вход по аккаунту


Scaling the Alignment of Small Organic Molecules in Substituted Polyglutamates by Variable-Angle Sample Spinning.

код для вставкиСкачать
NMR Spectroscopy
Scaling the Alignment of Small Organic
Molecules in Substituted Polyglutamates by
Variable-Angle Sample Spinning**
Christina M. Thiele*
The determination of the relative configuration of small
organic molecules by NMR spectroscopy usually relies on the
analysis of NOE data and 3J coupling constants. Recently the
approach of using residual dipolar couplings (RDCs) has
come into focus, as RDCs contain both distance and angle
information. It is, for example, possible to assign all diastereotopic protons in strychnine simultaneously without the use
of NOE data[1] and to determine the complete relative
configuration of a fluorinated compound containing five
stereocenters.[2] In order to observe residual dipolar couplings, however, it is necessary to induce partial alignment of
the molecule in the magnetic field. Alignment media inducing
weak alignment for organic molecules are still scarce. The
liquid-crystalline polyglutamates[1, 3–6] and deuterated PCBP
(4-n-pentyl-4’-cyanobiphenyl)[7] are used for this purpose.
Recently the use of crosslinked polystyrene (PS) and crosslinked poly(dimethylsiloxane) (PDMS) was described, and
here the degree of alignment is dependent on the degree of
The disadvantage of liquid-crystalline media is that below
a minimum concentration of the liquid crystal in the solvent,
the phase becomes isotropic and no longer induces alignment.
Thus the degree of alignment can be adjusted within specific
regions only. Scalability is possible with polymer sticks,
although the degree of alignment cannot be adjusted after
sample preparation, which might be problematic in the
structure elucidation of natural products due to limited
sample amounts. An alternative is provided by variableangle sample spinning (VASS), where the liquid-crystalline
sample is spun at an angle Vr to the magnetic field in order to
reduce anisotropic interactions.[11–15] We present here a
straightforward approach to scaling residual dipolar couplings
in the chiral orienting media poly(g-benzyl-l-glutamate)/
CDCl3 (PBLG) and poly(g-ethyl-l-glutamate)/CDCl3
(PELG) by applying variable-angle sample spinning.
If a liquid crystal with molecular positive magnetic
anisotropy (Dcm > 0) is spun at an angle Vr to the magnetic
field which is smaller than the magic angle (Vr < 54.78) and
[*] Dr. C. M. Thiele
Institut fr Analytische Chemie
Universitt Leipzig
Johannisallee 29, 04103 Leipzig (Germany)
Fax: (+ 49) 341-97-36115
[**] This work was supported by the Fonds der Chemischen Industrie. I
thank Prof. Dr. S. Berger, Dr. P. Lesot, Dr. M. Zweckstetter, and
Prof. Dr. D. Freude for their help.
Supporting information for this article is available on the WWW
under or from the author.
Angew. Chem. Int. Ed. 2005, 44, 2787 –2790
with a spinning speed considerably higher than its critical
spinning speed wc, then the director of the liquid crystal aligns
parallel to the spinning axis. The anisotropic interactions of
the solute (and the liquid crystal) are directly related to the
angle Vr since they are proportional to the second Legendre
polynomial (P2(cos Vr) = (3 cos2Vr1)/2). Thus by changing
the angle Vr the dipolar interactions can be scaled. The total
spin–spin coupling constant T (also called the effective
coupling constant Jeff) in spectra can obtained from Equation (1).[11]
T ¼ J eff ¼ J iso þ
3 cos2 Vr 1
ð2 D þ J aniso Þ
Here, T is the total spin–spin coupling constant, Vr the
angle to the magnetic field, Jiso the isotropic (scalar) coupling
constant, D residual dipolar coupling, and Janiso the anisotropy
of the scalar coupling, which is usually neglected.
The two polyglutamates used in this study, PBLG and
PELG, are commercially available, and the samples were
prepared as described previously with strychnine[5, 1] and other
compounds.[16] The samples were checked for homogeneity in
standard 5-mm liquid-state NMR tubes. If the lines of the
solvent CDCl3 displayed a sharp doublet in the deuterium
spectrum due to quadrupolar splitting, the viscous liquid
crystal was transferred into a 4-mm solid-state rotor and
closed with a cap without a hole to limit solvent evaporation.
