NMR studies of the bioenergetics and metabolism of microorganisms

NMR studies of the bioenergetics and metabolism of

microorganisms

Citation for published version (APA):

Nicolaij, K. (1983). NMR studies of the bioenergetics and metabolism of microorganisms. Rijksuniversiteit

Groningen.

Document status and date:

Published: 04/03/1983

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be

important differences between the submitted version and the official published version of record. People

interested in the research are advised to contact the author for the final version of the publication, or visit the

DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page

numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne

Take down policy

If you believe that this document breaches copyright please contact us at:

openaccess@tue.nl

...

NMR STUDIES OF THE BIOENERGETICS

AND METABOLISM OF MICROORGANISMS

Rijksuniversiteit te Groningen

NMR STUDIES OF THE BIOENERGETICS

AND METABOLISM OF MICROORGANISMS

Proefschrift

ter verkrijging van het doctoraat in de

Wiskunde en Natuurwetenschappen

aan de Rijksuniversiteit te Groningen

op gezag van de Rector Magnificus Dr. L. J. Engels

in het openbaar te verdedigen op

vrijdag 4 maart 1983 des namiddags te 4.00 uur

door

KLAAS NICOLAY

geboren te Stiens

BIBLIOTHEEK

8 302657

~

T.H.ElNDHOVEN

Eerste promotor: Prof. dr. R. Kaptein Tweede promotor: Prof. dr. W.N. Konings

Referent: Dr. K.J. Hellingwerf

This research was carried out at the Netherlands national NMR facility at the University of Groningen, supported by the Netherlands Foundation for Chemical Research (S.O.N.) with financial aid from the Netherlands Organization for the Advancement of Pure Research (Z.W.O.).

aan Gielie, Esther en Martine

aan mijn ouders

Een groot aantal mensen hebben een essentiele bijdrage aan dit boekje

geleverd. Graag grijp ik de gelegenheid aan hen daarvoor te bedanken.

Rob Kaptein,

promotor, bedank ik voor zijn stirrrulerende begeleiding.

De vrijheid, die hij me geboden heeft om eigen wegen in te slaan, heb ik zeer

op prijs gesteld.

Zonder de onmisbare hulp en ondersteuning van Klaas Dijkstra bij de NMR

experimenten, zou dit proefschrift er niet zijn gekomen. Of er nou

groentesoep dan wel tomatensoep op het menu stond, altijd was hij als "chef

de cuisine" paraat.

Wil Konings bedank ik voor zijn interesse en inbreng m.b.t. de

bioenergetische aspecten van het onderzoek.

Aan Klaas HeUingwerf heb ik zeer veel te danken. Door zijn enthousiaste

bijdrage is het werk, na een moeizame start, steeds voorspoediger verlopen.

Verder denk ik met veel plezier terug aan de samenwerking met Hans van

Gemerden. Aan zijn grote Uefde,

Chromatium vinosum,

ben ook ik gehecht

geraakt.

Lex Scheffers en Peter Bruinenberg wil ik bedanken voor hun initiatief om

NMR onderzoek aan gisten te beginnen. Zonder hun inzet en enthousiasme

zouden Hoofdstuk 9 en 10 ontbroken hebben.

Met plezier denk ik terug aan de periode van samenwerking met Jan Snel.

Helaas wilden, ondanks zijn creatieve aanpak, de spinazie-chloroplasten en

de algen hun geheimen niet aan het NMR apparaat kwijt.

Naast hen, die ik al genoemd heh wil ik voor de prettige samenwerking de

volgende mensen bedanken: Lodewijk Tielens, Jos van den Heuvel, Piet van

Dijck, Niek Vermue, Peter Hore, Erik Zuiderweg, Ruud Scheek, Sijtze Stob,

Ijsbrand van der Leeuw, Herman Berendsen, Fons Starns, Theo Hansen, Juke

Lolkema, George Robillard, Roel Riegman, Rob Beudeker, Johannes Boonstra,

Wim Harder, Marten Veenhuis, Ger Telkarrrp, Hidde Prins and Ad Schapendonck.

Greetje Lap en Anita Severijnse wil ik bedanken voor de nauwgezette wijze,

waarop zij de manuscripten hebben getypt. Berend Kwant bedank ik voor zijn

inspanningen op synthetisch terrein; Klaas Gillissen voor het fantastische

foto-werk. Willem Zevenberg, Bert van Dammen en Henk Bruinenberg dank ik

voor hun onmisbare hulp bij het bouwen van de benodigde opstellingen, terwijl

Bernard van Meurs altijd berid was voor ondersteuning op het electronische

vlak te zorgen. Jaap Deen wil ik bedanken voor het maken van het vereiste

CHAPTERS

I II III IV

v

C 0 N T E N T S Introduction 31 .

P nuclear magnetic resonance studies of energy

transduction in

Rhodopseudomonas sphaeroides

[ Eur.

J.

Biochem. 116

(1981) 191-197 ]

Quantitative agreement between the values for the

light-induced ~pH in

Rhodopseudomonas sphaeroides

measured with automated flow-dialysis and 31P NMR

[ FEES Lett. 123

(1981) 319-323 ]

The light-induced pH gradient in chromatophores

of

Rhodopseudomonas sphaeroides

as visualised by

31 P NMR

[ Photobiochem. Photobiophys. 2

(1981) 311-319 ]

Carbon-13 nuclear magnetic resonance studies of

acetate metabolism in intact cells of

Rhodopseu-domonas sphaeroides

[ Biochim. Biophys. Acta 720

(1982) 250-258 ]

9

31

41

49

CHAPTERS

VI 31 P NMR studies o . f p otophosp ory ation in intact h h 1 . . . 73

cells of

Chromatium vinosum

[ FEES Lett. 138

(1982) 249-254

VII

In vivo

phosphorus-31 and carbon-13 NMR studies 81

of acetate metabolism in

Chromatium vinosum

[ J. Bacterial.,

submitted for publication)

VIII 31P-nuclear magnetic resonance and freeze-fracture 103 electron microscopic studies of reconstituted

bac-teriorhodopsin vesicles

[ Eur.

J.

Biochem. 11?

(1981) 639-645 )

IX Phosphorus-31 nuclear magnetic resonance studies of 113

intracellular pH, phosphate compartmentation and phosphate transport in yeasts

[ Arch. Microbial.

133 (1982) 83-89 )

x The dynamics of phosphate pools and intracellular pH in 123

. d b 31 1 .

yeasts as studie y P nuc ear magnetic resonance

[Arch. Microbial.,

submitted for publication)

CHAPTER

I

CHAPTER I

INTRODUCTION

I. General Considerations

In the past fifteen years nuclear magnetic resonance (NMR) has become one of the most important experimental techniques for the research on molecular structures and intermolecular interactions in biological systems. The

applications of NMR to biological research include studies of the structural and dynamic properties of proteins, nucleic acids, membranes and a large number of low molecular weight compounds of biological interest. A more recent development is its application to intact cells and tissues. The state of the art at present is such that very large objects can be accomodated by wide bore magnets making studies on human beings feasible.

