recent developments in solid-state nuclear magnetic resonance...

REVIEW / SYNTHÈSE

Recent developments in solid-state nuclearmagnetic resonance of quadrupolar nuclei andapplications to biological systems

Gang Wu

Abstract: Recent advances in nuclear magnetic resonance (NMR) methodology and improvements in high-field NMRinstrumentation have generated a new wave of research interests in the application of solid-state NMR to the study ofquadrupolar nuclei. These developments now permit increasingly complex biological systems to be probed byquadrupolar NMR. In this review I describe a few recent developments in NMR studies of quadrupolar nuclei anddemonstrate the potential of solid-state quadrupolar NMR in the study of biological systems. In particular, I discuss theapplication of solid-state NMR of17O, 67Zn, 59Co, 23Na, and39K nuclei with a prognosis for future work.

Key words: nuclear magnetic resonance, quadrupole, solid state, biological system.

Résumé: Des développements récents de la méthodologie de RMN et des améliorations de l’instrumentation de RMNà champ de forte intensité ont généré une nouvelle vague d’intérêt de recherche dans l’application de la RMN de l’étatsolide à l’étude de noyaux quadripolaires. Ces développements permettent maintenant d’étudier des systèmesbiologiques de plus en plus complexes à l’aide de la RMN quadripolaire. Dans cette revue, nous décrivons quelquesdéveloppements récents résultant d’études de RMN de noyaux quadripolaires et nous démontrons le potentiel de laRMN quadripolaire de l’état solide pour l’étude des systèmes biologiques. En particulier, nous discutons del’application de la RMN de l’état solide des noyaux17O, 67Zn, 59Co, 23Na et 39K et nous envisageons de futurs travauxde recherche.

Mots clés: résonance magnétique nucléaire, quadripôle, état solide, système biologique.

[Traduit par la Rédaction] Review / Synthèse 442

Nuclear magnetic resonance (NMR) spectroscopy is apowerful and versatile analytical technique that can providesite-specific information about chemical bonding, structure,and dynamics in molecular systems. NMR applications havemade a major impact in a variety of disciplines ranging frommaterials science to molecular biology. Solid-state NMR is abranch of NMR spectroscopy that deals with solid orsolidlike systems and is presently undergoing rapid expan-sion as a result of the significant advances in both NMRmethodology and instrumentation that have occurred re-cently. To date most successful solid-state NMR applicationsto biological systems have utilized spin-1/2 nuclei such as

13C, 15N, and 31P (Griffiths and Griffin 1993; Evans 1995;Garbow and Gullion 1995). For a large number of biologi-cally important elements, however, the only NMR activeisotopes are those with nuclear spins greater than one half.Such nuclei have a nonspherical charge distribution and areknown as quadrupolar nuclei, for example,17O (S = 5/2),23Na (S= 3/2), 25Mg (S= 5/2), 39K (S= 3/2), 43Ca (S= 7/2),67Zn (S = 5/2), and59Co (S = 9/2), to mention just a few. Itshould be noted that all of these examples are half-integerspins. In fact, there is another type of quadrupolar nuclei forwhich the spin number is an integer such as2H (S = 1), 6Li(S = 1), 14N (S = 1), and10B (S = 3). The integer spins havequite different NMR properties compared with those ofhalf-integer nuclei; consequently, the techniques used to re-

Biochem. Cell Biol.76: 429–442 (1998) © 1998 NRC Canada

429

Received April 14, 1998. Revised June 10, 1998. Accepted June 11, 1998.

Abbreviations: CP, cross polarization; CT, central transition; DAS, dynamic-angle spinning; DOR, double-rotation; NMR, nuclearmagnetic resonance; MAS, magic-angle spinning; MO, molecular orbital; MQMAS, multiple-quantum magic-angle spinning; QCT,quadrupolar central transition; REAPDOR, rotational echo adiabatic-passage double resonance; SFC, sample-fixed coordinate;TRAPDOR, transfer of populations in double resonance.G.Wu. Department of Chemistry, Queen’s University, Kingston, ON K7L 3N6, Canada (e-mail: [email protected]).

cord solid-state NMR spectra of integer nuclei are unique.This review deals only with half-integer quadrupolar nuclei.

The major problem associated with solid-state NMR stud-ies of quadrupolar nuclei stems from the fact thatquadrupolar interactions are often several orders of magni-tude larger than other nuclear spin interactions such as thedipolar or chemical shift interactions. Each of these spin in-teractions depends on the microscopic orientation of themolecule in a magnetic field and as such they may lead tovery broad NMR peaks when polycrystalline samples are ex-amined. On the other hand, there are tremendous advantagesof carrying out NMR studies in the solid state. First, bothchemical shift and quadrupole interactions are anisotropicand best described by second-rank tensors. Characterizationof anisotropic nuclear spin interactions is capable of yieldingmore complete information about three-dimensional elec-tronic structure. In contrast, because of rapid molecular tum-bling motion in solutions, only averaged NMR parameterscan be measured. Second, another effect of rapid moleculartumbling in solutions is to cause efficient quadrupolar relax-ation leading to poor intrinsic resolution, which is a directconsequence of very short lifetimes of the quadrupole en-ergy levels. In solids, quadrupole relaxation times are muchlonger, resulting in NMR spectra of high intrinsic resolution.Therefore, the dilemma of NMR studies of quadrupolar nu-clei is that, although solid-state NMR spectra in principlecontain more information and exhibit higher intrinsic resolu-tion, it has been a challenge for experimentalists to deviseNMR techniques that can yield such high-resolution spectra.

A great deal of effort has been devoted in the last decadeto developing new techniques such as dynamic-angle spin-ning (DAS) (Llor and Virlet 1988; Mueller et al. 1990) anddouble rotation (DOR) (Samoson et al. 1988; Chmelka et al.1989; Wu et al. 1990) to overcome the difficulties encoun-tered in obtaining high-resolution NMR spectra forquadrupolar nuclei. In the meantime, traditional techniques

such as single-crystal NMR (Spiess et al. 1969; Kroeker etal. 1997), wide-line NMR (Baugher et al. 1969; Mooibroeket al. 1986; Power et al. 1990; Cheng et al. 1990),magic-angle spinning (Kundla et al. 1981; Samoson et al.1982; Oldfield et al. 1982a, 1982b), and nutation NMR(Man 1986) continue to contribute to the studies ofquadrupolar nuclei. All of these NMR techniques have se-vere limitations in one way or another; their direct applica-tions to biological systems have been difficult.

Recently, Frydman and Harwood (1995) developed a newmethod for obtaining high-resolution NMR spectra ofquadrupolar nuclei. The technique is known as multiple-quantum magic-angle spinning (MQMAS) NMR. TheMQMAS method has made it possible to obtainhigh-resolution NMR spectra for quadrupolar nuclei underordinary MAS conditions. The development of MQMASmay represent one of the most significant new approaches insolid-state NMR. It is an important step towards the ultimategoal of routinely obtaining solutionlike NMR spectra forquadrupolar nuclei in the solid state and has consequentlystirred tremendous excitement among researchers in theNMR community. It opens the door for extensivehigh-resolution quadrupolar NMR studies of a great varietyof solids including biologically important macromolecules.Recent MQMAS results have clearly added a new dimensionto solid-state NMR spectroscopy (Fernandez and Amoureux1995; Medek et al. 1995; Amoureux et al. 1996; Brown etal. 1996; Massiot 1996; Massiot et al. 1996; Rocha et al.1996; Wu et al. 1996a, 1996b, 1997a, 1997b; Youngman etal. 1996; Brown and Wimperis 1997; Dirken et al. 1997;Duer and Stourton 1997; Fernandez et al. 1997).

Partly inspired by Frydman’s beautiful experiment andpartly because of recent technological advances related tohigh-field NMR instrumentation, a new direction of researchis emerging. Solid-state NMR spectroscopy of quadrupolarnuclei is becoming a realistic and practical approach in thestudy of biological systems. In this review I describe a fewrecent developments in this rapidly expanding field, anddemonstrate the potential of solid-state quadrupolar NMR inthe study of biological systems. In particular, I discuss theapplication of solid-state, NMR of17O, 67Zn, 59Co, 23Na,and 39K nuclei with a prognosis for future work.

In this section I focus only on the recently developedMQMAS approach (Frydman and Harwood 1995). For othertopics related to solid-state NMR of quadrupolar nuclei,readers are referred to several excellent articles (Freude andHaase 1993; Jäger 1994; Chmelka and Zwanziger 1994).

