See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/41700843

A simple extractive technique for honey flavonoid HPLC

analysis

Article  in  Apidologie · March 1994

DOI: 10.1051/apido:19940103

CITATIONS READS

95 790

6 authors, including:

Federico Ferreres Francisco A Tomás-Barberán

Spanish National Research Council Spanish National Research Council
383 PUBLICATIONS   17,001 CITATIONS    493 PUBLICATIONS   34,023 CITATIONS   

SEE PROFILE SEE PROFILE

Garcia-Viguera Cristina
Spanish National Research Council
247 PUBLICATIONS   11,106 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

BEE SHOP View project

Evaluación de componentes químicos volátiles y no volátiles asociados a la calidad del cacao ecuatoriano View project

All content following this page was uploaded by Federico Ferreres on 31 May 2014.

The user has requested enhancement of the downloaded file.

 Original article

A simple extractive technique for honey flavonoid

HPLC analysis

F Ferreres FA Tomás-Barberán C Soler

C García-Viguera A Ortiz F Tomás-Lorente
1 CEBAS
(CSIC), PO Box 4195, Murcia 30080;
2 Laboratorio de Mieles, Centro Regional Apícola, SIA, JJ-CC Castilla-La Mancha, Marchamalo,
Guadalajara 19180, Spain

(Received 1 st March 1993; accepted 20 May 1993)

Summary — A simple technique for routine analysis of flavonoids from honey has been described
utilising a combination of filtration through the resin Amberlite XAD-2 and extraction with ethyl ether.
The proposed method is less complex than other methods for honey flavonoid analysis reported pre-
viously. The HPLC conditions for flavonoid analysis have also been improved. This technique was
applied to the analysis of flavonoids in 27 honey samples from the La Alcarria region (Spain). The to-
tal flavonoid content of the different samples ranged between 5 and 20 μg flavonoid/g honey. The
major flavonoids in these samples were the flavanones pinocembrin and pinobanksin and the fla-
vone chrysin. A total of 18 different flavonoids were detected in the honey samples analysed.

