Rapid communication Received: 11 July 2013

Revised: 9 October 2013

Accepted: 24 October 2013

Published online in Wiley Online Library: 26 November 2013

(wileyonlinelibrary.com) DOI 10.1002/mrc.4032

BASHD-J-resolved-HMBC, an efficient method for measuring proton–proton and heteronuclear long-range coupling constants Kazuo Furihataa* and Mitsuru Tashirob Natural products often possess various spin systems consisting of a methine group directly bonded to a methyl group (e.g. –CHa–CHb(CH3)–CHc–). The methine proton Hb splits into a broadened multiplet by coupling with several vicinal protons, rendering analysis difficult of nJC–H with respect to Hb in the J-resolved HMBC-1. In purpose of the reliable and easy measurements of nJC–H and nJH–H in the aforesaid spin system, we have developed a new technique, named BASHD-J-resolved-HMBC. This method incorporates band selective homo decoupled pulse and J-scaling pulse into HMBC. In this method, high resolution cross peaks can be observed along the F1 axis by J-scaling pulse, and band selective homo decoupled pulse simplified multiplet signals. Determinations of nJC–H and nJH–H of multiplet signals can easily be performed using the proposed pulse sequence. Copyright © 2013 John Wiley & Sons, Ltd. Keywords: BASHD; BASHD-J-resolved HMBC; J-resolved HMBC; selective J-resolved HMBC; heteronuclear long range coupling constants

Introduction

Experimental

Determination of the long-range 13C–1H couplings (nJC–H) and vicinal H–H couplings (3JH–H) provides valuable clues in the stereochemical studies of natural products. Various proton-detected methods were presented to determine long-range heteronuclear coupling constants.[1–5] We previously reported J-resolved HMBC-1 (Fig. 1(a)) and HMBC-2 for measuring nJC–H.[6] In the J-resolved HMBC-1, splittings of cross peak signals were observed owing to both nJH–H and nJC–H, resulting in difficult analysis of nJH–H and nJC–H in the complicated spin systems. Although the J-resolved HMBC-2 that incorporated the constant time method, together with 1 1 H- H decoupling, provided only nJC–H, the constant time delay caused considerable reduction of the signal intensities due to the short T2 effects in the analysis of the relatively large natural compounds. Generally, the analysis of these coupling constants is very difficult and unpractical for the complicatedly split cross peaks in these methods. For example, natural products, such as polyketides, often possess several spin systems, consisting of a methine group directly bonded to a methyl group (e.g. –CHa–CHb(CH3)–CHc–). In such a spin system, the methine proton Hb splits into a broadened multiplet by coupling with several vicinal protons, rendering analysis difficult of 3JH–H and nJC–H of Hb in the J-resolved HMBC-1. In the analysis of the relative configurations of such stereocenters, measurements of these coupling constants are essential. In observation of nJC–H and 3JH–H in the aforementioned spin system, one of the feasible solution is to make the complicated spin system simple by decoupling methyl groups. Considering this solution and the aforementioned problems associated with J-resolved HMBC-1, we have developed a new pulse sequence, the band selective homo decoupled (BASHD)-J-resolved HMBC. The key of this method is to selectively decouple methyl groups, which cause broadened multiplets. The effectiveness of the proposed pulse sequence has been investigated using portmicin.

The 20 mg of portmicin[7] was dissolved in 400 μl of C6D6. All spectra were recorded at 20 °C on a Varian Inova 500 spectrometer. The experimental parameters of all experiments were as follows: data size in t1 = 256 points, data size in t2 = 2048, spectral width in f1 = 14000 Hz, f2 = 2500 Hz, number of transients per increment = 328, n (scaling factor) = 25, nt1 max = 457 ms, and recycle time = 1.5 s. The selective 180o pulse widths were 6.1 ms (re-burp)[8] with an excitation range of 0.3–2.0 ppm in the BASHD-J-resolved-HMBC and 24.4 ms with an excitation range of 2.24–2.64 ppm in the selective J-resolved-HMBC.[9] The sinebell window function was used for both axes.

