Abstract
The relative magnitudes of the chemical shift differences (Δδs) in the two diastereomers of menthyl esters of known chiral derivatizing agents (CDAs) were compared to those of the 〈-methoxy-〈-trifluoromethyl-1-naphthylacetyl (MTN(1)A) analogs I. Discrimination of the terminal diastereotopic methyl resonances in esters of the homologous, symmetrical carbinols II was evaluated. Remarkably, the methyls differed in the MTN(1)A esters III even when n = 15; an unexpected crossover in the sign of the Δδ values was also observed.
Mosher ester/amide anaylsis for deducing the absolute configuration of stereogenic carbinol/amino centers of organic compounds, when properly applied, is a very powerful and widely used method.1 The Mosher method is the prototype of a larger set of analogous 1H-NMR-based methods that use chiral auxiliaries that are structurally related to the α-methoxy-α-trifluoromethylphenylacetyl (MTPA) moiety.2 All of these related methods3 rely on installation of a chiral derivatizing agent (CDA) into the analyte of interest; the newly installed chiral entity introduces local magnetic anisotropy that differentially influences diastereotopic proton resonances in the derivative. Most typically, one makes a complementary pair of diastereomeric derivatives of the analyte of interest using enantiomerically enriched samples of each antipode of the CDA. Comparison of the sets of differential chemical shifts (Δδs) for analogous proton resonances in the spectrum of each diastereomer allows one to deduce the configuration of the parent alcohol/amine. These empirical methods rely on an understanding, or at least a validated mnemonic construct, of the dominant conformation adopted by the ester/amide derivative.
A related interesting issue is the distance over which the anisotropic differential shielding effects exert themselves. The inherent nature of both the analyte and the derivatizing agent play a role. The studies we report here shed light on some of these issues. In particular, they provide insight about the relative ability of various CDAs (from among 1–11, Figure 1) to discriminate proton resonances distal to the point of attachment. Greater anisotropic reach presumably correlates with an inherently more useful CDA.4
Figure 1.
Known Mosher-like chiral derivatizing agents (CDAs).
Perspective on the choice of CDA (more specifically, on its ability to discriminate analogous protons in a pair of diastereomeric derivatives of an analyte) is gained from consideration of the Δδ data for the (–)-menthol derivatives (#-m) summarized in Table 1. First, the magnitude of the Δδs is enhanced by the replacement of the phenyl with a naphthyl or anthracenyl moeity in the CDA. This is easily seen, for example, in the relative magnitude of the “mean|Δδ|” values in the bottom line of Table 1. Within each of the sets MPA vs. either MN(1)A or MA(9)A, MPP vs. MN(1)P, and MTPA vs. MTN(1)A the discriminating power of naphthyl or anthracenyl based CDA is greater [by factors of approximately three (cf. 1-m vs. either 4-m or 9-m), ten (cf. 2-m vs. 5-m), and five (cf. 3-m vs. 6-m), respectively]. Second, whereas MPA shows greater discrimination than MTPA across the mean |Δδ| values for the same set of menthyl protons (cf. 1-m vs. 3-m) the opposite is true for the MN(1)A vs. MTN(1)A pairs (cf. 4-m vs. 6-m). Third, the difference in the position of substitution on the CDA aromatic moiety (i.e., C1 vs. C2 positions for naphthyl and C9 vs. C2 positions for anthracenyl groups) also affects the discriminating power of MN(1)A/MN(2)A, MN(1)P/MN(2)P, and MA(9)A/MA(2)A. Specifically, mean |Δδ| values are lower for the “C2” substituted positional isomers [by factors of approximately three (cf. 4-m vs. 7-m), ten (cf. 5-m vs. 8-m), and two (cf. 9-m vs. 10-m), respectively].
Table 1.
