of WIUU - weizmann.ac.il · E.R. ANDREW LAWRENCE BERLINER ROBERT BLINC University of Florida Ohio...

WIUU '.Irof

T . f r- ' •/ i

Christo -Wrapped Reichstag, Berlin 1995

BULLETIN OF MAGNETIC RESONANCEThe Quarterly Review Journal of the

International Society of Magnetic Resonance

Editor:

DAVID G. GORENSTEIN

Sealy Center for Structural BiologyThe University of Texas Medical BranchGalveston, Texas 77555-1157 U.S.A.

Phone: 409-747-6800Fax: 409-747-6850

INTERNET: [email protected]

Editorial Board:

E.R. ANDREW LAWRENCE BERLINER ROBERT BLINCUniversity of Florida Ohio State University University of LjubljanaGainesville, Florida, U.S.A. Columbus, Ohio, U.S.A. Ljubljana, Slovenia

H. CHIHARA GARETH R. EATON DANIEL FIATOsaka University ' University of Denver University of Illinois at ChicagoToyonaka, Japan Denver, Colorado, U.S.A. Chicago, Illinois, U.S.A.

SHIZUO FUJIWARA DAVID GRANT ALEXANDER PINESUniversity of Tokyo University of Utah University of CaliforniaBunkyo-Ku, Tokyo, Japan Salt Lake City, Utah, U.S.A. Berkeley, California, U.S.A.

M. MIK PINTAR CHARLES P. POOLE, JR. BRIAN SYKESUniversity of Waterloo University of South Carolina University of AlbertaWaterloo, Ontario, Canada Columbia, South Carolina, U.S.A. Edmonton, Alberta, Canada

The Bulletin of Magnetic Resonance is a quarterly review journal by the International Society ofMagnetic Resonance. Reviews cover all parts of the broad field of magnetic resonance, viz.. thetheory and practice of nuclear magnetic resonance, electron paramagnetic resonance, and nuclearquadrupole resonance spectroscopy including applications in physics, chemistry, biology, andmedicine. The BULLETIN also acts as a house journal for the International Society of MagneticResonance.

CODEN: BUMRDT ISSN: 0163-559X

Bulletin of Magnetic Resonance, The Quarterly Journal of International Society of MagneticResonance. 1999 copyright by the International Society of Magnetic Resonance. Rates: Librariesand non-ISMAR members $80.00, members of ISMAR, $25.00. All subscriptions are for a volumeyear. All rights reserved. No part of this joumai may be reproduced in any form for any purpose or byany means, abstracted, or entered into any data base, electronic or otherwise, without specificpermission in writing from the publisher.

Council of the International Society of Magnetic Resonance

President: M. GOLDMAN, France

Vice-President: J. Waugh, U.S.A.

Founding Chairman: D. FIAT, U.S.A.

Secretary-General: P. SERVOZ-GAVIN, France

Treasurer: R.R. VOLD, U.S.A.

Past President: A. PINES, U.S.A.

J. ANGLISTERIsrael

G. BODENHAUSENSwitzerland

R DESLAURIERSCanada

E.L. HAHNU.S.A.

H-J. JAKOBSENDenmark

C.L. KHETRAPALIndia

H.C. PANEPUCCIBrazil

M. PUNKKINENFinland

V. SKLENARCzechoslovakia

H.W. SPIESSGermany

J.S. WAUGHU.S.A.

K. WUTHRICHSwitzerland

E.D. BECKERU.S.A.

P.T. CALLAGHANNew Zealand

S. FORSENSweden

R.K. HARRISEngland

RKAPTEINThe Netherlands

B. MARAVIGLIAItaly

V. PETROSIANC.I.S.

' J. REISSEBelgium

C.P. SLICHTERU.S.A.

H.STERKAustria

A.B.WIECKOWSKIPoland

C.S. YANNONIU.S.A.

M.R. BENDALLAustralia

S. CLOUGHEngland

C.FYFECanada

M.J.R. HOCHS. Africa

D.KELLYAustralia

M. MEHRINGGermany

M.M. PINTARCanada

G.C.K. ROBERTSEngland

P. SOHARHungary

T.TERAOJapan

X-W.WUChina

The aims of the International Society of Magnetic Resonance are to advance and diffuse knowledgeof magnetic resonance and its applications in physics, chemistry, biology, and medicine, and toencourage and develop international contacts between scientists.

The Society sponsors international meetings and schools in magnetic resonance and its applicationsand publishes the quarterly review journal. The Bulletin of Magnetic Resonance, the house journal ofISMAR.

The annual fee for ISMAR membership is $20 plus $25 for a member subscription to the Bulletin ofMagnetic Resonance.

Send subscription to: International Society of Magnetic ResonanceProfessor Regitze R. Void, TreasurerDepartment of Chemistry & BiochemistryUniversity of California, San Diego9500 Gilman DriveLa Jolla.CA 92093-0359(619) 534-0200; FAX (619) 534-6174e-mail: [email protected]

Vol. 20, 1/4

Joint 29th AMPERE - 13th ISMAR ConferenceAugust 2-7,1998, Berlin, Germany

Organizing CommitteeD. Ziessow, Chairman

W. Lubitz, Co-ChairmanF. Lendzian, Secretary

D. Haberland, Local Organization

D. Ziessow, (Berlin)W. Lubitz, (Berlin)H. H. Limbach, (Berlin)K. Mobius, (Berlin)H. Oschkinat, (Berlin)D. Stehlik, (Berlin)B. Bliimich, (Aachen)

Program Committee

K. P. Dinse, (Darmstadt)C. Griesinger, (Frankfurt)H. Gunther, (Siegen)A. Haase, (Wurzburg)U. Haeberlen, (Heidelberg)H. Kessler, (Miinchen)R. Kimmich, (Ulm)

G. Kothe, (Freiburg)M. Mehring, (Stuttgart)D. Michel, (Leipzig)H. Riiterjans, (Frankfurt)J.-M. Spaeth, (Paderborn)H. W. Spiess, (Mainz)D. Blechert, (Leipzig)

In consultation withM. Goldman (France) President ISMAR

P. Servoz-Gavin (France) Secretary General ISMARB. Maraviglia (Italy) President Groupement AMPERE

R. ICind (Switzerland) Secretary General Bureau AMPERE

International Advisory BoardAll members of the ISMAR Council and Committee AMPERE

CO-EDITORS OF THE PROCEEDINGS

D.G. GorensteinDepartment of HBC&G

Sealy Center for Structural BiologyGalveston,TX 77555-1157

D. ZiessowIwan-N.-Stranski-InstitutBerlin, Germany 10623

Bulletin of Magnetic Resonance

ACKNOWLEDGMENTS

The organizers are grateful to the following agencies, institutions, and companies for financialsupport:support:

Deutsche Forschungsgemeinschaft,Deutsche Physikalische Gesellschaft/Heraeus-Stiftung,Gesellschaft Deutscher Chemiker,Deutscher Akademischer Austauschdienst,Technische Universitat Berlin,Freunde des Weizmann-Instituts,International Society of Magnetic Resonance,Department of Energy (USA),NMR Concepts (USA),Bruker Analytik GmbH,Schering AG Berlin,Beiersdorf AG,Martek Biosciences Corp. (USA),Jeol GmbH Germany, andDory Scientific Inc. (USA)

Vol. 20, No. 1/4

Regitze R. Void: 1937-1999

This issue of the Bulletin of Magnetic Resonance is dedicated

to the memory of Gitte.


Regitze R. Void: 1937 -1999

Gitte Void died in an accident in Cambridge,England on April 11. The news, spread rapidly by e-mail, stunned and saddened the worldwide NMRcommunity, who admired Gitte for her manyscientific accomplishments and adored her for herpersonal warmth and charm. I was asked to preparethis personal reflection for the Newsletter.

Gitte Shoup showed up in my lab at the NationalInstitutes of Health in 1965, eager to work withNMR, a technique that she had become familiarwith in graduate school in Denmark, and one that shecould see as a fascinating area for her research. Theimagination, resourcefulness and research drive thatcharacterized her later career were immediatelyapparent. She spent over five years in a jointresearch program at NIH and at the National Bureauof Standards, where she learned from Tom Farrar agreat deal about then new Fourier transformmethods. She made important discoveries on C-13 relaxation processes, nucleoside hydrogenbonding and internal rotation in amides, while alsodeveloping improved methods for studying C-13NMR.

In 1971 Gitte went to San Diego to marry and worktogether with Bob Void. For many years their labwas an outstanding center for developing newNMR methods and for investigations of dynamicsand structure of molecules in isotropic liquids, liquidcrystals and solids. During recent years,Gitte concentrated on studies of membrane-boundpeptides. She and her coworkers wereespecially successful in investigating magneticallyaligned phospholipid bicelles and laying thegroundwork for the widespread use of these systemsin structural NMR studies.Gitte served ISMAR very well as Treasurer for nineyears, not only handling fiscal matters but also

serving as a sparkplug for stimulatingactivities within the Society. She was a long-termmember of the Board of Trustees of the ENC. Mostpeople find the chairmanship of one ENC a task theywouldn't repeat; Gitte is the only person to agree toorganize two ENC's! She was also to chair theforthcoming Gordon Conference on MagneticResonance.

Gitte was an excellent scientist, a conscientiousmentor, an outgoing and dedicated spokesperson forNMR, and a close friend to many of us in theNMR community. Her vibrant personality made heran unforgettable figure at many conferences. Wemiss her.

Ted BeckerNational Institutes of Health

Vol. 20, No. 1/4

Thanks from the Editor

David G. Gorenstein

The Bulletin of Magnetic Resonance has served asthe house journal for the International Society ofMagnetic Resonance for 20 years. It has also servedas a quarterly review journal dealing with all aspectsof magnetic resonance. After 17 years as Editor ofthe Bulletin, I believe it is time for new directionsfor the journal and ISMAR, and I will be steppingdown with this issue. Over the years, the Bulletinhas published numerous ISMAR Proceedings issues,Waterloo Summer School issues as well as manysuperb review articles. Some of my fondest issuesincluded the Felix Bloch Memorial issue, the RichardErnst Nobel Prize issue and issues celebrating the50th anniversaries of NMR and EPR.

I want to express my gratitude to past officers ofISMAR, members of the Editorial Board and co-editors for help with these issues and in particularDan Fiat who has been so supportive of the Bulletinand ISMAR. I also want to thank the staff over theyears in Illinois, Indiana and Texas, includingPatricia Campbell, Donna Beiswanger, Jacky Luxonand Debbie and Jennifer Gorenstein.

My hope is that the Bulletin and the new ISMARNewsletter will increasingly provide a newelectronic connection to its readers and ISMARmembers via the WEB. Lyndon Emsley hasestablished a new WEB site (http://www.ens-lyon.fr/ISMAR/) for these purposes, and I hopereaders take advantage of this new medium.

Clearly as we move into the 21st century, the field ofmagnetic continues as vibrant a field as it has everbeen. It has been an exciting adventure, and I haveappreciated the opportunity to help communicateour field over these years.

Finally, I join with Ted Becker and the manyMends and colleagues of Gitte Void, in expressingmy deep regret at her death last month. She workedtirelessly as treasurer for ISMAR and the Bulletinfor the past nine years, always supporting ourefforts to be responsive to the magnetic resonancecommunity. She was a dear friend to many of usand we will sadly miss her bright spirit.


Proceedings of the Joint 29th AMPERE and 13th ISMAR Conference

Part II

Contents

Recording Heteronuclear Quantitative J-Correlation Spectra with Internal Reference Peaks, F. Lohr, C. Perez,J.M. Schmidt and H. Riiterjans 9

Self-Consistent Relaxation-Rate Analysis applied to Fatty-Acid Binding Protein, J.M. Schmidt, C. LudwigandH. Ruterjans 15

C-F Bonding in Fluorine-Graphite Intercalation Compounds and Graphite Fluorides: 13C and 19F NMR Study,A.M. Panich 20

A 2H-NMR Relaxation Study of Surfactant Order In Surface Aggregates and Bulk Structures, M. Schonhoff,S. Soderman, Z.X. Li and R.K. Thomas 25

Anomalies of Inhomogeneously Broadened Resonance Line Shape Due to Nonlinear Effects Contribution,M.D. Glinchuk and S.N. Nokhrin 30

Structural Investigations on the Activation of Plasminogen by Staphylokinase, O. Ohlenschlager, R.Ramachandran, K.-H. Giihrs, B. Schlott and L.R. Brown 35

EPR of Incommensurate Phases in Ferroelastic MgSiF6*6H2O:Mn2+: Lineshare Simulation with Variation ofthe Spin-Lattice Relaxation Rate, P.G. Skrylnik and A.M. Ziatdinov 39

NMR Studies of the Halotolerant Bacterium Holomonas Israelensis: Sodium-23, Cesium-133 andPhosphorus-31 studies,. H. Gilboa and A. Sakhnini 43

Fluorine Multiquantuni Coherences as Indicators of the Line-width Transition in CF3COOAg, A. Kaikkonen,E.E. YlinenaandM. Punkkinen 47

Calendar of Forthcoming Conference 50

1999 ISMAR Membership Address List 52

Vol. 20, No. 1-4

Recording Heteronudear Quantitative/-CorrelationSpectra with Internal Reference Peaks

F. Lohr, C. Perez, J.M. Schmidt, and H. RuterjansInstitut fur Biophysikalische Chemie, J.W. Goethe-Universitat Frankfurt

Biozentrum N230, Marie Curie-Str. 9, D-60439 Frankfurt am Main, Germany

ure

Quantitative /-correlation pulse sequences are modifiedto allow for determination of heteronudear 3 / couplingconstants from intensity ratios between cross peaks andauto peaks within a single spectrum. A new phase cyclingscheme gives rise to internal reference signals by retainingthe magnetization which is not transferred between the twocoupled nuclei of interest The new method is used to meas-

H>), 3/(N,HaM), 3/(C',Hp), 3/(C\Ha

M), 3/(C',HN),3 / (C',O, 3/(C',C»w), 3ma,V), and 3/(H»,C5)

coupling constants in a small I3C/15N enriched flavoprotein.

Introduction

Quantitative /-correlation experiments are well-establishedtools for the determination of vicinal coupling constants inisotopically labeled proteins [1]. They have been successfullyapplied to measure homonuclear /-couplings such as 3/(HN,Ha)[2], V(Ha,Hp) [3], or 3 / (C\C) [4]. The two nuclei are corre-lated in an 'out-and-back' manner yielding in-phase diagonaland cross peaks the intensity ratio of which depends on themagnitude of the active /-coupling. Heteronuclear couplingconstants have been obtained by comparing signal intensitiesin two spectra recorded with and without evolution of therelevant (passive) coupling in so-called spin-echo differenceexperiments [5]. This method, however, fails in the presenceof scalar interactions with more than one nucleus of a species.Alternatively, the chemical shifts of the different nuclei havebeen spread into an additional (third) spectrum dimensionwhile actively (rather than passively) evolving the desired/-coupling the value of which is finally encoded in the crosspeak intensity [6,7]. Magnetization components being nottransferred are cancelled by appropriate phase cycling leadingto 3D spectra void of 'auto' peaks, a fact which requires aseparate two-dimensional reference spectrum be recorded toselect for the said components [8]. Hence, the intensities ofcross peaks and auto peaks are measured in two spectra ofdifferent dimensionality and taking the volume integrals istherefore susceptible to errors. Otherwise, if peak heights areevaluated, the different number of time domain data points andacquisition times as well as the apodization function and theeffect of transverse relaxation in the third dimension have to beaccounted for correctly. Moreover, difficulties can aris% due tochanges of experimental conditions between the two data sets.

Recently, an elegant method has been described, permittingthe measurement of long-range ''N-'H coupling constants in asingle 3D experiment [9]. It relies on the rather uniform

'/(N,HN) interaction to provide internal reference peaks at theexpense of a moderate cross peak attenuation. Here wepropose a different strategy which does not require thepresence of one-bond and three-bond couplings to nuclei of thesame species and is therefore more generally applicable. Thecritical modification made to the original pulse sequences onlyinvolves the phase cycling scheme.

Results and Discussion

Apart from the homonuclear 3/(H°,Hp) coupling there arefive heteronuclear vicinal interactions that depend on the sidechain torsion angle x1 in amino acid spin systems (Fig. 1) viaKarplus-type relations ^/(C'.C1) is considered heteronuclearhere since carbonyl and aliphatic carbons are readily distin-guished by semi-selective pulses). Determination of 3/(N,Hp),3/(C',Hp), ^(N.C1), 3/(C',C0, and V(Ha,C) coupling constantswas accomplished by quantitative /-correlation experiments asdescribed previously [1,10-13]. Each of these experiments ismodified here to produce cross peaks as well as auto peaks in asingle spectrum. The evaluation is then based on intensityratios between cross and auto peaks of the same 3D spectrumand is less prone to experimental error.

The underlying principle shall be described for the HNHBexperiment [6] depicted in Fig. 2A. After the initial INEPTsequence [17] the density operator at point a is given by aa =-2HNzNy. During the delay 8 which is adjusted to an odd

'NC*

X=H, C, OH, SH

Figure 1. Spin coupling topology of an amino acid residuerelevant for the xl torsion angle. Solid arrows indicate thevicinal couplings investigated in this study.

10 Bulletin of Magnetic Resonance

multiple of (2I/NH)'1, antiphase I5N coherence becomes in-phase with respect to the amide proton directly attached to thenitrogen but dephases with respect to long-range coupledprotons. Simultaneously, 15N chemical shifts evolve as a func-tion of ti in a constant time fashion [18] but are disregarded inthe following. At time point b the relevant operators are

Ob = N x cos(ji/NHi5) Ilj COS(7C7NHJ5)

+2N y H' z sin(jc/NHiS) FIj COS(TL/NHJ5),

where Hj denotes a proton which has a vicinal coupling to thenitrogen and U are further protons linked via 2J or 37. Thefollowing lH 90° pulse converts the second term into heteronu-clear multiple-quantum coherence, leaving the first termunchanged, i.e.

<Jc = N x

-2NyH'y sin(7t/NHi5) Oj COS(JL/NHJ5).

In t2 the second term becomes modulated with the protonchemical shifts and the discrimination of positive and negative

frequencies is accomplished by recording two FIDs for each t2

value with the phase fa of the preceding proton pulse alternat-ing between x and y [19]. Usually the first term is cancelled bya phase inversion of one of the proton 90° pulses flanking t2along with the receiver. Here this two step phase cycle isapplied only to the. imaginary part of the t2 increments (fa = y)but not to the real part (fa = x). As a result, the fraction of thenitrogen magnetization that has not been transferred to anylong-range coupled proton gives rise to a signal of constantamplitude (apart from a slight decay due to transverse 15Nrelaxation) in the real part and of zero amplitude in the imagi-nary part of each t2 increment. After complex Fourier transfor-mation these signals appear symmetric about the F2 carrierposition at zero frequency (i.e. as axial peaks) with oppositephase with respect to cross peaks. The procedure is somewhatrelated to the quantitiative HMBC, where a separate 'pseudo2D' reference spectrum was recorded in a similar way [20].

15

H I I I ; • M2

N • i 4 i « 4 l iG,

i • i • i • ^1'I f | T|TgT'|-t'[T'^9l t.

G,xa b A cI I I !

G2G3 G3G4 G+ Gg

G

Figure 2. Improved pulse sequences of the HNHB (A) and the HN(CO)HB (B) experiments. Narrow and wide bars denote 90°and 180° pulses, respectively. Proton decoupling in B is achieved with a 4.2 kHz DIPSI-2 [14] sequence while nitrogens aredecoupled by a 0.75 kHz GARP [15] modulation. The proton carrier is placed on the water resonance throughout sequence A butis temporarily switched to the center of the amide region in B. Sine-shaped carbonyl pulses, applied at 176 ppm, have a durationof 124 us. Pulses on aliphatic carbons are shifted to 46 ppm by phase modulation [16] with their length adjusted to avoid excita-tion in the 13CO region. The delays x, x', T" , TJ, and e = TN are adjusted to 2.3, 2.5, 0.7, 5.4, and 28 ms, respectively. \ is a shortdelay to compensate for the duration of the l5N and 13C 180° pulses in the first t2 increment. The 5 delays are set to 48.6 ms in Aand to 26 ms in B. Quadrature in F2 is achieved by changing the phase fa in the States manner. Different phase cycles were usedfor the hatched 90° pulses (x or <j>5) and the receiver (fa or fy), depending on whether the real (fa = x) or the imaginary (<t>4 = y)part is recorded. Phase cycling: (A) fa = y,-y, fa = 2(x),2(-x), fa = 4(x),4(y), <t>5 = 8(x),8(-x), fa = y,fa = R,-R, 4>i = R,2(-R),R and(B) fa = y,-y, <|)2 = 2(x),2(-x), <t>3 = 4(x),4(-x), <t>5 = 8(x),8(-x), fa = x,fa = y, fa = R,-R, ft = R,2(-R), R, where R = x,2(-x),x. Allgradients are sine-bell shaped and have the following durations and strengths: (A) Gi = 1.5 ms, 10 G/cm; G2 = 1 ms, -39.4 G/cm;G3 = 0.5 ms, 5 G/cm; G4 = 0.5 ms, 4 G/cm; Gs = 0.5 ms, 8 G/cm (all along z); (B) G, = 1 ms, 10 G/cm (y); G2 = 1 ms, -39.4 G/cm(z); G3 = 0.5 ms, 4 G/cm (x) 5 G/cm (y); G4 = 0.5 ms, 5 G/cm (x) 4 G/cm (y); Gs = 0.5 ms, 8 G/cm (z) (directions are indicated).For each ti increment N- and P-type signals are recorded alternately by changing the sign of G2 along with phases fa (A) or fa(B). Axial peaks in F l are shifted to the edge of the spectrum by inverting fa (A) / fa (B) and the receiver for each ti value.

Vol. 20, No. 1-4 11

3D heteronuclear/-correlationF3<

/i

/

'

•

y

2D reference spectrumF24

3D with preservation of axial peaks

/ .

•

A

c : 1lliilillliitilBy

F2

Q2=0

Figure 3. Schematic representation of quantitative./-correlation spectra with the modification proposed here(bottom) and their relation to the conventional approach (top).

In the subsequent second 8 period the events of the firstone are reversed, yielding '

Oi = -2HN,Ny COS2(jrJNHi8) rij COS2(7L/NHJ8)

+2HNJSTy cos(QHit2) sin2(7t/NHi8) 1 } COS2(TL/N H J8).

Finally, the antiphase nitrogen magnetization is converted intoobservable proton magnetization using a gradient selected,sensitivity enhanced reverse INEPT step [21,22]. The sameprocedure was applied to the HN(CO)HB [7] pulse sequenceshown in Fig. 2B. Insertion of a relay step via carbonylcarbons leads to a dependence of signal intensities on theevolution of 13C'-'H couplings during the delays 8.

The three-dimensional spectra thus obtained contain thesame cross peaks as the original quantitative /-correlationexperiments but, as illustrated in Fig. 3, the recording oftwo-dimensional reference spectra is no longer necessary, asthe corresponding peaks are already present in the F1.F3 planeat the center of the F2 axis. Coupling constants can be calcu-lated from the ratio of cross peak and axial peak volumeswhich equals -tan2(jrJ8), where / is the desired three-bondcoupling constant and 8 is the delay in the pulse sequences forits evolution. Note that integrated signal intensities rather thanpeak heights have to be measured due to the differentlinewidths of cross peaks and reference peaks along the F2dimension.

The application of the modified HNHB and HN(CO)HBexperiments to Desulfovibrio vulgaris flavodoxin, a bacterialredox protein consisting of 147 amino acids and a non-covalently bound flavin mononucleotide cofactor, is demon-strated in Fig. 4. The strips along F2 contain several crosspeaks involving protons scalar coupled to the 13N and 13C

spins, respectively, of a particular residue and an auto peak atthe 'H carrier frequency providing the desired reference inten-sity. The coupling constants 37(N,Hfl) and ' / (CEP) are impor-tant to the stereospecific assignment of the P-protons and tothe determination of %' dihedral angles. Additional signals inthe HNHB spectrum are due to intraresidual 2J(N,Wi) andinterresidual VCNJTVi) couplings, while additional signals inthe HN(CO)HB spectrum are due to intraresidual V(C',Haj)and V(C',HNi) and interresidual 37(C',Ha

itl) and 2/(C',HNi+I)

couplings. The three-bond couplings are relevant to the \|f and(j) torsion conformation.

N14

N14H"

N14H"

IJ(N.H''')=2.4Hz

"J(N,H")=4.6Hz

E25,172

'J(N,H')=1.8Hz

'J(N,H-)=0.9Hz

E25H" »

E2SH"

I72H-e

"J(N.HPI)=2.1Hz

"J(N.H;,)=I.7HZ

W60

W60H"©

W60H"O

W60H"

3J(N,H")=2.0Hz

"J(N,H")=3.6H2

T81

T81H'

T81H"o

V(N,H')=3.1 Hz

B

9.9 9.7L5

7.3 7.1

F71

8.7 8.5

L78* - I6 H"

L5H"

L S H " -

I 6 H - *

'J(C\HB)=6.5Hz"J(Cf.H")=2.4Hz"J(C',H;,)=2.6HZ

*F71 H"

F71H"

_I72H"

'J(C,H")=5.5Hz"J(Cf.H»)=3.9Hz

L78H"^

E79H"

L78H-

L78H"

'J(C-.H»)=3.2Hz'J(C.H»)=3.4Hz•j(C.H")=3.0Hz

7.4 7.2

D127

1J(C',H")=6.SHz<J(C1,H»)=3.4Hz

D127H"

D127 H"

G128HT3

G128W

D127H"

G128H"

1.0

2.0

3.0

4.0

5.0

6.0

1.0

2.0

EQ.Q.

