Determination of free and glucosidically-bound volatiles in plants. Two case studies: L-menthol in peppermint (Mentha x piperita L.) and eugenol in clove (Syzygium aromaticum (L.) Merr. & L.M.Perry)

Article abstract of DOI:10.1016/j.phytochem.2015.06.017

Abstract This study arises from both the today's trend towards exploiting plant resources exhaustively, and the wide quantitative discrepancy between the amounts of commercially-valuable markers in aromatic plants and those recovered from the related essential oil. The study addresses the determination of both the qualitative composition and the exhaustive distribution of free and glucosidically-bound L-menthol in peppermint aerial parts (Mentha x piperita L., Lamiaceae) and of eugenol in dried cloves (Syzygium aromaticum (L.) Merr. & L.M.Perry, Myrtaceae), two plants known to provide widely ranging essential oil yields. The two markers were investigated in essential oils and residual hydrodistillation waters, before and after enzymatic hydrolysis. Their amounts were related to those in the headspace taken as reference. The results showed that the difference between marker compound in headspace and in essential oil amounted to 22.8% for L-menthol in peppermint, and 16.5% for eugenol in cloves. The aglycones solubilised in the residual hydrodistillation waters were 7.2% of the headspace reference amount for L-menthol, and 13.3% for eugenol, respectively representing 9.3% and 15.9% of their amounts in the essential oil. The amount of L-menthol from its glucoside in residual hydrodistillation waters was 20.6% of that in the related essential oil, while eugenol from its glucoside accounted for 7.7% of the amount in clove essential oil. The yield of L-menthol, after submitting the plant material to enzymatic hydrolysis before hydrodistillation, increased by 23.1%, and for eugenol the increase was 8.1%, compared to the amount in the respective conventional essential oils. This study also aimed to evaluate the reliability of recently-introduced techniques that are little applied, if at all, in this field. The simultaneous use of high-concentration-capacity sample preparation techniques (SBSE, and HS-SPME and in-solution SPME) to run quali-quantitative analysis without sample manipulation, and direct LC-MS glucoside analysis, provided cross-validation of the results.

Full text of DOI:10.1016/j.phytochem.2015.06.017

Phytochemistry 117 (2015) 296–305
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Determination of free and glucosidically-bound volatiles in plants. Two
case studies: L-menthol in peppermint (Mentha x piperita L.) and eugenol
in clove (Syzygium aromaticum (L.) Merr. & L.M.Perry)
Barbara Sgorbini a, Cecilia Cagliero a, Alberto Pagani b, Marla Sganzerla a,1, Lorenzo Boggia a,
Carlo Bicchi a, , Patrizia Rubiolo a
⇑
a Dipartimento di Scienza e Tecnologia del Farmaco, Università degli Studi di Torino, Via Pietro Giuria 9, I-10125 Torino, Italy
b Dipartimento di Scienze del Farmaco, Università degli Studi del Piemonte Orientale ‘‘Amedeo Avogadro’’, Largo Donegani 2/3, I-28100, Novara, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
This study arises from both the today’s trend towards exploiting plant resources exhaustively, and the
wide quantitative discrepancy between the amounts of commercially-valuable markers in aromatic
plants and those recovered from the related essential oil. The study addresses the determination of both
the qualitative composition and the exhaustive distribution of free and glucosidically-bound L-menthol
in peppermint aerial parts (Mentha x piperita L., Lamiaceae) and of eugenol in dried cloves (Syzygium
aromaticum (L.) Merr. & L.M.Perry, Myrtaceae), two plants known to provide widely ranging essential
oil yields. The two markers were investigated in essential oils and residual hydrodistillation waters, before
and after enzymatic hydrolysis. Their amounts were related to those in the headspace taken as reference.
The results showed that the difference between marker compound in headspace and in essential oil
amounted to 22.8% for L-menthol in peppermint, and 16.5% for eugenol in cloves. The aglycones solubilised
in the residual hydrodistillation waters were 7.2% of the headspace reference amount for L-menthol, and
13.3% for eugenol, respectively representing 9.3% and 15.9% of their amounts in the essential oil. The amount
of L-menthol from its glucoside in residual hydrodistillation waters was 20.6% of that in the related essential
oil, while eugenol from its glucoside accounted for 7.7% of the amount in clove essential oil. The yield of
L-menthol, after submitting the plant material to enzymatic hydrolysis before hydrodistillation, increased
by 23.1%, and for eugenol the increase was 8.1%, compared to the amount in the respective conventional
essential oils.
Received 6 March 2015
Received in revised form 12 May 2015
Accepted 14 June 2015
Keywords:
Mentha x piperita L. (Lamiaceae)
Syzygium aromaticum (L.) Merr. & L.M.Perry
(Myrtaceae)
Essential oils
Residual distillation waters
Monoterpene glucosides
Enzymatic hydrolysis
L-menthol
Eugenol
Quantitative determination
This study also aimed to evaluate the reliability of recently-introduced techniques that are little applied, if
at all, in this field. The simultaneous use of high-concentration-capacity sample preparation techniques
(SBSE, and HS-SPME and in-solution SPME) to run quali-quantitative analysis without sample manipulation,
and direct LC–MS glucoside analysis, provided cross-validation of the results.
Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction
species belonging to 50 families (Crouzet and Chassagne, 1999;
Stahl-Biskup et al. 1993; Winterhalter and Skouroumounis,
Volatile compounds are often present in plants as glycosidi-
cally-bound components. These compounds were first described
by Bourquelot and Bridel (1913) who identified a geranyl-b-D-
glucoside in Pelargonium odoratissimum. To date, glycosidically-
bound volatiles have been found to occur in almost 170
1997) not only in the aerial parts but also in roots, rhizomes,
petals, fruits and seeds. Most of them are O-glycosides whose
aglycone moieties belong to different classes of metabolites
(mainly phenols and terpenoids). Conversely, the commonest gly-
cone in nature is b-D-glucose, directly bound to the aglycone. The
glucose moiety may also be replaced by one or more additional
sugar units (Winterhalter and Skouroumounis, 1997). The large
number of b-D-glucose derivatives, and the roles they play in
plants, were the background for the development of a rapid and
effective method, based on enzymatic hydrolysis by b-glucosi-
dase, to determine the content of these compounds in several
matrices of vegetable origin, including wines, fruits, and other
⇑
Corresponding author.
E-mail addresses: barbara.sgorbini@unito.it (B. Sgorbini), cecilia.cagliero@unito.
it (C. Cagliero), lorenzo.boggia@unito.it (L. Boggia), carlo.bicchi@unito.it (C. Bicchi),
patrizia.rubiolo@unito.it (P. Rubiolo).
1
On leave from Laboratory of Food Analysis, Department of Food Science, Faculty
of Food Engineering, University of Campinas (UNICAMP), P.O. Box 6121, CEP 13083-
862 Campinas, SP, Brazil.
http://dx.doi.org/10.1016/j.phytochem.2015.06.017
0031-9422/Ó 2015 Elsevier Ltd. All rights reserved.

B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
297
List of acronyms
HD
CEO
conventional hydrodistillation
clove essential oil
Micro-HD micro-hydrodistillation
MHS-SPME multiple headspace solid phase micro-extraction
PEO peppermint essential oil
PEO-R-EH peppermint EO from the distillation of the waters
residual after enzymatic hydrolysis
CEO-R-EH clove EO from the distillation of the waters residual
after enzymatic hydrolysis
CR
clove residual distillation water
EH-EO
EO
HS
essential oil obtained after plant enzymatic hydrolysis
essential oil
headspace
PR
R(s)
SBSE
peppermint residual distillation water
residual hydrodistillation water(s)
stir bar sorptive extraction
HS-SPME headspace solid phase microextraction
substrates
Ananthakumar et al., 2006).
(Winterhalter
and
Skouroumounis,
1997;
when quantitated in essential oils and via headspace solid phase
microextraction (HS-SPME) was observed (Sgorbini et al., 2015).
The aim of this study was therefore (i) to investigate exhaus-
tively the volatile composition of the aerial parts of peppermint
(Mentha x piperita L., Lamiaceae) and dried cloves (Syzygium
aromaticum (L.) Merr. & L.M.Perry, Myrtaceae) (ii) to evaluate the
distribution of eugenol and L-menthol in free form, in the essential
oils and in the water phase resulting from hydrodistillation (resid-
ual distillation water – R) and to compare these distributions to the
‘‘nominal’’ total amounts in the original matrices, measured by
multiple headspace extraction-SPME-GC-FID/MS, and (iii) to
analyse the two compounds in their glucosidically-bound forms.