NMR applications to living material have largely developed along two lines. High resolution studies at high magnetic field strengths are mostly

( . 31 d 13 )

concerned with the study of metabolic phenomena mainly P an C NMR .

Anatomical studies are performed by the method of spin-imaging or "zeugmatography" which is carried out at relatively low magnetic fields. Since the subject of this thesis is the application of high resolution NMR methods to intact biological systems the imaging approach will not be dealt with in this introduction. For a review of recent developments in this area the reader is referred to Ref. 1.

A large number of nuclei give rise to the NMR phenomenon, i.e. all nuclei possessing non-zero spin. However, only a limited number of nuclei have proved so far to be informative in biological high resolution NMR. These nuclei, together with their NMR properties, are listed in Table 1. It is clear from the table that, apart from the intrinsic sensitivity of a specific nucleus, also its natural abundance may set limits to its

13

usefulness in NMR. For example, C NMR at natural abundance isotope levels

is almost invariably useless for intact biological systems because of a combination of low intrinsic sensitivity, low natural abundance and the metabolite concentrations usually encountered. However, if isotope

enrichment is feasible 13

c

NMR is a very attractive method in metabolic

Table 1. NMR properties of the nuclei employed in NMR of intact cells and Nucleus lH 13c 15N 19F 23Na 31p tissues Resonance frequency at 8.46 Tesla (MHz) 360 90.5 36.5 338.7 95.2 145.8 Natural abundance (%) 99.98 1. 1 0.37 100 100 100

Relative sensitivity for equal number of nuclei at constant field 1.000 1.6 x 10-2 1.0 x 10-3 8.3 x 10-1 9.3 x 10-2 6.6 x 10-2

Hence, the metabolism of the enriched substrate fed to the intact preparation can be followed specifically without interference with the rest of the

metabolic machinery.

NMR in relation to other analytical techniques

The major advantage of NMR over more classical analytical techniques is that i t is non-invasive, i.e. the system is monitored while still being intact. This allows the continuous monitoring of the same sample under a variety of different conditions without the need of quenching the preparation. Thus, the metabolic activity as observed by NMR can be directly related to the physiological state of the system.

Furthermore, the recordings made by NMR contain many important

parameters in a single spectrum. Thus, 31P NMR spectra from a suspension of

+

compartmentation of H and phosphate metabolites. Parallel arguments hold for

13

c

NMR studies of substrate metabolism since all compounds labeled with 13

c

that are present in high enough concentration will be manifested in the

spectra. This implies that not only

a priori

anticipated metabolites will be

observed but also unexpected intermediates formed during substrate utilization may be encountered. The latter is exemplified by the massive

formation of butyrate as a major endproduct of the metabolism of 2-13

c-acetate in the phototrophic bacterium

Rhodopseudomonas sphaeroides

as

described in Chapter 4. The accumulation of butyrate by this organism had not been described previously and, therefore, was totally unexpected.

Small metabolites bound to macromolecules (e.g. a ligand complexed to a high molecular weight protein) can be regarded immobilized on the time scale of the NMR experiment. Consequently, only "free", i.e. metabolically available, molecules give rise to narrow peaks in conventional high

resolution NMR. In this respect, NMR is clearly superior to classical extraction methods which measure total metabolite levels.

The maximum sample volume that can be used in NMR is restricted by the dimensions of the bore of the magnet. Modern spectrometers suitable for

in vivo

NMR of whole cells can accomodate tubes with a diameter of 20 mm

for 31P and 13

c

NMR experiments. All experiments described in this thesis

have been carried out with "10 mm-probes". Since NMR is a relatively insensitive spectroscopic technique, this necessitates a high density of biological material to be present within the boundaries of the radio frequency coil. For whole cell NMR this requires the application of dense cell suspensions. This is one of the major drawbacks of NMR because such high concentrations of cells are not encountered under physiological conditions. Therefore, i t must be ascertained that the parameters which are measured have physiological meaning and do not represent artifacts arising from the rather extreme experimental conditions. For example, when

studying yeast cells under aerobic conditions i t must be proven that oxygen concentrations in the cell suspension are well above the Km for oxygen utilization by this organism. Chapter 10 describes an aeration-system which allows adequate oxygenation of yeast cell suspensions under NMR

conditions (see also Ref. 2). The viability of

Chromatium vinosum

cells just

after harvest was the same as after a period of up to 6 hours under NMR circumstances (Chapter 7). Moreover, the metabolic behaviour of these cells

observations. This means that under NMR conditions these cells display a physiologically relevant behaviour.

The problem of using high density suspensions becomes even more severe, when light is required as a "substrate". Thus, light-dependent phenomena in photosynthetic microorganisms can usually only be studied by NMR under limi ted conditions. This is exemplified in Chapters 2 and 3 by the light-induced pH gradient (6pH) across the cytoplasmic membrane of the

photosynthe-tic bacterium

Rps. sphaeroides.

The 6pH generated by light is, although

homogenous over the entire sample because of efficient mixing, suboptimal due to the high optical density suspensions required for NMR measurements. In this specific case, this problem could be overcome by measuring the 6pH as a function of decreasing cell density, i.e. by extrapolation to infinite dilution. This allows an estimation of the 6pH under light-saturated conditions. Clearly, such a solution to light-limitation will only be successful, if the light-generated state is rather long-lived on the time scale of the mixing procedure. This condition is easily fulfilled for the

light-induced 6pH in

Rps. sphaeroides

since upon darkening the pH gradient

decays very slowly in this organism (Chapter 2).

Paramagnetic broadening effects

Another problem that may be encountered in studies of intact biological samples is the broadening of resonances in NMR spectra due to the

interactions of the metabolites to be observed with paramagnetic species. A

number of ions, in particular those from transition metals (e.g. Mn, Co, Ni),

may cause desastrous broadening of NMR lines at relatively low concentrations.

2+ .

