Food Chemistry 81 (2003) 159168

www.elsevier.com/locate/foodchem

Flavonoids, phenolic acids and abscisic acid in Australian

and New Zealand Leptospermum honeys
Lihu Yaoa, Nivedita Dattaa,*, Francisco A. Tomas-Barberanb, Federico Ferreresb,
Isabel Martosb, Riantong Singanusongc
a

Food Science and Technology, School of Land and Food Sciences, The University of Queensland, Gatton, Queensland 4343, Australia
b
Laboratorio de Fitoqumica, Department of Food Science and Technology, CEBAS (CSIC), PO Box 4195, Murcia 30080, Spain
c
Department of Agro-Industry, Faculty of Agriculture, Natural Resources and Environmental Sciences, Naresuan University,
Muang, Phitsanulok 65000, Thailand
Received 27 June 2002; accepted 29 July 2002

Abstract
Flavonoids, phenolic acids and abscisic acid of Australian and New Zealand Leptospermum honeys were analyzed by HPLC.
Fifteen avonoids were isolated in Australian jelly bush honey (Leptospermum polygalifolium), with an average content of 2.22 mg/
100 g honey. Myricetin (3,5,7,30 ,40 ,50 -hexahydroxyavone), luteolin (5,7,30 ,40 -tetrahydroxyavone) and tricetin (5,7,30 ,40 ,50 -pentahydroxyavone) were the main avonoids identied. The mean content of total phenolic acids in jelly bush honey was 5.14 mg/100
g honey, with gallic and coumaric acids as the potential phenolic acids. Abscisic acid was quantied as twice the amount (11.6 mg/
100 g honey) of the phenolic acids in this honey. The avonoid prole mainly consisted of quercetin (3,5,7,30 ,40 -pentahydroxyavone), isorhamnetin (3,5,7,40 -tetrahydroxyavone 30 -methyl ethyl), chrysin (5,7-dihydroxyavone), luteolin and an unknown
avanone in New Zealand manuka (Leptospermum scoparium) honey with an average content of total avonoids of 3.06 mg/100 g
honey. The content of total phenolic acids was up to 14.0 mg/100 g honey, with gallic acid as the main component. A substantial
quantity (32.8 mg/100 g honey) of abscisic acid was present in manuka honey. These results showed that avonoids and phenolic
acids could be used for authenticating honey oral origins, and abscisic acid may aid in this authentication.
# 2002 Published by Elsevier Science Ltd.
Keywords: Honey; Leptospermum; Botanical origin; Flavonoids; Phenolic acids; Floral markers; Abscisic acid

1. Introduction
The quality of honey is judged by the botanical or
oral origin and chemical composition (Cherchi, Spanedda, Tuberoso, & Cabras, 1994) and price of honey is
based on its quality and, hence, its oral origin
(Andrade, Ferreres, & Amaral, 1997). Traditionally, the
oral source of a honey has been identied by the analysis of bee pollens present in the honey. However, Tan,
Wilkins, Molan, Holland, and Reid (1989) suggested
that chemical approaches might be more accurate and
easily undertaken in the characterisation of the oral

* Corresponding author. Tel.: +61-754-601017; fax: +61-754601171.

E-mail address: [email protected] (N. Datta).

source of a honey. The use of phenolic compound analysis, including avonoids, in the identication of
honeys has been suggested (Amiot, Aubert, Gonnet, &
Tacchini, 1989) and has since been used as a tool for
studying the oral and geographical origins of honeys.
Earlier, researchers tried to use the analysis of amino
acids (Bosi & Battaglini, 1978; Davies, 1976; Davies &
Harris, 1982) to complement pollen analysis in the
determination of the oral origins of honey. In recent
years, volatile compounds (Bonaga, Giumanini, &
Gliozzi, 1986; DArcy, Rintoul, Rowland, & Blackman,
1997), degradation products of phenylalanine (Speer &
Montag, 1987), aromatic acids and their esters (Steeg &
Montag, 1988a, 1988b, 1988c), aromatic and degraded
carotenoid-like substances (Tan, Wilkins, Holland, &
McGhie, 1989, 1990; Tan, Wilkins, Molan et al., 1989;
Wilkins, Lu, & Molan, 1993) and aromatic aldehydes
and heterocycles (Hausler & Montag, 1989, 1990, 1991)

0308-8146/02/$ - see front matter # 2002 Published by Elsevier Science Ltd.