L-menthol and eugenol were selected as model compounds
because (a) they are the most abundant markers of the plants
investigated, (b) they may be taken as representative of monoter-
pene alcohols and phenolic compounds, i.e. two of the groups of
secondary metabolites often present in glycoside form in several
species, (c) they may be taken as representative of compounds
with different water solubility, their octanol/water partition coeffi-
cient, Ko/w, being 3.38 for L-menthol and 2.73 for eugenol (Episuite,
2012), and (d) the two plant matrices are known to provide widely
differing essential oil yields.
In plant tissue, glycosylation is a common protective mecha-
nism that prevents lipophilic volatile compounds (mainly phenols
or alcohols) from damaging labile cellular components, such as cell
membranes, improves the storage of volatile metabolites within
the plant vacuoles, and protects plants from any form of toxicity
due to the aglycone (Stahl-Biskup et al., 1993). Moreover, glyco-
conjugates are considered to be important for accumulation and
storage, as well as providing transportable forms of some
hydrophobic substances involved in several processes of plant
metabolism, and, in particular, as intermediates in the formation
of secondary metabolites (Hosel, 1981). As a consequence, it is
quite common to find the glycosides of phenolic or monoterpenic
alcohols alongside the main components of the essential oil of a
plant (Stahl-Biskup et al., 1993). This tendency however cannot
be generalised because glycoside-released-aglycones may or may
not be present in the corresponding essential oil, as shown by
Den Van Dries and Baerheim Svendsen (1989) in a study on free
and glucosidically-bound volatiles in fresh needles of Juniperus
chinensis var. pfitzeriana and J. communis, in leaves of Cupres-socyparis
leylandii, and in fresh aerial parts of Mentha piperita citrata and Salvia
officinalis.
Glycoconjugates are generally water soluble, less reactive, non-
volatile and odourless compounds, and are considered as potential
natural sources of odorous volatile aglycones, releasable by enzy-
matic or chemical hydrolysis (Nirmala Menon and Narayanan,
1992) during maturation, industrial pretreatment or processing
(Stahl-Biskup et al., 1993).
The study of the free and glycosidically-bound volatile compo-
sition is useful for a correct evaluation both of the total contents of
odorous components that are potentially available from an aro-
matic plant, and of their total recovery, to be added to that conven-
tionally obtained with essential oil isolation, not least in view of
their use in the food and cosmetic fields. Moreover, residual distil-
lation waters after essential oil isolation have also recently been
tested to evaluate their effect on growth, productivity, and essen-
tial oil content and composition of peppermint (Mentha ꢀ piperita
L. ‘Black Mitcham’) plants, with positive results. These treatments
showed significant effects on plant height and weight, essential
oil content and yield, as well as essential oil composition
(Zheljazkova and Astatkieb, 2012).
In spite of this interest, relatively few publications have dealt
with the chemistry of glycosidically-bound volatiles and their dis-
tribution in plants. Most such studies have concerned quantitative
analysis of hydrolytically-liberated aglycones (Ananthakumar
et al., 2006).
During a study aimed at developing a rapid but reliable method
for quality control of spices, a marked inconsistency between men-
thol and eugenol contents in peppermint aerial parts and in cloves,
2. Results and discussion
This study comprised two main parts: the first part focused on
L-menthol and eugenol in their free forms, and the second on their
glucosidically-bound forms; each part consists of a number of
steps that are summarised in Fig. 1, in particular for L-menthol
and eugenol free forms:
(1) Headspace analysis of dried plant material by MHS-SPME-
GC–MS (HS-1) and quantitation of L-menthol and eugenol.
Headspace sampling with solid phase microextraction
(HS-SPME) makes it possible to evaluate the qualitative
composition of the volatile fraction of peppermint aerial
parts and cloves, and to quantify L-menthol and eugenol pre-
sent in free form in the original dried plant material (Arthur
and Pawliszyn 1990). Quantitation was achieved by the mul-
tiple headspace approach (MHS-SPME) (Ezquerro et al.,
2003; Bicchi et al., 2011) in consideration of its effectiveness
with solid matrices, since it is not influenced by the matrix
effect (Kolb and Ettre, 1997). HSSE was also tested (Bicchi
et al., 2000; Tienpont et al., 2000) and gave comparable
results to HS-SPME; HS-SPME was chosen as being easier
to be run automatically. Theory and use of MHS-SPME is
reported in paragraph 1SM in the supplementary material.
(2) Hydrodistillation of plant material (EO-1) and determination
of normalised % composition, and true quantitation of
L-menthol and eugenol in the resulting EOs by GC-FID and

298
B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
Plant
material
HS-1
EH+
micro-HD
HD//micro-HD
MHS-SPME-GC-MS
EH-EO
EO-1
GC-FID-MS
GC-FID-MS
R1
EH
R1-EH
SBSE-GC-MS
LC-UV-MS
SBSE-GC-MS
EO-R1-EH
GC-FID-MS
EO-2
GC-FID-MS
R2
EH
EH
R2-EH
SBSE-GC-MS
LC-UV-MS
SBSE-GC-MS
EO-R2-EH
GC-FID-MS
EO-3
GC-FID-MS
R3
R3-EH
SBSE-GC-MS
LC-MS
SBSE-GC-MS
EO-R3-EH
GC-FID-MS
Fig. 1. Scheme of free and glycosidically-bound volatile compounds analysis. EO: essential oil; HS: Headspace; HD: hydrodistillation; EH: enzymatic hydrolysis; EO-EH: EO
hydrodistilled after enzymatic hydrolysis of water-soluble glucosides; R1, (2 or 3): Residual distillation water 1 (2 or 3); R1- (2 or 3)-EH: Residual distillation water 1 (2 or 3)
submitted to enzymatic hydrolysis; EO-R1 (2 or 3)-EH: EO from R1 (or 2) after enzymatic hydrolysis.
GC–MS. After the first hydrodistillation, both samples were
submitted to repeated hydrodistillation, by re-suspending
plant materials in ‘‘new’’ freshly distilled water, to isolate
EOs exhaustively (EO-2 and EO-3). Residual hydrodistillation
waters (R1, R2 and R3) were also stored for further analysis.
(3) All residual waters from the hydrodistillations (R1, R2, R3)
were submitted to enzymatic hydrolysis of solubilised glu-
cosides, and again hydrodistilled, and the resulting EOs
(EO-R1-EH, EO-R2-EH, EO-R3-EH) analysed by GC-FID and
GC–MS to determine normalised % composition, and to
quantify L-menthol and eugenol. These results are related
to the previous steps, because they also provide an indica-
tive recovery of volatiles deriving from hydrolysis of gluco-
sides solubilised in each R.
The second part, concerning glucosidically-bound L-menthol
and eugenol, involved:
(6) Quantitation of total L-menthol and eugenol (i.e. free + glu-
cosidically-bound) in residual distillation waters after enzy-
matic hydrolysis (R1-EH, R2-EH, R3-EH) of glucosides that
was run by SBSE-GC–MS and GC-FID. Acidic hydrolysis was
abandoned because in the preliminary tests carried out on
octyl-b-glucoside taken as model compound, significant dif-
ferences in the 1-octanol abundance depending on the pH of
hydrolysis were measured, and extensive formation of arte-
facts when applied to the first peppermint residual distilla-
tion water (PR1) were detected, in agreement with
Williams et al. (1982).
(7) Quali- and quantitative results of glucoside analyses in Rs
were confirmed by analysing them directly by LC–MS.
Eugenyl glucoside was synthesised as reported in paragraph
4.5, while a 78% pure L-menthol glucoside sample was avail-
able from the collection of standards in the authors’
laboratory.
(4) Quantitation of free L-menthol and eugenol in residual dis-
tillation waters (R1, R2 and R3) was run by Stir Bar
Sorptive Extraction (SBSE) in combination with GC-FID and
GC–MS (Baltussen, 1999).
(5) Micro-hydrodistillation (micro-HD) of plant material after
maceration and enzymatic hydrolysis (EH) of water solu-
bilised glucosides (EH-EO) and determination of normalised
% composition, and quantitation of L-menthol and eugenol in
the resulting EOs by GC-FID and GC–MS. These analyses
aimed to compare yield and % composition of the marker
compounds in the samples obtained from the plant material,
as such (EO-1) and after maceration/hydrolysis (EH-EO);
they also sought to quantify total amounts in free and gluco-
sidically-bound forms and, indirectly, amounts of the related
glucosides, and to evaluate the free/glucosidically-bound
aglycone ratios using the procedures described here.