To illustrate the principles of MQMAS spectroscopy, con-sider a quadrupolarS= 3/2 spin system in a strong magneticfield. In the rotating frame, the spin Hamiltonian can bewritten as

[1] H H H HQ Q= + +CS( ) ( )1 2

where HCS describes the chemical shielding andHQ( )1 and

HQ( )2 describe the first- and second-order quadrupole interac-

tions, respectively. An energy level diagram forS = 3/2 isshown in Fig. 1. Here we are only concerned with the sym-

© 1998 NRC Canada

430 Biochem. Cell Biol. Vol. 76, 1998

Fig. 1. Energy level diagram of a quadrupolarS = 3/2 nucleus.

metric (m, –m) coherences; as the evolution of such coher-ences are unaffected by the first-order quadrupoleinteraction, HQ

( )1 can be neglected. Using the fictitiousspin-1/2 formalism (Wokaun and Ernst 1977; Vega 1978),we can writeHCS and HQ

( )2 as (Wu et al. 1996a)

[2] H mA Ilml

Zm m

CSCS= ∑∑

=

−2 00 2,

,

[3] H C A IQ lmml

lQ

Zm m( )

, ,

,2

0 2 40= ∑∑

=

−

where

[4] C S a C llmQ

m m( ) [ ( , , , , ),( )= −−

ωω

2

0

1 2 2 1 1

+ −−a C lm m,( ) ( , , , , )]2 2 2 2 2

and

[5] a m S S mm m,( ) [ ( ) ]− = + − −1 24 1 8 1

[6] a m S S mm m,( ) [ ( ) ]− = + − −2 22 1 2 1

[7] ωQe qQ hS S

=+

2

2 1( )

In the above equations,C(2, 2, l, 1, –1) andC(2, 2, l, 2, –2)are the Clebsch–Gordan coefficients.Al0

CS and AlQ0 describe

the orientation dependence of the chemical shielding andquadrupolar interactions, respectively. Under the sample ro-tation condition, bothAl0

CS and AlQ0 become time dependent

and can be written using the Wigner rotation matrices as

[8] Α τ ∆ τ Αλ µλ

µ λ

λ

λµΘ

0 0CS, SFC CSQ( ) ( ( )),

( ) ,= ′′=−

′∑ Ω

whereΩSFC(t) symbolizes the three Euler angles of the trans-formation to the laboratory frame from a sample fixed coor-dinate (SFC). In general,Al

Q0CS, (t) contains second- and

fourth-rank Legendre polynomials forl = 2 and l = 4, re-spectively. Under the MAS condition,A Q

20CS, vanishes, and

eqs. 2 and 3 become

[9] H mA Im

Zm m

CSCS= ∑ −2 00

,

[10] H C A C A IQ mQ

mQ

m

m

Zm m( ) ,[ ]2

0 00 4 40= +∑ −

Now one can obtain the resonance frequency for the (m, –m)transition as

[11] ν ν ν( , ) ( , ) ( , )m m m m m m− = − + −iso aniso

where

[12] νisoCS( , )m m mA C Am

Q− = +2 00 0 00

[13] νaniso( , )m m C AmQ− = 4 40

It is clear from the above equations thatνaniso is the source ofthe so-called second-order broadening observed in the cen-tral transition (1/2, –1/2) MAS spectra, sinceAQ

40 is not aver-aged to zero by MAS.

Frydman and coworkers noted thatC4m may have an op-posite sign asC4,1/2, which means that the evolution owingto the anisotropic part of the symmetric multiple-quantum(MQ) coherence may be reversed if the MQ coherence istransferred to the central transition (CT). In other words, theMQ and CT spectra are related by a pseudoreflection sym-metry with a scaling factor,k = |C4m/C4,1/2|, as also illus-trated in Fig. 1. (ForS = 3/2, k = 7/9.)

The basic two-pulse pulse sequence for MQMAS experi-ments is shown in Fig. 2, together with the coherence trans-fer pathway. The first pulse, P1, is to excite the desirableMQ coherence. During the time,t1, the MQ coherenceevolves with the frequency ofν(m, –m). The MQ coherenceis then converted by the P2 pulse to the CT for detection.Since the CT evolves in an opposite sense from that of theMQ, the dephasing of the MQ coherence can be refocusedafter the MQ is converted to the CT coherence. Conse-quently, an echo will form att2 = kt1 and the echo intensitywill be independent of the second-order quadrupolebroadenings. Most importantly, whereas the dephasing ow-ing to the anisotropic part of the second-order quadrupoleinteraction,νaniso(m, –m), is refocused at the echo maximum,information about the isotropic part of the second-orderquadrupole interaction and the isotropic chemical shift is re-tained in the MQMAS experiment. Therefore, isotropicpeaks with different peak positions will be observed inMQMAS spectra for chemically or crystallographically non-equivalent sites once the echo maxima are recorded andFourier transformed. The coherence transfer pathway is se-lected by a proper phase cycling (Frydman and Harwood1995). Of course, it is also possible to use the field-gradienttechnique to select the desirable coherence transfer pathway.

An example is shown in Fig. 3. For a stationarypolycrystalline sample of sodium citrate trihydrate, the23Na(S = 3/2) NMR spectrum consists of a featureless line shapeapproximately 6 kHz wide. Under the MAS condition, theNMR spectrum becomes narrower but exhibits some over-lapping features, from which very little useful informationcan be extracted. With the MQMAS technique, the NMRspectrum exhibits three isotropic peaks, indicating threecrystallographically distinct sodium sites. MQMAS can alsobe applied toS = 5/2 nuclei. Solid-state17O (S = 5/2) NMR

© 1998 NRC Canada

Review / Synthèse 431

Fig. 2. The basic two-pulse sequence and coherence transferpathway in MQMAS experiments

© 1998 NRC Canada


spectra of several inorganic phosphates are shown in Fig. 4.Clearly, the dramatic resolution improvement in theMQMAS spectra permits the detection of subtle differencesin various oxygen sites. The conclusion derived from the17O MQMAS spectra is consistent with that from the crystal-lographic symmetry (Wu et al. 1997a).

Solid-state NMR spectroscopy of quadrupolar nuclei, inparticular MQMAS, is rapidly expanding. Despite its over-whelming initial success, the frontiers of solid-statequadrupolar NMR are replete with unsolved problems, manyof which are related to the fundamentals of NMR. Future re-search on the development of new methodology for obtain-ing high-resolution NMR spectra for quadrupolar nuclei maybe focused in the following two areas: (i) sensitivity im-provement of MQMAS experiments and (ii ) internuclear dis-tance determination involving quadrupolar nuclei.

The utility of solid-state NMR spectroscopy as a tool forstructure determination in biological systems has produced anew direction of research during the last decade, but suc-cessful applications have been limited to spin-1/2 systemssuch as those of13C, 15N, and 31P (Griffiths and Griffin1993; Garbow and Gullion 1995). Although there have beenrecent efforts to address the problem associated withquadrupolar nuclei such as the development of transfer ofpopulations in double resonance (TRAPDOR; Grey andVeeman 1992) and rotational echo adiabatic-passage doubleresonance (REAPDOR; Gullion 1995), it still remains achallenge to determine internuclear distances in spin pairsinvolving quadrupolar nuclei (e.g.,1H–17O and 13C–17O).Research in this direction would be extremely important,since solid-state NMR potentially can pinpoint the localstructure around a metal-binding pocket for noncrystallinebiological samples.

Oxygen is a key constituent of many important functional

groups in biological systems. Oxygen-17 (S = 5/2, naturalabundance 0.037%) is the only naturally occurring, stableoxygen isotope with nonzero nuclear spin, which makes itsuitable for NMR studies. Despite the enormous importanceof oxygen-containing compounds in chemistry and biology,relatively few17O NMR studies have been reported (Boykin1991) compared with the applications of1H and 13C NMR.The primary impediments of17O NMR spectroscopy includelow natural abundance of17O nuclei and the fact that17O isa quadrupole nucleus (Q = –2.6 × 10–30 m2). These make17O an extremely insensitive NMR nucleus (approximately105 times less sensitive than1H). Therefore, it is not surpris-ing that most of the17O NMR studies so far reported havedealt only with solution samples of small molecules.

Solid-state17O NMR applications have been rare andlargely limited to studies of inorganic materials in which17O quadrupole coupling constants are reasonably small,e2qQ/h < 6 MHz (Schramm and Oldfield 1984; Walter etal. 1988; Chmelka et al. 1989; Mueller et al. 1991;Grandinetti et al. 1995; Youngman et al. 1995; Bastow et al.1996). Meanwhile,17O NMR of organic–biological solidsremains essentially an undeveloped research field. Besidesan early single-crystal17O NMR study of an organic com-pound, benzophenone (Scheubel et al. 1985), there havebeen only a few reports on systems of biological interest.Oldfield and coworkers investigated the interactions between17O-labeled CO and O2 ligands and hemoproteins and syn-thetic model compounds (Park et al. 1991; Oldfield et al.1991). For the Fe17O2 moiety, they observed remarkably large17O chemical shift tensors but rather small quadrupole ten-sors for both the bridging and terminal oxygen atoms. A similartrend was observed for the CO ligand upon coordination tothe Fe center.Gann et al.(1994) demonstrated theutility ofcross-polarization(CP) DAS in resolving two nonequivalent ox-ygen atoms in17O-labeled L-alanine. The two nonequivalentoxygen atoms are due to the differenthydrogen-bonding envi-ronments. In this study, CP from1H to 17O was achievedwhen the sample was spun about a direction parallel to the

Fig. 3. Solid-state23Na NMR spectra of sodium citrate trihydrateat 9.4 T (G. Wu, D. Rovnyak, and R.G. Griffin, unpublishedresults).

