Original Papers
gree to which the measurement of 13C/12C ratios in several alka-
loids differs with origin. We also highlight how labeling regula-
tions may potentially lead to ambiguities in the designation of
“natural” and “synthetic” for a given compound.
Results and Discussion
Nicotine 1 (▶ Fig. 1) is the dominant alkaloid in the leaves of a
number of Nicotiana species (Solanaceae), including smoking to-
bacco, Nicotiana tabacum L. The positional 13C/12C ratios for four
samples of nicotine from different sources (see ▶ Table 1) were
determined using irm-13C NMR spectrometry under quantitative
conditions. The profiles obtained are presented in ▶ Fig. 2. Nic-3
is an authentic sample of (S)-nicotine guaranteed extracted from
plants of Nicotiana. Nic-1 is designated by the supplier (Sigma-Al-
drich) as “synthetic” and “manufactured from materials of plant
origin” on the Certificate of Origin, while Nic-2 (Fluka) is desig-
nated “synthetic” and it is stated that “only synthetic materials
were used in the manufacturing process.” Nonetheless, Nic-1,
Nic-2, and Nic-3 show very similar profiles, indistinguishable in a
number of positions. Only the C-2 position of Nic-2 differs signifi-
cantly: at all other positions, the maximum difference is 3.8‰.
Notably, Nic-1 and Nic-2 are especially similar. Considering the
similarity of the three profiles of these samples, it is difficult to
conceive that their origins differ so much. All three samples show
two key features that are also found in nicotine of natural origin,
Nic-3 [6]. First, the C-2′ position is richer than the C-5′ position by
a few ‰ (▶ Fig. 2A), a differentiation suggested as being intro-
duced by the isotope effect associated with the nicotine synthase
enzyme [6]. Second, the C-6 is very impoverished: again, a feature
that can be explained by the biosynthetic route, as seen in the nat-
ural sample [6].
▶ Fig. 1 Structures of the alkaloids discussed in this paper: nicotine
(1); atropine (2); tropine (3); caffeine (4).
This is because the isotope distribution pattern in a target
compound is determined by the isotopic fractionation associated
with steps in the (bio)synthesis of the compound [3]. This intra-
molecular variation can be due to a combination of factors, in-
cluding flux variation and competition for precursors. These as-
pects of isotope fractionation are developed in [4]. The most ef-
fective method for accessing this intramolecular pattern is NMR
2
spectrometry, using which both H/1H and 13C/12C ratios can be
obtained [5]. Monitoring 2H/1H ratios with isotope ratio monitor-
ing 2H NMR (irm-2H NMR) has been successfully exploited for
some time, mainly in the food industry, although a smaller num-
ber of applications to pharmaceuticals has been made. The major
2
drawbacks to H monitoring are that the nucleus has poor NMR
properties with a relatively limited dispersion in the frequency
range and the isotope occurs at only low natural abundance
(0.017%). In addition, some H-positions are exchangeable, lead-
ing to modifications that might depend on, for example, the ex-
traction and work-up conditions. As an alternative, monitoring
13C/12C ratios using isotope ratio monitoring 13C NMR (irm-13
C
NMR) has been developed [2], with the advantages that the iso-
tope has a greater dispersion in the frequency range and occurs
at much higher abundance. In addition, provided certain condi-
tions are respected, the positional 13C/12C ratios should not be
modified during purification.
This similarity is emphasized when these samples of pure (S)-
nicotine are compared with Nic-4, a pure synthetic sample of (R,
S)-nicotine made in the laboratory using the method of Felpin et
al. [11]. While the profile of (R,S)-nicotine will clearly vary depend-
ing on the exact source of the precursors used, the difference in
its profile is very evident, with a variation of δ13Ci values relative
to the other samples of up to 25‰. A particularly notable differ-
Irm-13C NMR has now been developed successfully to analyze a
number of alkaloids, including nicotine 1, tropine 3, caffeine 4
(▶ Fig. 1), and tramadol. For the solanaceous alkaloids, nicotine
and tropine, it has proved possible to link the natural 13C/12C ratio
patterns to the biosynthetic pathways [6]. Furthermore, for the
recently discovered natural compound racemic (1R,2R)-2-[(dime-
thylamino)methyl]-1-(3-methoxyphenyl)cyclohexanol, known by
the trivial name tramadol, it has been possible to make a series of
predictions as to its probable biosynthetic precursors based on
the observed 13C/12C ratio in the natural product [7]. Within the
pharmaceutical context, it has been clearly demonstrated that
positional 13C/12C ratios can distinguish between different syn-
thetic origins for ibuprofen [8], aspirin, and paracetamol [9]. How-
ever, two aspects remain underexplored. The first is the relation-
ship during synthesis between the primary products, the syn-
thetic process, and the final drug. The second is the degree to
which natural compounds might be distinguished from those of
synthetic origin on the basis of their position-specific 13C/12C ra-
tios. Within this latter context, caffeine of synthetic origin has
been shown to differ considerably from that of natural origin
[10]. In this paper, we extend this approach to investigate the de-
13
13
ence is that in Nic-4 δ
C
C‑5
> δ
C
C‑4
, in complete contrast to the
samples of natural origin.
A straightforward explanation of these observations is that the
samples Nic-1 and Nic-2 are of natural origin but have undergone
processing in such a way that they can no longer be labeled as
“natural.” The designation of Nic-1 as “manufactured from mate-
rials of plant origin” would fit with the suggestion that is an exam-
ple of a case where the supplier is obliged to label a compound
“synthetic” because the process used in its manufacture involved
one or more chemical treatments so it cannot be labeled “natu-
ral”. In this case, it can be reasonably assumed that the compound
is, to all intents and purposes, extracted from a plant. The case of
Nic-2 is more problematic, as the labeling specifically states that
“only synthetic materials were used in the manufacturing pro-
cess.” Nonetheless, its profile is very similar to those of Nic-1 and
Nic-3. This problem of the assignation of origin remains to be re-
solved. It should be noted, however, that it is commercially im-
practical to produce (S)-nicotine by laboratory synthesis (see [11]
Robins RJ et al. Difficulties in Differentiating… Planta Med