Pulse sequences of the J-resolved-HMBC-1 and BASHD-J-resolved-HMBC are shown in Fig. 1. In the BASHD-J-resolved-HMBC pulse sequence (Fig. 1(b)), the BASHD scheme,[10–12] which is a combination of a selective and a hard 180o pulses, is incorporated into the J-resolved portion of the J-resolved-HMBC-1. In the evolution period of nt1, the consecutive selective 180o and hard 180o pulses, applied only to the methyl group, has an effect of 0o flip, and the gradient pulses G1 eliminate its magnetizations by dephasing. The target protons, corresponding to the methine

* Correspondence to: Kazuo Furihata, Division of Agriculture and Agricultural Life Sciences, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113–8657, Japan. E-mail: [email protected] a Division of Agriculture and Agricultural Life Sciences, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan b Department of Chemistry, College of Science and Technology, Meisei University, Hino, Tokyo 191-8506, Japan

Copyright © 2013 John Wiley & Sons, Ltd.

27

Magn. Reson. Chem. 2014, 52, 27–31

Results and Discussion

K. Furihata and M. Tashiro

Figure 1. Pulse sequences of (a) J-resolved HMBC-1 and (b) band selective homo decoupled J-resolved HMBC. The thin and thick bars represent 90 and 180° pulses, respectively, and the sine shape is a selective 180° pulse. All pulses were along x unless otherwise shown. The phase cycling was Φ1 = x, x, 1 x, x; Φr = x, x, x, x. The G3 and G4 were set to a ratio of 2 : 1 (amplitude of 7.0 and 3.5 G cm , respectively, and duration of 1.0 ms) for CH coherence 1 selection. Both G1 and G2 were set to amplitude of 1 G cm and duration of 1 ms. The selective 180° pulse was applied as re-burp type profile. Low pass J-filter delay (Δ) was set to 3.5 ms. The scaling factor (n) must be set so as to give the nt1 max value larger than 1/JC-H.

protons in the aforementioned spin system ( CHa–CHb(CH3)– CHc–), are turned over by a hard 180o pulse and refocused at the end of nt1 with modulation by 1H-1H spin couplings. Homodecoupling between the methyl group, experienced 0o flip, and the rest including methine protons, experienced 180o flip, can be carried out in the evolution period of nt1.[10–12] In the constant-

time-HMBC portion, the 1H–1H and 1H–13C spin couplings can be measured as the displacements of n*JC–H and n*JH–H, whereas these couplings can be measured as those of n*JC–H and ( n + 1) *JH–H in the J-resolved HMBC-1. The confusing measurements of JC–H and JH–H can be avoided in the present method, because both amplitudes of displacements are the exact scaling factor n.

28

Figure 2. Schematic drawings of the cross peak patterns of (a) J-resolved HMBC-1, (b) selective J-resolved HMBC, and (c) band selective homo decoupled J-resolved HMBC. The selective decouple of methyl groups is applied in (b) and (c). Scaling factor is n, and F1 and F2 axes are exchanged.

wileyonlinelibrary.com/journal/mrc

Copyright © 2013 John Wiley & Sons, Ltd.