For example, free Mn concentrations above 10 µM would lead to unobservable

31

P NMR peaks from phosphorylated metabolites (3). Especially, ATP resonances are very sensitive to the presence of paramagnetic ions because of the high

31

binding constant of metal ion to ATP. In P NMR spectra of whole cells of

Rps. sphaeroides

this led to the apparent absence of resonances from ADP and

ATP while these nucleotides could be easily demonstrated chemically (Chapter 2). A combination of two approaches has proven to be of value to overcome this paramagnetic broadening effect. When dealing with microorganisms grown in liquid culture, the concentrations of paramagnetic trace elements in the growth medium can usually be lowered without affecting its growth. Thus, the

for several generations without any observable effects upon their growth

characteristics. For the purple sulfur bacterium

Chr. vinosum

this

. . 31

procedure suffices to produce high quality P NMR spectra. However,

Rps. sphaeroides

cells still contain rather high concentrations of 31

paramagnetic species leading to severely broadened P NMR peaks. In this

case a time-consuming washing procedure with buffer containing the

chelating agent EDTA was necessary. As pointed out in Chapter 2, this EDTA treatment must be carried out with care. In the absence of divalent cations

2+

(e.g. Mg ) EDTA deteriorates the structural integrity of the membranes of

Rps. sphaeroides

which is accompanied by a greatly increased proton permeability of its cytoplasmic membrane.

The solution to the problem of paramagnetic broadening will depend upon 31

the nature of the organism studied. In spite of serious attempts, P NMR

studies of chloroplasts from

Spinacia oleracea

and the cyanobacterium

2+

Anacystis nidulans

were unsuccessful due to too high internal levels of Mn

(J.F.H. Snel and K. Nicolay, unpublished experiments). In this case, the

concentration of Mn2+ in the growth medium cannot simply be reduced since

manganese is involved in the water splitting reaction of these photosynthetic systems.

The type of information available from NMR

Before discussing a number of applications of whole cell NMR relevant to this thesis, the type of information which can be extracted from NMR is summarized below.

The first problem that has to be solved before doing detailed metabolic studies by NMR is that of the assignment of the resonances in spectra. In many cases this is rather straightforward. The preparation of cell-free extracts and the recording of their corresponding NMR spectra are a great help in this respect. The most important method of assignment using extracts is to accumulate spectra as a function of pH. Spectra obtained before and after the addition of expected compounds may provide further evidence that

assignments are indeed correct. As an example Figure shows a 31P NMR

spectrum from a cell-free extract of the yeast

Candida utilis.

The spectrum

contains resonances of phosphorylated metabolites which are frequently

2

1

5

3

6

8

-5

6,ppm

-15

17

-25

FIG. 1:

31

P NMR spectrum from a perchloric acid extract of the yeast

Candida utilis.

1, glucose-6-phosphate; 2, fructose-6-phosphate;

3,

inorganic

phosphate; 4, glycerophosphorylethanolamine; 5, glycerophosphorylserine;

6, glycerophosphorylcholine; ?, terminal phosphates from polyphosphate

chains longer than tripolyphosphate; 8, the y-phosphate of ATP; 9, the

S-phosphate of ADP; 10, pyrophosphate and terminal phosphates of

tripolyphosphate; 11, the a-phosphate of ADP; 12, the a-phosphate of ATP;

13, NAD; 14, uridinediphosphoglucose; 15, penultimate phosphates of

polyphosphate, including middle phosphates of tripolyphosphate; 16, the

S-phosphate of ATP; 17, inner phosphates from polyphosphate chains longer

than tripolyphosphate. (K. Nicolay, unpublished observation).

NMR can be employed to assess the following parameters:

(i) internal and external pH;

(ii) internal and external metabolite concentrations;

(iii) kinetics of cellular processes, both under equilibrium and non-equilibrium conditions;

(iv) intracellular compartmentation.

A detailed description of the parameters monitored by NMR of whole cells can be found in the excellent monograph by Gadian (4). Information on more specific applications of NMR to the study of cellular metabolism is given in a number of recent reviews (2, 5-11).

II. Applications to Living Cells

Concentration measurements

Most applications of NMR on whole cells involve the measurement of

~bsolute

or relative concentrations of metabolites. 31P NMR applications are

' ' ' b 13 d ' 1 d ' '

in the maJority ut also C an , to a minor extent, H NMR are use in this

' 13 1

respect. For experimental reasons, C and H NMR usually are only applied to

measure relative concentrations. The major advantage of NMR over extraction methods is that NMR measures free concentrations of metabolites in their natural environment. By monitoring resonance areas as a function of time, the kinetics of reactions can be followed as they proceed within the living cell. The time resolution of such kinetic measurements is limited by the time required to record a NMR spectrum with sufficient signal-to-noise. Although dependent upon many factors, the time resolution will roughly be of the order

of 0.5 - 5 min for 31P and up to 15 min for 13

c

NMR. This thesis describes

the measurement of metabolite concentrations under different external

conditions in

Rps. sphaeroides

(Chapter 5),

Chr. vinosum

(Chapters 6 and 7)

and in yeasts (Chapters 9 and 10) .

Measurement of internal pH

A special kind of measurement of free metabolite levels by NMR involves the determination of the hydrogen ion concentration (i.e. pH). The knowledge of intracellular pH is very important in the understanding of bioenergetics

sensitive to pH can serve as a pH probe. In practice, however, only a limited number of endogenous metabolites are usable in this respect. Phosphorus-31 is the most powerful nucleus for pH determination. Most cells contain readily observable levels of, in particular, inorganic phosphate while in addition the Pi chemical shift is very sensitive to pH in the physiologically relevant range.

2-and HP0

4 around neutral pH. These two

species would give rise to two resonances separated by about 2.4 ppm in the absence of chemical exchange. However, the two species exchange very rapidly with each other in solution. As a result the experimental spectrum consists of a single resonance, the frequency of which is set by the relative amounts of the two species. The chemical shift of the signal measured as a function of pH therefore produces the usual type of pH titration curve (see for example Chapter 2, Fig. 3). In principle, the measurement of internal pH should be possible by simply measuring the Pi

chemical shift

in vivo

and determining the pH from a standard titration

curve. However, first a number of potential problems must be considered. It is well known that a number of factors other than pH affect the chemical shift of Pi (12). Among these the most important are variations in

ionic strength (5) and metal ion binding (especially Mg2+ will be important

in vivo)

(13). These effects largely arise through their influence on the pK

2 of Pi (12). Furthermore the binding of Pi to intracellular components

(proteins, etc.) is another factor that could influence its chemical shift

in vivo.

However, a comparison of titration curves in simple aqueous solution and in dog heart homogenates demonstrated that this effect is of no

importance (14).

A number of different approaches have been followed to overcome the problems of Pi chemical shift calibration. When the ionic composition of the intracellular medium is approximately known, the calibration curve can best be constructed using a solution of that composition. For example, in yeast cells the pH of the cytoplasm and the vacuole have been determined from curves constructed in media made up to approximate the concentrations of the

major intracellular ionic components in yeasts (15,16, Chapters 9 and 10)

Of course, some uncertainty is bound to remain since the precise ionic environment within the cell is not known exactly and may vary for different compartments.

dissipate the pH gradient (6pH) across the cellular membrane by the action of protonophorous uncouplers. By the elimination of the 6pH through the action of the uncoupler, the internal pH can be set at will by manipulation of the pH of the extracellular medium. By measuring the chemical shift of internal

31

Pi from P NMR spectra taken at different external pH values, a calibration

curve is obtained. The possible contribution of a residual Donnan potential across the cell membrane to the observed chemical shifts is probably small at high ionic strength (17, Chapter 3). The above procedure is described in

Chapters 2 and 3 for the measurement of cytoplasmic pH in

Rps. sphaeroides

and offers the closest simulation of the intracellular situation in those cases where no data on the intracellular ionic composition are available.

