3.4 Structure determination of γ-T1 by 1H NMR spectroscopy and GC/MS data
The bulk of isolated γ-T1 (5.4 mg) was analyzed by 1H NMR analysis (section 2.7). In the downfield range, the singlet at 6.30 ppm confirmed the presence of one proton on C-5 in the aromatic part of the 6-chromanol ring (Fig. 9a ). This chemical shift was slightly upfield compared to 6.40 ppm in the 1H NMR spectrum of γ-T3 which was measured under the same conditions for comparison. This is in accordance with literature data for a proton on C-5 at 6.38 ppm (γ-T3) (Ohnmacht et al., 2008) and 6.37 ppm (γ-T), respectively (Baker and Myers, 1991). Compared to that, a proton on C-7 (β-T3) was expected more downfield at 6.48 ppm (Ohnmacht et al., 2008) or at 6.46 (β-T), respectively (Baker and Myers, 1991). This confirmed our assignments by GC/MS via (the silylated) M+ atm/z 486 and m/z 223 > m/z 222 (section 3.2).
Furthermore, only one signal (triplets of triplet, J = 7.13 Hz, 1.25 Hz) was detected in the range of olefinic protons at 5.033 ppm. Accordingly, the second sp2 hybridized carbon did not carry a proton (which is true when located on a branching point). Also, large coupling indicated a regular 3J -coupling between C-10´and C-11´ and a long-range coupling between C-9´and C-11´. Such a scenario can only be found when the double bond is located in either C-3´-, C-4´-, C-7´-, C-8´- or C-11´-position, respectively. Assuming a natural substitution pattern, the double bond would be expected in C-11´-position while exclusive presence in C-7´-position was rather unlikely. However, the presence on C-3´-position could not be fully excluded at this point. In the case of α-T2, db positions could be distinguished by means of the signal of C-3 at ~2.6 ppm for no double bond in C-3´ and ~2.7 ppm for one db on C-3´ (Müller et al., 2020). Contrary to (expected) triplets as in the case of α-tocochromanols, the multiplet at ~2.6 ppm looked more like a quartet in γ-tocochromanols, although the integral (I = 2) was the same as expected. The fact that the same multiplet was also observed in the 1H NMR spectrum of γ-T3 (and γ-T) produced evidence that both protons on C-3 of γ-tocochromanols were not isochronous. Non-constant distances between the individual signals confirmed this difference between α- and γ-tocochromanols. Most importantly, however, compared to γ-T3, the mulitplet of γ-T1 was shifted highfield by ~0.1 ppm (Fig . 9cand 9d ) which verified the absence of a db in C-3´-position (Müller et al., 2020). This was further supported by a singlet at ~1.62 ppm originating from the methyl group in C-13´ position. For α-tocochromanols, Müller et al. (2020) observed a singlet at ~1.70 ppm when it is close to a double bond in C-11´-position. In case of γ-T1, this singlet was slightly shifted upfield but the ratio of the integrals of 2:3 compared to the multiplet at 2.70 ppm agreed well with literature data. Therefore, the db was confirmed to be on C-11´-position. This is in agreement with GC/MS data of silylated γ-T1 which featured the diagnostic fragment ion atm/z 69. Hitherto, m/z 69 was detected in all unsaturated tocochromanols detected in plants (tocomonoenols, tocodienols and tocotrienols) but not in tocopherols and marine-derived tocopherols with the db on C-12´. Hence, the terminal double bond seems to be obligate in unsaturated plant tocochromanols.