O4

EQ.Q.

CM

3.0 X U-

4.0

5.0

6.0

7.0

8.0

9.2 9.0 7.2 7.0 10.3 10.1 8.3 8.1

F3 [5,H» ppm]

Figure 4. Sections from F2/F3 slices of HNHB (A) and HN(CO)HB (B) spectra of flavodoxin. The slices are taken at theI5N (Fl) chemical shifts of the residues indicated at the top (A)or the respective sequentially following residues (B). Positiveand negative intensities are respresented by solid and brokencontours, respectively. The level spacing factor is 21/2and 2 forpositive (cross peaks) and negative levels (axial peaks),respectively. Peaks labeled with an asterisk are aliased in F2.


Pulse schemes for the determination of coupling constantsbetween backbone 1SN or 13C nuclei and aliphatic y-carbonsare depicted in Fig. 5. They include a magnetization transferbetween nitrogens and y-carbons of the same residue (HNCG)[10,12] or of the preceding residue in the carbonyl-relayedHN(CO)CG [11,12]. The new phase cycling scheme applies tothe 90° pulses on aliphatic 13C spins, again retaining axialpeaks as internal reference. Care has been taken in the presentimplementations to prevent any dephasing of the water mag-netization by Bo field gradients [24] or by RF inhomogeneityduring proton composite-pulse decoupling [25]. Strip plotsfrom both spectra are shown in Fig. 6. The intensity ratios ofcross peaks involving nC resonances and axial peaks dependon the x'- r e l a t ed ^(N.C) and 37(C',C) couplings, re-

A ,

spectively, while the intensity of interresidual 13CP cross peaksin the HN(CO)CG correlates with the f-angle [20]. Except forthe very weak cross peak between Ilel48-15N and Alal47-13Cp

of flavodoxin (Fig. 6A) signals corresponding to 3J(N,CKi)interactions relevant to the determination of \|/- angles were notobserved in the HNCG as they are invariably small [10].

Finally, %x information can be derived from 3J(Ha,Cr)coupling constants. The HCGHA sequence presented in Fig. 7essentially is the long-range 13C-'H correlation experiment(LRCH) [13], supplemented with the WATERGATE scheme[27], allowing its application to proteins in H2O solution.Following the initial heteronuclear NOE-enhancement, I3Cmagnetization is transferred to long-range coupled protons inan 'out-and-back' manner. Our modified phase cycle applies to

15

13

13

H AJN

CY

CO

•»I • Ih|x|

1

i

! 1

l

y

1 18n

82

DIPSI-2*

i 1 i; x* 4 4"5

1 t2 1

82

y

c

2

J_t,

A

•±1 1

y•d1

• • •X' g X' |X'g

•

1

rec:4.r

i GARP

| GARP .

13ca

Gi Gt G2

B i , , «,

G4 G 4 G 5 G5

I I i A

H Al I I15

N13

i.\kp

aQaliph,

G1 Gi G2 G3G4G4G5G5 G6

LJLJLJ L

Figure 5. Experimental schemes for measuring ^(N.C11) (A) and 3/(C',Cp'r) (B) coupling constants. All proton pulses are appliedat the water resonance. The initial selective 90° pulse is Gaussian shaped and has a duration of 10 ms. RF field strengths for 'H,l5N and 13CO CPD sequences are 4.5, 0.75, and 1.0 kHz, respectively. Carbon carrier positions are 45 ppm in the HNCG (A) and50 ppm in the HN(CO)CG (B) experiment. The power level for the rectangular 90° 13C7 pulses (at 28 ppm) in A is adjusted toprovide zero excitation in the 13CO region. Pulses on 13C (at 60 ppm) are 1.2 ms G3 Gaussian cascades [23]. During ti, "C and13CO spins are inverted simultaneously by a 124 \is cosine-modulated sine-pulse. In sequence B, 90° and 180° pulses on aliphaticcarbons are centered at 33 and 58 ppm, respectively. Delays and phase cycles are: (A) x = 2.3 ms, x' = 2.5 ms, x" = 0.7 ms, TJ =5.4 ms, 5 = 64 ms, <j>, = -y,y, 2̂ = y,-y. <j>3 = 2(x),2(-x), <|>5 = 4(x),4(-x), <t>6 = y, <t>r = R, <|>i = R.-R; (B) x = 2.3 ms, x' = 2.5 ms, x" =0.7 ms, r| = 5.4 ms, e = TN = 28 ms, 8 = 37 ms, $1 = y,-y, <t>2 = 4(x),4(-x), <t>3 = x,-x,y,-y, <t>5 = 8(x),8(-x), fa = x, fa = y $t = R,-R, <!>•= R,2(-R),R (R = x,2(-x),x). The phase of the hatched pulses is either x or $5 for the real (<t>4 = x) and imaginary (((u = y) part ofeach t2 increment. In both sequences A and B gradients are: Gi = 1 ms, 7.5 G/cm (x); G2 = 1 ms, 10 G/cm (y); G3 = 1 ms,-39.4/+39.4 G/cm (z); G4 = 0.5 ms, 4 G/cm (x) 5.5 G/cm (y); G5 = 05 ms, 5.5 G/cm (x) 4 G/cm (y); G6 = 0.5 ms, 8 G/cm (z).

Vol. 20, No. 1-4 13

the two central 90° 'H pulses, similar to the HNHBexperiment. During the second half of the pulse sequence, 13Cmagnetization antiphase with respect to directly bondedprotons builds up which is readily converted into observable'H magnetization using a reverse INEPT prior to acquisition.As shown in Fig. 8, the resulting spectra contain I3CV1Ha/IHr

(F1/F2/F3) cross peaks and axial peaks at F2 = 0. Due to thelong period of transverse I3C magnetization (60 ms) sufficientsignal intensity is obtained only for the slowly relaxing methylcarbons and thus 3J0P,C) coupling constants can be measuredin valine, isoleucine, and threonine residues. Three-bondcouplings involving leucine and isoleucine 5-carbons as well

135-

122

122 C

3

122 C

'J(N,C)=2.0Hz

Q84, E110

'J(N,Cl=1.5Hz

1o Q84C"

E110 C"

0 Q84Ce V

E110CT

tV(N,C)=1.4Hz

L124

0L124 C

G123 C0

11481148 C i

A147C

1148 C

1148 C

'JtN.C'l'O.SHzI4N,Cr>)=1.2Hz

B

7.9 7.7V41

©V41 C"

oV41 C"

&

E42C

o'J(C'.C")=2.BHz'J(C'.C")=1.0Hz

8.2 8.0P73

V(C\C')=2.0Hz>J(C',C;,)=2.8Hz

P73C

o

L74C'

7.8 7.6L1120

L112C

' C>O ®

K113C

"J(C\C)=3.6Hz

7.0 6.8R131

oA132C"

aR131 C

i

V(C'.C)=3.SHz

8.3 8.1 6.9 6.7 8.6 8.4•4-

7.3 7.1

20

25

30

35

40

45

50

20

25

30

35

40

45

50

F3 [S,H, ppm]

a.a.

CMU -

Q.Q.

to.

CM

Figure 6. F2/F3 strips from HNCG (A) and HN(CO)CG (B)spectra of flavodoxin, taken at the 1SN frequencies of theresidues given at the top of each panel (A) or of the followingresidues (B). Positive (solid lines) and negative (dotted lines)contours are drawn on an exponential scale where each level is2m times higher or 2 times lower than the preceding one.

15N,13CO

JLFigure 7. Pulse scheme employed to measure VOEF.C) and37(HP,CS) coupling constants. Narrow and wide bars represent90° and 180° pulses, respectively, applied along the x-axis,unless otherwise noted. The proton carrier is set to the waterresonance at the beginning of the sequence, jumped to 5.15ppm before the pulse labeled fo, and returned after the pulsewith phase x or (j>4. Low power rectangular 90° pulses in theWATERGATE sequence have a duration of 1.5 ms. The I3Ccarrier is positioned at 26 ppm. RF fields of pulses on "C"8 are18 and 13 kHz for 90° and 180° flip angles, respectively.Delays A, Tc, T and x' are adjusted to 10, 60, 1.9, and 1 ms,respectively. The length of £ is 3.2 ms, such that JCH evolvesfor a period of 23.6 ms ( = 3/'JCH) before t2, while being activefor 21.6 ms ( = 2.75/'JCH) with K = 4.2 ms after t2, where '/CHis the one-bond coupling constant in methyl groups. Phasecycling is as follows: <(>i = x, fa = x,y,-x,-y, <|>4 = 2(x),2(-x), <t>5

= 4(x),4(-x), fc = 4(y),4(-y), <t>r = 2(x,-x),2(-x,x), fc = x,2(-x),x,-x,2(x),-x. The hatched pulse is applied along x for the realpart (<1>3 = x) and with phase §* for the imaginary part (<t>3 = y)of each t2 increment. Quadrature detection in Fl is achieved byStates-TPPI [26] of (j>i. Durations and strengths of pulsed fieldgradients are: Gi = 1 ms, 10 G/cm (x); G2 = 1 ms, 10 G/cm(y); G3 = 0.8 ms, 17.5 G/cm (x) 12.5 G/cm (z).

as VOEF'.C2) couplings in isoleucines are relevant for thedetermination of x2-angles. Note that vicinal interactions withlabile protons such as H1" of Thrl5 in flavodoxin can also bedetected, providing information about thex2I-angle.

Conclusions

A novel phase cycling scheme has been proposed for incor-poration into a variety of heteronuclear /-correlation experi-ments. Quantitative evaluation of heteronuclear couplingconstants takes advantage of internal reference signals locatedin axial 2D planes of the 3D spectra without recourse toseparate calibration experiments.

Experimental

The HNHB experiment was carried out on a 4 mM sampleof 15N enriched Desulfovibrio vidgaris flavodoxin dissolved in0.5 ml 10 mM potassium phosphate buffer (pH = 7.0). It wasrecorded at a Broker DMX 500 spectrometer equipped with aself-shielded z-gradient triple-resonance probe. All other pulse


T15y

2.3 Hz

@©

n ^*

#>o

L26 5"H V

©LjB

•

4.9 Hz]J(H",Cl)=

1.6 Hz

I65y

©

.«•t"

2.1 HzJ(H"'.C*)=4.2/5.3Hz

1119 5

H"

H'

>J(H',C)=4.4 Hz

V144y2

o

Jjk.

o

o"

V(H-,C)=3.6Hz

o

0.0

1.0

2.0

3.0

4.0

5.0

6.0

Q.a.

CM

1.2 1.0 0.2 0.0 0.7 0.5 0.9 0.7 -0.2 -0.4

F3 [8W.6 ppm]

Figure 8. Expansions from HCGHA slices of flavodoxin,taken at the I3C chemical shifts of the methyl groups specifiedat the top. Cross peaks and axial peaks are distinguished bydense (solid) and sparse (broken) contours, respectively. Thesuperscripts u and d denote upfield and downfield resonatingprotons located in methylene or isopropyl groups, respectively,indicating that stereospecific assignments are not available.

sequences were performed at a Bruker DMX 600 spectrometerwith xyz-gradient accessory, using a 1.4 mM 13C/15N labeledflavodoxin sample. The temperature was set to 300 K.

Eight scans per FID were accumulated in the HNCG andHCGHA experiments, whereas 16 scans were acquired other-wise. Total recording times were 63 h (HNHB), 74 h (HN(CO)HB), 36 h (HNCG), 68 h (HN(CO)CG), and 42.5 h (HCGHA).All data sets were processed with the XWIN-NMR program(Bruker). Since the lineshapes of cross peaks and axial peaksare identical along Fl in all spectra, peak volumes were deter-mined by 2D integration in the F2/F3 slices at the maximumintensity in Fl or from the sum of two adjacent slices.

Acknowledgements

C.P. acknowledges a grant from the Deutscher Akade-mischer Austauschdienst (DAAD). S.G. Mayhew (Dublin) andM. Knauf (Frankfurt) are thanked for help in sample prepara-tion. Support from the Deutsche Forschungsgemeinschaft (Ru145/11-2) is acknowledged.

References

[1] A. Bax, G.W. Vuister, S. Grzesiek, F. Delaglio, A.C.Wang, R. Tschudin, and G. Zhu, Methods Enzymol. 239,

79-105 (1994).[2] G.W. Vuister and A. Bax, J. Am. Chem. Soc. 115,7772-

7777 (1993).[3] S. Grzesiek, H. Kuboniwa, A.P. Hinck, and A. Bax,

J. Am. Chem. Soc. 117,5312-5315 (1995).[4] J.-S. Hu and A. Bax, J. Am. Chem. Soc. 118, 8170-8171

(1996).[5] P.R. Blake, B. Lee, M.F. Summers, M.W.W. Adams,

J.-B. Park, Z.H. Zhou, and A. Bax, J. Biomol. NMR 2,527-533 (1992).

[6] SJ. Archer, M. Ikura, D.A. Torchia, and A. Bax,J. Magn. Reson. 95,636-641 (1991).

[7] S. Grzesiek, M. Ikura, G.M. Clore, A.M. Gronenborn,and A. Bax, J. Magn. Reson. 96, 215-221 (1992).

[8] G.W. Vuister, T. Yamazaki, D.A. Torchia, and A. Bax,J. Biomol. NMR 3,297-306 (1993).

[9] P. Dux, B. Whitehead, R. Boelens, R. Kaptein, and G.W.Vuister, J. Biomol. NMR 10,301-306 (1997).

[10] J.-S. Hu and A. Bax, J. Biomol. NMR 9,323-328 (1997).[11] J.-S. Hu and A. Bax, J. Am. Chem. Soc. 119,6360-6368

(1997).[12] R. Konrat, D.R. Muhandiram, N.A. Farrow, and L.E.

Kay, J. Biomol. NMR 9,409-412 (1997).[13] G.W. Vuister and A. Bax, J. Magn. Reson. B102,228-

231 (1993).[14] A.J. Shaka, C.J. Lee, and A. Pines, J. Magn. Reson. 77,

274-293 (1988).[15] AJ. Shaka, P.B. Barker, and R. Freeman, J. Magn.

Reson. 64,547-552 (1985).[16] S. Part, J. Magn. Reson. 96,94-102 (1992).[17] G.A. Morris and R.Freeman, J. Am. Chem. Soc. 101,

760-762 (1979).[18] J.C. Madsen, O.W. S0rensen, P. S0rensen, and

F.M. Poulsen, J. Biomol. NMR 3,239-244 (1993).[19] D.J. States, R.A. Haberkorn, and DJ. Ruben, J. Magn.

Reson. 48,286-292 (1982).[20] G. Zhu and A. Bax, J. Magn. Reson. A 104, 353-357

(1993).[21] A.G. Palmer JH, J. Cavanagh, P.E. Wright, and M. Ranee,

J. Magn. Reson. 93,151-170 (1991).[22] L.E. Kay, P. Keifer, and T. Saarinen, J. Am. Chem. Soc.

114,10663-10665 (1992).[23] L. Emsley and G. Bodenhausen, Chem. Phys. Lett. 165,

469-476 (1990).[24] S. Grzesiek and A. Bax, J. Am. Chem. Soc. 115,12593-

12594 (1993).[25] L.E. Kay, G.Y. Xu, and T. Yamazaki, J. Magn. Reson.

A 109,129-133 (1994).[26] D. Marion, M. Ikura, R. Tschudin, and A. Bax, J. Magn.

Reson. 85,393-399 (1989).[27] M. Piotto, V. Saudek, and V. Sklenaf, J. Biomol. NMR 2,

661-665 (1992).

Vol. 20, No. 1-4 15

Self-Consistent Relaxation-Rate Analysis applied toFatty-Acid Binding Protein

J. M. Schmidt, C. Ludwig, and H. RiiterjansInstitut fur Biophysikalische Chemie, J.W. Goethe-Universitat Frankfurt

Biozentrum N230, Marie-Curie-Str. 9, D-60439 Frankfurt am Main, Germany

IntroductionMolecular internal mobility gains increasing importancein modeling macromolecular structure on the basis ofNMR spectroscopy. Of particular utility are dipolar-relaxation effects that probe the reorientation of bondvectors between specified pairs of nuclei. Modernstable-isotope enrichment techniques applied to proteinsalleviate sensitive determination of heteronuclearrelaxation-rate constants involving 15N being now usedin routine studies on molecular motion in solution [1].Simple models of N—H bond-vector reorientation areusually fit to a set of 15N longitudinal and transverserelaxation times, Tx and T2, respectively, as well as 15N-{'H} nuclear Overhauser enhancement (NOE) effects.Series of internal-fluctuation order parameters S2 andassociated life times t are then determined in the contextof the so-called Lipari-Szabo analysis [2,3]. However, itis only on a per-residue basis that details on localangular dynamics along the protein chain are retrieved,whereby critical assumptions on the global constants ofthe overall molecular-tumbling correlation time tc, theN-H bond length rNH, and the nitrogen chemical-shiftanisotropy AaN must be made.

The information content in relaxation data is not yetexhausted in these standard procedures, andcharacterization of molecular mobility might beimproved by a novel self-consistent relaxation-dataevaluation strategy. Self-consistency entails the jointinterpretation of all available relaxation rates from, forexample, the majority of amino-acid residues in aprotein. Redundant information inherent to large sets ofrelated relaxation-rate constants is then exploited notonly to assess local N-H bond reorientation but also torefine initial estimates of global properties critical torelaxation. The concept involves simultaneous least-squares regression of model parameters while referingto experimental relaxation rates, using the extendedPeng-Wagner type of spectral-density mapping [4,5].

Most critically, accurate values emerge for the globalfactors persistent to all spectral densities, without priorassumptions on xc, rNH, and Aa^, respectively.

TheoryTo a first approximation, molecules in solution tumbleisotropically with overall reorientational correlationtime xc giving rise to the spectral density./(GO) = (2/5)xc(l + CO2T;2)-I. Subject to a magneticfield of proton Larmor frequency <BH, relaxation ofnitrogen nuclei senses the spectral-density function atthe five frequencies / 0 = / ( 0 ) , JN = ./(<%),J'H = /(coH), JA = / ( © H - C O J , ) , andJx = /(coH + ooN). The predominant nitrogenrelaxation processes in a 15N-enriched protein aredipole-dipole (DD) interaction of the amide 15N nucleuswith the attached proton, and, to a lesser extent,chemical-shielding anisotropy (CSA) due to theplanarity of the peptide linkage. The dipolar portion isaffected by the interatomic distance rNH, while thechemical-shift anisotropy A<j = |cr±-C||| depends onthe magnetic flux density Bo, finally leading to theinteraction constants£DD = (l/8)(Y2Y2/r6H)(/z/2Ji)

2(|X0/47i)2 and, respectively.

Experimental parameters measured to probe the N-Hdipole reorientation are single-spin nitrogenlongitudinal and transverse relaxation rates, /?(NZ) and/?(NX ) , respectively, as well as the HN-crossrelaxation rate R(HZ -> N z) . These are supplementedby the rates of HN-antiphase relaxation /?(HZNX y ) ,HN-longitudinal 2-spin order /?(HZNZ), and H-longitudinal relaxation R(HZ).The system of simultaneous equations relatingrelaxation rates and spectral densities casts into a matrixequation, also known as spectral-density mapping [4,5],so that the set of rates determined for each residue at asingle field strength is given by


<-,y

/?(HNV)z-"x, y'

7?(HZNZ)

0 6c 2d 0 12d 0

4c 3c d 6d 6d 0

Ac 3c d 0 6d 1

0 6c 0 6d 0 1

0 0 2 6rf 12d 1

0 0 -Id 0 12d 0 PHH

where coefficients involving c = + ^CSA

andd = t,DD indicate the relative contributions of power-spectral density to NH dipolar interaction and N-chemical shift anisotropy. While relaxation interferencedue to cross-correlation of dipolar and CSA interactionis experimentally excluded, leakage pathways ascribedto potential homonuclear proton-dipolar interaction isaccounted for by the non-specific compound relaxationrate p H H - ^.r^HJ(J0 + 3 / H + 6/2 H) .

MethodThe evaluation strategy proposed exploits redundantinformation inherent in a large set of relaxation rateconstants. As the set of the six 15N-related relaxationrates never can adopt all possible value, combinations,local angular mobility is assessed from the uniquepattern in a few relaxation rate values already [8,9].Nevertheless, overdetermination in multipleexperiments allows more complicated models of motionto be applied. In the context of a rigid molecular model,the six experimental determinants are opposed to asingle adjustable global correlation time xc, thus givingrise to five degrees of freedom. At the expense of twodegrees of freedom, internal NH bond mobilityaccording to the model-free approach is assessed inaddition. However, as perceived from the coefficientmatrix, both transverse-relaxation rates /?(NX y) andR(HzNxy) show a very similar dependence on thespectral density, such as to reduce the availableinformation in the data set. Measurements at variousfield strengths add only limited information to the dataset as no new spectral-density function is sampled.Advantage is taken here from linking redundantrelaxation properties in the manifold of amino-acidresidues in the studied molecule using a self-consistentanalysis. In other words, residues are no longer analyzedseparately. Redundant relaxation information in thisway forms a pool of excess degrees of freedom, which,apart from fitting descriptors of local dynamics, allowsfor refining initial estimates of a number of globalparameters critical to relaxation as well.

»60C

*80C

•Hri

'60C

I I l II I I I

Experiment

'80C

n I I III I I I

C50C

0^ 6 0 0

0

^ 8 0 0

Simulation Coefficients

Figure 1. Matrix arrangement in iterative simultaneous back-calculation of self-consistent model parameters of m amino-acid residues based on 3 x 6 relaxation rates each. Mixingcoefficients and spectral densities Jo for fields of 500,600, and800 MHz are merged into one row and column, respectively.

Figure 1 depicts the crucial data array organization of allrelaxation-rate values related to the NH directors in thepolypeptide. So far, variation of order parameters andinternal-motion correlation times only affect therespective matrix rows, i.e. those entries associated witha specific residue. Data redundance is exploited byrealizing that variation of global properties like theoverall rotational correlation time, the NH bond length,and the CSA factor give rise to simultaneous changes inall entries in the R matrix. Thus, self-consistentrelaxation-rate evaluation is driven by the effect ofredistributing fit deviations over the complete array.Transformation of spectral density into relaxation ratesis accomplished by applying the coefficient matrix to allresidue-related data sets at the same time, in shorthandnotation R = CJ = JCL Multiple-field relaxationmeasurements allow to concatenate data into biggerarrays R and J while the C matrix becomes block-diagonal as depicted.

Experiments and Relaxation DataThe method is exemplified with recombinant 15N-enriched bovine-heart fatty-acid-binding protein (FABP,pi 5.1 isoform, 133 aa, 15 kDa) which specifically bindsC16 and C18 fragments [10]. Rate constants of auto-relaxation of longitudinal (Nz, 2NZHZ, and Hz) andtransverse (Nx and 2NXHZ) magnetization componentsas well as those of heteronuclear NOE cross-relaxation(HZ-»NZ) originate from a variety of heteronuclearHSQC-type pseudo-3D NMR experiments [5-7] carriedout in aqueous solution at 310 K and at three differentproton Larmor frequencies (500, 600, and 800 MHz).Each relaxation rate was accurately determined fromleast-squares exponential fits to at least 8 multipletintensities collected at experimental delay periods from3 to 800 ms, depending on the type of experiment.

Vol. 20, No. 1-4 17

Table 1: Self-Consistent FABP-Mobility Parameters from the Analysis of Nitrogen Relaxation Rates0

Model -,2

isotropic444052421123181231732310823085

axial symmetry2304823026

P

1

101

201

202

203

. 211

204

206

[ns]

5.596.076.146.146.136.12

6.496.54

*c2

[ns]

——

—

—

——

5.955.91

(Sf)%

100. ± 0.088.3 ±5.187.2 ± 5.285.8 ± 5.282.9 ± 5.382.8 ± 5.2

83.0 ± 5.383.0 ± 5.3

<T;>

[ps]

——

28 ± 4525 ± 4354 ± 32754 ± 326

56 ± 34655 ± 340

rNH

[pm]

——

—

101.6101.4101.4

101.4101.4

AaN {RJ 0,<D

[ppm] [s1] [deg]

— — —— — —— — —— — —

168.57 — —168.58 4.7 + 4.0 —

168.72 — —168.78 — 12, 18

a% » summed squared deviation in the relaxation rate constants in = 1800 observables) weighted by their experimental errors, p,number of model parameters so that n — p is the number of degrees of freedom, xc and %c2, principal and two-fold degeneratemolecular rotational correlation times of the oblate rotor, S2 and x, order parameters and correlation times of internal motion givenas averages over the 100 residues included in the fit, rNH, apparent length of the covalent polypeptide backbone amide NH bond,AaN, nitrogen chemical-shift anisotropy, Rex, chemical-exchange rates given as average over selected residues to which they apply,0,O, tilt angles between the principal-axis systems of rotation-diffusion and the inertia tensor obtained from cartesian coordinates.