honey / flavonoid / botanical origin / geographical origin / HPLC

INTRODUCTION flavonoids. The main problem in the analy-

sis of flavonoids from honey is the latter’s
The occurrence of flavonoids in honey has very high sugar content, which renders dif-
been reported a number of times over the ficult the extraction of these metabolites
last 10 yr (Bogdanov, 1984; Amiot et al, and sample preparation for HPLC analy-
1989; Ferreres et al, 1991; Sabatier et al, sis. Liquid-liquid partitions produce incon-
1992). Flavonoid analysis is a very prom- venient interphases which do not permit
ising technique in studies of the botanical the complete recovery of flavonoids. This
(Amiot et al, 1989; Ferreres et al, 1992) problem has recently been solved by us-
and geographical (Ferreres et al, 1991, ing the non-ionic polymeric resin Amberlite
1992; Tomás-Barberán et al, 1993a) ori- XAD-2 (Ferreres et al, 1991; Tomás-
gins of honey. It is well established that Barberán et al, 1992). In a previous paper
HPLC is a method of choice for analysis of we reported the identification of 16 flavo-
noids in honey via HPLC analysis of sam- (25 x 2 cm) of Amberlite XAD-2 (Fluka Chemie;
ples prepared with a combination of Am- pore size 9 nm, particle size 0.3-1.2 mm). The
various phenolic compounds remained in the
berlite XAD-2 and Sephadex LH-20 col-
column while sugars and other polar com-
umn chromatography (Ferreres et al, pounds were eluted with the aqueous solvent
1991).However, the limitations of this (Ferreres et al, 1991). The column was washed
technique were that it was rather complex, with acid water (100 ml) and subsequently with
especially as regards the Sephadex LH-20 distilled water (= 300 ml). The whole phenolic
fraction was then eluted with methanol (≈ 300 ml
chromatography, and unsuitable for rou- until no more colour was eluted) and concentrat-
tine analyses in quality control determi-
ed under reduced pressure (40°C). Although the
nations. In addition, it did not allow the main proportion of honey sugars had been re-
quantification of flavonoids, since the re- moved by filtration through the Amberlite col-
covery of flavonoids from the Sephadex umn, some sugars still contaminated the phenol-
LH-20 column was not accurate owing to ic fraction. This phenolic compounds fraction
was analysed via HPLC and divided into 4 ali-
the fact that the separation of the flavonoid
fraction from the previous eluting phenolic quots. The first aliquot was treated by filtration
through Sephadex LH-20 column as reported
derivatives was not clear-cut. Thus, the previously (Ferreres et al, 1991).The second ali-
aim of the present work was to improve quot of the eluate from the Amberlite column
upon this analytical technique, to avoid was redissolved in 10 ml distilled water and fil-
tered through a Maxi-Clean RP-C-18 (900 mg)
utilising the expensive Sephadex LH-20
chromatography, and to apply this modi- cartridge to retain the phenolics. The cartridge
was washed with 50 ml distilled water and then
fied technique to the qualitative and quan- with 50 ml of the following solutions: 20% meth-
titative analysis of flavonoids in La Alcarria anol, 30% methanol, 40%, methanol, 50% meth-
honey. anol 60% methanol and 80% methanol. The
phenolics eluting with the different methanol-
water mixtures were analysed via HPLC. The
third aliquot was redissolved in 4 N NaOH and
MATERIALS AND METHODS
left overnight in a stoppered test tube under a ni-
trogen atmosphere to complete saponification.
This was then taken to pH 2 with HCl and ex-
Honey samples tracted with ethyl ether (5 ml x 3). The ether ex-
tracts were combined, concentrated under re-
duced pressure and redissolved in 0.5 ml
The different honey samples used in this study methanol for HPLC analysis. The last aliquot
came from the La Alcarria region (Cuenca and was taken to dryness under reduced pressure
Guadalajara provinces, Spain) and were directly and the residue redissolved in 5 ml distilled wa-
provided by the bee-keepers. The honey sam- ter. This water extract was partitioned with ethyl
ples had not been industrially processed. To ether (5 ml x 3), the ether extracts combined
minimize any alterations, the samples were and the ether removed under reduced pressure.
stored at -20°C in the dark. In addition 1 kg The residue was dissolved in 0.5 ml methanol
commercial honey from La Alcarria was used in and analysed via HPLC.
the modified technique.

