TOTAPOL: A biradical polarizing agent for dynamic nuclear polarization experiments in aqueous media

Article abstract of DOI:10.1021/ja061284b

In a previous publication, we described the use of biradicals, in that case two TEMPO molecules tethered by an ethylene glycol chain of variable length, as polarizing agents for microwave driven dynamic nuclear polarization (DNP) experiments. The use of biradicals in place of monomeric paramagnetic centers such as TEMPO yields enhancements that are a factor of approximately 4 larger (ε ～ 175 at 5 T and 90 K) and concurrently the concentration of the polarizing agent is a factor of 4 smaller (10 mM electron spins), reducing the residual electron nuclear dipole broadening. In this paper we describe the synthesis and characterization by EPR and DNP/NMR of an improved polarizing agent 1-(TEMPO-4-oxy)-3-(TEMPO-4-amino)propan-2-ol (TOTAPOL). Under the same experimental conditions and using 2.5 mm magic angle rotors, this new biradical yields larger enhancements (ε ～ 290) at lower concentrations (6 mM electron spins) and has the additional important property that it is compatible with experiments in aqueous media, including salt solutions commonly used in the study of proteins and nucleic acids.

Full text of DOI:10.1021/ja061284b

Published on Web 08/16/2006
TOTAPOL: A Biradical Polarizing Agent for Dynamic Nuclear
Polarization Experiments in Aqueous Media
Changsik Song,† Kan-Nian Hu,†,‡ Chan-Gyu Joo,‡ Timothy M. Swager,† and
Robert G. Griffin,*,‡
Contribution from the Francis Bitter Magnet Laboratory and Department of Chemistry,
Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
Received March 3, 2006; E-mail: rgg@mit.edu
Abstract: In a previous publication, we described the use of biradicals, in that case two TEMPO molecules
tethered by an ethylene glycol chain of variable length, as polarizing agents for microwave driven dynamic
nuclear polarization (DNP) experiments. The use of biradicals in place of monomeric paramagnetic centers
such as TEMPO yields enhancements that are a factor of approximately 4 larger (ꢀ ∼ 175 at 5 T and 90
K) and concurrently the concentration of the polarizing agent is a factor of 4 smaller (10 mM electron
spins), reducing the residual electron nuclear dipole broadening. In this paper we describe the synthesis
and characterization by EPR and DNP/NMR of an improved polarizing agent 1-(TEMPO-4-oxy)-3-(TEMPO-
4-amino)propan-2-ol (TOTAPOL). Under the same experimental conditions and using 2.5 mm magic angle
rotors, this new biradical yields larger enhancements (ꢀ ∼ 290) at lower concentrations (6 mM electron
spins) and has the additional important property that it is compatible with experiments in aqueous media,
including salt solutions commonly used in the study of proteins and nucleic acids.
1. Introduction
frequency (140-600 GHz), high-power (∼10 W) microwave
source is required to drive the continuous-wave (CW) DNP
transitions associated with the second-order electron-nuclear
dipolar interactions. To date this requirement has been satisfied
by utilizing gyrotrons13-16 since they operate in the requisite
frequency range and produce the required microwave powers.
Second, the relaxation times of the spin systems in the
experiment dictate that it be optimally performed at low
temperatures (usually e 90 K), and to obtain high-resolution
spectra of solids, magic-angle spinning (MAS) is incorporated
The sensitivity in solid-state NMR (ssNMR) experiments can
be enhanced by 2-3 orders of magnitude by dynamically
polarizing the nuclear spin system prior to recording the NMR
spectrum.1-12 This procedure involves microwave irradiation
of the electron paramagnetic resonance (EPR) spectrum of either
an endogenous or exogenous paramagnetic species present in
the sample and results in the transfer of the greater spin
polarization of the electrons to the nuclei. However, to perform
dynamic nuclear polarization (DNP) experiments at magnetic
fields used in contemporary NMR experiments (5-20 T), the
following three requirements must be satisfied. First, a high-
1
into the experiment.4-6,9,11 Thus, multiple resonancesi.e., H,
13C, 15N, and e-slow-temperature MAS probes are required
for proper execution of DNP experiments.
The third requirement is the presence of a suitable paramag-
netic center that acts as source of polarization. This paramagnetic
center should (a) be compatible with the polarization mechanism
that yields the optimal signal enhancement, namely the three-
spin thermal mixing (TM)4,17 or cross effect (CE),18-23,24 (b)
† Department of Chemistry.