Fig. 4. Solid-state17O NMR spectra of several inorganicphosphates: (a) Ca5(P

17O4)3(OH), (b) CaHP17O4·2H2O,(c) KH2P

17O4, and (d) NH4H2P17O4.

external magnetic field. It should be emphasized that sensi-tivity enhancement by CP from1H to quadrupolar nuclei un-der the MAS condition remains one of the most importantbut unsolved problems in solid-state NMR (Vega 1992a,1992b). Kuroki et al. (1994) used solid-state17O NMR tocharacterize both17O quadrupole and chemical shift tensorsin a series of dipeptides and polypeptides. However, the ori-entation of the17O chemical shift tensor in Gly-Gly appearsto be inconsistent with the orientation found in benzophe-none (Scheubel et al. 1985). Further studies are clearly re-quired to resolve this discrepancy. Zhang et al. (1996) alsoreported a single-crystal17O NMR study of crystalline hy-drates. Since17O chemical shift anisotropy is known to besmall in water molecules, only the17O quadrupolar interac-tion was considered in the analysis of the single-crystalspectra.

Recently, several17O MQMAS studies have also beenpublished (Wu et al. 1996a, 1997a; Dirken et al. 1997;Stebbins and Xu 1997). Figure 5 shows17O NMR spectra ofBa(ClO3)2·H2

17O obtained under static MAS and MQMASconditions. At 9.4 T (400 MHz for1H nuclei), the static17ONMR spectrum exhibits a 55-kHz broad line shape arisingfrom the second-order quadrupole interaction. Rapid samplespinning at 20 kHz significantly reduces the second-orderquadrupole broadening. Analysis of the static and MASspectra yieldse2qQ/h = 6.8 MHz andη = 1.0. As seen inFig. 5, the17O MQMAS spectrum exhibits an isotropic peakindicating only one water molecule in the asymmetric unitcell. The 100-Hz line width of the17O MQMAS spectrum isabout 150 times narrower than that of the17O MAS spec-trum. My collaborators and I also obtained the17O MQMASspectrum ofL-alanine (Wu et al. 1997c), which exhibits twoisotropic peaks indicating two nonequivalent oxygen atoms,consistent with the DAS result noted earlier (Gann et al.1994).

It is clear from the aforementioned studies that solid-state17O NMR studies are indeed feasible, but there also exist se-rious sensitivity problems especially in MQMAS experi-ments. It should be noted that all of these17O MQMASstudies were performed at low field strength, 9.4 T, on sam-ples with an17O enrichment level of≤ 50%. We anticipate

that with a combination of high magnetic fields (e.g., 18.8 T),fast sample spinning (e.g., >25 kHz), and NMR probes thatcan deliver high radiofrequency (RF) power,17O MQMASNMR will be feasible and useful in the study of biologicallyimportant macromolecules. The great advantage of17ONMR spectroscopy lies in the remarkable sensitivity of17ONMR parameters to molecular structure and hydrogen bond-ing effects. Future solid-state17O NMR studies in the fol-lowing areas are likely to be important.

Oxygen-17 chemical shift tensorsDespite numerous reported empirical correlations con-

cerning17O isotropic chemical shift and molecular structure,very little is known about the tensorial nature of this funda-mental NMR parameter, the17O chemical shift tensor. Aknowledge of the complete17O chemical shift tensor (ratherthan only its trace) is essential for understanding the rela-tionship between17O chemical shift and molecular structure.Furthermore, recent developments in ab initio molecular or-bital (MO) shielding calculations have made calculations of17O chemical shielding possible. However, the quality of17Ochemical shielding results calculated by the state-of-the-artMO methods is uncertain, simply because very little experi-mental data are available for17O chemical shift tensors.Clearly it is necessary to carry out systematic measurementsof 17O chemical shift tensors in systems with biologicallyrelevant functional groups such as amide, carboxyl, phenol,and hydroxyl groups. It is of fundamental importance to es-tablish correlation between17O chemical shift tensor andmolecular structure. A similar trend has emerged insolid-state13C NMR studies of proteins (Le et al. 1995). Todetermine an17O chemical shift tensor completely (i.e., bothits magnitude and its orientation in the molecular frame) andto understand better this NMR parameter, both single-crystalNMR and ab initio quantum mechanical shielding calcula-tions will be helpful.

Hydrogen-bonded systemsHydrogen bonding effects are fundamental to the structure

and function of many biological systems. The fact that bothisotropic 17O chemical shift and quadrupole coupling aresensitive to intra- and inter-molecular hydrogen bonding hasbeen previously observed (Boykin 1991). However, I expectthat characterization of the dependence of the entire17Ochemical shift tensor on hydrogen-bonding effects will rep-resent a promising route to delineating completethree-dimensional hydrogen-bond networks. A recent theo-retical study has indicated that the17O chemical shift tensorin a model dipeptide, Gly-Gly, is sensitive to the state ofhydration, whereas only small effects are noted in the13Cand 15N chemical shift tensors (Chestnut and Phung 1993).Of particular interest is the possibility of using17O NMR todetect the protonation state of amino acid side groups (e.g.,the carboxylates) in proteins.

Binding of small ligands to biomoleculesApp l i ca t i on o f so lu t i on17O NMR to the s tudy

o f e n z y m e – s u b s t r a t ecomplexes was demon-strated some time ago (Wisner et al. 1985). However,17O NMR spectra of enzyme–substrate complexes are of-ten plagued by broad lines resulting from efficientquadrupole relaxation, especiallywhen the l igand is

© 1998 NRC Canada


Fig. 5. Solid-state17O NMR spectra of Ba(ClO3)2·H217O (with

50% 17O enrichment) at 9.4 T.

tightly held. Solid-state NMR is anideal technique forstudying small ligands bound to macromolecules and is notas severely limited by the molecular weight of the complexunder study as is solution17O NMR. It will be potentiallyimportant to explore the utility of17O-labeled ligands as areporter to probe the local chemical environment at the bind-ing site. For example, solid-state17O NMR studies of nucle-otide–enzyme complexes will be intriguing and feasible,since17O-labeled nucleotides can be prepared.

The extremely low natural abundance of the17O isotope(0.037%) excludes the possibility of obtaining NMR spectraof biological systems with naturally occurring17O nuclei.Future research will require site-specific17O labeling.Therefore, it is necessary to develop efficient syntheticroutes for introducing17O onto important functional groups.Fortunately, many methods have been developed over theyears for17/18O labeling a variety of organic and biologicalcompounds such as amino acids (Steinschneider et al. 1985;Kabalka and Goudgaon 1991), amino acid side chains(Eckert and Fiat 1986), polypeptides (Gilboa et al. 1984;Sakarellos et al. 1989), nucleic acid bases (Petersheim et al.1983; Chandresekaran et al. 1985; Amantea et al. 1996), andduplex DNA (Wilde et al. 1989).

The future promise of solid-state17O NMR lies in thepossibility of its becoming an additional nuclear probe forstudying biological systems. This is a new and exciting fron-tier of considerable potential. Research along this line mayprovide a unique approach to the study of molecular struc-ture and dynamics.

It has long been recognized that metal ions are vital tomany biological processes. Recent discoveries about the cru-cial role of metal ions in gene expression, in RNA catalysis,and in anticancer drugs have had tremendous impact on avariety of problems ranging from metalloprotein andribozyme chemistry to medicine (Lippard and Berg 1994).With only a few notable exceptions, quadrupolar metal ionstightly bound to biological macromolecules have been NMRinvisible in liquid media (Forsén et al. 1987; Sanders andTsai 1989). This is because efficient nuclear quadrupole re-laxation in solutions often results in undetectably broadNMR peaks. Therefore, it is commonly accepted lore thatsite resolution from solution NMR spectra of quadrupolarnuclei is always elusive.

However, Vogel and coworkers recently demonstratedthat for large metalloproteins (ca. 80 kDa), it is actuallypossible to achieve sufficient site resolution in solutionNMR spectra of quadrupolar nuclei at high magnetic fields(Aramini et al. 1993a, 1993b, 1994; Aramini and Vogel1994; Germann et al. 1994). For large metalloproteins at theso-called slow-motion regime (i.e.,ω0τc > > 1), the linewidth of the central transition decreases with increasing ap-plied magnetic field. Consequently, relatively sharp signalscan be obtained at high magnetic fields (greater than 11.7 T)for metal ions tightly bound to a large protein. The approachwas referred to as quadrupolar central transition (QCT)spectroscopy. It is interesting to compare QCT withsolid-state NMR experiments. As mentioned earlier, in

solid-state NMR of quadrupolar nuclei, one is also focusedon the central transition whose line width is influenced byquadrupole interactions only to the second order. Further-more, the second-order quadrupole broadening is inverselyproportional to the applied magnetic field. Thus it is not sur-prising that both QCT and solid-state NMR experimentsshould be performed at the highest magnetic field possible.Solid-state NMR experiments may yield more informationabout other anisotropic nuclear spin interactions such aschemical shift ansiotropy and heteronuclear dipolar interac-tions, which may contain valuable structural information. Onthe other hand, additional information may overcrowdsolid-state NMR spectra, making their interpretation diffi-cult. Nevertheless, it may prove advantageous to apply bothQCT and solid-state NMR techniques in the study of largemetalloproteins. In this section, I briefly discuss the presentsituation of solid-state metal NMR with emphasis on studiesof 67Zn, 59Co, 23Na, and39K nuclei.