Results and DiscussionMolecular motion in FABP was studied dn the basis ofextensive 15N-relaxation data subject to self-consistentrate evaluation. Based on a selection criterion of themean Tx/T2 ratio and ±3 standard deviations thereof,100 non-glycine and non-proline residues were includedin non-linear regression of a total of 1800 experimentalrate constants. To match back-calculated andexperimental relaxation rates, only the overall rotationalcorrelation time t c was initially adjusted with theunderlying assumption of a rigid isotropic rotor model[11]. In the following instances, the model complexitywas increased as to make use of the large number ofdegrees of freedom available to study the effect ofabandoning various simplifications initially made.

Molecular reorientation correlation timeFrom self-consistent 15N-relaxation rate evaluation,overall molecular reorientation of a hypotheticallyspherical FABP molecule would occur with an isotropiccorrelation time of xc = 6.12 ± 0.40 ns. However, thestudy also revealed that FABP more likely possessesaxial symmetry [12] of an oblate rotor with an axis-length ratio of 1:0.9:0.9 and respective correlation timesof tc = 6.54 ± 0.60 ns and xc2 = 5.91 ± 1.26 ns, the lattervalue being associated with the degenerate axes. Neglectof internal NH bond mobility would have yielded asignificantly faster reorientation with xc = 5.59 ns from

which it is stated that the NH bond vectors experienceangular and/or radial fluctuation.

Order parameters ofNH bond fluctuationConsidering a bond vector (of fixed length) libratingwithin a cone of rotational symmetry about a mean localorientation (0), its motion is convenientlycharacterized by the angular order parameter given by

1

= <p2(coseaP))f

where P2(x) is the second-rank Legendre polynomialand 6ap is the jump angle difference between any twoorientations of the NH bond vector with associated stateprobabilities pa and pp. Accordingly, radial fluctuationsare described by time-averaged radial order parameters

c2 _ /r-3 \2/r-6 \-l

The formal separation of angular and radial changes inthe NH bond geometry [13] affects the dipolarrelaxation rates in a manner that R °= S%Sf. Sinceadjustment of any of the two order parameters may fully


Figure 2. Protein fold of FABP, view of the fatty-acid bindingpocket, and orientation of the oblate-disk model of thediffusion tensor. Picture created from X-ray coordinates [14](2HMB) using RasMol by Roger Sayle, University ofEdinburgh.

0.4

[ns]0.3

0.2

0.1

0.0 ill JuliiiAJ JiUuli .llu..,iL20 40 60 80 100 120

residue number

Figure 3. Order parameters and local correlation times of NHbond reorientation in individual amino-acid residues of FABPas obtained from self-consistent relaxation rate evaluation.

compensate for effects from the other, both valuescannot be fitted independently. In our analysis, onlyvalues of S2, were associated with individual amino-acidresidues, while S2 was considered a global parameterbeing however completely absorbed into the value of theNH bond distance determined.All of the angular order parameters S2, are in theexpected range of 0.7-0.9, i.e. re-orientation angles are10-20°, and hardly correlate with secondary structurepointing out that internal mobility is likely to be uniformalong the protein chain. The details obtained under thedifferent model assumptions are summarized in Table 1.

Internal motion correlation timesNuclear relaxation is most sensitive to motionalprocesses on the nanosecond time scale of molecularreorientation [13]. Internal mobility on a shorter timescale reduces the relaxation efficacy by the S2 orderparameters discussed above.

Having the effect of overall tumbling properly modelledand subtracted from the spectral-density function, theremaining portion superimposed on /(co) is due tointernal mobility. This is conveniently expressed interms of the associated time correlation function [2,3]

C(0 = S2 + ( l -S 2 )exp( - f /x )

from which the average residence time of NH bondorientation x is retrieved.However, x values vary wildly in the analysis with meanand median values of approximately 50 and 15 ps,respectively, indicating that internal NH bond mobilityin FABP is of significantly higher frequency and occurson a picosecond time scale far off the moleculartumbling time. A connection with secondary structure isnot perceived.

Vol. 20, No. 1-4 19

X2

12

10

8

6

4

2

n

i

j.>

V

\

/,500 MHz.'//

JL0.100

rNH [nm]-160

Figure 4. Dependence of the fit error on variation of theinternuclear bond length and of the chemical-shift anisotropy.Note the enlarged CSA error scale.

Polypeptide amide NH bond lengthThe critical NH bond length often used in routine 15N-relaxation studies ranges from 100 to 104 pm [1]. Self-consistent analysis allows to adjust the optimal value atthe same time the residue-specific internal mobility isassessed. Note that optimizing localNHbond lengths iscounteracted by variations in the angular orderparameters S% with which a full anti-correlationpersists. For similar reasons, fitting the global NH bondlength in the rigid-molecule approximation (S2 = 1) isexpected to lead to an unacceptably large bond length.The final value of the average NH bond lengthconverged at rNH = 101.4 ± 1.3 pm. The narrowstandard error points out that misadjustment of thisparameter might bias the results in a conventionalrelaxation study considerably.

Nitrogen chemical-shift anisotropyWith increasing magnetic fields, chemical-shieldinganisotropy becomes non-negligible. Based on smallmolecules, routine I5N-relaxation studies of polypeptidebackbones use a AaN value of 160 ppm [15]. Self-consistent relaxation rate analysis again allows forconcomittent adjustment of the optimal value whichconverged at AcN = 168.6 ± 16.2 ppm and agrees betterwith recent experiments [16]. However, error marginsare much wider compared with those of the NH bondlength determination, implying that the CSA valueactually chosen affects the results less critically. Similarreasons justify the assumption of a uniform CSA valuefor all of the amide groups in the polypeptide, although

values might depend on amino-acid composition andpolypeptide fold.

Chemical exchangeTransverse relaxation rates of approximately 10% of theresidues were found to not agree with the range ofexpected T2 values. These sites are likely to be affectedby chemical-exchange phenomena. Additionalparameters were applied to account for the increase ini?(Nx y) and/?(HzNxy) rates, but the results obtainedcannot be assigned to a unique physical or chemicalprocess.

ConclusionsOverall and intramolecular dynamics in bovine-heartfatty-acid binding protein was assessed by graduallyincreased levels of motional model complexity. FABP insolution is an oblate rotor exhibiting restricted rapidinternal motion on the picosecond time scale.Relaxation relevant parameters have been refined whichare useful to other relaxation studies. Work targeted atfully anisotropic modelling of motion in FABP is inprogress.

References[ 1 ] J. W. Peng and G. Wagner, Meth. Enzymol. 239,563-

596 (1994)[2] G. Lipari and A. Szabo, J. Am. Chem. Soc. 104,4546-

4559(1982)[3] G. Lipari and A. Szabo, J. Am. Chem. Soc. 104,4559-

4570 (1982)[4] J. W. Peng and G. Wagner, Biochemistry 31, 8571-

8586 (1992)[5] J. W. Peng and G. Wagner, J. Magn. Reson. 98, 308-

332(1992)[6] L. E. Kay, L. K. Nicholson, F. Delaglio, A. Bax, and

D. A. Torchia, J. Magn. Reson. 97, 359-375 (1992)[7] S. Grzesiek and A. Bax, J. Am. Chem. Soc. 115,

12593-12594 (1993)[8] J.-F. Lefevre, K. T. Dayie, J. W. Peng, and G. Wagner,

Biochemistry 35,2674-2686 (1996)[9] N. A. Farrow, O. Zhang, A. Szabo, D. Torchia, and L.

E. Kay, J. Biomol. NMR 6,153-162 (1995)[10] N. M. Bass, Int.Rev. Cytol. I l l , 143-185 (1988).[11] D. E. Woessner, J. Chem. Phys. 37, 647-654 (1962)[12] D. E. Woessner, B. S. Snowden Jr., and G. H. Meyer,

J. Chem. Phys. 50,719-721 (1969)[13] R. Briischweiler, B. Roux, M. Blackledge, C.

Griesinger, M. Karplus and R. R. Ernst, J. Am. Chem.Soc. 114,2289-2302 (1992)

[14] G. Zanotti, G. Scapin, P. Spandon, J. H. Veerkamp,and J. C. Sacchettini, J. Biol. Chem. 267,18541-18550(1992)

[15] Y. Hiyama, C. Niu, J. V. Silverton, A. Bavaso and D.A. Torchia, J. Am. Chem. Soc. 110,2378-2383(1988)

[16] N. Tjandra, P. Wingfield, S. Stahl, and A. Bax, J.Biomol. NMR 8,273-284 (1996)


C-F Bonding in Fluorine-GraphiteIntercalation Compounds and Graphite

Fluorides: 13C and 19F NMR Study

A. M. PanichDepartment of Physics, Ben-Gurion

University, Be'er Sheva 84105, Israel

Abstract

13C and 19F NMR chemical shift datain fluorine-graphite intercalation compoundsand graphite fluorides are presented. Thenature of C-F bonding in these compounds isdiscussed.

Introduction

Graphite reacts with fluorine gas in awide range of temperatures, yielding twokinds of compounds with different electricalproperties: fluorine-graphite intercalationcompound. ̂ CXF, and graphite fluorides(CF^ and (C2F)n [1]. Graphite fluorides(CF)n and (C2F)n are prepared by thefluorination of graphite at high temperature,from 300 to 600 oc. Their carbon skeletonsare not planar but consist of trans-linkedcyclohexane chairs with sp3 bonding [2-6].The in-plane symmetry of both (CF)n and(C2F)n belongs to a hexagonal system [2-4].Graphite fluorides are electrical insulators;C-F bonds are considered to be completelycovalent [1].

Fluorine-graphite intercalationcompound is usually synthesized at ambienttemperatures in the presence of traces ofmetal fluorides or HF as a catalyst [1]. It isnow considered a unique family of graphiteintercalation compounds (GICs) whichexhibits a behavior that is strikingly different

from all other known GICs [7-10]. For mostGICs the in-plane electrical conductivityinitially increases upon intercalation due tothe increase in carrier density resulting frominjection of carriers from the intercalatespecies to the graphite layers. This results inan increase in conductivity by an order ofmagnitude and transformation of thesemimetallic graphite into a metal. Influorine-intercalated graphite, CXF, thein-plane conductivity also increases with theincrease of fluorine content for dilutecompounds, x>8, has a plateau for 8>x>6,but then, in contrast to the other GICs,decreases sharply with further intercalation[7-10]. Based on XPS data, it was suggestedthat the nature of C-F bonding changes fromionic to more covalent character (calledsemicovalent or semi-ionic) with increasedfluorine content [1, 8-10]. Such an increasein the covalent character of the C-F bondwas expected to reduce the carrier densityand cause a rapid decrease in conductivity.

Recently, we applied 13C and 19FNMR to study the origin of semimetal-metaland metal-insulator transformations influorine-intercalated graphite CxF for a widerange of fluorine content, 3.8 <x< 12.7 [11].We have shown that fluorine spectra forsmall fluorine content, x>8, show a narrowline attributed to mobile fluorine acceptorspecies which are responsible for the increaseof electric conductivity in the dilutecompound. When increasing the fluorinecontent to x~8 corresponding to themaximum electric conductivity, a broad linetypical for covalent C-F bond starts to occur.The position and shape of the broad line arepractically independent on fluorine contentindicating that the character of C-F bonds isalmost unchanged. On the other hand, theintensity ratio of broad to narrow line and,thus, the number of covalent bonds growswith fluorine content resulting in thedecrease in conductivity which is caused by a

Vol. 20, No. 1-4 21

percolation mechanism rather than by achange in bond length. The distribution in 19FNMR chemical shifts has also been obtainedand associated with a disorder in F sites forthe intermediate fluorine content.

In the present paper, we analyze 13Cand I9F chemical shift data influorine-graphite intercalation compoundsand graphite fluorides and discuss the natureof C-F bond in these compounds. Wesupposed to get the information of interestbecause chemical shifts of NMR lines aresensitive to local bonding as well as toelectronic structure of compounds.Paramagnetic contribution to chemical shift,(8p)a ~ 2{<M/o|Lz/r

3|ynxv(/n|Lz|v)/o>/An},results from the mixing between theoccupied ground state and the unoccupiedexcited states. (Here A is the difference inenergy of the ground and excited state). Thisterm strongly depends on the structure ofexternal molecular orbitals taking part inchemical bonding and thus is sensitive to thenature of the chemical bond.

13C and 19F chemical shifts and C-Fbonding

Our recent measurements of 13Cchemical shift [11] along with literature data[12] on fluorine-graphite intercalationcompounds are presented in the diagram Fig.1. (We note that all 13C shifts discussedbelow are given relative to tetramethylsilane(TMS), and all 19F shifts are given relative toCFCI3). One can see that 13C resonances fallinto two regions, which are assigned tographitic carbons (112-129 ppm) and C-Fcarbons (86 ppm), respectively. The formerone's show the position close to thatdetermined in graphite and reflects the factthat, at least for carbons which are not boundto fluorine, the planar structure of the carbon

graphitei1

graphiticC-F carbons

1diamond

140 120 100 80 60 40 20ppm, relative to TMS

13 r r

Fig. 1. C chemical shifts in fluorine-graphiteintercalation compounds [11, 12].

layers is almost preserved during theintercalation. The chemical shift of C-Fcarbons is intermediate between sp2 graphiticcarbons and sp3 aliphatic carbon atoms indiamond which is shown for comparison.Because of cubic symmetry of the carbon sitein diamond, the paramagnetic contribution isabsent, and highfield shift 8=35 ppm causedonly by diamagnetic contribution is observed.When going from diamond to benzene andgraphite, hybridization is changed from sp3

to sp2, and reduced symmetry of the carbonelectron shell gives rise to the paramagneticcontribution and therefore to downfield shift.From Fig. 1, one can suggest that, duringintercalation, fluorine attaches the carbonplane and partially destroys the delocalizedelectronic state of graphite in its immediatevicinity creating C-F bonds with adiamond-like geometry, although the localsymmetry of this carbon site is reduced incomparison to diamond yieldingparamagnetic contribution to the chemicalshift and thus downfield shift of this line. Themore corrugated the carbon plane is, thehigher the local symmetry of the carbon siteand the less the value of 6p are, and the moreupfield chemical shift should be.


It is noteworthy that l3C chemicalshifts of C-F bound carbons in insulatorgraphite fluorides lie in the range from 65ppm in (Ci.igFIo.o23)n, [13] to 85 and 86 ppmin C2.27F(HF)o.23 and (C1.04FIo.02On,respectively [14] and are close to the shiftsof C-F carbons in fluorine-GICs mentionedabove. Graphitic carbons in two lattercompounds exhibit chemical shift +132 and+138 ppm, respectively [14]. No continuoustransformation of the chemical shift and nointermediate regime which should becharacteristic for a smooth transition fromionic to semi-ionic (or semicovalent) andthen to covalent C-F bonding has beenobserved. Thus one can conclude that C-Fbonding in fluorine-GICs and in graphitefluorides does not show significantdifference. At least the present 13C spectra(as well as physical properties [11]) can, inprinciple, be explained in the frames of amodel which does not involve uncommonsemi-ionic (or semicovalent) C-F bond.

Some difference in the C-F bond forfluorine-graphite intercalation compound andgraphite fluoride can be seen from 19F NMRdata (Fig. 2). In CxF, asymmetric broad lineof covalent C-F bond lies in the rangebetween -30 and -270 ppm with its center of

200 100 0 -100 -200 -300

ppm, relative to CFC13

Fig. 2. Room temperature ^ spectra of fluorine-GICC^F and graphite fluoride ( C F \ at 282.4 MHz.

gravity at 5 —150 ppm [11]. Our recentmeasurements of 19F chemical shifts in fully

fluorinated compound (CF)n which is aninsulator and, according to the literature [1,10], is supposed to exhibit completelycovalent C-F bond, show a chemical shift8=-174±10 ppm [11] (Fig. 2). Measurementsof Hamwi et al. show 8(CF)n=-197±20 ppm[13]. The difference in 19F chemical shiftbetween CXF and (CF)n indicates a differencein chemical bond. However, the (CF)nresonance shows a shift to high field incomparison to CxF. This is in contradictionwith the common fluorine chemical shiftscale where a more covalent compoundexhibits a downfield shift [15, 16]. Thus thisdifference in chemical shift can hardly bedescribed in terms of ionicity and covalencyand does not accord with an idea of ionic tocovalent transformation of C-F bonding withincreasing fluorine content based on XPSdata. From our data, we can conclude thatinterior CF-groups adjacent to aromatic sp2

carbons in nonsaturated CxF compoundexhibit a downfield chemical shift, whilethose adjacent to sp3~hybridized CF speciesin the fully fluorinated and saturatedcompound (CF)n exhibit highfield chemicalshift. We also note that the absence ofconductivity in the fluorine sublatticeexcludes the existence of 19F Knight shift.

To understand the nature of C-Fbonding in these compounds, we note thatPauling [17] related partial ionic character ofthe single bond between atoms A and B tothe difference in their electronegativities xA

andx5:I=l-exp[-(xA-xBf/4] (1)

Pauling calculated that C-F bond is expectedto have 44 percent ionic character becausethe electronegativity of fluorine (XF=3.95) islarger than that of carbon (xc=2.5).Following Pauling, we conclude that C-Fbond in fully fluorinated compound (CF)n ispartially ionic and thus shows highfield 19Fshift characteristic for ionic compound.

Vol. 20, No. 1-4 23

On the language of molecular orbital(MO) theory, hybrid MO of the C and Fatoms is <F=XCM'C+A,FV|/F- Pure covalent bondshows XC=A-F. Based on the idea of Pauling,one can suggest that fully fluorinatedgraphite (CF)n shows A,F>XC, and C-F bondelectrons are preferably located at F atomrather than at C atom because of largerelectronegativity of fluorine.

Fluorine-GICs, CxF, show the otherX?fkc ratios and charge distribution at C-Fbond in comparison to (CF)n. According tothe 1£F chemical shift scale and chemical shiftdata mentioned above, one can conclude thatthe electron distribution on the C-F bond influorine-GICs is such that A* is closer to Xcthan in (CF)n, with the electrons partiallymoved from fluorine to carbon atom incomparison to graphite fluorides. In thissense, the broad ^ line in fluorine-GICsmay be attributed to more covalent C-F bondthan in graphite fluorides. Thus in F-GICs,most of fluorine atoms bound to carboncannot withdraw significant electron densityfrom the graphite a or 7t-system.

Measurements of the transportproperties showed that the CXF systemundergoes a transition from a well-orderedionic compound at low fluorineconcentrations to a weak-localization regimefor intermediate fluorine concentrations, andfinally, to a strong localization regime forhigh fluorine concentrations [10]. Theseweak localization effects have been obtainedin the resistivity measurements (slightlogarithmic increase in resistivity withdecreasing temperature at low temperatures).NMR data do not contradict these results. Inour recent paper, the broad 19F line wasattributed to a distribution of chemical shiftsand associated with a disorder in F sites forthe intermediate fluorine content [11]. This isin agreement with transmission electronmicroscopy measurements which haveestablished the presence of disorder in

fluorine-intercalated graphite [18]. In aslightly disordered lattice, quantuminterference effects become important whenconsidering multiple scattering of conductionelectrons. Phase coherence of electron wavefunction leads to the theory of weaklocalization, generated by the interferencebetween elastically backscattered carrierwaves [10].

In addition, it is interesting tomention the 19F MAS and 19F-13C CP/MASstudy of the charcoal fluorinated at -80 to350°C [19]. The basic structural units ofcharcoals are microcrystallites composed ofstacked sp2 carbon platelets. While graphiteconsists of microcrystallites composed ofparallel-oriented platelets, charcoals exhibitdisorder in both microcrystallite orientationand platelet composition. UC spectra showthree distinct resonances at 129, 112 (weak)and 86 ppm assigned to C, CF2 and CFspecies. The 5 values for C and CF speciesare close to that in fluorine-GICs andgraphite fluorides mentioned above.Deconvolution of the broad I9F featuresyields the resonances assigned to CF3 -groups (-55 to -90 ppm), CF2 -groups (-100to -145 ppm), and CF-groups (-130 to -200ppm). As fluorination temperature increases,F/C ratio also increases, and resonanceintensity within the CF range shifts upfield assp2 carbon is consumed. Thesemeasurements show that chemical shift ofCF-groups adjacent to aromatic carbon arefound in the low field extreme of the CFrange, while adjacent to other fluorocarbonspecies at the high field extreme. In thespectrum of the CxF prepared at -80°C,signal is centered at -144 ppm, while in thespectrum of the C1.2F prepared at 350°Cresonance intensity is shifted upfield to -182ppm, in agreement with our measurementsmentioned above.


References

1. T. Nakajima, in: Fluorine-Carbon andFluoride-Carbon Materials, ed. T.Nakajima, Marcel Dekker, Inc., New York,1995,p.l-31.2. Y. Kita, N. Watanabe and Y. Fujii, J. Am.Chem. Soc. 101 (1979) 3832.3. V. K. Mahajan, R. B. Badachape and J. L.Margrave, Inorg. Nucl. Chem. Lett. 10(1974) 1003.4. N. Watanabe, Physica 105B (1981) 17.5. A. M. Panich, A. M. Danilenko and S. P.Gabuda, Dokl. ANSSSR 281 (1985) 389.6. A. M. Panich, Colloids Surf. 72 (1993)19.7. D. Vaknin, I. Palchan, D. Davidov, H.Selig, and D. Moses, Synth. Met. 16 (1986)349.8. L. Piraux, V. Bayot, J. P. Issi, M. S.Dresselhaus, M. Endo, and T. Nakajima,Phys. Rev. B 41 (1990) 4961.9. S. L. di Vittorio, M. S. Dresselhaus, M.Endo, and T. Nakajima, Phys. Rev. B 43,1313(1991).10. M. S. Dresselhaus, M. Endo and J.-P.Issi, in: Fluorine-Carbon andFluoride-Carbon Materials, ed. T.Nakajima, Marcel Dekker, Inc., New York,1995, p.95-186.11. A. M. Panich, T. Nakajima, H.-M.Vieth, A. Privalov and S. Goren, J. Phys.:Condens. Matter 10 (1998) 7633.12. T. Mallouk, B. L. Hawkins, M P.Conrad, K. Zilm, G. E. Maciel and N.Bartlett, Phil.Trans. R. Soc. London A 314,(1985) 179.13. A. Hamwi, M. Daoud, D. Djurado, J.C.Cousseins, Z. Fawal and J. Dupuis, Synth.Met, 44 (1991) 75.14. A. Hamwi, J. Phys. Chem. Solids, 57(1996) 677.15. H. Gunther, NMR Spectroscopy, JohnWiley & Sons, Chichester-NewYork-Brisbane-Toronto, 1980.

16. S.P. Gabuda, Yu.V. Gagarinsky and S.A.Politschuk, NMR in inorganic fluorides,Atomizdat, Moscow, 1978.17. L. Pauling. The nature of the chemicalbond. Cornell University Press, 1960.18. K. Oshida, M. Endo, T. Nakajima, S. L.di Vittorio, M. S. Dresselhaus, and G.Dresselhaus, / . Mater. Res., 8 (1993) 512.19. E.W. Hagaman, D.K. Murray and G.D.Del Cul, Energy and Fuel, 12 (1998) 399.

Vol. 20, No. 1-4 25

A 2H-NMR Relaxation Study of Surfactant Orderin Surface Aggregates and Bulk Structures

M. Schonhoff1'2*, O. Soderman1, Z. X. Li3 and R. K. Thomas3

Physical Chemistry 1, University of Lund, 22100 Lund, SwedenMax-Planck-Institute of Colloids and Interfaces, 12489 Berlin, Germany

3Physical and Theoretical Chemistry Lab, Oxford 0X1 3QZ, U.K.

AbstractNonionic surfactant adsorption layers

on colloidal silica are investigated byapplying 2H-NMR on selectively deuterateddodecyl penta(ethylene oxide), C12E5. Localmolecular order and timescales of surfactantmotional modes in surface aggregates arecompared to different types of bulk aggre-gates, such as micelles, and aggregates inlamellar and hexagonal phases. Surfaceaggregates show an isotropically averagedlorentzian line shape with relaxation ratesR̂ of the order of several kHz, indicating aslow motion correlation time of t ~ \is.Possible mechanisms of isotropic averagingin surface aggregates are discussed.

Relative order parameter profiles Sre]

are obtained for surface and different bulkaggregates from quadrupolar splittings and2H-relaxation, respectively. An increase ofSrel with increasing distance of the 2H labelposition from the headgroup was found forall bulk aggregate types, with the slopedepending on aggregate curvature.Comparison of the results for surfaceaggregates to these profiles gives anindication of their curvature.

The results are consistent with the for-mation of large anisotropic surfaceaggregates, with adsorption leading tomotions slowed down compared to bulkmicelles, and isotropic averaging achieved

by exchange and/or dynamic micelle re-formation processes.

Introduction

Surfactants at interfaces can form avariety of different aggregate structures,such as monolayers, bilayers, or adsorbedspherical or anisotropic micelles. Nonionicalkylpoly(ethylene oxide) CnEm surfactantaggregates show curvatures that arestrongly dependent on n and m. Adsorptionlayers of C12E5 on flat hydrophilic silicasubstrates for example have been charac-terized by reflection methods: ellipsometryresults pointed at the formation of large flatsurface aggregates1, and neutron reflectivityexperiments are consistent with a bilayermodel2. An AFM approach revealed thelateral nanoscale structure for a series ofCnEm surfactants, but did not lead toreproducable results for C12E5

3. In a small-angle neutron scattering study of C12E5 oncolloidal silica however, the determinedlayer thickness was interpreted by smallmicelle formation4.

