glucose catabolism (Figure 1), its inactivation is particularly useful for studying general metabolic network behavior because glucose catabolism must then proceed via the pentose phosphate (PP) and/or the Entner-Doudoroff

of the (ED) pathway. This flux rerouting has a profound influence on the balance reducing

in excess when catabolism power budget because NADPH is formed large proceeds exclusively via these pathways. In contrast to NADH, NADPH is solely used in anabolic membrane- reactions and not in respiration (9). If and to which extend the energy-coupled, bound transhydrogenase PntAB and/or the soluble transhydrogenases UdhA are involved in maintaining the balance between NADH and NADPH in E. coli is currently an exciting matter of debate.

57 S ADPH NADPH

*- GLUCOSE G6P * 6PG -—— 1 * PEP PYR PP pathway

- FRUCTOSE * F6P ** A GI,colT PYR DAP-»-#*GAP 1 Crmetabolism mJ^h,xm pathway V •+ SER -?—^ *- PGA SA DPH

hADH \ADPH

GL) •M VIP// CQ?

AIP SOC V ,V ACETATE , ACoA' Transhydrogenase(s) NADH soluble ,V/tD+ + AttD/7/ -*- NADH + NADP"

membuiue -bound TCA cycle and NAD+ + NADPH ** r*""* NADH + \ADP+ -H+ glyoxylate shunt

Figure 1. Reaction network of E. coli central carbon metabolism. The arrows indicate physiological reaction Abbreviations: directionality and key enzymes are indicated by their 3-letter code in the grey ellipses. G6P, glucose-6-phosphate; F6P, fructose-6-phosphate; 6PG, 6-phosphogluconate; KDPG, 2-keto-3-deoxy 6PG; P5P, pentose-5-phosphate; E4P, erythrose-4-phosphate; S7P, seduheptulose-7-phosphate; GAP, glyceraldehyde-3- phosphate; DAP, dihydroxy acetone-phosphate; PGA, 3-phosphoglycerate; SER, serine; GLY, glycine; PEP, phosphoenolpyruvate; PYR, pyruvate; ACoA, acetyl coenzyme A; CIT, citrate; ICT, isocitrate; OGA, oxoglutarate; SUC, succinate; FUM, fumarate; MAL, malate; OAA, oxaloacetate; GOX, glyoxylate; Eda, KDPG aldolase; Edd, 6-phosphogluconate dehydratase; Ppc, PEP carboxylase; PckA, PEP carboxykinase; Mdh, malate dehydrogenase; MaeA, NAD-lmked malic enzyme; MaeB, NADP-lmked malic enzyme; transhydrogenase, pyridine nucleotide transhydrogenase; Zwf, G6P dehydrogenase.

58 Materials and methods

MG1655 Strains and growth conditions. Here we used the two E. coli strains wild-type

of G. (X, F", rph-T) and a Pgi mutant (K, Y, IN(rrnD-rrnE)\, rph-1, pgi::TnlO) (a gift PI Sprenger, Biotechnology, Forschungszentrum Jülich, Germany) that was constructed by and M9 phage transduction of the pgi-TnlO marker (2) into W3110. Luria-Bertani (LB)

M9 medium was minimal medium were prepared as described previously (13). supplemented

Aerobic cultivation was with either glucose or fructose at a final concentration of 0.5% (w/v). performed in 1-1 baffled shake flasks with maximally 150 ml medium at 30°C on a gyratory shaker at 200 rpm. with 0.45% For 13C labeling experiments, cultures were grown in M9 medium (w/v) unlabeled glucose and 0.05% (w/v) uniformly labeled [U-13C6]-glucose (13C >98%, Isotech).

of the culture so that the In these cases, the inoculum volume was well below 1% volume,

for the of the presence of unlabeled biomass could be neglected analysis 13C-labeling patterns.

at an These cultures were always harvested in the mid-exponential growth phase optical density at 600 nm (ODôoo) of about 1.

of and Analytical procedures. Cell growth was monitored by determination ODôoo determined cellular dry weight (cdw) was calculated from previously ODôoo-to-cdw commercial correlations. Glucose, acetate, and protein concentrations were determined with

calculated as described kits (Beckman) or by HPLC. Physiological parameters were previously (13).

cultures were washed and in one To prepare crude cell extracts, LB-grown resuspended for volume 0.9% (w/v) NaCl, 10 mM MgS04, and disrupted by two sonication steps at 100 W

transferred to a 1 min each. After centrifugation at 10,000 g for 30 min, the supernatant was

concentration and activities. new tube and used directly for determination of protein enzyme described Specific activities of Pgi (6) and transhydrogenase (5,12) were determined as Edd and Eda previously. Specific ED pathway activity was determined from the combined

mM and 200 mM Tris-HCl reactions in a 750 ul mixture containing 5 mM 5PG, 10 MgS04,

and was incubated at (pH 7.2). The reaction was started by adding 50 ul crude cell extract mM 30°C for 30 min. After addition of 750 ul 0.2% (w/v) 2,4-dinitrophenylhydrazine in 500

4 M the HCl, incubation at RT for 10 min, and stopping the reaction by adding 500 ul NaOH,

at 450 nm reaction product pyruvate was detected by recording the extinction (1).

59 and NMR spectroscopy and data analysis. Preparation of protein hydrolyzates recording

1 "\ 1 of 2D C- H correlation NMR spectra for aliphatic and aromatic amino acid resonances were

The FCAL was used for performed as described previously (13,16). program (8,18) of relative integration of 13C-13C scalar coupling fine structures and the calculation molecule of abundances, f of intact carbon fragments originating from a single source from the glucose (16). Briefly, these/values were calculated with probabilistic equations fine relative intensities, I, of the superimposed multiplets in the 13C-13C scalar coupling

in amino acids structures of resonances in 48 carbon atoms of the amino acids. The/values

of their molecules carbon atoms provide then information on the metabolic origin precursors in central metabolism, i.e. P5P, E4P, PGA, PEP, PYR, OGA, and OAA (16).

UdhA Genetic manipulations. Inducible overexpression of the soluble transhydrogenase

of udhA from chromosomal DNA of E. coli MG1655 (3) was achieved by PCR amplification with the primers 5'-CGGGATCCGA TGCCATAGTA ATAGG-3' and 5'-CCCAAGCTTI The PCR TTTAAAACAG GCGGTT-3' (chromosomal sequences are underlined). resulting

and cloned under the control of the IPTG- product was digested with BamHI and Hindlll inducible trc promoter of pTrc99A (Pharmacia).

60 Results and discussion

activities of 966±5 and 5±5 in Phenotypic characterization . Pgi enzyme U/mg protein crude cell extracts from control and mutant batch cultures, respectively, proved complete absence of Pgi activity in the mutant. Under aerobic conditions, mutant cultures grew well on

rate of 0.40 h" as fructose as the sole carbon source with a maximum specific growth (umax) ,

that G6P is not an essential of E. coli was reported previously (6). This indicates component biomass because it cannot be synthesized from fructose in pgi mutants. Providing glucose as

slower and consumed less the sole carbon source, the Pgi mutant grew significantly glucose

(Table 1). The biomass yield of the Pgi mutant on glucose, however, was slightly improved when compared with the wild-type, which is consistent with previous observations (6). This

which is the indicates a kinetic limitation of glucose catabolism, supported by complete absence of the metabolic overflow product acetate (data not shown).

metabolic of Metabolic flux response to pgi knockout. To investigate consequences

to a blocking the first step of glycolysis, E. coli wild-type and the Pgi mutant were subjected [U-13Cé]-glucose labeling experiment in batch culture. Visual inspection of the resulting C-

13C scalar coupling fine structures in the amino acids immediately revealed several striking

differences between wild-type and mutant (Figure 2). Using probabilistic equations

and then (METAFoR analysis), these fine structures were then quantitatively analyzed

recruited to derive metabolic flux ratios (16,18).

Parameter Wild-type Pgi mutant

u(h"') 0.74±0.02a 0.16 ±0.01 <7gic(ggV)b 2.4 ±0.1 1.1 ±0.1 Yx/s (g g"')c 0.46 ± 0.03 0.54 ± 0.02

carbon source. Table 1. Exponential growth parameters of aerobic E. coli cultures with glucose as the sole

a Standard deviation from triplicate experiments

b Specific glucose consumption rate

c Yield of biomass on glucose during the exponential growth phase

61 MG1655 Pgi mutant

Püe-a x Phc-a * A A iC2«fPF-P> (tjtrfPEP)

f«,J-lC«t5« « .511 [llz| iii'VltiO * -SU Ittzl

1% 2^ $% m**

AI»-« jj Ala-« jj

iCjrfPYRt CjOfPYJÜ ,

..in ii .»il illxi i.MI « MI |il/|

)Kï 5% l*r M*** K< 2J<* 7<5r MW

Asp-a q 4t'2ufÜAAj K:2ofüAA)

22f* 2« H% i*k

Glu-o D Glu-a [J ic2ofo<;A]i

A*r 52*5- 25e* 2im I9*> 22^ 5»% 9%

His-p | ;i E Hh-p j | . y Ti1) fCjftf HiSPl ^ r1! iT ri fRi5P) iT ri

|>M II .Sl |EIül +50 fl -Sil IHjuI

1** 61 <* (K< 38% »2<$ 313 Üt S?«X-

His-5 l" His-5 F

tSO 0 .51» [Hil «it I» .51) [Hi]

| 7* wt 1« 99%

13C-labeled amino acids obtained from Figure 2. 13C scalar coupling fine structures observed for fractionally alanine wild-type E.coli MG1655 (left) and the Pgi mutant (right): (A) phenylalanine Ca, (B) C, (C) aspartate derived from the metabolic Ca, (D) glutamate Ca, (E) histidine Cß, and (F) histidine C5. The carbon atoms are shown intermediates given in the panels. In each panel, the relative abundances of the different carbon fragments that remained intact amino acid biosynthesis from in (A) are indicated, where carbon-carbon bonds during of doublets from l3cY-13Cp-13Ca is further split by the two- glucose are shown in bold. In (E), the doublet arising from a source bond scalar coupling VCßcs. which reveals the presence of intact C5-fragments originating single molecule of glucose in ribose, which arise from the action of the oxidative branch of the PP pathway (14,16). the of The relative abundancies were calculated as described (16,17) and served to determine biosynthetic origin metabolic intermediates (Figure 3).

62 for E. The ratios obtained for MG1655 (Figure 3) were very similar to those reported

In accordance with textbook coli JM101 when grown under the identical conditions (13). knowledge (9), the glyoxylate shunt (Figure 3H) and the gluconeogenic reactions catalyzed by PckA and MaeA/B (Figure 3 F and G) were inactive during aerobic exponential growth on glucose. The only significant difference between MG1655 and JM101 was the relative flux through the anaplerotic Ppc reaction, 72% (Figure 3E) and 45% (13), respectively. The comparably high value of MG165 5 demonstrates that it uses the TCA cycle predominantly for generating the building blocks OAA and OGA, and to a lesser extent for generating ATP via

TCA oxidative phosphorylation. E. coli JM101, on the other hand, exhibited a more balanced cycle use for building block synthesis and energy generation (13).

Fraction of total pool (%)

Pentose pho^hate pathway Glucaneageaeiia and TCA. cycle C^metaboliim

Figure 3. Origins of metabolic intermediates (A to K) during aerobic exponential growth of E. coli MG1655 the determination of (white bars) and the Pgi mutant (grey bars). In certain cases, the NMR data permit only metabolites. The error was estimated from upper bounds (ub) or lower bounds (lb) on the origin of experimental the analysis of redundant 13C scalar coupling fine structures and the signal-to-noise ratio of the [13C,'H]COSY the total for a spectra employing the Gaussian law of error propagation. The fraction of pool particular all other metabolite quantifies the ratio of this metabolite that is derived from a specified substrate to the sum of substrates that contribute to the pool of this metabolite. In cases where only two reactions contribute to one fraction of the total can be attributed to metabolite, e.g., OAA from PEP and PEP from OAA, the remaining pool the competing reaction. Abbreviations are explained in Figure 1.

The METAFoR data of the Pgi mutant differed significantly from those of the wild- type, most pronounced for the flux ratios that are related to the PP pathway (Figure 3A-D).

that were derived at Most telling is the upper bound of the fraction of PEP molecules through

for the but least one transketolase reaction (Figure 3B), which is at around 5-10% wild-type almost 100% for the mutant. This strongly suggests that glucose catabolism proceeds predominantly via the PP pathway in the mutant. Consistently, the fraction of P5P molecules

in the mutant that were not affected by fast exchange via transketolase is increased (Figure

3A). Furthermore, the fraction of OAA originating from PEP (Figure 3E) was reduced more

63 TCA than 2-fold in the mutant, when compared to the wild-type. This shows that the cycle operates predominantly for the generation of ATP in the mutant.

Consistent with previous results (6), the primary flux response to Pgi inactivation

PP METAFoR NMR, however, appears to be the flux rerouting via the pathway. analysis by

of PEP molecules that were derived via provides an upper bound for the fraction

catabolism via the ED transketolase, and we thus cannot exclude residual glucose pathway

in vitro of 6PG to based on the labeling data (16). In fact, ED pathway activity (conversion

cultures of both strains ±0.1 PYR) was identically low in glucose-grown (1.2 U/mg minor fraction of (protein)). Hence, we conclude that the ED pathway likely catalyzes a

glucose catabolism in both wild-type and mutant strain. Consequently, the physiological

knock-out is manifested in consequences of PP pathway flux rerouting due to a Pgi apparently

significantly reduced specific growth and glucose uptake rates, and thus a kinetic limitation of

glucose catabolism. limitations of UdhA overexpression increases umax of the Pgi mutant. Kinetic

catabolism might be due to (i) the difficulty of E. coli metabolism to reoxidize NADPH, (ii)

the inability of the PP pathway to support higher fluxes, or (iii) the inhibition of one or more

essential reactions by accumulation of metabolites. Since have no evidence for the latter

possibility (iii) from HPLC analysis of culture supematants, the first two possibilities appear

be achieved three reactions in E. coli, more likely. Reoxidation of NADPH can potentially by

the (i) the NADPH-dependent malic enzyme (either via backflux through NADH-dependent

via and PEP to the malic enzyme or via a futile cycle from PYR MAL, OAA, PYR), (ii)

membrane-bound transhydrogenase PntAB (4), and (iii) the soluble transhydrogenase UdhA

(3) (Figure 1). Significant involvement of the malic enzyme can be excluded because virtually

a small fraction of PEP from no PYR originates from MAL (Figure 3G), and only originates

OAA (Figure 3F). Similarly, PntAB is unlikely to catalyze sufficient reoxidation of NADPH

the additional because pgi" pntAB" double mutants grow as slowly as the pgi" mutant (10), i.e., be knock-out of PntAB does not further reduce the rate of glucose catabolism, which would

metabolic expected if reoxidation through PntAB would play an important role. The primary

because zwf mutants function of PntAB appears to be in the generation of NADPH pnf grow

NADPH in the absence slower than a zwf mutant (10), as is expected when PntAB produces

of PP pathway fluxes (in the zwf mutant).

To verify that insufficient reoxidation of NADPH is relevant for the kinetic limitation of UdhA glucose catabolism in Pgi mutants, we thus overexpressed the soluble transhydrogenase

in both wild-type and Pgi mutant. The IPTG-induced expression level was chosen such that

64 when to two- to three-fold increased transhydrogenase activity was achieved, compared

and to the controls of the same strain that were transformed with the empty pTrc99a subjected

had no effect on of the same IPTG concentration. While this moderate overexpression umax

increased about 25% 0.22 ±0.00 to 0.27 ±0.01 wild-type, umax of the Pgi mutant was by (from the soluble h"1). This observation provides the first evidence for a physiological role of in UdhA transhydrogenase UdhA in the reoxidation of NADPH. A further increase

ten-fold the control did not lead overexpression to a transhydrogenase activity about activity

of the mutant cannot be to a further increase in umax (data not shown). Thus, growth Pgi

NADPH reoxidation. restored fully to wild-type rates by further improving the capacity for

both the increased demand This suggests that reduced growth of the Pgi mutant is caused by

for NADPH reoxidation and the limited capacity of the PP pathway.

Acknowledgments

the mutant with us. T. S. is indebted to We are grateful to G. Sprenger for sharing Pgi fund. the University at Buffalo, The State University of New York, for a start-up

65 REFERENCES

in the 1. Allenza, P., and Lessie, T.G. (1982). Pseudomonas cepacia mutants blocked Entner-Doudoroff pathway. J Bacteriol 150, 1340-1347. UDP- 2. Bohringer, J., Fischer, D., Mosler, G., and Hengge-Aronis, R. (1995). molecule in the control of of glucose is a potential intracellular signal expression J Bacteriol 413-422. sigma S and sigma S-dependent genes in Escherichia coli. 177, udhA 3. Boonstra, B., French, C. E., Wainwright, I., and Bruce, N. C. (1999). The

a soluble nucleotide J gene of Escherichia coli encodes pyridine transhydrogenase. Bacteriol 181, 1030-1034.

4. Clarke, D. M., Loo, T. W., Gillam, S., and Bragg, P. D. (1986). Nucleotide the nucleotide sequence of the pntA and pntB genes encoding pyridine transhydrogenase of Escherichia coli. Eur JBiochem 158, 647-653. riboflavin- 5. Dauner, M., and Sauer, U. (2001). Stoichiometric growth model for producing Bacillus subtilis. Biotechnol Bioeng 76, 132-143.

in an 6. Fraenkel, D. G., and Levisohn, S.R. (1967). Glucose and gluconate metabolism Escherichia coli mutant lacking phosphoglucose isomerase. J Bacteriol 93, 1571- 1578.

in bacteria. 7. Fraenkel, D. G., and Vinopal, R.T. (1973). Carbohydrate metabolism Annu Rev Microbiol 27, 69-100.

8. Glaser, R. (1999). FCAL 2.3.0

9. Gottschalk, G. (1986). Bacterial metabolism, Springer-Verlag, New York.

10. Hanson, R. L., and Rose, C. (1980). Effects of an insertion mutation in a locus affecting pyridine nucleotide transhydrogenase {pntr.TnS) on the growth of Escherichia coli. J Bacteriol 141, 401-404.

T. 11. Maaheimo, H., Fiaux, J., Çakar, Z. P., Bailey, J. E., Sauer, U., and Szyperski, (2001). Central carbon metabolism of Saccharomyces cerevisiae explored by biosynthetic fractional 13C labeling of common amino acids. Eur J Biochem 268, 2464-2479.

12. Park, S. M., Sinskey, A.J., and Stephanopoulos, G. (1997). Metabolic and physiological studies of Corynebacterium glutamicum mutants. Biotechnol Bioeng 55, 864-879.

13. Sauer, U., Lasko, D.R., Fiaux, J., Hochuli, M., Glaser, R., Szyperski, T., Wüthrich, K., and Bailey, J.E. (1999). Metabolic flux ratio analysis of genetic and environmental modulations of Escherichia coli central carbon metabolism. J Bacteriol 181, 6679-6688.

14. Sauer, U., Hatzimanikatis, V., Bailey, J. E., Hochuli, M., Szyperski, T., and Wüthrich, K. (1997). Metabolic fluxes in riboflavin-producing Bacillus subtilis. Nat Biotechnol 15, 448-452.

15. Sauer, U., Szypersky, T., Bailey, J.E. (2000) Future trends in complex microbial reaction studies, in NMR in microbiology: theory and applications, pp. 483-494, Horizon scientific press, Wymondham, UK.