The results of the set of analyses of each plant matrix are, for
clarity, collected in two separate paragraphs (2.1 for peppermint
aerial parts and 2.2 for clove). L-Menthol and eugenol were taken
as representative markers of the investigated plant materials; the
quantitative results of this study are all expressed as ppm or mg
of L-menthol and eugenol, i.e. as absolute amounts in 100 g of plant
material or in 2 l of residual distillation waters; they are sum-
marised in Table 1a–c. Concentration range, linear regression equa-
tions, and determination coefficients for L-menthol and eugenol

B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
299
Table 1
Free and glucosidically-bound L-menthol and eugenol quantitative results obtained by GC–MS or GC-FID and LC–MS in the various steps of the present study: (a) free L-menthol
and eugenol quantitative data in HS, EO and Rs expressed as mg in 100 g of plant material; (b) glucosidically-bound L-menthol and eugenol quantitative data, after enzymatic
hydrolysis and as such, expressed as ppm and mg in 100 g of plant material; (c) summarising quantitative data. C: clove; P: peppermint aerial parts; EO: essential oil; HS-1:
Headspace of plant material; EH: enzymatic hydrolysis; EO-1: exhaustively isolated EO; EH-EO: EO isolated after enzymatic hydrolysis of water-soluble glucosides; R1 (2 or 3):
Residual distillation water 1 (2 or 3); R1- (2 or 3)-EH: Residual distillation water 1 (2 or 3) after enzymatic hydrolysis; EO-R1 (or 2)-EH: EO from R1 (or 2) after enzymatic
hydrolysis; Rs: Residual distillation waters.
Mentha x piperita
Syzygium aromaticum
L-menthol
Eugenol
1a
mg
mg
Headspace sampling
HS-SPME-GC–MS
Exhaustive EO isolation
GC–MS
PHS-1
PEO-1
516.0
CHS-1
CEO-1
7890
6588
398.2
Residual distillation waters (Rs)
SBSE-GC–MS
PR1
PR2
PR3
23.7
9.6
3.9
CR1
CR2
CR3
600
316
130
Total free
37.2
1046
1b
mg
mg
EO isolated after EH
GC-MS
EH-PEO
490.1
EH-CEO
7120
EO isolated after EH of Rs
GC-MS
PEO-R1-EH
PEO-R2-EH
n.q.
n.q.
CEO-R1-EH
CEO-R2-EH
1009
52.5
ppm
Free + bound
mg
Glucoside
mg
ppm
Free + bound
mg
Glucoside
mg
Residual distillation waters (Rs)
SBSE-GC-MS
PR1-EH
PR2-EH
PR3-EH
Total
36.8
17.0
5.8
73.6
33.9
11.6
119.1
50.0
24.3
7.7
CR1-EH
CR2-EH
CR3-EH
Total
226
115.6
46.8
904
462
188
1554
304
146
58
508
82
Glucoside
mg
Glucoside
mg
Glucoside in Rs
Direct LC–MS
PR1
PR2
PR3
Total
n.q.
n.q.
n.q.
n.q.
CR1
CR2
CR3
Total
330.8
158
64
552.8
1c
Free
mg
Bound
mg
Free
mg
Bound
mg
Headspace
516
EO
398.2
Rs
37.2
R-EH
119.1
R
82
Headspace
7890
EO
6588
Rs
1046
Rs-EH
1554
R
508
quantitation, in HS, EO and R, are reported in Table 1SM.
Repeatability data for the various approaches and methods
adopted in this study are collected in Table 2SM and always
acceptable with RSD% ranging between 0.9 to 7.6 for L-menthol
and 1.2–6.9 for eugenol.
components and the chemical composition of PEO expressed as
relative percentage abundances are included in supplementary
material (Fig. 2SM and Table 3SM). Twenty-nine components were
identified, the most abundant of them being L-menthol (51.2%), L-
menthone (17.1%) and 1,8-cineole (5.0%). The EO obtained from
10 g of plant material, submitted to micro-hydrodistillation in
the dedicated apparatus (PEO-1.1) (Bicchi et al., 1983) gave yield
and composition comparable to those of PEO-1 (Table 3SM).
These results meant that micro-hydrodistillation (micro-HD) could
be applied to all experiments involving enzymatic hydrolysis (EH),
and, as a consequence, experiments could be run with comparable
volumes resulting from consistent amounts of plant material.
Processed plant material re-submitted to two successive
hydrodistillations with ‘‘new’’ water did not produce a measurable
EO amount in either case, meaning that it had been exhaustively
isolated in the first step. Conversely, the resulting PR-2 and PR-3
were processed together with PR-1 as reported below.
A non-quantifiable amount of peppermint EO, resulting from
the distillation of the waters residual from micro-hydrodistillation
after enzymatic hydrolysis (PEO-R1-EH), was recovered with cyclo-
hexane and analysed by GC-FID/MS, giving a significant profile.
Twenty-nine compounds were identified in PEO-R1-EH, some of
them not found in PEO-1 or EH-PEO – chiefly 3-hexenol (0.09%)
– some others absent in PEO-R1-EH, although present in both
PEO-1 and EH-PEO (e.g. trans-sabinene hydrate). Conversely, ade-
quate amounts of PEO-R2-EH and PEO-R3-EH to carry out
2.1. Peppermint aerial parts
2.1.1. Determination of free L-menthol
2.1.1.1. Quantitation of L-menthol in the headspace of peppermint
aerial parts. The total area of L-menthol was estimated through
three consecutive MHS-SPME extractions of a suitable amount of
peppermint aerial parts. The resulting decay equation of L-menthol
in peppermint, linear regression coefficient average quotient Q
value are reported in Table 1SM. A calibration curve was built up
by analysing a set of L-menthol standard solutions under the same
conditions (i.e. three consecutive extractions) as reported in para-
graph 4.3.5. The resulting amount of L-menthol in peppermint aer-
ial parts, determined by MHS-SPME-GC–MS, was 516.0 mg, with
an RSD of 3.4% (PHS-1) (Table 1a).
2.1.1.2. L-Menthol in peppermint essential oil. Table 2 reports the list
and relative percentage abundances normalised versus
a-thujone
of hydroxylated components of interest for this study. The GC-
FID profile of the EO obtained by conventional hydrodistillation
(HD) from peppermint aerial parts (PEO-1) and list of all identified

300
B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
Table 2
Normalised % composition versus
a
-thujone of peppermint (PEO) and clove (CEO) essential oils after GC-FID analysis. PEO-1/CEO-1: peppermint/clove EO obtained with
conventional hydrodistillation from 100 g of leaves; PEO-1.1/CEO-1.1: peppermint/clove EO obtained by micro-hydrodistillation from 10 g of leaves; EH-PEO/EH-CEO:
peppermint/clove EO obtained from plant material after enzymatic hydrolysis of water-soluble components (micro-hydrodistillation
peppermint/clove EO obtained from residual distillation waters (Rs) after enzymatic hydrolysis (micro-HD – 200 ml of R1).
–
10 g); PEO-R1-EH/CEO-R1-EH:
Mentha ꢀ piperita
PEO-1
PEO-1.1
EH-PEO
PEO-R1-EH
Components
LRI
Av. %
RSD %
Av. %
RSD %
Av. %
RSD %
Av. %
RSD %
3-Hexenol
3-Octanol
1-Octenol
/
/
/
0.2
/
0.1
4.1
1.1
52.1
1.3
/
3.6
/
8.0
1.3
3.5
0.1
2.9
/
/
0.1
2.0
1.8
0.1
4.0
1.1
50.7
2.0
2.4
0.1
1.5
1.8
1.1
1.0
0.1
0.1
1419
1423
1549
1557
1646
1654
2081
0.2
0.1
0.1
4.3
1.2
51.2
1.3
2.7
4.7
4.0
0.3
0.9
0.05
0.7
0.8
0.8
0.1
4.2
1.2
52.5
1.2
1.6
2.7
13.7
0.7
0.8
0.04
1.2
i-Pulegol
neo-Menthol
neo-i-Menthol
L-menthol
Viridiflorol
Syzygium aromaticum
CEO-1
CEO-1.1
EH-CEO
CEO-R1-EH
Components
Eugenol
LRI
2169
Av. %
80.3
RSD %
0.006
Av. %
81.2
RSD %
0.1
Av. %
88.8
RSD %
1.3
Av. %
91.7
RSD %
0.6
Table 3
qualitative analysis could not be recovered. Table 1a–c is a synop-
sis of the free and glucosidically-bound L-menthol quantitative
results, expressed as mg of L-menthol in 100 g of plant material.
PEO-1 and PEO-1.1 were obtained in a yield of 0.75%, i.e. 750 mg
of EO, corresponding to 398.2 mg of L-menthol; this was confirmed
by GC-FID/MS (Table 1a).
Main components/menthone I.S.-normalised area ratios in the peppermint EOs
obtained under different conditions. PEO-1.2: peppermint EO from micro-hydrodis-
tillation; EH-PEO: peppermint EO from micro-hydrodistillation after hydrolysis of
plant material suspended in water with b-glucosidase.