Magn. Reson. Chem. 2014, 52, 27–31

BASHD-J-resolved-HMBC, an efficient method for measuring JC-H and JH-H

Figure 3. The structure of portmicin

The schematic spectral patterns of the BASHD J-resolved HMBC are illustrated in Fig. 2. In the J-resolved HMBC-1 spectrum (Fig. 2(a)), all cross peaks are observed as tilted doublet-multiplet signals owing to the 1H–1H and 13C–1H J-modulations. Each cross peak is separated by n*JC–H, where n is a scaling factor.[13–15] In the selective J-resolved HMBC spectrum (Fig. 2(b)), the 1H–1H decoupling, which can eliminate undesirable splitting by nJH–H, enables simple measurements of nJC–H. As can be easily imagined, measurements of nJH–H are inapplicable. In the BASHD J-resolved HMBC, the 1H–1H decoupling between methyl group and Ha results in a simple spectral pattern. The cross peaks are split by n*JC–H and n*JH–H as shown in Fig. 2(c). The n*JC–H corresponds to the parallel displacement along the F1 axis, and n*JH–H corresponds to the tilted displacement. This spectral pattern provides the clear distinction between JC–H and JH–H. The full spectrum of portmicin (Fig. 3) acquired using the BASHD-J-resolved HMBC pulse sequence is shown in Fig. 4. The BASHD region by re-burp pulse is shown in the attached 1H spectrum. The selective pulse excited a range from 0.3 ppm to 2.0 ppm to decouple a methyl group. The expanded spectra for H4 of portmicin acquired using the J-resolved HMBC-1, selective J-resolved HMBC, and BASHD J-resolved HMBC are compared in Fig. 5. The methine proton H4, directly bonded to a methyl group, is split into a broadened multiplet owing to the couplings

with H3, H5, and methyl protons. This complicated couplings, so far, resulted in the impractical analysis of the stereo study of H4. In the J-resolved HMBC-1 spectrum, H4 cross peaks appeared as tilted multiplet due to H–H and C–H J modulation. In the analysis of these cross peaks, nJC–H can be measured; however, nJH–H cannot be analyzed (Fig. 5(a)). In the selective J-resolved HMBC, all nJH–H in the F1 dimension are decoupled, and lengthwise cross peaks are observed (Fig. 5(b)). The cross peaks that split by n*JC–H are observed as doublets. In the projection spectrum, cross peaks are simplified, and the magnitude of nJC–H can be clearly determined from this splitting. However, nJH–H cannot be determined in the selective J-resolved HMBC. In the BASHD J-resolved HMBC, 1H–1H couplings of methyl group are suppressed by decoupling, and the cross peaks of H4 appear as tilted doublet owing to H–H and C–H J modulation (Fig. 5(c)). In the projection spectra, cross peak patterns are more simplified than those of the J-resolved HMBC-1. In the BASHD J-resolved-HMBC, the 3JH3–H4 was determined to be less than 3.0 Hz by H4-C6 cross peak (Fig. 5(c)), indicating the conformation of H3–H4 to be gauche as shown in Fig. 6(a). The 2JH4–C3 was determined to be ca. 6.3 Hz, indicating the conformation of H4 and a methoxy group at C3 to be gauche. The 3JH4–C2 and 3 JH4–H5 were determined to be 5.2 and 10.9 Hz, respectively, indicating the both conformations of H4–C2 and H4–H5 to be anti (Fig. 6). The 3JH4–C6 was determined to be less than 3.0 Hz, indicating the conformation of H4–C6 to be gauche. The 2JH4–C5 was determined to be 7.8 Hz, indicating the conformation of H4 and an oxygen at C5 to be gauche (Fig. 6(b)). In the BASHD J-resolved HMBC, both nJC–H and nJH–H can be measured along the F1 axis. The J-scaling method was introduced, where nt1 max was set to the values from 300 ms to 500 ms. In observation of the coupling around nJC–H = 2 Hz, the digital resolution must be smaller than (n × 2) Hertz and

Magn. Reson. Chem. 2014, 52, 27–31

Copyright © 2013 John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/mrc

29

Figure 4. The band selective homo decoupled J-resolved HMBC spectrum of portmicin. The boxed region is expanded in Fig. 5(c). In the attached 1H spectrum, the selectively decoupled region is indicated.

