1624
E. Caytan et al. / Tetrahedron: Asymmetry 17 (2006) 1622–1624
HPLC experiments.6,7 The resolution of the above Eqs. 1–
3 led to
References
1. Hays, P. A.; Remaud, G. S.; Jamin, E.; Martin, Y.-L.
J. Forensic Sci. 2000, 45, 552–562.
2. Robins, R. J.; Billault, I.; Duan, J.-R.; Guiet, S.; Pionnier, S.;
Zhang, B.-L. Phytochem. Rev. 2003, 2, 87–102.
3. Remaud, G. S.; Martin, Y.-L. G.; Martin, G.; Martin, G. J.
J. Agric. Food Chem. 1997, 45, 859–866.
2
2
A1 ¼ ð½Sꢀ ꢁ Ac ꢂ ½Rꢀ ꢁ AaÞ=ð½Sꢀ ꢂ ½Rꢀ Þ
ð4Þ
ð5Þ
2
2
A3 ¼ ð½Rꢀ ꢁ Ac ꢂ ½Sꢀ ꢁ AaÞ=ð½Rꢀ ꢂ ½Sꢀ Þ
Since A2 corresponded to the central carbon, no correction
was needed: the origin of this carbon was the same in
R-MBG and S-MBG. The isotopic deviation d& of each
carbon could then be calculated from the global 13C con-
tent measured by IRMS (Isotope Ratio, Mass Spectro-
metry) on extracted glycerol, as has already been proposed
for glycerol5 and vanillin.12 The glycerol samples studied
included several commercial origins (pure chemicals, vege-
table oils, animal fats) and authentic wines. In Table 1, the
results from the three types of experiment are displayed: (i)
the repeatability of the 13C NMR measurement alone was
assessed by five replicates over 3 months on the same MBG
(sample 3). The standard deviation found was of the same
magnitude as for the application of 13C-SNIF NMR on
other molecules.13 (ii) The repeatability of the whole
approach, from the glycerol to the 13C NMR analysis on
MBG via the synthesis and purification of MBG, was good
enough for a routine application, as demonstrated by the
three replicates of the methodology on the same glycerol
(sample 2). (iii) The isotopic deviation observed for the
three carbons of glycerol was very different for samples
of a given origin and also between samples of different ori-
gins. This observation is in agreement with the previous
work, which has shown that, either by NMR5,14 or by
IRMS,11 in biogenic glycerol carbon 2 has a higher 13C
content than carbons 1 and 3. In synthetic glycerol (sample
1), the 13C distribution was more homogeneous. The
instrumental precision of the chiral HPLC allowing the
determination of the relative percentage of (R)-MBG
([R]) was described by a relative standard deviation of
0.5%. Such small variations in [R] measurements had very
little influence on the calculated isotopic abundance, as can
be tested using Eqs. 4 and 5.
4. Tenailleau, E. J.; Lancelin, P.; Robins, R. J.; Akoka, S. J.
Agric. Food Chem. 2004, 52, 7782–7787.
5. Zhang, B.-L.; Buddrus, S.; Trierweiler, M.; Martin, G. J. J.
Agric. Food Chem. 1998, 46, 1373–1380.
6. Kato, Y.; Fujiwara, I.; Asano, Y. J. Mol. Catal. B: Enzym.
2000, 9, 193–200.
7. Xu, J.-H.; Kato, Y.; Asano, Y. Biotechnol. Bioeng. 2001, 73,
493–499.
8. Collet, A.; Brienne, M.-J.; Jacques, J. Chem. Rev. 1980, 80,
215–230.
´
9. Rabiller, C.; Maze, F.; Mabon, F.; Martin, G. J. Analusis
1991, 19, 18–22.
10. Akoka, S; Tenailleau, E. J. Presented in part at the 46th
ENC, Providence, Rhode Island, USA, April 2005.
11. Weber, D.; Kexel, H.; Schimdt, H.-L. J. Agric. Food Chem.
1997, 45, 2042–2046.
12. Tenailleau, E. J.; Lancelin, P.; Robins, R. J.; Akoka, S. Anal.
Chem. 2004, 76, 3818–3825.
13. Caytan, E.; Remaud, G. S.; Tenailleau, E. J.; Akoka, S.
Talanta, in press.
14. Zhang, B.-L.; Trierweiler, M.; Jouitteau, C.; Martin, G. J.
Anal. Chem. 1999, 71, 2301–2306.
15. Gleixner, G.; Schmidt, H.-L. J. Biol. Chem. 1997, 272, 5382–
5387.
16. Chemicals: Typically, 1 g of glycerol was dissolved in 110 mL
of 1,4-dioxane (VWR, France) with 3.7 g of benzoic anhy-
dride (Alfa Aesar, France) and 0.9 g of carrier-fixed Candida
antarctica Novozym 435 (Sigma, France). The mixture was
stirred until there was complete consumption of glycerol,
monitored by GC. After filtration (for enzyme recovery), the
reaction mixture was evaporated under reduced pressure. The
residue was dissolved in ethyl acetate (VWR, France)/
n-hexane (VWR, France) (50/50) and loaded onto a silica gel
(VWR, France) column. Dibenzoyl glycerol, benzoic acid and
the remaining benzoic anhydride were eluted first by ethyl
acetate/n-hexane (50/50). Then, MBG was obtained with
pure ethyl acetate. To complete elution, methanol (VWR,
France) may be added when monitored by GC. After
evaporation to dryness, the refined MBG was dissolved in
the minimum of boiling 2-propanol (VWR, France)/n-hexane
(5/95). One crystal of (R)-MBG was added to initiate
preferential crystallisation. The mixture was further chilled
overnight at ꢂ15 to ꢂ20 °C. The white crystals of MBG were
filtered and finally dried in an oven (40 °C max).
17. NMR experiments: Samples were prepared by dissolving
400 mg of purified MBG in 800 lL of acetone-D6 (Euriso-
top, France). Quantitative 13C NMR spectra were recorded
using a Bruker DRX 500 spectrometer fitted with a 5-mm-i.d.
dual probe 13C/1H carefully tuned at the recording frequency
of 125.76 MHz. The temperature of the probe was set at
303 K. The experimental parameters for 13C NMR spectral
acquisition were the following: pulse width 7.8 ls (90°),
spectral width 30,000 Hz, sampling period 1 s, repetition
delay 17 s, number of scans 400. Inverse-gated decoupling
techniques were applied in order to avoid NOE. The
decoupling sequence employed a cosine adiabatic pulse with
appropriate phase cycles, as described in Ref. 10. Each
sample was measured four times.
3. Conclusion
Work is currently in progress to delineate the significance of
such an approach to study the isotope profile of glycerol
wherever it is found. The present preliminary results indi-
cate that large isotopic fractionations occur during the
metabolism of glycerol, either in the fatty acid pathway or
in the bio-transformation from glucose. The huge depletion
of site 1 has been explained by both (i) an isotope balance
where depletion is very large when a by-product, such as
glycerol, is concerned and (ii) the kinetic isotope effect on
the aldolase reaction,15 supporting the origin of glycerol
from the upper part of glucose.11 By increasing the data
on authentic glycerol samples, the origin of glycerol should
be well characterised. The present methodology was easy to
use and is robust. A similar strategy could be applied to
other molecules, such as citric or tartaric acids, and to other
isotopic determinations, such as site-specific 2H/1H ratios.