PII: S0308-8146(02)00388-6

160

L. Yao et al. / Food Chemistry 81 (2003) 159168

have been analysed using gas chromatography connected

with mass spectroscopy (GCMS) for the assessment of
oral origins of honeys. The objective studies have been
extended to better HPLC methods for the analysis of
non-volatile phytochemical constituents of honeys, such
as phenolic compounds (Amiot et al., 1989; Ferreres,
Andrade, & Tomas-Barberan, 1994a; Ferreres, GarciaViguera, Tomas-Lorente, & Tomas-Barberan, 1993;
Ferreres, Giner, & Tomas-Barberan, 1994; Sabatier,
Amoit, Tacchini, & Aubert, 1992). These HPLC analyses have proven to be very useful in determining the
authenticity of the oral origin of honey (Andrade,
Ferreres, & Amaral, 1997; Andrade, Ferreres, Gil, &
Tomas-Barberan, 1997b; Martos, Ferreres, & TomasBarberan 2000a; Martos, Ferreres, Yao, DArcy, Can,
& Tomas-Barberan, 2000b).
Analysis of phenolic compounds has been regarded as
a very promising technique for studying the oral and
geographical origins of honeys (Amiot et al., 1989; Ferreres, Ortiz, Silva, Garcia-Viguera, Tomas-Barberan, &
Tomas-Lorente, 1992; Sabatier et al., 1992; TomasBarberan, Ferreres, Garcia-Viguera, & Tomas-Lorente,
1993a; Tomas-Barberan, Garcia-Viguera, Vit-Olivier,
Ferreres, & Tomas-Lorente, 1993). In these studies, the
avanone hesperetin has been used as a marker for
citrus honey (Ferreres et al., 1993; Ferreres, Blazquez,
Gil, & Tomas-Barberan 1994; Ferreres, Giner, &
Tomas-Barberan, 1994); the avone, kaempferol, for
rosemary honey (Ferreres, Blazquez et al., 1994b; Ferreres, Juan, Perez-Arquillue, Herrera-Marteache, Garcia-Viguera, Tomas-Barberan, 1998), and quercetin for
sunower honey (Tomas-Barberan, Martos, Ferreres,
Radovic, & Anklam, 2001). Moreover, some phenolic
acids such as ellagic acid in heather honey have also
been used as oral markers (Andrade, Ferreres, &
Amaral, 1997; Ferreres, Andrade, Gil, & Tomas-Barberan, 1996; Ferreres, Andrade, & Tomas-Barberan,
1996), and the hydroxycinnamates (caeic, p-coumaric
and ferulic acids) in chestnut honey (Andrade, Ferreres,
& Amaral, 1997). Pinocembrin, pinobanksin and chrysin are the characteristic avonoids of propolis, and
these avonoid compounds have been found in most
European honey samples (Tomas-Barberan et al., 2001).
In some honey samples, such as lavender and acacia
honeys, no specic phenolic compounds have been
found suitable as oral markers (Tomas-Barberan et al.,
2001). Other possible phytochemicals markers may be
found, such as abscisic acid for heather honey (Ferreres,
Andrade, & Tomas-Barberan, 1996). Abscisic acid has
also been detected in rapeseed, limetree and acacia
honeys (Tomas-Barberan et al., 2001).
This study employs the HPLC analyses of avonoids
and phenolic acids, along with abscisic acid, in Australian and New Zealand Leptospermum honeys, and
correlates these constituents with the honey oral
origins as potential biochemical markers.