The samples were spun at various angles relative to the
magnetic field (at the same spinning speed). For this purpose
a Bruker DRX-600 spectrometer equipped with a standard
dual (1H,13C) Bruker HR-MAS probe with z-gradients was
used. The deuterium spectra were recorded (using the lock
channel) to get an estimate of the reduction in the observed
couplings (Figure 1). As there is no goniometer on the probe,
the angle was calibrated using the quadrupolar splitting of the
solvent within the liquid crystal. To record spectra of the static
case, the measurement was performed without rotation. In
this case the liquid crystal aligns parallel to the magnetic field
at whatever angle the rotor stands. For PBLG this was a
spontaneous process; for PELG, however, equilibration
within the probe for 20 min was necessary to get sharp lines.
Care must be taken when adjusting the spinning speed. The
spinning speed, which has to be exceeded to avoid spinning
side bands and guarantee alignment of the director with the
rotation axis, shows a square dependence on the magnetic
field. In theory the alignment of the director parallel to the
spinning axis should improve with higher spinning frequencies.[11] A different observation was made here, which is in
accordance with results published by Lesot et al.,[17] where a
second critical spinning speed was observed, above which the
director is distributed at various angles towards the magnetic
field. Therefore the spinning speed was chosen to be 1180 Hz
for PBLG and 1002 Hz for PELG in order to stay within these
two borders at a magnetic field strength of 14.10 T (corresponding to a 1H NMR frequency of 600 MHz). At spinning
speeds of less than 1000 Hz rotation was not stable with our
HR-MAS probe, thus 1000 Hz was the lower limit. At
spinning speeds of more than 1400 Hz for PBLG and
DOI: 10.1002/anie.200461532
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 1. One-dimensional 2H NMR spectra (92.09 MHz) of CDCl3 in
PBLG. Vr (on the left side of each trace) is varied to give DnQ values
between 0 Hz (top, magic angle) and 481 Hz (bottom, static case, 08).
Spinning speed 1180 Hz, temperature 298 K.
3500 Hz for PELG the behavior of the lines was the same as
that described by Lesot et al.[17]
Two measurements were performed to extract one-bond
C-1H RDCs. In one measurement an isotropic solution of
strychnine was used to give the scalar coupling constant J (or
the nematic sample is spun at the magic angle); the other
measurement was conducted in the nematic phase using a
solid-state rotor at an angle where the size of total couplings is
convenient. The measurement of RDCs for strychnine at
angles close to the magic angle (54.7–48.58 for PBLG and all
angles used for PELG, respectively) HSQC spectra without
decoupling during acquisition could be used without problems (see Figure 2), as the dipolar interactions remain small
(as expected). At angles farther from the magic angle gateddecoupled carbon spectra were used to extract residual
dipolar coupling. As can be seen from Figure 3 b, 13C-1H
RDCs can be scaled according to requirements. At the top the
isotropic spectrum (spinning at the magic angle) is depicted,
thus giving the scalar couplings. From top to bottom the
degree of anisotropic interaction increases leading, for
example, from a triplet for C15 (methylene group) in the
isotropic case (Vr = 54.78) to a doublet of doublets (Vr =
40.38) with both D values smaller than Jiso. This then develops
further to a doublet-like signal for C15 in the static case,
where 2 D for one proton is equal to Jiso, giving a spin–spin
coupling constant T of 0 Hz, and one D is of the size of
15.1 Hz giving a T value of 108.8 Hz for the other proton.
The considerable broadening of the lines, due to among other
things long-range C–H couplings, leads to large errors in D for
the static case as described previously.[5] This example of the
two protons on C15 clearly shows the power of this approach,
as it was not considered advisable to extract couplings of this
size from static spectra only.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Figure 2. Two-dimensional HSQC spectrum of strychnine in PBLG/
CDCl3 at Vr = 50.38 recorded without 13C decoupling during acquisition
(1180 Hz, 298 K). As can be seen from the 1D slice at a 13C chemical
shift of 42 ppm the total spin–spin C–H couplings can be extracted
easily. No signals of PBLG are seen in the 2D spectrum.