In this work we apply 2H-NMR as analternative method to study the localmolecular order and the local dynamics insuch surface aggregates. The surfactantC12E5, forming large rodlike micelles insolution, or liquid crystalline phases athigher concentrations, is adsorbed tocolloidal silica. 2H-quadrupolar splittings


and 2H-relaxation on selectively deuteratedC12E5 reveal information about local orderparameters and timescales of surfactantmotional modes. The results for surfaceaggregates are compared to different bulkstructures as reference systems.

Materials and Methods

C12E5 in the selectively deuterated formwith the label in the a, p\ and y- position ofthe alkyl chain, respectively, is syntesizedfrom suitably labelled dodecanol and penta-ethylene glycol. The degree of deuterationin the respective position was determinedfrom 1H spectra and amounts to 93 % forthe a-deuterated form, 16 % for (3, and 43% for the y- deuterated form. As substrate,colloidal silica Cab-O-Sil (Fluka) with aspecific surface area of 200 m2/g is used.

Samples of micellar solutions orordered binary phases are prepared bymixing with deuterium-depleted water. Insamples with high surfactant concentrationthe deuterated form is diluted withprotonated C12E5 obtained from Nikko(Japan). Adsorption samples on silica areprepared by mixing deuterated surfactantand Cab-O-Sil in dilute aqueous solution.The surfactant amount corresponds to fullsurface coverage, which is 45 A2/molecule,as determined from ellipsometry measure-ments1. After equilibration under gentleshaking for several days the samples arefilled into NMR tubes and centrifugeddirectly prior to measurement. Spectra aretaken on the centrifugation pellet containingCab-O-Sil. 2H solids spectra are obtainedby applying a quadrupolar echo sequenceusing a 2H solids probe head with asolenoid coil, and liquids spectra are takenwith a 90-acq sequence in a liquids probehead. All experiments are performed on aBruker DSX 200 spectrometer.


In the lamellar and hexagonal phase (at69 % and 50 % wt. surfactant, respec-tively), the spectra show a Pake pattern. Infig. (1) the order parameter S of the C-Dbond relative to the aggregate surface isshown, which is calculated from thequadrupolar splitting AQ according to

AQ = 3/n S x, (1)where % is the quadrupolar coupling

constant, n = 4 in the lamellar phase and n= 8 in the hexagonal phase on account ofthe reduction of AQ due to the averaging ofthe interaction by fast diffusion in theangular direction in hexagonal rods.

0.12

0.1CO

.2l o . O 8

8.5 0.06•so 0.04

0.02

0

1 4 , ^ • *

-

- —A— 69 % wt. C12E5, lamellar 1

y 50 % wt. C12E5, hexagonal |

Position of D label

Fig. 1: Order parameters in the lamellar andhexagonal phase, obtained from quadrupolarsplittings. Positions 1, 2, 3 indicate deuteration onthe a, p, and y carbon of the alkyl chain.

The order parameters are increasingtowards the hydrophobic region. This resultis opposed to the typical behaviour of ionicsurfactants in a lamellar phase5'6, where Sis decreasing with distance from theinterface due to a large hydrophilic/hydro-phobic contrast which is causing a sharpand ordered headgroup/chain interface. Itcan be concluded that in ethyleneoxidesurfactant aggregates where the hydropho-

Vol. 20, No. 1-4 27

bicity of the headgroup and the alkyl chainis less different, order is mainly driven bythe aggregation of the chains. Additionally,the absolute values of S are lower than inionic surfactants, where they reach valuesof up to S = 0.25, which is again proof ofthe headgroup/alkyl interface being lessordered in ethyleneoxide surfactants. Con-sistent with our results, for perdeuteratedC12E4 in the lamellar phase, a maximum ofS is found at the fourth carbon atom7.

In micellar solutions, the spectralshape is lorentzian, since the quadrupolarinteraction is averaged by intramicellardiffusion and rotational tumbling. Accor-ding to the two-step model for NMR re-laxation in surfactants8, the relaxation ratedifference AR = R2 - R1; is determined bythe slow isotropic motion correlation time

AR = R2-R, = 9/20 7t2x2'S2Ts (2)

This equation assumes <BTS » 1.Measurements of R, in concentratedmicellar samples show that it is R , « R2.Since Rj relates to the fast anisotropicmotions, which are independent onaggregation state, it follows that for Rj at allconcentrations:

R 2 = AROCTSS 2 (3)

In micellar solutions, R2 evaluatedfrom the linewidth is increasing with con-centration (see fig. 2). This reflects mainlyan increase of ts, which is due to micellargrowth causing slower tumbling motions.This is consistent with micellar growth seenin PFG-NMR diffusion experiments9. Dueto the dependence of R2 on both xs and Sabsolute values for S cannot be obtainedfrom the relaxation data. However,applying equ. (3), relative order parametersSrel = S/Sa can be calculated, making use ofthe independence of t s on label position.

7T 3 -

2 -

1 -

1 %wtC12E56.4 % wt C12E515%wtC12E5on surface after centrif.on surface after redisp.

Position of D label

Fig. 2: Relaxation rates R2 of bulk micellesat different concentrations and surface aggregatesafter different treatments.

Srel values are given in fig. (3), andcompared with Srel in the anisotropicphases. For all aggregate types, an increasewith position is observed, and it isinteresting to find that the slope of therelative order parameter profile is dependingon aggregate curvature: In flat aggregateslike the lamellar phase Srel is almostindependent on label position, whereas withdecreasing concentration and decreasingaggregate size, the mobility gradient ismonotoneously increasing. Thus, the Srel

profile can be taken as a measure ofcurvature.

In adsorption samples, the spectralshape is found to be lorentzian. Thisindicates the presence of an isotropicmotional mode, fast enough to average thequadrupolar interaction by a motionalcorrelation time of t s < 150 (is. On the otherhand, rather large relaxation rates R2 on theorder of several kHz were deduced (see fig.2). This shows the existence of a very slowmotional mode, which will be discussedbelow. Relaxation rates are found to beslightly dependent on water content in thesamples, and are increasing after successive


cycles of centrifugation and silica redis-persion in solution.

i i i

1.35

1.3

1.25

1.2 -

'i.is

1.1

1.05 :

1 .

— 9 — 1 % wt C12E5—O— 6.4 % wt C12E5— • — 1 5 % wtC12E5— • — 50 % wt., hexagonal— * — 69 % wt., lamellar

Position of D label

Fig. 3: Relative order parameter profiles ofbulk micelles and anisotropic phases determinedfrom R2 and quadrupolar splittings, respectively.

Relative order parameters Srel= S/Sa ofsurface aggregates show an increase oforder with label position as well (see fig.4). By comparison of the slope to the datain fig. (3) conclusions on the curvature ofsurface aggregates can be drawn: Surfaceaggregates after centrifugation (full squaresin fig. 4) are more similar to the lamellar orhexagonal phase than to small micelles,indicating surface structures in the rangebetween large anisotropic micelles andbilayers. This is consistent with the resultsfrom neutron reflection and ellipsometry1'2.After redispersion and centrifugation cyclesthese large structures are probablydestroyed and the Srel profile becomes moresimilar to small micelles.

Additional indications about thesurface aggregate structure can be obtainedfrom the absolute relaxation rate values:Assuming Sa to be the same as in thehexagonal phase, from R2 = 1.7 kHz, forcentrifuged adsorption samples a correlationtime of ts ~ 1.1 (is (redispersion samples:1.8 us) is deduced.

— • — on surface after centrif. 1—D— on surface after redisp. |

: /

-. r ]

1

1.35

1.3

1.25

1.2

2

" 1 . 1 5 -

1.1

1.05

1 -

Position of D label

Fig. 4: Relative order parameter profiles ofsurface aggregates treated by centrifugation only,and treated by centrifugation- redispersion cycles.

Motional modes that could possiblycause isotropic averaging in surface aggre-gates and determine R2 are: I. intramicellardiffusion in case of small spherical surfacemicelles, II. diffusion along the silicasurface in case of closed layer aggregates,III. rotation of the particle, or IV. othermechanisms like exchange and breakingand re-formation of aggregates.

An estimate of the expected correlationtimes xs for these motions results in fastnanosecond dynamics for type I motions.In large anisotropic surface micelles,diffusion would partly average thequadrupolar interaction and additionalisotropic modes would be necessary forcomplete averaging. In the case of closedbilayer formation, diffusion around theparticle will average AQ. For the inner andouter layer correlation times of 0.3 (is and0.45 jis, respectively, are estimated, whichare faster than the xs- value obtained fromRj, but the diffusion might be slowed downin a disrupted bilayer.

The particle rotation will not play amajor role, since the silica consists ofspherical particles which are interconnected,

Vol. 20, No. 1-4 29

thus preventing rotation. For motions oftype IV, the time scale is difficult toestimate, one reason being that it willdepend strongly on the local surfactantconcentration. The exchange time scalebetween free surfactant and a colloidalsurface in a dilute system was found to beon the order of 12 ms9, but is expected tobe much faster in these concentratedsamples.

Finally, it follows from this discussionthat the motions with correlation timesconsistent with the R2 data are type IV.motions, they will therefore dominate xs,and cause isotropic averaging. This ispossibly achieved in combination with typeII. diffusion, if the surface aggregate is adisrupted bilayer.

ConclusionsWe have determined the order para-

meter profiles of C12E5 in various kinds ofaggregates. In all bulk and surface aggre-gates order parameters are increasing withlabel position, indicating order to beinduced by the hydrophobic aggregation,whereas the hydrophobic/hydrophilic inter-face is less ordered. The slope of therelative order parameter profile is found toreflect the aggregate curvature in bulkaggregates of known shape. In comparison,Srel in surface aggregates corresponds tolarge flat structures.

Additional information is given by theR2 relaxation rates and a discussion ofpossible motional modes in the system. Thequadrupolar interaction is averaged by aslow isotropic motional mode. The shape ofthe surface aggregates is large anisotropicmicelles or disrupted bilayers, with iso-tropic averaging achieved by exchangeand/or dynamic aggregate re-formation pro-cesses.

AcknowledgementsThe authors would like to thank

Fredrik Tiberg for helpful discussions. M.S. was supported by a grant, partly fromthe European Commission TMR programand the DFG (Scho 636/2-1). The spectro-meter was sponsored by the SwedishCouncil for Planning and Coordination ofResearch.

References1 F. Tiberg, B. Jonsson, J.-a. Tang,

and B. Lindman, Langmuir 10, 2294-2300(1994).

2 P. N. Thirtle, Z. X. Li, R. K.Thomas, A. R. Rennie, S. K. Satija, andL. P. Sung, Langmuir, 5451-5458 (1997).

3 L. M. Grant, F. Tiberg, and W. A.Ducker, J. Phys. Chem. B 102, 4288-4294 (1998).

4 P. G. Cummins, J. Penfold, and E.Staples, J. Phys. Chem. 96(20), 8092-4(1992).

5 J. H. Davis, Biochim. Biophys.Acta 737, 117-71 (1983).

6 U. Henriksson, L. Odberg, and J.C. Eriksson, Mol. Cryst. Liq. Cryst. 3 0 ,73-78 (1975).

7 A. J. I. Ward, H. Ku, M. A.Phillippi, and C. Marie, Molec. Cryst. Liq.Cryst. 154, 55-60 (1988).

8 H. Wennerstrom, B. Lindman, O.Soderman, T. Drakenberg, and J. B.Rosenholm, J.Am.Chem.Soc. 101, 6860-6864 (1979).

9 M. Schonhoff and O. Soderman, J.Phys. Chem. B 101, 8237-42 (1997).


ANOMALIES OF INHOMOGENEOUSLY BROADENED RESONANCE

LINES SHAPE DUE TO NONLINEAR EFFECTS CONTRIBUTION

M.D.Glinchuk, S.N.NokhrinInstitute for Problems of Materials Science, NASc of Ukraine,

252180 Kiev, Ukraine

ABSTRACTThe theory of inhomogeneously broadened resonancefines shape is developed in the case when spin-packet res-onance frequency shift is square random field function.It was shown that the line shape /(&>) is narrow 5-typecurve with broad wings. Mechanisms of homogeneousbroadening lead to the line broadening and Imax{u) de-creasing. The theory fitted pretty good anomalies ofobserved NMR lines shape of 93Nb in PbMgxfiNbîzOz(PMN) and KTa0t9i»Nbofi\7Qz (KTN). This made itpossible to discuss the peculiarities of structure of thedisordered ferroelectrics PMN and KTN.

1 INTRODUCTIONThe investigations of anomalies of resonance lines shapein ESB. and NMR spectra draw the scientists attentionin many years. It is because the analysis of lines shapemakes it possible to obtain a valuable information aboutthe distribution of random fields in solids, about the con-centrations and characteristics of the defects, spin-latticeinteraction etc. (see e.g. [1] and ref. therein).

Static and dynamic (in the scale of measurement)characteristics of the internal fields can be obtained onthe base of analysis of inhomogeneous and homogeneousmechanisms contribution into resonance lines shape andwidth. In particular, in the disorder systems magneticresonance lines usually are strongly inhomogeneouslybroadened so that their shapes correspond to distribu-tion function of random field in lattice. This function isknown to be especially important for the physics of phasetransitions and calculation of the system phase diagram[2,3].

Quantitative information about random field sourceswhich denned the inhomogeneous broadening of reso-nance lines can be obtained on the base of comparison ofobserved and calculated resonance Hne shape. The cal-culation can be carried out in the statical theory frame-work allowing for both linear and nonlinear contributionof random fields. The latter was shown [4,5] to changeessentially resonance line shape calculated on the base of

linear contribution only (6ee [1,6]). Meanwhile the pre-cise analytical expression for resonance line shape wasobtained for the case of sum of linear and nonlinearterms, in the case of pure quadratic term only approxi-mate solution in the framework of the statical theory ofthe II order was proposed [7]. Note, that II order staticaltheory gives line shape in complex enough integral formwhich can be solved only numerically without possibilityof analytical representation of the line shape.

The quadratic random field contribution to resonancefrequency shift is frequent event in radiospectroscopy,in particular, in NMR spectra of nuclei with largequadrupole moments and spin I > 1 (e.g. ^Nb,1*1 To),where the lines width and form determines mainly by nu-cleus quadrupole moment interaction with electric fieldgradient, the line shape of transition ±1/2 —* 71 /2 isdenned by quadratic gradient of random electric fieldcontribution.

In present work the theory of resonance lines shape inthe case of squared random field contribution to inhomo-geneously broadened fine in developed. The comparisonof the calculated and early measured 93iV6 NMR spec-tra in incipient ferroelectric KToOz doped by Nb andrelaxor ferroelectric PbMg1/3Nb2f3O3 is carried out.

2 THEORYIntensity of inhomogeneously broadened Hne at the fre-quency wo+Ao> is known to be proportional to the num-ber of local perturbation sources which lead to the shiftof resonance frequency w0 on Aw value. In dependenceon magnitude of paramagnetic center spin and its char-acteristics such as spin-phonon, spin-electric constants,quadrupole moments at nuclei magnitude of hu> can bedefined by random magnetic, electric and elastic fieldsand their gradients, created by different imperfectionsas these fields sources. Quantity of random field sourcesis known to be extremely large in the disordered fer-roelectric (ferromagnetic) systems due to substitutionaldisorder, vacancies in cation and anion sublattice as wellas due to random site and orientation electric (magnetic)

I

Vol. 20, No. 1-4 31

dipoles.Allowing for linear and nonlinear contribution of afore-

mentioned fields, one can write down the shift of reso-nance frequency in the form

Au = ae + be2(1)

where a is dimensionless constant, 6 has the dimensionof inverse frequency.

In the case when only the linear contribution is essen-tial (a / 0, > = 0) line shape / i (») can be calculated inthe I-st order statistical theory framework and is knownto have Gaussian, Lorentzian or Holtzmarkian form independence of random field type [lj.

The problem of resonance shape calculation in the casewhen linear and nonlinear contribution up to min powerare essential was solved in the statistical theory frame-work recently [4,5] and lead to the following equation

(2)

Here u>t are the real roots of the algebraic equation

<p(e) = u,-ae-be2-cea- ... - fem = 0 (3)

Note that Eq.(2) can be obtained also in the frameworkof probability theory as the distribution of probabilityfor the function of random quantity [8].

In the case we are interestedin, when there is no linearcontribution , i.e. a = 0, 6 ̂ 0 in Eq.(l), tine shape onthe base of Eqs. (2), (3) can be written in the form

ftW=^qK\/f)+''(-VD] <«It is seen from Eq.(4), that nevertheless /i(<w = 0)

has maximum value f2 —* oo at u> —• 0, i.e. at reso-nance frequency because zero point in u> scale coincideswith resonance frequency o> = ua. The divergency of/2(a> = 0) can be transformed into maximum by takinginto account any even small mechanism of homogeneousbroadening, which do exist in any real crystal. In manycases homogeneous broadening leads to Lorentzian tineshape, so that the tine shape has to be the convolutionof /2(4>) and Lorentzian. The same results can be ob-tained on the base of more simple procedure with thesubstitution in Eq.(4) «—•««> + ifr (r"1 is halfwidth onhalfheigh of Lorentzian), the tine shape being the realpart of /2(o> + i/r), namely

(5)

Eq.(5) is general form of tine shape allowing for bothinhomogeneous and homogeneous broadening.

Let us suppose that / i (» ) has Gaussian form,i.e. A H = l /(AV^)exp(-a;/(2A2)) , where A ish&lfwidth. In this case Eq.(5) can be rewritten as

expf - cos2 6 T A 2

(6)It can be expected from Eq.(6) that with the increas-ing of dimensionless parameters 6A and ( A T ) " 1 , whichcharacterize inhomogeneous nonlinear contribution andhomogeneous broadening respectively maximal intensityhmax = /2OO has to decrease and halfwidth has to in-crease due to normalization condition f f% (w)dw = 1.

The exact expression for halfwidth wi/a can be ob-tained from Eq.(6) and expressed in the form

(7)

The results of numerical calculations of line shape /zO^)with the help of Eq.(6) for several values of oA and( A T ) " 1 are represented in fig.la,b. It is seen, that thetine shape essentially differs from Gaussian form, whichcorresponds to linear case, and with 6A and ( T A ) " 1

increasing /jmax does decreases and its halfwidth in-creases. Note, that the existence of several mechanismsof homogeneous broadening, each having Lorentzianform, leads to 1/T = ^ 1/T,-, where i numerate the in-

idependent mechanisms of homogeneous broadening.

3 DISCUSSION AND COMPARISON

WITH EXPERIMENTResonance line shape depicted in fi.g.1 draws an attentionby unusual form which looks like ^-function nearby res-onance frequency w = 0, but it has very broad wings at\w\ > (ifi/2, their intensity being very small at &>/A > 2.The tine is completely symmetrical relatively resonancefrequency. It is because the peculiar point connectedwith zero value of denominator in Eq.(4) coincides withresonance frequency w = 0 of tine shape calculated inlinear approximation. In the case when both linear andsquared contributions are essential (« # 0, b ^ 0 inEq.(l)) the aforementioned coincidence does not takeplace as one can see from Eqs. (2) and (3). This leads tostrongly asymmetrical tine shape with two or one maxi-mum in dependence on a/6 ratio as it was shown recently[4,5]. Therefore peculiar form of line shape representedin fig.l can be considered as the direct consequence oflinear contribution absence and existence only squared


random field contribution. Under experimental condi-tion the linear terms may be absent at special externalmagnetic field orientation and everywhere for the transi-tion ±1/2 ++ Tl/2 (I > 1) which is known to be sensitiveonly to nonlinear contributions of random electric fieldand its gradient. In fig.2 we represented the compari-son of observed early [9] and calculated one shape of thetransition +1/2 -»• -1/2 of 93Nb NMR spectrum in thedisordered ferroelectric PbMgi/3Nb2/3O3. Let us con-sider the procedure of fitting in more the details. It iswell known [9], that nucleus of 93Nb has large quadrupolemoment, so that its line shape has to be sensitive torandom electric field gradient due to interaction of thequadrupole moment with the gradient. This interactionwill lead to linear and nonlinear terms in resonance fre-quency shift for all the transition (93Nb spin I = 9/2),but the central one ±1/2 *-* =fl/2. For the transition±1/2 *-+ T I / 2 the shift is calculated usually in the Il-ndorder of perturbation theory so that it is squared electricfield gradient function (see e.g. [10]). Therefore the lineshape can be represented by Eq.(6), where parameter ofhomogeneous broadening 1/r would be connected withmagnetic spin-spin interaction of Nb nuclei. Calculationof 1/r value on the base of conventional formulas (seee.g. [11]) in supposition that in macroregions the rationof Mg to Nb ions concentration corresponds in averageto 1:2 gives 1/r Rf 3 kHz. This value made it possi-ble to obtaine the values of parameters 6A == 0,25 andA = 0,25 kHz by fitting Eq.(7) with observed halfwidthvalue and measured maximal intensity of the line withthe equation

f . (8)

It is seen from fig.2 that Eq.(6) for aforementionedvalues of parameters fits pretty good observed NMRline shape. Some asymmetry of observed line shapecan be connected with small additional to quadrupoleand spin-spin contributions which were not taken intoaccount in theoretical consideration. Note that if itcould be the macroregions with 1:1 type of ordering andthe regions enriched by Nb (see e.g. [12]), ^Nb NMRspectrum would contain two lines contrary to observedone line spectrum. On the other hand spin-spin inter-action in the regions of 1:1 type gives ]/T = 1 kHzwhich leads to physically unreasonable result «j/2 < 0as one can see from Eq.(7). Therefore our considerationgives evidence about the existence only of 1:2 macrore-gions in PMN and absence of 1:1 regions and those en-riched by Nb. NMR line of 93Nb looks like that infig.l was observed in another disordered ferroelectricKTai-zNbxO3 with x = 0,012 [13]. The increasing of1/2 —• —1/2 transition line maximum intensity and the

line narrowing was observed at temperature decreasingfrom T = 19,9 to T = 7,9 K. Such behavior follows fromEq.(6) when the value of homogeneous broadening 1/rdecreases (see figs. la,b). In KTN homogeneous broad-ening can be denned by spin-spin interaction betweenNb and lslTa?9 K nuclei as well as reorientation of Nbbetween its electric dipole (Nb is known to be off-centerimpurity in KTaOz) eight equivalent orientations. Inthis case 1/r = 1/TS-S + l/rm, where the first and thesecond term is the contribution of two aforementionedmechanisms. Since reorientation frequency l/rm has todecrease with temperature decreasing one can expect 1/rdecreasing. This has to result in the line narrowing andits maximal intensity increasing (see fig.l).

The detailed fitting of the theory and experimentaldata is represented in fig.3 for T = 19,9 (a); 18,1(b); 7>9 (c)- Th« value of l/rm for all the tempera-tures was calculated on the base of Arrenius formulal/rm = l/r0 exp(-(7/T) with U - 200 K, l/r0 = 7 • 10s

Hz [14] which correspond to reorientation of elastic mo-ment of Nb. Contribution of nuclei spin-spin interaction

. was calculated on the base of conventional formula andgave l/rs_j = 1,5 kHz. One can see from fig.3, thatEq.(6) fits pretty good observed line shape for all tem-peratures with the following parameters: A = 4 kHz;6A = 0,14 and with the values of l /rm , calculated fromaforementioned Arrenius law, l/rm = 300 kHz; 110 kHzand 0,1 kHz at T = 19,9 K; 18,3 K and 7,9 K respec-tively.

The authors of [13] tried to explain unusual line shapeof 93Nb in KTN in supposition that the origin of broadwings of the line at T > Tc {Tc = 10 K was supposedto be a temperature at which Nb ions become off-centerones) is forbidden for cubic symmetry satellite transi-tions like 1/2 -> 3/2, 3/2 ->• 5/2 etc., meanwhile withsymmetry lowering at T < Tc the satellites are shifted farfrom a resonance field of the transition ±1/2 *-* ^ 1/2 i.e.they contribute nothing to this central transition. Thismodel however contradicts to the fact of approximatelyconstant integral intensity of spectra observed at T < Tc

and T > Tc. The explanation proposed in our model isfree from supposition about Nb'ions shift at T = Tc RJ 10K, which was not confirmed later (see e.g. [15]). We tookinto consideration the reorientation of elastic moment ofAr6 ions, measured early independently by ESR [14] andNMR Q technique. Note, that reorientation frequencyof elastic and electric moment of Nb is strongly differ-ent [14] so that it seems to be cumbersome to discussoff-center position of Nb ions at T < Tc on the base ofNMR measurement performed in the work [13].

References

Vol. 20, No. 1-4

0.8 r-r-

33

&c

0.6

• a 0.4(0

0.0

0.81

CO

0.6

• 0 2

0.00.8

•90.4

f£0.2

0.0

-2

= 1.05

- 6 - 4 - 2 0 2 4 6

2 4 6bA=1.751 - 1/TA = 0.12-1/tA = 0.2

2 3 -1 /TA = 0.4

0a>/A

«»>,, kHz

Fig.2 "Nb NMR spectrum mPoints - experiment, solid Hne - theory.

0.5

0.4

0.3

02

0.1

0.0O25

§ 0.15

^ 0.10

I

Fig.1. Numerical calculations of inhomogeneously

broadened resonance Hne shapes. Parameters bA and (AT)"1

are characteristic of inhomogeneous nonlinear contribution

and homogeneous broadening, respectively. The dotted line

represents Gaussian Hne shape calculated in the 1* orderstatistical theory framework.