66 of 16. Szyperski, T. (1995). Biosynthetically directed fractional 13C-labeling proteinogenic amino acids. An efficient analytical tool to investigate intermediary metabolism. Eur JBiochem 232, 433-448.

17. Szyperski, T. (1998). 13C-NMR, MS and metabolic flux balancing in biotechnology research. Q Rev Biophys 31, 41-106. J. and 18. Szyperski, T., Glaser, R. W., Hochuli, M., Fiaux, J., Sauer, U., Bailey, E., ratio Wüthrich, K. (1999). Bioreaction network topology and metabolic flux analysis 1 "\ by biosynthetic fractional C labeling and two-dimensional NMR spectroscopy. MetabEngl, 189-197.

19. Wiechert, W. (2001). 13C metabolic flux analysis. Metab Eng 3, 195-206.

67 Seite Leer / Blank leaf CHAPTER 3

Dissecting the role of soluble and membrane-bound transhydrogenases UdhA and PntAB in Escherichia coli

FABRIZIO CANONACO, SYLVIA HERI, ELIANE FISCHER, AND

UWE SAUER

69 Abstract

membrane-bound To elucidate the metabolic role for the soluble and the

constructed mutants transhydrogenases UdhA and PntAB in Escherichia coli, we

conditions where NADPH was lacking one or both enzymes under formation

isomerase or artificially modified. Specifically, we eliminated phosphoglucose (Pgi) glucose 6-phosphate dehydrogenase (Zwf), thereby forcing carbon flux exclusively

through the NADPH-generating pentose phosphate pathway in the former mutant,

mutant. The obtained mutants were analyzed or avoiding this pathway in the latter

sole carbon source. The main of the on glucose, acetate, or glycerol as the function

NADPH to NADH, as soluble transhydrogenase appears to be the conversion of and shown by the lethality of the combined deletion o/*pgi and udhA on glucose by A role the beneficial effect of UdhA overexpression in the pgi knockout mutant. of

UdhA in NADPH formation is excluded, as shown by the unchanged membrane-bound transhydrogenase flux if UdhA is knocked out. The the increase transhydrogenase, in contrast, converts NADH to NADPH, as shown by

and the reduced Zwf is knocked use ofPP pathway in PntAB mutants by growth if

be NADPH reoxidation out additionally. The normal function of UdhA appears to

NADPH than acetate. on substrates that lead to higher formation of glucose, e.g.

evidence that Quantification of transhydrogenase transcript levels provides first reoxidation expression of UdhA is upregulated in conditions that require of

NADPH, and PntAB is downregulated. In conditions that require reoxidation of

NADH, UdhA is downregulated.

70 Introduction

the macromolecular A large number of anabolic reactions are required to produce

but the core of this intricate anabolic reaction components that make up functional cells, metabolism and the cofactors ATP, network are eleven intermediates of central carbon

be carbon NADH, and NADPH (23). These intermediates and cofactors must supplied by hence anabolism and catabolism at appropriate stoichiometries for balanced growth;

on a wide of substrates. The catabolism are delicately balanced to enable growth variety functions. While reduction aequivalents NADH and NADPH serve distinct biochemical is the usual reductant in NADH is the main electron donor for the respiratory chain, NADPH

of detailed biochemical studies, the of carbon and anabolic processes. After decades pathways

known and well understood (15, 23, 27). energy metabolism (ATP and NADH) are relatively such Despite its important role in linking catabolism and anabolism, however, quantitative understanding of NADPH metabolism is still missing for most microbes. be those of the The primary NADPH-generating reactions are typically considered to isocitrate oxidative pentose phosphate (PP) pathway and the NADPH-dependent E. coli book) dehydrogenase in the tricarboxylic acid (TCA) cycle (15) (REF. Fraenkel, be involved, but (Figure 1). Additionally, nicotinamide nucleotide transhydrogenases may

role of has been a source of ever since their discovery, the physiological transhydrogenases

In the reversible speculation and often a matter of controversy (16, 17, 19). principle, transhydrogenase reaction

[NADPH] + [NAD+] + [H+in] <- [NADP±] + [NADH] + [H+out]

or a soluble, that may be catalyzed by either a membrane-bound, proton-translocating

reoxidize a of energy-independent isoform (1, 17) could function as a valve to surplus In the NADPH from catabolism or as a NADPH-generating reaction. eukaryotes, proton-

a flexible function as a buffer against translocating transhydrogenase appears to have In microbial dissipation of either the mitochondrial redox potential or the energy supply (17).

of remains an systems, however, the physiological role of the two types transhydrogenases

exciting matter of debate.

Organisms without functional transhydrogenases, such as the yeast Saccharomyces catabolic NADPH and cerevisiae, cannot tolerate significant imbalances between production In the anabolic NADPH consumption, hence must delicately balance both processes (2, 10). that catabolic NADPH bacterium Escherichia coli, however, metabolic flux analyses revealed

conditions as was also formation exceeds the anabolic requirement under certain growth (9),

E. coli contains both shown for other bacteria (8, 22, 33). In sharp contrast to most microbes,

71 the udhA and a soluble and a membrane-bound transhydrogenase isoform, encoded by (4)

role of the soluble isoform pntAB (6) genes, respectively. In particular the physiological

remains obscure, but heterologous overexpression of UdhA or its analogs in yeast was shown

to catalyze the conversion of NADPH to NADH (10, 25), since the [NADPH]-to-[NADP+]

ratio is generally in a more reduced state than is the [NADH]-to-[NAD±] ratio (16).

Here we quantitatively elucidate the physiological role of the two transhydrogenases

UdhA and PntAB in E. coli NADPH metabolism, more generally, and the extent to which

catabolism is coupled to anabolism via NADPH. The experimental strategy is based on perturbing NADPH metabolism by defined mutations. Metabolic consequences of these perturbations are monitored by metabolic flux analysis based on 13C-labeling experiments,

which is at present the exclusive methodology for quantification of NADPH formation (8, 9,

22, 32, 33).

72 GLUCONATE

NADPH I NADPH co2

ZW, t ) GLUCOSE * G6P scL 6PG , Pentose 5-phosphate

TP& Glycolysis PP ÇCttt Pathway F6F GLYCEROL I FADH tl, KDPG '"""'7 "' \"*^ eda ^ Erythros« Glycerol-3P -L GAP ^ r 4-phosphate ) ED _^.Seduheptiilose NADH-*—4 7-phosphate 3-phosptto-^^ Pathway glycerate "NADH 1

PEP *r?^ ^\ ca FADH

NADH-*—^-^C02 NADH ACETATE-« ACoA

Pyridine nucleotide transhydrogenase(s) UdhAXsoluble NADPH TCA cycle and glyoxylate shunt NADPH + NAD+ NADH + NADP*

PntAFi/4ncm|jrane_j)0un(j

Figure 1. Reaction network of E. coli central carbon metabolism. The arrows indicate physiological reaction Abbreviations: F6P, directionality and enzymes are indicated by their 3-letter code. G6P, glucose 6-phosphate; fructose 6-phosphate; 6PG, gluconate 6-phosphate; KDPG, 2-keto-3-deoxy 6PG; GAP, glyceraldehyde 3- phosphate; PEP, phosphoenolpyruvate; ACoA, acetyl coenzyme A; OGA, oxoglutarate; OAA, oxaloacetate; Eda, KDPG aldolase; Edd, 6-phosphogluconate dehydratase; UdhA, soluble transhydrogenase, PntAB, membrane-bound transhydrogenase; Zwf, G6P dehydrogenase; Pgi, phosphoglucose isomerase; PP pathway, acid. pentose phosphate pathway; ED pathway, Entner-Doudoroff pathway; TCA, tricarboxylic

73 EXPERIMENTAL PROCEDURES

Strains, plasmids and growth conditions. All genetic manipulation were performed in E. coli strain DH5oc and physiological experiments were done with wild-type E. coli MG1655

(E. coli Genetic Stock Center, Yale University, New Haven, MT) and mutants derived thereof

(Table 1). The knockout construction plasmids pKD13, pKD46, and pCP20 were obtained from B.L. Wanner (Purdue University, Lafayette, IN, USA) (7). pKD13 contains a kanamycin cassette (kan) flanked by FLP recombinase recognition targets, as well as the ampicillin

sensitive of resistance gene (bla). pKD46 carries the bla gene, a temperature origin replication

recombinase which enhances with a non-permissive temperature of 37°C and the Red system, the recombination rate under the control of the arabinose-inducible promoter ParaB. pCP20

the bla it has a contains the gene for chloramphenicol resistance (cat) and gene. Moreover, temperature sensitive origin of replication and expresses the yeast FLP recombinase. pUdhA The contains the E. coli udhA gene under the control of the IPTG-inducible trc promoter (5).

expression was induced with 100 |iM IPTG.

Strain Known genetic markers Reference

DH5oc F' I endAl hsdRl7 (rk"mk+) glnV55 thi-1 recAl gyrA (Nalr) (34)

relAl A(lacIZYA-argF)U169 deoR (<\>80dlacA(lacZ)Ml5)

MG1655 YIXrph-1 (20)

Pgi MG165S pgi this work

Zwf-EDP MG1655 zwfedd eda this work

UdhA MG1655 udhA this work

PntAB MG1655 pntAB this work

UdhA-PntAB MG1655 udhA pntAB this work

Pgi-UdhA MG1655 pgi udhA this work

Zwf-EDP-UdhA MG1655 zwfedd eda udhA this work

Zwf-EDP-PntAB MG1655 zwfedd eda pntAB this work

1. Table 1. Used E. coli strains. For proteins encoded by the deleted genes see also the legend to Figure

74 Luria-Bertani (LB) complex medium and M9 minimal medium were prepared as The described previously (29) and all physiological experiments were done in M9 medium.

either or M9 medium was supplemented with 5.0 g/1 of glucose, acetate, glycerol, gluconate

batch cultures were in as the sole carbon source. Unless indicated otherwise, aerobic grown

shaker at 500 ml baffled shake flasks with maximally 50 ml medium at 37°C on a gyratory

50 50 or 25 mg/1 200 rpm. When necessary, mg/1 ampicillin, mg/1 kanamycin, chloramphenicol were added.

For 13C-labeling experiments, cultures were grown in 30 ml M9 medium, supplemented and 2.4 unlabeled with 3 g/1 [l-13C]-glucose or with a mixture of 0.6 g/1 [U-13C]-glucose g/1

and to an glucose. After inoculation with maximally 1% (v/v) of an M9 culture growth up

at optical density at 600 nm (ODôoo) of 1.0-1.5, culture aliquotes were harvested, centrifuged

with 10 mM and 9 and stored at -20°C. 2,500 g, and 4°C for 8 min, washed MgS04 g/1 NaCl,

and Genetic modifications / Strains construction. Plasmid preparation and purification, commercial kits Madison DNA extraction from agarose gels were performed using (Promega,

to standard WI, USA). All other recombinant DNA techniques were done according protocols

of a (28). Colony polymerase chain reaction (PCR) was performed by dissolving part colony

and incubation at 99°C for 10 min. 0.5-2.0 ul of the from a LB plate into 50 ul of ddH20

lysate were then used as the template for PCR.

from start to codon Defined knockout mutants were generated by deleting genes stop

(7). Briefly, PCR-fragments of the FRT-flanked kanamycin gene from plasmid pKD13 were of amplified with primers that contained sequence extension homologous to flanking regions

used: target genes or opérons. The following primers were

for the pgi gene 5'-aatcgtgctatgtattaattgccgaatacgatgtacatatattccggggatccgtcgacc-3'

and 5'-aacattacgctaacggcactaaaaccatcacatttttgtgtaggctggagctgcttc-3';

for the [zwfedd eda] Operon 5'-caatggaaaacgatgatttttttatcagttttgccggtgtaggctggagctgcttc-3'

and 5'-gacttttacagcttagcgccttctacagcttcacgcattccggggatccgtcgacc-3';

for the udhA gene S'-aaaggtggtgttgacgaagtactcaatagtgttccattccggggatccgtcgacc-S'

and 5'-aacggtaacgcaataaaacatatttactgcaattctgtgtaggctggagctgcttc-3';

for the [pntAB] operon 5'-ctcagcagaggccgtcagggttacagagctttcaggattccggggatccgtcgacc-3'

and S'-acacaggcaaaccatcatcaataaaaccgatggaaggtgtaggctggagctgcttc-S'

(chromosomal sequences underlined). into the strain, The Z)/7«I-digested PCR-fragments were purified and transformed target insertion of containing the temperature-sensitive Red recombinase plasmid pKD46. Correct

75 PCR fragments in kanamycin-resistant transformants was verified by colony PCR using Red- primers homologous to regions adjacent the knocked-out gene or Operon. The pKD46

inserted PCR at 37°C helper plasmid was cured by growing clones with correctly fragments

verified at 30°C. To on LB plates with kanamycin. Plasmid loss was by ampicillin sensitivity excise the chromosomally-integrated kanamycin cassette, FLP-recombinase plasmid pCP20 and was transformed to cured clones at 30°C. Correct excision in chloramphenicol-resistant

as before. kanamycin-sensitive clones was verified by colony PCR using the same primers without and Finally, pCP20 was cured by growing clones at 37°C on LB plates antibiotics, chloramphenicol sensitivity was verified at 30°C.

with 1.5 ml 6 GC-MS analysis. The biomass pellets were resuspended and hydrolyzed

M HCl at 105°C for 24 h in sealed glass tubes. Hydrolysates were dried in a vacuum

u.1 centrifuge at room temperature and derivatized at 85°C in 50 tetra-hydrofuran (Fluka,

Switzerland) and 50 ul of N-(/er?-butyldimethylsilyl)-N-methyl-trifluoroacetamide amino (MTBSTFA) (Fluka Switzerland) for 60 min. The mass isotope distribution in the

described acids was then analyzed by GC-MS of the derivatized samples as (12).

corrected for The GC-MS-derived mass distributions of proteinogenic amino acids were

METAFoR naturally occurring isotopes, and the corrected mass distributions were used for

(metabolic flux ratio) analysis (12). Calculation of the metabolic flux ratios was done with the

MATLAB-based (Mathworks, Natick, MA, USA) software Fiat Flux (N. Zamboni, unpublished) using the equations described in (12). The relative abundances of intact carbon

were calculated from the fragments originating from a single source molecule of glucose

corrected mass distributions of the amino acids.

Following the principles outlined by Sauer et al. (30), metabolic net fluxes were

fluxes and determined using Fiat Flux by considering a stoichiometric matrix containing 25

22 metabolite balances (including balances for the measured substrates and products glucose,

network acetate and C02 and the cofactors NADH and NADPH) derived from the bioreaction

considered in Figure 1. Since this system of linear equations is underdetermined, the

previously accumulated flux ratios were used as additional constraints (N. Zamboni,

the constraints from both unpublished). Briefly, the sum of the weighed square residuals of The metabolite balances and flux ratios was minimized using the MATLAB functionfmincon.

For flux ratios residuals were weighed by division with the experimental error. implemented

from or deviations from the as upper and lower bounds only error residuals positive negative

experimental value, respectively, were considered.

76 monitored Determination ofphysiological parameters and errors. Cell growth was by

and determination of optical density at 600 nm (ODöoo), glucose, acetate, protein kits concentrations were determined enzymatically using commercial (Beckman-Coulter,

described Zürich, Switzerland or Dispolab, Dielsdorf, Switzerland) as previously (9).

Physiological parameters in batch cultures were calculated as described previously (29).

Briefly, in batch cultures, the exponential growth phase was identified by log-linear

the regression of OD6oo versus time, with maximal growth rate (umax) as regression coefficient. To obtain specific production rates, ODôoo values were converted to cellular dry

0.51 cdw unit The weight (cdw) using a predetermined correlation factor of g/1 per ODôoo (5).

of substrate versus specific substrate uptake rate (qs) was calculated by log-linear regression biomass concentration, dividing the regression coefficient by the maximal growth rate. Errors all of metabolic net fluxes and metabolic flux ratios were calculated as described in (12).

of the mean. For a other errors are calculated either as standard deviations or as average error

= general functionf(x,y) the latter error is calculated as Af(x,y) J(fxAx)2 + (fyAy)2 .

from M9 medium batch In vitro enzyme activities. Crude cell extracts were prepared

washed once with 9 cultures that were harvested at around ODôoo of 1. Cell pellets were g/1

NaCl and 10 mM MgS04, and disrupted by sonication at 100 W on ice for 2 min. Enzyme

crude cell extracts or activities and protein concentrations were determined either directly in activities of in supematants after centrifiigation at 10,000 g for 30 min. Specific The phosphoglucose isomerase were determined as described previously (14).

the in absorbance at transhydrogenase protocol was slightly modified (26). Briefly, change Tris-HCl 2 mM 375 nm was monitored at 25°C in a mixture containing 50 mM (pH 7.6),

MgCl, 500 uM NADPH (Fluka, Switzerland), 1 mM 3-acetylpyridine adenine dinucleotide

this (Sigma, Switzerland), and 10-100 ul crude cell extract. Since PntAB is membrane-bound,

then obtained activity was determined in cell extracts. The specific activity was by dividing

the measured slope by the protein concentration.

culture with an Quantitative RT-PCR. Total RNA was purified from 1 ml of a batch

OD600 of about unity with the RNeasy Mini Kit (Qiagen). The obtained RNA solution was

frozen in liquid nitrogen and stored at -70°C for further use. Aliquots were analyzed on

and determination of RT- agarose gels containing formaldehyde for quantification purity (28).

RT-PCR Kit to the PCR was then performed using the OneStep (Qiagen), according suppliers

77 total RNA and 0.6 uM of instructions. Reaction mixtures were setup with 100 ng of gene-

50°C and the cDNA was specific primers. Reverse transcription was done for 30 min at

at and 1 min/kb at 72°C, using amplified by amplification cycles of 30 s at 94°C, 30 s 55°C, the following primers: for udhA 5'-AAAATGTTGGCGGCGGTTGC-3'

' and 5 ' -CATCGTCGGGTAGTTAAAGGTGGTGTT-3 ;

' for pntA 5'-CCAAGAGAACGGTTAACCAATGAAA-3 and5'-ATAAGCACCGCGATAAAACTAAGGAA-3'; for rpoD 5'-GATCAACGACATGGGCATTCAGGTG-3' and 5'-CTTCTTCCAGCGTGTAGTCGGTGTTCATA-3'.

the we tested 10, To stop amplification of the obtained cDNA during exponential phase, the best trade-off between 15, 20, 25, and 30 amplification cycles. Since 20 cycles gave all other RT-PCR amplification yield and easiness of quantification (data not shown), the Total RNA were routinely tested for experiments were done with 20 cycles. samples outside absence of chromosomal DNA contaminations by using control primers hybridizing

of the RNA was also confirmed by the open reading frame. The quality preparations

A 2 ul was analyzed on an spectrophotometric and electrophoretic assays (28). aliquot

to the visual Results were normalized according agarose gel and quantified by inspection. amplification product with rpoD-specific primers.

78 Results

Construction and physiological characterization of knockout mutants. To elucidate the of function of the two transhydrogenases in E. coli, we constructed single and double mutants both transhydrogenases and two families of knockout mutants with increased or reduced formation of NADPH during glucose catabolism. For this purpose, a marker-free deletion

Increased NADPH formation method was chosen (7), thus minimizing potential side effects.

which encodes the in the first mutant family was achieved by deleting the pgi gene, phosphoglucose isomerase, thereby forcing carbon flux from glucose to proceed exclusively

with a reduced through the PP or ED pathways (Figure 1). The Pgi mutant grew significantly

As a maximal specific growth rate of 0.22 h"1 on glucose, as was described previously (14).

of isomerase in the consequence of the pgi deletion, the in vitro activity phosphoglucose Pgi

while the strain was reduced to 10.6 ±0.1 IU/mg during growth on glucose, wild-type exhibited an activity of 966.3 ± 0.5 IU/mg.