Because of the uncertainties inherent to the methods of shift calibration described above, i t is important to establish the reliability of NMR

measurements of pH by independent methods. Chapter 3 describes the excellent

agreement between the internal pH in

Rps. sphaeroides

as determined by

31

P NMR and by flow-dialysis under similar conditions.

pH homeostasis has been studied by NMR methods in

Escherichia coli

(18)

( ) . . 31 1 d 1 h . 1

and yeast 16 . In this thesis P NMR was emp oye to eva uate t e interna

pH in both procaryotic (Chapters 2, 3, 6 and 7) and eukaryotic

microorganisms (Chapters 9 and 10) as well as in model systems (Chapters 4 and 8).

The absolute accuracy of measuring internal pH by NMR is approximately 0.1 pH unit. Changes in pH, however, can often be measured to better than 0.05 pH unit (Chapter 2).

Bioenergetics of intact cells by

31

P NMR

Among the parameters that can be measured by 31P NMR are the

concentrations of ATP, ADP, and P. while in addition the chemical shift of

i

the latter can be used to evaluate internal pH. Moreover, the chemical shift of the middle phosphate of ATP provides information on the intracellular free

2+

Mg levels (Chapter 6). The above parameters determine the energy status of

whole cells.

The intracellular phosphorylation potential, i.e. the free energy of ATP

hydrolysis, can be determined with 31P NMR (19,20, Chapter 6). This

parameter was used to determine the proton stoichiometry of ATP synthesis in respiring mitochondria (20). According to the chemiosmotic theory of

related to the electrochemical potential gradient of the proton (or proton motive forve, pmf) across the mitochondrial membrane. The electrical

component of the pmf was determined from the K+ distribution in the presence of valinomycin as measured by a K+-specific electrode while the pH gradient

contribution to the pmf was determined by 31P NMR (20).

Ugurbil et al. (22) have studied bioenergetic phenomena in wild type and

ATPase-deficient strains of

E.

coli.

Their results are in agreement with the

role of the membrane-bound ATPase as proposed by the chemiosmotic hypothesis (21).

Oxidative (22, Chapters 9 and 10), substrate level (Chapter 10) and

photophosphorylation (Chapter 6) can be studied by 31P NMR in the intact

cell. Effects of uncouplers and inhibitors on these processes can be monitored on the same preparation.

Another contribution of 31P NMR to discussions on bioenergetics has been

the unequivocal demonstration of the existence of a bulk-to-bulk pH gradient across energy transducing membranes. This is illustrated in Chapter 4 by the

light-induced pH gradient in chromatophores of

Rps. sphaeroides.

By measuring the distribution of Pi across the mitochondrial membrane in parallel with the pH gradient, Ogawa et al. (17) have confirmed that transport of Pi across this membrane takes place through electroneutral exchange of H

2Po4 and OH .

Compartmenlation

Metabolic compartmentation is an important phenomenon in cell physiology and regulation; i t appears to be a universal phenomenon in lower as well as higher eukaryotic cells.

In favourable cases 31P NMR can be used to distinguish different

compartments inside intact cells. The success of the approach greatly depends upon a number of prerequisites. Inorganic phosphate usually is the most

abundant phosphorus compound monitored by 31P NMR. Consequently, the

chemical shift of Pi is mostly used for the determination of internal pH. Its concentration in the compartment of interest should be high enough in order to determine its chemical shift accurately. Moreover, the fraction of the cellular volume occupied by this compartment must be large enough to contain such absolute levels of Pi to allow its registration by NMR. Also, a pH gradient must exist between the cytoplasm and the organelle. In practice,

In rat liver cells cytosolic and mitochondrial pH were determined (25). In yeast cells the chemical shifts of cytoplasmic and vacuolar Pi are sufficiently different to determine the pH in both compartments under a variety of different conditions (2,16, Chapters 9 and 10).

Since in yeast cells cytoplasmic and vacuolar Pi can be monitored, this allows the distribution of Pi between its major pools to be measured

(Chapters 9 and 10) . Since also polyphosphate is observed, the

in vivo

regulation of phosphorus metabolism in yeasts can be studied with 31P NMR

(Chapter 10) .

Often the NMR method can discriminate between intracellular and extracellular pools of metabolites. For example, transport of metabolites

across cell membranes has been studied (23). Also, i t can be useful to

distinguish internal and external metabolite pools by adding paramagnetic ions to broaden or shift the signal from the extracellular medium

(Chapter 9). Of course, this procedure can only be employed if the ions do not penetrate into the cells. Internal and external ATP and ADP can usually be discriminated on the basis of the significant shift of these nucleotides

. 2+ 2+ .

between their Mg -free (external) and Mg -bound (internal) forms (3,20,24).

31

P NMR chemical shifts of titratable phosphates not only give

information on internal pH but also on the uniformity of this pH over the ensemble of cells measured. This can be deduced from the line width of, in particular, internal inorganic phosphate. Thus, i t was established that

whole cells of

Rps. sphaeroides

maintain a very uniform cytoplasmic pH in

the light as manifested by the width of the intracellular Pi peak (Chapter

2). By contrast, chromatophores of

Rps. sphaeroides

develop a very broad

range of internal pH values upon illumination. This heterogeneity can be largely abolished by the K+-ion6phore valinomycin (Chapter 4).

A more extreme example of heterogeneity was encountered in the case of bacteriorhodopsin vesicles (Chapter 8) . As a function of the reconstitution procedure used, the fraction of vesicles containing the light-driven protonpump (i.e. bacteriorhodopsin) and the overall orientation of the protein could be determined from spectra taken during illumination.

Transport

Transport of metabolites across cell membranes can be studied if the pools of metabolites on both sides of the membrane can be distinguished from each

discriminated as a consequence of the difference in pH in both compartments. Hence, internal transport of Pi across the vacuolar membrane can be followed

by 31P NMR in yeasts (Chapter 9).

In suspensions of red blood cells, Campbell and coworkers have measured the transport of various metabolites across the cell membrane with the use

of 1H spin-echo NMR (20,23). Internal and external resonances were

distinguished through their difference in spin-spin relaxation time T

2 which

is significantly shorter for the extracellular metabolites (23). This corresponds to broader peaks for the external compounds. For example, after adding alanine to the medium, there was a gradual growth of signal which provided a direct measure of the rate of transport of alanine into the red blood cells.