Zinc is among the most important metal ions in biology.Zinc(II), as a diamagneticd10 ion, is often referred to as be-ing spectroscopically silent, since one cannot use EPR, UV,and visible spectroscopies to probe the nature of the Zn(II)site in an enzyme. In addition, the unfavorable NMR proper-ties of 67Zn (S = 5/2, natural abundance 4.11%,Q = 0.16 ×1028 m–2) nuclei have prevented researchers from exploiting67Zn as a nuclear probe to biological systems. In fact,67ZnNMR studies are scarce even for small compounds (Granger1991). Direct NMR detection of the metal binding sites inzinc-containing proteins has entirely relied on the utility of asurrogate nuclear probe,113Cd, which is an NMR-friendlyspin-1/2 nucleus (McAteer et al. 1996). However, the ques-tion as to the validity of using cadmium to model the zincion environment remains partially unanswered, because theconnection between zinc ion environments and67Zn NMRparameters in native zinc metalloproteins has not yet beenestablished.

For several zinc-containing compounds in which the cubiccrystallographic symmetry at the zinc site results in negligi-ble effects from quadrupolar interactions,67Zn NMR param-eters were reported in the solid state (Haller et al. 1980; Wuet al. 1995). In the case of K2Zn(CN)4, indirect spin–spin (J)coupling between13C and 67Zn nuclei is observed in the67Zn MAS spectrum,1J(13C, 67Zn) = 88 Hz, which appearsto be the onlyJ-coupling constant known involving67Zn(Wu et al. 1995). There is only one solid-state67Zn NMRstudy on compounds with noncubic symmetry. Using aspin-echo technique, Oldfield and coworkers recorded astatic 67Zn NMR spectrum of zinc acetate dihydrate at11.75 T (Kunwar et al. 1986). It is seen in Fig. 6 that thestatic line shape arising from the second-order quadrupolarinteraction is about 50 kHz wide, indicating a largequadrupole coupling constant. As also shown in Fig. 6, thecrystal structure of zinc acetate dihydrate (van Niekerk et al.1953) exhibits a very distorted octahedral zinc site, which isresponsible for the large, asymmetric67Zn quadrupole ten-sor, e2qQ/h = 5.3 MHz andη = 0.87. In an early solution67Zn NMR study of the zinc(II)-insulin complex, Shimizuand Hatano (1983) estimated the67Zn quadrupolar coupling

© 1998 NRC Canada


constant for the zinc sites in insulin to be 1.86 MHz. How-ever, there is serious suspicion about this estimated value,considering that each of the two zinc sites in the two-zincinsulin hexamer is coordinated by three imidazolyl nitrogenatoms and three water molecules (Adams et al. 1969). Re-gardless, assuminge2qQ/h = 1.86 MHz and a magnetic fieldof 11.7 T, one would expect a line width of approximate4 kHz in static67Zn NMR spectra of insulin. This is signifi-cantly narrower than that observed for zinc acetatedihydrate.

The unique feature of the zinc active site in allmononuclear zinc metalloenzymes characterized to this dateis an unsymmetrical tetrahedral geometry with one of theligands being a water molecule. However, the67Zn quad-rupole coupling constant is unavailable for any complexwith unsymmetrical tetrahedral coordination. Pentacoordi-nated zinc sites have also been found in enzymes containingtrinuclear metal clusters. To assess the feasibility of usingsolid-state67Zn NMR in the study of zinc metalloproteins, itis necessary to obtain fundamental67Zn NMR parameterssuch as quadrupolar and chemical shift tensors in model systemsof typical ion coordination geometries. To this end, single-crystal

67Zn NMR studies are underway in our laboratory. Tocircumvent the low sensitivity of67Zn NMR, isotopic67Znenrichment may be necessary in future solid-state67Zn NMRstudies, especially for large zinc–biomolecule complexes.Given the similar NMR properties of67Zn and 43Ca (S =7/2) nuclei, it is perhaps worth mentioning that natural abun-dance solid-state43Ca NMR studies of inorganic compoundswere recently reported by Dupree et al. (1997).

Cobalt is one of the so-called ultratrace metals found inmetalloenzymes. Cobalt is present in vitamin B12 (Fig. 7)and its coenzyme, which is the first naturally occurringorganometallic compound ever identified. Cobalt-59 (S =7/2, natural abundance 100%) is the only stable isotope ofcobalt and is also one of the most sensitive nuclei for NMRmeasurements. One would think naturally to apply59CoNMR to biological systems because of the apparent interestin vitamin B12 chemistry. However, except for an early sin-gle-crystal 59Co NMR study by Spiess et al. (1969),applications of solid-state59Co NMR have been largely re-stricted to simple compounds with highly symmetric cobaltenvironments (Reynhardt 1974a, 1974b; Eaton et al.1987;Chung et al. 1993; Hayashi 1996; Eichele et al. 1997). Thisis because both quadrupole interaction and chemical shift of59Co nuclei are highly anisotropic, often resulting in verybroad NMR lines for polycrystalline samples.

Not until very recently have researchers attempted totackle more complicated systems with solid-state59CoNMR. In a study of hexacoordinated cobalt porphyrin com-plexes, Medek et al. (1997a) reported59Co NMR spectra fora series of polycrystalline compounds under both static andfast MAS conditions at different magnetic fields. SinceCo(III) porphyrin complexes are isoelectronic to Fe(II)hemochromes, they may be used as model systems in under-standing the electronic structures in the latter compounds.Cobalt-59 NMR is capable of yielding useful informationabout the electronic structure at the metal center. It wasfound that the Co(III) porphyrin complexes exhibit rathersmall 59Co quadrupole coupling but very large chemicalshift anisotropy. Interestingly, the relation between59Coquadrupole coupling and chemical shift tensors found byMedek et al. (1997a) deviates considerably from that pre-viously known for simple octahedral cobalt complexes.The origin of such a discrepancy is unclear at this time.

Medek et al. (1997b) also investigated a series of natu-rally occurring cobalamins and synthetic cobaloximes. Inparticular, they obtained solid-state59Co NMR spectra ofstatic samples of vitamin B12, the B12 coenzyme,methylcobalamin, and dicyanocobyrinic acidheptamethy-lester.They found that the NMR parameters are sensitive tothe type of ligands attached to the metal and to the crystalli-zation history of the sample. In addition, they studied sev-eral synthetic cobaloximes of alkyl, cyanide, aquo, andnitrogenated axial ligands. All of the reported solid-state59Co NMR spectra exhibit broad lines whose widths rangebetween 100 and 400 kHz as a result of a combination ofchemical shift anisotropy and second-order quadrupolar in-teractions.

© 1998 NRC Canada


Fig. 6. (Top) Static67Zn NMR spectrum of Zn(CH3COO)2·2H2Oobtained at 11.7 T (Kunwar et al. 1986). Reproduced withpermission. (Bottom) Coordination geometry of the zinc ion inZn(CH3COO)2·2H2O (van Niekerk et al. 1953). Selected bonddistances and angles are as follows: Zn–O(1) = 2.18, Zn–O(2) =2.17, Zn–O(3) = 2.14 Å (1 Å = 0.1 nm); O(1)–Zn–O(2) = 61°,O(1)–Zn–O(3) = 103°, O(1)–Zn–O(1A) = 158°, O(2)–Zn–O(3A) =152°.

As Spiess et al. (1969) demonstrated, single-crystal59CoNMR spectra exhibit relatively narrow lines and can be usedto study systems with large anisotropic spin interactions. Inaddition, single-crystal NMR is the method of choice for un-ambiguous determination of a nuclear spin interaction tensorin the molecular frame. A beautiful example is the recentsingle-crystal59Co NMR study of vitamin B12 by Power etal. (1998). Analysis of the single-crystal NMR spectra yieldsboth the chemical shift and quadrupole coupling tensors ofthe 59Co nucleus:δ11 = 5075, δ22 = 4670, δ33 = 3901 ppm;e2qQ/h = 27.3 MHz andη = 0.24. These results are in excel-lent agreement with those obtained by Medek et al. (1997b)from a study of polycrystalline vitamin B12 samples. Mostimportantly, the single-crystal NMR study of Power et al.(1998) also yields the orientations of the two59Co NMR ten-sors in the molecular frame. As shown in Fig. 8, the mostshielded component of the chemical shift tensor,δ33, isfound to point approximately through the center of one tri-angular face of the octahedral cobalt coordination. The leastshielded component,δ11, is directed toward the phosphorusatom, whereas theδ22 component is in the corrin plane. Theorientation of the quadrupole coupling tensor is such that thelargest component of the quadrupole coupling tensor devi-ates by approximately 10° from the Co–N direction, whereasthe other two components lie within the plane of the corrinring. This is the first report of a single-crystal NMR study ofa metal center in a biological molecule. The exciting resultsfrom these solid-state59Co NMR studies will surely encour-age further NMR investigations of quadrupolar metal iso-topes in biological systems.