‡ Francis Bitter Magnet Laboratory.
(1) Overhauser, A. W. Phys. ReV. 1953, 92, 411-415.
(2) Carver, T. R.; Slichter, C. P. Phys. ReV. 1953, 92, 212-213.
(3) Abragam, A.; Goldman, M. Nuclear magnetism: order and disorder;
Clarendon Press: Oxford, U.K., 1982.
(4) Wind, R. A.; Duijvestijn, M. J.; Vanderlugt, C.; Manenschijn, A.; Vriend,
J. Prog. Nucl. Magn. Reson. Spectrosc. 1985, 17, 33-67.
(5) Singel, D. J.; Seidel, H.; Kendrick, R. D.; Yannoni, C. S. J. Magn. Reson.
1989, 81, 145-161.
(13) Becerra, L. R.; Gerfen, G. J.; Bellew, B. F.; Bryant, J. A.; Hall, D. A.;
Inati, S. J.; Weber, R. T.; Un, S.; Prisner, T. F.; McDermott, A. E.; Fishbein,
K. W.; Kreischer, K. E.; Temkin, R. J.; Singel, D. J.; Griffin, R. G. J.
Magn. Reson., Ser. A 1995, 117, 28-40.
(6) Afeworki, M.; McKay, R. A.; Schaefer, J. Macromolecules 1992, 25, 4084-
4091.
(7) Gerfen, G. J.; Becerra, L. R.; Hall, D. A.; Griffin, R. G.; Temkin, R. J.;
Singel, D. J. J. Chem. Phys. 1995, 102, 9494-9497.
(8) Rosay, M.; Lansing, J. C.; Haddad, K. C.; Bachovchin, W. W.; Herzfeld,
J.; Temkin, R. J.; Griffin, R. G. J. Am. Chem. Soc. 2003, 125, 13626-
13627.
(14) Becerra, L. R.; Gerfen, G. J.; Temkin, R. J.; Singel, D. J.; Griffin, R. G.
Phys. ReV. Lett. 1993, 71, 3561-3564.
(15) Joye, C. D.; Griffin, R. G.; Hornstein, M. K.; Hu, K.-N.; Kreischer, K. E.;
Rosay, M.; Shapiro, M. A.; Sirigiri, J. R.; Temkin, R. J.; Woskov, P. P.
IEEE Trans. Plasma Sci. 2006, 34, 518-523.
(9) Rosay, M.; Weis, V.; Kreischer, K. E.; Temkin, R. J.; Griffin, R. G. J.
Am. Chem. Soc. 2002, 124, 3214-3215.
(16) Hornstein, M. K.; Bajaj, V. S.; Griffin, R. G.; Kreischer, K. E.; Mastovsky,
I.; Shapiro, M. A.; Sirigiri, J. R.; Temkin, R. J. IEEE Trans. Electron
DeVices 2005, 52, 798-807.
(10) Rosay, M.; Zeri, A. C.; Astrof, N. S.; Opella, S. J.; Herzfeld, J.; Griffin,
R. G. J. Am. Chem. Soc. 2001, 123, 1010-1011.
(17) Goldman, M. Spin Temperature and Nuclear Magnetic Resonance in Solids;
Clarendon Press: Oxford, U.K., 1970.
(11) Bajaj, V. S.; Farrar, C. T.; Mastovsky, I.; Vieregg, J.; Bryant, J.; Elena,
B.; Kreischer, K. E.; Temkin, R. J.; Griffin, R. G. J. Magn. Reson. 2003,
160, 85-90.
(18) Hwang, C. F.; Hill, D. A. Phys. ReV. Lett. 1967, 19, 1011-1014.
(19) Hwang, C. F.; Hill, D. A. Phys. ReV. Lett. 1967, 18, 110-112.
(20) Kessenikh, A. V.; Lushchikov, V. I.; Manenkov, A. A.; Taran, Y. V. SoV.
Phys. Solid State 1963, 5, 321-329.
(12) Hu, K.-N.; Yu, H.-h.; Swager, T. M.; Griffin, R. G. J. Am. Chem. Soc.
2004, 126, 10844-10845.