For PELG the measurement of virtually all RDCs was
possible at all chosen angles including the static case (see the
Supporting Information). A linear correlation of the size of
the dipolar coupling of the solute (D/Hz) vs. the quadrupolar
splitting of the organic solvent CDCl3 (DnQ/Hz) shows
excellent fits (R2 larger than 0.94) for 13 out of 22 couplings.
Due to the limited amount of data for PELG (only three
different angles were examined in addition to the static case)
no statement can be made concerning the nine other
couplings. An extrapolation of the values for the static case
from those obtained by variable-angle sample spinning gives a
good estimate of the size of all couplings. For PBLG, however,
very few couplings could be measured for the static case as
described previously.[5] Using the approach of variable-angle
sample spinning virtually all couplings could be measured at
the angles between 408 and 508, and linear behavior for D vs.
DnQ was observed for most of the couplings (see Figure 4).
The slope of this straight line can then be used to extrapolate
the values for the static case; these values can be confirmed by
the spectra ( 10 Hz is, for example, only 3 % error for C1 and
C4) for 17 out of 22 couplings, thus also revealing earlier
misinterpretations of the spectra.
The alignment tensors for strychnine in PBLG and PELG
were calculated using the program package PALES using the
modules “bestfit” and “addition of dipolar coupling
noise” [18, 19] Significant changes in the degree of alignment
within each sample were found, which increased upon
increasing degree of anisotropy introduced (increasing DnQ)
as expected (Table 1).
The degree of orientation in PBLG and PELG is different, as previously described.[1] With these new experiments,
however, the orientation of alignment for PBLG can be
determined more reliably, thus revealing differences in the
orientation of strychnine in PBLG and PELG. The x, y and z
Angew. Chem. Int. Ed. 2005, 44, 2787 –2790
Figure 4. Plot of observed residual dipolar coupling constants D [Hz]
of strychnine in PBLG vs. quadrupolar splitting DnQ [Hz] of the solvent
CDCl3 (as a measure for Vr). Linear behavior can be observed for all
RDCs (not all lines shown).
Figure 3. a) Structural formula of strychnine and numbering as used in
the text. b) Part of 13C gated decoupled 1D NMR spectra of strychnine
(signal of carbon atoms denoted C14 and C15) in PBLG/CDCl3 at a
magnetic field strength of 14.1 T. Vr is varied from the magic angle
(top) to 08, the static case (bottom).
axes differ by 118, 128 and 68, respectively (see the Supporting
Information for details). Whether this difference in the
orientation of alignment is strong enough to consider these
two alignment media as linearly independent remains to be
seen. If this is indeed the case, they could then be used for the
determination of the relative configuration at highly substituted stereocenters as well as for the study of dynamic
processes, as has been shown impressively for biomolecules.[20–22]
In summary, we have demonstrated the usefulness of
variable-angle sample spinning to scale partial alignment of
small organic molecules. Extraction of RDCs becomes now
convenient since it is possible to reduce the RDCs according
to requirements from one single sample. Even the isotropic
spectrum can be obtained from the anisotropic sample by
spinning at the magic angle for the liquid crystals used here.
One drawback might be the need for a probe that can be spun
at different angles. We used a standard Bruker HR-MAS
probe available to many organic chemistry groups; a specially
built VASS probe is not required.
The alignment media used here induce orientations of the
solute complimentary to those possible with the recently
described PS sticks. This is important as it broadens the
applicability of using RDCs in structure elucidation of small
organic molecules, an approach that will certainly remove
many ambiguities in the structure determination of natural
Received: August 4, 2004
Revised: January 10, 2005
Published online: March 30, 2005
Keywords: NMR spectroscopy · polyglutamates ·
residual dipolar couplings · structure determination
Table 1: Alignment properties of PBLG and PELG at various angles Vr.[a]
DnQ [Hz]
Vr [8]
Da [104 Hz]
[a] Da is the axial component and Dr/Da the rhombicity of the alignment tensor. n(RDC) is the number of RDCs used for the calculation, R is Pearson’s
correlation coefficient of experimental versus back-calculated dipolar couplings within PALES.