In the wild-type background, the deletion of transhydrogenases affects the maximal specific growth rate only in the case of the membrane-bound transhydrogenase. Surprisingly,

In the mutant double mutant grew as fast as the wild-type strain (Table 2). Pgi background, however, knockout of UdhA precludes growth on glucose.

that The second mutant family was constructed in the background of Zwf-EDP mutant lacks the first reaction of the PP pathway, glucose 6-phosphate dehydrogenase, and the entire

is reduced when to ED pathway (Figure 1). As a consequence, NADPH formation compared

than the wild-type situation. The Zwf-EDP strain grew slightly slower on glucose wild-type

knockout of (Table 2), as was described before (13). Differently from the Pgi mutant family, the soluble transhydrogenase in strain Zwf-EDP-UdhA had no negative effects. In sharp contrast, deletion of the membrane-bound transhydrogenase in strain Zwf-EDP-PntAB, was detrimental to growth, resulting in very poor growth rates (Table 2).

It should be noted that all transhydrogenase mutants showed significantly lower

strain maximum biomass concentration and yield of biomass on glucose than their control

(Table 2).

79 Strain Maximum Maximum Yield of biomass on Specific glucose

specific biomass substrate during uptake rate

growth rate concentration exponential phase (mmol-g^-h"1) (IT1) (g-r1) (g/g)

±0.5 MG1655 0.67 ±0.01 2.24 ± 0.02 0.47 ± 0.03 7.9

±1.5 UdhA 0.67 ±0.01 1.62 ±0.03 0.78 ±0.24 4.8

±1.7 PntAB 0.61 ±0.03 1.89 ±0.02 0.71 ±0.25 4.8

±1.5 UdhA-PntAB 0.69 ±0.02 1.77 ±0.03 0.40 ±0.06 9.7

determined Pgi 0.22 ± 0.02 not determined not applicable not

not Pgi-UdhA no growth not applicable not applicable applicable

±1.3 Zwf-EDP 0.41 ±0.01 2.12 0.24 ±0.03 9.6

Zwf-EDP-UdhA 0.48 ±0.01 1.71 0.74 ±0.27 3.6 ±1.3

Zwf-EDP-PntAB 0.05 ± 0.03 not determined not applicable not determined

batch on Table 2. Physiological parameters of knockout mutants during exponential growth glucose.

measured it in To get insights on the relevance of transhydrogenase activity, we glucose Zwf-EDP strains cultures (Figure 2). Transhydrogenase activity does not vary in the Pgi and

the is reduced by during growth on glucose. In the UdhA strain transhydrogenase activity

43%and in the PntAB strain by 71%. In the UdhA-PntAB strain, transhydrogenase activity was zero.

is deleted in the Surprisingly, transhydrogenase remains at a high level when UdhA that the loss Zwf-EDP strain, indicating that either UdhA is not active in this background, or in in activity is compensated by upregulation of PntAB. In the Zwf-EDP-PntAB strain,

is that of contrast, an increase in transhydrogenase activity observed, indicating upregulation

UdhA is required to partially compensate for the missing membrane-bound transhydrogenase.

80 two S 0.042 u

0.04 -

è u 0.030 0028 °028 g 0.03 1 0.028

a

TJ 0.02 >-> 0.016 J3

a M !-

0.01 - 0.008

4) a, 0.000

n nn

wild UdhA PntAB UdhA- Pgi Zwf- Zwf- Zwf- EDP- type PntAB EDP EDP- UdhA PntAB

was determined in crude cell Figure 2. Specific transhydrogenase activity in knockout mutants. The activity with 5 Values are the extract, harvested from exponentially growing batch cultures in M9 medium g/1 glucose. deviations were smaller than 3% means of independent duplicate experiments. Maximal always

Physiological behaviour of strains overexpressing the soluble transhydrogenase. Since of the Pgi-UdhA mutant is not viable on glucose, we tested here whether overexpression

UdhA, improves growth of the Pgi mutant. Plasmid-based overexpression of the soluble

not transhydrogenase UdhA led to a 2-fold increased transhydrogenase activity (data shown). the soluble In the absence of phosphoglucose isomerase on glucose, overexpression of transhydrogenase increased the growth rate by 13% and the substrate uptake rate by 40% as

of the Zwf-EDP was shown before for a similar strain (5). The growth rate mutant, however,

Growth on causes a was not affected by overexpression of UdhA (Table 3). gluconate reduction of the net formation rate of both NADH and NADPH, leaving the NADPH:NADH ratio unchanged (see Appendix A). Under these conditions, the overexpression of the soluble

reduction of rate and a 5% transhydrogenase on wild type background causes a 10% growth reduction of the maximal biomass concentration (Table 3).

81 Maximum specific Maximum biomass Specific substrate

growth rate concentration uptake rate

, , Strain , (h1) (g/1) (mmol-g^h1)

Glucose Gluconate Glucose Gluconate Glucose

11.4 ± 1.9 MG1655 [p7rc99A] 0.70±0.02 0.64±0.00 2.13±0.02 1.63±0.02 ±1.1 Pgi[p7Vc99A] 0.23 ±0.01 n.d. 1.70 ±0.02 n.d. 2.8 ±0.6 Zwf-EDP [p7>c99A] 0.44 ±0.02 n.d. 1.75 ±0.03 n.d. 6.6

± 0.7 MG1655 [pUdhA] 0.65 ± 0.02 0.58 ± 0.00 2.11 ±0.01 1.55 ±0.04 9.5 3.9 ±1.3 Pgi[pUdhA] 0.26 ±0.01 n.d. 1.65 ±0.02 n.d.

± 2.2 Zwf-EDP [pUdhA] 0.44 ±0.02 n.d. 1.80 ±0.02 n.d. 7.4

UdhA. Cultures Table 3. Physiological parameters of strains overexpressing the soluble transhydrogenase IPTG and 5 or 5 g/1 gluconate (n.d.: were grown in M9 medium with 100 uM (final concentration) g/1 glucose non determined).

into the Determination of metabolic net fluxes and flux ratios. To obtain insights

of we determined relative usage of pathways and the formation rates reducing equivalents,

metabolic flux ratios (Table 4) and metabolic net fluxes (Table 5) in the constructed mutants

the UdhA the of the ED (for complete flux data see Appendix B and C). In strain, usage

the flux to PP pathway is doubled and the PP pathway is reduced, while in the PntAB strain

pathway is increased by 22%. Similarly to the PntAB strain, the double transhydrogenase decrease in mutant UdhA-PntAB showed a 34% increase in PP pathway and a 17% glycolysis

of the causes in the net usage. This modification in the usage fueling pathways changes

formation rates for NADPH and NADH, as estimated by net flux analysis (Table 5).

formation In the wild-type, transhydrogenases contributes significantly to NADPH

in but in PntAB and (Table 5). In the UdhA mutant we observe the same flux as wild-type,

is a UdhA-PntAB strains the transhydrogenase flux is zero. This indicates that PntAB

significant NADPH producing pathway in wild-type E. coli during batch growth on glucose. is Metabolic fluxes in the Zwf-EDP mutant reveal that reducing equivalent metabolism

NADH as a significantly perturbed, with a very high formation rate of and, consequence,

reduced flux through the transhydrogenase to NADH. Surprisingly, however, there is still a

small but significant flux through the PP pathway. This may be catalyzed by the periplasmic

82 radical glucose oxidase. Deletion of phosphoglucose isomerase leads to an even more

flux is now metabolic perturbation it is the only case where the transhydrogenase reversed, also the actually requiring NADPH to NADH flux. Moreover, these flux results demonstrate

or the PP biochemically expected flux responses of blocking flux into glycolysis pathway.

E. coli strains

Zwf-EDP Metabolic flux ratios MG1655 UdhA PntAB UdhA MG1655 Pgi

PntAB pUdhA

0 75 ±0 01 0 00 ± 0 00 0 98 ±0 01 Serine through glycolysis J 0 76 ±0 01 0 80 ±0 01 0 67 ±0 01 0 63 ±0 01

±0 0 08 ±0 05 0 12 ±0 06 0 28 ± 0 05 0 00 ±0 01 Pyruvate through ED Pathway % 0 00 ± 0 07 0 18±007 0 12 05

± 0 ±0 07 0 69 ± 0 07 0 35 ±0 06 0 37 ±0 06 OAA from PEP 0 76 ± 0 05 0 67 ± 0 04 0 54 0 04 91

0 02 ± 0 02 0 01 ±001 001 ±001 PEP from OAA -0 04 ±0 01 0 00 ± 0 05 0 00 ±0 01 0 00 ±0 18

0 06 ±0 19 0 01 ±0 02 0 02 ± 0 02 0 02 ± 0 02 Pyruvate from malate (lb) 0 02 ± 0 02 0 04 ±0 01 0 08 ±0 01

±0 0 66 ±0 61 0 02 ± 0 07 0 07 ± 0 05 0 06 ± 0 05 Pyruvate from malate (ub) 0 07 ± 0 06 0 13 ±0 05 0 16 03 ± ± 0 ± 0 08 0 10±0 17 0 85 ± 0 07 0 00 0 07 PEP through PP pathway (ub) 010±017 0 09 ± 0 07 0 25 0 05 08

Table 4. Metabolic flux ratios (METAFoR) obtained by GC-MS analysis from [U-13C]-and [l-,3C]-glucose labeled ($) cultures. For abbreviations refer to Figure 1

E. coli strains

Zwf-EDP Net flux MG1655 UdhA PntAB UdhA MG1655 Pgi

(mmol-g (cdw)"1^"1) PntAB pUdhA

= = = = = 0 41 = = h1) (u 067h') (u. 067h') (n 062h') ((x 0 69 h ') ((i 065h') (u 022h') (u.

7 ±0 11 7 75 ±0 94 2 50 ±0 01 8 80±0 05 Glucose uptake rate 7 50±0 01 6 90 ±0 61 8 11 ±0 10 38

5 03 ± 0 66 0 01 ±0 00 7 82 ±0 05 G6P -» F6P 4 95 ±0 03 4 60 ±0 43 4 76 ±0 06 4 04 ±0 06

06 2 11 ±0 19 2 06 ±0 04 0 94 ±0 01 6PG -> P5P 2 12 ±0 04 155 ±0 01 2 70 ±0 05 2 98 ±0

05 0 58 ±0 11 0 42±0 04 0 00±0 01 6PG -» KDPG 0 39±0 07 0 71 ±0 25 0 62±0 05 0 33 ±0

±0 6 77 ± 0 48 6 49 ±0 17 8 45 ±0 21 NADPH formation rate 6 95 ±0 46 6 15 ±0 65 9 89 ±0 24 10 91 09

±0 23 71 ±2 54 10 98 ±0 39 42 51 ±0 05 NADH formation rate 23 55 ±0 99 21 27 ±2 12 30 86 ±0 65 29 20 25

transhydrogenase ±0 03 -2 26 ±0 05 -3 63 ±0 47 -4 42 ±0 86 0 01 ±0 23 0 08 ±0 09 -3 52 ±0 19 3 54 NADPH+NAD* -> NADH+NADP*

2 and and the Table 5. Selected net fluxes obtained by flux balancing of physiological data (Table 3) using the sum of the net METAFoR values (Table 4). Net NADH and NADPH formation rates are calculated as calculated as error of the mean fluxes of reactions by which cofactors are formed and their errors are average For abbreviations refer to Figure 1

83 To elucidate the normal function Growth ofknockout mutants on acetate and gluconate.

mutants on acetate and of the two transhydrogenases, we grew the generated knockout

less gluconate, substrates which are expected to generate more and NADPH, respectively.

of while the NADH formation is Growth on acetate strongly increases the formation NADPH,

ratio. Gluconate only slightly increased, resulting in an increase of the NADPH:NADH ratio decreases formation of both NADPH and NADH, but leaves the NADPH:NADH unchanged (see Appendix A).

Knockout of the soluble transhydrogenase appears to be lethal if cells are grown on

While PntAB acetate (Table 6), the substrate that generates a surplus of NADPH. grows

strain is because undisturbed on acetate, knockout of pntAB in the Zwf-EDP lethal, probably

both this mutant cannot generate sufficient NADPH. On gluconate, all strains lacking one or

control strain transhydrogenases showed the same growth behaviour as the respective (Table

6).

Strain Maximum specific growth rate Maximal biomass concentration (h1) (g cdw.r1)

Acetate Gluconate Acetate Gluconate

± ±0.01 1.71 ±0.01 wild type 0.24 ±0.02 0.65 0.03 1.16

1.43 ±0.02 UdhA no growth 0.65 ± 0.04 no growth

PntAB 0.20 ±0.02 0 62 ± 0.03 1.19 ±0.03 1.48 ±0.04

UdhA-PntAB 0.03 ± 0.00 0.61 ±0.02 0.31 ±0.02 1.57 ±0.03

not determined Zwf-EDP-UdhA no growth not determined no growth

1.31 ±0.04 Zwf-EDP-PntAB no growth 0.67 ± 0.02 no growth

Table 6. Physiological parameters of knockout mutants on acetate and gluconate. Maximal specific growth Cultures were rate and maximal biomass concentration were determined in strains as described in Table 1 grown 1 in M9 supplemented with 4 9 g/1 acetate (Ac) or 5 g/1 gluconate (Glue)

how Determination of the udhA and pntA transcript levels. To address the question of

mRNA were determined for all transhydrogenases are regulated, the levels of udhA and pntAB carbon transhydrogenase mutants and for wild-type by quantitative RT-PCR on different

subunit was used as an internal sources. The rpoD gene, encoding the RNA polymerase o70,

84 and standard (24). Aliquots of the RT-PCR reactions were separated on agarose gels for and udhA quantified by visual inspection (Figure 3). At this point the results obtained pntA

for The normalized values were as fold- were divided by the intensity found rpoD. expressed changes of the transcript level in wild type grown on glucose (Figure 4).

M E F G

2 3 12 3 23 123 123123 123 M12312312312 3 2 3 1 2 3 M 1 2 3

RT-PCR was with total RNA Figure 3. Quantification of RT-PCRs on agarose gels. Quantitative performed were as the sole carbon source. Reaction aliquots isolated from mutants grown on glucose, acetate, or glycerol marker 1 the of 2 the amplification then loaded onto a 1% agarose gel. M indicates the lane, amplification rpoD, MG1655 on acetate; of udhA, and 3 the amplification ofpntA. Conditions analyzed: A, MG1655 on glucose; B, MG1655 UdhA on UdhA-PntAB on glucose; G, C, MG1655 on glycerol; D, UdhA on glucose; E, glycerol; F, on PntAB on PntAB on glycerol; K, Pgi glucose; L, on glucose; H, PntAB on glucose; I, acetate; J, Zwf-EDP-UdhA on O, Zwf-EDP-PntAB on MG1655:pUdhA on glucose; M, Zwf-EDP on glucose; N, glucose; glucose.

3.0 B 2.5

1.7 1.7 2.0

1.5

1 1.0 - 0.5 0.5 0.3

0.0 J

determined RT-PCR. Figure 4. Relative transcript levels of udhA (A) and pntA (B) by quantitative or 5.1 either 5.0 0.49 acetate (white bars), Cultures were grown in batch culture with g/1 glucose (grey bars), g/1 Values normalized to the transcript level on glucose in g/1 glycerol (black bars) as the sole carbon source. MG1655.

considered as Only changes exceeding 0.5- to 1.5-fold of the wild type level were with increased significant (Figure 4). Transcription ofpntA was upregulated under conditions

85 PP fluxes in the NADPH formation, i.e. growth on acetate (Table 1) or with high pathway Pgi

showed a 3.3-fold decrease in strain. The Pgi mutant on glucose and wild-type on acetate decreased pntA mRNA levels. Under conditions, where the NADPH formation was (growth

fluxes in the Zwf-EDP we detected on glycerol (Table 1) or with high glycolysis strain), of udhA either a moderate upregulation ofpntA or a significant downregulation transcription.

in mRNA level and a 3.3- The Zwf-EDP mutant showed on glucose a 1.7-fold increase pntA 5.0- fold decrease in udhA mRNA level, whereas the wild type grown on glycerol showed a

fold decrease of udhA transcription.

increased 1.7-fold on If udhA was deleted, pntA transcription was by during growth

on This would speak glycerol, whereas it showed a 2.0-fold decrease during growth glucose. of in favour of a compensation for the missing transhydrogenase by transcriptional regulation

in udhA level was not the intact one. However, an analogous compensation transcription

sole carbon source. in observed in the PntAB mutant grown on glucose and acetate as Finally, level both Zwf-EDP-UdhA and Zwf-EDP-PntAB we observed an increase in the transcription

of the intact transhydrogenase during growth on glucose.

86 Discussion

soluble and a Escherichia coli is at present the sole organism known to contain both a membrane-bound transhydrogenase (3). We show here that both isoforms have distinct is the physiological functions. The main role of the membrane-bound transhydrogenase of the formation of NADPH by NADH reoxidation. In the wild-type, the knockout

in the use of NADPH- membrane-bound transhydrogenase causes a significant increase of generating PP pathway. Moreover, the Zwf-EDP-PntAB mutant is almost incapable

that the two NADPH growth, which can be rationalized by the fact primary generating

the PP and PntAB contribute to reactions are missing in this strain. In wild-type pathway Zwf-EDP-PntAB about 75% of the NADPH production. The transhydrogenase activity in the

Zwf-EDP most due to increased mutant was 2.5 fold higher than in the mutant, likely

of PntAB be expression of UdhA (Figure 4). Thus, the activation of UdhA in the absence may

NADH reoxidation However, considering an attempt to compensate for the missing activity.

the extremely low growth rate of 0.03 h"1 of the Zwf-EDP-PntAB mutant argues against a

NADPH formation activity of UdhA. NADPH The primary role of the soluble transhydrogenase UdhA appears to be

strain with a reoxidation, because the knockout of UdhA in the Pgi mutant, which is the only

of UdhA in the mutant, in surplus of NADPH, is lethal on glucose. Overexpression Pgi The contrast, is advantageous and increases the maximal specific growth rate significantly.

NADPH reoxidation because catalytic properties of UdhA are also more appropriate for consistent with UdhA is repressed by NADP+ and activated by NADPH (3). These results are

in S. cerevisiae of E. coli the observed recovery of the lethal Pgi mutation by expression

as the rate of UdhA (11). The in vitro rate of NADPH reoxidation by PntAB is much slower in NADPH NADPH formation (21,31). We expect therefore no important role of PntAB

reoxidation.

batch on what then is Since UdhA appears to have no function during growth glucose, which its normal role? The results suggest that UdhA is important for growth on substrates

since the UdhA knockout generate increased NADPH formation, such as for example acetate,

acetate is of is lethal on this substrate. The ability to optimally behave during consumption

after of outmost importance for E. coli, since this by-product is exploited consumption

to reoxidize NADPH soluble glucose. Acetate is only an example, and the ability by formation transhydrogenase is applicable in all other situations where increased NADPH UdhA-like in occurs. This result is consistent with the overexpression of an transhydrogenase

S. cerevisiae, which likewise catalyzed the conversion ofNADPH into NADH (25).

87 of Determination of transcriptional levels of transhydrogenases grown in presence Concomitant with acetate and glycerol provide first insights into regulation of these enzymes. increased NADPH availability in the Pgi mutant on glucose or in wild-type on acetate, we

of udhA occurs if NADPH observe a downregulation ofpntAB. Analogously, downregulation formation is reduced in the Zwf-EDP mutant on glucose or in wild-type grown on glycerol.

level. This indicates the presence of a regulatory mechanism at transcriptional

and Acknowledgements. We thank Barry Wanner for the gift of pKD13, pKD46, pCP20.