9
10.1021/ja061284b CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 11385-11390
11385

A R T I C L E S
Song et al.
mixture was stirred overnight at room temperature. It was then poured
into water and extracted with ethyl acetate. The organic layer was
washed with brine, dried over MgSO4, and evaporated under reduced
pressure. The resulting crude product was purified by column chro-
matography (dichloromethane, methanol), providing a quantitative yield
of red, viscous oil. IR (KBr disk, cm-1): 2975, 2937, 1636, 1464, 1384,
1363, 1244, 1179, 1095. 1H NMR (300 MHz, DMSO-d6): δ 7.12 (bs,
1H), 3.69 (dd, 1H, J ) 11, 2.7 Hz), 3.60 (m, 1H), 3.24 (dd, 1H, J )
11, 6.6 Hz), 3.04 (m, 1H), 2.70 (pseudo-t, 1H, J ) 5.1 Hz), 2.52 (dd,
1H, J ) 5.1, 2.7 Hz), 1.86 (dd, 2H, J ) 12, 3.6 Hz), 1.24 (pseudo-t,
2H, J ) 11 Hz), 1.05 (d, 12H, J ) 11 Hz).
be useful in polarizing a large array of samples ranging from
small molecules to proteins, (c) produce large signal enhance-
ments at a reduced concentration of paramagnetic species, and
(d) be soluble in aqueous media. In a previous publication,12
we described the use of biradicals that satisfy the first three of
these criteria and yield improved DNP enhancements. In
particular, we reported that bis-TEMPO-n-ethylene glycol
(BTnE), where TEMPO is 2,2,6,6-tetramethylpiperidin-1-oxyl
and n ) 2 indicates a tether of two ethylene glycol units,
produced DNP enhancements of ∼175 at 90 K and 5 T. Further,
this was accomplished at a reduced radical concentration (∼5
mM biradicals or 10 mM electron spins, as opposed to ∼40
mM monomeric TEMPO), thus reducing the electron nuclear
dipolar broadening. The design, synthesis, and characterization
of an improved polarizing agent, 1-(TEMPO-4-oxy)-3-(TEMPO-
4-amino)propan-2-ol (TOTAPOL), satisfying the first three as
well as the fourth requirement, is the topic of this paper.
Specifically, we have prepared a biradical consisting of two
TEMPO molecules tethered with a three carbon chain that
increases the average electron-electron dipole coupling constant
from ∼0.5 MHz, in the typical 40 mM solution of TEMPO
that we use for DNP experiments, to ∼30 MHz in the biradical.
At 140 GHz this yields a maximum enhancement of ∼290, and
again the electron concentration is reduced by a factor of 6 from
the typical level of 40 to 6 mM. Very importantly, TOTAPOL
has hydroxyl and secondary amine moieties on the tether and
these functional groups increase the solubility of the biradical
in aqueous media so that it is compatible with a variety of
biological systems where DNP experiments are currently
performed.
2.1.c. 1-(2,2,6,6-Tetramethyl-1-oxy-4-piperidinyl)oxy-3-(2,2,6,6-
tetramethyl-1-oxy-4-piperidinyl)amino-propan-2-ol [1-(TEMPO-4-
oxy)-3-(TEMPO-4-amino)propan-2-ol] (TOTAPOL). In a 100 mL
round-bottom flask equipped with a stirring bar were combined 4-(2,3-
epoxy-propoxy)-TEMPO (3) (1.62 g, 7 mmol), LiClO4 (0.745 mg, 7
mmol), and 10 mL of anhydrous CH3CN under Ar. To the mixture
was added a CH3CN (3 mL) solution of 4-amino-TEMPO (1.20 g, 7
mmol), and the mixture was stirred overnight at room temperature.
Most of the solvent was evaporated under reduced pressure, and the
crude product was purified by column chromatography (dichlo-
romethane, methanol). Yield: 1.98 g of orange-red solid. IR (KBr disk,
1
cm-1): 3446, 2977, 2938, 1635, 1465, 1364, 1244, 1178, 1098. H
NMR (400 MHz, DMSO-d6): δ 7.31 (bs, 1H), 7.15 (bs, 1H), 4.57 (bs,
1H), 3.93 (m, 1H), 3.69 (m, 1H), 3.57 (m, 1H), 3.41 (m, 2H), 3.34 (m,
1H), 3.01 (pseudo-d, 1H, J ) 11 Hz), 2.80 (m, 1H), 1.96 (m, 2H),
1.86 (m, 2H), 1.52 (m, 2H), 1.23 (pseudo-t, 2H, J ) 11 Hz), 1.04 (dd,
24H, J ) 17, 3.9 Hz). HR-MS (ESI): calcd for C21H41N3O42‚, [M +
H]+, 400.3170; found, 400.3161.