Unlike divalent metal ions, monovalent ions such as alkalimetal cations generally exhibit weak association to biologi-cal macromolecules. Their roles have been thought primarilyto be as bulk electrolytes that stabilize surface charges on

proteins and nucleic acids. However, recent findings re-vealed a unique structural role of potassium and sodium ionsin the form of G-quartet structures in telomeric DNAs, in-spiring new views about the role that alkali metals may playin biological processes (Williamson 1994). Telomeres arelocated at the ends of linear eukaryotic chromosomes andconsist of tandem arrays of G-rich sequences. The mainstructural motif of such G-rich DNAs is known as theG-quartet (Williamson 1994). The G-quartet structure, as de-picted in Fig. 9, consists of four guanine base pairs con-nected by the Hoogsteen pattern of hydrogen bonds. Theremarkable stability of the G-quartet structure arises fromthe presence of monovalent cations, usually K+ or Na+. Thecation-binding site in the G-quartet is thought to be at thecenter of a cavity formed by the eight O6 oxygen atomsfrom two stacking G-quartets (Sundquist and Klug 1989),similar to the K+-stabilized structure of 5′-GMP (Detellierand Laszlo 1980). A recent X-ray crystallographic study ofd(GGGGTTTTGGGG) (Kang et al. 1992) essentially con-firmed this model of ion coordination environment; how-ever, the crystal structure of d(TGGGGT) also revealed

© 1998 NRC Canada


NH2

HO

O

N

NH2

O

N

NH2

O

O O

OP

O

HN

H2N

O

O

H2N

O

2N

O

N

N

N

N

Co

NC

OH

Fig. 7. Vitamin B12. Fig. 8. Orientations of the59Co chemical shift tensor (in green)and EFG tensor (in blue) in vitamin B12 (courtesy of W.P.Power, University of Waterloo, Waterloo, Ont.).

some variations in ion-binding environments (Laughlan etal. 1994). Interestingly, a total of seven Na+ ions are foundin d(TGGGGT), as shown in Fig. 10. The seven sodium ionsare located between and within planes of hydrogen-bondedguanine quartets. Each Na+ ion, except Na(1) and Na(1′), iscoordinated by eight guanine O6 atoms from the two stack-ing G-quartet planes. Na(1) and Na(1′) each is coordinatedby four guanine O6 atoms and one water molecule at the ax-ial position, giving rise to a square pyramidal geometry.

Another interesting aspect of ion binding to G-DNAs isthe ion-binding stoichiometry. It is conceivable that alkalimetal ions can bind to a G-DNA with differentstoichiometries. For example, while there is only one K+ ionlocated at the center of four stacking G-quartets ind(GGGGTTTTGGGG) (Kang et al. 1992), two Na+ ions arededuced from solution23Na NMR study of the same DNA(Deng and Braunlin 1996). As noted earlier, there are sevenNa+ ions interacting with the eight G-quartets ind(TGGGGT) (Laughlan et al. 1994). Recently, Hud et al.(1996) showed that there are at least two ions (either Na+ orK+) bound to the three G-quartets in d(GGGTTTTGGG). Itremains unclear what factors determine the ion-bindingstoichiometry in G-DNAs.

Although solution NMR of alkali metal ions was used inthe early NMR applications to self-assembled systems(Detellier and Laszlo 1980), solid-state alkali metal NMRhas found very little application in biological systems. Thisis partially because the small chemical shift range andquadrupole broadening of alkali metal nuclei often preventone from observing separate signals from different bindingsites. However, with a combination of high magnetic fieldsand the recently developed MQMAS technique, it may bepossible to obtain solid-state NMR spectra of resolution suf-ficient to resolve different ion-binding sites.

Because of the striking similarity in ion coordination en-vironment between G-DNAs and antibiotic ionophores(Dobler 1981), the latter should serve as excellent modelsystems for establishing a correlation between alkali metalNMR parameters and ion-binding geometry. This will be ofobvious importance to further NMR studies of G-DNAs andrelated self-assembled systems. For instance, the Na+ ion innonactin–NaSCN complex (Dobler and Phizackerley 1974)is coordinated by eight oxygen atoms as depicted in Fig. 11.An early solution NMR relaxation study of nonactin–Na+

complex by Saitô and Tabeta (1987) yielded an estimate forthe 23Na quadrupole coupling constant,e2qQ/h ≈ 0.8 MHz.Since Na(4) of d(TGGGGT) is equidistantly coordinated byall eight O6 atoms, it is expected thate2qQ/h < 0.8 MHz forthis site. It is also known that for Na+ ions with square py-ramidal coordination geometry,23Na quadrupole couplingconstants are-2.8 MHz (Koller et al. 1994), which shouldbe a reasonable estimate for the Na(1) and Na(1′) sites. Sev-eral years ago, Saitô and coworkers also carried out exten-sive solid-state23Na NMR studies of sodium complexeswith ionophores (Tabeta et al. 1986); unfortunately, thequality of the experimental data, obtained with techniquesavailable at the time, not only prevented them from obtain-ing any useful23Na quadrupole parameters, but also led to

© 1998 NRC Canada


Fig. 9. The G-quartet structure.

Fig. 10. Sodium ion binding in d(TGGGGT) (Laughlan et al.1994). Na(4) is equidistantly coordinated by eight oxygen atomswith an average Na–O distance of 2.75 ± 0.02 Å. Na(2), Na(2′),Na(3), and Na(3′) are between two stacked G-quartets withaverage Na–O distances of 2.89 ± 0.07 and 2.60 ± 0.07 Å tofarther and closer G-quartets, respectively. Na(1) and Na(1′) arelocated within the G-quartet planes, and each is coordinated by awater molecule at the axial position (crystallographic dataobtained from the Brookhaven Data Bank).

erroneous23Na chemical shifts. Clearly, a reexamination ofalkali metal – ionophore complexes is in order.

On the basis of the crystal structures of d(TGGGGT) andd(GGGGTTTTGGGG) described above, it is anticipated thatK+ binding sites in G-DNAs are generally of high symmetry,thus having a small39K quadrupole coupling constant, andthat these K+ ions may be readily accessible by solid-state39K NMR. It is worth noting that solid-state39K NMR stud-ies have been successful in systems with more distorted ioncoordination environments,e2qQ/h = 0.87–3.2 MHz(Kunwar et al. 1986; Kim et al. 1996). Since similar alkalimetal stabilized G-quartet structures also exist in RNAs(Kim et al. 1991; Cheong and Moore 1992), it is also possi-ble to apply solid-state39K and 23Na NMR to the study ofion-binding environments in G-rich RNAs.

Recently, a K+-specific binding site was structurally char-acterized for the first time in a protein, dialkylglycine decar-boxylase (Toney et al. 1993). Its high-resolution crystalstructure revealed that the K+ ion is coordinated by six oxy-gen atoms in octahedral geometry, similar to those found inmany K+-selective ionophores such as valinomycin (Hamil-ton et al. 1981). Interestingly, the39K quadrupole couplingconstant in valinomycin is only 1.2 MHz (Neurohr et al.1983), which suggests that it may be possible to use

solid-state39K NMR to study ion binding in K+-transportingproteins.

The recent invention of MQMAS spectroscopy byFrydman and coworkers has triggered a new wave of re-search interest in the study of quadrupolar nuclei. The exam-ples described in this review are a survey of what ispresently occurring in this research area. These initial resultsindicate that new areas of research are becoming available.One of the most promising aspects of solid-state NMR ofquadrupolar nuclei, especially of MQMAS, is to provide ameans of directly observing individual binding sites inmetalloproteins consisting of multiple sites. Since mostquadrupolar nuclei, especially metal ions, have been previ-ously inaccessible in biological solids, future research in thisarea may provide a unique approach to the study ofion-binding structures, and ultimately to the revelation ofbinding specificity and functional roles (structural or cata-lytic) of metal cofactors in biological systems. In the mean-time, since quadrupolar nuclei occupy over 70% of theNMR periodic table, effort in developing new NMR method-ology for studying quadrupolar spin systems will have sig-

© 1998 NRC Canada


Fig. 11. Ion-coordination geometry of the Na+ (a) and K+ (b) complexes of nonactin. The carbonyl and ether oxygen atoms are labeledby shaded and open circles, respectively.

nificant impact on many research areas in chemistry andbiology. In any case, researchers should not limit themselvesto only one particular type of NMR technique. As I haveshown in this review, different solid-state NMR techniquesranging from single-crystal NMR and MAS to MQMAScould all be useful to the study of biological systems.

Significant advances in NMR methodology and continu-ing development of high-field, high-sensitivity NMR instru-mentation will permit increasingly complex biologicalsystems to be probed by quadrupolar NMR. We anticipate arapid evolution of this new research field. At high magneticfields, NMR of quadrupolar nuclei may possibly become asroutine as NMR of spin-1/2 nuclei such as13C or 15N. Thisis a truly compelling prospect, since the latter has alreadyhad a major impact on chemistry, biology, and materials sci-ence.

The author thanks Prof. Bob Griffin and all his other col-leagues at the Massachusetts Institute of Technology, Cam-bridge, Mass., where the author’s interest in the researcharea covered in this review was initiated. The author is alsograteful to Prof. Bill Power (University of Waterloo,Waterloo, Ont.) for providing single-crystal59Co NMR re-sults of vitamin B12 prior to publication. Financial supportfrom Queen’s University in the form of a Faculty InitiationGrant is acknowledged.