Individual and mean Δδ values of menthyl esters 1-m–11-m derived from CDAs 1–11.
 |
MPA5 1-m |
MPP6 2-m |
MTPA7 3-m |
MN(1)A8 4-m |
MN(1)P5 5-m |
MTN(1)A 6-m |
MN(2)A5 7-m |
MN(2)P6 8-m |
MA(9)A8,9 9-m |
MA(2)A9 10-m |
MPhen(9)P6,1 11-m |
|---|---|---|---|---|---|---|---|---|---|---|---|
| proton # | ΔδSR (= δS – δR) | ||||||||||
| 1 | +0.09 | +0.05 | +0.02 | +0.15 | +0.16 | -0.19 | +0.16 | +0.06 | +0.21 | +0.08 | +0.14 |
| 2 | +0.11 | +0.06 | -0.03 | – | +0.26 | -0.30 | +0.14 | +0.06 | +0.42 | +0.14 | +0.28 |
| 3eq | +0.07 | +0.01 | -0.02 | – | +0.14 | -0.20 | +0.10 | +0.01 | +0.23/+0.26 | +0.10/+0.13 | +0.16 |
| 3ax | +0.07 | +0.02 | -0.02 | – | +0.12 | -0.14 | +0.10 | +0.02 | +0.23/+0.26 | +0.10/+0.13 | +0.13 |
| 4eq | 0.00 | +0.01 | 0.00 | – | +0.14 | +0.02 | 0.00/-0.01 | 0.00 | 0.00/-0.10 | 0.00/-0.02 | 0.00 |
| 4ax | +0.01 | -0.03 | +0.01 | – | 0.00 | 0.00 | 0.00/-0.01 | -0.01 | 0.00/-0.10 | 0.00/-0.02 | -0.03 |
| 5 | -0.04 | -0.01 | +0.02 | – | -0.03 | +0.03 | -0.03 | 0.00 | -0.04 | -0.05 | -0.04 |
| 6eq | -0.22 | +0.01 | +0.05 | – | -0.28 | +0.28 | -0.15/-0.23 | +0.02 | -0.50/-0.51 | -0.18/-0.23 | -0.30 |
| 6ax | -0.14 | +0.05 | +0.14 | – | -0.31 | +0.34 | -0.15/-023 | -0.04 | -0.50/-0.51 | -0.18/-0.23 | -0.37 |
| 7 | +0.56 | +0.07 | -0.32 | +0.93 | +1.21 | -1.48 | +0.64 | +0.09 | +1.8 | +0.58 | +1.20 |
| 8/9 | +0.26 | +0.02 | -0.10 | +0.51 | +0.55 | -0.50 | +0.35 | +0.04 | +0.79 | +0.36 | +0.52 |
| 9/8 | +0.22 | -0.02 | -0.13 | +0.46 | +0.46 | -0.67 | +0.36 | +0.05 | +0.85 | +0.35 | +0.57 |
| 10 | -0.06 | -0.01 | +0.03 | -0.06 | -0.10 | +0.18 | -0.08 | 0.00 | -0.19 | -0.10 | -0.12 |
| mean |Δσ| | 0.14 | 0.03 | 0.07 | 0.42 | 0.29 | 0.33 | 0.16 | 0.03 | 0.39 | 0.16 | 0.30 |
Having earlier noticed some surprising long-range effects in several Mosher ester derivatives, we decided to investigate the ability of three CDAs [the commonly used MPA (1) and MTPA (3), as well as MTN(1)A (6)] to differentially shield distant protons. Specifically, we have examined the resonances of the terminal methyl groups in the 1H NMR spectra of the series of homologous derivatives 13, derived from the symmetrical secondary carbinol precursors 12 (Figure 2). These methyls, which are enantiotopic (pro-R and pro-S) in 12, become diastereotopic in 13. Notice that for this study, it makes no difference whether or not the CDAs used were racemic.11
Figure 2.
Trends in Δδ values of the terminal methyl groups of the CDA esters 13 derived from symmetric carbinols 12.