Springer and coworkers (26) have measured by 23Na NMR the transport of

Na+ across the cell membrane of yeasts and erythrocytes. A shift reagent, the

23 +

dysprosium nitrilotriacetate ion, was used to distinguish between Na

inside and outside the cells.

Metabolic pathways

13

C NMR has proven to be a valuable method in the study of the biosynthesis and metabolism of specific compounds as carried out by

microorganisms (for reviews see Refs. 7,10,11). 13c NMR studies at natural

abundance have so far been limited to systems which are stable as a function of time and which contain high concentrations of the metabolites of interest. For example, the osmoregulation in a blue-green alga has been studied by this method (27). Shulman and coworkers have studied the regulation of trehalose breakdown during dormancy and the induction of germination in yeast

13

ascospores by natural abundance C NMR (28).

Most 13c NMR applications have dealt with the fate of specifically

. h d b . . f 13

enric e su strates. There are many reports describing the use o C NMR to

characterize the end-products of bacterial metabolism in cell-free extracts (29-32). This has allowed quantitative descriptions of the pathways involved

in substrate utilization such as the relative flow of 13c label from

13 13

1- C-glucose and 1- c-acetate through the Krebs cycle, the phosphogluconate

pathway, and the glyoxylate shunt into L-glutamate excreted by

Microbacterium

ammoniaphilum

(32).

13

Early work on 1- C-glucose consumption by anaerobic suspensions of the

yeast

Candida utilis

demonstrated the feasibility of 13

c

NMR studies

in

vivo

(33). Den Hollander et al. studied anaerobic glycolysis (34) and

acetate utilization (35) by the yeast

Saccharomyces cerevisiae

using

specifically labeled glucose and acetate, respectively. The latter studies demonstrated the existence of a futile cycle in which phosphoenolpyruvate, formed from oxalacetate, returns to the Krebs cycle through pyruvate and

acetyl-CoA (35) . Such futile cycles are very difficult to assess by 14

c

tracer methods. Scott and coworkers have made an important contribution to the knowledge of glucose utilization by bloodstream forms of the African

. . 13

trypanosomes by

&n V&VO

C NMR (36). By using both wild-type and

genetically altered strains of

E. coli,

Ogino et al. (37) demonstrated that

the flow of carbon through the biosynthetic pathway of aromatic amino acids is controlled by feedback inhibition in this organism.

Chapter 5 of this thesis deals with the kinetics and pathways of 2 -13 C-acetate meta olism in the p otosynt etic bacterium b . . h h . Rp

s. sp aero& es.

h "d I

n V&VO

. 13 C NMR studies were performed under a variety of different . physiological conditions. Similar studies on acetate utilization by

Chromatium vinosum

are described in Chapter 7 which deals with the combined

13 31

approach of C and P NMR to probe metabolic and energetic phenomena,

respectively. By a combination of NMR and other analytical methods for the

measurement of reserve materials, the metabolic activity of

Chr. vinosum

as

13 31

monitored by C and P NMR could be related to the physiological state of

the organism.

In vivo

enzyme kinetics

As described above whole cell NMR has been widely used to measure the kinetics of cellular processes under non-steady state conditions. However, NMR spectra are also sensitive to chemical exchange processes that occur under steady-state or equilibrium conditions. Of particular interest for

studies of enzyme-catalyzed reactions

in vivo

are the saturation transfer

and the isotope-exchange techniques.

The saturation transfer method was originally developed by Forsen and Hoffman (38). It involves the selective saturation of an NMR peak causing the reduction in intensity of resonance(s) coupled to the saturated resonance by chemical exchange. The method can only be successful, if the rate of the

spin-lattice relaxation times T

1 of the nuclei involved. Using this approach

Shulman and coworkers have measured by 31P NMR the kinetics of the enzyme

adenosinetriphosphatase (ATPase) in aerobic

E. coli

(39) and aerobic and

anaerobic

S. cerevisiae

cells (40) .

A method of more general applicability is that of isotope exchange

monitored by 1H NMR. For example, if red blood cells are suspended in 2H

2

o,

the 1H NMR signals from the C-3 protons of added lactate and pyruvate

gradually disappeared because the C-3 position was deuterated (41,42). The rate of interconversion of lactate and pyruvate could be measured from the observed exchange rates, and hence the activity of lactate dehydrogenase

could be determined

in vivo

(42). By comparing the

in vivo

activity of the

enzyme with that of the purified form, Campbell and coworkers determined the free (NAD+ + NADH) concentration in erythrocytes (43). The isotope-exchange method has also been employed to assess the metabolism of malate and fumarate in red blood cells (44).

Future prospects

The progress of whole cell NMR in the future will greatly depend on two factors: (i) the progress of NMR technology and methodology; (ii) the development of methods to maintain cells under physiologically optimal

conditions. The availability of wide bore magnets has allowed the utilization

20 31 13 . . . d h

of -mm tubes for P and C NMR. This has significantly increase t e

sensitivity of NMR detection, thus allowing more dilute cell suspensions to be used. Thus, Shulman and coworkers were able to analyze synchronous growth

of the yeast

S. cerevisiae

by 31P NMR (45).

A factor of 2.7 in signal-to-noise is gained if a solenoid coil is used

instead of the standard Helmholtz coil (2,46). This allowed 31P NMR studies

of dilute suspensions of anchorage-dependent mammalian cells (2,46). As can be seen in Table 1, the proton is the most sensitive nucleus for NMR detection. This has stimulated the development of methods to study whole

b . 1 . 1 . . 1 . 1 1 f

cell meta olism by H NMR. Since H

2

o

is the physio ogica so vent or most

cells, this poses as enormous dynamical range problem: compounds present at

the typical concentrations (~ 5 mM) found in living cells have to be detected

in the presence of 55 M water. The 1H spin-echo NMR method (23) overcomes

this problem. A simple two-pulse sequence 90°-T-180° produces an echo at time 2T after the 90° pulse. By choosing a suitable delay-time T the water

observed, especially at very high magnetic fields (47). 1H spin-echo NMR spectra of whole cells may be complicated since essentially all metabolites contain protons. However, with some effort most lines can be assigned (48).

In proton NMR, spectral overcrowding often severely limits the

possibilities for analysis. However, a simplification can be obtained by employing a recently developed method for multiplet selection in crowded 1

H NMR spectra (49). The method relies on the momentary creation of double quantum coherence between coupled protons. By a suitable choice of the transmitter frequency and delays between pulses, the procedure can be made to select specific multiplets, with concomitant suppression from the

singlets and other multiplets. Using this approach in the 1H NMR spectrum

of a yeast cell extract the CH

2 protons of citrate which are largely

obscured normally can be observed specifically (49). Although i t may prove to be of value in other systems, the method so far was unsuccessful when applied to intact yeast cells (P.J. Hore and K. Nicolay, unpublished experiments).