2. Materials and methods

2.1. General
All chemicals used were of HPLC grade. Pinocembrin, chrysin and quercetin were purchased from Extrasynthese, 69726 Genay Cedex, France; kaempferol,
kaempferol 8-methyl ether, isorhamnetin, and luteolin
were kindly provided by the Laboratory of Phytochemistry, Department of Food Science and Technology,
CEBAS(CSIC), Murcia, Spain; gallic, coumaric and
ellagic acids were purchased from Sigma Chemicals Co.,
St Louis, MO, USA; ferulic acid was purchased from
Aldrich Chemicals Co., Milwaukee, WI, USA; abscisic
acid was kindly provided by the Laboratory of Plant
Physiology, Department of Botany, The University of
Queensland, Brisbane, Australia. Amberlite XAD-2 was
from Supelco, Bellefonte, PA, USA and the 0.45 mm
membrane lter was supplied by Alltech Associates
(Aust) Pty. Ltd., Baulkham Hills, NSW, Australia.
2.2. Honey samples
During the 19971998 and 19981999 owering seasons, 12 jelly bush (Leptospermum polygalifolium) honey
samples were collected and supplied by individual Australian apiarists from the North Coast, New South
Wales (NSW). Two manuka (Leptospermum scoparium)
honey samples were sourced from New Zealand. All the
samples were stored in a freezer of 18 to 24  C
before analysis. Details of these Leptospermum honey
samples used in this experiment are described in Table 1.
2.3. Extraction (column chromatography)
Extraction was performed according to methods
described previously (Martos, Cossentini, Ferreres, &
Tomas-Barberan 1997; Martos, Ferreres, & TomasBarberan, 2000; Martos, Ferreres, Yao et al., 2000).
Namely, honey samples (100 g each) were thoroughly
mixed with ve parts (500 ml) of distilled water, adjusted to pH 2 with concentrated HCl, until completely
uid by stirring with a magnetic stirrer at room temperature. The uid samples were then ltered through
cotton wool to remove solid particles. The ltrate was
mixed with 150 g Amberlite XAD-2 (pore size 9 nm,
particle size 0.31.2 mm) and stirred in a magnetic stirrer for 10 min, which was considered enough to absorb
honey phenolics with a recovery rate more than 80%
(Martos et al., 1997; Tomas-Barberan, Blazquez, Garcia-Viguera, Ferreres, & Tomas-Lorente, 1992). The
Amberlite particles were then packed in a glass column
(423.2 cm) and the column was washed with acidied
water (pH 2 with HCl, 250 ml) and subsequently rinsed
with distilled water (300 ml) to remove all sugars and
other polar constituents of honey. The phenolic

161

L. Yao et al. / Food Chemistry 81 (2003) 159168

Table 1
Unioral Leptospermum honey samples analysed in this study
Sample code

Common name

Floral origin

Year

Geographical origin

JB1052
JB1291
JB0041
JB0539
JB5367
JB0481
JB5364
JB0566
JB2174
JB1053
JB0140
JB0484
JB040
MIA01
MA01

Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Jelly bush
Manukab
Manukab

L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. polygalifolium
L. scoparium
L. scoparium

1997
1997
1997
1997
1998
1997
1998
1997
1998
1997
1997
1997

North Coast, NSW

North Coast, NSW
North Coast, NSW
North Coast, NSW
North Coast, NSW
North Coast, NSW
North Coast, NSW
North Coast, NSW
North Coast, NSW
North Coast, NSW
North Coast, NSW
North Coast, NSW

1998
1998

New Zealand
New Zealand

a
b

Date and location could not be conrmed.

Supplied by Honey Research Unit, Department of Biological Sciences, University of Waikato, New Zealand.

compounds remained absorbed on the column (Ferreres, Tomas-Barberan, Gil, & Tomas-Lorente, 1991)
and were eluted with methanol (400 ml). The methanolic extract was concentrated to dryness under reduced
pressure in a rotary evaporator at 40  C. The residue
was resuspended in distilled water (5 ml) and extracted
with diethyl ether (5 ml3). The ether extracts were
combined, and the diethyl ether was removed by ushing with nitrogen. The dried residue was then redissolved in 1 ml of methanol, ltered through a 0.45 mm
membrane lter, and analysed by HPLC.
2.4. HPLC analysis
The HPLC analyses of avonoids and other phenolic
acids were performed by using a Shimadzu Class-VP
HPLC system, a computer-controlled system with an
upgraded Class-VP 5.03 software. Separations were
carried out on a reversed phase column LiChroCART
RP-18 (Merck, Darmstadt, Germany; 12.5 cm0.4 cm,
particle size 5 mm), using a mobile phase of 5% (v/v)
aqueous formic acid (solvent A) and methanol (solvent
B) at a constant solvent ow rate of 1 ml/min. The
temperature of the column oven was set at 35  C.
The gradient elution was established according to the
method of Martos et al. (1997): 30% methanol (B) owed through the column isocratically with solvent A for
15 min; this was increased to 40% methanol at 20 min,
45% methanol at 30 min, 60% methanol at 50 min,
80% methanol at 52 min, and 90% methanol at 60 min.
Finally, isocratic elution with 90% methanol was done
until 65 min.
The honey extracts were injected with a SIL-10A XL
Auto Injector and the avonoids were detected using a