Mentha ꢀ piperita Essential oils
Analyte ratios⁄
PEO-1.1 (mean SD)
EH-PEO (mean SD)
1,8-Cineole/menthone
3-Octanol/menthone
1-Octenol/menthone
Menthyl acetate/menthone
Neomenthol/menthone
Neoisomenthol/menthone
L-menthol/menthone
Viridiflorol/menthone
0.3437 0.0014
0.0097 0.0004
0.0047 0.0002
0.2535 0.0009
0.2480 0.0033
0.0692 0.0024
3.0056 0.0026
0.0866 0.0026
0.3548 0.0012
0.0495 0.0007
0.0531 0.0014
0.2704 0.0012
0.2636 0.0018
0.0766 0.0009
3.3014 0.0129
0.0786 0.0013
2.1.1.3. Quantitation of free L-menthol in peppermint residual distil-
lation waters (PR). L-menthol was first analysed in the residual dis-
tillation waters by SBSE-GC-FID/MS. Free L-menthol concentration
accounted for 11.8 ppm in PR-1, corresponding to 23.7 mg, while
its concentration in PR-2 was 4.8 ppm, corresponding to 9.6 mg,
and in PR-3 it was 2.0 ppm, corresponding to 3.9 mg (Table 1a).
The above results were confirmed by in-solution SPME-GC-
FID/MS applied to all PR samples (data not reported). Moreover,
L-menthol determined by HS-SPME-GC–MS, and that resulting
from the sum of its contents in the essential oil and as free form
in the residual distillation waters, were in a good agreement
(Table 1c).
additional amount of L-menthol deriving from enzymatic hydroly-
sis in combination with hydrodistillation was therefore 91.9 mg.
2.1.2.2. Quantitation of glucosidically-bound L-menthol in peppermint
residual distillation waters (PR). The aqueous extracts resulting
from enzymatic hydrolysis (EH) were analysed by SBSE-GC-
FID/MS. Table 4 reports the areas of the PR1, PR2 and PR3 main
2.1.2. Determination of glucosidically-bound L-menthol
2.1.2.1. Peppermint aerial part essential oil after enzymatic
hydrolysis. The same twenty-nine compounds as for PEO-1 and
PEO-1.1 were identified in the essential oil obtained by micro-hy-
drodistillation, after hydrolysis of the water-soluble components of
peppermint aerial parts with b-glucosidase, although, as expected,
components, normalised vs. the a-thujone internal standard, after
enzymatic hydrolysis, together with their% increase compared to
areas for aglycones. These results clearly show that enzymatic
hydrolysis markedly increases the abundance of hydroxylated
aglycones in PR1-EH, PR2-EH and PR3-EH, while keeping the
amount of carbonyl derivatives constant. Some alcohol compo-
nents were only present after enzymatic hydrolysis, e.g. 3-octanol,
the abundance normalised versus
a-thujone of some alcohols
increased, due to enzymatic hydrolysis, compared to PEO-1.1;
these included 3-octanol, 1-octenol, and L-menthol (Table 2).
The enzymatic hydrolysis effect on plant material suspended in
water, and on the residual distillation water, was evaluated
through both the abundance of components in their hydrodistilla-
tion products (Table 2), and the normalised area ratios of those
components that can exist in glucosidic form (alcohols) vs. men-
thone, i.e. a components whose abundance should not vary when
submitted to enzymatic hydrolysis (Table 3). Comparison of PEO-
1 and EH-PEO results showed that only alcohols/menthone ratios
increased because of enzymatic hydrolysis, although to different
extents because of their different amounts in the original matrix,
while 1,8-cineole/menthone and menthyl acetate/menthone ratios
were almost constant, as expected, these two compounds not
being involved in enzymatic hydrolysis. The EO yield after prelim-
inary enzymatic hydrolysis (EH-PEO) was 0.93%, i.e. 930 mg of EO,
in its turn corresponding to 490.1 mg of L-menthol (Table 1a). The
linalool,
a-terpineol, benzyl alcohol, p-menthadienol in PR1-EH,
while 3-octenol was only detected in PR2-EH. Further components
not found in the EOs were also identified, e.g. benzyl alcohol,
p-menthadienol.
Glucosidically-bound L-menthol was quantitated by applying
the same conditions as for the aglycone. Its concentration
accounted for 36.8 ppm in PR1-EH, corresponding to 73.7 mg,
while its concentration in PR2-EH was 17.0 ppm that is 33.9 mg,
and in PR3-EH it was 5.8 ppm that is 11.6 mg (Table 1b). The above
results were confirmed by in-solution-SPME-GC-FID/MS applied to
all PR-EH samples.
These analyses showed that the total L-menthol deriving from
enzymatic hydrolysis of its glucoside, in PR1, PR2 and PR3,
accounted for 81.9 mg in 100 g of plant material, which is in agree-
ment with EH-PEO (91.9 mg).

B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
301
Table 4
Areas normalised vs.
a-thujone as internal standard and % increase of the peppermint and clove residual distillation water main components before and after enzymatic
hydrolysis resulting from SBSE-GC-FID analysis. PR1, PR2 and PR3: peppermint first, second and third residual distillation waters, PR1-EH, PR2-EH, PR3-EH: peppermint first,
second and third residual distillation waters after enzymatic hydrolysis. CR1, CR2 and CR3: clove first, second and third residual distillation waters, CR1-EH, CR2-EH, CR3-EH:
clove first, second and third residual distillation waters after enzymatic hydrolysis.
Mentha ꢀ piperita
Compounds
PR1
RSD %
PR1-EH
RSD %
% Increase
PR2
RSD %
PR2-EH
RSD %
% Increase
PR3
RSD %
PR3-EH
RSD %
% Increase
1,8-Cineole
Menthone
Neo-menthol
Neo-i-menthol
L-menthol
0.1
0.3
0.1
0.02
0.7
13
0.1
0.2
0.2
0.03
2.0
18
10
10
12
3.9
0.4
ꢁ19
ꢁ45
167
80
176
26
0.04
0.04
0.03
0.01
0.3
6.3
0.9
7.2
15
8.8
/
0.03
0.05
0.2
0.02
1.0
/
8.7
0.4
1.7
3.9
1.1
/
ꢁ13
12
529
119
226
/
0.02
0.01
0.01
/
0.12
/
6.0
1.5
6.9
/
7.2
/
0.01
0.01
0.1
/
0.35
/
8.7
0.4
1.7
/
1.1
/
ꢁ50
10
800
/
192
/
5.7
1.9
2.0
3.7
3.5
Viridiflorol
0.01
0.02
/
Syzygium aromaticum
RSD % CR2-EH RSD % % increase CR3
Compounds
CR1
RSD % CR1-EH RSD % % increase CR2
RSD % CR3-EH RSD % % increase
t-Caryophyllene 0.01 4.0
0.01
1.8
0.03
1.3
0.5
7.3
ꢁ21
30
18
0.01 1.0
0.8 0.1
0.06 5.4
0.004
1.0
0.04
8.9
0.8
0.6
ꢁ58
27
ꢁ44
/
0.3
0.01 2.5
/
1.2
/
/
1.5
1.9
/
35
10
Eugenol
Chavicol
1.4
0.03 0.4
0.02
0.4
0.01
2.1.2.3. LC-MS analysis of L-menthol glucoside in peppermint residual
distillation waters (PRs). L-Menthol glucoside in PR1, PR2 and PR3
was identified by direct injection of the PRs in an HPLC-MS-MS sys-
tem with a triple quadrupole analyser, through its fragmentation
pattern and by comparison with a 78% pure standard from the
authors’ standard collection of reference compounds and extracts.
The ESI + precursor ion was 341.00 m/z [M + Na]+ while the transi-
tions monitored were m/z 341.00 ? 179.00 and m/z
341.00 ? 161.00, these two ions respectively corresponding to
The yield of the essential oil obtained by hydrodistillation for
the investigated clove sample (CEO-1) was 8.8 g of EO in 100 g of
plant material, corresponding to 6588 mg of eugenol; these results
were confirmed by quantitation through GC-MS (Table 1a).
2.2.1.3. Quantitation of free eugenol in clove residual distillation
waters (CR). Free eugenol was first quantitated in CR1, CR2 and
CR3 by SBSE-GC-FID/MS; the results are summarised in Table 1a.
Free eugenol concentration in CR1 was 150 ppm, corresponding
to 600 mg, while its concentration in CR2 was 78.9 ppm, i.e.
316 mg, and in CR3 it was 32 ppm, i.e. 130 mg. These analyses
showed that the residual amount of free eugenol in CR1, CR2 and
CR3 was quite significant (about 13%, i.e. 1046 mg). In this case,
too, the results for eugenol were confirmed by in-solution-SPME-
GC–MS (data not reported). Moreover, the amount of free eugenol
contained in the essential oil and in the residual distillation waters
was in a good agreement with that determined by HS-SPME-GC–
MS in cloves. (Table 1c).
the adducts [L-menthol (m/z 156) + Na]+ and [L-menthol
–
H2O + Na (m/z 161)]+. The amount of L-menthol glucoside in PRs
was not reliably measurable, making its quantitation unsuitable
for consideration for comparison with the other methods applied.