Adams, M.J., Blundell, T.L., Dodson, E.J., Dodson, G.G., Vijayan,M., Baker, E.N., Harding, M.M., Hodgkin, D.C., Rimmer, B.,and Sheat, S. 1969. Structure of rhombohedral 2 zinc insulincrystals. Nature (London),224: 491–495.

Amantea, A., Henz, M., and Strazewski, P. 1996. Synthesis of(15N2)[

17O]urea, (15N2)[O2,O4-17O2]uridine, and (15N3)[O

2-17O]cytidine. Helv. Chim. Acta,79: 244–254.

Amoureux, J.-P., Fernandez, C., and Frydman, L. 1996a. Opti-mized multiple-quantum magic-angle spinning NMR experi-ments on half-integer quadrupoles. Chem. Phys. Lett.259:347–355.

Amoureux, J.-P., Fernandez, C., and Steuernagel, S. 1996b. Z fil -tering in MQMAS NMR. J. Magn. Reson. Ser. A,123: 116–118.

Aramini, J.M., and Vogel, H.J. 1993. Aluminum-27 and carbon-13NMR studies of aluminum(3+) binding to ovotransferrin and itshalf-molecules. J. Am. Chem. Soc.115: 245–252.

Aramini, J.M., and Vogel, H.J. 1994. A scandium-45 NMR studyof ovotransferrin and its half-molecules. J. Am. Chem. Soc.116:1988–1993.

Aramini, J.M., Germann, M.W., and Vogel, H.J. 1993.Field-dependent aluminum-27 NMR studies of the transferrins:an approach for the study of metal ion binding sites in large pro-teins. J. Am. Chem. Soc.115: 9750–9753.

Aramini, J.M., McIntyre, D.D., and Vogel, H.J. 1994. Gallium(3+)binding to ovotransferrin and its half-molecules: a multinuclearNMR study. J. Am. Chem. Soc.116: 11 506 – 11 511.

Bastow, T.J., Dirken, P.J., Smith, M.E., and Whitfield, H.J. 1996.Factors controlling the17O chemical shift in ionic mixed metaloxides. J. Phys. Chem.100: 18 539 – 18 545.

Baugher, J.F., Taylor, P.C., Oja, T., and Bray, P.J. 1969. Nuclearmagnetic resonance powder patterns in the presence of com-

pletely asymmetric quadrupole and chemical shift effects:application to metavanadates. J. Chem. Phys.50: 4914–4925.

Boykin, D.W. 1991. Oxygen-17 NMR spectroscopy in organicchemistry. CRC Press, Inc., Boca Raton, Fla.

Brown, S.P., and Wimperis, S. 1997. Two-dimensional MASmultiple-quantum NMR of quadrupolar nuclei. A comparison ofmethods. J. Magn. Reson.128: 42–61.

Brown, S.P., Heyes, S.J., and Wimperis, S. 1996. Two-dimensionalMAS multiple-quantum NMR of quadrupolar nuclei. Removalof inhomogeneous second-order broadening. J. Magn. Reson.Ser. A, 119: 280–284.

Chandresekaran, S., Wilson, W.D., and Boykin, D.W. 1985.17ONMR studies on 5-substituted uracils. J. Org. Chem.50:829–831.

Cheng, J.T., Edwards, J.C., and Ellis, P.D. 1990. Measurement ofquadrupolar coupling constants, shielding tensor elements, andthe relative orientation of quadrupolar and shielding tensor prin-cipal axis systems for rubidium-87 and rubidium-85 nuclei inrubidium salts by solid state nuclear magnetic resonance. J.Phys. Chem.94: 553–561.

Cheong, C., and Moore, P.B. 1992. Solution structure of an unusu-ally stable RNA tetraplex containing G- and U-quartet struc-tures. Biochemistry,31: 8406–8414.

Chestnut, C.B., and Phung, D.G. 1993. A model study of chemicalshielding in a partially hydrated dipeptide.In Nuclear magneticshieldings and molecular structure.Edited by J.A. Tossell.NATO ASI Series C. No. 386. pp. 221–242.

Chmelka, B.F., and Zwanziger, J.W. 1994. Solid state NMR linenarrowing methods for quadrupolar nuclei: double rotation anddynamic-angle spinning.In NMR basic principles and progress.Vol. 33. Edited byP. Diehl, E. Fluck, H. Günther, R. Kosfeld,and J. Seelig. Springer-Verlag, Berlin. pp. 79–124.

Chmelka, B.F., Mueller, K.T., Pines, A., Stebbins, J., Wu, Y., andZwanziger, J.W. 1989. Oxygen-17 NMR in solids by dy-namic-angle spinning and double rotation. Nature (London),339: 42–43.

Chung, S.C., Chan, J.C.C., Au-Yeung, S.C.F., and Xu, X. 1993.Polycrystalline59Co NMR studies of metal–ligand interaction inaxially symmetric diamagnetic Co(III) complexes: correlation ofδ(59Co) with NQCC/∆Eav. J. Phys. Chem.97: 12 685 – 12 690.

Deng, H., and Braunlin, W.H. 1996. Kinetics of sodium ion bind-ing to DNA quadruplexes. J. Mol. Biol.255: 476–483.

Detellier, C., and Laszlo, P. 1980. Role of alkali metal and ammo-nium cations in the self-assembly of the 5′-guanosinemonophosphate dianion. J. Am. Chem. Soc.102: 1135–1141.

Dirken, P.J., Kohn, S.C., Smith, M.E., and Van Eck, E.R.H. 1997.Complete resolution of Si-O-Si and Si-O-Al fragments in analuminosilicate glass by17O multiple quantum magic anglespinning NMR spectroscopy. Chem. Phys. Lett.266: 568–574.

Dobler, M. 1981. Ionophores and their structures. Wiley, NewYork.

Dobler, M., and Phizackerley, R.P. 1974. The crystal structure ofthe NaSCN complex of nonactin. Helv. Chim. Acta,57:664–674.

Duer, M.J., and Stourton, C. 1997. Further developments inMQMAS NMR spectroscopy for spin-3/2 nuclei. J. Magn.Reson.124: 189–199.

Dupree, R., Howes, A.P., and Kohn, S.C. 1997. Natural abundancesolid state43Ca NMR. Chem. Phys. Lett.276: 399–404.

Eaton, D.R., Buist, R.J., and Sayer, B.G. 1987. The59Co nuclearmagnetic resonance spectra of polycrystalline complexes. Can.J. Chem.65: 1332–1335.

Eckert, H., and Fiat, D. 1986. Isotopic labeling of tyrosine, fol-lowed by 17O NMR. Int. J. Pept. Protein Res.27: 613–616.

© 1998 NRC Canada


Eichele, K., Chan, J.C.C., Wasylishen, R.E., and Britten, J.F. 1997.Single-crystal cobalt-59 NMR study of tris(2,4-pentanedionato-O,O′)cobalt(III). J. Phys. Chem. A,101: 5423–5430.

Evans, J.N.S. 1995. Biomolecular NMR spectroscopy. Oxford Uni-versity Press Inc., New York.

Fernandez, C., and Amoureux, J.-P. 1995. 2D multiquantumMAS-NMR spectroscopy of 27Al in aluminophosphate molecu-lar sieves. Chem. Phys. Lett.242: 449–454.

Fernandez, C., and Amoureux, J.-P. 1996. Triple-quantumMAS-NMR of quadrupolar nuclei. Solid State Nucl. Magn.Reson.5: 315–321.

Fernandez, C., Delevoye, L., Amoureux, J.-P., Lang, D.P., andPruski, M. 1997.27Al 1H cross polarization triple-quantummagic angle spinning NMR. J. Am. Chem. Soc.119:6858–6862.

Forsén, S., Drakenberg, T., and Wennerström, H. 1987. NMR stud-ies of ion binding in biological systems. Q. Rev. Biophys.19:83–114.

Freude, D., and Haase, J. 1993. Quadrupole effect in solid state nu-clear magnetic resonance.In NMR basic principles and prog-ress. Vol. 29.Edited by P. Diehl, E. Fluck, H. Günther, R.Kosfeld, and J. Seelig. Springer-Verlag, Berlin. pp. 1–90.

Frydman, L., and Harwood, J.S. 1995. Isotropic spectra ofhalf-integer quadrupolar spins from bidimensional magic-anglespinning NMR. J. Am. Chem. Soc.117: 5367–5368.

Gann, S.L., Baltisberger, J.H., Wooten, E.W., Zimmermann, H.,and Pines, A. 1994. Cross polarization and dynamic-angle spin-ning of 17O in L-alanine. Bull. Magn. Reson.16: 68–72.

Garbow, J.R., and Gullion, T. 1995. Chap. 3.In Carbon-13 NMRspectroscopy of biological systems.Edited by N. Beckmann.Academic Press, San Diego, Calif.

Germann, M.W., Aramini, J.M., and Vogel, H.J. 1994. Quadrupolarmetal ion NMR study of ovotransferrin at 17.6 T. J. Am. Chem.Soc.116: 6971–6972.