multichannel photodiode-array detector (SPD-M10A

VP) to obtain the UV spectra of the various phenolic
compounds. The chromatograms were monitored at 290
nm and 340 nm, since the majority of the honey avonoids and phenolic acids show their UV absorption
maxima around these two wavelengths (Martos et al.,
1997).
2.5. Identication and quantication of avonoids
The phenolic compounds (avonoids) were identied
and quantied as reported previously (Martos et al.,
1997; Martos, Ferreres, & Tomas-Berberan, 2000;
Martos, Ferreres, Yao et al., 2000), with reference to
their standard compounds. For some of the minor
honey avonoids and phenolic acids, when their
authentic compounds were unavailable, the stored UV
spectra extracted from the same HPLC methods for
honey analysis and their corresponding retention times,
were utilised as library data for the comparison, and for
identication.
The avonoids were quantied by their absorbance in
the HPLC chromatograms against external standards,
and four standard avonoids were used: pinocembrin at
290 nm (for avanone), chrysin at 340 nm (for avones
with unsubstituted ring B), kaempferol at 340 nm (for
avone kaempferol and its methyl ether), quercetin at
340 nm (for the rest of avones).
The phenolic acids such as gallic and coumaric acids,
were quantied against their standards at 290 nm.
Ellagic acid was quantied against its standards at 340
nm. The abscisic acid, both of trans,trans- and cis,
trans- isomers, was determined against the standard at
290 nm.

162

L. Yao et al. / Food Chemistry 81 (2003) 159168

compounds. These compounds were labelled F01 and

F02 (Fig. 2).
Myricetin, luteolin and tricetin represented 27.2, 22.3
and 20.5% of the total avonoids, respectively (Table 2)
and pinobanksin, the unknown avonoid (F01), and
quercetin are the secondary avonoids, representing
14.3, 12.6, and 11.5% of the total avonoids, respectively (Table 2). Tricetin levels were variable amongst
the samples studied; most of the samples did not contain
this avonoid. However, most of the samples show a
common avonoid prole, comprising myricetin, quercetin, luteolin, kaempferol and the unknown avonoid
(F01), suggesting that they could be used as oral markers for Australian jelly bush honeys.
The main similarity between Australian jelly bush
honeys and Eucalyptus honeys (Martos, Ferreres, Yao
et al., 2000) lies in the fact that they contain quercetin,
luteolin and kaempferol in their avonoid proles. The
main dierence between these two types of honeys is
that the avonoid proles in most of the jelly bush
honeys are devoid of tricetin, instead, they have the
unknown avonoid (F01) as one of the main components in their avonoid prole. Thus, the jelly bush

3. Results and discussion

3.1. Flavonoid proles of Leptospermum honeys
3.1.1. Flavonoids in Australian jelly bush (L.
polygalifolium) honey
HPLC analysis of avonoids in samples of Australian
jelly bush honey showed that myricetin and luteolin are
main avonoids present in most of the samples analysed
(Fig. 1a and Table 2). About 15 avonoids have been
found in this type of honey and most of them occur in
small amounts. The content of total avonoids was 2.22
mg/100 g honey, with myricetin being 0.37 mg/100 g
honey (Table 2).
There were two minor compounds present in the jelly
bush honey that showed similar avonoid spectra and
chromatographic behaviour but could not be identied,
due to lack of availability of authentic compounds and
literature data. These compounds had retention times of
24 and 30 min, respectively, with UV spectra (Fig. 2)
similar to pinocembrin and pinobanksin (or the derivatives), where the major absorption peaks occur at ca 283
and ca 270 nm, respectively, suggesting avanone