005

0.00

20J

IS

1J0

O 5

OJO

T=18.1 K

T-7.8K

-10 10

Fig.3 "Nb NMR specrum in KTa^^Nb^Ô,.Points - experiment, solid line - theory.


1. M.D.Glinchuk, V.G.Grachev, M.F-Deigen,AJBJEloitsin, L.A.Suslin, Elektricheskie effecty v ra-diospectroskopii, Nauka, Moskva (1989).

2. M.D.GEnchuk, V.A.Stephanovich,JPhys.:Condens.Matter, v.6, N3, p.6317 (1994).

3. M.D.Glinchuk, R.Farhi, J.Phys.:Condens.Matter,v.«, N9, p.6985 (1996).

4. MD.Glinchuk, I.V.Kondakova, Fiz.Tv.Teia (inpress).

5. M.D.Glinchuk, I.V.Kondakova, J.MolecPhys. (inpress).

6. A.ILStoneham, Rev.ModPhys., v.41, Nl , p.82(1969).

T. A.U.Stoneham, Proc.Phys.Soc, v.l, N2, p.565(1968).

8. D.Hudson, Statistica dlya fizikov, Mir, Moskva(1967).

9. V.Viaguta, MJ>.Glinchuk,IP.Bykov, AJT.Titov,EJ»l.Andreev, Fiz.Tv.Tela, v.32, NlO, p.3132 (1990).

10. M.D.Glinchuk, I.P.Bykov, V.Viapi ta , Ferro-electrics, v.143, Nl , p.39 (1993).

11. A.Abraliam, The Principles of Nuclear Mag-netism, Oxford, Clarendon Press (1961).

12. E.Husson, L.Abello, A.Morell, Mat.Res.Bull.,v.25, N2, p.539 (1990).

13. J.J.van der Klink, S.Rod, A.C3iatelian,Phys.Rev.B, v.33, N4, p.2084 (1986).

14. T.V.Antimirova, M.D.Glinchuk, A.P.Pecheniy,I.M.Smolyaninov, Fiz.Tv.Tela, v.32, N l , p.208 (1990).

15. B-E.Vugmeister, M.D.Glinchuk, Rev.Mod.Phys.,v.82, p.993 (1990).

Vol. 20, No. 1-4 35

Structural Investigations on the Activation of Plasminogen by Staphylokinase

Oliver Ohlenschlager*, Ramadurai Ramachandran*, Karl-Heinz Giihrs*, Bernhard Schlott* and Lany R. Brown*

Department of Molecular Biophysics /NMR Spectroscopy,Department of Biochemistry,

Institute for Molecular Biotechnology,Postfach 100813, D-07708 Jena, Germany

Introduction

Thromboembolic disorders have emerged to be one maincause of mortality in the western world. In order to dissolvethe fibrin clots by the protease plasmin, an activation of plas-minogen has to take place (1). The 15.5 kDa bacterial proteinstaphylokinase (Sak) isolated from Staphylococcus aureus, anon-enzymatic cofactor like streptokinase (2), forms a stochi-ometric protein-protein complex with plasmin(ogen). Cleav-age of the peptide bond LyslO-Lysl 1 of Sak in this complexis the trigger for the formation of an enzyme species capableof activating plasminogen (3). This is in contrast to enzy-matic cofactors like tissue plasminogen activator or uroki-nase, which activate plasminogen by ' their inherentproteolytic activity (4, 5). The Sak-plasmin(ogen) complexshows a high degree of specificity for cleavage of blood clotswith reduced side effects because, in contrast to streptokinaseactivation, the Sak-plasmin(ogen) complex is strongly inhib-ited by (X2-antiplasmin circulating in the blood (6,7,8). Theprotein staphylokinase is currently undergoing clinical trialsfor the therapy of myocardial infarction (6,9) and peripheralthrombosis (10) because of its profibrinolytic properties.

We have recently determined the NMR solution structureof the full-length protein by multidimensional heteronuclearNMR spectroscopy (11,12). Staphylokinase exhibits a welldefined global structure of the (J-grasp like fold. A single cen-tral oc-helix of 13 residues is flanked by a two-stranded 13-sheet, both of which are located above a five-stranded P-sheet. The N-terminus folds back onto the protein core. Dur-ing activation of plasminogen the N-terminal sequence oftenresidues is proteolytically cleaved, thereby creating a newcharged N-terminus. NMR data on this N-terminal processedvariant SakANIO was additionally acquired allowing to dis-cuss the structure and dynamics during the initial activationstep. The investigation of the conformations of native andmutant staphylokinases serves as a starting point for under-standing the changes in the enzymatic specificity of plasminand the mechanisms of plasminogen activation and will allowthe design of improved drugs for lysis of blood clots.

Materials and Methods

NMR samples of staphylokinase were expressed in

E. coli TGI transformed with plasmid pMEX602sakB andpurified as recently described (11,13). The concentration ofthe Sak sample was 1.1 mM (15N Sak). N-terminal proc-essed, 15N-labeled staphylokinase (15N SakANIO) wasobtained at a concentration of 1.3 mM after treatment of N-labeled Sak with plasminogen immobilized on a CNBr-Sepharose CL-4B matrix.

NMR spectra were recorded at 300 K on a Varian INOVA750 MHz four channel NMR spectrometer equipped withpulse field gradient accessories and a triple resonance probewith an actively shielded Z gradient coil. The NMR data wasanalyzed with the program XEASY (14) on Silicon GraphicsINDY & INDIGO2 workstations. Chemical shifts were refer-enced as described previously (12). Unless indicated, the *Hand I5N carriers were set at 4.74 ppm and 119.5 ppm respec-tively. For both samples a 15N-edited HSQC and a 3D 15N-edited NOESY-HSQC (15,16) experiment were carried outemploying a mixing time of 50 ms. Heteronuclear ^ N ^ H } -NOE measurements were performed according to thesequence of Kay et al. (17). The steady-state heteronuclearNOEs were calculated by normalizing the difference betweenthe cross peak integrals in the presence and absence of the^-saturation with the integral of the latter.


As can be seen from the ^N-'H HSQC spectra shown inFigure 1, only minor differences were observed in the 15Nand *HN chemical shifts of Sak and SakANIO thereby facili-tating spectral assignments. Most of the differences in thechemical shift values lie within the limits defined by thespectral resolution. While the complete assignment of theHSQC cross peaks in Sak has already been reported (11,12)in Figure 1 (left panel) only those peaks experiencing minorchemical shift changes in SakANIO are indicated in additionto a few N-terminal residues. The removal of the first ten res-idues unavoidably disturbs chemical shifts in the residual N-terminal section between Asp 13 and Tyr24. Only for residuesVal45 and Glu65 a consistent difference in both 15N and *Hchemical shifts could be observed. In addition, protons ofresidues Tyr63, Leu68, Tyr73 and Glu75 in the a-helix andthe adjacent loop are shifted besides Aspll5, Hel20 andLysl21. The latter reside in a loop before a P-strand and


Sak

G7 •

G12»

• •

G22 fr •

10

15N [ppm]

110.0

115.0

120.0

125.0

130.0

SakANIO

•

• ••

* J* 4

.4*-* • •

•

•

« •• •

• •

• • •

• # • • •

«, ••

10 9 8H [ppm]

1: N-'Figure 1: N-'H HSQC spectra of full-length (left) and N-terminal processed staphylokinase (right).

come close to each other (12).N{ H}-heteronuclear Overhauser enhancement experi-

ments were performed to obtain an overview on possiblechanges in the dynamical features. This data, presented in

Figure 2, reveals no major dynamical differences of the twospecies.

The characteristic proton-proton NOE pattern in the oc-helical and P-sheet regions seen in full-length Sak (11) have

I i

ffli in

-0.2-1

| -0.4J

-0.6-

-0.8-

Sak SakANIO

20 40 60 80 100 120 20 40 60 80 100 120

Figure 2: 15N{ *H}-NOE data for full-length and N-terminal processed Sak as a function of the residue number.

Vol. 20, No. 1-4 37

E58 K I Y Y V E W66

4.00

5.00

6.00

7.00

8.00-

9.00

ppm 8.60 7.43 8.35 8.69 7.48 7.88 8.48 8.34 8.43

Figure 3: Strip plots of the 15N-edited NOESY-HSQC ofSakANlO. a-Helical NOE connectivities seen between Glu58and Tip66 are also indicated. Peaks only present at a lowerthreshold are marked by boxes.

also been observed in the 15N-edited 3D NOESY-HSQC ofthe N-terminal processed Sak. Figure 3 displays a part of thehelical NOE connectivities in SakANlO. But for the occur-rence of an additional cross peak between the amide protonsof Val27 and Val45 located in strands I and II (for numberingof the strands see ref. 11), the paucity of NOE contacts forthe bulge region between strands II and III is also noticed inSakANlO (data not shown).

The essential activation step in the proteolytic processingof Sak is the cleavage of the conformationally labile N-termi-nal residues Serl-LyslO involved in interactions with resi-dues 40-46. The similarities in the chemical shifts, dynamicalfeatures as judged by heteronuclear NOEs and identical NOEpattern observed in Sak and SakANlO indicate almost identi-cal overall solution structural characteristics of the two spe-cies. The global fold of SakANlO is also similar to therecently published X-ray structure of a proteolyticallydegraded Sak (18; residues 16-136), although a detailed anal-ysis of structural differences will need a l3C,15N-labeledsample.

The removal of residues 1-10 allows access to the bulgeresidues Leu40-Glu46 in a direction parallel to the cc-helixaxis with an increase of the mean solvent accessible surfaceby more than a factor of 2 as deduced from molecular model-

ling (12). Two of three mutation sites (Pro48 and Ala67)identified by the alanine scan (19) lead to inactive mutantswith impaired plasminogen binding. Other single sitemutants of staphylokinase (20) showed that charged residuesaround position 68 of the helix and an intact C-terminalstructure were necessary for activation of plasminogen.Exactly in these region around amino acids 48 and 67 the res-idues with the highest chemical shift differences betweenfull-length and N-terminal shortened staphylokinase areobserved. Both of the sites are shielded from surface expo-sure by the flexible N-terminal region of the staphylokinasestructure and show NOE contacts to that region. The thirdsite identified by the alanine scan covered the cleavage site atLyslO and led to a Sak mutant which retained binding, butwas incapable of activation. This emphasizes the importanceof demasking a lysine residue by the cleavage betweenLyslO-Lysl 1 while the global 3D structure remains virtuallyunchanged.

Further work focussing on the plasmin-bound solutionstructure of staphylokinase is in progress.

Acknowledgements

This work has been supported by the Fonds der Che-mischen Industrie. The authors thank Mrs. B. Pohle, Mrs. I.Tiroke, Mr. J. Leppert and Mr. M. Semm for excellent techni-cal assistance.

References

'D. Collen and H.R. Lijnen, Blood 78, 3114-3124 (1991).2W.J. Brockway and F.J. Castellino, Biochemistry 13,

2063-2070 (1974).3B. Schlott, K.-H. Guhrs, M. Hartmann, A. Rocker and D.

Collen, J.BioLChem. 272,6067-6072.(1997).4D. Collen, Circulation 72, 18-20 (1985).5C. Zamarron, H.R. Lijnen, B. Van Hoef and D. Collen,

Thromb.Haemost. 52, 19-23 (1984).6D. Collen, Nature Medicine 4,279-284 (1998).7H.R. Lijnen, B. Van Hoef, F. De Cock, K. Okada, S.

Ueshima, O. Matsuo and D. Collen, J.BioLChem. 266,11826-11832(1991).

8K. Silence, D. Collen and H.R. Lijnen, J.BioLChem.268,9811-9816(1993).

9D. Collen and F. VandeWerf, Circulation 87, 1850-1853(1993).

10S. Vanderschueren, L. Stockx, G. Wilms, H. Lacroix, R.Verhaeghe, J. Vermylen and D. Collen, Circulation 92,2050-2056 (1995).

"O. Ohlenschlager, R. Ramachandran, J. Flemming, K.-H. Guhrs, B. Schlott and L.R. Brown, J.Biornol.NMR 9, 273-286(1997).

I2O. Ohlenschlager, R. Ramachandran, K.-H. Guhrs, B.Schlott and L.R. Brown, Biochemistry 37, 10635-10642(1998).

13B. Schlott, M. Hartmann, K.-H. Guhrs, E. Birch-Hir-


schfeld, H.-D. Pohl, S.B. Vanderschueren, F. Van de Werf, A. 18A. Rabijns, H.L. DeBondt and C. De Ranter,Michoel, D. Collen and D. Behnke, Biotechnology 12, 185- Nature Struct.Biol. 4, 357-360 (1997).189(1994). 19K. Silence, M. Hartmann, K.-H. Gtthrs, A. Gase, B.

14C. Bartels, T. Xia, M. Billeter, P. Guntert, P. and K. Schlott, D. Collen and B.R.U}nen,J.Biol.Chem. 270,27192-W\iihnch,J.Biomol.NMR 6, 1-10(1995). 27198(1995).

15S.W. Fesik and E.R.P. Zuiderweg, J.MagruReson. 78, 20A. Gase, M. Hartmann, K.-H. Guhrs, A. Rocker, D.588-593 (1988). Collen, D. Behnke and B. Schlott, Thromb.Haemostasis 76,

16E.R.P. Zuiderweg and S.W. Fesik, Biochemistry 28, 755-760(1996).2387-2391 (1989).

I7L.E. Kay, D.A. Torchia and A. Bax, Biochemistry 28,8972-8979(1989).

Vol. 20, No. 1-4 39

EPR of Incommensurate Phases in Ferroelastic MgSiF6* 6H2O : Mn2+ :Lineshape Simulation with Variation of the Spin-Lattice Relaxation Rate

P.G. Skrylnik and A.M. ZiatdinovInstitute of Chemistry, Far Eastern Branch of the Russian Academy of Sciences,

159, Prosp. 100-letija, Vladivostok 690022, Russia, e-mail: [email protected], Fax:(4232)311889

The availability of angular dependence of the EPR spectra line splitting indicates that the angleof complex ions Mg[H20]6 disorientation around the C3 - axis may be chosen as a primary or-der parameter for the phase transition from paraelastic phase to incommensurate phase. Thischoice of order parameter determines the presence of only even terms in expanding of the EPRspectra fine structure parameter on the powers of order parameter. A successful description ofexperimental lineshapes may be reached under the following additional assumptions: 1) solitonsexisting within the wide temperature range; 2) variation of the spin - lattice relaxation rate Tf1

over the spectral distribution; 3) presence of nearly independent subsystems of solitons anddomains within the temperature range from 300 K to 310 K.

IntroductionMagnesium fluorosilicate hexahydrate (MFSH)belongs to isomorphous compounds of the typeABF6»6H2O (where A and B are di- andfourvalent metals, respectively) in which twocomplex ions A[H20]6 and BF6 can bedistributed between two orientations around the3-fold axis [1,2]. A number of these compoundsare characterized by an improper ferroelasticphase transition from a rhombohedralmodification to a monoclinic phase, which isstable at low temperature [2]. According to thecrystallographic studies [3] for MFSH thespace groups are R3m and P2i/c, respectively.The presence of intermediate states betweenmentioned phases in crystals MgBF6«6H2O(where B - Si, Ge, Ti) have been established andinvestigated by various methods includingelectron paramagnetic resonance (EPR) [4-8].However, the nature of this phase is notunderstood finally. In this paper we present thenew approaches to describing the mentionedEPR experimental data on MgSiF6«6H2Ocrystals and the ideas concerning motifs ofintermediate phase crystal structure.

ExperimentalThe single crystals of improper ferroelasticMgSiF6*6H2O have been investigated by means

of X - band EPR on admixture ions Mn2+.At Tii s370K and Tc s299K the crystalsundergo phase transitions from paraelastic phaseto incommensurate phase and fromincommensurate phase to ferroelastic phase,respectively [4]. The temperature decreasebelow Tii is accompanied by smoothinhomogeneous broadening of the Mn2+

hyperfine structure lines followed by splitting ofthese lines (Fig.l). Moreover, the EPR lineshapeis typical for incommensurate one-dimensionalmodulated systems. The evidences for theincommensurate structure of MFSH wereobtained by optical studies [6] too. Theinhomogeneous phase similar to that inmagnesium fluorosilicate have been observedin mixed single crystals MgxZni.xSiF6»6H2O. Itis worth mentioning that this phase forms at x «20%, indicating on a special role of Mg2+ ions inthe incommensurability emersion.

Results and Discussionhi crystals under investigation the lineshape ininhomogeneous phase is formed mainly by themodulation AD of the fine structure parameterD. However, the presence of angular dependenceof EPR lineshape shows that the angle <p ofcomplex ions orientation around crystal C3 axismust be a primary order parameter [5]. This


statement seems to be physically reasonable andqualitatively accounts for the principalexperimental data. Further, we suppose thatparameter D is connected with <p. For thereasons of symmetry, we should conclude thatAD = AD (cp2) (quadratic case for AD ~ cp2).

T = 369K

T = 301 K

Fig. l . The temperature evolution of the EPR Mn2+

spectra lineshape (scattered points correspond toexperimental spectra, solid line - theoretical simulatedspectra).

Experimental Mn2+ EPR lineshape has beeninterpreted in the terms of model analogous tothat of Blinc [9] for the interpretation of

magnetic resonance spectra of crystalincommensurate phases. The resonance field ofa given paramagnetic centre was expanded inpowers of order parameter (holding even termsup to fourth power), soliton density dependingon temperature has been taken into account.Multiplicity parameter of the superstructure pwas chosen to be equal to 3 according to theRaman spectroscopy data above Tc for relatedcrystals [10].

Basing on these assumptions we calculatedthe temperature dependence (Fig. 2) ofmodulation parameter h.2 (ha ~(Tn - T)2p) whichallows to determine the critical index p = 0,36 ±0,02 of a transition to incommensurate phase.This value of P is a typical magnitude for thesystem with one-dimensional incommensuratemodulation and close to p = 0,345 ± 0,002 forthe 3d-XY model [11]. This fact may beconsidered as indirect evidence for thecorrectness of our order parameter choice.

-10290 300 310 320 330 340 350 360 370

Fig. 2. The temperature dependence of incommensuratemodulation parameters of the second (h2) and fourth (h4)order calculated from the experimental spectra.

The calculations show that smooth evolutionof incommensurate phase follows to the predic-tions of a classical theory [9] between Tu and TJ2(T;2 = 343K) : the modulation follows from aplane-wave regime to multisoliton regime withdecreasing soliton density ns at T ->Tj2 (Fig. 3);Below Tj2 the spectra lineshape changes differessentially from magnetic resonance spectra

Vol. 20, No. 1-4 41

evolution in conventional incommensurate sys-tems with one-dimensional modulation (Fig. 1).The calculations performed in particular showthat ns undergoes a step-wise decrease at Tato a small value about« 0,1 (Fig. 3).

0.0290 300 310 320 330 340 350 360 370

Fig. 3. The temperature dependence of soliton density n$for the model taking into account variation of Tf1 over thespectrum.

0 '-

-5290 300 310 320 330 340 350 360 370

Fig. 4. The temperature dependence of the parametersWo, W2, and 6W. Parameters Wo and W2 indicate thevalue of Lorentzian linewidth at the points of incommen-surate spectral distribution singularities, 8W is a deviationfrom linear function in a proposed parabolic dependenceW(H).

A successful description of the lineshape(Fig. 1) may be in principle obtained between T-aand Tc + 10K taking into account the variationof spin - lattice relaxation rate Tf * over the in-commensurate spectral distribution. This phe-nomenon was predicted theoretically and ob-served experimentally by direct Tf1 measure-ments in some compounds [9, 12]. The reasonfor such variation may be different contributionsof amplitudon and phason fluctuations to theTf at different parts of the inhomogeneousmagnetic resonance lines [9, 12]. However, be-low T« Tc + 10K the value of variation indi-cated becomes too large (Fig. 4).

Therefore, we have proposed and examinedanother reason for EPR spectra evolution belowTi2. We supposed the reversible crystal divisioninto subsystems of domains and soliton-like ar-eas at this temperature. The model allows to de-scribe rather satisfactorily the experimentalspectra with two distinct components from T;2to Tc. It is notable that soliton-like compo-nent lineshape is typical for multidimensional2q (or 3q) incommensurate modulation (absenceof edge singularities) rather than one-dimensional case at T > Ta. The step-wisechanges in parameters describing EPR spectrumlineshape at Ta may be attributed to the presenceof nearly independent aforementioned subsys-tems below TJ2. The soliton density ns is almostinvariable within some temperature range. How-ever, the interaction between these subsystemsprobably restores at T -» Tc resulting in solitondensity ns decrease (close by Tc) followed by atransition to commensurate ferroelastic phase.

ConclusionThe results presented show that the EPR spectraof Mn2+ in MgSiF6»6H2O intermediate phasemay be qualitatively described as incommensu-rate system spectra. On the other hand, the tem-perature evolution of experimental spectra hassome peculiarities. They, in principle, may beaccounted for under the next additional assump-tions: structural solitons existing and presence ofspin - lattice relaxation rate Tf1 variation over


the incommensurate spectral distribution si-multaneously. However, taking into account toolarge values of these variation parameters nearTc, we suppose, as alternative model, thepresence of two nearly independent subsystemsof soliton-like areas and domains. All presentedresults give evidence for non-trivial structure ofMgSiFg*6H2O crystals in its intermediate phaseand, we hope, will stimulate further investiga-tions of this compound.

AcknowledgementsWe would like to thank T.F. Antohina for thegrown MgSiF6«6H2O single crystals andV.G. Kuryavyi for presenting some experimen-tal EPR spectra.

References[1] W.C. Hamilton, Acta Cryst., 15, 353 (1962).[2] D.C. Price, Can. J. Phys., 65, 1280 (1987).[3] S. Syoyama and K. Osaki, Acta Cryst., B28,

2626-2627 (1972).[4] A.M. Ziatdinov , V.G. Kuryavyi and

R.L. Davidovich, Sov. Phys. Solid State, 27,1288-1290 (1985).

[5] A.M. Ziatdinov and V.G. Kuryavyi, Fer-roelectrics, 143, 99 - 107 (1993).

[6] R. Hrabanski , V. Kapustianik, V. Kardashand S. Sveleba, Phys. Stat. Sol. (a), 142,509(1994).

[7] R. Hrabanski and A. Kassiba, Ferroelec-trics, 172,443 - 448 (1995).

[8] M.Suhara, T.Bandoh, T.Kitai, T.Kobayashiand H.Katsuda, Phase Transitions, 37, 111-119(1992).

[9] R. Blinc, P. Prelovsek, V. Rutar, J. Seligerand S. Zumer, in Incommensurate Phasesin Dielectrics, edited by R. Blinc and A.P.Levanyuk, North - Holland, Amsterdam,143-277 (1986).

[10] V.V. Eremenko, A.V. Peschanskii and.V.I. Fomin, Solid State Physics (Russia),39,929-939 (1997).

[11] J.C. Le Guillon and J. Zinn-Justin, Physi-cal Review, B21, 3976 (1980).

[12] Blinc, T. Apih, J. Dolinsek, U. Mikac,D.CAilion and P.-H. Chan, Physical Re-view, B51, 1354-1357, (1995).

Vol. 20, No. 1-4 43

NMR Studies of the Halotolerant Bacterium Holomonas Israelensis:Sodium-23, Cesium-133 and Phosphorus-31 studies.

Haggai Gilboa and Asad SakhniniDepartment of Chemistry, TECHNION- Israel Institute of Technology, Haifa 32000, Israel

Holomonas Israelensis ( previous name used wasBa^ were isolated from the evaporation pools ofthe Dead Sea. They can grow on various saltconcentrations from 0.2M-4M NaCl. The optimalgrowth concentration was found to be 0.8M [1].Measurements have shown that the bacteriaaccumulate the so called "compatible solutes",compounds that retain the high osmotic pressurebut do not interfere with cell function.Holomonas Israelensis accumulate glycogen-betaine or ectoine when grown on high salt media[2]. The adaptation of halotolerant bacteria to thechanging growth conditions was the subject ofmany studies [3-5]. In this research we havestudied various parameters that influence thegrowth of the bacteria and tried to learn about theadaptation of the bacteria to the various growthconditions. The nmr measurements of the variousnuclei enlighten different aspects of theadaptation process.ExperimentalHolomonas Israelensis bacteria were grown at37°C on a medium containing nutrient broth(Difco) as described elsewhere [2] they wereharvested at the stationary phase. The bacteriawere resuspended in solution according to thenmr experiment to be performed.All nmr studies were carried out on a BrukerAM400 WB, the nmr frequencies were 161MHz,105.8 MHz and 52 MHz for phosphorus sodiumand cesium respectively using multinuclei probes,31P studies were carried out on bacterialsuspensions introduced to a 20mm glass tube.Aerobic conditions were obtained by using a'double bubbler' apparatus similar to that used byGadian et. al. [6]. The phosphorus chemical shiftswere calibrated against 85% H3PO4 serving as anexternal reference.

Sodium-23 nmr signals in the cytosol wereobtained by using the shift reagent dysprosiumtripolyphosphate to separate between the intra-and extracellular sodium. The sodium nmr studieswere carried out on resting cells placed in a10mm tube, without nutrition or oxygen. Singlequantum and double quantum filtered spectrawere measured [7].Cesium-133 spectra were measured using aspecial perfusion device where we had control onthe oxygen and the nutrition in the sample [8].There was a good separation of the intracellularcesium from the extracellular without the use ofshift reagents.