2.2. EPR Experiments. For 9 GHz solution EPR experiments, the
samples consisted of a capillary containing ∼10 µL of 0.5 mM
TOTAPOL in absolute ethanol since it has a relatively low dielectric
constant that facilitates observation of solution EPR spectra. To increase
the resolution of powder EPR spectra from frozen solutions, TOTAPOL
was synthesized with 15N, 2H-labeled TEMPO’s (CDN Isotope, Quebec,
Canada). Samples for the X-band experiments were typically 60 µL of
In this article we describe the synthesis of TOTAPOL and
several EPR and DNP-enhanced NMR experiments that char-
acterize the molecule and illustrate its utility as an effective
polarizing agent. These include a comparison of the enhance-
ments obtained with TEMPO and several other biradicals, as
well as its compatibility with low-temperature MAS experiments
in glycerol/water mixtures.
2
2
0.5 mM 15N, H-TOTAPOL in the glass forming solvent H6-DMSO/
2H2O (6:4 w/w) in a 4 mm o.d. quartz tube. The 9 GHz CW EPR spectra
were recorded with a Bruker EMX spectrometer with the sample
immersed in liquid nitrogen (77 K) in a finger dewar (Wilmad WG-
819-B-Q). Samples for 140 GHz experiments consisted of 0.4 µL of
TOTAPOL in the same deuterated glass forming solution contained in
a quartz capillary immersed in a helium cryostat (Oxford Instruments).
The 140 GHz spectra were recorded with echo-detected experiments
using a custom-designed spectrometer.13 The rigid-limit for EPR spectra
of TOTAPOL is reached below 100 K in the glass forming mixture;
thus, the 20 and 77 K temperatures chosen for the above EPR
experiments are sufficient for observing rigid limit EPR spectra.
2. Experimental Section
2.1. Synthesis of TOTAPOL. 2.1.a. General Experimental Condi-
tions. 4-Hydroxy-TEMPO (1) and 4-amino-TEMPO (2), containing
g97.0% free radical, were purchased from Sigma-Aldrich (St. Louis,
MO) and used without further purification. Anhydrous CH3CN was
purchased from the same company as a Sure-Seal bottle. All other
chemicals were of reagent grade and used as received. For NMR
analysis, TEMPO radicals were reduced to N-hydroxy compounds by
ascorbic acid in methanol. NMR spectra were recorded on a Bruker
Advance-400 or Varian Mercury-300 spectrometer, and chemical shifts
were referenced to residual solvent peaks. IR spectra were obtained on
a Nicolet 8700 FT-IR spectrometer, in which the sample was drop-
casted on a KBr disk. High-resolution mass spectra were obtained on
a Bruker Daltonics APEX II 3T FT-ICR-MS.
2.1.b. 4-(2,3-Epoxypropoxy)-2,2,6,6-tetramethyl-1-piperidin-1-
oxyl [4-(2,3-Epoxypropoxy)-TEMPO] (3). In a 100 mL round-bottom
flask equipped with a stirring bar were combined tetrabutylammonium
hydrogen sulfate (0.136 g, 4 mol %), 50% w/w aqueous NaOH (10
mL), and epichlorohydrin (3.91 mL, 50 mmol). To the mixture was
added 4-hydroxy-TEMPO (1.73 g, 10 mmol) in portions, and the
2.3. DNP-Enhanced NMR Experiments with TOTAPOL. Solu-
tions for the DNP/NMR experimentss2 M 13C-urea or 0.2 M U-13C-
15N-prolineswere prepared in 2H6-DMSO/2H2O/H2O (60:34:6 w/w/w)
or 2H8-glycerol/2H2O/H2O (60:25:15 w/w/w) and doped with 3-5 mM
TOTAPOL (6-10 mM electron spins). The reduced 1H concentration
was required to optimize the signal enhancements and chosen to
maintain effective proton homonuclear spin diffusion. DNP experiments
using TOTAPOL as a polarizing agent were performed on a custom-
1
designed DNP/NMR spectrometer and triple-resonance (e-, H, and
13C/15N) cryogenic (90 K) MAS probe (with a 4 or 2.5 mm rotor
1
stator)9,25 operating at 5 T (140 GHz EPR and 211 MHz H NMR).