There are a number of noteworthy features seen in the data for these series of compounds. The magnitude of the Δδ values is greatest for MTN(1)A and least for MTPA. For each CDA series, the maximum Δδ value is observed when n = 1 (i.e., 13CDA-1 or the 3-pentanol derivative). There is evidence of an “even-odd” effect12 in the step function (or sawtooth) nature within each of the three series. At values of n = 6–8 (i.e., chain lengths of 15–19), the absolute value of Δδ reaches a minimum but, surprisingly, reemerges; that is, the curves show a double maximium. We cannot help but be reminded of the similar, textbook trends for rates of cyclization vs. ring size for α,ω-difunctional substrates,13 wherein transannular effects, maximal at medium ring sizes, mitigate against the reactant residing in the reactive conformation (i.e., with its termini in close proximity). This situation is mitigated as the chain length is further increased.
An additional feature of the data intrigued us. At first glance there is no obvious reason for the crossing of the blue (for 13MTN(1)A-n) with the orange and green curves (for 13MPA-n and 13MTPA-n, respectively) (see Figure 2 inset). A different way to state this is that there is a discontinuity in the smoothness of the curves, most strongly evident in the 13MTN(1)A-n data set (blue). However, recall that the data are plotted as the absolute value of Δδ. This consideration led us to hypothesize that the relative deshielding effect of the pro-R vs. pro-S methyl groups reverses for each of the series (a crossover), causing the sign of Δδ to change for each series. This was testable, assuming we could access (i) non-racemic MTN(1)A-OH (6-OH, Figure 1) and (ii) an enantiomerically enriched and strategically deuterated analog of one of the larger (n ≥ 7) carbinol precursors 12n.
We elected to prepare a non-racemic sample of partially deuterated 10-nonadecanol 16 (the precursor to 13MTN(1)A-8) via the route summarized in Scheme 1. Racemic alcohol (±)-14 was resolved via PDC oxidation and asymmetric Noyori reduction14 to give back (S)-14 having 95% ee. Alkyne isomerization to the terminal alkyne 15 was followed by deuteration to provide the d4-alcohol 16.
Scheme 1.
Synthesis of (S)-1,1,2,2-d4-10-nonadecanol [(S)-16].
Acids (R)-6-OH and (S)-6-OH (Scheme 2) are the non-racemic versions of MTN(1)A acid. These were prepared from alkenol (±)-17, which was O-methylated and oxidatively cleaved to give the primary alcohol (±)-19. Partial kinetic resolution was achieved with a lipase induced acetylation. Although the levels of enantiopurity of the derived samples of (R)-19 [via acetate (R)-20] and (S)-19 were marginal [86% ee and 28% ee, respectively (from MTPA analysis)], they were sufficiently high to serve our purposes. Oxidation of each gave the carboxylic acids (R)-6-OH and (S)-6-OH. Alternatively, we prepared racemic acid 6-OH (by the method of Bourissou:15 1-NphthMgBr + CF3COCO2Et; K2CO3, MeI; KOH, EtOH), derivatized it as the diasteromeric menthyl esters 6-m, and separated these by MPLC (silica gel) to provide (R)-6-m (>99.8% de) and (S)-6-m (94% de), which were used to collect the data summarized in Table 1.
Scheme 2.
Enantioselective synthesis of MTN(1)A acid.a
The preparation of the diastereomeric MTN(1)A esters (S,R)- and (S,S)-21, each derived from non-racemic acid chlorides prepared in situ and derived from non-racemic samples of acids (R)-6-OH or (S)-6-OH, is outlined in Scheme 3. With these two esters in hand, we were in position to evaluate the hypothesis laid out above.
Scheme 3.
Synthesis of diastereomeric C-19 carbinyl MTN(1)A esters (S,R)-21 and (S,S)-21.a
The proton NMR data for the terminal methyl regions of (S,R)-21 and (S,S)-21, each having a diastereomeric purity reflective of the level of enantiopurity of its precursors, are shown in Figure 3. For comparison, the same spectral region of the non-deuterated analog of 21 (i.e., 13MTN(1)A-8, blue data point, n = 8, Figure 2) is also shown. From the method of synthesis we know that the chain with the deuterated terminus occupies the pro-S position in the structures in Scheme 3. From the spectral data in Figure 3 we learn that the non-depleted methyl resonance in the unlike diasteromer [(S,R)-21] is further upfield and that in the like diastereomer [(S,S)-21] further downfield. This confirms that indeed there is a crossover–a change in the sign of Δδ for 13MTN(1)A-8.