HPLC analysis of honey flavonoids

Flavonoid extraction from honey
This was carried out on a reversed-phase col-
Two hundred g commercial honey from La Al- umn LiCrochart RP-18 (Merck, Darmstadt)
carria were thoroughly mixed with 5 parts of wa- (12.5 x 0.4 cm, 5 μm particle size), using water-
ter (pH 2 with HCl) until completely fluid and fil- formic acid (19:1) (solvent A) and methanol
tered through cotton to remove solid particles. (solvent B) as solvents. Elution was performed
The filtrate was then passed through a column at a solvent flow rate of 1 ml/min, starting with
30% methanol which remained isocratic until 15 tifiedby absorbance of their corresponding
min, then installing a gradient to obtain 40% peaks in the chromatograms, the flavanones as
methanol at 20 min, 45% methanol at 30 min, pinocembrin detected at 290 nm, the flavones
60% methanol at 50 min, and 80% methanol at with an unsubstituted B ring (chrysin, galangin
52 min, and which then become isocratic until and tectochrysin) as chrysin detected at 340 nm
60 min. Detection was performed with a diode- and the rest of flavonols and flavones as querce-
array detector, and chromatograms were record- tin detected at 340 nm.
ed at 340 and 290 nm. The retention times for
the different flavonoids identified are shown in
table I.
Analysis of flavonoids from La Alcarria
honey samples
Flavonoid identification
and quantification The different honey samples (50 g each) were
filtered through an Amberlite XAD-2 column.
The methanolic eluate was then concentrated
The different flavonoidswere identified by chro- and extracted with ethyl ether 3 times as de-
matographic comparisons with authentic mark- scribed above. The 3 ether extracts were com-
ers (commercial or previously isolated and iden- bined, concentrated under reduced pressure
tified from honey) (Ferreres et al, 1991, 1992) and redissolved in 0.5 ml methanol for HPLC
and by their UV spectra. Flavonoids were quan- analysis.
RESULTS AND DISCUSSION A previously described technique (Fer-
reres et al, 1991) was used for this pur-
pose, ie a second chromatography on Se-
Improvement of the analytical phadex LH-20 column with methanol. The
technique purification of the flavonoid fraction by this
technique has already been described in
detail (Ferreres et al, 1991) and deserves
The first step in the extraction of flavo-
no further comment apart from a brief men-
noids from honey is the sorption of the
tion of some of its disadvantages. First, it
phenolic compounds on a non-ionic poly- is a time-consuming technique, which is
meric resin Amberlite XAD-2. This was
carried out by filtration of a solution of hon-
experimentally complex and which re-
quires a UV lamp to follow the chromatog-
ey in acid water through a column contain- raphy development in a dark room; this
ing the resin, and washing the sugars and step is therefore unsuitable for routine
other polar compounds with water. The re-
analyses. Secondly, the separation of the
tained phenolics were then eluted with flavonoid fraction from the previous eluting
methanol. The recovery of flavonoids in
phenolic acid derivatives and brown poly-
distilled water solutions (pH 5.5) is≈ 75% mers is not clear-cut, and in most cases
(Tomás-Barberán et al, 1992), whereas in part of the flavanones pinobanksin and pin-
acid solutions (pH 2) the flavonoid aglyc- ocembrin which elute very close to the
one recovery is > 95% (García-Viguera, phenolic acid fraction are discarded; thus
1991).For this reason, the use of acid wa- the amount of flavanones detected in the
ter (pH 2 with HCl) to dilute honey before chromatograms may not accurately repre-
passing through the Amberlite XAD-2 col- sent the real flavonoid composition of hon-
umn is highly recommended. The HPLC ey. This prevents an accurate quantifica-
chromatogram of the honey phenolic frac- tion of these flavanones from being made.
tion obtained with the Amberlite XAD-2 Thirdly, but no less important, is the fact
column (fig 1) shows that some of the flav- that Sephadex LH-20 is rather expensive,
onoids previously reported from honey are although it can be used for a large number
detected, but that there are also some oth- of assays. The chromatogram obtained for
er non-flavonoid phenolic compounds the flavonoid fraction of the same honey
which contaminate the flavonoids peaks. purified by Sephadex LH-20 chromatogra-
The overlapping of peaks is clearly ob- phy is shown in figure 1. However, there
served with a diode-array detector. In addi- are some advantages to using the Sephad-