Results and DiscussionPhosphorus-31 nmr. The 31P spectrum of intactcells in aerobic conditions with enough nutritionis presented in Figure 1. HEPES was used as thebuffer. Intracellular pH values were determinedfrom the inorganic phosphate chemical shifts. For31P studies bacteria were grown on 0.8M total saltconcentration and were resuspended at the samesalt concentration. The biological active 31Pcompounds were detected under aerobicconditions. The nmr signals were assigned withthe aid of the bacterial extract. The spectrum ofthe extract is presented in figure 2. Changing thepH levels of the solution and comparing the shiftsand their pH dependence to the literature, theassignment was accomplished [6]. The presenceof ATP, ADP and AMP was also proved byHPLC. Only in the aerobic extract ATP wasdetected. Figure 3 show the spectrum of ananaerobic suspension in HEPES buffer, no ATP isdetected under these conditions.


20 10 0 -10 -20 -30PPM

20 10 -10 -20 -30

PPM

3 ITFigure 1. P spectrum of intact cells at aerobic conditions.The medium contains nutrient broth.

20 10 0 -10 -20 -30PPM

Figure 2.3IP spectrum of the aerobic bacteria extract.

Figure 3. 3IP Spectrum of bacteria of bacteria, suspended inHEPES buffer, at anaerobic conditions.

When the bacteria were grown on HEPES bufferwith small amounts of inqrganic phosphate (Pi) inthe medium, it was possible to distinguishbetween the intracellular and extracellular Pi andto obtain the intracellular pH values. The signal at~10ppm is attributed to pyrophosphate in thecells. On changing the extracellular pH from 6 to8, the pH in the cytosol of the bacteria remainedconstant 6.3-6.4, in aerobic conditions and inresting cells, as well. When the buffer solutionwas phosphate buffer it was not possible to detectthe intracellular phosphate. A strong line of the Piin the medium was detected, it covered the smallintracellular signal. Bacteria suspended inphosphate buffer and in aerobic conditionswithout nutrition show accumulation ofpolyphosphates, as shown in figure 4 the signal at~23ppm is attributed to soluble polyphosphte.When the cells were in anaerobic conditions the

Vol. 20, No. 1-4 45

polyphosphate signal disappeared and a spectrumsimilar to that in figure 3 was measured, this wasa reversible process, on introducing oxygen againthe polyphosphate signal reappeared. Since nonutrition are present and the cells in aerobicconditions produce polyphosphate from theinorganic phosphate it seems as if the cells arefunctioning. Here the polyphosphate may serveas the energy source instead of ATP.

iO -10 -20 -30 -40PPM

Figure 4. Spectra of bacteria suspended in phosphate bufferin aerobic conditions without nutrient broth.

Sodium -23 nmr. Single quantum (SQ) anddouble quantum (DQ) sodium nmr spectra weremeasured. From the single quantum studies theintracellular sodium concentration was calculated[9]. In the range between 0.2M NaCl in themedium and up to 1.2M NaCl, the intracellularconcentration was almost constant -0.2Msodium. Relaxation times were measured onresting cell at two media concentrations. Tirelaxation times were obtained from the 180-T-90pulse sequence. From the DQ filtered spectrumthe relaxation times T^ T2f and T2DQ of thesodium in the cytosol were calculated [7]. Table 1represent the results obtained for the intracellularrelaxation times in the cytosol of the bacterium.From T2s, T2f, and T2OQ , the correlation times tc

for sodium ions in the two growth media were

calculated, they are presented in the last columnof the table.

Table 1The relaxation times of sodium in the cytosol of holomonasIsraelensis at two media concentrations. The values weredetermined from SQ and DQ nmr studies. Calculated TC

values are presented in the last column.

NaClM0.40.8

/msec6.9+1.18.6±1.2

T*/msec4.6±1.03.5+0.7

T2f

/msec-35±.l.6±-2

T2D0/msec.49±.O6.87±,2

Tc/10"*sec3.311.12.0+.8

The relaxation times measured show that Ti >T2s

» T2f wT2D0. The values of tc for the sodium ionsare in the range of 10" sec compared to -10 inaqueous solutions. This indicate that the sodiummotions in the cytosol are restricted. However thevalues of xc in the two concentrations studied arevery close and within experimental error. Thisresult may indicate that the bacteria try to keep aconstant environment in the cytosol regardless ofthe extracellular medium.

Cesium-133 nmr. In bacteria grown on a mediumcontaining CsCl, or for bacteria where CsCl wasadded to the medium after harvesting, there was avery good separation between the intracellularand extracellular cesium [8]. Only at 0.4M saltconcentration in the medium the shift was verysmall. Cesium chemical shifts in the cytosol ataerobic conditions are presented in Figure 5. Thechange in chemical shift of the cesium signal inthe cytosol is moderate and much smaller thanthat of the signal in the medium. This result maypoint that when there are drastic changes in thesalinity of the medium, the cells producecompounds that their effect on the cesiumchemical shift is much smaller. It may alsoindicate that the cells try to keep the intracellularenvironment unchanged.Cesium intensity studies show that theintracellular cesium concentration was «100mMwhere the CsCl concentration in the medium wasonly 25 mM. Influx and efflux occurred only inaerobic samples containing enough nutrition. Theresults indicate active transport of cesium into the


cell. Probably replacing potassium by the relevantpotassium pumps.

I

salt concentraion (M)

Figure 4. Chemical shifts of the intracellular (•) andextracellular (•) cesium ions as a function of mediumsalinity. The chemical shifts were calibrated against 50mMCsCl that served as an external reference.

Influx and efflux of cesium were studied. It wasdemonstrated again that the cesium enters the cellby active transport replacing the potassium ions.A typical graph for the influx of cesium into thecell is presented in figure 5.

•3OO 0 200 600 SCO 1000 1200 1400 1&

Time(minutes)

Figure 5. Influx of cesium into the cell as a function of time.CsCl concentration in the medium 25mM the total saltconcentration 0.8M.

ConclusionsThe nmr study of the halophilic halotolerantbacterium holomonas Israelensis show some ofthe adaptation properties of that bacterium. Thebacteria adapt to the various properties of thedifferent media by keeping a constant pH level inaerobic and anaerobic media. The intracellular

sodium concentration is also unchanged and ismuch lower than the extracellular sodium. Thesodium relaxation times indicate that the dynamicproperties of the sodium in the cytosol areunaffected by the varying conditions. The resultsobtained from the cesium nmr measurements alsoindicate an adaptation capacity of holomonasIsraelensis to the varying growth conditions. Italso may indicate that the cesium nucleus mayserve as a tool for studying potassium pumps.

references1. Rafaeli-Eshkol, D., and Avi-Dor, Y. Biochem.J. 109,679-685 (1968).2. Regev, R., Peri, L, Gilboa, H., and Avi-Dor, Y.Arch. Biochem. Biophys. 278,106-112 (1990).3. Huval, J. H., Lotta, R., Kushner, J. D. andVreeland, R. H. Can. J. Microbiol. 41, 1124-1131(1995).4. Brown, A. D. Bacteriol. Rev. 40, 803-846(1976).5. Forsyth, M.P. and Kushner, J. D. Can. J.Microbiol. 16, 253-261 (1970).6. Gadian, D. G., Radda, G. K., Richards, R. E.,and Seeley, P. J. in "Biological Applications ofMagnetic Resonance" R. G. Shulman Ed.,Academic Press, Inc., New York, (1979).7. Sskhnini, A., and Gilboa, H. Biophys. Chem.46, 21-25 (1993).8. Sakhnini, A., and Gilboa, H. NMR Biomed. 1180-86 (1998).9. Gilboa, H., Kogut, M., Chalamish, S., Regev,R., Avi-Dor, Y. and Russell, N. J. J. Bacteriolog.173,7021-7023(1991).

Vol. 20, No. 1-4 47

Fluorine multiquantum coherences as indicators of the line-width transition in

CF3COOAg

A. Kaikkonen, E.E. Ylinen and M. Punkkinen.

Wihuri Physical Laboratory, Department of Physics, University of Turku, FIN-20014, Finland

ABSTRACTFluorine 3Q and 6Q coherences were followedas functions of temperature in a single crystal ofCF3COOAg. When the external magnetic fieldmakes the magic angle with normal to thefluorine plane of CF3 groups, both the signalsare weak down to 130 K. Around thistemperature the signal amplitudes experience asteplike increase which is indicative of slowingdown of CF3 reorientations. The fluorine secondmoment was observed to vary similarly withtemperature.

INTRODUCTIONThe temperature dependence of the

amplitude of a multiquantum coherence can beused as a tool for investigating dynamicproperties of symmetric molecular groups insolids. Particularly, observation of the six (6Q)and three (3Q) quantum coherences in twoproximate coaxial methyl groups can giveinformation about the transition region, wherewith lowering temperature the fast-reorientinggroups become practically static. This method isan alternative to observe the so-called line-widthtransition temperature.

Excitation of multiquantum coherencescan be achieved by selective or non-selective RFpulses (1,2). Non-selective or strong pulses areusually applied on systems where the RF fieldcan be made considerably larger than theinteractions dominating the line width. Toexcite multiquantum coherences in systems withstrong interactions one has to use selectivepulse(s) at multiquantum transition frequencies(3-6).

In the present work we employ thepulse sequence (jc/2)a - xxl2 - (jt)p - i\l2 - (id2)§ -evolution - (n/2)T - xjl - (n)y - x2/2 - detection inorder to observe 3Q and 6Q coherences in thesingle crystal of CF3COOAg. The appropriatecoherence transfer pathways were separated byphase cycling, which consisted of 12 phaseincrements with a = 3 for the 6Q coherence andof 6 phase increments with a = 3 + jt/2 for 3Qexperiments (1,2). Temperature dependence of

the 3Q and 6Q signals was followed from 290 Kdown to liquid nitrogen temperature, also thesecond moment was observed.

We chose a single crystal of CF3 as oursample, since in it the CF3 groups have theirfluorine planes parallel to each other (8).Besides, the CF3 groups are arranged inproximate pairs, which causes a splitting of thecentral part of the spectrum (7). Griffin et. al(8,9) determined the fluorine chemical shifts bycoherent averaging techniques. Also, thefluorine spin-lattice relaxation and spindiffusion have been studied (10).

THEORYThe theoretical description of

trifluoromethyl groups in a single crystal ofsilver trifluoroacetate can be split into two parts.Above the transition temperature of 130 K (8,9)the CF3 groups reorient with a frequency greaterthan the frequency with which their position ismeasured in the NMR experiment, while belowthis temperature these groups can be treated asstatic.

The secular part of the Hamiltonian forthe dipolar interaction of the fluorines in twoneighbouring static CF3 groups can be written

[1]

whereBy - (HO/4JI) u [1 -3cos20y]

4 4

Under fast reorientation of triads the effectivepart of Hd can be expressed in a symmetryadapted fashion. Its fully symmetric part equals,

where f © and 01)

[2]

contain interactionsbetween the nuclei within a triad (intra) andHa0'1" contains those between the nucleibelonging to different triads (inter). The


following equalities express these parts ofHamiltonian (in units of ft.):

[3]

with the intra-dipolar shift

ooa =[(no/4jt)3Y2 ft fcr^Qcas2*® -1)=Dr(3cos2ea)-1)

Here r is the nucleus-nucleus distance in thetriad and 9® the angle between its threefold axisand the field Bo . Similarly the inter part of theHamiltonian is

-4

[4]

The operators are defined by the equations likeIz0) = Ikz +112+ I™ and Lai) = Iu. + Iv.+ Iw., wherek,l,m and u,v,w refer to the fluorines in the CF3

groups I and n, respectively. The inter-dipolarshift is given by

3](l-3cos20R)=DR(l-3cos2eR)

where R is the average distance between thenuclei belonging to the neighbouring triads and9R is the angle between Bo and Ro , the axisjoining the centres of the triangles. The value ofR, averaged over the orbits of the apices of thetwo triangles, can be calculated numerically (7).

For coaxial triads both the inter andintra parts of Hd(s) vanish if the angle betweenthe external field Bo and the axis joining thecentres of triangles is 54.7° . Thus it ispractically impossible to excite multiquantumcoherences above the line-width transitiontemperature in the CF3COOAg crystal orientedin such a way. Multiquantum coherences relatedto interactions between fluorines of more distantCF3 groups can still be observed but they areweak and require longer delays between the RFpulses. However, under cooling the CF3reorientation slows down and Hd(s) no moreaverages to zero. Then one can observe thesignals originating from multiquantumcoherences below the line-width transitiontemperature.

Silver trifluoroacetate crystallises in themonoclinic system with four dimers per unit cell(8). The opposite CF3 groups of a dimer form acoaxial pair along the a axis with RQ = 0.30 nmand r = 0.21 nm, which give R = 0.35 nm, R/r

= 1.67 (10), D/2re = (Po/47i) yfhltf- 4.31kHz and DR/2jt - (jio/4rt) f ft /R3= 2.48 kHz.

Numerical calculations were performedfor the excitation and detection of 3Q and 6Qcoherences for static and fast-reorientingmolecules. Simulations were performed withSun4d Sparc computer using the C++ NMRlibrary GAMMA (11,12).

RESULTS AND DISCUSSIONExperiments were carried out by using

the Bruker MSL 300 pulsed NMR spectrometeroperated at the fluorine resonance frequency35.5 MHz. The single crystal of CF3COOAgwas oriented at room temperature in such a waythat 9R = 9 ~ 54,7° which corresponds to thenarrowest spectrum. We used the maximalavailable RF field strength v, = yBi/2Jt = 88.7kHz in all experiments.

DrT,,D,T2

Fig.l. 6Q coherence after the three-pulse excitationversus DtT, or the 6Q related detected signal at theend of the pulse sequence versus Drt2 (—); 3Qcoherence after the three-pulse excitation versus D,T,for isolated triads (—); and the 3Q related detectedsignal at the end of the pulse sequence versus D,t2

(-•-•-) with Drx, = 0.87 corresponding to the maximalsignal. R/r- 1.67 and 6 R - 9-54.7°.

Fig.l. represents 6Q and 3Q signals forstatic CF3 groups as a function of the pulseintervals DrXi and Drx2 in the excitation ordetection sequences, respectively, for 9R = 9 =54,7° and R/r = 1.67. Because signals decay veryfast at low temperatures we used the firstmaxima which correspond to Xi = x2 = 40.6 usfor 6Q and Ti = x2 = 32.3 us for 3Q coherences.According to our calculations the excitation of6Q coherence by the (7r/2)a - Xi/2 - (jt)p - ti/2 -(jr/2)p sequence has the same behaviour as itstransformation to the observable 1Q signal by(71/2), - T2/2 - (ri)y - T2/2 if co, » Dr, DR.Therefore the excitation and detection of the 6Qcoherence are described-by the single curve inFig. 1.

Vol. 20, No. 1-4 49

0.6

0.4

0.2

0.0

-0.2

-0.4

-0.6

50 52 54 60

Fig.2.6Q coherence after the three-pulse excitationversus 6R for R/r= 1.67, D,T| = 1.1 and 8R = 6. Fast-reorienting methyl groups (—), static groups (—).

In practice it is impossible to orient thecrystal in such a way that 6R and 8 are exactlymagic. Therefore it is important to consider inwhich range around the magic angle the 6Q and3Q signals vanish for fast-reorienting CF3

groups. The dependences of the excited 6Qcoherence and of the detected 3Q-related signalon the crystal orientation are shown on figures 2and 3, respectively.

2.5

2.0

"TO

& 1.5O« 1.0•gI 0.5&

0.0

-0.5

1

1//

y\\ /

20 40 60 80 100

Fig.3. 3Q related 1Q signal at the end of pulsesequence versus 6R for R/r = 1.67, Drx, - Drt2 = 0.87and 6R= 0. Fast-reorienting methyl groups (—), staticgroups (—).

The usual way to detect the line-widthtransition temperature is to follow the secondmoment or line-width of the spectrum. In Fig.4open triangles represent the second moment as afunction of temperature. The second momentincreases rapidly with decreasing temperaturearound 130 K which is in accordance with theresults of M. Mehring et al. (9). The shape ofthe spectrum changes only within the transitionregion, but remains constant above and below it.

Our results show that the 3Q and 6Qsignals can alternatively be used as indicatorsfor fast reorienting or static trifluoromethylgroups. In Fig.4 the circles and squaresrepresent experimentally observed 3Q and 6Q

coherences, detected at the end of the pulsesequence as 1Q signals. The multiquantumcoherence amplitudes are seen to follow thesecond moment fairly well.

120 160 200 240Temperature (K)

Fig.4. 3Q (squares) and 6Q (circles) related,experimentally detected 1Q signals as functions oftemperature. For comparison the second moment data(open triangle) are also shown. 6R= 6 = 54.7°.

ACKNOWLEDGEMENTSAuthors thank A.H.Vuorimaki and P.Malmi forhelpful discussion.

REFERENCES1. R.R. Ernst, G. Bodenhausen, and A. Wokaun:"Principles of Nuclear Magnetic Resonance inOne and Two Dimensions", Claredon Press,Oxford (1987).2. G. Bodenhausen, Prog, in NMR Spectroscopy14,137 (1981).3. J.P. Amoureux, C. Fernandez, and L .Fryd-man, Chem. Phys. Lett 259,347 (1996).4. S.P. Brown, SJ. Heyes, and S. Wimperis, J.Magn. Reson. 119,280 (1996).5. A. Kaikkonen, E.E. Ylinen, and M. Punkki-nen, Solid State NMR 10,129 (1998).6. A. Kaikkonen E.E. Ylinen, and M. Punkki-nen, to be published in J. Magn. Reson.7. A.H. Vuorimaki and M. Punkkinen, J. Magn.Reson. 91,539 (1991).8. R.G. Griffin, J.D. Ellett, Jr., M. Mehring, J.G.Bullitt, and J.S. Waugh, J. Chem. Phys. 57,2147 (1972).9. M. Mehring, R.G. Griffin, and J.S. Waugh, J.Chem. Phys. 55,746 (1971).10. A.H. Vuorimaki, Chem. Phys. Letters 202,253 (1992).11. S.A. Smith, T.O. Levante, B.H. Meier, andR.R. Ernst, J. Magn. Reson. 106,75 (1994).12. http://gamma.magnet.fsu.edu/index.html

50 Bulletin'of Magnetic Resonance

Calendar of ForthcomingConferences in Magnetic

Resonance

May 22-28, 1999Seventh Scientific Meeting and Exhibition of theIntl. Soc. for Magnetic Resonance in Medicine(ISMRM), Philadelphia, Pennsylvania (USA)

For information contact:

ISMRM2118 Milvia St.Suite 201Berkeley, CA 94704

June 6-11, 1999Gordon Research Conference on: ComputationalAspects of Biomolecvlar NMR, Conference Center"II Ciocco", Barga (Italy)


Gordon Research ConferencesUniversity of Rhode IslandP.O. Box 984West Kingston, Rhode Island 02892-0984 USAPhone: (401) 783-4011Fax: (401) 783-7644E-mail: [email protected]: [email protected] (send/requeste-mail application)Registration http://www.grc.uri.edu

May 25 - June 5, 1999International School of Structural Biology andMagnetic Resonance, 4th Course: Dynamics,Structure and Function of BiologicalMacromolecules, Erice, Sicily (Italy)


Ms. Robin HolbrookStanford Magenitc Resonance LaboratoryStanford UniversityStanford, CA, 94305-5055 Phone: (650) 723-6270Fax: (650) 723-2253E-mail: [email protected]

May 30 - June 5, 19998th Chianti Workshop on Magnetic Resonance,Nuclear and Electron Relaxation, San Miniato(PISA), (Italy)


Ivano BertiniUniversity of FlorenceDepartment of ChemistryVia Gino Capponi 750121 Firenze - ItalyFax: +39-055-2757555E-mail: [email protected]

June 14-18, 1999Specialized Colloque AMPEREEPR, NMR, NQR in Solid State Physics - RecentTrends, Pisa (Italy)


Conference Secretariat, IFAM-CNR, Viadel Giardino, 756127 Pisa ItalyFax: +39 50 3139035E-mail: [email protected]

June 27 - July 2, 199914th International Meeting on NMR Spectroscopy,Royal Society of Chemistry, University ofEdinburgh, Edinburgh (U.K.)


Dr. John GibsonThe Royal Society of ChemistryBurlington HousePiccadillyLondonW1V 0BNUnited KingdomPhone: +44 (0) 171 437 8656 Fax: +44 (0) 171 7341227 E-mail: [email protected]

Vol. 20, No. 1-4 51

July 10-15, 1999NMR in Molecular Biology - Structure, Bindingand Molecular Recognition, Granada, (Spain)


Dr. Josip HendekovicPhone: +33 3 88 76 71 35Fax: +33 3 88 36 69 87E-mail: [email protected]

Augustl-5, 199941st Rocky Mountain Conference on AnalyticalChemistry, Denver, Colorado (USA)


Phone: 1-800-996-3233E-mail: [email protected]

August 6-7, 1999Varian/Chemagnetics 10th Annual Workshop onSolid State NMR, Estes Park, Colorado (USA)


Jim FryeChemagnetics/Varian NMRI2607 Midpoint Dr.Ft. Collins, CO 80525 USAPhone: (888) VARIANS or (970) 493-7007 x244Fax: +1-970-482-0570E-mail:

September 12-16, 1999The Alpine Conference on Solid-State NuclearMagnetic Resonance, Chamonix-Mont Blanc(France)


Alpine Conference SecretariatLaboratoire SUMEcole Normale Superieure de Lyon46, Allee d'ltalie69364 Lyon cedex 7, France

April 9-14, 200041st ENC (Experimental NMR Conference),Asilomar Conference Center, Pacific Grove,California (USA)


ENC1201 Don Diego AvenueSanta Fe, NM 87505 (USA) Phone: (505) 989-4573Fax: (505) 989-1073 E-mail:[email protected]

The editor would be pleased to receivenotices of upcoming meetings in the field ofmagnetic resonance to include in future vol-umes.

August 14-19, 199937th IUPAC Congress, Frontiers in Chemistry:Molecular Basis of the Life Sciences, GesellschaftDeutscher Chemiker, Berlin (Germany)


Gesellschaft Deutscher ChemikerIUPAC 99P. O. Box 90 04 40D-60444 Frankfurt am Main, GermanyPhone: +49-69-7917-358/ -360/ -366Fax: +49-69-7917-475E-mail: [email protected]: http://www.gdch.de


1999 ISMAR MEMBERSHIP ADDRESS LIST

ABE, AMY

PO BOX 978WAUKEGAN, IL 60079-0978 USA

ANGLISTER, JACOB

DEPARTMENT OF STRUCTURAL BIOLOGYWEIZMANN INSTITUTE OF SCIENCEREHOVOT 76100, ISRAEL972-8-343393

AHN, MYONG-KU

DEPARTMENT OF CHEMSITRYINDIANA STATE UNIVERSITYTERRE HAUTE, IN 47809 [email protected]

ARMITAGE, IAN M.

DEPARTMENT OF BIOCHEMISTRYUNIVERSITY OF MINNESOTA4-225 MILLARD HALL435 DELAWARE STREET, SE

MINNEAPOLIS, MN 55455-0347612-624-5977ian @ dimer. biochem. umn. edu

AILION, DAVID C.

DEPARTMENT OF PHYSICSUNIVERSITY OF UTAH304 JAMES FLETCHER BLDGSALT LAKE CITY, UT 84112 USA

(801) 581-6973ailion @ mail.physics. Utah, edu

ANDREW, E. RAYMOND

DEPARTMENT OF PHYSICS215 WILLIAMSON HALLPO BOX 118440UNIVERSITY OF FLORIDA

GAINESVILLE, FL 32611 USA(904) 392-6691andre w @phys. ufl. edu

AURENTZ, DAVID J.

152DAVEYLAB, BOX 55

DEPT. OF CHEMISTRY

PENNSYLVANIA STATE UNIVERSITY

UNIVERSITY PARK, PA 16802 USA

[email protected]

BASOSI, RICCARDO

DEPARTMENT OF CHEMUNIVERSITY OF SIENAPIAN DEI MANTELLINI4453100 SIENA, ITALY

0577/47054

Vol. 20, No. 1-4 53


BATCHELDER, LYNNE S.CAMBRIDGE ISOTOPE LABORATORIES50 FRONTAGE ROADANDOVER, MA 01810-5413 USA(508) 749-8000

BECKER, EDWIN D.

BLDG 5, ROOM 124NATIONAL INSTITUTES OF HEALTHBETHESDA, MD 20892 USA(301) [email protected]

BATES, JR., RICHARD D.

DEPARTMENT OF CHEMISTRYGEORGETOWN UNIVERSITYWASHINGTON, DC 20057 USA(202) 687-5970,bates@guvax

BEERY, JAMES W.

BASF CORPAGRICUTURAL PRODUCT CENTER26 DAVIS DRRESEARCH TRIANGLE PARK, NC 27709

[email protected]

BATLEY, MICHAEL

SCHOOL OF CHEMISTRYMACQUARIE UNIVERSITYNORTH RYDE, NSW 2109, AUSTRALIAmbatley@ocs 1. ocs.mq. edu.au

BENE, GEORGES J.