Gilboa, H., Steinschneider, A., Valentine, B., Dhawan, D., andFiat, D. 1984. Hydrogen bonds in the tripeptidePro-Leu-Gly-NH2:

17O and1H studies. Biochim. Biophys. Acta,800: 251–257.

Grandinetti, P.J., Baltisberger, J.H., Farnan, I., Stebbins, J.F.,Werner, U., and Pines, A. 1995. Solid-state17O magic-angle anddynamic-angle spinning NMR study of the SiO2 polymorphcoesite. J. Phys. Chem.99: 12 341 – 12 348.

Granger, P. 1991. Copper to mercury.In Transition metal nuclearmagnetic resonance.Edited byP.S. Pregosin. Elsevier SciencePublishers B.V., Amsterdam. pp. 265–346.

Grey, C.P., and Veeman, W.S. 1992. The detection of weakheteronuclear coupling between spin 1 and spin ½ nuclei inMAS NMR; 14N/13C/1H triple resonance experiments. Chem.Phys. Lett.192: 379–385.

Griffiths, J.M., and Griffin, R.G. 1993. Nuclear magnetic reso-nance methods for measuring dipolar couplings in rotating sol-ids. Anal. Chim. Acta,283: 1081–1101.

Gullion, T. 1995. Measurement of dipolar interactions be-tween spin-1/2 and quadrupolar nuclei by rotationalecho,adiabatic-passagedouble-resonance NMR. Chem.Phys.Lett. 246: 325–330.

Haller, M., Hertler, W.E., Lutz, O., and Nolle, A. 1980.67Zn and33S nuclear magnetic shielding in zinc chalcogenides. SolidState Commun.33: 1051–1053.

Hamilton, J.A., Sabesan, M.N., and Steinrauf, L.K. 1981. Crystalstructure of valinomycin potassium picrate: anion effect onvalinomycin cation complexes. J. Am. Chem. Soc.103:5880–5885.

Hayashi, S. 1996.51V and 59Co off-MAS NMR spectra: determina-tion of quadrupole coupling, chemical shift anisotropy and theirrelative orientation. Magn. Reson. Chem.34: 791–798.

Hud, N.V., Smith, F.W., Anet, F.A., and Feigon, J. 1996. The selec-tivity for K + versus Na+ in DNA quadruplexes is dominated byrelative free energy of hydration: a thermodynamic analysis by1H NMR. Biochemistry,35: 15 383 – 15 390.

Jäger, C. 1994. Satellite transition spectroscopy of quadrupolar nu-clei. In NMR basic principles and progress. Vol. 31.Edited byP.Diehl, E. Fluck, H. Günther, R. Kosfeld, and J. Seelig.Springer-Verlag, Berlin. pp. 133–170.

Kabalka, G.W., and Goudgaon, N.M. 1991.17O-enrichment meth-ods. In Oxygen-17 NMR spectroscopy in organic chemistry.Edited by D.W. Boykin. CRC Press, Inc., Boca Raton, Fla.pp. 21–37.

Kang, C., Zhang, X., Ratliff, R., Moyzis, R., and Rich, A. 1992.Crystal structure of four-strandedOxytricha telomeric DNA.Nature (London),356: 126–131.

Kilbourn, B.T., Dunitz, J.D., Pioda, L.A.R., and Simon, W. 1967.Structure of the K+ complex with nonactin, a macrotetrolide an-tibiotic possessing highly specific K+ transport properties. J.Mol. Biol. 30: 559–563.

Kim, J., Cheong, C., and Moore, P.B. 1991. Tetramerization of anRNA oligonucleotide containing a GGGG sequence. Nature(London),351: 331–332.

Kim, J., Eglin, J.L., Ellaboudy, A.S., McMills, L.E.H., Huang, S.,and Dye, J.L. 1996.87Rb, 85Rb, and 39K NMR studies ofalkalides, electrides, and related compounds. J. Phys. Chem.100: 2885–2891.

Koller, H., Engelhardt, G., Kentgens, A.P.M., and Sauer, J. 1994.23Na NMR spectroscopy of solids: interpretation of quadrupoleinteraction parameters and chemical shifts. J. Phys. Chem.98:1544–1551.

Kroeker, S., Eichele, K., Wasylishen, R.E., and Britten, J.F. 1997.Cesium-133 NMR study of CsCd(SCN)3: relative orientation ofthe chemical shift and electric field gradient tensors. J. Phys.Chem. B,101: 3727–3733.

Kundla, E., Samoson, A., and Lippmaa, E. 1981. High-resolutionNMR of quadrupolar nuclei in rotating solids. Chem. Phys. Lett.83: 229–232.

Kunwar, A.C., Turner, G.L., and Oldfield, E. 1986. Solid-statespin-echo Fourier transform NMR of39K and 67Zn salts at highfield. J. Magn. Reson.69: 124–127.

Kuroki, S., Takahashi, A., Ando, I., Shoji, A., and Ozaki, T. 1994.Hydrogen-bonding structural study of solid peptides andpolypeptides containing a glycine residue by17O NMR spectros-copy. J. Mol. Struct.323: 197–208.

Laughlan, G., Murchie, A.I.H., Norman, D.G., Moore, M.H.,Moody, P.C.E., Lilley, D.M.J., and Luisi, B. 1994. Thehigh-resolution crystal structure of a parallel-stranded guaninetetraplex. Science (Washington, D.C.),265: 520–524.

Le, H., Pearson, J.G., de Dios, A.C., and Oldfield, E. 1995. Proteinstructure refinement and prediction via NMR chemical shiftsand quantum chemistry. J. Am. Chem. Soc.117: 3800–3807.

Lippard, S.J., and Berg, J.M. 1994. Principles of bioinorganicchemistry. University Science Books, Mill Valley, Calif.

Llor, A., and Virlet, J. 1988. Towards high-resolution NMR ofmore nuclei in solids: sample spinning with time-dependentspinner axis angle. Chem. Phys. Lett.152: 248–253.

Man, P.P. 1986. Investigation of the central line of55Mn in KMnO4by a two-dimensional NMR method. J. Magn. Reson.67: 78–90.

Massiot, D. 1996. Sensitivity and lineshape improvements ofMQ-MAS by rotor synchronized data acquisition. J. Magn.Reson. Ser. A,122: 240–244.

© 1998 NRC Canada


Massiot, D., Touzo, B., Trumeua, D., Coutures, J.P., Virlet, J.,Florian, P., and Grandinetti, P.J. 1996. Two-dimensionalmagic-angle spinning isotropic reconstruction sequences forquadrupolar nuclei. Solid State Nucl. Magn. Reson.6: 73–83.

McAteer, K., Lipton, A.S., and Ellis, P.D. 1996. Cadmium-113NMR: a surrogate probe for zinc and calcium in proteins.In En-cyclopedia of nuclear magnetic resonance.Edited by D.M.Grant and R.K. Harris. Wiley, Chichester, U.K. pp. 1085–1091.

Medek, A., Harwood, J.S., and Frydman, L. 1995. Multiple-quantum magic-angle spinning NMR: a new method for thestudy of quadrupolar nuclei in solids. J. Am. Chem. Soc.117:12 779 – 12 787.

Medek, A., Frydman, V., and Frydman, L. 1997a. Cobalt-59 NMRstudies of diamagnetic porphyrin complexes in the solid phase.J. Phys. Chem. B,101: 8959–8966.

Medek, A., Frydman, V., and Frydman, L. 1997b. Solid and liquidphase59Co NMR studies of cobalamines and their derivatives.Proc. Natl. Acad. Sci. U.S.A.94: 14 237 – 14 242.

Mooibroek, S., Wasylishen, R.E., Dickson, R., Facey, G., andPettitt, B.A. 1986. Simultaneous observation of shieldinganisotropies and quadrupolar splittings in solid state133Cs NMRspectra. J. Magn. Reson.66: 542–545.

Mueller, K.T., Sun, B.Q., Chingas, G.C., Zwanziger, J.W., Terao,T., and Pines, A. 1990. Dynamic-angle spinning of quadrupolarnuclei. J. Magn. Reson.86: 470–487.

Mueller, K.T., Wu, Y., Chmelka, B.F., Stebbins, J., and Pines, A.1991. High-resolution oxygen-17 NMR of solid silicates. J. Am.Chem. Soc.113: 32–38.

Neurohr, K.J., Drakenberg, T., Forsén, S., and Lilja, H. 1983. Po-tassium-39 nuclear magnetic resonance of potassium–ionophorecomplexes: chemical shifts, relaxation times, and quadrupolecoupling constants. J. Magn. Reson.51: 460–469.

Oldfield, E., Schramm, S., Meadows, M.D., Smith, K.A., Kinsey,R.A., and Ackerman, J. 1982a. High-resolution NMR spectros-copy of quadrupolar nuclei in solids: sodium salts. J. Am.Chem. Soc.104: 919–920.

Oldfield, E., Kinsey, R.A., Montez, B., Ray, T., and Smith, K.A.1982b. High-resolution solid-state NMR spectra of quadrupolarnuclei: magic-angle and off-axis spinning of vanadium-51 (I =7/2) in sodium and ammonium metavanadates. J. Chem. Soc.Chem. Commun. 1982: 254–256.