tion, there is another problem which ex LH-20 step, since a fraction containing
makes this extract unsuitable for direct flavonoids only is prepared for analysis,
analysis of flavonoids: in spite of the large and the chromatograms obtained are clean
amount of sugars eliminated in the Amber- and suitable for the identification of flavo-
lite XAD-2 step, some sugars still contami- noid markers which indicate the botanical
nate the phenolic fraction which is eluted origin. This step seems to be especially
with methanol. When this extract is con- useful in flavonoid analysis of honey sam-
centrated to be redissolved in methanol for ples that are rich in phenolic acid deriva-
HPLC analysis, the sugars present give tives and brown polymers.
the extract a syrupy texture, rendering diffi- Some attempts were than made the
cult its injection in the HPLC. For these avoid to use of Sephadex LH-20 and to ob-
reasons additional treatment of the sample tain a suitable chromatogram of honey
is necessary. flavonoids.
 First, filtration was attempted through a acid derivatives which appear in the first
reversed-phase cartridge to retain flavo- part of the HPLC chromatogram. The phe-
noids and elute the remaining contaminant nolic fraction was passed through a C-18
sugars which were not eliminated via the cartridge and washed with distilled water.
Amberlite XAD-2 step. In addition, this was The flavonoids were then successively
used to obtain selective elution of the flav- eluted with different methanol-water solu-
onoid fraction, cleaning the sample of poly- tions ie: 20% MeOH (only pinobanksin
meric brown phenolics and other phenolic eluted), 30% methanol (pinobanksin and
quercetin), 40% methanol (quercetin, 8- the ether removed under reduced pressure
methoxykaempferol, kaempferol, pinocem- and the residue dissolved in a minimum
brin and chrysin), 50% methanol (querce- amount of the methanol to be analysed via
tin, 8-methoxykaempferol, kaempferol, HPLC. The ether preferentially extracted
chrysin and galangin), 60% methanol the flavonoids with a recovery of > 95% af-
(quercetin, 8-methoxykaempferol, kaemp- ter the 3 extractions, and left sugars and
ferol, chrysin and galangin) and 80% other polar compounds in the water layer.
methanol (chrysin, galangin, genkwanin In addition, the more lipophilic compounds
and tectochrysin). The results indicated were eliminated when the dry residue was
that in order to purify the whole flavonoid redissolved in water. This extraction tech-
fraction, the cartridge should be washed nique was considered suitable, and the
with 50 ml water and 50 ml 15% methanol chromatogram obtained for honey flavo-
to remove sugars and phenolic acid deriv- noids obtained by this method is shown in
atives, after which the flavonoids should figure 1. This chromatogram shows no sig-
be eluted with 80% methanol (50 ml). This nificant differences with that obtained after
flavonoid fraction was then concentrated, filtration through Sephadex LH-20, and is
redissolved in methanol and analysed via the best of the 3 alternatives assayed in
HPLC. The contaminant sugars had been this work to avoid the costly Sephadex LH-
removed, but the chromatogram obtained 20 step.
still showed some of the lipophilic phenolic
acid derivatives and brown polymers
which, in some cases, eluted with the Application of the simplified technique
same retention times as the flavonoids, to flavonoid analysis
rendering this step unsuitable. in La Alcarria honey
A second attempt was made by treating
the phenolic fraction with an alkaline solu- The flavonoids from 27 honey samples
tion to hydrolyse the phenolic acid esters from the La Alcarria region were analysed
which were presumably present in honey
by the proposed technique in order to vali-
as they are important constituents of date this procedure and assess its applica-
propolis (Wollenweber et al, 1987). This tion to the routine flavonoid analysis of
was carried out as described in the Materi-
honey. Flavonoids were extracted from the
als and Methods, and the ether extracts
samples in only 2 steps: firstly filtration of
obtained after acidification of the saponi- the honey through Amberlite XAD-2; and
fied extract were analysed via HPLC but
secondly, extraction of the flavonoids re-
although the chromatograms appeared tained in the Amberlite column with ethyl
somewhat cleaner than those obtained for ether. The flavonoid fractions of the differ-
the Amberlite-eluting fraction, they did not ent honey samples were dissolved in
show any significant improvement. methanol (0.5 ml) and analysed via HPLC
The third method attempted was the ex- under the chromatographic conditions
traction of flavonoids from the Amberlite described in the Materials and Methods.
XAD-2 purified phenolic compounds frac- The results obtained have been shown in
tion with ethyl ether, to preferentially ex- table II. It is interesting that the amount of
tract the flavonoids and leave dark phenol- total flavonoids present in the honey sam-
ic polymers and contaminant sugars in the ples analysed ranged between 5 and 20
aqueous layer. The partitioning took place μg of flavonoid per g honey. In previous
3 times to ensure recovery of the flavo- studies on the flavonoids from French sun-
noids, and the extracts were combined, flower honeys, higher amounts (50-100
μg/g honey) were reported (Ribeiro- CONCLUSION
Campos et al, 1990). The main flavonoid
components in the honey samples ana- The proposed simplified technique for hon-
lysed here were the flavanones pinobank- ey flavonoid analysis avoids the use of fil-
sin (A) and pinocembrin (I) and the flavone
tration through Sephadex LH-20 by extrac-
chrysin (M). Generally, those flavonoids tion of the flavonoids eluting from the
from propolis were present in all the sam-
Amberlite XAD-2 column with ether. This is
ples analysed (compounds A, I, K, M and a suitable technique for routine analysis of
N). However, tectochrysin (P), which is flavonoids from honey, as has been dem-
also known to be present in propolis, was
onstrated by the successful analysis of a
only detected in some of the samples and number of honey samples in the present
in very small amounts. This is probably
due to the fact that this is an extremely lip- study. The reproducibility of the analysis
was ± 5%. The HPLC chromatographic
≈