88 Appendix A

as sole carbon source show Cells grown on glucose, acetate, glycerol, or gluconate

of such we different net NADPH and NADH formation rates. To estimate the extent changes, calculated the expected NADPH:NADH ratio for growth on these carbon sources (Table Al),

fluxes The net formation of both NADPH and on the basis of previously calculated (18).

extent. On the net NADH is increased on acetate, but of NADPH on a greater glycerol, formation of NADH is strongly increased, since one molecule of NADH is synthesized per

GAP metabolized molecule of glycerol (24). Moreover, since glycerol is directly converted to

(Figure 1), less NADPH is available.

Specific formation rates

(mmol • (g cdw)"1 • h"1) during growth on

Cofactor Reaction Glucose Glycerol Acetate Gluconate

0.92 NADPH G6P -> 6PG 1.86 0.72 0.44 ' * 6PG -* P5P 1.80! 0.70 0.43 0.86l

Isocitrate -* OGA 3.36 4.68 9.84 4.12

Sum 7.02 6.10 10.71 5.90

8.65 NADH GAP -^ PGA 17.29 11.41 -1.58

OGA -> Succinate 2.37 3.94 9.39 3.17 2 Succinate -> Fumarate 2.372 3.94 13.98 3.172 2 Malate -* OAA 2.372 3.94 17.57 3.172 12.98 Pyruvate -> AcCoA 10.73 6.53 0.00 PGA -> Serine 1.28 0.95 0.59 1.23 Glycerol -* Glyc-3-P 0.00 14.00 0.00 0.00 Malate -> Pyruvate 0.00 0.00 1.00 0.00

32.37 Sum 36.41 44.71 41.05

NADPH:NADH ratio (U9 (U4 (K26 0.18

abbreviations refer to 1 Table Al. Fluxes through NADPH and NADH producing reactions. For Figure Malic enzyme is considered as NADH-dependent.

1 ED as on of the flux from On glycerol and acetate, we consider the same flux through pathway glucose (3% ' ' mmol • • h of G6P to 6PG) (18) G6P to 6PG) On gluconate this flux corresponds to 10 62 (g cdw) (93%

2 if shunt is not active A flux This values are considered to be the same as for OGA-> Succinate, glyoxylate is considered on and through ED pathway corresponding to 3% of the flux from G6P to 6PG glucose, glycerol acetate, 93% on gluconate

89 Appendix B

• In vivo reaction rates (mmol • g (cdw)"1 h"1)

Zwf-EDP Reaction MG1655 UdhA PntAB UdhA MG1655 Pgi

PntAB pUdhA

Glucose rate 2 50±0 01 8 80±0 05 uptake 7 50±001 6 90±0 61 8 1U0 10 738±011 7 75 ±0 94

-> 6PG 2 48±0 01 0 94±0 01 G6P 2 51±0 03 2 26±0 21 3 32±0 04 331±0 05 2 69±0 28

6PG » P5P + 19 2 06±0 04 0 94±0 01 C02 2 12±0 04 1 55±0 10 2 70±0 05 2 98±0 06 2 11 ±0

G6P -> F6P OOliOOO 7 82±0 05 4 95±0 03 4 60±0 43 4 76±0 06 4 04±0 06 5 03 ±0 66

6PG -» T3P + 11 0 42±0 04 OOOiOOl Pyruvate 0 39±0 07 071±0 25 0 62±0 05 0 33±0 05 0 58±0

F6P + ATP -» 2T3P 78 1 28±0 03 7 88±0 05 5 80±0 05 5 07±0 45 6 05±0 08 5 44±0 09 5 90±0

-> S7P + T3P 0 68±0 01 0 18±0 00 2 P5P 057±001 0 38±0 03 0 79±0 01 0 85±0 02 0 58±0 06

P5P + E4P -> F6P + T3P 0 60±0 01 -0 07±0 00 033±001 0 14±0 03 0 56±0 01 0 60±0 02 0 34±0 06

+ E4P -> E4P + F6P 68±0 01 0 18±0 00 P5P 0 57±0 01 0 38±0O3 0 79±0 01 0 85±0 02 0 58±0 06 0 -> 57±0 03 15 60±0 11 T3P PGA 12 25±0 06 10 91±1 06 13 2Ü0 27 11 73±0 18 12 64±1 70 3

PGA -> PEP 3 29±0 04 14 38±0 11 11 04±0 07 9 71±1 07 12 10±0 26 10 48±0 16 1148±1 69

PEP -> 2 37±0 09 11 13±0 13 Pyr 7 93±0 37 6 88±0 77 8 97±0 26 4 69±0 16 8 86±1 88

-» AcCoA + 2 36±0 10 9 84±0 15 Pyruvate C02 6 95±0 12 5 93±1 28 8 50±0 25 6 23±0 12 7 67±174

OAA + AcCoA -» ICT 195±0 17 6 56±0 21 2 32±0 46 2 34±0 61 3 87±0 23 4 63±0 05 1 98±0 34

ICT -» OGA + 34 195±0 17 656±021 C02 2 32±0 46 2 34±0 61 3 87±0 23 4 63 ±0 05 198±0

-> Fumarate + 1 68±0 17 5 69±0 21 OGA C02 145±0 46 148±0 62 3 05 ±0 24 3 75 ±0 04 1 13±0 35

-> Malate 1 68±0 17 5 69±0 21 Fumarate 145±0 46 1 48±0 62 3 05 ±0 24 3 75 ±0 04 1 13±0 35

Malate -» OAA 12 150±0 12 497±0 18 0 84±0 12 1 15±0 22 2 27±0 23 051±0 08 0 96±0

-> + 0 19±0 09 0 72±0 05 Malate Pyruvate C02 061±035 0 33±0 40 0 78±0 13 3 24±0 07 0 18±0 26

OAA + ATP -> PEP + 03 005±001 0 20±0 01 C02 0 00±0 01 0 00±0 05 0 00±0 01 0 00±0 02 0 02±0

PEP + -> OAA 24 081±0 08 2 92±0 06 C02 2 59±0 36 231±0 40 2 64±0 13 5 26±0 08 2 13±0

-> Acetate 0 00±0 16 1 71±024 AcCoA 3 08±0 49 2 04±0 82 3 19±0 37 OOOiOlO 4 18±2 03

NADPH -» NADH 19 3 54±0 26 -2 26±0 25 -3 63±0 47 -4 42±0 86 001±0 23 0 08±0 09 -3 52±0

16 95 1231 7 96 22 42 Net C02 formation rate 12 23 10 69 17 58

mutants obtained GC-MS of from Table B. In vivo reaction rates in transhydrogenase by analysis hydrolysates cofactors For abbreviations refer to Figure 1 13C-glucose labeling experiments Reactions are indicated without Errors are calculated as described in (12)

90 Appendix C

Metabolic net fluxes normalized to the glucose uptake rate

Zwf-EDP Reaction MG1655 UdhA PntAB UdhA MG1655 Pgi

PntAB pUdhA

009 0 346 ±0 055 0 992 ±0 006 0 107 ±0 002 G6P -> 6PG 0 335 ±0 004 0 328 ±0 042 0 409 ±0 007 0 448 ±0

±0 007 0 403 ±0 010 0 272 ±0 041 0 825 ±0 016 0 107 ±0 002 6PG -> P5P + CO, 0 283 ±0 006 0 225 ±0 024 0 333

011 0 649 ±0 116 0 005 ±0 000 0 889 ±0 008 G6P -> F6P 0 660 ±0 004 0 667 ±0 086 0 586 ±0 010 0 547 ±0

±0 007 0 074 ±0 017 0 166 ±0 016 0 000 ±0 001 6PG -> T3P + Pyruvate 0 052 ±0 009 0 103 ±0 037 0 077 ±0 006 0 045 0 737 ±0 016 0 761 ±0 136 0513 ±0011 0 895 ±0 008 F6P + ATP -» 2T3P 0 774 ±0 007 0 736 ±0 093 0 746 ±0 013

±0 003 0 074 ±0 012 0 272 ±0 005 0 020 ±0 000 2 P5P-» S7P + T3P 0 076 ±0 002 0 055 ±0 007 0 097 ±0 002 0 115

0 082 ±0 002 0 044 ±0 009 0 241 ±0 005 -0 008 ±0 000 P5P + E4P -> F6P + T3P 0 044 ±0 002 0 020 ±0 005 0 068 ±0 002

002 0 115 ±0 003 0 074 ±0 012 0 272 ±0 005 0 020 ±0 000 P5P + E4P ^ E4P + F6P 0 076 ±0 002 0 055 ±0 007 0 097 ±0

1 630 ±0 294 1427 ±0 014 1773 ±0 016 T3P -» PGA 1 633 ±0 009 1 584 ±0 209 1 628 ±0 038 1 589 ±0 033

1480 ±0 281 1 316 ±0016 1 634 ±0016 PGA -> PEP 1 472 ±0 010 1409 ±0 200 1492 ±0 036 1 420 ±0 030

±0 023 1 143 ±0 279 0 948 ±0 035 1265 ±0016 PEP -» Pyr 1 058 ±0 049 0 999 ±0 143 1 106 ±0 034 0 635 0 843 ±0 020 0 989 ±0 254 0 946 ±0 041 1 118 ±0 019 Pyruvate -> AcCoA + C02 0 926 ±0 016 0 861 ±0 201 1 048 ±0 034 ±0 011 0 255 ±0 054 0 781 ±0 066 0 746 ±0 024 OAA + AcCoA -» ICT 0 309 ±0 061 0 339 ±0 094 0 477 ±0 029 0 627

±0 011 0 255 ±0 054 0 781 ±0 066 0 746 ±0 024 ICT -> OGA + C02 0 309 ±0 061 0 339 ±0 094 0 477 ±0 029 0 627 009 0 146 ±0 048 0 674 ±0 068 0 647 ±0 024 OGA -> Fumarate + C02 0 193 ±0 061 0 214 ±0 091 0 376 ±0 029 0 508 ±0

009 0 146 ±0 048 0 674 ±0 068 0 647 ±0 024 Fumarate -> Malate 0 193 ±0 061 0214 ±0 091 0 376 ±0 029 0 508 ±0 ±0 022 0 598 ±0 048 0 565 ±0 021 Malate -> OAA 0112 ±0015 0 166 ±0 035 0 279 ±0 028 0 069 ±0 011 0 123

0 023 ±0 033 0 075 ±0 036 0 081 ±0 006 Malate -> Pyruvate + C02 0 082 ±0 047 0 048 ±0 059 0 096 ±0 016 0 439 ±0 011 ±0 002 0 003 ±0 004 0 018 ±0 005 0 023 ±0 001 OAA + ATP -» PEP + C02 0 000 ±0 002 0 000 ±0 008 0 000 ±0 002 0 000

±0015 0 275 ±0 045 0 322 ±0 032 0 332 ±0 007 PEP + C02 -» OAA 0 345 ±0 047 0 335 ±0 065 0 325 ±0 016 0713 ±0 269 0 000 ±0 065 0 194 ±0 027 AcCoA -> Acetate 0411 ±0 065 0 297 ±0 121 0 394 ±0 045 0 000 ±0 014 0 540

454 ±0 060 1416 ±0 104 -0 257 ±0 029 NADPH -> NADH -0 484 ±0 063 -0 642 ±0 138 0 001 ±0 029 0011 ±0012 0

normalization of data from Table C. Metabolic net fluxes (%) in transhydrogenase mutants obtained by cofactors For abbreviations refer to Appendix B to the glucose uptake rate Reactions are indicated without Figure 1 Errors are calculated as average error of the mean

91 REFERENCES

1. Bizouarn, T., Althage, M., Pedersen, A., Tigerström, A., Karlsson, J., Johansson, its C, and Jackson, J. B. (2002). The organization of the membrane domain and interaction with the NADP(H)-binding site in proton-translocating transhydrogenase from Escherichia coli. Biochim Biophys Acta 1555, 122-127.

2. Boles, E., Lehnert, W., and Zimmermann, F. K. (1993). The role of the NAD- dependent glutamate dehydrogenase in restoring growth on glucose of a Saccharomyces cerevisiae phosphoglucose isomerase mutant. Eur JBiochem 217, 469-477.

3. Boonstra, B., French, C. E., Wainwright, L, and Bruce, N. C. (1999). The udhA nucleotide J gene of Escherichia coli encodes a soluble pyridine transhydrogenase. Bacteriol 181, 1030-1034.

4. Boonstra, B., French, C. E., Wainwright, I., and Bruce, N. C. (1999). The udhA soluble nucleotide J gene of Escherichia coli encodes a pyridine transhydrogenase. Bacteriol 181, 1030-1034.

5. Canonaco, F., Hess, T. A., Heri, S., Wang, T., Szyperski, T., and Sauer, U. (2001). knock-out in Escherichia coli Metabolic flux response to phosphoglucose isomerase and impact of overexpression of the soluble transhydrogenase UdhA. FEMS Microbiol Lett 204, 247-252.

6. Clarke, D. M., Loo, T. W., Gillam, S., and Bragg, P. D. (1986). Nucleotide the nucleotide sequence of the pntA and pntB genes encoding pyridine transhydrogenase of Escherichia coli. Eur J Biochem 158, 647-653.

7. Datsenko, K. A., and Wanner, B. L. (2000). One-step inactivation of chromosomal Proc Natl Acad Sei U S A 91, genes in Escherichia coli K-12 using PCR products. 6640-6645.

8. Dauner, M., Bailey, J. E., and Sauer, U. (2001). Metabolic flux analysis with a comprehensive isotopomer model in Bacillus subtilis. Biotechnol Bioeng 76, 144-156.

9. Emmerling, M., Bailey, J.E., and Sauer, U. (1999). Glucose catabolism of Escherichia coli strains with increased activity and altered regulation of key glycolytic enzymes. Metabol Eng 1, 117-127.

10. Emmerling, M., Dauner, M., Ponti, A., Fiaux, J., Hochuli, M., Szyperski, T., Wüthrich, K., Bailey, J. E., and Sauer, U. (2002). Metabolie flux responses to pyruvate kinase knockout in Escherichia coli. J Bacteriol 184, 152-164.

11. Fiaux, J., Cakar, P., Sonderegger, M., Wüthrich, K, Szyperski, T., and Sauer U. (2003). Metabolic-flux profiling of the yeasts Saccharomyces cerevisiae and Pichia stipitis. Eukaryotic cell in press

12. Fischer, E., and Sauer, U. (2003). Metabolic flux profiling of Escherichia coli mutants in central carbon metabolism using GC-MS. Eur J Biochem 270 (in press)

13. Fraenkel, D. G. (1968). The accumulation of glucose 6-phosphate from glucose and isomerase and its effect in an Escherichia coli mutant lacking phosphoglucose glucose 6-phosphate dehydrogenase. J Biol Chem 243, 6451-6457.

14. Fraenkel, D. G., and Levisohn, S.R. (1967). Glucose and gluconate metabolism in an Escherichia coli mutant lacking phosphoglucose isomerase. J Bacteriol 93, 1571- 1578.

92 New York. 15. Gottschalk, G. (1986). Bacterial metabolism. 2nd Ed., Springer-Verlag,

Freeman and 16. Harold, F. M. (1986). The vitalforce: A study ofbioenergetics, Company, New York. nicotinamide nucleotide 17. Hoek, J. B., and Rydström, J. (1988). Physiological roles of transhydrogenase. Biochem J 254, 1-10. in 18. Holms, H. (1996). Flux analysis and control of the central metabolic pathways Escherichia coli. FEMS Microbiol Rev 19, 85-116.

19. Jackson, J. B., Quirk, P. G., Cotton, N. P. J., Venning, J. D., Gupta, S., Bizouarn, transfer in T., Peake, S. J., and Thomas, C. M. (1998). Interdomain hydride proton- translocating transhydrogenase. Biochim Biophys Acta 1365, 79-86. W3110 and MG1655 20. Jensen, K. F. (1993). The Escherichia coli K-12 "wild types" that leads to starvation due to low have an rph frameshift mutation pyrimidine pyrE expression levels. J Bacteriol 175, 3401-3407. membrane-bound 21. Liang, A., and R.L. Houghton. (1980). Structural aspects of the Escherichia coli pyridine nucleotide transhydrogenase. FEBS Letters 109, 185-188. L. 22. Marx, A., Eikmanns, B. J., Sahm, H., de Graaf, A. A., and Eggeling, (1999). to the use of an Response of the central metabolism in Corynebacterium glutamicum NADH-dependent glutamate dehydrogenase. MetabolEng 1, 35-48. the 23. Neidhardt, F. C, Ingraham, J. L., and Schaechter, M. (1990). Physiology of Mass. bacterial cell: a molecular approach, Sinauer Associates, Inc., Sunderland, molecular 24. Neidhart, F. C. (1996). Escherichia coli and Salmonella: cellular and D.C. biology. 2nd Ed. 2 vols., American Society for Microbiology, Washington, M. 25. Nissen, T. L., Anderlund, M., Nielsen, J., Villadsen, J., and Kielland-Brandt, in cerevisiae C. (2001). Expression of a cytoplasmic transhydrogenase Saccharomyces results in formation of 2-oxoglutarate due to depletion of the NADPH pool. Yeast 18, 19-32.

Metabolic and 26. Park, S. M., Sinskey, A.J., and Stephanopoulos, G. (1997). physiological studies of Corynebacterium glutamicum mutants. Biotechnol Bioeng 55, 864-879.

balance of 27. Russell, J. B., and Cook, G. M. (1995). Energetics of bacterial growth: anabolic and catabolic reactions. Microbiol Rev 59, 48-62.

- manual. 28. Sambrook, J., and Russell, D. W. (2001). Molecular cloning a laboratory 3rd Ed. 3 vols., Cold Spring Harbor Laboratory Press, Cold Spring Harbor (NY). K. 29. Sauer, U., D.R. Lasko, J. Fiaux, M. Hochuli, R. Glaser, T. Szyperski, and Wüthrich, and J.E. Bailey. (1999). Metabolic flux ratio analysis of genetic environmental modulations ofEscherichia coli central carbon metabolism. JBacteriol 181, 6679-6688. and 30. Sauer, U., Hatzimanikatis, V., Bailey, J. E., Hochuli, M., Szyperski, T., subtilis. Nat Wüthrich, K. (1997). Metabolic fluxes in riboflavin-producing Bacillus Biotechnol 15, 448-452.

are two different 31. Voordouw, G., van der Vies, S.M., Themmen, A.P.N. (1983). Why found in Eur J types of pyridine nucleotide transhydrogenase living organisms. Biochem 131, 527-533.

32. Wiechert, W. (2001). 13C metabolic flux analysis. Metabol Eng 3, 195-206.

93 strain 33. Wittmann, C, and Heinzle, E. (2002). Genealogy profiling through metabolic flux of improvement by using metabolic network analysis: genealogy Environ Microbiol 68, several generations of lysine-producing Corynebacteria. Appl 5843-5859.