Shulman and coworkers (50) have greatly increased the resolution and

specificity of 1H NMR in studies of cellular metabolism by feeding

13

c-labeled substrate and observing 1H difference spectra in the presence and

absence of 13c decoupling fields. This amounts to the specific observation of

13

c-labeled metabolites with the sensitivity of 1H NMR. With 2-13c-acetate

as a substrate for an aerobic suspension of

S. cerevisiae,

the build-up of

labeled glutamate and aspartate was monitored by the use of this method (50).

III. Outline of This Thesis

This thesis describes the application of high resolution NMR to study the bioenergetics and metabolism of intact microorganisms and reconstituted bacteriorhodopsin vesicles. The major part (Chapters 2-8) is concerned with

31

light-dependent procaryotic systems while in Chapters 9 and 10 P NMR

studies of eukaryotic yeasts are described.

In Chapter 2 the use of 31P NMR to determine the light-induced

pH gradient (~pH) across the cell membrane of the purple non-sulfur

bacterium

Rhodopseudomonas sphaeroides

is introduced. In Chapter 3 the

independent methods show good agreement. pH Homeostasis in

Rps. sphaeroides

appears to be strongly influenced by the ionic composition of the

extracellular medium. 31

Chapter 4 treats P NMR measurements of the light-induced 6pH in

chromatophores of

Rps. sphaeroides.

These membrane-structures are derived

from the cytoplasmic membrane of the intact organism by French pressure cell treatment and have an inverted orientation as compared to the native

membrane. 13

_c

NMR measurements of acetate utilization by intact

Rps. sphaeroides

cells are described in Chapter 5. The destination of labeled acetate appears to be strongly dependent upon the physiological conditions of the cells.

Chapters 6 and 7 describe 31P and 13

c

NMR experiments on the phototrophic

sulfur bacterium

Chromatium vinosum.

Dark starvation is studied both at the

level of carbon metabolism and energy metabolism.

A combination of freeze fracture electron microscopy and 31P NMR has been

used to characterize bacteriorhodopsin vesicles (Chapter 8) . Various

reconstitution methods are compared with respect to their ability to reach vesicle preparations which are homogenous both in their protein orientation and in their protein distribution.

h 9 d 10 h · · 31 d' f 1

C apters an deal wit ~n v~vo P NMR stu ies o yeast eel s.

Cytoplasmic and vacuolar pH are monitored during aerobic and anaerobic

substrate utilization (Chapter 9). The regulation of phosphate metabolism as

. 31 . 'b d . 10

evidenced from P NMR is decri e in Chapter .

REFERENCES

1. Bottomley, P.A. (1982)

Rev. Sci. Instrum., 53,

1319-1337.

2. Gillies, R.J., Alger, J.R., Den Hollander, J.A., and Shulman, R.G. (1982)

in:

Intracellular pH: its measurement, regulation, and utilization in

cellular functions, pp.

79-104.

Alan. R. Liss, Inc., New York.

3. Ogawa, S., Rottenberg, H., Brown, T.R., Shulman, R.G., Castillo, C.L., and

Glynn, P. (1978)

Proc. Natl. Acad. Sci. USA, 75,

1796-1800.

4. Gadian, D.G. (1982)

Nuclear magnetic resonance and its applications to

living systems. Oxford University Press, Oxford.

463-535.

Academic Press, New York.

6. Ugurbil, K., Shulman, R.G., and Brown, T.R. (1979) in:

Biological

applications of magnetic resonance (Shulman, R.G., ed.), pp.

537-589.

Academic Press, New York.

7. Norton, R.S. (1980)

Bull. Magn. Res.,

3, 29-48.

8. Gadian, D.G., and Radda, G.K. (1981)

Annu. Rev. Biochem., 50,

69-83.

9. Roberts, J.K.M., and Jardetzky,

o.

(1981)

Biochim. Biophys. Acta, 639,

53-76.

10.Scott, A.I., and Baxter, R.L. (1981)

Annu. Rev. Biophys. Bioeng., 10,

151-174.

11.Matwiyoff, N.A., London, R.E., and Hutson, J.Y. (1982) in:

NMR

spectroscopy: new methods and applications (Levy, G.C., ed.), pp.

157-186.

ACS symposium series 191, American Chemical Society, Washington.

12.Roberts, J.K.M., Wade-Jardetzky, N., and Jardetzky, 0. (1981)

Biochemistry, 20,

5389-5394.

13.Jacobson, L., and Cohen, J.S. (1981)

Bioscience Reports, 2,

141-150.

14.Hollis, D.P. (1980) in:

Biological magnetic resonance, vol.2 (Berliner,

L.J., and Reuben, J., eds.), pp.

1-44.

Plenum Press, New York, London.

15.Gancedo, J.M., and Gancedo,

c.

(1973)

Biochimie, 55,

205-211.

16.Den Hollander, J.A., Ugurbil, K., Brown, T.R., and Shulman, R.G. (1981)

Biochemistry, 20,

5871-5880.

17.0gawa,

s.,

Boens, C.C., and Lee, T-M. (1981)

Arch. Biochem. Biophys., 210,

740-747.

18.Slonczewski, J.L., Rosen, B.P., Alger, J.R., and MacNab, R.M. (1981)

Proc.

Natl. Acad. Sci. USA, 78,

6271-6275.

19.Shen,

c.,

Boens, C.C., and Ogawa,

s.

(1980)

Biochem. Biophys. Res. Commun.

' 93,

243-249.

20.0gawa,

s.,

and Lee, T-M. (1982)

Biochemistry, 21,

4467-4473.

21.Mitchell, P. (1966)

Biol. Rev. Cambridge Philos. Soc., 41,

445-502.

22.Ugurbil, K., Rottenberg, H., Glynn, P., and Shulman, R.G. (1982)

Biochemistry, 21,

1068-1075.

23.Brindle, K.M., Brown, F.F., Campbell, I.D., Grathwohl, C. and Kuchel, P.W.

(1979)

Biochem. J., 180,

37-44.

24.Casey, R.P., Njus, D., Radda, G.K., and Sehr, P.A. (1977)

Biochemistry,

16,

972-977.

26.Balshi, J.A., Cirillo, V.P., and Springer, C.S. (1982)

Biophys. J., 38,

323-326.

27.Borowitzka, L.J., Dernrnerle, S., MacKay, M.A., and Norton, R.S. (1980)

Science, 210,

650-651.

28.Thevelein, J.M., Den Hollander, J.A., and Shulman, R.G. (1982)

Proc. Natl.

Acad. Sci. USA, 79,

3503-3507.