Angew. Chem. Int. Ed. 2005, 44, 2787 –2790
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
[1] C. M. Thiele, J. Org. Chem. 2004, 69, 7403 – 7413, and references
[2] J. Yan, F. Delagio, A. Kaerner, A. D. Kline, H. Mo, M. J. Shapiro,
T. A. Smitka, G. A. Stephenson, E. R. Zartler, J. Am. Chem. Soc.
2004, 126, 5008 – 5017.
[3] C. Aroulanda, V. Boucard, F. Guib, J. Courtieu, D. Merlet,
Chem. Eur. J. 2003, 9, 4536 – 4539.
[4] L. Verdier, P. Sakhaii, M. Zweckstetter, C. Griesinger, J. Magn.
Reson. 2003, 163, 353 – 359.
[5] C. M. Thiele, S. Berger, Org. Lett. 2003, 5, 705 – 708.
[6] C. Aroulanda, P. Lesot, D. Merlet, J. Courtieu, J. Phys. Chem. A
2003, 107, 10 911 – 10 918.
[7] B. Bendiak, J. Am. Chem. Soc. 2002, 124, 14 862 – 14 863.
[8] B. Luy, K. Kobzar, H. Kessler, Angew. Chem. 2004, 116, 1112 –
1115; Angew . Chem. Int. Ed. 2004, 43, 1092 – 1094.
[9] J. C. Freudenberger, P. Spiteller, R. Bauer, H. Kessler, B. Luy, J.
Am. Chem. Soc. 2004, 126, 14 690 – 14 691.
[10] Very recently published polymer alignment media for polar
substances: J. C. Freudenberger, S. Knr, K. Kobzar, D. Heckmann, T. Paululat, H. Kessler, B. Luy, Angew. Chem. 2005, 117,
427 – 430; Angew. Chem. Int. Ed. 2005, 44, 423 – 426; P. Haberz, J.
Farjon, C. Griesinger, Angew. Chem. 2005, 117, 431 – 433;
Angew. Chem. Int. Ed. 2005, 44, 427 – 429.
[11] J. Courtieu, J. P. Bayle, B. M. Fung, Prog. Nucl. Magn. Reson.
Spectrosc. 1994, 26, 141 – 169.
[12] R. R. Havlin, G. H. J. Park, T. Mazur, A. Pines, J. Am. Chem.
Soc. 2003, 125, 7998 – 8006.
[13] G. Zandomeneghi, M. Tomaselli, P. T. F. Williamson, B. H.
Meier, J. Biomol. NMR 2003, 25, 113 – 123.
[14] J. Courtieu, D. W. Alderman, D. M. Grant, J. P. Bayle, J. Chem.
Phys. 1982, 77, 723 – 730.
[15] N. Lancelot, K. Elbayed, A. Bianco, M. Piotto, J. Biomol. NMR
2004, 29, 259 – 269.
[16] C. Aroulanda, M. Sarfarti, J. Courtieu, P. Lesot, Enantiomer
2001, 6, 281 – 287.
[17] P. Lesot, J. W. Emsley, J. Courtieu, Liq. Cryst. 1998, 25, 123 – 127.
[18] M. Zweckstetter, A. Bax, J. Am. Chem. Soc. 2000, 122, 3791 –
[19] M. Zweckstetter, A. Bax, J. Biomol. NMR 2002, 23, 127 – 137.
[20] J. Meiler, J. J. Pompers, W. Peti, C. Griesinger, R. Brschweiler,
J. Am. Chem. Soc. 2001, 123, 6098 – 6107.
[21] J. R. Tolman, J. Am. Chem. Soc. 2002, 124, 12 020 – 12 030.
[22] T. S. Ulmer, B. E. Ramirez, F. Delagio, A. Bax, J. Am. Chem.
Soc. 2003, 125, 9179 – 9191.
2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2005, 44, 2787 –2790
Без категории
Размер файла
137 Кб
spinning, alignment, variables, organiz, molecules, small, samples, angl, scaling, substituted, polyglutamates
Пожаловаться на содержимое документа