DPMC - SECTION DE PHYSIQUE24, QUAI E. ANSERMETCH 1211 GENEVE 4, SWITZERLAND

BATTA, GYULA

L KOSSUTH UNIVERSITYPOB 70H-4010 DEBRECEN, HUNGARY

36-52316666X2370batta ©tigris.klte.hu

BERGER, STEFAN

INST FUR ANALYTISCHE CHEMIEUNIVERSITAT LEIPZIGLINNESTRASSE 304103 LEIPZIG, GERMANY

[email protected]



BERGTER, LOTHARSOUZA CRUZ S.A.AV. SUBURBANA 206621050-450 RIO DE JANEIRORIO DE JANEIRO, BRAZIL

21-5827908Ibergter© Unisys, com. br

BLAZEK, ALMIRA

DEPARTMENT OF CHEMISTRYUNIVERSITY OF BRITISH COLUMBIA2036 MAIN MALLVANCOUVER, B.C. CANADA V6T 1Z1

(604) 822-2293almira @ unixg. ubc. ca

BERLINER, LAWRENCE J.

DEPARTMENT OF CHEMOHIO STATE UNIVERSITY120 W. 18TH AVENUECOLUMBUS, OH 43210 USA

(614) [email protected]

BLINC, ROBERT

J. STEFAN INSTITUTEPO BOX 3000JAMOVA 391111 LJUBLJANA, SLOVENIA

386-61-1773-900robert,blinc@ ijs.si

BERNHARD, WILLIAM A.DEPARTMENT OF BIOPHYSUNIVERSITY OF ROCHESTER601 ELMWOOD AVENUEROCHESTER, NY 14642 USA

(716) 275-3730

BLOOM, MYER

DEPARTMENT OF PHYSICS

UNIVERSITY OF BRITISH COLUMBIA

6224 AGRICULTURAL RD

VANCOUVER, B.C., CANADA V6T 1Z1

(604) 822-2136 OR 3853

BERTHIER, CLAUDE

LABORATOIRE DE SPECTROMETRIEUNIVERSITE JOSEPH FOURIER GRENOBLEBP 87 38402 SAINT MARTIN D'HERESCEDEX, FRANCE

[email protected]

BLUMICH, BERNHARD

LEHRSTUHL FUR MAKROMOLEKULAREWORRINGER WEG 1, D52074 AACHEN,49-241-80-6420bluemich @rwth-aachen. dc

Vol. 20, No. 1-4 55


BODDENBERG, BRUNO

LEHRSTUHL FUR PHYSIKALISCHE CHEMIE IIUNIVERSITAT DORTMUNDOTTO-HAHNSTR. 6D-44227 DORTMUND, GERMANY

231-755-3910

BUDINGER, THOMAS F.LAWRENCE BERKELEY LABUNIVERSITY OF CALIFORNIA1 CYCLOTRON RD., MS 55-121BERKELEY, CA 94720 USA

BREY, WALLACE S.

DEPARTMENT OF CHEMUNIVERSITY OF FLORIDAPO BOX 117200GAINESVILLE, FL 32611-7200 USA

(352) 392-0520

BULL, THOMAS E.

BIOPHYSICS LABORATORYDAPP/CBER/FDA HFM-4191401 ROCKVILLE PIKEROCKVILLE, MD 20852-1448 USA

[email protected] or bull©a 1 .cber.fda.gov

BROWN, ROBERT J.S.

515 W. 11THST.

CLAREMONT, CA 91711-3721 USA

(714) 626-7593

rjsbmeb @cyberg8t.com

CALLAGHAN, PAUL T.

DEPARTMENT OF PHYSICS & BIOPHYSMASSEY UNIVERSITYPALMERSTON NORTH, NEW ZEALAND

[email protected]

BUBB, WILLIAM A.

DEPT OF BIOCHEMISTRYUNIVERSITY OF SYDNEYSYDNEY NSW 2006, [email protected]

CAROLAN, JAMES L.

NALORAC CORPORATION814A ARNOLD DRIVEMARTINEZ, CA 94517 [email protected]



CARPER, W. ROBERT

CHEM DEPARTMENT, BOX 51WICHITA STATE UNIVERSITYWICHITA, KS 67260-0051 USA(316) 689-3120

CHEN-LOUNG, CHEN

DEPARTMENT OF WOOD & PAPERNORTH CAROLINA STATE UNIVERSITYBOX 8005RALEIGH, NC 27695-8005 USA

(919)515-5749

CASE, DAVID A.

DEPT OF MOLECULAR BIOLOGY MB1THE SCRIPPS RESEARCH INSTITUTELA JOLLA, CA 92037 [email protected]

CHOH, SUNG HO

DEPARTMENT OF PHYSICSKOREA UNIVERSITYSEOUL 136-701, KOREA(822) 927-3292

CHANCE, BRITTON

D501 RICHARDS BLDGUNIVERSITY OF PENNSYLVANIAPHILADELPHIA, PA 19104 USA

CODRINGTON, ROBERT S.

26815 ST. FRANCIS DRIVE

LOS ALTOS HILLS, CA 94022 USA

(650) 941-0127

CHAZIN, WALTER J.

DEPT OF MOLECULAR BIOLOGY, MB2THE SCRIPPS RESEARCH INSTITUTE10666 N. TORREY PINES RDLA JOLLA, CA 92037 USA

CONTI, LUIGI G.

ISTITUTO Dl CHIMICA GENERALECITTA UNIVERSITARIA1-00185 ROMA, ITALY

Vol. 20, No. 1-4 57

1999ISMAR MEMBERSHIP ADDRESS LIST

CONWAY, THOMAS F.

4144 HARVEY AVENUEWESTERN SPRINGS, IL 60558-1244 USA

DE LA CAILLERIE, JEAN-BAPTISTE

LAB DE PHYSIQUE QUANTIQUEESPCI10, RUEVAUQUELIN75231 PARIS CEDEX 05, FRANCE

COOK, IAIN

ICI RESEARCH GROUPGATE 2, NEWSOM ST ASCOT VALEVICTORIA, AUSTRALIA 3032613-9283-6383; 613-9283-6408

DESLAURIERS, ROXANNE

NRC INSTITUTE FOR BIODIAGNOSTICS435 ELLICE AVE

WINNIPEG, MB R3B 1Y6, CANADA

roxanne. desla uriers @nrc. ca

COZZONE, PATRICK J.CRMBM-CNRSFACULTE DE MEDECINE27 BD J. MOULIN13005 MARSEILLE, FRANCE

33-4-912-56529

DIEZ, ERNESTO

FACULTAD DE CIENCIAS C2-103UNIVERSIDAD AUTONOMA DE MADRID28049 MADRID, SPAINfabian @ vm I.sdi. uam. es

CROSS, KEITH J.

BAMBU, SCH OF DENTAL SCIENCEUNIVERSITY OF MELBOURNE711 ELIZABETH STMELBOURNE, VICTORIA, AUSTRALIA 3000

[email protected]

DOBSON, CM.

NEW CHEMISTRY LABORATORYUNIVERSITY OF OXFORDSOUTH PARKS RDOXFORD OX1 3QR, ENGLAND

[email protected]



DOMMISSE, R.

UNIVERSITY OF ANTWERP (RUCA)DEPARTMENT OF CHEMISTRYGROENENBORGERLAAN 1712020 ANTWERP, BELGIUM

32-3-218-0229dommisse @ruca. ua.be

DYSON, H. JANE

DEPARTMENT OF MOLECULAR BIOLOGYTHE SCRIPPS RESEARCH INSTITUTE10550 N. TORREY PINES ROADLA JOLLA, CA 92037 USA

619-784-2223dyson @scripps.edu

DOTY, F. DAVID

DOTY SCIENTIFIC INC.700 CLEMSON RDCOLUMBIA, SC 29223 USA803-788-6497

EASTLAND, GEORGE

DEPARTMENT OF CHEMSAGINAW VALLEY STATE UNIVERSITYUNIVERSITY CENTER, Ml 48710 USA


DUFOURC, ERICK J.

CENTRE DE RECHERCHE PAUL PASCAL,AV. A. SCHWEITZER33600 PESSAC, FRANCE(33) [email protected]

EATON, GARETH R.

DEPARTMENT OF CHEMUNIVERSITY OF DENVERDENVER, CO 80208 USA(303) 871-2980geaton @ cair. do. edu

DYBOWSKI, CECIL

DEPARTMENT OF CHEM & BIOCHEMISTRYUNIVERSITY OF DELAWARENEWARK, DE 19716 -2522 USA(302) 831-2726dybowski @ udel. edu

EGAN, WILLIAM M.

CENTER FOR BIOLOGICS EVALUATION .1401 ROCKVILLE PIKEROCKVILLE, MD 20852 USA

Vol. 20, No. 1-4 59


ERNST, MATTHIASLABORATORY FOR PHYSICAL CHEMISTRYUNIVERSITY OF NIJMEGENTOERNOOIVELDNL-6525 ED NIJMEGEN

THE [email protected]

FAIRBROTHER, WAYNE J.

GENENTECH, INC.460 POINT SAN BRUNO BLVD.SOUTH SAN FRANCISCO, CA 94080-4990

[email protected]

ERNST, RICHARD R.

LABORATORIUM FUR PHYSIKAL. CHEMIEETH-ZENTRUMCH-8092 ZURICH, SWITZERLAND

41-1-632-4366maer@ nmr.phys.chem.ethz.cn

FERREIRA, ANTONIO GILBERTO

AV. DAS HORTENCIAS 13913566 533-SAO CARLOS - SP, BRAZIL

16-2608208giba @ dg. uffscar/br

EVANS, FREDERICK E.

11340SOUTHRIDGEDRLITTLE ROCK, AR 72212-1832 USA

FIAT, DANIEL

DEPT OF PHYSIOLOGY & BIOPHYSICS (M/CUNIVERSITY OF ILLINOIS AT CHICAGO901 S. WOLCOTT AVENUECHICAGO, IL 60612-7342 USA

HOME: 1014 WOODBINE AVE, OAK PARK, IL312-996-7609

EVANS, JEREMYDEPT OF BIOCHEMISTRY & BIOPHYSICSWASHINGTON STATE UNIVERSITYPULLMAN, WA 99164-4660 USA

FORIS, ANTHONY

JACKSON LABDUPONT CENTRAL RESEARCH &JACKSON LABORATORY, CHAMBERSDEEPWATER, NJ 08023 USA

609-540-2034anthony. foris @ usa. dupont. com



FORSEN, STUREPHYSICAL CHEM 2CHEMICAL CTR, PO BOX 124UNIVERSITY OF LUND22100 LUND, SWEDEN

46-4610-8245sture @ hlin.fkem2.lth.se

FREEMAN, RAYDEPT OF CHEMUNIV OF CAMBRIDGELENSFIELD RDCAMBRIDGE CB2 1EW, ENGLAND

223-336-450

FRAHM, JENS

BIOMEDIZINISCHE NMR FORSCHUNGS

PO BOX 2841D-37018 GOETTINGEN, GERMANY

49-551-201-1721

FRIEDRICH, DIRK

HOECHST MARION ROUSSEL, INC.ROUTE 202/206, PO BOX 6800BRIDGEWATER, NJ [email protected]

FRAISSARD, JACQUESLABORATOIRE DE CHIMIE DES SURFACESUNIVERSITE PIERRE ET MARIE CURIE4 PLACE JUSSIEU75252 PARIS, CEDEX 05, FRANCE

FUJIWARA, FRED Y.

INSTITUTO DE QUIMICACAIXA POSTAL 6154UNIVERSIDADE ESTADUAL DE CAMPINAS13.081 CAMPINAS, SP, BRAZIL

[email protected]

FRECHET, DENISE

RHONE-POULENC RORER, CRVA, FRIEDEL13 QUAI JULES GUESDE94403 VITRY-SUR-SEINE, FRANCE

33-1-55-71-8463

denise. frechet@rp. fr

FUJ1WARA, HIDEAKI

FAC OF MEDICINE/SCH APPLIED HEALTHOSAKA UNIVERSITY2-2, YAMADA-OKA, SUITA, OSAKA 565JAPAN

Vol. 20, No. 1-4 61


FUKUI, H.

KITAMIINST OF TECHNOLOGY165KOENCHOKITAMI 090, JAPAN

GEOFFROY, MICHAEL

CHIMIE PHYSIQUE - SCIENCES II30, QUAI ERNEST-ANSERMET1211 GENEVE 4, [email protected]

FULEA, ADRIAN O.

CALEA PLEVNEI NO. 127

BUCHAREST 12, ROMANIA

40-0-637-4014

GERIG, JOHN T.

DEPT OF CHEMUNIV OF CALIFORNIASANTA BARBARA, CA 93106 USA

805 893-2113gerig @nmr. ucsb.edu

GAFFNEY, BETTY J.

NATIONAL HIGH MAGNETIC FIELD LABFLORIDA STATE UNIVERSITYTALLAHASSEE, FL 32310410-516-7451

GETTINS, PETERDEPT BIOCHEMISTRY M/C 536UNIVERSITY OF ILLINOIS AT CHICAGO1819-53 W. POLKCHICAGO, IL 60612 USA

312-996-5534pgetiins @ uic. edu

GARROWAY, ALLEN N.

CODE 6122NAVAL RESEARCH LABWASHINGTON, DC 20375-5342 USA

[email protected]

GILBOA, HAGGAI

DEPT OF CHEMTECHNION - ISRAEL INSTITUTE OFTECHNION CITY, HAIFA 32000, ISRAEL

[email protected]



GIURGIU, LIVIU

INST OF ISOTOPIC & MOLECULARPO BOX 7003400-R, CLUJ-NAPOCA, ROMANIA

[email protected]

GOLDMAN, MAURICE

SPECCE /CE SACLAYF 91191 GIF SUR YVETTE CEDEX, FRANCE

33-169-087-513goldman @ amoco.saclay.cea.fr

GLONEK, THOMAS

MR LAB5200 SOUTH ELLIS AVECHICAGO COL. OSTEOPATHIC MEDCHICAGO, IL 60615 USA

[email protected]

GRAHAM, LACHLAN JOHN WILLIAM

CSIRO DIV OF BLDG, CONSTRUCTION &PO BOX 56, HIGHETTVICTORIA 3190, AUSTRALIA

61-3-9252-6000lachlan @ mel.dbce.csiro.au

GMEINER, BILL

EPPLEY CANCER INSTITUTE

600 S. 42ND ST

OMAHA, NE 68198-6805 USA

402-559-4257

bgmeiner© unmc. edu

GRANT, DAVID M.

CHEMISTRY-314 S. 1400 E. #1320UNIV OF UTAHSALT LAKE CITY, UT 84112 USA

801- 581-8854grant© chem. Utah, edu

GOLDFARB, DANIELLA

DEPT OF CHEMICAL PHYSICSWEIZMANN INSTITUTE OF SCIENCEFÊHOVOT 76100, ISRAELcigoldfa @ wis. weizmann. ac.il

GRANT, HAMISH

VARIAN AUSTRALIA PTY LTD679 SPRINGVALE RDMULGRAVE VIC 3170, AUSTRALIA

61-3-9566-1138

[email protected]

Vol. 20, No. 1-4 63


GRUTZNER, JOHN B.CHEM DEPT1393BRWNBLDG.PURDUE UNIVW. LAFAYETTE, IN 47907-1393 USA


HAASE, A.LEHRSTUHL FUR EXPERIMENTELLE PHYSIKUNIVERSITAT WURZBURGAM HUBLANDD-97074 WURZBURG, GERMANY

[email protected]

GUNTHARD, HANS H.

PHYSICAL CHEM LABETH ZENTRUMCH-8092 ZURICH, SWITZERLAND

41-01-632-4170

HALL, L.D.

HERCHEL SMITH LAB FOR MEDICINALUNIVERSITY FORVIE SITEROBINSON WAYCAMBRIDGE, ENGLAND CB2 2PZ

44-1223-336-805Idh11 @ hslmc.cam.ac.uk

GUNTHER, HAROLD

UNIVERSITAT SIEGENFACHBEREICH 8, OC IID-57068 SIEGEN, GERMANY

HARRIS, ROBIN K.DEPT OF CHEMUNIV OF DURHAMSOUTH RDDURHAM DH1 3LE, ENGLAND

[email protected]

GUTOWSKY, H.S.

SCHOOL OF CHEMICAL SCIENCES177NOYESLAB, BOX 25UNIV OF ILLINOIS505 S. MATHEWS AVE

URBANA.IL 61801 USA

HASHA, DENNIS L.

DOW CHEMICAL USAANALYTICAL SCIENCES, BLDG. 1897 EMIDLAND, Ml 48667 USA517-636-4856



HASL1NGER, ERNST

INSTITUT F. PHARMAZEUTISCHE CHEMIEKARL-FRANZENS-UNIVERSITAT GRAZSCHUBERT STR. 1A-8010GRAZ, AUSTRIA

[email protected]

HIKICHI, KUNIODEPT OF POLYMER SCIENCEFACULTY OF SCIENCEHOKKAIDO UNIVSAPPORO, 060 JAPAN

[email protected]

HAUPT, ERHARD T.K.

INSTITUT FUR ANORGANISCHEUND ANGEWANDTE CHEMIEUNIVERSITAT HAMBURGMARTIN-LUTHER-KING-PL. 6

20146 HAMBURG-FRG, GERMANY49 404 123 3135haupt@ chemie. uni-hamburg. 6400. de

H1MMELRE1CH, UWE

INST FOR MAGNETIC RESONANCEBLACKBURN BLDG (D06)UNIVERSITY OF SYDNEYSYDNEY, NSW 2006, AUSTRALIA

61-2-9351-6168uwe @med. usyd. edu.au

HAUSSER, KARL H.MAX-PLANCK-INSTITUTFUR MEDIZINISCHE FORSCHUNGJAHN-STRASSE 296900 HEIDELBERG, GERMANY

48 62 46

HIRAOKI, TOSHIFUMI

DEPT OF APPLIED PHYSICSFACULTY OF ENGINEERINGHOKKAIDO UNIVSAPPORO 060, JAPAN

81-11-706-6640

hiraoki @ sun2. huap. hokudai. ac.jp

HIGINBOTHAM, JOHN

DEPT OF APPLIED CHEMICALAND PHYSICAL SCIENCESNAPIER UNIVERSITY10COUNTONRD

EDINBURGH EH10 5DT, SCOTLAND131-455-2517/[email protected]

HO, CHIEN

DEPT OF BIOLOGICAL SCIENCESCARNEGIE MELLON UNIVERSITY4400 FIFTH AVENUEPITTSBURGH, PA 15213-2683 USA

[email protected]

Vol. 20, No. 1-4 65


HOCH, M.J.R.DEPT OF PHYSICSUNIV OF THE WITWATERSRANDP.O. WITS 2050JOHANNESBURG, SOUTH AFRICA

[email protected]

ISOYA, J.UNIV OF LIBRARY & INFO SCIENCE1-2 KASUGA, TSUKUBA-CITYIBARAKI 305 JAPAN81-298-52-0511

HYNNINEN, PAAVO H.

DEPT OF CHEM, DIV ORG CHEMUNIV OF HELSINKIPO BOX 55FIN-00014 HELSINKI, FINLAND

358-0-191-40358

JACKSON, GRAHAM E.

DEPT OF CHEMUNIVERSITY OF CAPE TOWNRODEBOSCH 7700, SOUTH AFRICA21-6502531jackson ©psipsy. uct.ac.za

IKEDA, RYUICHI

DEPT OF CHEMUNIV OF TSUKUBATSUKUBA305 JAPAN

81-298 53 4250ikeda ©staff, chem. tsukuba. ac.jp

JAMES, THOMAS L.DEPT OF PHARMACEUTICAL CHEMBOX 0446UNIV OF CALIFORNIASAN FRANCISCO, CA 94143-0446 USA

(415)444476-1569james ©picasso. ucsf. edu

INGWALL, JOANNE S.

NMRLABBRIGHAM & WOMEN'S HOSPITAL221 LONGWOOD AVEBOSTON, MA 02115 USA

617 732-6994

JANES, NATHAN

DEPT OF PATHOLOGY, ANATOMY & CELLTHOMAS JEFFERSON UNIVERSITY1020 LOCUST STREET, RM 264PHILADELPHIA, PA 19107 USA

[email protected]



JARDETZKY, OLEGSTANFORD MAGNETIC RESONANCE LABSTANFORD UNIVSTANFORD, CA 94305-5055 USA

415 723-6153

JOHNSON, JR., CHARLES S.

CB #3290 DEPT. OF CHEMISTRYUNIVERSITY OF NORTH CAROLILNACHAPEL HILL, NC 27599-3290 USA919-966-5229charlesjohnson @ unc. edu

JEENER, J.

ULB-PLAINE (CPI-232)BD DU TRIOMPHEB-1050 BRUSSELS, BELGIUM

[email protected]

JOSEPH-NATHAN, PEDRO

DEPT OF CHEMPO BOX 14-740CENTRO DE INVESTIGACION IPNMEXICO CITY 07000 MEXICO

525-747-7112

JELICKS, LINDA A.DEPARTMENT OF PHYSIOLOGY &ALBERT EINSTEIN COLLEGE OF MEDICINE1300 MORRIS PARK AVENUEBRONX, NY 10461 USA


KALB1TZER, H.R.

MAX-PLANCK-INSTITUTFUR MED. FORSCHUNGJAHNSTR. 29D-69028 HEIDELBERG, GERMANY

[email protected]

JINSART, WANIDA

DEPT GEN SCI

FAC OF SCICHULALONGKORN UNIVERSITYBANGKOK 10330, THAILAND

[email protected]

KAWAMORl, ASAKO

FACULTY OF SCIENCEKWANSEI GAKUIN UNIVNISHINOMtYA 662, [email protected]

Vol. 20, No. 1-4 67


KEIFER, PAUL A.VARIAN NMR INSTRUMENTS3120 HANSEN WAY, D-298PALO ALTO, CA 94304-1030650-424-6609paul.keifer@nmr. varian.com

KHETRAPAL, C.L.INDIAN INST OF SCIENCEBANGALORE-560 012, INDIA91-80-3344550elk @ sit Use. ernet. in

KEIRE, DAVID

DIVISION OF IMMUNOLOGYBECKMAN RES INSTOF THE CITY OF HOPE1450 E. DUARTE RDDUARTE, CA 91010 USA

818-359-8111 X2601dak @ ernst. con. org

KIND, R.

INSTITUT FUR QUANTENELEKTRONIKETH HONGGERBERG (HPF)CH-8093 ZURICH, SWITZERLAND

411-633-2331kindrd© iqe.phys.ethz.cn

KERTESZ, JEAN C.

2329 W. 2ND ST. #7LOS ANGELES, CA 90057 USA

213 383-5602

KISPERT, LOWELL D.

CHEM DEPTBOX 870336UNIV OF ALABAMATUSCALOOSA, AL 35487-0036 USA

(205) 348-7134Ikispert@ua1vm

KEVAN, LARRY

DEPTARTMENT OF CHEMISTRYUNIVERSITY OF HOUSTONHOUSTON, TX 77204-5641 USA

713 [email protected]

KISTLER, J. PHILIP

STROKE SERVICE/VBK 802MASSACHUSETTS GENERAL HOSPITALBOSTON, MA 02114 USA


1999 ISMAR MEMBERSHIP ADDRESS UST

KOBORI, YASUHIRO

INST FOR CHEMICAL REACTION SCIENCETOHOKU UNIVERSITYKATAHIRA 2-1-1, AOBAKU, SENDAI980-8577 JAPAN

[email protected]

KUCHEL, PHILIP W.

DEPT OF BIOCHEMUNIV OF SYDNEYSYDNEY, NSW 2006, AUSTRALIA

[email protected]

KOSHINO, HIROYUKI

THE INST. OF PHYS. & CHEM. RES. (RIKEN)WAKO, SAITAMA 351-01 [email protected]

KULAGINA, TATIANA

MOSCOW REGION CHERNOGOLOVKA142432 INSTITUTSKI PR. 4 APP. [email protected]

KOWALEWSKI, VALDEMAR J.

FACULTAD DE CIENEIAS EXACTAS1428 BUENOS AIRES, ARGENTINA

KUMAR, ANIL

DEPT OF PHYSICS & SIFINDIAN INST OF SCIENCEBANGALORE 560012, INDIA

[email protected]

KOZAKI, MASATOSHI

DEPT OF CHEM - GRAD SCH OF SCIENCEOSAKA CITY UNIVERSITYSUGIMOTO, SUMIYOSHIOSAKA 558-8585 JAPAN

KURODA, YOSHIHIRO

FACULTY OF PHARMACEUTICAL SCIENCESKYOTO UNIVERSITYSAKYO-KU, KYOTO, 606-01 JAPAN

81-75-753-4530

Vol. 20, No. 1-4 69


LANGE, DAVID G.DEPT OF ANESTHESIOLOGY, BLALOCK

JOHNS HOPKINS HOSPITAL601 N. WOLFE ST"BALTIMORE, MD 21205 USA

LEWIS, ANDREW

CHEM DEPTUNIVERSITY OF BRITISH COLUMBIA2036 MAIN MALLVANCOUVER, B.C. V6T 1Z1, CANADA

[email protected]

LAUTERBUR, PAUL C.

DEPT OF MEDICAL INFORMATION SCIENCEBIOMEDICAL MAGNETIC RESONANCE LABUNIV OF ILLINOIS2100S. GOOWDINAVE

URBANA.IL 61801 [email protected]

LIM, AE-RAN

DEPARTMENT OF PHYSICS

JEONJU UNIVERSITY

HYOJA-DONG 3KA 1200

CHONJU 560-759, KOREA

0652-220-2514

LED, JENS J.