K. Furihata and M. Tashiro

Figure 5. The expanded spectra of portmicin acquired using (a) J-resolved HMBC-1, (b) selective J-resolved HMBC, and (c) band selective homo decoupled J-resolved HMBC pulse sequences. Attached are the conventional 1H spectra along the F2 axis and projections of the expanded region along the F1 axis.

(n × t1 max) must be larger than 500 ms. In the present method, the scaling factor (n) was set to allow nt1 max being larger than 1/J as in the case of the conventional 2D J-resolved spectra to amplify the spin coupling constants by a factor of n for both

n

JH–H and nJC–H. Although the broadband X nucleus for longrange couplings (XLOC) technique was proposed for measurements of nJH–H in ethyl trans-cinnamate,[16] its application to the multiplet signal would be limited. In the present method, the BASHD scheme simplified multiplet signals to be measured. In conclusion, the BASHD J-resolved HMBC is a useful technique to measure nJC–H and nJH–H of the multiplet proton signals. This method provides the more simplified cross peaks than J-resolved HMBC-1 with an advantage of the reliable measurement of coupling constants, which cannot be easily determined using other pulse sequences. Acknowledgements

30

Figure 6. Newman projections along the (a) C3-C4 and (b) C4-C5 bonds of portmicin showing several JC-H and JH-H.

wileyonlinelibrary.com/journal/mrc

This study was supported by two Grant-in-Aids for Scientific Research (No. 20580108 for Kazuo Furihata and 21550092 for Mitsuru Tashiro) from the Ministry of Education, Culture, Sports, Science, and Technology.

Copyright © 2013 John Wiley & Sons, Ltd.

Magn. Reson. Chem. 2014, 52, 27–31

BASHD-J-resolved-HMBC, an efficient method for measuring JC-H and JH-H

References [1] [2] [3] [4] [5] [6] [7]

J. Keeler, D. Neuhaus, J. J. Titman. Chem. Phys. Lett. 1988, 146, 545. W. Bermel, K. Wagner, C. Griesinger. J. Magn. Reson. 1989, 83, 223. D. G. Davis. J. Magn. Reson. 1989, 83, 212. J. J. Titman, D. Neuhaus, J. Keeler. J. Magn. Reson. 1989, 85, 111. V. V. Krishnamurthy. J. Magn. Reson. A 1996, 121, 33. K. Furihata, H. Seto. Tetrahedron Lett. 1999, 40, 6271. H. Nakayama, K. Furihata, H. Seto, N. Otake. Tetrahedron Lett. 1981, 22, 5217.

[8] G. Otting, L. P. M. Orbons, K. Wüthrich. J. Magn. Reson. 1990, 89, 423. [9] K. Furihata, M. Tashiro, H. Seto. Magn. Reson. Chem. 2009, 47, 814. [10] H. Nakayama, K. Furihata, H. Seto, N. Otake. Tetrahedron Lett. 1981, 22, 5217. [11] H. Geen, R. Freeman. J. Magn. Reson. 1991, 93, 93. [12] R. V. Hosur, M. Ravikumar, A. Sheth. J. Magn. Reson. 1985, 65, 375. [13] V. V. Krishnamurthy. J. Magn. Reson. A 1996, 121, 33. [14] R. V. Hosur, M. Ravikumar, A. Sheth. J. Magn. Reson. 1985, 65, 375. [15] V. V. Krishnamurthy. J. Magn. Reson. B 1996, 113, 46. [16] A. Meissner, O. W. Sørensen. Magn. Reson. Chem. 2001, 39, 49.

31

Magn. Reson. Chem. 2014, 52, 27–31

Copyright © 2013 John Wiley & Sons, Ltd.

wileyonlinelibrary.com/journal/mrc

BASHD-J-resolved-HMBC, an efficient method for measuring proton-proton and heteronuclear long-range coupling constants.

Natural products often possess various spin systems consisting of a methine group directly bonded to a methyl group (e.g. -CHa-CHb(CH3)-CHc-). The met...
3MB Sizes 0 Downloads 0 Views