The enhanced 1H polarization developed by the microwave irradiation
was detected indirectly via observation of the cross-polarized (CP) 13
C
signals. The 140 GHz microwaves were generated by a gyrotron, a
vacuum electron device capable of producing high-power (>10 W)
(21) Kessenikh, A. V.; Manenkov, A. A.; Pyatnitskii, G. I. SoV. Phys. Solid
State 1964, 6, 641-643.
(22) Wollan, D. S. Phys. ReV. B 1976, 13, 3686-3696.
(23) Wollan, D. S. Phys. ReV. B 1976, 13, 3671-3685.
(24) Atsarkin, V. A. SoV. Phys. Usp. 1978, 21, 725-744.
(25) Joo, C.-G.; Hu, K.-N.; Bryant, J. A.; Griffin, R. G. J. Am. Chem. Soc.
2006, 128, 9428-9432.
9
11386 J. AM. CHEM. SOC. VOL. 128, NO. 35, 2006

Biradical Polarizing Agent for DNP in Aqueous Media
A R T I C L E S
Figure 1. 9 and 140 GHz EPR spectra of TEMPO, BT2E, and TOTAPOL with the molecular structures shown in the top line. (a)-(c) are 9 GHz solution
spectra illustrating the extra two lines in the spectrum from the transient proximity of two TEMPO moieties with a strong electron-electron J-coupling
[compare (a) with (b)]. These lines in TOTAPOL are severely broadened in (c) by shorter lifetime of the transient proximity due to rigidity of the tether,
which cannot bend the biradical easily. (d)-(f) illustrate the 9 GHz spectra obtained from frozen solutions at 77 K, and (g)-(i) illustrate the 140 GHz spectra
from the same solutions frozen at 20 K.
Scheme 1. Synthesis of TOTAPOLa
millimeter waves.15 Sapphire, rather than zirconia, rotors are preferred
for the DNP experiments since they transmit microwaves with ∼30%
less attenuation, and we have used rotors of two diameters, 2.5 and 4
mm, in the experiments reported here. As will be seen below, we find
larger enhancements in the smaller diameter system probably because
the microwave penetration is more complete. Finally, the 5 T magnet
has a superconducting sweep coil used to vary the field by (750 G
for EPR and DNP experiments, and that facilitates locating the
maximum and minimum in the DNP enhancement curve from its field
dependence.
3. Results and Discussion
The design of TOTAPOL was dictated by two considerations.
First, following our successful experiments on the BTnE series
of biradicals whose structures are illustrated in Figure 1 (top
row), we wanted to prepare a radical with as short a linker as
possible. Second, the new species was supposed to be soluble
in aqueous media so that it could be used to polarize biological
solids. These considerations motivated us to attempt to prepare
a polarizing agent with a two carbon linkage, such as TEMPO-
O-CH2-CH2-O-TEMPO which would be BT1E (vide infra),
but despite exploration of several approaches, we were not
successful. Subsequently, we did synthesize a biradical with a
three carbon linker with asymmetric bridging atomssN and O
rather than two O’s. In addition we added an -OH moiety to
the central carbon to increase the solubility of the molecule
-TEMPO-NH-CH2-CH(OH)-CH2-O-TEMPO. The fact
that we obtained an enhancement factor of ∼165 (in 4 mm
rotors) is probably due to the short electron-electron distance
and the fortuitously correct orientation of the two TEMPO
g-tensors. In a separate paper we describe radicals such as
a Conditions: (a) epichlorohydrin, tetrabutylammonium hydrogen sulfate,
50% w/w NaOH(aq), room temperature; (b) LiClO4, CH3CN, room
temperature.
BTurea (two TEMPOs tethered by -NH-CO-NH-) that have
a shorter linker but yield enhancements of only about 100 (again
in 4 mm rotors). We believe the lower enhancement is due to
a suboptimal orientation of the g-tensors of the two TEMPO
moieties. In 2.5 mm rotors we observe larger enhancements from
all of the biradicals (up to ∼290 from TOTAPOL), and we
believe that this effect is due to more efficient microwave
penetration of the sample as discussed further below.
1-(TEMPO-4-oxy)-3-(TEMPO-4-amino)propan-2-ol (TO-
TAPOL) was prepared according to the two-step reaction
illustrated in Scheme 1. Note that the molecule is an asymmetric
biradical with an ether and a secondary amino linkage in contrast
to the symmetrical bis-TEMPO-ethylene oxide biradicals that
we described previously.12
9
J. AM. CHEM. SOC. VOL. 128, NO. 35, 2006 11387

A R T I C L E S
Song et al.