DEPT OF CHEMUNIVERSITY OF COPENHAGENUNIVERSITETSPARKEN 5DK-2100 COPENHAGEN 0, DENMARK

[email protected]; [email protected]

LIN, YUNG-YA

C/O PROF. ALEX PINESDEPARTMENT OF CHEMISTRYUNIVERSITY OF CALIFORNIA, BERKELEYBERKELEY, CA 94720 USA510-559-8358

LEE, MAN-HO

DEPARTMENT OF INDUSTRIAL CHEMISTRYKYUNGPOOK NATIONAL UNIVERSITYTAEGU 702-701, SOUTH KOREA82-53-950-5584; 82-53-950-6594mhlee @bh.kyungpook.ac.kr

LITT, LAWRENCE

DEPT OF ANESTHESIOLOGYUCSFBOX 0648SAN FRANCISCO, CA 94143 USA

[email protected]



LUNDBERG, PETERMR-UNIT, DEPT OF RADIATION PHYSICSUNIVERSITY HOSPITAL OF LINKOPINGS-581 85 LINKOPING, [email protected]

LUTZ, OTTO

PHYSIKALISCHES INSTITUTDER UNIVERSITAT TUBINGENMORGENSTELLED-72076 TUBINGEN, GERMANY

7071-2976265, 290

LUZ, ZEEVDEPT OF CHEMICAL PHYSICSWEIZMANN INSTITUTEREHOVOT 76100, ISRAELciluz@wiz-ofer. weizmann.ac.il

MACURA, SLOBODAN

MAYO CLINIC"GUGGENHEIM C061Z200 FIRST STREET, SWROCHESTER, MN 55905 USA

507284-5917macura @mayo. edu

MAFEE, MAHMOOD F.

DIRECTOR, MRI FACILITYUNIV OF ILLINOISMRICTR, M/C711830 S. WOOD STREET

CHICAGO, IL 60612-7336 USA(312) 996-7295

MARAVIGLIA, BRUNO

DIPARTIMENTO DI FISICAUNIVERSITA "LA SAPIENZA"P.LE ALDO MORO 200185 ROMA, ITALY

[email protected]

MARINO, ROBERT A.

DEPT OF PHYSICSHUNTER COLLEGE OF CUNY695 PARK AVENEW YORK, NY 10021 USA

212-772-5391/[email protected]

MARKLEY, JOHN L.

BIOCHEM DEPTUNIV OF WISCONSIN420 HENRY MALLMADISON, Wl 53706-1569 USA

(608) 263-9349

i

Vol. 20, No. 1-4 71


MATSUI, SHIGERU

INSTITUTE OF APPLIED PHYSICSUNIVERSITY OF TSUKUBATSUKUBA, IBARAKI 305JAPAN

[email protected]

MEHRING, MICHAEL

PHYSIKALISCHES INSTITUT, TEILINSTITUT 2UNIVERSITAT STUTTGARTPFAFFENWALDRING 577000 STUTTGART 80, GERMANY

mm2 ©physik. uni-stuttgart. de

MCCARTHY, MICHAEL J.

DEPT OF FOOD SCIENCE & TECHNOLOGYUNIVERSITY OF CALIFORNIADAVIS, CA 95616-8598 USA916-752-8921mjmccarthy@ ucda vis. edu

MEI, ELIZABETH H.

PO BOX 1128MILLBRAE, CA 94030 USA

MCGARVEY, BRUCE R.

DEPT OF CHEMISTRY & BIOCHEMISTRYUNIVERSITY OF WINDSORWINDSOR, ONTARIO, CANADA N9B 3P4

519-253-4232beprm @ uwindsor. ca

MESSERLE, BARBARA A.

DEPT OF ORGANIC CHEMISTRYTHE UNIVERSITY OF SYDNEYSYDNEY, 2006 [email protected]

MCTIGUE, MIKE

3 BUNDEERA RDCAULFIELD SOUTHVICTORIA 3162, [email protected]

METZ, KENNETH R.

400 BROOKLINE AVE. #22ABOSTON, MA 02215 USA617 [email protected]



MISRA, SUSHIL K.

PHYSICS DEPTCONCORDIA UNIVERSITY1455 DE MAISONNEUVE BLVD. WESTMONTREAL, QUEBEC H3G 1M8, CANADA


MOORE, GEOFF

SCHOOL OF CHEMICAL SCIENCESUNIVERSITY OF EAST ANGLIANORWICH NR4 7TJ, [email protected]

MIYAGAWA, ICHIRO

DEPARTMENT OF PHYSICS & ASTRONOMYUNIVERSITY OF ALABAMATUSCALOOSA, AL 35487-0324 USA205-348-3787jharrell© uaivm. ua.edu

MOTA DE FREITAS, DUARTE

CHEM DEPTLOYOLA UNIV6525 NORTH SHERIDAN RDCHICAGO, IL 60626 USA


MONDELLI, ROSANNA

DIPARTIMENTO Dl SCIENZEMOLECOLARI AGROALIMENTARI (DISMA)UNIVERSITA Dl MALANOVIA CELORIA 2

20133 MILANO, ITALY02.2663662

MULLER-WARMUTH, W.

INSTITUT FUR PHYSIKAL CHEMIEWESTFALISCHE WILHELMS UNIVSCHLOTZPLAZ 44400 MUNSTER, GERMANY

MONTELIONE, GAETANO T.

CABM - RUTGERS UNIVERSITY679 HOES LANEPISCATAWAY, NJ 08854-5638 USA

[email protected]

MURPHY-BOESCH, JOSEPH

DEPT OF NMRFOX CHASE CANCER CTR7701 BURHOLME AVEPHILADELPHIA, PA 19111 USA


Vol. 20, No. 1-4 73


NAIRN, KATHERINE

DEPT OF MATERIALS ENGINEERINGMONASH UNIVERSITYCLAYTON, VICTORIA, AUSTRALIA 3168

[email protected]

NETZEL, DANIEL A.

2526 SKYVIEW LANE

LARAMIE, WY 82070 USA

(307) 745-3510

NAKAMURA, NOBUODEPT OF CHEMGRADUATE SCHOOL OF SCIENCEOSAKA UNIVTOYONAKA, OSAKA, 560 JAPAN

[email protected]

NEWMARK, RICHARD A.

3M.BLDG201-BS-07ST. PAUL, MN 55144 USA612-733-7679ranewmark@mmm. com

NARUTIS, VYTAS

390 LONGCOMMON RD

RIVERSIDE, IL 60546-1748 USA

mokslas @ CompuServe, com

NORBERG, RICHARD E.

7134 PRINCETON

ST. LOUIS, MO 63130 USA

374 935-4837

NAVON, GIL

SCHOOL OF CHEMTEL-AVIV UNIVTEL AVIV 69978, ISRAEL

972-3-640-8156

[email protected]

OHNO, KEIICHI

DEPT OF INFO ENGTHE POLYTECHNIC UNIVERSITY4-1-1 HASHIMOTODAI, SAGAMIHARA 229KANAGAWA, JAPAN

[email protected]



OKADA, KEIJI

DEPT OF CHEM, GRAD SCH OF SCIENCEOSAKA CITY UNIVERSITYSUGIMOTO, SUMIYOSHI-KUOSAKA 558-8585 JAPAN

[email protected]

PILBROW, JOHN R.

DEPT OF PHYSICSMONASH UNIVERSITYCLAYTON, VICTORIA, AUSTRALIA 3168

[email protected]

PACKER, KEN

DEPARTMENT OF CHEMISTRYUNIVERSITY OF NOTTINGHAMNOTTINGHAM NG7 2RO, UNITED [email protected]

PINES, ALEXANDER

DEPT OF CHEMUNIV OF CALIFORNIABERKELEY, CA 94720-1460 USA

(510) 642-1220pines© cchem.berkeley. edu

PANEPUCCI, HORACIO CARLOS

RUA JOSE DUARTE DESOUZA, 203INSTITUTO DE FISICA E QUIMICA DE SAOCAIXA POSTAL 36913560-970 SAO CARLOS, SP, BRAZIL

55-162-720-299

[email protected]

PINTAR, M. MIKPHYSICS DEPARTMENTUNIVERSITY OF WATERLOOWATERLOO, ONTARIO N2L3G1 [email protected]

PEEMOELLER, H.

PHYSICS DEPTUNIV OF WATERLOOWATERLOO, ONTARIO, CANADA N2L 3G1

(519)885-1211/2633

POGNI, REBECCA

DEPT OF CHEMUNIV OF SIENAPIAN DEI MANTELLINI4453100 SIENA, ITALY

0577-47065

Vol. 20, No. 1-4 75


POOLE, JR., CHARLES P.

UNIV OF SOUTH CAROLINACOLUMBIA, SC 29208 USA

PUNKKINEN, MATTI

WIHURI PHYSICAL LABUNIV OF TURKUFIN-20500 TURKU, [email protected]

PORTE, ANDREW L.

DEPTOFCHEMUNIV OF GLASGOWGLASGOW G12 8QQ, SCOTLAND

RAJAGOPAL, PONNI

DEPT OF BIOCHEM, BOX 357742HEALTH SCIENCES BLDG, RM K-410CUNIVERSITY OF WASHINGTONSEATTLE, WA 98195-7742

206-616-4508ponjan @u. washington.edu

PRESTON, CAROLINE M.

PACIFIC FORESTRY CENTRENATURAL RESOURCES CANADA506 W. BURNSIDE RDVICTORIA, B.C., CANADA V8Z 1M5

604-363-0720cpreston @pfc. forestry, ca

RAMANATHAN, K.V.

SOPHISTICATED INSTRUMENTS FACILITYINDIAN INSTITUTE OF SCIENCEBANGALORE 560-012, [email protected]

PROCTOR, WARREN G.

GAMMEL STRANDVEJ 3003050 HUMLEBAEK, DENMARK

RAO, B.D. NAGESWARA

DEPT OF PHYSICS, LD 154DINDIANA UNIV/PURDUE UNIV (IUPUI)402 N. BLACKFORD STREETINDIANAPOLIS, IN 46202-3273 USA

(317)274-6901brao @ indyvax.iupui.edu



REDFIELD, ALFRED G.DEPT OF BIOCHEMBRANDEIS UNIVWALTHAM, MA 02250 USA

ROBERTS, J.D.CALTECH 164-30PASADENA, CA 91125 USAHOME 213-681-9024

REISSE, J.

CHIMIE ORGANIQUE (CP 165)UNIVERSITE LIBRE DE BRUXELLES50, AVENUE F.D. ROOSEVELB 1050 BRUXELLES, BELGIQUE

32-2-650-3605jreisse @ ulb.ac.be

RODGER, CHARLES

SANOFI WINTHROP25 GREAT VALLEY PRKWYMALVERN, PA 19355 USA

610-889-8873/8811

RICO, MANUEL

INSTITUTO DE ESTRUCTURA DE LECSICSERRANO 11928006 MADRID, SPAIN

ROSSI, CLAUDIOCHEMICAL & BIOSYSTEM SCIENCESUNIV OF SIENAPIAN DEI MANTELLINI 4453100 SIENA, ITALY

39 [email protected]

ROBERTS, GORDON C.K.

BIOLOGICAL NMR CENTREUNIV OF LEICESTERPO BOX 138MEDICAL SCIENCES BLDG UNIVERSITY

LEICESTER LE1 9HN, UNITED KINGDOM44-533-522938

ROURICK, ROBYN

BRISTOL-MYERS SQUIBB5 RESEARCH PARKWAYWALLINGFORD, CT 06492 USA203-284-6983

Vol. 20, No. 1-4 11

1999ISMAR MEMBERSHIP ADDRESS LIST

RUDOWICZ, CZESLAW

DEPT OF PHYSICS & MATERIALS SCIENCECITY UNIVERSITY OF HONG KONG83 TAT CHEE AVEKOWLOONITONG, HONG KONG

(852) 788-7787apceslaw© cityu. edu.hk

SANCHEZ-FERRANDO, FRANCISCO

DEPARTAMENT DE QUIMICAUNIVERSITAT AUTONOMA DE BARCELONA08193 BELLATERRABARCELONA, SPAIN

34-3-581-1271iqofsanchez @ cc. uab. es

SAGSTUEN, EINAR

DEPT OF PHYSICSUNIV OF OSLOPO BOX 1048 BLINDERNN-0316 OSLO 3, NORWAY

4722855653einarsa @ fys. uio. no

SARATHY, K.P.

NMR FACILITYCHEM BLDG #203AUBURN UNIVAUBURN, AL 36849-5312 USA


SAITO, KOJI

ADVANCED MATERIALS & TECHNOLOGYNIPPON STEEL CORPORATION3-35-1 IDA. NAKAMARA-KUKAWASAKI 211, JAPAN

44-797-1274saito @lab I.nsc. co.jp

SATO, KAZUNOBU

FACULTY OF SCIENCEOSAKA CITY UNIVERSITY3-3-138 SUGIMOTO, SUMIYOSHI-KUOSAKA 558, JAPAN

[email protected]

SAKELLARIOU, DIMITRIOS

27, QUAI CLAUDE BERNARD

69007 LYON, FRANCE

33-4-37280504; LAB 33-4-72728395

[email protected]

SCHMIDT, JAN

HUYGENS LABUNIV OF LEIDENPO BOX 95042300 RA LEIDEN, THE NETHERLANDS




SCHMIDT-ROHR, KLAUS

DEPT OF POLYMER SCIENCE &UNIVERSITY OF MASSACHUSETTSAMHERST, MA 01003413-577-1417

SEVILLA, MICHAEL D.DEPT OF CHEMOAKLAND UNIVROCHESTER, Ml 48309 USA(313) 370-2328

SEPAROVIC, FRANCES

SCH OF CHEMISTRYUNIVERSITY OF MELBOURNEPARKVILLE VIC 3052AUSTRALIA

[email protected]

SHABANOVA, ELIZABETH S.

CISMI, SYMBION SCIENCE PARKUNIVERSITY OF COPENHAGENFRUEBJERVEJ 3, DK-2100, COPENHAGEN 0DENMARK

45-3532-1898shabanov@xray. ki. ku. dk

SERIANNI, ANTHONY S.

DEPT OF CHEM & BIOCHEMUNIV OF NOTRE DAMENOTRE DAME, IN 46556 USA(219) 631-7807serianni. 1 ©nd.edu

SHAFER, JULES A.

MERCK RESEARCH LABORATORIESWP44-L202WEST POINT, PA 19486 USA

215 661-7203

SERVOZ-GAVIN, PIERRE

ECOLE NORMALE SUPERIEURE DE LYONRELATIONS INTERNATIONALES ET46 ALLEE D'lTALIE69364 LYON CEDEX 07, FRANCE

33+-472728517; HOME 33-4768-77918pservoz @ ens-lyon. fr

SHARP, ROBERT R.

DEPT OF CHEMUNIV OF MICHIGANANN ARBOR, Ml 48109-1055 USA

Vol. 20, No. 1-4 79


SHIMOKAWA, SHIGEZO

NMR LABFACULTY OF ENGINEERINGHOKKAIDO UNIVSAPPORO 060, JAPAN

81-11-706-6795sige @e3.hines.hokudai.ac.jp

SKLENAR, VLADIMIR

LAB OF BIOMOLECULAR STRUCTURE &FACULTY OF SCIENCEMASARYK UNIVERSITYKOTLARSKA 2

CZ-611 37 BRNO, CZECH [email protected]

SHIOMI, DAISUKE

DEPT OF MATERIAL SCIENCEOSAKA CITY UNIVERSITYSUGIMOTO, SUMIYOSHI-KUOSAKA 558, JAPAN

81-6-605-3149shiomi © sci. osaka-cu. ac.jp

SLICHTER, CHARLES P.

DEPT OF PHYS., 311 LOOMIS LABUNIV OF ILLINOIS1110W. GREEN STURBANA, IL 61801 USA

(217) 333-3834cps ©physics, uiuc. edu

SHIRLEY, WILLIAM M.

CHEM DEPTPITTSBURGH STATE UNIVPITTSBURGH, KS 66762 USA(316)[email protected]

SMITH, J.A.S.

CHEMISTRY DEPARTMENTKING'S COLLEGESTRAND, LONDON WC2R 2LSUNITED KINGDOM

PHONE/FAX 171-873-2699

SIMINOVITCH, DAVID J.

DEPARTMENT OF PHYSICSUNIVERSITY OF LETHBRIDGE4401 UNIVERSITY DRIVELETHBRIDGE, ALBERTA, CANADA T1K 3M4

403-329-2359siminovitch ©hg.uleth.ca

SOHAR.PALDEPT OF GENERAL & INORGANIC CHEMLORAND EOTVOS UNIVERSITYPO BOX 32H-1518 BUDAPEST-112, HUNGARY

835-340/162



SPIESS, HANS W.

MAX-PLANCK-INSTITUT FURPOSTFACH 31 48D-55021 MAINZ, GERMANY

49 6131 379 120spiess @ mpip-mainzmpg. de

ST1LBS, PETER

PHYSICAL CHEMROYAL INST OF TECHNOLOGYSTOCKHOLM S-10044, SWEDEN

[email protected]

SPISNI, ALBERTO

ISTITUTO Dl CHIMICA BIOLOGICAUNIVERSITA Dl PARMAVIA GRAMSCI14-43100 PARMA, ITALY

[email protected]

STIMSON, EVELYN R.

4098 CALIFORNIA ROADOKEANA, OH [email protected]

SPRINGER, CHARLES S.

DEPARTMENT OF CHEMISTRYBROOKHAVEN NATIONAL LABORATORYUPTON, NY 11973-5000 [email protected]

STRANGE, J.H.

PHYSICS LABTHE UNIV OF CANTERBURYCANTERBURY, KENT CT2 7NRUNITED KINGDOM

[email protected]

STEINSCHNEIDER, M.A.

RESEARCH RESOURCES CTRUNIV OF ILLINOIS835 S. WOLCOTT AVECHICAGO, IL 60612 USA

STUART, S.N.

1 KENILWORTH STREETBALWYNVIC. 3103AUSTRALIA

61-3-9857-8318

Vol. 20, No. 1-4 81


SUGIURA, MAKIKO

KOBE PHARMACEUTICAL UNIVERSITY4-19-1, MOTOYAMAKITA-MACHI,KOBE 658-8558, [email protected]

TERPINSK1, EVA A.

INTERFERON SCIENCES, INC.783 JERSEY AVENEW BRUNSWICK, NJ 08901 USA

[email protected]

TAKUI, TAKEJI

DEPARTMENT OF CHEMISTRYFACULTY OF SCIENCEOSAKA CITY UNIVERSITY3-3-138 SUGIMOTO

SUMIYOSHI-KA, OSAKA 558, [email protected]

TOKITA, M.

FUKUOKA INST OF TECHNOLOGYWAJIRO, HIGASHI-KUFUKUOKA 811-01, JAPAN

TEKI, YOSHIO

DEPT OF MATERIALS SCIENCE

FACULTY OF SCIENCE

OSAKA CITY UNIVERSITY

SUGIMOTO 3-3-138

SUMIGOSHI-KU, OSAKA 558, JAPAN

TOWNER, RHEAL A.DEPT OF PHYSIOLOGY & PHARMACOLOGYSCH OF MOLECULAR SCIENCESJAMES COOK UNIV OF NORTHTOWNSVILLE 4811 QUEENSLAND,

TERAO, TAKEHIKO

DEPT OF CHEM

GRADUATE SCHOOL OF SCIENCEKYOTO UNIVKYOTO 606-01, JAPAN

[email protected]

TROMMER, WOLFGANG

FACHBEREICH CHEMIE DER UNIVERSITATPOSTFACH 304967653 KAISERSLAUTERN, GERMANY

[email protected]



UKLEJA, PAUL

PHYSICS DEPTUNIV OF MASSACHUSETTS DARTMOUTHNORTH DARTMOUTH, MA 02747 USA

(508) 999-8353pukleja @ umassd. edu

VAN HALBEEK, HERMAN

GLYCOBIOLOGY PROGRAM, UCSD CANCERUNIVERSITY OF CALIFORNIA, SAN DIEGOCMM-EAST BUILDING, ROOM 10509500 GILMAN DRIVE

LA JOLLA, CA [email protected]

UZAWA, JUN

THE INSTITUTE OF PHYS & CHEMHIROSAWA2-1.WAKOSAITAMA 351-01, JAPAN

81-48-467-9361uzawa @ rikeunmr.rikeu.go.jp

VAN HECKE, PAUL E.

E. GILBERTSTRAAT 4B-3000 LEUVEN, [email protected]

VAN CALSTEREN, MARIE ROSE

AGRICULTURE & AGRI-FOOD CANADA3600 CASAVANT BLVD WESTST. HYACINTHE, QUEBEC, CANADA J2S 8E3

[email protected]

VENTURELLA, VINCENT S.

VENTURA ASSOCIATES9 RING PLACECEDAR GROVE, NJ 07009 USA

[email protected]

VAN ETTEN, ROBERT L.

DEPARTMENT OF CHEMISTRYPURDUE UNIVERSITYWEST LAFAYETTE, IN 47907-1393 USA

[email protected]

VISINTAINER, JAMES

316 W. STREETSBORO STREET

HUDSON, OH 44236 USA

(330) 796-4673

[email protected]

Vol. 20, No. 1-4 83


VOLD, REGITZE R.DEPT OF CHEMISTRY & BIOCHEMISTRYUNIV OF CALIFORNIA, SAN DIEGO9500 GILMAN DRLA JOLLA, CA 92093-0359 USA


WASYLISHEN, RODERICK E.

DEPT OF CHEMDALHOUSIE UNIVHALIFAX.NOVA SCOTIA, CANADA B3H 4J3

[email protected]

VON PHILIPSBORN, WOLFGANG

INST OF ORGANIC CHEMUNIV OF ZURICHWINTERHURERSTRASSE190CH-8057 ZURICH, SWITZERLAND

41-1-25742 60egysi@oci. unizh. ch

WAUGH, JOHN S.

MIT 6-235CAMBRIDGE, MA 02139 USA

(617)253-1901

VON ZELEWSKY, A.

INST OF INORGANIC CHEMUNIV OF FRIBOURGPEROLLES, 1700 FRIBOURG,41-26-300-87-30

WERNER-ZWANZIGER, ULRIKE M.

DEPARTMENT OF CHEMISTRYINDIANA UNIVERSITYBLOOMINGTON, IN 47405 [email protected]

WASSALL, STEPHEN R.

DEPT OF PHYSICSINDIANA UNIV-PURDUE UNIV402 N. BLACKFORD ST.INDIANAPOLIS, IN 46202-3273 USA

(317)[email protected]

WESTERMAN, PHILIP

DEPARTMENT OF BIOCHEMISTRYNORTHEASTERN OHIO UNIVERSITIESCOLL OF MED4209 STATE ROUTE 44 - PO BOX 95

ROOTSTOWN, OH 44272-0095 [email protected]



WHIPPLE, EARL B.7 FOREST HILLS DRIVEMADISON, CT 06443 USA

WONG-MOON, KIRBY C.

GENEVA PHARMACEUTICALS, ARD DEPTPO BOX 4462655 W. MEDWAY BLVD.BROOMFIELD, CO 80038-0446

[email protected]

WIECKOWSKI, ANDR2EJ B.

DIV EXPER PHYSICS/INST OF PHYSICSFAC MATH, PHYS, & TECHTADEUSZ KOTARBINSKI PEDAGOGICALPLAC SLOWIANSKI 6

PL-65-069 ZIELONA GORA, POLAND48-68-202439

WOSINSKI, TADEUSZ

INST OF PHYSICSPOLISH ACADEMY OF SCIENCESAL LOTNIKOW 32/4602-668 WARSAW, POLAND

22-437001wosin @ delta 1. if pan. edu.pl

WIJMENGA, S.

DEPT OF MEDICAL BIOCHEM & BIOPHYSUNIV OF UMEAS-90187 UMEA, [email protected]

WU, XUEWEN

200062 EAST CHINA NORMAL UNIVERSITY3663 ZHONGSHAN BEI LUSHANGHAI, P.R. CHINAxwwue ©online.sh.cn

WILLEM, RUDOLPH

HIGH RESOLUTION NMR CENTREVRIJE UNIVERSITEIT BRUSSELPLEINLAAN 2 B-1050 BRUSSEL, BELGIUM

WUTHRICH, KURT

INSTITUT FUR MOLEKULARBIOLOGIE UNDETH-HONGGERBERG HPMCH-8093 ZURICH, SWITZERLAND

41-1-377-2473

Vol. 20, No. 1-4 85


YANNONI, COSTANTINO S.

IBM ALMADEN RESEARCH CTR650 HARRY RDSAN JOSE, CA 95120 -6099 USA


YE, CHAOHUI

PO BOX 71010WUHAN 430071, P.R. CHINA

86-27-7876842ye @nmr. whcnc.ac.cn

YEOM, TAE HO

DEPT OF PHYSICSCHONGJU UNIVERSITYCHONGJU 360-764, KOREA

82-431-229-8472thyeom @alpha94. chongju. ac. kr

ZOLNAI, ZSOLT

MAYO FOUNDATION200 FIRST STREET SWROCHESTER, MN [email protected]

Date post:	17-Sep-2018
Category:	Documents
Upload:	lequynh
View:	216 times
Download:	0 times

of WIUU - weizmann.ac.il · E.R. ANDREW LAWRENCE BERLINER ROBERT BLINC University of Florida Ohio...

Documents