2.2. Cloves
2.2.1. Determination of free Eugenol
2.2.1.1. Quantitation of eugenol in the headspace of cloves. Eugenol
was quantitated in cloves by MHS-SPME-GC–MS, with three con-
secutive extractions of 1 mg of cloves dispersed in Celite at a
1:20 ratio, to achieve linear decay. Decay equation, linear regres-
sion coefficient and average quotient Q of eugenol in cloves are col-
lected in Table 1SM. A calibration curve was built up by analysing a
set of eugenol standard solutions under the same conditions (i.e.
three consecutive extractions) as reported in paragraph 4.3.5. The
amount of eugenol in clove, determined by MHS-SPME-GC–MS
(CHS-1), was 78.9 mg per g of plant material, corresponding to
7890 mg in 100 g of cloves (Table 1a).
2.2.2. Determination of glucosidically-bound eugenol
2.2.2.1. Clove essential oil after enzymatic hydrolysis. The clove EO
obtained after enzymatic hydrolysis (EH-CEO) gave a 11.6 g of EO
in 100 g of cloves, corresponding to 7120 mg of eugenol
(Table 2a). The amount of eugenol deriving from the combination
enzymatic hydrolysis-hydrodistillation was therefore 532 mg.
The same five compounds as for CEO-1 were identified and moni-
tored, but the GC-FID/MS profile changed, especially for eugenol,
which accounted for 88.8% of the EH-CEO composition.
2.2.1.2. Eugenol in clove essential oil. The same procedure adopted
for peppermint aerial parts was applied to cloves. Table 2 reports
2.2.2.2. Quantitation of glucosidically-bound eugenol in clove residual
distillation waters (CR). Glucosidically-bound eugenol was quanti-
tated in CR-1, CR-2 and CR-3 by SBSE-GC-FID/MS after enzymatic
hydrolysis under the same conditions reported for the aglycone.
The eugenol quantitative results are summarised in Table 1b.
Eugenol concentration in CR-1 after enzymatic hydrolysis (CR-1-
EH) was 226 ppm, corresponding to 904 mg, while its concentra-
tion in CR-2-EH was 115.6 ppm, i.e. 462 mg, and in CR3-EH it
was 46.8 ppm, i.e. 188 mg.
the list and relative percentage abundances normalised versus
a-
thujone of eugenol in clove. The GC-FID profile of the EO obtained
by conventional hydrodistillation (HD) from clove (CEO-1) and list
of all identified components and the chemical composition of CEO
expressed as relative percentage abundances are included in sup-
plementary material (Figure 3SM and Table 4SM). Five components
were identified, the most abundant of them being eugenol (80.3%),
eugenyl acetate (10.3%) and trans-b-caryophyllene (8.1%).
Table 1a–c lists the quantitative results for eugenol, expressed as
mg of eugenol in 100 g of plant material. As for L-menthol in pep-
permint aerial parts, processed plant material was re-submitted to
two successive hydrodistillations with ‘‘new’’ water, without
obtaining a measurable EO amount in either case. The resulting
CR2 and CR3 were processed together with CR1, as indicated
below.
Table 4 reports the areas of the main components of CR1, CR2
and CR3 and CR1-EH, CR2-EH and CR3-EH i.e. before and after
enzymatic hydrolysis, normalised vs.
a-thujone as internal stan-
dard, together with their % increase. These results clearly show
the increase of eugenol content in CR1 and CR2 due to enzymatic
hydrolysis, whereas the amount of hydrocarbons remained con-
stant. In this case too, the results were confirmed by in-solution-
SPME-GC-FID/MS (data not reported).

302
B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
the environment to a minimum, and at exploiting plant resources
Again, eugenol determined by HS-SPME-GC–MS and that result-
exhaustively.
ing from the sum of its contents in the essential oil and in the
residual waters were in a good agreement.
The strategy adopted here also aimed to evaluate the impact of
recently-introduced techniques that are rarely applied to investi-
gations in this field, if at all. The combined use, for the first time,
of high-concentration-capacity sample preparation techniques
(SBSE, and HS-SPME and in-solution SPME) making it possible to
run reliable quali-quantitative analysis directly, i.e. without sam-
ple manipulation, and direct LC–MS glucoside analysis, provided
cross-validation and increased the reliability of the reported
results. This strategy has been shown to be reliable, since the quan-
titative results obtained by different approaches agreed: e.g. (1) the
sum of the marker amounts contained in the EO and the total agly-
cones in R, compared to the amount determined by HS sampling,
accounted respectively for 84.4% of L-menthol and for 96.8% in
the case of eugenol, and (2) with cloves, the amount and percent-
ages of glucosides expressed as eugenol was closely consistent for
EH-CEO (532 mg), CR-EH (508 mg) and by direct CR LC–MS
(552 mg), varying by up to 8.0%. Similar results were obtained with
peppermint aerial parts, where the glucoside amounts were closely
comparable, being 92.1 mg for EH-PEO, and 82 for PR-EH, i.e. dif-
fering by about 11%.
Unlike peppermint aerial parts, hydrodistillation of clove resid-
ual distillation waters after enzymatic hydrolysis (CEO-R1-EH and
CEO-R2-EH) produced measurable essential oil yields: (a) CEO-R1-
EH yield was 1.1%, calculated on the initial mass of plant material,
and contained 91.7% of eugenol, corresponding to 1009 mg, while
(b) for CEO-R2-EH, it was 0.055% and contained 95.1% of eugenol,
corresponding to 52.5 mg (Table 1b). Conversely, the amount of
CEO-R3-EH was negligible, and could not be recovered for analysis.
2.2.2.3. LC–MS analysis of eugenol glucoside in clove residual distil-
lation waters (CRs). Eugenol glucoside was identified by its UV
spectra (maximum at 276 nm) and mass spectral data in MRM
mode in both ESI+ and ESIꢁ). The MRM transitions were selected
on the basis of the fragment ions obtained by analyzing the euge-
nyl glucoside standard, using 349.00 m/z [M + Na]+ for ESI+ and
371.00 m/z [M + Formic Acid-H]ꢁ for ESIꢁ as precursor ions. The
monitored transitions were: (a) for ESI+: m/z 349.00 ? 187.00 cor-
responding to the adduct [eugenol (m/z 164) + Na]+, (b) for ESI-:
m/z 371.00 ? 163.00 corresponding to the fragments [eugenol
(m/z 164) – 1]ꢁ. The quantitation of eugenyl glucoside in CRs was
carried out on the PDA-UV profiles (at 276 nm) using the external
calibration method. These analyses confirmed its presence in the
three CRs in amounts very similar to those obtained indirectly after
enzymatic hydrolysis (Table 1b).
4. Experimental
4.1. Plant material
Plant material: Mentha x piperita L. (Lamiaceae). Aerial parts of
peppermint were harvested in 2012 and dried at room tempera-
ture in the shade; they were kindly supplied by Dr. Franco
Chialva (ChialvaMenta, Pancalieri, Turin Italy).
3. Conclusions
The results clearly show that the amount of characterising
markers, in both free and glucosidically-bound forms that is not
recovered when an EO is isolated by hydrodistillation is quite con-
siderable. The aglycone(s) not recovered by hydrodistillation can
indicatively be calculated as a percentage through the following
equation:
Dried cloves (Syzygium aromaticum (L.) Merr. & L.M.Perry,
Myrtaceae) were purchased from Cannamela (Zola Predosa (BO),
Italy). Cloves were finely powdered before analysis.
4.2. Reagents and chemicals
All chemicals were analytical grade. Octyl-b-D-glucopyra-
noside, b-glucosidase and Bis-(trimethylslyl)-trifluoroacetamide
QHS ꢁ QEO=QHS ꢀ 100
ð3Þ
(derivatization grade) were from Sigma Aldrich (Milan, Italy).
a-
where QHS is the quantity naturally occurring in the plant material
measured through headspace (HS) and QEO is the amount recovered
from the EO. In the matrices investigated here, these percentages
accounted for 22.8% for L-menthol in peppermint aerial parts, and
16.5% for eugenol in cloves. Of these two compounds, a significant
amount is solubilised as aglycone in the residual distillation waters
(R); in the investigated matrices, free L-menthol in R accounted for
7.2% of the HS reference amount, and free eugenol for 13.3%; at the
same time, the dissolved aglycone represents 9.3% and 15.9% com-
pared to the amounts in their respective EOs. The differences in per-
centages between the two markers, in the two plant matrices
investigated, are fully explained by their significantly different
water solubility. The amount of related glucosides are, of course,
not included in the HS-determined aglycone, although they may
provide an extra yield of the marker if submitted to hydrolysis. In
this case, L-menthol from its glucoside is 20.6% of that in the EO
of peppermint aerial parts, while eugenol deriving from its glu-
coside accounted for 7.7% of the eugenol in clove EO. Another pos-
sibility to increase the yield of the investigated markers is to submit
plant material to enzymatic hydrolysis prior to hydrodistillation:
when this was done, the increase of L-menthol accounted for
23.1%, and that of eugenol for 8.1%, compared to the amount in
the EOs without enzymatic hydrolysis. These results indicates that
the residual distillation waters are by-products exploitable as a
source of useful compounds; this is in line with today’s general
trends aiming at reducing the dissipation of active components in
Thujone and eugenol (99%) were from Sigma Aldrich (Milan,
Italy). Silica gel 60 (70–230 mesh) was from Macherey–Nagel.