34. Woodcock, D. M., Crowther, P. J., Doherty, J., Jefferson, S., DeCruz, E., Noyer- W. Weidner, M., Smith, S. S., Michael, M. Z., and Graham, M. (1989). Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine 3469-3478. methylation in plasmid and phage recombinants. Nucleic Acids Res 17,

94 CHAPTER 4

Functional expression of phosphagen kinase systems

confers resistance to transient stresses in Saccharomyces

cerevisiae by buffering the ATP pool

FABRIZIO CANONACO, UWE SCHLATTNER, PAMELA S. PRUETT,

THEO WALLIMANN, AND UWE SAUER

Journal ofBiological Chemistry 277(35), 31303-31309 (2002)

95 Abstract

with Phosphagen kinase systems provide different advantages to tissues high and fluctuating energy demands, in particular an efficient energy buffering system.

two In this study we show for the first time functional expression of phosphagen contain such kinase systems in Saccharomyces cerevisiae, which does not normally

addition to systems. First, to establish the creatine kinase system, in overexpressing creatine creatine kinase isoenzymes, we had to install the biosynthesis pathway of by co-overexpression of L-arginine:glycine amidinotransferase and guanidinoacetate

creatine kinase methyltransferase. Although we could achieve considerable activity

this was not to together with more than 3 mM intracellular creatine, sufficient

conditions examined here. confer an advantage to the yeast under the specific stress intracellular Second, using arginine kinase, we successfully installed an phosphagen pool of about 5 mMphosphoarginine. Such arginine kinase expressing

that drain cellular yeast showed improved resistance under two stress challenges

and starvation. While transient energy, which were transient pH reduction

starvation led to 50% reduced intracellular ATP concentrations in wild type yeast,

arginine kinase overexpression stabilized the ATP pool at the pre-stress level. Thus,

is an intrinsic our results demonstrate that temporal energy buffering property of

distant phosphagen kinases that can be transferred to phylogenetically very

organisms.

96 Introduction

is The availability of biochemical energy, with ATP as the primary energy currency,

its are involved in fundamental to most cellular processes. Although ATP and congeners

ATP is literally hundreds of biochemical reactions, the intracellular concentration of generally

rate on and tissues, with a turnover kept very constant at about 2-5 mM, depending organisms metabolic ATP of the ATP pool that is in the range of a few seconds (21). Hence, generation Small derivations from the in a cell must be tightly balanced with ATP consuming processes. roles standard cellular concentrations of free ATP, ADP, and AMP serve important regulatory in fine tuning this delicate balance. is This balance between energy consuming and producing processes particularly

such as challenged in tissues that experience periods of high and fluctuating energy demand,

ATP these tissues brain, heart or skeletal muscle. To maintain constant levels, express inert of creatine kinase (CK; EC 2.7.3.2) that uses creatine (Cr) to create a metabolically pool

phosphocreatine (PCr). Among other functions, this PCr pool serves as a temporal energy

of to the buffer that can rapidly replenish ATP during phases high energy demand, according

following reaction (42):

MgADP" + PCr2-+ H+ <=* MgATP2" + Cr

and is the best-known The CK system is found in many vertebrate tissues probably

kinase The example of what is more generally referred to as the a phosphagen system (6).

a inert of common feature of these kinases is their capability to synthesize metabolically pool

and to phosphorylated compounds (phosphagens) during normal metabolic conditions

replenish the ATP from this pool during periods of high energetic demand. Eight phosphagen

kinase EC as a kinases are found in the animal kingdom (6), with arginine (AK; 2.7.3.3) unicellular prominent example, occurring in insects (13), crustaceans (45) and in certain

organisms (30). In analogy to CK, AK catalyzes the reaction:

MgADP" + PArg2" + H+ *=± MgATP2" + Arg.

In addition to their supposedly primary role as temporal energy buffer, phosphagen

which include of the intracellular kinase systems serve a number of other functions, buffering

that would metabolic pH, and preventing a rise in intracellular ADP levels trigger multiple the CK/PCr has the responses (6,42). In addition to the above general functions, system well This unique capability to establish a spatial energy buffering as (11). complex of the CK functionality is also reflected by oligomeric composition and compartmentalization mitochondrial isoforms, with cytosolic isoenzymes (B-CK, M-CK) forming dimers and of CK isoenzymes (sMtCK, uMtCK) forming dimers and octamers (34). In mice, expression

97 in the naturally CK-deficient liver led to multiple, beneficial effects for this organ, including tolerance to hypoxia and endotoxins (15,23). Although microorganisms are often exposed to rapidly changing environmental conditions and fluctuating availability of energy, phosphagen caudatum and kinase systems occur only in few unicellular organisms, e.g. Paramecium

such as Trypanosoma cruzi (28,30). Thus, we hypothesized that lower unicellular eukaryotes, with the yeast Saccharomyces cerevisiae, would potentially benefit if artificially equipped

is a such phosphagen kinase systems. In particular for S. cerevisiae, intracellular acidification

to weak acids well-known consequence of various environmental stresses, including exposure

of stress like desiccation and and copper (10,17,41) but also as a secondary effect challenges heat (1,29). This acidification is counteracted by a greatly increased activity of the plasma membrane H -ATPase, which expels protons from the cytoplasm at the expense of ATP

to hydrolysis (17). The energetic expense for this proton pumping activity may require up

60% of the total ATP production (35), thus constituting an ATP drain similar to muscle contraction. Likewise starvation and several other stress challenges are known to exert high

ATP demands (14).

in lower In this study, we attempted to install phosphagen kinase systems eukaryotes, thereby addressing the hypothesis that functional expression of such phosphagen kinases could potentially improve resistance to those stress challenges that constitute a significant energetic burden. This work is also a first step towards a more detailed investigation on molecular functions and mechanisms of the phosphagen kinase systems in a biological background that is free of endogenous phosphagen kinases and which can be easily genetically manipulated.

98 EXPERIMENTAL PROCEDURES

Strains, plasmids, and media. S. cerevisiae CEN.PK 113-7D (MATa) was used for Cr adaptation experiments. All other experiments were performed with strain CEN.PK 113-6B

(MATa ura3-52 leu2-3,\\2 trpl-289). Construction of expression plasmids was done in E. coli DH5a (F / endAl hsdRl7(rk-mk+) glnV44 thi-1 recAl gyrA(Naï) relAl A(lacZYA- argF)U169 deoR (<\>S0dlacA(lacZ)M15)). The plasmids p424HXT7 (TRP1), p425HXT7

(LEU2), and p426HXT7 (URA3) were used for heterologous gene expression. Constitutive expression is driven from the truncated promoter of the high-affinity glucose transporter gene

HXT7 and is terminated by the CYC1 terminator (16,26).

All physiological experiments were done in yeast minimal medium (YMM), containing

0.85 g/1 KH2P04, 0.15 g/1 K2HP04, 0.5 g/1 MgS04, 0.1 g/1 NaCl, 0.1 g/1 CaCl2, 500 ug/l

H3BO3, 63 |ig/l CuS04-5H20, 100 ug/l KI, 200 ug/l FeCl3, 450 u.g/1 MnS04H20, 235 ug/l

Na2Mo04-2H20, 712 ug/l ZnS04-7H20, 20 ug/l biotin, 2 mg/1 calcium pantothenate, 2 ug/l folic acid, 10 mg/1 inositol, 0.4 mg/1 nicotinic acid, 0.2 mg/1 p-aminobenzoic acid, 0.4 mg/1 pyridoxine hydrochloride, 0.2 mg/1 riboflavin, and 0.4 mg/1 thiamin hydrochloride. If not specified otherwise, the medium was also supplemented with 5 g/1 (NH4)2S04 and 5 g/1 glucose as nitrogen and carbon sources respectively. Uracil (20 mg/1), tryptophane (50 mg/1), or leucine (240 mg/1) were added, if necessary.

Growth conditions. Aerobic batch cultivations were performed in 500 ml baffled shake flasks with maximally 50 ml medium at 30°C on a gyratory shaker at 300 rpm. Cr adaptation experiments were performed by growing S. cerevisiae CEN.PK 113-7D for up to 100 generations in consecutive 3 ml batch cultures at 30°C and 300 rpm. Two types of carbon and nitrogen source combinations were used, with either 0.1% (w/v) glucose, 33 mM Cr, and

0.5% (w/v) ammonium sulfate, or 0.5% (w/v) glucose, 0.1% (w/v) ammonium sulfate, and

22.5 mM Cr, respectively. As positive and negative control, cultures with combinations of zero or 0.5% (w/v) glucose as well as zero or 0.5% (w/v) ammonium sulfate were also used.

After 24 h, the next batch culture was inoculated with 1.5% (v/v) in fresh medium and the intracellular Cr concentration was measured. To detect metabolism of Cr, 3 ml yeast minimal medium cultures were grown for up to 6 days at 30°C and 300 rpm, with either 40 mM Cr (as the sole carbon source), and 0.5% (w/v) ammonium sulfate, or 0.5% glucose (w/v) and 28 mM Cr (as the sole ammonium source).

99 tubes with 3 ml Transient pH stress experiments were performed in 10 ml culture

at medium at 30°C on a gyratory shaker at 300 rpm. Cultures were grown to an optical density

2 with 10% After 1 600 nm (OD600) of about 0.5, before the pH was set to (v/v) H3P04. h, pH

time was determined as the 5 was re-established using 100 mM KOH. The recovery period

5. required to reach a growth rate of at least 0.1 h"1 after re-establishing pH

cultures at a Starvation stress experiments were performed in glucose-limited chemostat dilution rate of 0.1 h"1. The culture volume was kept constant at 1 1 using a weight-controlled

the addition of 2M KOH. The airflow was pump and the pH was controlled at 5.0 by kept constant at 1.0 L/min and the agitation speed was adjusted to l'OOO rpm. The temperature was kept constant at 30°C. Oxygen and carbon dioxide concentration in the culture effluent gas

Prima After cultures were determined with a quadrupole mass spectrometer (Fisons 600, UK).

with constant CO2 and reached a stable steady-state, defined as at least 5 volume changes O2,

for ODôoo readings, the medium feed-pump was programmed so that the feed was interrupted

until a new was 150 s and initiated for 30 s. These 180 s cycles were maintained steady-state achieved, usually within 5 volume changes. For determination of intracellular metabolite and concentrations, aliquots were withdrawn at least 30 s after the feeding was interrupted before it was started again.

Constructions ofplasmids and molecular biology procedures. pBCK was constructed

Stolz and T. by PCR amplification of the chicken B-CK gene from plasmid pT23-l (M.

Wallimann, unpublished) with the primers 5'-CGCACTAGTATGCCCTCCTCAAA-3' and

5'-CGCGAATTCTTATTTCTGAGCTGG-3'. The resulting Spel I EcoRi fragment was then

of the chicken cloned into p424HXT7. pMtCK was constructed by PCR amplification

added as in-frame fusion for sarcomeric MtCK gene with a cytochrome ci pre-sequence mitochondrial targeting from plasmid pFG8 (R. Furter and T. Wallimann, unpublished) with the primers 5'-GACTAGTATGTTTTCAAATCTATCTAA-3' and 5'-

CCGCTCGAGTCACTTCCTGCCAAACT-3'. The resulting Spel I Xhol fragment was then

of the cloned into p426HXT7. pAGAT was constructed by PCR amplification L-arginine- with the glycine amidinotransferase gene from pig (18) from plasmid pBS/SK":RGAT-7 primers 5'CGCGGATCCATGCTGCGGGTGC-3' and 5'-

CGCGAATTCTCAGTCCAAGTAGGAC-3'. The resulting BamRl I EcoKl fragment was

the human then cloned into p425HXT7. pGAMT was constructed by PCR amplification of

with the guanidino acetate methyltransferase gene (20) from pGEM-T-4-hGAMT-7 primers

5'-CGCACTAGTATGAGCGCCCCCA-3' and 5'-CGCGAATTCTCAGCCTTTGGTCAC-

100 constructed 3'. The resulting Spel I EcoRl fragment was cloned into p426HXT7. pAK was by

PCR amplification of the Limulus polyphemus AK gene (38) from plasmid pET-22b(+):AK with the primers 5'-GGAATTCATGGTGGACCAGGCAACATTG-3' and 5'-

TGCGGTCGACTTAGGCAGCAGCCTTTTCCATC-3'. The resulting EcoRl / Sail fragment

PCR of the Rattus was cloned into p424HXT7. pCreaT was constructed by amplification

The norvegicus creatine transporter gene (22) from plasmid pRC/CMV-HCRT. resulting

was checked EcoRl / EcoRl fragment was cloned into p424HXT7 and correct insertion by

S.c. kit digestion analysis. Plasmids were transformed in S. cerevisiae using the EasyComp

(Invitrogen).

Crude cell Analytical procedures. Cell growth was monitored by the increase in ODôoo- extracts for determination of intracellular metabolite concentrations were prepared by

in 2 of buffer 20 mM washing cells in 4 pcv of cold water and resuspending pcv containing

Tris-HCl pH 7.9, 10 mM MgCl2, 1 mM EDTA, 5% (w/v) glycerol, 4 mM dithiothreitol, 0.3 M

(NH4)2S04, and 1 mM phenylmethanesulfonylfluorid (PMSF) at 0°C. 4 pcv ice-cold, acid-

vortexed for 1 min and washed 0.45-0.55 mm glass beads were added, the mixture was

times until incubated for 2 min on ice. This procedure was repeated up to 10 complete

for 60 disruption, as was verified by microscopic control. The liquid phase was centrifuged

and stored at -70°C for min at 12'000 g and 4°C, the supernatant frozen in liquid nitrogen, further analysis. For determination of protein activity, samples were analyzed within a week.

Phosphagen kinase activities. CK and AK activity were determined using pH-stat

mM analysis (24,43). Briefly, crude cell extracts were incubated at pH 7.0 and 25°C with 6.3

KCl, 8.3 mM MgCl2, 83 uM EGTA, 1 mM ß-mercaptoethanol, 4 mM ADP, and 10 mM of either PCr (CK) or PArg (AK), and the rate of stoichiometric proton consumption by conversion of PCr and ADP to Cr and ATP was monitored by titration.

Intracellular metabolite concentrations. Cr concentrations in supematants and crude

at cell extracts were determined with the above pH-stat analysis by incubating cell extracts pH

5 mM 83 8.0 and 25°C in a nitrogen atmosphere in a buffer containing 63 mM KCl, MgCb, uM EGTA, 1 mM ß-mercaptoethanol, 4 mM ADP, and 0.1 U/ml rabbit B-CK. Calibration was done with Cr concentrations between 1 to 20 mM. For Cr measurement, cells were washed three to five times with water prior to crude cell extract preparation.

101 mM Arg concentrations were determined enzymatically by adding 200 triethanolamine,

12 mM oc-ketoglutarate, 130 uM NADH, 2 mM ADP, 1.3 U/ml glutamate dehydrogenase, 10

and the in U/ml urease, and 6 U/ml arginase to crude cell extracts monitoring change absorbance at 340 nm (25).

cultures PArg concentrations were determined as described previously (28). Briefly,

Cells were in 50 mM aliquots were frozen in liquid nitrogen and stored at -70°C. resuspended

KCl, 2 mM EDTA, and 10 mM Tris-maleate pH 7.0 and disrupted by boiling for 1 min at

95°C with 0.6%o (v/v) trichloroacetic acid. This boiling step was necessary since incubation at lower temperatures did not result in complete disruption of the yeast cells (data not shown).

incubation at Stability of PArg during this treatment was experimentally verified. While acid

95°C for 2 min lead to significant hydrolysis of PArg, PArg concentrations after 1 min

incubation at 95°C were within 10% of those found after incubation at 0°C. The supernatant

8.2. To was then transferred to a new tube and the pH was set to precipitate adenyl

nucleotides, 1.5% (w/v) barium acetate was added and the mixture was incubated overnight

and for 5 min at on ice. The supernatant was supplemented with 100 uM Na2S04 centrifuged

acid was added to 2.3% and 13'000 rpm to remove residual barium ions. Trichloroacetic (v/v)

The liberated was samples were hydrolyzed for 1 min at 100°C. organic phosphate

mM determined colorimetrically at 623 nm after addition of 1.5% (v/v) H2S04, 10 Na2Mo04,

alcohol and 185 malachite and 10%) (v/v) of a color reagent containing 10 g/1 polyvinyl mg/1

green.

Intracellular ATP and ADP concentrations were determined as described previously

on acid-washed (7,40). Briefly, 1 ml culture samples were quickly withdrawn (within 1 s)

culture broth were glass beads that were pre-cooled at -20°C. 50 ul aliquots of supplemented

with 200 ul dimethylsulfoxide and 750 ul 25 mM Hepes (pH 7.75). Aliquots were then frozen

ATP at -70°C for further analysis. ATP concentrations were determined using an

bioluminescence kit HS II (Roche Biochemicals). ADP was converted to ATP by incubating

another aliquot for 30 min at 37°C with 1 mM phosphoenolpyruvate and 1 '250 U/ml pyruvate

calculated as kinase, and the total ATP level was then determined. ADP concentration were

the difference between total ATP and the ATP level in the other aliquot. Calibration was done

with ATP and ATP/ADP mixtures, in the concentration range of 1 and 1 '000 nM.

Intracellular metabolite concentrations were calculated from the experimentally

ml and a concentration of 108 determined concentration using a cellular volume of 1.162T0"11

cells/ml at an OD600 of 1.5 (44).

102 Determination of physiological parameters. In batch cultures, the exponential growth

with maximal rate phase was identified by log-linear regression of ODooo versus time, growth

to calculate rates, OD6oo (Umax) as the regression coefficient. In order specific production correlation factor values were converted to cellular dry weight (cdw) using a predetermined

concentration was calculated as cdw volume (0.52 g/1 cdw per OD6oo unit). The biomass per unit. The C02 evolution rate (CER) was defined as the relative C02 production (ACO2)

• = of the CER ÀCO2 F . multiplied by the effluent gas flow rate (F), on the basis relationship

the biomass The specific production rate for C02 (qco2) was calculated dividing the CER by

concentration (X) and the culture volume (V), on the basis of the relationship

qco2 = CER/(X-V).

103 Results

creatine Establishing a functional creatine kinase system by co-expression of

To establish a functional CK in S. biosynthetic enzymes plus creatine kinase. system cerevisiae (Figure 1), we first over-expressed the cytosolic brain-type isoenzyme (B-CK) and/or the ubiquitous mitochondrial isoenzyme (uMtCK). Transformants exhibited CK activity of 0.5 - 0.9 IU/mg protein (Figure 2), corresponding to the CK activity found in many

Cr from the chicken tissues (32). Based on a previous report (3), we expected uptake medium,

of a Cr although yeast genome data suggested the absence specific transporter gene (27,36).

During growth in YMM supplemented with 10 mM Cr, however, neither CK-expressing nor

of intracellular control yeast accumulated any appreciable amounts

Cr (Table 1).

supplemented Arginine Arginine transporter.

plasma membrane\

CITRATE trg[mnu ^ cyci f D10syntnesis pathway I GLYCOLYSIS PROTEINS < Arginine .— Glycine

urea ^^y Nsion Suffi CYCLE <—"" EKKUr^Guanidinoacetate /\> Creatine / t \_— SAM SAH

ATP-—¥ ^^ J ^-^ Phosphoarginine 133» "*"- ÀTP \ \T)X Phosphocreatine-^—^ ADP> ^ADP

Scheme Figure 1. S. cerevisiae engineered with either arginine kinase or creatine kinase systems. showing that are cellular uptake and biogenesis of the phosphagens analyzed. Black boxes indicate heterologous enzymes kinase; not naturally present in S. cerevisiae. Abbr.: AGAT, L-arginine:glycine amidinotransferase; AK, arginine S- CK, creatine kinase; GAMT, guanidinoacetate methyltransferase; SAM, S-adenosyl methionine; SAH, adenosyl homocysteine.