The EPR spectra of flexible biradicals such as BTnE, two
TEMPO’s tethered by n ethylene glycol units,12 usually have
two additional peaks yielding a quintet spectrum with the lines
spaced at about half the 14N hyperfine splitting normally
observed in TEMPO, 16.7 G. This five-line solution EPR
spectrum is a result of the proximity of the two TEMPO radicals
and arises when the average J-coupling (exchange integral) is
g10 times stronger than the hyperfine coupling.26 This phe-
nomenon is illustrated in Figure 1a,b which shows the solution
EPR spectra of 4-hydroxy-TEMPO and the biradical BT2E.
There are three lines present in the TEMPO spectrum, but in
the BT2E spectrum there are two additional lines arising from
the strong J-coupling between the electrons. In contrast, the
spectrum of TOTAPOL (Figure 1c) shows that the two
additional biradical lines are broadened into the baseline due
to the short tether (-O-CH2-CHOH-CH2-NH-) that re-
stricts the proximity of two TEMPO moieties.
Figure 2. Illustration of the growth of the nuclear polarization as a result
of microwave irradiation. Integration of the spectral intensities with and
1
without irradiation yields a H enhancement of ∼290 measured indirectly
Moreover, the EPR spectrum of 0.5 mM TOTAPOL in 2H6-
DMSO/2H2O (6:4 w/w) glass-forming solution at 77 K (Figure
1f) reveals a significantly broadened line shape with a resolved
dipolar splitting at 3360 G. Note for comparison that neither
the broadening nor the splitting is present in the spectrum of
frozen monomeric TEMPO at 1 mM (Figure 1d). For TO-
through 13C CP signal using the pulse sequence shown in the inset. The
measurements were performed on a sample of 3 mM TOTAPOL and 2 M
2
13C-urea in H6-DMSO/2H2O/H2O (60:34:6 w/w/w) at 90 K, 5 T, and ωr/
2π ) 7 kHz MAS (spinning sidebands marked by asterisks). The time
constant associated with the growth is ∼9 s, approximately the nuclear T1
of the sample.
27
TAPOL, the ∼8 G (Figure 1f,i) splitting at gzz results from
an intramolecular electron-electron interaction, which is mainly
attributed to a dipolar coupling. As a comparison, the powder
EPR spectra of BT2E reveal broadened features (Figure 1e,h).
Analyses of the powder EPR line shapes yielded similar
electron-electron distances (∼12.8 Å) for both BT2E and
TOTAPOL, even though the two molecular tethers are com-
posed of different numbers of atoms. Since an electron-electron
dipolar splitting can be observed in the powder EPR spectra
(Figure 1f,i), we surmise that the conformational flexibility of
the tether in TOTAPOL may be decreased by its short length.
In contrast, the longer and more flexible tether in BT2E probably
permits a distribution of orientations of the g-tensors of the two
TEMPO moieties and thereby transforms the dipolar splitting
visible at 3360 G to a broad peak (Figure 1e). Similar dipolar
splitting and broadening are present in the 140 GHz biradical
spectra (Figure 1h,i), where inhomogeneous broadening due to
the larger g-anisotropy dominates the line shape. Both the
solution and solid-state EPR spectra of TOTAPOL provide
information on the distance between the two TEMPO moieties,
and a detailed analysis of these line shapes will be the subject
of a future publication.
Figure 3. Top: EPR absorption of TOTAPOL measured on the 140 GHz
spectrometer. The line width is much greater than the 1H Larmor frequency
(ωn/2π) and therefore can encompass two coupled electron spins with the
correct EPR frequency separation, ωn, required for the cross effect.
1
Bottom: Solid circles show the field dependence of the H enhancement
(in a 4 mm rotor) in a DNP experiment using the biradical polarizing agent
TOTAPOL. The field dependence from BT2E, shown as open circles, is
essentially identical with that of TOTAPOL. The lines connecting the data
points are for guidance of the eye.