Table 2
Flavonoid content of Australian jelly bush (Leptospermum polygalifolium) and New Zealand manuka (Leptospermum scoparium) honeys
Sample

Content of avonoids (mg/100 g honey)a

Myr

JB1052
JB1291
JB0041
JB0539
JB5367
JB0481
JB5364
JB0566
JB2174
JB1053
JB0140
JB0484
JB040
Mean
S.D.
%b
MIA01
MA01
Mean
S.D.
%b

0.58
0.22
0.08
0.25
0.12
0.74
0.06

Tri
0.68

0.14
0.08
0.07

0.33
0.22
0.11
0.10
0.85
1.15
0.37
0.36
27.2
0.14
0.07
3.0

PB

0.06
0.22

F01
0.41
0.37
0.01

F02

0.07
0.36
0.24

0.16

0.26
0.26
20.5

0.05
0.15
0.12
14.3

0.07
0.22
0.15
12.6
0.38
0.39
0.39
0.01
12.7

Lut

0.08
0.43
0.07
0.13
0.06
0.17

0.24

0.07
0.23
0.37
0.04

Que

0.11
0.12
0.06

0.36
0.15
0.12
8.1
0.25
0.12
3.7

0.13
0.09
0.11
0.14
0.21
0.04
0.14
0.11
11.5
0.55
0.31
0.43
0.17
13.8

0.45
0.08
0.36
0.06
0.41
0.03
0.23
0.21
0.22
0.26
0.57
0.26
0.17
22.3
0.33
0.43
0.38
0.07
12.6

3MQ

Kae

8MK

PC

DMQ

IRM

0.19
0.33
0.03

0.13
0.10
0.01
0.12
0.01

0.13
0.06
0.09

0.10
0.06

0.12

0.08

Chr

0.04

Tch

0.23

0.07
0.05
0.06
0.03
0.02

0.04
0.04
0.08
0.03
0.02

0.09
0.10
0.11
5.6
0.13
0.18
0.16
0.03
5.2

0.05
0.06
0.04
6.5
0.13
0.17
0.15
0.02
5.0

0.06
0.10

0.05

0.02

0.03

0.09
0.08
0.04
6.3
0.26
0.13
0.20
0.09
6.3

0.26
0.12
0.09
5.4
0.23
0.15
0.19
0.05
6.2

0.09
0.07
0.04
4.3
0.21
0.18
0.19
0.02
6.4

0.05
0.11

0.06
0.07
0.03
4.0
0.47
0.32
0.40
0.10
12.9

0.09

0.05

0.12

2.8
0.40
0.37
0.38
0.02
12.6

5.5

Total
2.30
2.23
0.51
1.23
0.37
1.85
0.14
1.20
1.67
1.15
0.78
1.62
2.32

3.34
2.77

a
Myrmyricetin (3,5,7,30 ,40 ,50 -hexahydroxyavone), Tritricetin (5,7,30 ,40 ,50 -pentahydroxyavone), Quequercetin (3,5,7,30 ,40 -pentahydroxyavone), Lutluteolin (5,7,30 ,40 -tetrahydroxyavone), 3MQquercetin 3-methyl ether (5,7,30 ,40 -tetrahydroxy-3-methoxyavone), Kaekaempferol (3,5,7,40 -tetrahydroxyavone), 8MKkaempferol 8-methyl ether (3,5,7,40 -tetrahydroxy-8-methoxyavone), PCpinocembrin (5,7dihydroxyavanone), DMQquercetin 3,30 -dimethyl ether (5,7,40 -trihydroxy-3,30 -dimethoxyavone), IRMisorhamnetin (3,5,7,40 -tetrahydroxy30 -methoxyavone), Chrchrysin (5,7-dihydroxyavone), PBpinobanksin (3,5,7-trihydroxyavanone), Tchtectochrysin (5-hydroxy-7-methoxyavone), F01unknown avonoid 01, F02unknown avonoid 02.
b
Percentage of each individual avonoid in the total avonoids.