Reactions were monitored by TLC on Merck 60 F254 (0.25 mm)
plates, visualised by UV inspection and/or staining with 5% H2SO4
in ethanol and heating. Organic phases were dried with Na2SO4
before evaporation. Solvents (acetic acid, dichloromethane, ether,
ethyl acetate, methanol, petrolether, tetrahydrofuran) and other
reagents were from Sigma Aldrich. All solvents were dried follow-
ing standard procedures.
4.3. Sample preparation and analysis
Each step of the procedure described in the following para-
graphs was repeated three times, and the mean areas of the GC-
FID and GC–MS analysis were considered for further processing.
All steps are summarised in Fig. 1.
4.3.1. Isolation of essential oils (EOs)
EOs were isolated from 100 g of homogeneous peppermint aer-
ial parts, and from 50 g of powdered cloves, by hydrodistillation
using 2000 mL of distilled water. Conventional hydrodistillation
(HD) was performed in a Clevenger-type apparatus for four hours
as indicated in the European Pharmacopoeia (2008). The resulting
EO was separated from the aqueous layer, dried over anhydrous
sodium sulphate, and stored in a sealed vial at low temperature

B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
303
(–18 °C) until analysis. A similar procedure was applied with a
micro-hydrodistillation apparatus (micro-HD) developed in the
authors’ laboratory, in which 10 times less plant material (i.e.
10 g peppermint aerial part and 5 g cloves) were submitted to
hydrodistillation (Bicchi et al., 1983).
SPME device and CAR/PDMS/DVB fused silica fibres were sup-
plied by Supelco (Bellafonte, PA, USA). Before use, all fibres were
conditioned as recommended by the manufacturer, and tested to
evaluate the consistency of their performance versus reference
peppermint and clove samples. In addition, fibre extraction relia-
bility was evaluated by means of full evaporation HS-SPME sam-
4.3.2. Glucosidically-bound volatile extraction
pling of the vapor phase from a standard solution of
(5 l of a 2000 ppm solution in di-butyl phtalate – DBP) and of
l of a 5000 ppm peppermint EO solution in DBP (Bicchi et al.,
a-thujone
The investigated glucoside fractions were solubilised in the
aqueous phase during the hydrodistillation process (HD) to isolate
the EOs. The resulting aqueous phase was separated by decanta-
tion (first residual distillation-water, R1); the residual plant mate-
rial was re-suspended in another 2 l of water and re-hydrodistilled,
obtaining the second essential oil (EO-2) and the second residual
distillation water (R2). This operation was repeated a third time
under the same conditions, to give the third essential oil (EO-3)
and the third residual distillation water (R3).
l
l
1
2007).
4.3.5.1. Headspace sampling conditions. 1 mg of plant material and
10
ll of a 10 ppm solution of a-thujone in 1:1 water:ethanol
(v/v), used as internal standard, were submitted to HS-SPME sam-
pling with a CAR/PDMS/DVB fused silica fibre (2 cm–50/30 lm) for
30 min at 35 °C.
4.3.3. Isolation of essential oils after enzymatic hydrolysis
The plant materials were also submitted to enzymatic hydroly-
sis (EH) prior to hydrodistillation, roughly following the method
described by Zhiping et al. (2006) to isolate the volatile fraction
from roses. 10 g of plant material from Mentha x piperita and 5 g
from Syzygium aromaticum were suspended in a buffer solution
(pH 5.5) consisting of 250 ml distilled water, 3.608 g anhydrous
sodium acetate, and 0.35 ml acetic acid, to which 0.1 g b-glucosi-
dase from almonds were added. The activity of b-glucosidase was
P6 U/mg. The resulting sample was kept at 37 °C overnight under
stirring, and subsequently submitted to micro-hydrodistillation in
the above microapparatus (Bicchi et al., 1983) as described in para-
graph 4.3.1, to give the corresponding essential oil (EH-EO).
4.3.5.2. Multiple headspace solid phase microextraction (MHS-
SPME). L-menthol and eugenol were quantitated with the multiple
headspace extraction approach (MHS-SPME). The total area of
eugenol and menthol was estimated with three consecutive sam-
plings of a suitable amount, 1 mg of peppermint aerial part sample,
and 1 mg of a mixture of powdered cloves and Celite (1:20, w/w),
in order to obtain correct decay trend. A calibration curve was built
up by analysing two sets of solutions of L-menthol and eugenol in
acetone under the same conditions (i.e. three consecutive extrac-
tions). Individual stock solutions of L-menthol and eugenol were
prepared in a 20 ml vial by adding 500 mg of pure standard to an
appropriate volume of HPLC grade acetone (10 ml) to obtain ana-
lyte concentrations of about 50000 and 15000 ppm, respectively.
The calibration solutions of L-menthol were 7500, 5000, 3000
and 1000 ppm, while those of eugenol were obtained by diluting
the stock solution to 25000, 10000, 5000 and 2500 ppm. The cali-
4.3.4. Glucoside hydrolysis in the residual distillation waters (R)
Acidic and enzymatic hydrolysis procedures were experi-
mented directly on the residual water samples resulting from
hydrodistillation (after pH adjustment). Preliminary hydrolysis
assays were carried out using a standard of octyl-b-glucoside as
model compound, to tune both the acidic and the enzymatic
hydrolysis conditions of glucosidically-bound volatiles.
bration curve was built up by introducing 1 ll of each solution in
a 20 ml vial and submitting each vial to three consecutive extrac-
tions (in triplicate for each concentration). The resulting standard
solutions were stored at 0 °C and renewed weekly (Bicchi et al.,
2011).
4.3.4.1. Acidic hydrolysis. Diluted hydrochloric acid was added to
4 ml of R1, to pH 3.0 and to pH 1.0, in sealed vials, and then heated
for 60 min at 100 °C (Maicas and Mateo 2005).
4.3.6. SBSE and in-solution-SPME
SBSE for in-solution-sampling was run with commercial PDMS-
EG twisters™ (df 500 lm–2 cm) supplied by Gerstel (Mülheim a/d
4.3.4.2. Enzymatic hydrolysis. b-glucosidase (16 mg) was added to
4 ml of R1, at pH 5.5, in a sealed vial and submitted to incubation
under stirring at 37 °C overnight. Enzymatic hydrolysis conditions
were in agreement with those reported in the existing literature
(Morales et al., 2000; Boulanger and Crouzet, 2001; Mastelic and
Jerkovic, 2003; Radonic and Mastelic, 2008). The incubation times
giving the highest hydrolysis yield (99 to 100%) ranged from 12 to
24 h while free L-menthol or eugenol in PR1 and CR1 under the
same conditions did not increase after 18 h.
The hydrolysis process was monitored, taking n-octyl-b-
glucoside as internal standard. R1 were first controlled to check
whether free 1-octanol and n-octyl-b-glucoside were absent.
These operations were repeated on R2 and R3. The hydrolysed
volatile aglycones were recovered from the aqueous phase, by
hydrodistillation and solid phase microextraction (SPME) and stir
bar sorptive extraction (SBSE) operating both in headspace and
in-solution modes (for analysis conditions par. 4.3.5 and 4.3.6).
Ruhr, Germany). Sampling was carried out in a thermostatic bath.
For SBSE sampling, twisters were introduced into the sample
solution in a 10 ml vial under the following conditions: sample
amount: 4 mL of R containing 10 lL of a 10 ppm solution of a-thu-
jone in water:ethanol, 1:1 (v/v), used as internal standard; temper-
ature: 35 °C; sampling time: 15 min with agitation at 250 rpm
(Bicchi et al., 2002). Quantitation of L-menthol and eugenol was
done by GC-FID and GC–SIM–MS from the corresponding calibra-
tion curves, built within a suitable range of concentrations, i.e. 10
to 250 ppm for menthol, and 25 to 500 ppm for eugenol resulting
from the dilution of a L-menthol and eugenol solutions in acetone
(HPLC grade) with a suitable volume of HPLC water. The following
target ions and qualifiers were chosen: L-menthol target ion:
71 m/z, qualifiers 123 and 138 m/z, eugenol: 164 m/z, qualifiers
149 and 77 m/z.