29.Runquist, E.A., Abbott, E.H., Armold, M.T., and Robbins, J.E. (1981)

Appl.

Environ. Microbial., 42,

556-559.

30.Paalme, T., Olivson, A., and Vilu, R. (1982)

Biochim. Biophys. Acta, 720,

303-310.

31.Paalme, T., Olivson, A., and Vilu, R. (1982)

Biochim. Biophys. Acta, 720,

311-319.

32.Walker, T.E., Han, C.H., Kollman, V.H., London, R.E., and Matwiyoff, N.A.

(1982) J.

Biol. Chem., 257,

1189-1195.

33.Eakin, R.T., Morgan, L.O., Gregg, C.T., and Matwiyoff, N.A. (1972)

FEES

Lett., 28,

259-264.

34.Den Hollander, J.A., Brown, T.R., Ugurbil, K., and Shulman, R.G. (1979)

Proc. Natl. Acad. Sci. USA, 76,

6096-6100.

35.Den Hollander, J.A., Behar, K.L., and Shulman, R.G. (1981)

Proc. Natl.

Acad. Sci. USA, 78,

2693-2697.

36.MacKenzie, N.E., Hall, J.E., Seed, J.R., and Scott, A.I. (1982)

Eur. J.

Biochem., 121,

657-661.

37.0gino, T., Garner, C., Markley, J.L., and Hermann, K.M. (1982)

Proc. Natl.

Acad. Sci. USA, 79,

5828-5832.

38.Forsen,

s.,

and Hoffman, R.A. (1963) J.

Chem. Phys., 39,

2892-2901.

39.Brown, T.R., Ugurbil, K., and Shulman, R.G. (1977)

Proc. Natl. Acad. Sci.

USA, 74,

5551-5553.

40.Alger, J.R., Den Hollander, J.A., and Shulman, R.G. (1982)

Biochemistry,

21,

2957-2963.

41.Brown, F.F., Campbell, I.D., Kuchel, P.W., and Rabenstein, D.C. (1977)

FEES Lett., 82,

12-16.

42.Simpson, R.J., Brindle, K.M., Brown, F.F., Campbell, I.D., and Foxall,

D.L. (1982)

Biochem.

J.,

202,

573-579.

43.Simpson, R.J., Brindle, K.M., Brown, F.F., Campbell, I.D., and Foxall,

D.L. (1982)

Biochem.

J.,

202,

581-587.

45.Gillies, R.J., Ugurbil, K., Den Hollander, J.A., and Shulman, R.G. (1981)

Proc. Natl. Acad. Sci. USA, 78,

2125-2129.

46.Ugurbil, K., Guernsey, D.L., Brown, T.R., Glynn, P., Tobkes, N., and

Edelman,

r.s.

(1981)

Proc. Natl. Acad. Sci. USA, 78,

4843-4847.

47.Rabenstein, D.L., and Isab, A.A. (1979) J.

Magn. Res., 36,

281-286.

48.Agris, P.F., and Campbell, I.D. (1982)

Science, 216,

1325-1327.

49.Hore, P.J., Zuiderweg, E.R.P., Nicolay, K., Dijkstra, K., and Kaptein, R.

(1982)

J.

Amer. Chem. Soc., 104,

4286-4288.

50.Sillerud, L.O., Alger, J.R., and Shulman, R.G. (1981)

J. Magn. Res., 45,

CHAPTER

II

31

P NUCLEAR MAGNETIC RESONANCE STUDIES OF ENERGY

TRANSDUCTION IN

RHODOPSEUDOMONAS SPHAEROIDES

K. Nicolay, R. Kaptein, K.J. Hellingwerf and W.N. Konings

Reprinted from:

31

_{P Nuclear Magnetic Resonance Studies}

of Energy Transduction in

Rhodopseudomonas sphaeroides

Klaa' NICOLAY. Robert KAPTEIN. Klaas J. HELLINGWERF, and Wt! N. KONINGS

Department of Physical Chemistry and Department of Microbiology, Univer;,1ly of Groningcn

(Received August 25 'December 18, 1980)

31 _{P nuclear magnetic resonance spectra of the phototrophic bacterium Rhodopseudomonas sphaeroides reveal}

the presence of inorganic phosphate, sugar phosphates and two non-identified P,P1

-diesterified pyrophosphate

compounds. Due to the presence of paramagnetic cations the resonances of these compounds can only be

detected after repeated washing of the bacterial cells with a buffer, containing EDTA plus excess Mgz+ Washing with Mg2 _{'-free EDTA buffer deteriorates the structural integrity of the membranes of Rps.}

sphae-roides. This is indicated by the appearance of an extra resonance peak in the spectra of these cells in a region

where the phospholipids absorb and by a fivefold increase in proton permeability of the cytoplasmic membrane of Rps. sphaeroides under these conditions. Upon illumination of the cell suspension in the NMR tube the generation of a transmembranc pH gradient can be inferred from the shift in the resonances of extracellular

and intracellular inorganic phosphate. Intracellular inorganic phosphate shows one homogeneous resonance peak upon illumination. This demonstrates that the mixing system, which has been developed for this application, functions efficientl;. The magnitude of' the light-dependent pH difference is 0.8 at an external pH 6.

The width at half height of the internal inorganic phosphate peak is essentially independent of internal pH from pH 5-8. remains unchanged upon addition of uncoupler and is inversely proportional to the number of EDTA washings applied. These observations indicate that the inorganic phosphate NMR peak width is predominantly determined by the presence of a residual amount of paramagnetic cations, rather than by a broad distribution of internal pH values over the cells.

Ionophores have an effect on the light-dependent pH-gradient in accordance with the chcmiosmotic theory: valinomycin increases, and carbonylcyanide p-triftuoromethoxyphenylhydrazone decreases. the magnitude of this gradient.

The number of' applications of NMR for the study of intact biological systems is rapidly increasing. Especially there are numerous reports of investigations of whole cells by

31_{P NMR. These include studies of phosphorus-containing} metaholitcs in suspensions of whole cells from mammalian

[1-7], microbial [8-12] and plant origins [13].

According to Mitchell's chemiosmotic hypothesis [14] a trammembranal electrochemical potential gradient for pro-tcms (or a protonmotive force) is the essential and obligatory intermediate in many membrane-bound energy-transducing processes. The protonmotive force is composed of a trans-membranal pH gradient (JpH) and an electrical potential (J rjJ). Several processes are driven or controlled by the proton-motive force, two of the most recently discovered examples being DNA transfer [15] and nitrogen fixation [16]. With 31_P

NMR it is possible to measure the intracellular pH [1,3,6, 8, 13] and the concentrations and distributions of phospho-rylated metabolites such as adenine nucleotides, sugar

phos-phates and inorganic phosphate in a non-invasive manner

[3,5- 7. 10. 11.13]. High-resolution 31_{P NMR studies have}

proven to be very useful in studying biocnergetic phenomena, since these studies can supply information about the internal

pH, the transmembranal pH gradient and the levels of phos-phorylated metabolites crucial to energy transduction (e.g. [7, 10]).