ophilic compound which is mainly present

in beeswax, and its presence in honey conditions for the analysis of flavonoids
from honey have been improved with re-
probably depends on the contamination of
honey with beeswax (Tomás-Barberán et spect to other previously reported chromat-
al, 1993b). The occurrence and amount of ographic conditions (Ferreres et al, 1992),
those flavonoids present mainly in nectar and allows a clear separation between
and/or pollen, such as quercetin (B), luteo- quercetin and pinobanksin, which were
lin (C), 8-methoxykaempferol (D), kaemp- eluted together in previous analyses. How-
ferol (E), apigenin (F) and isorhamnetin ever, the separation of luteolin and querce-
(G), are much more variable in these hon- tin 3-methyl ether, and apigenin and
ey samples, as could be expected, since kaempferol 3-methyl ether, was not possi-
their occurrence in honey depends on the ble under these chromatographic condi-
latter’s botanical origin. It is interesting that tions (table I).
quercetin (B), which was not detected in
10 of the samples analysed, was present
in honeys G-54 and G-116 in amounts of = ACKNOWLEDGMENTS
1 μg per g honey. Something similar oc-
curred with 8-methoxykaempferol (D) The authors are grateful to the Spanish CICYT
which was absent in samples G-112 and (Grants AL191-0486 and ALI92-0151) and to the
G-113 and present in =
1 μg/g honey in Consejeria de Agricultura de la Junta de Comu-
samples G-53, G-54 and G-122. This nidades de Castilla La Mancha for their financial
agrees with the different floral origin of support.
these samples collected from the same
geographical region. However, all the sam-
ples analysed contained apigenin (apige- Résumé — Technique simple d’extrac-
nin + kaempferol 3-methyl ether) in tion pour analyser en HPLC les flavo-
amounts ≥ 1 μg/g honey. On the contrary, noïdes du miel. L’analyse des flavonoïdes
quercetin 3,3’-dimethyl ether (H), luteolin du miel présente un intérêt pour étudier
7-methyl ether (L) and genkwanin (O) l’origine géographique et botanique du pro-
were detected only in some of the sam- duit. Nous décrivons ici une technique sim-
ples, and in very small amounts. Quercetin ple d’analyse des flavonoïdes. Les divers
3,7-dimethyl ether (J), was only detected échantillons de miel (50 g chacun) ont été
in trace amounts in some honey samples, soigneusement mélangés avec 5 parties
but its quantification was not possible and d’eau (à pH 2 avec HCl) et filtrés à travers
is not included in table II. une colonne Amberlite XAD-2 qui retient
les flavonoïdes et élimine les sucres. Les miel / flavonoïde / origine botanique /
flavonoïdes ont été ensuite élués au mé- origine géographique / HPLC
thanol et l’éluat a été concentré sous vide.
La fraction sortante des flavonoïdes a été
analysée par chromatographie liquide Zusammenfassung — Eine einfache Ex-