The pulse sequence for DNP-enhanced 13C-CPMAS NMR
experiments is shown in the inset of Figure 2. The 1H
polarization is initially saturated by a series of 90° pulses
followed by a delay of 3T2. Next, the microwave irradiation is
1
leading to an enhancement factor for H polarization ꢀ ) 290
1
applied to dynamically polarize the H’s or, in the absence of
( 30 determined by comparing the saturated NMR signals after
40 s delay with and without microwaves. Note that the error
for the enhancement factor was determined primarily by the
uncertainty in measuring the intensity of the unenhanced NMR
signal.
microwaves, the thermal equilibrium polarization develops.
1
Finally, the H polarization is transferred to 13C via cross-
polarization and observed in the presence of TPPM decoupling.28
The Fourier transforms of the FIDs are shown (Figure 2) as a
series of spectra with various microwave irradiation periods,
In the top half of Figure 3 we show the EPR absorption of
TOTAPOL recorded at 140 GHz/5 T. This inhomogeneous line
shape supports the cross-effect mechanism in which two dipole-
coupled electrons, separated by ωn/2π in the EPR spectrum,
execute a three-spin electron-electron-nucleus process involv-
ing the mutual flip of an electron and a second electron separated
(26) Luckhurst, G. R., Biradicals as Spin Probes. In Spin Labeling Theory and
Applications; Berliner, L. J., Ed.; Academic Press: New York, 1976; pp
133-181.
(27) Griffith, O. H.; Cornell, D. W.; McConnell, H. M. J. Chem. Phys. 1965,
43, 2909-2910.
(28) Bennett, A. E.; Rienstra, C. M.; Auger, M.; Lakshmi, K. V.; Griffin, R. G.
J. Chem. Phys. 1995, 103, 6951-6958.
9
11388 J. AM. CHEM. SOC. VOL. 128, NO. 35, 2006

Biradical Polarizing Agent for DNP in Aqueous Media
A R T I C L E S
Figure 4. Histogram of DNP enhancements (with error bars) in 4 mm
(light gray) and 2.5 mm (dark gray) rotors with the TOTAPOL, a series of
BTnE biradicals, and monomeric TEMPO. The data illustrate that TO-
TAPOL yields the largest enhancement, especially when the microwave
penetration is optimized in 2.5 mm rotors.
Figure 5. DNP enhanced spectrum of a 9 µL sample of 0.2 M proline
solution doped with 5 mM TOTAPOL. An enhancement of ∼240 was
observed in the experiment with 20 s of microwave irradiation. The sample
was prepared in 2H8-glycerol/2H2O/H2O (60:25:15 w/w/w), and the spectra
were recorded with ωr/2π ) 7.4 kHz MAS (2.5 mm rotor) at 90 K and 5
T. The asterisk indicates a rotational sideband.
by ωn/2π and a nuclear spin. The energy difference matches
the nuclear Larmor frequency and results in nuclear spin flips
that generate the enhanced nuclear polarization. The correct
frequency separation of the two electrons is provided by the
fact that the two TEMPO moieties have relative g-tensor
enhancement strongly depends on the electron-electron dipolar
interaction, it appears to be optimized for BT2E and TOTAPOL
at ꢀ ∼ 165 in a 4 mm rotor. In a 2.5 mm rotor, where the
microwave penetration is more complete, the maximum en-
hancement from TOTAPOL grows to ꢀ ∼ 290. Note that the
increase in ꢀ in going from 4 to 2.5 mm rotors is ∼1.75 for
TOTAPOL whereas for BT2E it is ∼1.36. The underlying
reason for this difference is not clear at present. It could be
related to differences in the electronic relaxation times of the
two biradicals. For systems which are sample-volume limited
the smaller rotor offers advantages in sensitivity. However, when
this is not the case then the 4 mm system could yield improved
signal-to-noise.
A very important feature of TOTAPOL is its solubility in
water due the -OH group on the central carbon of the tethering
chain and the -NH- linkage of the TEMPO moiety. Further,
TOTAPOL is soluble in 200 mM salt, solvent conditions that
are common in preparation of protein samples, and it is stable
in glycerol/water (6:4 w/w) at -10 °C for months. In contrast,
BT2E is insoluble in all of these media. To illustrate the utility
of TOTAPOL in these circumstances, we show in Figure 5 a
DNP-enhanced high-resolution 13C NMR spectrum of 0.2 M
proline in 2H8-glycerol/2H2O/H2O (60:25:15 w/w/w) which
exhibits an enhancement of 240 ( 40 using 5 mM TOTAPOL
at 90 K and 5 T. Note the enhancement with proline is slightly
smaller than the 290 observed with 13C-urea. In addition, we
have used TOTAPOL to polarize samples of membrane pro-
teins such as bacteriorhodopsin29 and amyloid peptide (e.g.,
GNNQQNY) nanocrystals30,31 in glycerol/water solvents, so we
believe that TOTAPOL should be a widely applicable polarizing
agent.