In-solution SPME was carried out under the same conditions as
for SBSE, adopting a CAR/PDMS/DVB fibre.
4.3.5. Headspace (HS) analysis of the volatile fraction
The HS composition of peppermint dried leaves, (Mentha x
piperita), and cloves (Syzygium aromaticum) was determined by
headspace solid phase microextraction (HS-SPME) sampling with
a DVB/Carboxen/PDMS (2 cm) fibre.
4.3.7. Analysis conditions
Each sample was analysed in triplicate, using
internal standard, and the FID mean area values for area % normal-
a-thujone as
isation and SIM-MS peak area for quantitative analysis.

304
B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
GC–MS analyses were carried out with an Agilent 6890 unit
provided with FID combined with a mass selective detector model
5973 N. GC operative conditions were:
(MW = 390.12; 7.69 mmol) dissolved in 10 mL of dichloromethane
and hydrogen bromine 33% in acetic acid (MW = 80.12;
38.46 mmol). The reaction was run for 6 h under stirring at room
temperature and monitored by TLC. Brine and sodium bicarbonate
were then added; the organic phase was extracted with dichloro-
methane, dried with sodium sulphate, filtered and recrystallised
from ether/petroleum ether 1:1 (yield: 86%). The product was used
without any further purification.
(a) EOs were analysed by injecting 1 ll of a solution diluted
1:200 in cyclohexane through a MPS-2 multipurpose sam-
pler (Gerstel, Mülheim a/d Ruhr, Germany) on-line com-
bined with the above GC–MS system in split mode, split
ratio 1:20.
(b) In HS-SPME and in-solution-SPME, the sampled volatiles
were recovered by direct desorption (10 min at 230 °C) in
the GC injector port through a MPS-2 multipurpose sampler
on-line combined with the above GC–MS system in split
mode, split ratio 1:20.
(c) In HSSE and SBSE, volatiles were on-line transferred to GC–
MS by a MPS-2 multipurpose sampler (Gerstel, Mülheim
a/d Ruhr, Germany) equipped with a Thermo Desorption
Unit (TDU) and a CIS-4 PTV injector (Gerstel, Mülheim a/d
Ruhr, Germany).
4.5.2. Synthesis of eugenyl peracetylglucoside
Eugenyl-2,3,4,6-tetra-O-acetyl-D-glucopyranoside
(Mw = 494.18) was prepared by reacting
a-acetobromoglucose
(0.487 mmol) and eugenol (0.975 mmol) in acetic acid and sodium
hydroxide 0.725 N to obtain a pH value around 9. After a few min-
utes the pH dropped to 7; it was thereafter maintained at around 9
by further additions of sodium hydroxide. The reaction mixture
was stirred for 24 h at room temperature and the reaction moni-
tored by TLC. Brine was then added; the organic phase was
extracted with tetrahydrofuran, dried with sodium sulphate, fil-
tered, and the resulting eugenyl peracetylglucoside purified by col-
umn chromatography with petroleum ether/ethyl acetate 8:2
(yield: 52%). White powder, 1H NMR (300 MHz, CDCl3): d 7.02
(1H, d, J = 7.95 Hz, 1’-H), 6.70 (1H, s, 3’-H), 6.68 (1H, d,
J = 8.25 Hz, 2’-H), 5.90 (1H, m, 5’-H), 5.70 (1H, d, J = 8.28 Hz, 1-H),
5.30–5.10 (3H, m, 2-, 3- and 4-H), 4.89 (2H, m, 6’-H), 4.25–4,06
(2H, dd, 6-H-6a,b), 3.81 (3H, s, -OMe), 3.78 (1H, m, 5-H), 3.32
(2H, d, J = 6.12 Hz, 4’-H), 2.10–2.02 (12H, s, 2-, 3-, 4-, 6-OAc).
TDU operative conditions: from 30 °C to 270 °C (5 min) at
60 °C/min; flow mode: splitless; transfer line: 270 °C, CIS-4 PTV
injector temp: ꢁ50 °C; coolant: liquid CO2; injection temperature
program: from ꢁ50 °C to 270 °C (10 min) at 12 °C/s; inlet operated
in split mode: split ratio 1:20.
EOs, HS, and in-solution sampled samples were analysed by GC
with a polyethylene glycol fused silica column (50 m ꢀ 200
lm
ꢀ 0.200 i.d.) (MEGA-WAX MEGA Legnano, Milan, Italy).
l
m
Injector and FID detector temp. were maintained at 250 °C and
280 °C, respectively. The column oven was programmed from
35 °C (1 min) to 250 °C (5 min) at 3 °C/min. Carrier gas: (He)
flow-rate: 1 mL/min; injection mode: split, ratio: 1:20. MS condi-
tions: ionisation mode EI at 70 eV; transfer line temp.: 250 °C;
ion source temp.: 250 °C; mass range: 35–350 mass units.
Individual peaks were identified by comparing their linear
retention indices calculated versus a C9–C25 hydrocarbon mixture
to those of authentic samples, as well as by comparing their mass
spectra to a set of commercial and in-house libraries. Percentage
composition was determined from GC-FID data through the peak
area normalisation approach, adopting a response factor for each
class or sub-class of compounds (hydrocarbons, aldehydes, alco-
hols, esters, etc.) in the investigated sample, calculated versus the
internal standard, taking one component representative of each
class (Costa et al., 2008).
4.5.3. Synthesis of eugenyl glucoside
Eugenyl peracetylglucoside (0.243 mmol) was dissolved in
methanol (4 mL) and deacetylated with sodium methoxide 0.5%
in methanol. The reaction was stirred at room temperature for
1 h and monitored by TLC. Brine was then added; the organic phase
was extracted with tetrahydrofuran, dried with sodium sulphate,
filtered, and the resulting eugenyl glucoside (MW = 326.14) was
purified by column chromatography with petroleum ether/ethyl
acetate 8:2 and ethyl acetate/methanol 8:2 (yield: 33%). White
powder, 1H NMR (300 MHz, DMSO-d6): d 7.07 (1H, d, J = 7.95 Hz,
1’-H), 6.80 (1H, s, 3’-H), 6.70 (1H, d, J = 8.25 Hz, 2’-H), 5.92 (1H,
m, 5’-H), 5.06 (1H, s, -OH), 5.02 (1H, d, 6’-H), 4.88 (1H, m, 1-H),
4.85 (1H, s, -OH), 4.55 (1H, s, -OH), 3.81 (3H, s, OMe), 3.68–3,25
(2H, dd, 6AB-H), 3.44 (2H, d, J = 6.12 Hz, 4’-H), 3.68–3,25 (4H, m,
2–3-, 4- and 5-H).
4.6. HPLC-MS analysis
4.4. Repeatability of the method
4.6.1. Analysis conditions
Repeatability was determined in different ways depending on
the sample analysed: (a) MHS-SPME of plant material: three sam-
ples of each matrix were submitted to MHE for a total of 9 analy-
ses; (b) each EO was analysed three times by split injection-GC-FID
after a 1:200 dilution in cyclohexane; (c) each R was sampled three
times by SBSE and analysed by GC–MS. Repeatability was evalu-
ated on L-menthol for peppermint aerial parts and on eugenol for
clove; it is expressed as Relative Standard Deviation% (RSD%) on
Analyses were carried out on a Shimadzu Nexera X2 system
equipped with a photodiode detector SPD-M20A in series to a tri-
ple quadrupole Shimadzu LCMS-8040 system provided with elec-
trospray ionisation (ESI) source (Shimadzu, Dusseldorf Germany).
An AscentisÒ Express C18 column (150 ꢀ 2.1 mm i.d., 2.7
lm par-
ticle size), (Supelco, Bellefonte, PA) was used. Analysis conditions:
mobile phase: eluent A: 0.1% formic acid in water; eluent B: 0.1%
formic acid in acetonitrile; mobile phase gradient was as follows:
5–25% B in 10 min, 25–40% B in 5 min, 40–100% B in 5 min, and
normalised peak areas of L-menthol and eugenol versus a-thujone
as internal standard.
100% B for 1 min. Injection volume: 5 ll; flow-rate 0.4 ml/min; col-
umn temperature: 30 °C. UV wavelength range: 210–450 nm. MS
operative conditions: heat block temperature: 400 °C; nebulizing
gas (nitrogen) flow-rate: 3 l/min; drying gas (nitrogen) flow-rate:
15 l/min; desolvation line (DL) temperature: 250 °C. Collision
gas: argon (230 kPa).