haute pression (HPLC), mais on s’est traktionsmethode für die HPLC-Analyse
rendu compte qu’une purification plus von Honig-Flavonoiden. Die Analyse von
poussée était nécessaire avant l’analyse Flavonoiden im Honig ist für Untersuchun-
en HPLC. Quatre méthodes différentes ont gen der geographischen und botanischen
été testées (fig 1).La plus efficace est la Herkunft des Honigs von Interesse. In der
méthode déjà décrite de la filtration à tra- gegenwärtigen Studie wird eine einfache
vers une colonne Sephadex LH-20, mais Methode zur Analyse der Honig-
elle ne permet pas de quantifier les flavo- Flavonoiden beschrieben. Die verschiede-
noïdes, est assez compliquée et ne nen Honigproben (zu je 50 g) wurden mit
convient pas pour des analyses de routine. fünf Teilen Wasser (bei pH 2 mit HCl)
Les 3 autres techniques testées sont plus gründlich gemischt und durch eine Amber-
simples et la meilleure consiste à redissou- lite XAD-2-Säule gefiltert, um die Flavonoi-
dre, dans de l’eau additionnée d’éther éthy- de zurückzuhalten und die Zucker zu ent-
lique, l’éluat obtenu à la sortie de la colon- fernen; die Flavonoide wurden dann mit
ne Amberlite XAD-2. L’éther extrait Methanol herausgelöst und das Eluat im
préférentiellement les flavonoïdes avec un Vakuum konzentriert. Die aus dem Amber-
taux de recouvrement supérieur à 95%. Ni lite XAD-2-Filter gewonnene Flavonoid-
la filtration à travers des cartouches à Fraktion wurde im HPLC analysiert; dabei
phase inversée ni la saponification de l’ex- zeigte sich, daß für die HPLC-Analyse eine
trait avant extraction n’ont donné de weitere Reinigung der Flavonoid-Fraktion
meilleurs résultats. Cette technique a été erforderlich war. Zu diesem Zweck wurden
appliquée à l’analyse des flavonoïdes de vier Methoden geprüft (Abb 1).Die wir-
27 échantillons de miel de la région de La kungsvollste Technik war die früher be-
Alcarria (Espagne) et 18 flavonoïdes diffé- schriebene Filtration durch eine Säule vom
rents ont été identifiés (pinobaksine, quer- Typ Sephadex LH-20, aber bei dieser Me-
cétine, lutéoline, éther de quercétin 3- thode war keine Quantifizierung der Flavo-
méthyle, 8-méthoxykaempférole, kaempfé- noide möglich, sie war technisch kompli-
role, apigénine, o-méthyl 3 kaempférol, ziert und für Routineuntersuchungen unge-
isorhamnétine, o-diméthyl 3,3’ quercétine, eignet. Die anderen drei geprüften Techni-
pinocembrine, o-diméthyl 7,3’ quercétine, ken waren einfacher; als beste Methode
o-diméthyl 3,7 quercétine, o-méthyl 7 lu- erwies sich die neuerliche Lösung des Elu-
téoline, chrysine, galangine, genkwanine ates aus der Amberlite XAD-2-Säule in
Wasser mit Äthyläther. Äther extrahierte
et tectochrysine) (tableau I). La quantité to-
tale de flavonoïdes varie entre 5 et 20 μg vor allem Flavonoide, mit einer Rückgewin-