Oldfield, E., Lee, H.C., Coretsopoulos, C., Adebodum, F., Park,K.D., Yang, S., Chung, J., and Philips, B. 1991. Solid-stateoxygen-17 nuclear magnetic resonance spectroscopic studiesof [17O2] picket fence porphyrin, myoglobin, and hemoglobin.J. Am. Chem. Soc.113: 8680–8685.

Park, K.D., Guo, K., Adebodun, F., Chiu, M.L., Sligar, S.G., andOldfield, E. 1991. Distal and proximal ligand interactions inheme proteins: correlations between C-O and Fe-C vibrationalfrequencies, oxygen-17 and carbon-13 nuclear magnetic reso-nance chemical shifts, and oxygen-17 nuclear quadrupole cou-pling constants in C17O- and13CO-labeled species. Biochemistry,30: 2333–2347.

Petersheim, M., Miner, V.W., Gerlt, J.A., and Prestegard, J.H.1983. 17O NMR as a probe of nucleic acid dynamics. J. Am.Chem. Soc.105: 6357–6358.

Power, W.P., Wasylishen, R.E., Mooibroek, S., Pettitt, B.A., andDanchura, W. 1990. Simulation of NMR powder line shapes ofquadrupolar nuclei with half-integer spin at low-symmetry sites.J. Phys. Chem.94: 591–598.

Power, W.P., Kirby, C.W., and Taylor, N.J. 1998. A single crystal59Co NMR study of vitamin B12. J. Am. Chem. Soc.120:9428–9434.

Reynhardt, E.C. 1974a. Experimental study of nuclear electricquadrupole splittings of59Co in Co(NH3)6Cl3. Can. J. Phys.52:1398–1404.

Reynhardt, E.C. 1974b. NMR of 59Co in trisacetylacetonatocobalt(III). J. Phys. C Solid State Phys.7: 4135–4142.

Rocha, J., Esculcas, A.P., Fernandez, C., and Amoureux, J.-P.1996. Two-dimensional triple-quantum27Al MAS NMR spec-troscopic study of the high-temperature phase transformation ofmicroporous VPI-5. J. Phys. Chem.100: 17 889 – 17 892.

Saitô, H., and Tabeta, R. 1987. A high resolution23Na NMR studyof sodium complexes with ionophores in solution. Bull. Chem.Soc. Jpn.60: 61–68.

Sakarellos, C., Gerothanassis, I.P., Birlirakis, N., Karayannis, T.,Sakarellos-Daitsiotis, M., and Marraud, M. 1989.17O NMRstudies of the conformational and dynamic properties ofenkephalins in aqueous and organic solutions using selectivelylabeled analogues. Biopolymers,28: 15–26.

Samoson, A., Kundla, E., and Lippmaa, E. 1982. High resolutionMAS-NMR of quadrupolar nuclei in powders. J. Magn. Reson.49: 350–357.

Samoson, A., Lippmaa, E., and Pines, A. 1988. High resolutionsolid-state NMR. Averaging of second-order effects by means ofa double rotor. Mol. Phys.65: 1013–1018.

Sanders C.R., II, and Tsai, M.-D. 1989. Ligand–protein interac-tions via nuclear magnetic resonance of quadrupolar nuclei.Methods Enzymol.117: 317–333.

Scheubel, W., Zimmermann, H., and Haeberlin, U. 1985.17Oquadrupole coupling and nuclear magnetic shielding tensors inbenzophenone. J. Magn. Reson.63: 544–555.

Schramm, S., and Oldfield, E. 1984. High-resolution oxygen-17NMR of solids. J. Am. Chem. Soc.106: 2502–2506.

Shimizu, T., and Hatano, M. 1983.67Zn NMR studies of aqueousZn2+ and Zn2+-insulin complex. Inorg. Chim. Acta,76:L177–L178.

Spiess, H.W., Haas, H., and Hartmann, H. 1969. Anisotropic chem-ical shifts in cobalt(III) complexes. J. Chem. Phys.50:3057–3064.

Stebbins, J.F., and Xu, Z. 1997. NMR evidence for excessnon-bridging oxygen in an aluminosilicate glass. Nature (Lon-don), 390: 60–62.

Steinschneider, A., Valentine, B., Burgar, M.I., and Fiat, D. 1985.NMR of carboxylic-17O in amino acids. Magn. Reson. Chem.23: 104–110.

Sundquist, W.I., and Klug, A. 1989. Telomeric DNA dimerizes byformation of guanine tetrads between hairpin loops. Nature(London),342: 825–829.

Tabeta, R., Aida, M., and Saitô, H. 1986. A high-resolutionsolid-state 23Na NMR study of sodium complexes with sol-vents, small ligand molecules, and ionophores. Bull. Chem. Soc.Jpn.59: 1957–1966.

Toney, M.D., Hohenester, E., Cowan, S.W., and Jansonius, J.N.1993. Dialkylglycine decarboxylase structure: bifunctional ac-tive site and alkali metal sites. Science (Washington, D.C.),261:756–759.

van Niekerk, J.N., Schoening, F.R.L., and Talbot, J.H. 1953. Thecrystal structure of zinc acetate dihydrate, Zn(CH3COO)2·2H2O.Acta Crystallogr.6: 720–723.

Vega, A.J. 1992a. MAS NMR spin locking of half-integerquadrupolar nuclei. J. Magn. Reson.96: 50–68.

Vega, A.J. 1992b. CP/MAS of quadrupolarS = 3/2 nuclei. SolidState Nucl. Magn. Reson.1: 17–32.

Vega, S. 1978. Fictious spin 1/2 operator formalism for multiplequantum NMR. J. Chem. Phys.68: 5518–5527.

© 1998 NRC Canada


© 1998 NRC Canada


Walter, T.H., Turner, G.L., and Oldfield, E. 1988. Oxygen-17cross-polarization NMR spectroscopy of inorganic solids. J.Magn. Reson.76: 106–120.

Wilde, J.A., Bolton, P.H., Mazumder, A., Manoharan, and Gerlt,J.A. 1989. Characterization of the equilibrating forms of the al-dehydic abasic site in duplex DNA by17O NMR. J. Am. Chem.Soc.111: 1894–1896.

Williamson, J. 1994. G-quartet structures in telomeric DNA. Annu.Rev. Biophys. Biomol. Struct.23: 703–730.

Wisner, D.A., Steginsky, C.A., Shyy, Y.-J., and Tsai, M.-D. 1985.Mechanism of adenylate kinase. Use of17O NMR to study thebinding properties of substrates. J. Am. Chem. Soc.107:2814–2815.

Wokaun, A., and Ernst, R.R. 1977. Selective excitation and detec-tion in multilevel spin systems: application of single transitionoperators. J. Chem. Phys.67: 1752–1758.

Wu, G., Kroeker, S., and Wasylishen, R.E. 1995. MultinuclearNMR study of dipotassium tetracyanometalates of the group 12metals in the solid state. Inorg. Chem. 34: 1595–1598.

Wu, G., Rovnyak, D., Sun, B., and Griffin, R.G. 1996a.High-resolution multiple quantum MAS NMR spectroscopy ofhalf-integer quadrupolar nuclei. Chem. Phys. Lett.249:210–217.

Wu, G. Rovnyak, D., and Griffin, R.G. 1996b. Quantitative multiple-quantum magic-angle-spinning NMR spectroscopy of quadrupolarnuclei in solids. J. Am. Chem. Soc.118: 9326–9332.

Wu, G., Rovnyak, D., Huang, P.C., and Griffin, R.G. 1997a.High-resolution oxygen-17 NMR spectroscopy of solids by multiple-quantum magic-angle-spinning. Chem. Phys. Lett.277: 79–83.

Wu, G., Kroeker, S. Wasylishen, R.E., and Griffin, R.G. 1997b.Indirect spin–spin coupling in multiple-quantum magic-angle-spinning NMR spectra of quadrupolar nuclei. J. Magn. Reson.124: 237–239.

Wu, G., Rovnyak, D., Huang, P.C., and Griffin, R.G. 1997c.High-resolution oxygen-17 NMR in solids.In Proceedings of38th Experimental NMR Conference (March 23–27, 1997, Or-lando, Fla.). ENC, Santa Fe, New Mexico. p. 161.

Wu, Y., Sun, B.Q., Pines, A., Samoson, A., and Lippmaa, E. 1990.NMR experiments with a new double rotor. J. Magn. Reson.89:296–309.

Youngman, R.E., Haubrich, S.T., Zwanziger, J.W., Janicke, M.T.,and Chmelka, B.F. 1995. Short- and intermediate-range struc-tural ordering in glass boron oxide. Science (Washington, D.C.),269: 1416–1420.

Youngman, R.E., Werner-Zwanziger, U., and Zwanziger, J.W.1996. A comparison of strategies for obtaining high-resolutionNMR spectra of quadrupolar nuclei. Z. Naturforsch. A Phys.Sci. 51: 321–329.

Zhang, Q.W., Zhang, H.M., Usha, M.G., and Wittebort, R.J. 1996.17O NMR and crystalline hydrates. Solid State Nucl. Magn.Reson.7: 147–154.

recent developments in solid-state nuclear magnetic resonance...

Documents