Figure 3.
Methyl resonances in the 1H NMR spectra (CDCl3, 500 MHz) for selected members of the MTN(1)A ester series.
In conclusion, trends in the relative discriminating power of a wide variety of Mosher-like esters were identified by analysis of the data in Table 1. Three homologous series of esters 13CDA-n (derived from the symmetrical carbinols 12n) were used to establish that the anistropic effects extend over remarkably long molecular distances. An interesting crossover effect was uncovered.
Supplementary Material
Acknowledgment
These studies were supported by the National Institute of General Medical Sciences (GM-65597) and the National Cancer Institute (CA-76497) of the United States National Institutes of Health.
Footnotes
Supporting Information Available: Detailed experimental procedures and spectroscopic characterization data for all new compounds. This material is available free of charge via the Internet at http://pubs.acs.org.
References
- 1.a Dale JA, Mosher HS. J. Am. Chem. Soc. 1973;95:512–519. [Google Scholar]; b Ohtani I, Kusumi T, Kashman Y, Kakisawa H. J. Am. Chem. Soc. 1991;113:4092–4096. [Google Scholar]
- 2.Wenzel Thomas J. Discrimination of Chiral Compounds Using NMR Spectroscopy. J. Wiley & Sons, Inc.; Hoboken, NJ: 2007. [Google Scholar]
- 3.Seco MJ, Quiñoá E, Riguera R. Chem. Rev. 2004;104:17–117. [Google Scholar]
- 4.Seco JM, Latypov Sh., Quiñoá E, Riguera R. Tetrahedron Lett. 1994;35:2921–2924. [Google Scholar]
- 5.Harada N, Watanabe M, Kuwahara S, Kasai Y, Ichikawa A. Tetrahedron: Asymmetry. 2000;11:1249–1253. [Google Scholar]
- 6.Kasai Y, Sugio A, Kuwahara S, Matsumoto T, Watanabe M, Ichikawa A, Harada N. Eur. J. Org. Chem. 2007;11:1811–1826. [Google Scholar]
- 7.Hoye TR, Jeffrey CS, Shao F. Nature Protocols. 2007;2:2451–2458. doi: 10.1038/nprot.2007.354. [DOI] [PubMed] [Google Scholar]
- 8.Latypov Sh., Seco JM, Quiñoá E, Riguera R. J. Org. Chem. 1995;60:504–515. [Google Scholar]
- 9.Kouda K, Kusumi T, Ping X, Kan Y, Hashimoto T, Asakawa Y. Tetrahedron Lett. 1996;37:4541–4544. [Google Scholar]
- 10.Ichikawa A, Ono. H, Harada N. Tetrahedron: Asymmetry. 2003;14:1593–1597. [Google Scholar]
- 11.For development of another unorthodox (“shortcut”) Mosher ester analysis that was used on “nearly symmetrical” carbinols see: Curran DP, Sui B. J. Am. Chem. Soc. 2009;131:5411–5413. doi: 10.1021/ja900849f.
- 12.a Baeyer A. Ber. Chem. Ges. 1877;10:1286–1288. [Google Scholar]; b von Sydow E. Ark. Kemi. 1955;9:231–254. [Google Scholar]; c Breusch FL. Fortschr. Chem. Forsch. 1969;12:119–184. [Google Scholar]
- 13.E.g., Smith MB, March J. March's Advanced Organic Chemistry. 5th Ed. J. Wiley & Sons, Inc.; New York, NY.: 2001. pp. 281–284. [Google Scholar]
- 14.Haack K-J, Hashiguchi S, Fujii A, Ikariya T, Noyori R. Angew. Chem. Int. Ed. Engl. 1997;36:285–288. [Google Scholar]
- 15.du Boullay OT, Alba A, Oukhatar F, Martin-Vaca B, Bourissou D. Org. Lett. 2008;10:4669–4672. doi: 10.1021/ol801930m. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.