4.5. Eugenol glucoside synthesis
The synthetic pathway is reported in Fig. 1SM (supplementary
material).
L-menthol glucoside was identified by comparison to a standard
available from the library of standards in the authors’ laboratory.
L-menthol glucoside in PRs was identified by its mass spectral data
in positive ionisation mode (ESI+) in Multiple Reaction Monitoring
4.5.1. Synthesis of
-Acetobromoglucose (2,3,4,6-tetra-O-acetyl-D-glucopyranosyl
bromide, MW = 410.02) was prepared from glucose penta-acetate
a-acetobromoglucose
a

B. Sgorbini et al. / Phytochemistry 117 (2015) 296–305
305
Bicchi, C., Ruosi, M.R., Cagliero, C., Cordero, C., Liberto, E., Rubiolo, P., Sgorbini, B.,
2011. Quantitative analysis of volatiles from solid matrices of vegetable origin
by high concentration capacity headspace techniques: determination of furan
in roasted coffee. J. Chromatogr. A 1218, 753–762.
Boulanger, R., Crouzet, J., 2001. Identification of the aroma components of acerola
(Malpighia glabra L.): free and bound flavor compounds. Food Chem. 74, 209–
216.
(MRM) mode. The MRM transitions were selected on the fragments
obtained by analysing the L-menthol glucoside standard in full-
scan mode (range: 150–500 m/z; scan speed: 1000 u/s) and then
in product ion scan mode in the range 100–400 m/z (scan speed:
1500 u/s).
Eugenyl glucoside was identified and quantitated versus a pure
standard synthesised as reported in paragraph 4.5. Its identifica-
tion was based on its UV spectrum (maximum at 276 nm) and
mass spectral data in both positive and negative ionisation mode
(ESI+ and ESI) in Multiple Reaction Monitoring (MRM) mode. The
MRM transitions were selected on the fragments obtained by ana-
lysing the eugenyl glucoside standard in full-scan in both ESI+ and
ESIꢁ (range: 150–500 m/z; scan speed: 1000 u/s) and then in
product ion scan mode in both ESI+ and ESIꢁ in the range 100–
400 m/z, scan speed: 1500 u/s (dwell time: 20 ms, collision energy:
35 V, event time: 0.096 s). Quantitation was done on the PDA-UV
profiles (at 276 nm) using the external calibration method.
Bourquelot, E., Bridel, M., 1913. Synthèse des glucosides d’alcools
à l’aide de
l’émulsine et réversibilité des actions fermentaires. Ann. Chim. Phys. 29, 145–
218.
Costa, R., d’Acampora Zellner, B., Crupi, M.L., De Fina, M.R., Valentino, M.R., Dugo, P.,
Dugo, G., Mondello, L., 2008. GC–MS, GC–O and enantio–GC investigation of the
essential oil of Tarchonanthus camphoratus L. Flavour Fragr. J. 23, 40–48.
Crouzet, J., Chassagne, D., 1999. Glycosidically bound volatiles in plants. In: Ikan, R.
(Ed.), Naturally Occurring Glycosides. John Wiley & Sons, Chichester, pp. 225–
274.
Den Van Dries, J.M.A., Baerheim Svendsen, A., 1989. A simple method for detection
of glycosidic bound monoterpenes and other volatile compounds occurring in
fresh plant material. Flavour Fragr. J. 4, 59–61.
Episuite ver. 4.11, November 2012.
European Pharmacopoeia, 8th ed., 2008.
Ezquerro, Ó., Pons, B., Tena, M.T., 2003. Multiple headspace solid-phase
microextraction for the quantitative determination of volatile organic
compounds in multilayer packagings. J. Chromatogr. A 999, 155–164.
Hosel, W., 1981. Glycosylation and glycosidases. In: Stumpf, P.K., Conn, E.E. (Eds.),
The Biochemistry of Plants, vol. 7. Academic Press, New York, p. 725.
Kolb, B., Ettre, L.S., 1997. Static Headspace-Gas Chromatography, Theory and
Practice. Wiley-VCH, New York (and references cited therein).
Maicas, S., Mateo, J.J., 2005. Hydrolysis of terpenyl glycosides in grape juice and
other fruit juices: a review. Appl. Microbiol. Biotechnol. 67, 322–335.
Mastelic, J., Jerkovic, I., 2003. Gas chromatography–mass spectrometry analysis of
free and glycoconjugated aroma compounds of seasonally collected Satureja
montana L. Food Chem. 80, 135–140.
Morales, A.L., Duque, C., Bautista, E., 2000. Identification of free and glycosidically
bound volatiles and glycosides by capillary GC and Capillary GC–MS in ‘‘Lulo del
Chocó’’(Solanum topiro). J. High Resolut. Chromatogr. 23, 379–385.
Nirmala Menon, A., Narayanan, C.S., 1992. Glycosidically bound volatiles of clove
Syzygium aromaticum (L.) Merr. et Perry (Myrtaceae). Flavour Fragr. J. 7, 155–
157.
Acknowledgements
This study was carried out within the project ‘‘Studio di meta-
boliti secondari biologicamente attivi da matrici di origine vege-
tale’’ financially supported by the Ricerca Locale (Ex 60% 2014) of
the University of Turin, Turin (Italy).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.phytochem.2015.
06.017.
Radonic, A., Mastelic, J., 2008. Essential oil and glycosidically bound volatiles of
Thymus pulegioides L. growing wild in Croatia. Croat. Chem. Acta 81, 599–606.
Sgorbini, B., Bicchi, C., Cagliero, C., Cordero, C., Liberto, E., Rubiolo, P., 2015. Herbs
and spices: characterization and quantitation of biologically-active markers for
routine quality control by multiple headspace solid-phase microextraction
combined with separative and nonseparative analysis. J. Chromatogr. A 1376,
9–17.
Stahl-Biskup, E., Intert, F., Holthuijzen, J., Stengele, M., 1993. Glycosidically bound
volatiles – a review 1986–1991. Flavour Fragr. J. 8, 61–80.
Tienpont, B., David, F., Bicchi, C., Sandra, P., 2000. High capacity headspace sorptive
extraction. J. Sep. Sci. 12, 577–584.
Williams, P.J., Strauss, C.R., Wilson, B., Massy-Westropp, R.A., 1982. Use of C18
reversed-phase liquid chromatography for the isolation of monoterpene
glycosides and nor-isoprenoid precursors from grape juice and wines. J.
Chromatogr. 235, 471–480.
Winterhalter, P., Skouroumounis, G.K., 1997. Glycoconjugated aroma compounds:
occurrence, role and biotechnological transformation. In: Scheper, T. (Ed.),
Advances in Biochemical Engineering/Biotechnology, vol. 55. Springer, Berlin,
Heidelberg, pp. 73–106.
Zheljazkova, V.V.D., Astatkieb, T., 2012. Distillation waste water can modify
peppermint (Mentha ꢀ piperita L.) oil composition. Ind. Crops Prod. 36, 420–
426.
Zhiping, Y., Weirong, Y., He, Q., 2006. Studies on vapor phase extraction of rose oil
enhanced by b-glucosidase. Flavour Fragr. J. 21, 776–782.
References
Ananthakumar, A., Variyar, P.S., Sharma, A., 2006. Estimation of aroma glycosides of
nutmeg and their changes during radiation processing. J. Chromatogr. A 1108,
252–257.
Arthur, C.L., Pawliszyn, J., 1990. Solid phase microextraction with thermal
desorption using fused silica optical fibers. Anal. Chem. 62, 2145–2148.
Baltussen, E., Sandra, P., David, F., Cramers, C., 1999. Stir bar sorptive extraction
(SBSE),
a novel extraction technique for aqueous samples: theory and
principles. J. Microcolumn Sep. 11, 737–747.
Bicchi, C., D’Amato, A., Nano, G.M., Frattini, F., 1983. Improved method for the
analysis of small amounts of essential oils by microdistillation followed by
capillary gas chromatography. J. Chromatogr. 279, 279–286.
Bicchi, C., Cordero, C., Iori, C., Rubiolo, P., Sandra, P., 2000. Headspace sorptive
extraction (HSSE) in the headspace analysis of aromatic and medicinal plants. J.
Sep. Sci. 23, 539–546.
Bicchi, C., Iori, C., Rubiolo, P., Sandra, P., 2002. Headspace sorptive extraction (HSSE),
stir bar sorptive extraction (SBSE), and solid phase microextraction (SPME)
applied to the analysis of roasted arabica coffee and coffee brew. J. Agric. Food
Chem. 50, 449–459.
Bicchi, C., Cordero, C., Liberto, E., Sgorbini, B., Rubiolo, P., 2007. Reliability of fibres in
solid-phase microextraction for routine analysis of the headspace of aromatic
and medicinal plants. J. Chromatogr. A 1152, 138–149.