104 Strain Intracellular Cr cone. (mM)

Control 0.0 ±0.0

pBCK 0.0 ±0.0

pAGAT 0.12 ±0.2

pGAMT 0.0 ±0.0

pAGAT pGAMT 3.2 ±0.2

pAGAT pGAMT pBCK 3.3 ±0.2

strains creatine Table 1. Intracellular creatine concentrations. Cr was determined in S. cerevisiae expressing and/or the creatine kinase synthesis genes (pAGAT and pGAMT) cytosolic brain-type (pBCK).

1.0

0.8

SI r-S— 0,85 s X i 0.73 5 0.6 0,69 —=E—

0,54 0.4

u 0.2

0.0 0.0 0.0 control pAGAT pBCK pMtCK pBCK pAG\T pGAMT pMtCK pGAMT pBCK

strains. activity was Figure 2. Specific creatine kinase activity in different CK-engineered yeast Enzymatic 113-6B and determined in crude cell extracts harvested from stationary phase of S. cerevisiae CEN.PK (control) medium with 0.5% The brain-type isoform transformed yeast strains grown on yeast minimal glucose. cytosolic isoform of CK are expressed alone, in (as pBCK) as well as the mitochondrial sarcomeric (as pMtCK) amidinotransferase and combination, or together with the Cr synthesis genes L-arginine:glycine (as pAGAT) alone is also shown. guanidinoacetate methyltransferase (as pGAMT). Control with the Cr synthesis genes

that To activate potential unspecific Cr transport systems or silent genes (33) may

S. cerevisiae to Cr facilitate some Cr uptake, we attempted to adapt consumption.

113-7D in successive batches with Cr as Specifically, we grew S. cerevisiae CEN.PK

with or additional carbon or nitrogen source in combination glucose ammonium, respectively.

Cr not nor in In no case, however, did we detect intracellular (data shown), any growth

cultures with Cr as the unique carbon or nitrogen source.

105 To provide intracellular Cr to CK-expressing S. cerevisiae, we tried two strategies. As

from R. first, we expressed the creatine specific transporter norvegicus (22). Unfortunately,

the expression from the plasmid pCreaT resulted deleterious for growth even at reduced level, not allowing cultured cells to attain ODôoo higher than 0.3 (data not shown). Anyway, we

in the cell extract of CreaT cells could not detect any increased amount of creatine expressing

(data not shown), and therefore the CreaT expression experiments were abandoned.

for Cr that does As a second strategy, we installed the biosynthetic pathway naturally amidinotransferase not occur in yeast (27,36). Expression of human L-arginine:glycine intracellular (AGAT) and rat guanidinoacetate methyltransferase (GAMT) yielded an

concentration of about 3.2 mM Cr (Table 1), presumably resulting from utilization of

endogenous glycine and Arg. The minor concentration of Cr detected in cells expressing only

AGAT is probably due to the accumulation of guanidinoacetate, which, at high

concentrations, is also detected by the Cr assay (data not shown).

Establishing afunctional arginine kinase system in S. cerevisiae. S. cerevisiae displays

the endogenous capability for Arg biosynthesis and transport (37). Thus, functional of establishment of an AK system should depend solely on heterologous expression AK,

which is not naturally present in yeast (45). Independent of the amount of exogenously

supplied Arg, AK activities in crude extracts of cells harboring pAK were 3.7 IU/mg, while

exclude that became control cultures did not exhibit any AK activity (data not shown). To Arg

concentration in cultures at limiting in our system, we determined intracellular Arg grown

different extracellular Arg concentrations. Increasing extracellular Arg supplementation up to

At concentration 6 mM was reflected by increasing intracellular Arg concentrations.

observed exceeding 6 mM, however, no further increase of intracellular Arg was (Figure 3).

observed for control and AK In this respect, an essentially identical behavior was expressing

strains. While intracellular PArg concentrations were below detection level in the control

strain, AK expressing strains exhibited intracellular PArg concentrations of about 5 mM at all

investigated extracellular Arg concentrations, whereas intracellular Arg concentrations in

these AK expressing strains were similar to those seen in controls.

106 10

B O

en 3 5

4 8 12 16

Arg cone, in medium (mM)

was on medium concentrations. Arginine Figure 3. Dependence of intracellular arginine levels arginine and strain PK 113-6B pAK, D) at determined in control (CEN PK 113-6B p424HXT7, ) AK-expressing (CEN different arginine concentrations in medium

We S. Expression of AK but not CK shortens the lag phase after pH stress. grew

AK and 10 mM in cerevisiae CEN.PK 113-6B expressing either (pAK Arg supplementation

and on minimal the medium) or the complete CK-system (pBCK, pAGAT, pGAMT) yeast observe medium in shake flasks. Under standard growth conditions, we did not any

not We then improvement in maximal cdw and umax with respect to the controls (data shown). therefore examined specific stress conditions which provoke a drop in ATP and possibly

kinase confer an advantage to cells expressing a phosphagen system (17,29).

of cultures an A transient acidic stress was applied by shifting the pH mid-log phase (at

5. Under these OD60o of 0.8-1.0) for 1 h from pH 5 to pH 2, and subsequently returning to pH

well as the conditions, the ability of the cultures in recovering to the original umax value, as Cultures time needed to recover growth (defined as 'recovery time'), were investigated.

with to harbouring the complete CK-system as above did not show any improvement respect time from the controls (Figure 4b) In contrast, AK conferred the ability to reduce the recovery

times not the 3.3 h to 2.3 h (Figure 4a). As the experiment was repeated two (data shown),

between 3 and 5 whereas the time was always recovery time of the control varied h, recovery

controls and cells AK at least 1 h shorter with AK-expressing cells. The umax of expressing of remained unchanged. Thus it became obvious that the AK system facilitated a shortening

the time needed to recover full growth after pH stress.

107 a 2.5 -i

00 1 1 1 1

4

time (h)

b 25

"D 0.5 Q- =U= ' ' :H- i 3 7 h (wt) i 4.7 h (CK)

0 0 -I 1 1 ,

0 4

time (h)

Growth of S Figure 4. Influence of intracellular acidification on growth of engineered yeast strains. is as the increase cerevisiae CEN PK 113-6B either harbouring (a) the AK (O) or (b) the CK-system (D) given Yeast cultures were in shake flasks in biomass concentration together with the relative controls ( or ) grown to transient stress on yeast minimal medium, either kept at normal pH (pH 5, continuous lines) or subjected pH the time needed to (1 h at pH 2; dashed lines) The stress experiment was initiated by addition of acid at t=0, bottom The shows one out recover full growth rate after return to normal pH is show at the figure representative of 3 independent experiments performed

Expression ofAK stabilizes intracellular A TP levels during short-term starvation. Since heterologous AK expression improved resistance to transient pH stress, we wanted to investigate whether this improvement could also be seen under starvation stress and directly defined related to the ATP buffering capacity of AK (5). In order to expose S. cerevisiae for periods to energetic stress, glucose-limited chemostat experiments were used, which place

108 and biomass from cultures in a metabolic state where they efficiently generate energy glucose

culture was at a D of as the limiting nutrient (4,31). The S. cerevisiae wild-type grown

5 reactor volume 0.1 h"1 until a stable steady-state was attained after about changes.

of and for defined Subsequently, the feed pump was programmed to cycles activity pause

of starvation and slow periods, so that the cultures experienced alternating periods growth on/off during the intervals of the discontinuous feed. Specifically, we tested the following each cycle periods (in seconds): 60/60, 60/120, 30/120, 30/180, 20/150, and 30/150. For stable biomass discontinuous feeding profile, we allowed cultures to attain a new

The were the concentration that was usually attained after 5 volume changes. only exceptions

which did not 30/180 profile, which led to a washout of the culture, and the 60/60 profile,

concentrations did not reflect a true allow for attaining a steady-state. The stable biomass biomass physiological steady-state, hence is referred to as a pseudo steady-state. Generally,

concentrations in these pseudo steady-states during discontinuous feeding were lower than the during continuous feeding (data not shown), which is consistent with the notion that

in biomass periods of starvation constitute an energetic burden. The strongest decrease

s at a D of 0.1 h" and concentration was observed with a discontinuous feeding profile of 30

was chosen for further 150 s at a D of 0 h"1 (Table 2). Hence, this profile analysis.

Feed type Plasmid OD60o qco2 (mmol g"1 h"1)

Continuous control 4.44 0.69 ±0.07

Continuous pAK 4.92 0.52 ±0.05

Discontinuous control 3.25 0.15 ±0.02

Discontinuous control 3.28 n.a.

Discontinuous pAK 4.28 0.07 ±0.01

Discontinuous pAK 4.34 0.07 ±0.01

Table 2. Growth and specific C02 production rate. OD60o and specific C02 production rate (qccc) were cultures of control and determined in steady-state before and in pseudo steady-state in continuous AK-expressing discontinuous was obtained with a D of ,S cerevisiae Continuous feeding was done at a D of 0 1 h"1 and feeding the 150 s starvation 0 h"1 for 150 s and 0 1 h"1 for 30 s For discontinuous feeding, the qC02 values during period for are given The data of two independent experiments are given pseudo steady-states

As with the wild-type control, AK-expressing cells were then grown in glucose-limited

was about 10% in chemostat at a D of 0.1 h"1. The steady-state biomass concentration higher

the AK-expressing culture (Table 2). Consistent with this higher biomass yield from the is available glucose in the medium, less substrate carbon was used for respiration, as

109 illustrated by the reduced specific CO2 production rate in the AK-expressing culture (Table

2). Upon imposing the above chosen discontinuous feeding profile, a new pseudo steady-state decreased biomass was attained after about 5 medium changes with a significantly concentration (Table 2). However, the AK-expressing culture was apparently less stressed by the harsh conditions, since its pseudo steady-state biomass concentration was about 30% higher than the concentration of the control strain under the same conditions.

To verify whether the improved biomass yield of the AK-expressing culture was indeed

determined intracellular ATP and related to the temporal energy buffering function of AK, we

ADP concentrations in these cultures. Within seconds after feed interruption, the intracellular

ATP pool dropped from 3.1 mM to 1.5 mM in wild-type S. cerevisiae and remained at this

level until the onset of feeding (Figure 5). The AK-expressing culture, however, did not

control and display such a drop in ATP levels. During the continuous feeding steady-state,

AK-expressing cultures exhibited rather similar intracellular ATP and ADP level (Figure 6). in A similar intracellular ATP concentration was observed during the 30 s feeding interval

pseudo steady-state (Figure 5). In contrast, ADP levels were found to be relatively constant

under all conditions in both strains (Figure 6).

0 30 60 90 120 150 180 210 Time (s)

starvation stress in AK- Figure 5. Time course of C02 production and intracellular ATP during intracellular ATP engineered yeast. The specific C02 production rate (qco2, O and , continuous lines) and 113-6B concentration (D and , dashed lines) during starvation in chemostats of S cerevisiae CENPK pAK (open symbols) and control (filled symbols) The bold line represents the discontinuous feeding profile (30 s on, a of two 150 s off) Values are given as mean ± SD of three independent measurements over period days

110 4.0

3.5 * 3.0 i

2.5

2.0

1.5

1.0 -f- 0.5

0.0

Ctrl, not Ctrl, AK,not AK, stressed stressed stressed stressed

Concentrations of ATP Figure 6. Intracellular ATP and ADP in control and AK-engineered yeast. (grey in bars) and ADP (white bars) in steady-state before stress begins (label 'not stressed') and pseudo steady-stress Concentrations (label 'stressed') in continuous chemostats of control (ctrl) and AK-expressing S. cerevisiae. 150 s feed The data the during pseudo steady-state were determined in the middle of the interruption. represent of two in The samples were mean of 3 independent measurements over a period days pseudo steady-state. withdrawn at least 30 seconds after the feed offset and before the feed onset.

Ill DISCUSSION

functional of The present study shows for the first time a successful expression could be phosphagen kinase systems in S. cerevisiae. The AK- and the CK-systems engineered in & cerevisiae. Both transgenic strains showed enzymatic activities that (i) are

and about 2 mM PCr as calculated from the able to generate a phosphagen pool (5 mM PArg

i.e. a rather fast flux from the Cr) and (ii) are high enough to allow equilibrium conditions, under phosphagen to ATP that theoretically could recover the cellular energy charge energy

stress.

intracellular Cr are a Since yeast does not naturally synthesize Cr, means to generate

Our results obtained with the major prerequisite for installing a functional CK system. yeast

model strain CEN.PK clearly demonstrate the absence of measurable intracellular Cr

of CK concentrations during growth or incubation with 15 mM Cr, independent expression. and utilization Long-term selection experiments were designed to possibly enable Cr uptake

not show of culture as an additional carbon or nitrogen source. They did any improvement

in which is sufficient physiology over a period of about 100 generations liquid media, usually

to known to identify such improved metabolic phenotypes (33). Thus, not only homologues

also Cr via Cr transporters are absent in the yeast genome (27,36), but unspecific transport

These results contradict an earlier NMR general amino acid permeases (12) does not occur.

study that expressed rabbit muscle CK in different S. cerevisiae strains to analyze intracellular

ADP concentrations (3). In these cultures, when maintained at 100 mM Cr, Brindle et al. (3)

reported the accumulation of 50-100 mM intracellular Cr, determined in a biochemical assay.

At these However, this cannot be taken as a final proof for cellular Cr-uptake. extremely high

the extracellular matrix and concentrations, Cr is at the solubility limit and may be trapped in

et at the whiskered cell wall of yeast cells, thus escaping washing procedures (Schlattner al.,

observed in these cells unpublished results). In fact, the very low concentration of PCr yeast

not rather suggests that the Cr is separated from the cytosolic CK and therefore mainly (if

is liberated entirely) extracellular. Remnant PCr may have been produced by CK possibly

exclude a Cr- from a few damaged cells. Although we can presently not faint, unspecific conclude that uptake at extremely high extracellular Cr concentrations, we nevertheless with AGAT and generally Cr-uptake does not occur in S. cerevisiae, and that transformation

GAMT is necessary to generate an intracellular Cr pool.

to Of the two phosphagen kinase systems that we installed in yeast, only AK appears

conditions that an confer an appreciable advantage to S. cerevisiae under impose energetic

after transient reduction stress. AK-yeast showed a clearly reduced lag phase in growth pH

112 Artificial decrease of the extracellular pH forces cells to counteract acidification of the cytoplasm, probably by increased activity of the plasma membrane H+-ATPase (17). An extra

maintain cells to boost of energy from the accumulated PArg pool may help AK-expressing

faster then the control. This resume growth upon relieve of the transient pH stress wild-type

because the conditions were phenotype is likely not a truly increased resistance to lower pH chosen such that little if any cell death occurred.

The hypothesis of PArg as a temporal energy buffer in AK-expressing yeast was

defined of addressed more specifically in chemostat cultures that were exposed to periods growth and starvation. These transient stress challenges reduced the biomass yield of wild- type S. cerevisiae significantly, but had only very little negative influence on AK-expressing

levels cells. Consistent with the above hypothesis, we found stable intracellular ATP during the starvation phase of the recombinant yeast. Since the intracellular level of ADP was similar

AMP. This is due in both cultures, it appears that most of the hydrolyzed ATP accumulated as

of to the presence of three isoenzymes and a rather high enzymatic activity yeast adenylate

AMP from two molecules of kinase (2). These enzymes catalyze the formation of ATP and concentration of ATP ADP, thus buffering the cellular energy charge (21). The intracellular

S. cerevisiae strain that and ADP determined here compare favourably with those of another

the time scale of the was grown under almost identical culture conditions (39). Moreover, rapid ATP decrease at the onset of starvation is in qualitative agreement with the data of

Theobald et al. (39), who investigated metabolic responses to a glucose pulse in glucose- limited chemostat culture.

Given the protective effects of AK in S. cerevisiae, the question remains why the CK

system did not confer similar advantages under the examined stress conditions, a phenomenon

account for this already seen with CK-transgenic tobacco (9). Several reasons may discrepancy. First, the synthesized pool of 3 mM intracellular Cr may give rise to not more than 2 mM PCr, which is much lower than the 5-30 mM PCr occurring in native animal

PCr concentrations tissues (42). In fact, with a Km(PCr) for chicken B-CK of 1.4 mM, these

do not favour fast conversions. Furthermore, the intracellular pH is much lower in yeast (pH

in the muscle 5.5-6.2) (17,19) as compared to the natural environment of CK, e.g. resting (pH

7.0-7.2) (42). The acidic intracellular pH of yeast reduces B-CK activity by about 50% and

leads to partial hydrolysis of PCr, which is less stable than PArg under acidic conditions (6).

Finally, the heterologous expression of three proteins (CK, AGAT, GAMT) on different

cultures that counteracts beneficial plasmids may impose a metabolic burden onto the yeast

effects. In contrast to CK, the AK system seems to combine several advantages for expression

113 it uses an intrinsic in unicellular organisms that can undergo intracellular pH fluctuations: substrate (Arg) and is able to accumulate a more acid-stable phosphagen (PArg) to higher concentrations.

of Our results with AK demonstrate that energy buffering is an intrinsic property

distant This is phosphagen kinases that can be transferred to phylogenetically very organisms.

in a that is free a first step towards the analysis of phosphagen kinases background naturally

of of those kinases. Yeast, with its ease of genetic manipulation and the availability specific mutants, is ideally suited to study in more detail the molecular physiology of phosphagen

a kinases, e.g. Cr-stimulation of oxidative phosphorylation (34). Moreover, installing functional phosphagen kinase system such as shown here for AK appears to be a pertinent metabolic engineering strategy to improve the biotechnological potential of microbes that are exposed to energetically stressful conditions. One such condition may be large-scale

of nutrients such as processes in which cells are often exposed to fluctuating availability carbon source and oxygen, due to imperfect mixing (8)

Eckhard Acknowledgements. We would like to express our gratitude to Boles,

Heinrich Heine Universität, Düsseldorf, Germany for the p42xHXT7 plasmids, Robert Huber,

Max-Planck-Institut für Biochemie, Martinsried, Germany for cDNA encoding for AGAT,

Dirk Isbrandt Universität Hamburg, Germany for cDNA encoding for GAMT, Patrick Schloss

Max-Planck-Institut für Brain Research, Frankfurt am Main, Germany for cDNA encoding

for the pRC/CMV-CHOT15 plasmid. and Ross Ellington, Florida State University, for the

AKcDNA.

114 REFERENCES

1. Attfield, P. V. (1997). Stress tolerance: the key to effective strains of industrial baker's yeast. Nat Biotechnol 15, 1351-1357.

2. Bandlow, W., Strobel, G., Zoglowek, C, Oechsner, U., and Magdolen, V. (1988). Yeast adenylate kinase is active simultaneously in mitochondria and cytoplasm and is required for non-fermentative growth. J Biochem 178, 451-457.

3. Brindle, K., Braddock, P., and Fulton, S. (1990). 3,P NMR measurements of the ADP concentration in yeast cells genetically modified to express creatine kinase. Biochemistry 29, 2395-3302.

4. Dauner, M., Storni, T., and Sauer, U. (2001). Bacillus subtilis metabolism and energetics in carbon-limited and excess-carbon chemostat culture. J Bacteriol 183, 7308-7317.

5. Ellington, W. R. (1989). Phosphocreatine represents a thermodynamic and functional Biol 177-194. improvement over other muscle phosphagens. J Exp 143,

6. Ellington, W. R. (2001). Evolution and physiological roles of phosphagen systems. Annu Rev Physiol 63, 289-325.

7. Emmerling, M., Bailey, J. E., and Sauer, U. (2000). Altered regulation of pyruvate fluxes in kinase or co-overexpression of phosphofructokinase increases glycolytic resting Escherichia coli. Biotechnol Bioeng 67, 623-627.