31 _{P NMR for the measurement of J pH has the advantage}

over other procedures such as ion distribution techniques

Ahhreviation. ppm, parts per million

f:'n:yme. Pyruvate kina~c (EC 2.7.1.40).

that it allows direct determinations of the absolute values of intracellular pH (pll;0 ) an<l extracellular pH (pH0) from the

chemical shift of internal and external inorganic phosphate. respectively, provided a calibration curve for the titration of P, in the appropriate medium can be obtained. Another

ad-vantage of Jip NMR is that the kinetics of internal and

ex-ternal~ pH changes are directly manifested by shifts in NMR

peak positions and no special exogenous probes are required

to monitor ,1pH.

Until now, JpH measurements in bacterial systems have been confined mainly to isolated membrane preparations (e.g. [17]). A limited number of JpH measurements in whole cells of bacterial species have been reported including aerobically and anaerobically grown Escherichia coli [10, 11, 18, 19],

re-spiring Staphylococcus aureus [19] and Bacillus subtilus [20,

21] and illuminated Halobacterium ha!ohium [22,23].

In this report results are presented of a 31_{P NMR study}

of the phototrophic bacterium Rhodopseudomonas .1phaeroides.

h this metabolically versatile organism a protonmotivc force can be generated by a respiratory chain or by light-induced cyclic electron transfer. This paper is the first report of 31 _P

NMR studies in a light-dependent intact biological system. Attention will be focussed mainly on JpH which develops during light-energisation.

MATERIALS AND METHODS

Cells

Rhodopseudomonas sphaeroides, .... train 2.4.1, was grov,..n

b) Sistrom [24] with omission of the paramagnetic cations Mn2

-and Co2

+_ at 30 C, as described previously [25 J. Cells

were grO\vn under high light intensity by illuminating the cultures with 8 x 150-W incandescent bulbs 10 cm from the culture bath.

Cells were harvested at the end of the logarithmic phase of growth and usually washed five times at room temperature in a medium containing 100 mM KCL 2 mM K-FDTA. 20 mM 4-morpholineethanesulphonic acid, 20 mM glycyl-glycine, 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesul-phonic acid pH 7.0, supplemented with 10 mg I chlorampheni-col and (unless otherwise stated) 5 mM MgS04. Finally the cells were washed and resuspended with medium supple-mented with 10",, 2_H

20. The cell density was typically

30-50 mg of protein/ml.

Jl PNM R A1easurements

145.7 MHz 31

P NMR measurements were performed at 22 C on a Bruker HX-360 spectrometer operating in the Fourier transform mode and equipped with quadrature

detec-tion. Experiments were carried out \vithout locking the

instrument and without proton decoupling. Spectra were accumulated in 4096 time domain addresses with a recycle time of 0.46 s and a spectral width of 4400 Hz. 90 radio-frequency pulses (pulse v.idth about 30 µs) were employed. NM R peak positions in spectra are given in parts per million (ppm) from the peak position of 85 ",, orthophosphoric acid

as an external reference, except for some experiments in which

the peak of exogenous glycerophosphoryl choline at - 0.49 ppm was used as a marker. Flat-bottomed 10-mm tubes were used (inner diameter 9 mm), with a sample volume of 1.5 ml.

Light Irradiation and Mixing System

Homogeneous illumination of the cell suspension was ensured by employing a mixing system (Fig. 1) consisting of a glass rod with propeller which, inserted into the suspension was kept in a fixed position by means of silicone tubing Rotation of the sample tube at a rate of ca. 30 Hz by the

normal air-driven spinner resulted in a homogeneou~ mixing

of the suspension (see Results). The experimental arrange-ment for light irradiation in the NMR probe has been de-scribed elsewhere [26]. A Spectra Physics model 171 argon ion laser with power mainly at 488 and 512 nm was used as light source. lllumination with optimal light intensity (350 mW) caused an increase in the temperature of the sample of about 2 C.

ATP-A DP /Jetenninations

ATP was determined with the luciferin-luciferase method [27]. ADP was converted to ATP by pyruvatc kinase (8 ~tg/ml)

in the presence of 5 mM sodium phosphoeno/p;ruvate. After

30 min at room temperature the samples were fixed with

ice-cold perchloric acid and subsequently neutralized with KOH. ADP was determined from the difference between ATP plus ADP and ATP.

Protein

Protein was determined according to Lowry et al. [28].

glass cod 'silicone

~i'"'

pf lug

lJ

air

~

_T

teflon plug 10mm I cell ~suspension

L__J

1

Fig.1. Schematic drawing of the stirrmg .1.r.1tem. Both the NMR !Ube and the bacterial suspension are capped \Vith tefton plug~ through which a glass rod with propeller runs. The rod is fixed by a flexible ~11icone connection which is fastened oubide the magnet. \11xing of the cell suspension is effected b) spinning the NMR tube a:-. de:-.cribed in Materials and Methods

Materials

3,5-Di-tert-butyl-4-hydroxybenzylidine malononitrite was a gift from Dr Y. Nishizawa (Sumitomo Chemical Industry, Osaka, Japan). All other materials were reagent grade and

obtained from commercial sources.

RESULTS

-''P NMR Spectra

Fig. 2 shows 145.7 MHz 31

P NMR spectra of

Rhodo-pseudomonas .\phaeroide.s cells washed in the presence

(spec-trum I) and in the absence (spec(spec-trum II) of magnesium sul-phate. Spectrum II was taken during steady-state illumination. Internal and external P, (peak B) are shown in spectrum I as one peak while in spectrum II a main peak with a shoulder on the high field side is observed. It will be shown below that illumination of the suspension results in a splitting of the P1

signal (peak B) into two peaks representing intracellular phosphate on the low field and external phosphate on the high field side.

An indication of the origin of the remaining peaks in spectrum I can be given on the basis of chemical shifts. Peak A fall in the sugar phosphate region but any assignments must await titration of cell extracts. Peak Cat 10.6 ppm and peak D at 12.3 ppm are in the P,P'-diesterified pyrophosphate region which can include contributions from NAD+. NADH and ADP-glucose [4, 9, 29]. the glycosyl donor for glycogen bio-synthesis in prokaryotes [30].

The peak indicated X in spectrum II at about 0 ppm has not yet been identified. In the short repetition time (0.46 s) employed, peak X was observed only in spectra of cells washed with EDT A-containing medium withouth Mg2

+ and dis-appeared completely after addition of excess MgS04 (data