par g de miel (tableau II). Les flavones pi- nungsrate von über 95%. Versuche mittels
nobanksine et pinocembrine et la flavone Filtration durch Patronen in Reverse-Phase
chrysine sont les constituants majoritaires oder Saponifizierung des Extraktes vor der
des flavonoïdes. Dans l’ensemble, les Extraktion brachten keine besseren Resul-
échantillons avaient tous une composition tate.
semblable, avec seulement de petites dif- Diese Technik wurde bei der Flavonoid-
férences dans les flavonoïdes qui provien- Analyse von 27 Honigen aus der Region
nent du nectar et du pollen. ’La Alcarria’ (Spanien) angewendet und es
 konnten 18 verschiedene Flavonoide be- Ferreres F, Tómas-Barberán FA, Gil MI, Tomás-
stimmt werden (Pinobanksin, Quercetin, Lorente F (1991) An HPLC technique for flav-
onoid analysis in honey. J Sci Food Agric 56,
Luteolin, Quercetin-3-Methyl-Äther, 8- 49-56
Methoxykaempferol, Kaempferol, Apige-
Ferreres F, Ortiz A, Silva C, García-Viguera C,
nin, Kaempferol 3-Methyl-Äther, Isorham-
Tómas-Barberán FA, Tomás-Lorente F
netin, Quercetin 3,3’-Dimethyl-Äther, Pino-
(1992) Flavonoids from La Alcarria honey. A
cembrin, Quercetin 7,3’-Dimethyl-Äther, study of their botanical origin. Z Lebensmitt
Quercetin 3,7-Dimethyl-Äther, Luteolin 7- Unters Forsch 194, 139-143
Methyl-Äther, Chrysin, Galangin, Genkwa- Garcia-Viguera C (1991) Caracterización de ali-
nin und Die Ge-
Tectochrysin) (Tabelle I). mentos vegetales mediante análisis de sus
samtmenge der Flavonoide in diesen Ho- polifenoles por HPLC. Doc thesis, Univ Mur-
nigproben schwankte zwischen 5 und 20 cia, Spain
μg per Gramm Honig (Tabelle II). Die Fla- Ribeiro-Campos MG, Sabatier S, Amiot MJ, Au-
vanone Pinobanksin und Pinocembrin und bert S (1990) Characterization of flavonoids
die Flavone Chrysin waren die Hauptantei- in three hive products: bee-pollen, propolis
le unter den gefundenen Flavonoiden. Im and honey. Planta Med 56, 580

allgemeinen enthielten alle Proben ein Sabatier S, Amiot MJ, Tacchini M, Aubert S
ähnliches Flavonoidmuster mit nur gerin- (1992) Identification of flavonoids in sunflow-
er honey. J Food Sci 57, 773-774
gen Unterschieden bei den Flavonoiden,
die aus dem Nektar und Pollen stammen. Tomás-Barberán FA, Blázquez MA, García-
Viguera C, Ferreres F, Tomás-Lorente F
(1992) A comparative study of different Am-
Honig-Flavonoide / botanische Herkunft berlite XAD resins in flavonoids analysis.
/ geographische Herkunft / HPLC Phytochem Anal 3, 178-181
Tomás-Barberán FA, Ferreres F, García-
Viguera C, Tomás-Lorente F (1993a) Flavo-
REFERENCES noids in honey of different geographical ori-
gin. Z Lebensmitt Unters Forsch 196, 38-44
Amiot MJ, Aubert S, Gonnet M, Tacchini M Tomás-Barberán FA, Ferreres F, Tomás-Lorente
(1989) Les composés phénoliques des F, Ortiz A (1993b) Flavonoids from Apis mellif-
miels : étude préliminaire sur l’identification era beeswax. Z Naturforsch 48C, (1-2), 68-72
et la quantification par familles. Apidologie Wollenweber E, Asakawa Y, Schillo D, Leh-
20, 115-125 mann V, Weigel H (1987) A novel caffeic acid

Bogdanov S (1984) Characterisation of antibac- derivative and other constituents of Populus

terial substances in honey. Lebensmitt Wiss bud excretion and propolis (bee-glue).
Technol 17, 74-76 Z Naturforsch 42C, 1030-1034

View publication stats