8. Enfors, S. O., Jahic, M., Rozkov, A., Xu, B., Hecker, M., Jürgen, B., Kruger, E., Schweder, T., Hamer, G., O'Beirne, D., Noisommit-Rizzi, N., Reuss, M., Boone, L., Hewitt, C, McFarlane, C, Nienow, A., Kovacs, T., Tragardh, C, Fuchs, L., Revstedt, J., Friberg, P. C, Hjertager, B., Blomsten, G., Skogman, H., Hjort, S., and Hoeks, F., Lin, H. Y., Neubauer, P., van der Lans, R., Luyben, K., Vrabel, P., J Manelius, A. (2001). Physiological responses to mixing in large scale bioreactors. Biotechnol 85, 175-185.

9. Farres, J., Holmberg, N., Schlattner, U., Bailey, J. E., Wallimann, T., and Kallio, P. T. (2002). Expressing creatine kinase in transgenic tobacco—a first step towards Res 49-59. introducing an energy buffering system in plants. Transgenic 11,

10. Fernandes, A. R., and Sa-Correia, I. (2001). The activity of plasma membrane H(+)-ATPase is strongly stimulated during Saccharomyces cerevisiae adaptation to acidification. Yeast growth under high copper stress, accompanying intracellular 18, 511-521.

11. Greenhaff, P. L. (2001). The creatine-phosphocreatine system: there's more than one song in its repertoire. J Physiol 537, 657.

12. Grenson, M., Mousset, M., Wiame, J. M., and Bechet, J. (1966). Multiplicity of the for a amino acid permeases in Saccharomyces cerevisiae. I. Evidence specific arginine-transporting system. Biochim Biophys Acta 127, 325-338.

13. Grieshaber, M. K., Hardewig, L, Kreutzer, U., and Portner, H. O. (1994). Rev Physiological and metabolic responses to hypoxia in invertebrates. Physiol Biochem Pharmacol 125, 43-147.

115 14. Hamilton, C. A., Good, A. G., and Taylor, G. J. (2001). Vacuolar H+-ATPase, but in not mitochondrial FIFO-ATPase, is required for aluminum resistance Saccharomyces cerevisiae. FEMS Microbiol Lett 205, 231-236.

15. Hatano, E., Tanaka, A., Iwata, S., Satoh, S., Kitai, T., Tsunekawa, S., Inomoto, T., Shinohara, H., Chance, B., and Yamaoka, Y. (1996). Induction of endotoxin 663- tolerance in transgenic mouse liver expressing creatine kinase. Hepatology 24, 669.

16. Hauf, J. (1998). Transcriptional regulation of the PDC1 gene and overexpression cerevisiae. PhD thesis studies of seven glycolytic enzymes in the yeast Saccharomyces studies of at the Transcriptional regulation of the PDC1 gene and overexpression PhD seven glycolytic enzymes in the yeast Saccharomyces cerevisiae, thesis, Technische Hochschule Darmstadt, Darmstadt, pp. 116.

17. Holyoak, C. D., Stratford, M., McMullin, Z., Cole, M. B., Crimmins, K., Brown, and A. J., and Coote, P. J. (1996). Activity of the plasma membrane H(+)-ATPase of optimal glycolytic flux are required for rapid adaptation and growth Saccharomyces sorbic acid. Environ cerevisiae in the presence of the weak-acid preservative Appl Microbiol 62, 3158-3164.

18. Humrn, A., Huber, R., and Mann, K. (1994). The amino acid sequences of human and pig L-arginine:glycine amidinotransferase. FEBSLetters 339, 101-107.

19. Imai, T., and Ohno, T. (1995). The relationship between Viability and Intracellular pH in the Yeast Saccharomyces cerevisiae. Appl Environ Microbiol 61, 3604-3608. of human 20. Isbrandt, D., and von Figura, K. (1995). Cloning and sequence analysis guanidinoacetate N-methyltransferase cDNA. Biochim. Biophys. Acta 1264, 264-267.

21. Lehninger, A. L. (1975). Biochemistry. 2nd edition Ed., Worth Publichers, New York.

novel 22. Mayser, W., Betz, H., and Schloss, P. (1991). Isolation of cDNAs encoding a 203-206. member of the neurotransmitter transporter gene family. FEBS Lett 295,

23. Miller, K., Halow, J., and Koretsky, A. P. (1993). Phosphocreatine protects and ischemia. Am J transgenic mouse liver expressing creatine kinase from hypoxia Physiol265, C1544-1551.

24. Milner-White, E. J., and Watts, D.C. (1971). Inhibition of adenosine 5'- triphosphate-creatine phosphotransferase by substrate-anion complexes. Biochem J 122, 727-740.

25. Mira de Orduna, R. (2001). Quantitative determination of L-arginine by enzymatic end-point analysis. J Agric Food Chem 49, 549-552.

26. Mumberg, D., Müller, R., and Funk, M. (1995). Yeast vectors for the controlled expression of heterologous proteins in different genetic backgrounds. Gene 156, 119- 122.

27. NCBI, National Center for Biotechnology Information. Entrez Nucleotide Database, http://www.ncbi.nlm.nih.gov/. the 28. Noguchi, M., Sawada, T., and Akazawa, T. (2001). ATP-regenerating system in cilia of Paramecium caudatum. J Exp Biol 204, 1063-1071.

29. Panaretou, B., and Piper, P. W. (1990). Plasma-membrane ATPase action affects several stress tolerances of Saccharomyces cerevisiae and Schizosaccharomyces

116 of the heat shock J Gen Microbiol pombe as well as the extent of duration response. 136,1763-1770. M. 30. Pereira, C. A., Alonso, G. D., Paveto, M. C, Iribarren, A., Cabanas, L., kinase Torres, H. N., and Flawia, M. M. (2000). Trypanosoma cruzi arginine Biol characterization and cloning. A novel energetic pathway in protozoan parasites. J Chem 275, 1495-1501. balance of 31. Russell, J. B., and Cook, G. M. (1995). Energetics of bacterial growth: anabolic and catabolic reactions. Microbiol Rev 59, 48-62.

32. Saks, V. A., Khuchua, Z.A., Vasilyeva, E.V., Yu Belikova, O., and Kuznetsov, muscle cells. A.V. (1994). Metabolic compartmentation and substrate channeling in Mol Cell Biochem 133/134, 155-192. microbial 33. Sauer, U. (2001). Evolutionary engineering of industrially important phenotypes. Adv Biochem Eng Biotechnol 73, 129-169. and 34. Schlattner, U., Forstner, M., Eder, M., Stochowiak, O., Fritz-Wolf, K., mitochondrial Wallimann, T. (1998). Functional aspects of the X-ray structure of Mol Cell Biochem 125-140. creatine kinase: a molecular physiology approach. 184,

a novel of 35. Serrano, R. (1984). Plasma membrane ATPase of fungi and plants as type 87-126. proton pump. Curr Top Cell Regul 23,

36. Snow, R. J., and Murphy, R. M. (2001). Creatine and the creatine transporter: a review. Mol Cell Biochem IIA, 169-181.

of lower 37. Sophianopoulou, V., and Diallinas, G. (1995). Amino acids transporter eukaryotes: regulation, structure and topogenesis. FEMS Microbiol 16, 53-75. of the 38. Strong, S. J., and Ellington, W. R. (1995). Isolation and sequence analysis the chelicerate Limulus insights gene for arginine kinase from arthropod, polyphemus: into catalytically important residues. Biochim Biophys Acta 1246, 197-200. In vivo 39. Theobald, U., Mailinger, W., Baltes, M., Rizzi, M., and Reuss, M. (1997). analysis of metabolic dynamics in Saccharomyces cerevisiae. Biotechnol Bioeng 55, 305-316.

and the cellular 40. Tran, Q. H., and Unden, G. (1998). Changes in the proton potential energetics of Escherichia coli during growth by aerobic and anaerobic respiration or by fermentation. Eur J Biochem 251, 538-543. The 41. Viegas, C. A., Almeida, P. F., Cavaco, M., and Sa-Correia, I. (1998). H(+)- ATPase in the plasma membrane of Saccharomyces cerevisiae is activated during in growth latency in octanoic acid-supplemented medium accompanying the decrease intracellular pH and cell viability. Appl Environ Microbiol 64, 779-783.

42. Wallimann, T., M. Wyss, D. Brdiczka, K. Nicolay, H.M. Eppenberger. (1992). in Intracellular compartmentation structure and fucntion of creatine kinase isoenzymes circuit' for tissues with high and fluctuating energy demands: the 'phosphocreatine 21-40. cellular energy homeostasis. Biochem J281, of M-line 43. Wallimann, T., Schlösser, T., and Eppenberger, H.M. (1984). Function at the end of the bound creatine kinase as intramyofibrillar ATP regenerator receiving phosphorylcreatine shuttle in muscle. J Biol Chem 259, 5238-5246. in The 44. Winge, Ö., and Roberts, C. (1958)Life, history, and cytology of yeast New York. chemistry and biology ofyeast (Ed. H. Cook), pp. 115 ff, Academic Press,

117 creatinine metabolism. 45. Wyss, M., and Kaddurah-Daouk, R. (2000). Creatine and Physiol Rev SO, 1107-1213.

118 CHAPTER 5

Functional expression of arginine kinase improves

coli recovery from pH stress of Escherichia

FABRIZIO CANONACO, UWE SCHLATTNER, THEO WALLIMANN,

AND UWE SAUER Abstract

and Acid stress in Escherichia coli involves a complex resource- energy- kinase from consuming response mechanism. By overexpression of arginine transient Limulus polyphemus in E. coli, we improved the recovery from a pH

a transient reduction 3 stress. While wild type E. coli resumed growth after pH topH

than the the kinase for 1 h with a rate that was 25% lower before stress, arginine

This is expressing strain continued to grow as rapidly as before. effect presumably caused by the physiological function of arginine kinase as a short term energy

cannot be buffer in the form of phosphoarginine, but a pH-buffering effect excluded.

120 Introduction

that serve as Cells of higher organisms often contain so-called phosphagen kinases the transient ATP buffering systems during high or fluctuating energy requirements. Perhaps

in all as well as most well-known of these is creatine kinase (CK) which is found vertebrates,

CK exert and temporal energy in sponges and in some worms (8). isoenzymes spatial homeostasis kinase buffering functions and are involved in intracellular pH (24). Arginine

which occurs in insects (13), (AK; EC 2.7.3.3) is a conceptually simpler phosphagen kinase, transient ATP crustaceans (25) and in certain unicellular organisms (17). Although buffering

be excluded. AK uses is the main function of AK (8), an involvement in pH buffering cannot

of according to arginine (Arg) to create a metabolically inert pool phosphoarginine (PArg), the following reaction (25):

MgADP" + PArg2-+ H+ <=± MgATP2" + Arg. environmental Although microorganisms are often exposed to rapidly changing

to have evolved conditions and fluctuating availability of energy, phosphagen kinases seem and only with metazoans (8). Exceptions are the unicellular eukaryotes Trypanosoma spp. transfer Since the Paramecium caudatum, probably as the result of horizontal gene (1,16-18).

of an high motility of these latter organisms requires large bursts of energy, the availability

ATP buffering system obviously conferred a significant evolutionary advantage. demand Bacteria frequently encountered environmental stresses that generate a severe

in the case of Escherichia coli The most for ATP (14), as for example exposure to low pH (3). in activation of an which variations prominent response to pH stress is ATPase, compensates of ATP cytoplasmic pH by extruding protons to the periplasm, with concomitant consumption ATP from fueling (14). Moreover, severe intracellular acidification impairs production

inhibition Further mechanisms include pathways as a consequence of enzyme (4). protective and amino acid proton extruding ion antiporters and symporters (4,11,26) proton-consuming

decarboxylases (7). of AK in the unicellular yeast Recently, we showed that heterologous expression

also to overcome a Saccharomyces cerevisiae provided a transient ATP buffer, which helped of transient, energy-demanding stress of short term starvation (6). Functional expression

from our own lab showed that creatine kinase was demonstrated in E. coli (15), but data

on cellular Since AK creatine kinase had no apparent beneficial effect physiology (5).

extend here our studies on transient ATP-buffering expression was successful in yeast, we in E. coli. systems to prokaryotes by overexpressing AK from Limulus polyphemus

121 Expérimental procedures

Strains and plasmids construction. E. coli wild-type strain MG1655 (X, F", rph-1) was used for all physiological experiments, and E. coli DH5a (F' / endAl hsdR17(r^-m^) glnV44

used for thi-1 recAl gyrA(Na\r) relAl A(lacZYA-argF)U169 deoR (<\»S0dlacA(lacZ)M15)) was used for genetic constructions. The expression vector p7rc99A (Pharmacia Biotech) was

driven from the IPTG-inducible trc heterologous gene expression. Expression was strong, from promoter (2). The AK-encoding gene of L. polyphemus was PCR-amplified plasmid pET-22b(+):AK (22) with the primers 5'-GGAATTCATGGTGGACCAGGCAACATTG-3' I Sail and 5'-TGCGGTCGACTTAGGCAGCAGCCTTTTCCATC-3'. The resulting EcoRl

which was transformed in E. coli by fragment was cloned into p7>c99A to yield pAK,

electroporation using a GenePulser (Biorad).

Media and growth conditions. Luria-Bertani (LB) and M9 minimal medium were

medium was with at a final prepared as described previously (21). M9 supplemented glucose of concentration of 5 g/1 and with 50 mg/1 ampicillin for plasmid maintenance. For induction

desired concentration. Cultivation AK expression, IPTG was added prior to inoculation at the

shake flasks with 50 ml medium at was performed aerobically in 500 ml baffled maximally

37°C on a gyratory shaker at 200 rpm.

cultures to an Transient pH stress experiments were performed by growing optical

in M9 medium with a of 6.5, before the was density at 600 nm (ODöoo) of about 0.5 pH pH 100 mM reduced to 3 with 10% (v/v) H3P04. After 1 h, pH 6.5 was re-established using

KOH.

cultures to an Benzoate and acetate stress experiments were performed by growing

added at OD6oo of about 0.5 in M9 medium, before sodium benzoate or sodium acetate were

final concentrations of 2 mM and 8 mM, respectively.

in Crude cell Analytical procedures. Cell growth was monitored by the increase ODôoo- from extracts for determination of AK activity and ATP concentration were prepared

and 10 mM stationary phase cells by 10 to 50-fold concentrated culture aliquots in 0.9% NaCl

W for 1 min on ice. MgSÛ4. Cells were then disrupted by sonication at 100

within a For determination of AK activity, samples were frozen at -70°C and analyzed

ATP and week. AK activity using pH-stat analysis, arginine concentration, and intracellular Intracellular ADP concentrations were determined as described previously (6,9,23).

concentration metabolite concentrations were calculated from the experimentally determined

using a cellular volume of 3.2 ul / mg of protein.

122 Determination of physiological parameters. In batch cultures, the exponential growth

of versus with maximum growth phase was identified by log-linear regression OD6oo time,

calculate rates, ODöoo values rate (|4,max) as the regression coefficient. To specific production correlation factor of 0.51 were converted to cellular dry weight (cdw) using a predetermined

concentration was calculated as cdw volume unit. g/1 (cdw) per OD60o unit. The biomass per

123 Results

transformation of into E. Establishing a functional arginine kinase system. Upon pAK concentration of coli MG 1655, the maximum in vitro activity of AK was attained at an IPTG

AK as 100 jxM (Figure 1). This value corresponds to about the same level of specific activity

found in muscles of L. polyphemus (18).

control 20 50 100 100 /Arg 200 400

IPTG concentration (jlM)

was Figure 1. Specific arginine kinase activity at different levels of IPTG induction. Enzymatic activity the control is E. coli determined in crude cell extracts harvested from stationary phase, pAK carrying E. coli, mM Error bars indicate the with p7>c99A. The hatched bar indicates cells grown in the presence of 10 Arg. standard deviation from triplicate determinations.

AK- To determine the impact of Arg feeding on intracellular Arg levels, we grew While expressing and control strains at different extracellular Arg concentrations (Figure 2). the the AK-expressing strain always had slightly lower intracellular Arg concentrations,

difference with respect to the control strain became smaller when 10 mM or more Arg were intracellular added to the medium. The control strain already reached the maximum Arg

the concentration of about 20 mM at 5 mM Arg supplementation, while the AK strain attained

maximum intracellular concentration at 10 mM Arg supplementation.

124 •rs S OD I» « c o t. ri -w « S U -w

u e u ai

s« e +* o s u

5 -

4 8 12 16 20 Extracellular arginine concentration (mM)

concentrations. Intracellular Arg Figure 2. Intracellular arginine levels at different extracellular arginine of E. coli MG1655 concentrations were determined at different levels of Arg supplementations during growth harboring the control plasmid () or pAK (D).

AK had no Expression of AK improves recovery after transient pH stress. expression

the biomass on appreciable influence on the maximum specific growth rate or yield glucose during batch growth, when compared to the control (Table 1). In fact, both values were

medium increased both and in slightly reduced. The presence of 10 mM Arg in the |i,max Y(X/s) control cultures but had little effect on AK expressing cells (Table 2).

IPTG concentration Maximum specific gi

(mM) (h'1) (g/g)

control 0.67 ±0.01 0.36 ±0.01

0 0.62 ±0.01 0.35 ±0.01

50 0.60 ±0.02 0.33 ±0.00

100 0.60 ±0.03 0.32 ±0.00

400 0.54 ±0.02 0.33 ±0.01

deviations on the indicated values are Table 1. AK induction level on growth rate and yield. The standard from duplicate experiments.

125 Strain Maximum specific growth Biomass yield on glucose

rate (h1) (g/g)

before pH stress

control, 0 mM arginine 0.61 ±0.01 0.34 ±0.02

±0.01 pAK, 0 mM arginine 0.60 ±0.00 0.32

control, 10 mM arginine 0.67 ± 0.00 0.36 ±0.01 ±0.01 pAK, 10 mM arginine 0.56 ±0.01 0.32

after pH stress

control, 10 mM arginine 0.51 ±0.00 0.32 ±0.00

±0.00 pAK, 10 mM arginine 0.67 ±0.01 0.31

1 h from 6.5 to 3. In all cases, Table 2. Growth rate and yield before and after a transient pH reduction for The standard deviations on the indicated values are from the M9 medium was supplemented with 100 uM IPTG. duplicate experiments.

AK transient stress, that Next, we analyzed the potential impact of expression during pH

other effects the of was expected to exert an ATP drain among (3,14). Specifically, pH and exponentially growing cultures (OD6oo of 0.5) was shifted for 1 h from pH 6.5 to pH 3,

afterwards followed by re-establishing pH 6.5 (Figure 3 and Table 2).

10 12 14 Time (h)

to 3 of AK-expressing (O) and Figure 3. Impact of a transient medium acidification pH during growth medium was with 10 mM control () strains. The induction level was 100 |xM IPTG and the M9 supplemented h at dashed or to transient pH stress (1 arginine. Cultures were either kept at normal pH (pH 6.5; lines) subjected pH 3; continuous lines). The figure shows one representative of 3 independent experiments performed.