orientations that satisfy the correct matching condition ω2e
-
ω1e ) ωn. The field dependence of the DNP enhancement with
TOTAPOL is shown in the bottom half of Figure 3, together
with the enhancement curve of BT2E. To facilitate observation
of NMR signals in the absence of microwave irradiation, the
field-dependent DNP experiments were performed using a 4
mm sapphire rotor. The DNP enhancement (ꢀ ∼ 165) in 4 mm
rotors (25 µL) yielded by TOTAPOL was smaller than that
observed (ꢀ ∼ 290) in 2.5 mm rotors (9 µL), presumably because
of limited microwave penetration by the larger sample volume.
However, the shape of the curve in Figure 3 and the optimal
external magnetic field for the maximum DNP enhancement
showed no dependence within experimental error on the average
microwave power at the sample.7 In particular, TEMPO-based
biradicals exhibit a universal DNP enhancement profile since
the electron-electron dipolar splittings are relatively small
compared to the inhomogeneous broadening of EPR line shapes.
As is illustrated in Figure 3, the enhancement profile curves
generally resemble the first derivative of the EPR absorptive
line shapes, having the maxima and minima of about 165 and
-140, respectively, separated by 107 G (∼300 MHz). The
enhancement curves can be compared with the EPR line shape
in the top half of Figure 3. Although the TOTAPOL chain is
shorter than in BT2E and probably less flexible, we nevertheless
observe essentially the same shape for the enhancement curve
for each case. The shape and amplitude of the two curves depend
on the average electron-electron distance and relative orienta-
tions of the two g-tensors, which must be very nearly the same
in the two molecules.
Figure 4 compares the enhancement obtained with TOTAPOL
with those from the BTnE (n ) 2, 3, 4) series and monomeric
TEMPO. The comparison was made using both 4 and 2.5 mm
rotors to demonstrate the effect of microwave penetration into
samples. Reducing the number of atoms separating the two
TEMPO moieties increases the electron-electron dipolar in-
teraction and the observed enhancement. Since the DNP
(29) Bajaj, V. S.; Mak, M.; Veshtort, M.; Herzfeld, J.; Griffin, R. G. In Dynamic
Nuclear Polarization-Enhanced Solid State NMR InVestigations of the
Photocycle Intermediates of Bacteriorhodopsin; 47th Experimental Nuclear
Magnetic Resonance Conference: Asilomar, CA, 2006.
(30) Nelson, R.; Sawaya, M. R.; Balbirnie, M.; Madsen, A. O.; Riekel, C.;
Grothe, R.; Eisenberg, D. Nature 2005, 435, 773-778.
(31) Van Der Wel, P.; Hu, K.-N.; Lewandowski, J.; Griffin, R. G. J. Am. Chem.
Soc. 2006, ASAP Article.
9
J. AM. CHEM. SOC. VOL. 128, NO. 35, 2006 11389

A R T I C L E S
Song et al.
4. Conclusions
has yielded interesting results in solid-state NMR studies of bR
and GNNQQNYsand observed larger enhancements than
obtained with monomeric TEMPO. Thus, it would appear
TOTAPOL will find wide applicability in DNP-enhanced NMR
spectroscopy of biological systems.
To summarize, we have synthesized a new TEMPO-based
biradical, TOTAPOL, which produces larger enhancements than
those observed with BT2E and, very importantly, is soluble in
aqueous media containing salt and glycerol at concentrations
typically found in protein solutions. These properties should
make this molecule widely applicable to DNP investigations
involving biological systems. The sizes of the enhancements
presently observed reach 290 in a 2.5 mm rotor or 165 in a 4
mm rotor at 90 K and 5 T. As mentioned above, we have already
used TOTAPOL in a number of other systemssin particular it
Acknowledgment. This research is supported by the National
Institutes of Health through Grants EB-002804 and EB-002026
to R.G.G. and by the Office of Naval Research through a grant
to T.M.S.. We gratefully acknowledge the earlier collaboration
and conversations with Dr. Hsiao-hua Yu.
JA061284B
9
11390 J. AM. CHEM. SOC. VOL. 128, NO. 35, 2006