126 acid A pH reduction to 3 is well below the value for which the induction of the

intracellular ATP level from 5.5 ± 0.2 tolerance response is reported (12). Indeed, the dropped the AK- mM in unstressed cells to 4.2 ± 0.2 mM during low-pH stress in both the control and

the expressing strain (Figure 4). While lag-phases were similar in both strains after the stress,

resumed at post-stress growth rates differed significantly. The AK-expressing strain growth slower. about the same rate as observed before the stress, but the control grew significantly

/.u

5.7 B 6.0 5.4 (3 T o

•-Ö 5.0 - es u 4.3 4.2 a T i. a T 1 u 4.0 - a ©

Oh H 3.0 < Im J3

- 's 2.0 =a 4) U es 1.0

on

pTrc99A pAK

concentrations were Figure 4. Intracellular ATP levels during the transient pH stress. Intracellular ATP MG 1655 determined in unstressed cultures (white bars) and during transient stress (grey bars) of E. coli with 10 mM The of one of harboring the control plasmid or pAK. The M9 medium was supplemented Arg. pH both from stressed and unstressed cultures. the cultures was reduced to 3 and after 1 h a sample was harvested measurements in two Intracellular ATP concentrations were determined as the mean from three independent parallel cultures.

in Expression ofAK during acetate and benzoate stress. By effect of the anion liberated

effect than the cytoplasm, weak organic acids are known to generate a larger stress strong acids (12,20). Acetate and benzoate stresses were applied here to control and AK-expressing and strains by incubating mid-exponential phase cultures with sub-lethal quantities of acetate and is benzoate (Figure 5). The anionic form of these acids becomes protonated at neutral pH, thus able to diffuse into the cytoplasm, where dissociation occurs.

strain No significant differences were observed between the control and AK concerning the maximum cell density attained. The concentrations of added acetate and benzoate were

127 We observed a chosen to reduce the maximal specific growth rate by about 20-30% (19). the reduction of the maximal growth rate by 11% and 21% in the strain transformed with if control plasmid after addition of acetate or benzoate, respectively (Table 3). However, pAK the is transformed and AK-expression is induced in presence of 10 mM arginine, negative

reduced 27% and 32% after effect was even stronger. Since the maximal growth rate was by

addition of acetate or benzoate, respectively.

a)

3.5

3.0

2.5

2.0 = ©

O O 1.5

1.0

0.5

0.0

5 7 9 11 13 Time (h)

b)

3.5

3.0

2.5

2.0 = ©

Q O 1.5

1.0

0.5

0.0

5 7 9 11 13 Time (h)

of strain. Figure 5. Effect of sub-lethal acetate and benzoate concentrations on growth AK-expressing the control Growth of E. coli MG1655 expressing AK (dashed lines) or just harboring plasmid p7>c99A in M9 medium (continuous lines) is given as the increase in cell density. Growth was performed supplemented were either allowed with 0.5% glucose (w/v), 50 mg/1 ampicillin, 100 uM IPTG, and 10 mM arginine. Cultures or 8 mM sodium or with 2 mM sodium benzoate (a; O) to grow undisturbed (control cultures; ) supplemented out of 2 acetate (b; ) after reaching OD60o 0.5. The figure shows one representative independent experiments performed.

128 maximum specific maximum specific maximum specific

growth rate (h1) growth rate (h1) growth rate (h"1)

non-stressed Ac-stressed Bz-stressed

Plasmid cultures cultures cultures

control 0.69 ± 0.02 0.61 ±0 01 0.54 ±0.01

pAK 0.61 ±0.02 0.41 ± 0 04 0.44 ±0.03

Table 3. Growth rates of AK cultures stressed with sub-lethal concentrations of acetate and benzoate. the control Maximal growth rates (Umax) of E coli MG 1655 expressing AK (dashed lines) or just harboring plasmid p7>c99A (continuous lines) during incubation with acetate or benzoate are given Growth was 50 100 uM IPTG, and 10 performed in M9 medium supplemented with 0 5% glucose (w/v), mg/1 ampicillin, undisturbed or with 8 mM arginine Cultures were either allowed to grow (non-stressed cultures) supplemented after mM sodium acetate (Ac-stressed cultures) or 2 mM sodium benzoate (Bz-stressed cultures) reaching OD60o 0 5 The figure shows the mean values from the two independent experiments performed

129 Discussion

transient and starvation Previously, we demonstrated an improved resistance to pH into the S. stress by metabolic engineering of a functional phosphagen kinase system yeast

in a cerevisiae (6). Although another phosphagen kinase was functionally expressed

beneficial effect of prokaryote before (15), we show here for the first time a phosphagen

from a transient stress kinase overexpression in a prokaryote; namely improved recovery pH

the since in E. coli. Apparently, AK expression has little if any protective effect during stress,

the AK strain. global intracellular ATP levels were similarly reduced in control and Although the AK- both strains resumed growth immediately upon reestablishment of the original pH, control. it expressing strain attained reproducibly a much higher growth rate than its Thus,

from a transient which reduced appears that AK expression improves recovery stress,

of intracellular ATP concentration. The lack of a beneficial effect of AK in presence organic

not involved in acids may be due to their inhibitory role on biosynthetic pathways strictly

ATP-generation(19).

Our study provides indirect evidence for the formation of the storage phosphagen PArg, when since the intracellular Arg level was always lower in strains overexpressing AK,

to compared to the control. Moreover, more than 10 mM extracellular Arg was necessary

achieve maximum intracellular Arg concentrations in the AK expressing strain, while 5 mM

of was converted was sufficient for the control, presumably because a significant portion Arg

to PArg by the AK.

The beneficial effect of arginine kinase may be related to overcome an energy-related or

kinase was so far not shown a low-pH-related buffering. While pH buffering by arginine results would also be explicitly in organisms that express arginine kinase naturally, our

consistent with proton scavenging by dephosphorylation of phosphoarginine (see equation 1).

lower The results presented here and in a previous study (6) show that even eukaryotes

transient and prokaryotes can be metabolically engineered with ATP-buffering systems

This be of working in the range of seconds and perhaps minutes. may biotechnological demands and relevance for industrial conditions that are characterized by fluctuating energy

with and thus availability, as is found for example in large-scale bioreactors imperfect mixing

local zones with oxygen or carbon source limitation (10).

Florida Acknowledgements. We would like to thank Ross Ellington and Pamela S. Pruett of

State University for providing the AK cDNA.

130 REFERENCES

1. Alonso, G. D., Pereira, C. A., Remedi, M. S., Paveto, M. C, Cochella, L., Ivaldi, M. S., Gérez de Burgos, N. M., Torres, H. N., and Flawia, M. M. (2001). Arginine and kinase of the flagellate protozoa Trypanosoma cruzi. Regulation of its expression catalytic activity. FEBS Lett 498, 22-25.

2. Amann, E., Ochs, B., and Abel, K. J. (1988). Tightly regulated tac promoter vectors useful for the expression of unfused and fused proteins in Escherichia coli. Gene 69, 301-315.

in 3. Bearson, S., Bearson, B., and Foster, J. W. (1997). Acid stress responses enterobacteria. FEMS Microbiol Lett 147, 173-180.

4. Booth, I. R., Cash, P., and O'Byrne, C. (2002). Sensing and adapting to acid stress.

Antonie van Leeuwenhoek 81, 33-42.

Institute of 5. Çakar, Z. P. (2000). Metabolic engineering of yeast. PhD thesis at the Biotechnology, ETH, Zurich, pp. 130.

6. Canonaco, F., Schlattner, U., Pruett, P. S., Wallimann, T., and Sauer, U. (2002). Functional expression of phosphagen kinase systems confers resistance to transient Biol Chem stresses in Saccharomyces cerevisiae by buffering the ATP pool. J 277, 31303-31309.

J. 7. Castanie-Cornet, M. P., Penfound, T. A., Smith, D., Elliott, J. F., and Foster, 3525-3535. W. (1999). Control of acid resistance in Escherichia coli. J Bacteriol 181,

8. Ellington, W. R. (2001). Evolution and physiological roles of phosphagen systems. Annu Rev Physiol 63, 289-325.

9. Emmerling, M., Dauner, M., Ponti, A., Fiaux, J., Hochuli, M., Szyperski, T., Wüthrich, K., Bailey, J. E., and Sauer, U. (2002). Metabolie flux responses to 152-164. pyruvate kinase knockout in Escherichia coli. J Bacteriol 184,

10. Enfors, S. O., Jahic, M., Rozkov, A., Xu, B., Hecker, M., Jürgen, B., Kruger, E., Schweder, T., Hamer, G., O'Beirne, D., Noisommit-Rizzi, N., Reuss, M., Boone, L., Hewitt, C, McFarlane, C, Nienow, A., Kovacs, T., Tragardh, C, Fuchs, L., Revstedt, J., Friberg, P. C, Hjertager, B., Blomsten, G., Skogman, H., Hjort, S., and Hoeks, F., Lin, H. Y., Neubauer, P., van der Lans, R., Luyben, K., Vrabel, P., scale bioreactors. J Manelius, A. (2001). Physiological responses to mixing in large Biotechnol 85, 175-185. coli K- 11. Epstein, W., and Kim, B. S. (1971). Potassium transport loci in Escherichia 12. J Bacteriol 108, 639-644.

12. Foster, J. W. (1999). When protons attack: microbial strategies of acid adaptation. Curr Opin Microbiol 2, 170-174.

13. Grieshaber, M. K., Hardewig, I., Kreutzer, U., and Portner, H. O. (1994). Rev Physiological and metabolic responses to hypoxia in invertebrates. Physiol Biochem Pharmacol 125, 43-147.

14. Kobayashi, H., Suzuki, T., and Unemoto, T. (1986). Streptococcal cytoplasmic pH J is regulated by changes in amount and activity of a proton-translocating ATPase. Biol Chem 261, 627-630.

131 is 15. Koretsky, A. P., and Traxler, B. A. (1989). The B isozyme of creatine kinase in vivo detection 31P NMR. FEBS active as a fusion protein in Escherichia coli: by Lett 243, 8-12. the 16. Noguchi, M., Sawada, T., and Akazawa, T. (2001). ATP-regenerating system in cilia of Paramecium caudatum. J Exp Biol 204, 1063-1071.

17. Pereira, C. A., Alonso, G. D., Paveto, M. C, Iribarren, A., Cabanas, M. L., Torres, H. N., and Flawia, M. M. (2000). Trypanosoma cruzi arginine kinase characterization and cloning. A novel energetic pathway in protozoan parasites. J Biol Chem215, 1495-1501.

18. Pereira, C. A., Alonso, G. D., Torres, H. N., and Flawia, M. M. (2002). Arginine in African and kinase: a common feature for management of energy reserves American flagellated trypanosomatids. JEukaryot Microbiol 49, 82-85. of 19. Roe, A. J., O'Byrne, C, McLaggan, D., and Booth, I. R. (2002). Inhibition and Escherichia coli growth by acetic acid: a problem with methionine biosynthesis homocysteine toxicity. Microbiology 148, 2215-2222.

20. Salmond, C. V., Kroll, R. G., and Booth, I. R. (1984). The effect of food Gen Microbiol 130 Pt preservatives on pH homeostasis in Escherichia coli. J ( 11), 2845-2850.

21. Sauer, U., D.R. Lasko, J. Fiaux, M. Hochuli, R. Glaser, T. Szyperski, K. Wüthrich, and J.E. Bailey. (1999). Metabolic flux ratio analysis of genetic and environmental modulations of Escherichia coli central carbon metabolism. J. Bacteriol. 181, 6679-6688. of the 22. Strong, S. J., and Ellington, W. R. (1995). Isolation and sequence analysis chelicerate Limulus insights gene for arginine kinase from the arthropod, polyphemus: into catalytically important residues. Biochim Biophys Acta 1246, 197-200.

23. Unden, G., and J. Bongaerts. (1997). Alternative respiratory pathways of Escherichia coli: energetics and transcriptional regulation in response to electron acceptors. Biochim Biophys Acta 1320, 217-234. 42- 24. Wallimann, T. (1994). Dissecting the role of creatine kinase. Current Biology 1, 46. metabolism. 25. Wyss, M., and Kaddurah-Daouk, R. (2000). Creatine and creatinine Physiol Rev 80, 1107-1213. coli is 26. Zakharyan, E., and Trchounian, A. (2001). K± influx by Kup in Escherichia Microbiol Lett 61-64. accompanied by a decrease in H+ efflux. FEMS 204,

132 Conclusions and outlook

133 metabolism. 2 Our investigations provided new insights in microbial cofactor Chapter

NADPH metabolism. and Chapter 3 show that transhydrogenases are important enzymes for efficient For the particular example of Escherichia coli, we could show that they provide

from the anabolic NADPH demand. To means of decoupling catabolic NADPH production

shows that the membrane-bound our knowledge, this is the first study that quantitatively

transhydrogenase contributes to 34 % of the anabolic NADPH requirement in exponentially

thus for radical redirection of growing E. coli on glucose. These enzymes are important An carbon flow in chemicals overproducing micro-organisms by metabolic engineering.

of the important question that needs to be addressed is how the cell regulates the activity

soluble transhydrogenase. It cannot be active during all conditions because it would then just

with the more oxidized NADH / NAD+ equilibrate the more reduced NADPH / NADP+ pool

for in vitro pool, which would be detrimental to growth. Hence, although we have evidence

- it should be in active activity of the soluble transhydrogenase on glucose a conditions where of the - The there is likely a mechanism that restricts the in vivo activity. complementation

to lethal transhydrogenase deletion phenotype under certain conditions, could be exploited

for in Bacillus subtilis, about detect enzymes with transhydrogenase-like activity, example

whose NADPH metabolism little is known.

in metabolic of micro¬ The use of heterologous phosphagen kinase systems engineering

be under organisms, confirmed how energy buffering systems may advantageous specific 4 and 5 of conditions of reduced ATP availability, such as during stress. In Chapter Chapter

has been but since reduced ATP this thesis, only a restricted palette of conditions applied, conditions availability is involved in a large number of microbial stress responses, more scale should be investigated, for example hypoxic stress. Such conditions are present in large

of nutrients such as processes in which cells are often exposed to fluctuating availability

carbon source and oxygen because of imperfect mixing. functional creatine Some difficulties were encountered in the successful expression of a

much This kinase system, while the establishment of the arginine kinase system was simpler.

distance to the host cells used. Creatine kinase may may be a consequence of the evolutionary

in Such an thus be more appropriate to achieve similar goals higher organisms. example

would be immobilized hepatocytes cultures, because of the reduced oxygen availability.

in E. coli we have not that Lastly, in the case of arginine kinase overexpression fully proved this should be indeed phosphoarginine was used as a transient energy buffer. Thus, question

also addressed.

134 LIST OF ABBREVIATIONS

6PG gluconate 6-phosphate

AcCoA acetyl coenzyme A

AdiA arginine decarboxylase

ADP adenosine di-phosphate

AGAT L-arginine:glycine amidinotransferase

AK arginine kinase

AMP adenosine mono-phosphate

cAMP cyclic AMP

AMPK AMP-activated protein kinase

AMPKK AMPK-activated protein kinase

ANT adenine nucleotide translocator

Arg arginine

ARS acid resistance systems

ATP adenosine tri-phosphate

cdw cell dry weight

CER CO2 evolution rate muscle CK creatine kinase (B-CK, cytosolic brain isoform, M-CK; cytosolic

isoform; sMtCK, sarcomeric mitochondrial creatine kinase; uMtCK,

ubiquitous mitochondrial creatine kinase)

Cr creatine

DTT dithiothreitol

E. coli Escherichia coli

E4P erythrose 4-phosphate

ED pathway Entner-Doudoroff pathway

Eda KDPG aldolase

Edd 6-phosphogluconate dehydratase

EDTA ethylenediamine tetraacetic acid

EGTA ethylene glycol-bis(ß-aminoethylether)-N,N,N',N'-tetraacetic acid

F6P fructose 6-phosphate

G6P glucose 6-phosphate

GABA y-aminobutyric acid

GadAB glutamate decarboxylase

135 GadC glutamate/GABA antiporter

GAMT guanidinoacetate methyltransferase

GAP glyceraldehyde 3-phosphate

GK glycocyamine kinase

Glu glutamate

HdeAB periplasmic chaperones

HTK hypotaurocyamine kinase

IPTG isopropyl-ß-D-1 -thiogalactopyranoside

KDPG 2-keto-3-deoxy 6PG

Kup H+/K+ symporter

LB medium Luria-Bertani medium

Leu leucine

LK lombricine kinase

MAL malate

Mmax maximum specific growth rate

METAFoR metabolic flux ratios

MTBSTFA N-(te^butyldimethylsilyl)-N-methyl-trifluoroacetamide

NAD(H) oxidized (reduced) nicotinamide adenine dinucleotide

NADP(H) oxidized (reduced) nicotinamide adenine dinucleotide phosphate

NhaA Na+/K+ antiporter

NMR nuclear magnetic resonance

OAA oxaloacetate

OD600 optical density at 600 nm

OGA oxoglutarate

OK ophaline kinase

P5P pentose 5-phosphate

PArg phosphoarginine

PCr phosphocreatine

PCR polymerase chain reaction

pcv packed cell volume

PEP phosphoenolpyruvate

PGA 3-phosphoglycerate

Pgi phosphoglucose isomerase

PKA AMP-dependent protein kinase

136 PMSF phenylmethanesulfonylfluorid

PntAB membrane-bound transhydrogenase

PP pathway Pentose phosphate pathway

PYR pyruvate

evolution rate / rate qcœ / qo2 specific CO2 O2 consumption

rate qs specific substrate-uptake

RpoS stationary-phase dependent sigma factor os

S. cerevisiae Saccharomyces cerevisiae

S7P sedoheptulose 7-phosphate

SAH S-adenosyl homocysteine

SAM S-adenosyl methionine

TCA cycle tricarboxylic acid cycle

ThK thalassemine kinase

TK taurocyamine kinase

TrkA H+/K+ antiporter

Trp tryptophane

UdhA soluble transhydrogenase

Ura uracile

YMM yeast minimal medium

Yx/s yield of biomass on substrate

Zwf G6P dehydrogenase

137 Seite Leer / Blank leaf

138 CURRICULUM VITAE

Name Fabrizio Canonaco

Birth date and place 22.04.1975, Locarno (Switzerland)

Domicile Via alia Peschiera 7, 6600 Locarno (Switzerland)

Nationalities Swiss and Italian

Training

1981-1990 primary and secondary school in Locarno

1990-1994 high school in Locarno, maturity certificate type C (scientific)

1994-1999 graduated at the ETH Zurich, faculty of biology, direction

biotechnology

Jul - Oct 1997 Semester work at the Institute of Biotechnology, ETH Zurich

Supervisors PD Dr U Sauer, Dr C Weikert

Oct 1998 - Apr 1999 Diploma work at the Institute of Biotechnology, ETH Zurich

Supervisors. PD Dr U Sauer, Prof J E Bailey

1999-2003 PhD thesis / assistant at the Institute of Biotechnology, ETH Zurich

Supervisors PD Dr U Sauer, Prof T Wallimann, PD Dr U Schlattner

List of publications

Christian Weikert, Fabrizio Canonaco, Uwe Sauer, and James E. Bailey (2000)

Co-overexpression of RspAB improves recombinant protein production in Escherichia coli

Metabol Engl 293-299

Fabrizio Canonaco, Tracy A. Hess, Sylvia Heri, Taotao Wang, Thomas Szyperski, and Uwe Sauer (2001)

coli and of Metabolic flux response to phosphoglucose isomerase knock-out in Escherichia impact overexpression of the soluble transhydrogenase UdhA

FEMS Microbiol Lett 204 247-252

Fabrizio Canonaco, Uwe Schlattner, Pamela S. Pruett, Theo Wallimann, and Uwe Sauer (2002)

Functional expression of phosphagen kinase systems confers resistance to transient stresses in Saccharomyces cerevisiae by buffering the ATP pool

J Biol Chem 277(35) 31303-31309

139