L. Yao et al. / Food Chemistry 81 (2003) 159168

honeys can be readily dierentiated from Australian

Eucalyptus honeys by their avonoid proles.
3.1.2. Flavonoids in New Zealand manuka
(L. scoparium) honey
In New Zealand manuka honeys, the avonoid prole
mainly consisted of quercetin (13.8%), isorhamnetin
(12.9%), unknown avonoid (F01) (12.7%), chrysin
(12.6%) and luteolin (12.6%), together representing
64.6% of the total avonoids (3.06 mg/100 g honey) in
this honey (Fig. 1b and Table 2). Since only two manuka honey samples were analysed, it is necessary to
analyse more samples to see whether this avonoid
prole could be used as a oral marker for this honey
species.
The main similarity between Australian jelly bush
honey and New Zealand manuka honey is that both
honeys contain similar and relatively stable avonoid
proles comprising quercetin, luteolin, quercetin 3methyl ether, kaempferol and an unknown avonoid
(F01) (Table 2). This result indicated that this similar
avonoid prole could be used as a oral marker for
Leptospermum honeys produced in the Oceanic areas.

163

The main dierence between these honeys is that

Australian jelly bush honey showed total avonoids as
2.22 mg/100 g honey, i.e. lower than that found in New
Zealand manuka honey (3.06 mg/100 g).
Weston, Mitchell, and Allen (1999) and Weston,
Brocklebank, and Lu (2000) identied pinobanksin,
pinocembrin, chrysin and galangin in New Zealand
manuka honeys at a level of (totally) about 0.01 mg/100
g honey, which was much lower than the levels found
for pinocembrin (0.19 mg/100 g honey) and chrysin
(0.38 mg/100 g honey) in the manuka honey samples
examined in this study (Table 2). The dierence in the
avonoid contents of manuka honey measured between
the present study and the studies by Weston et al. (1999,
2000) may be due to the dierences of the extraction
methods. In the method described by Weston et al.
(1999, 2000), only 5460 g of Amberlite XAD-2 was
used for 150236 g honey in the column chromatography, which could result in an incomplete extraction
of honey avonoids. Martos et al. (1997) advocated that
at least 100 g Amberlite XAD-2 should be used for the
extraction of avonoids from 100 g honey. In the present study, 150 g of Amberlite XAD-2 was applied to the

Fig. 1. HPLC chromatograms of avonoids in Leptospermum honeys (340 nm): (a) Australian jelly bush (L. polygalifolium) honey; (b) New Zealand
manuka (L. scoparium) honey. Flavonoids are: (1) myricetin, (2) tricetin, (3) quercetin, (4) luteolin, (5) kaempferol, (6) kaempferol 8-methyl ether,
(7) pinocembrin, and (8) chrysin.

164

L. Yao et al. / Food Chemistry 81 (2003) 159168

Fig. 2. Overlaid UV spectra of the unknown avonoids in Leptospermum honeys.

column chromatography for 100 g of honey, which

resulted in a similar analytical outcome to those analyses of other honeys by Martos et al. (1997), Martos,
Ferreres, and Tomas-Barberan (2000) and Martos,
Ferreres, and Yao et al. (2000).
Moreover, Weston et al. (2000) used the 280 nm
wavelength for identication and quantication of
avonoids by an HPLC method, whereas the present
methods, as well as other studies by Ferreres, TomasBarberan, Soler, Garcia-Viguera, Ortiz, and TomasLorente (1994d); Ferreres, Andrade, Gil, and TomasBarberan (1996) and Martos et al. (1997), Martos, Ferreres, and Tomas-Barberan (2000) and Martos, Ferreres, and Yao et al. (2000) quantied the avanones at
290 nm and the avones at 340 nm, and the identication of avonoids was based on both UV spectra and
retention time in the HPLC with photodiode array
detection. Therefore, further analysis of New Zealand
manuka honey, using a larger number of samples with
the same method under similar analytical conditions, is
required for selection of an individual avonoid or a avonoid prole as oral markers for this type of honey.
3.2. Phenolic acids in Leptospermum honeys
3.2.1. Phenolic acids in Australian jelly bush (L.
polygalifolium) honey
In the samples of Australian jelly bush (L. polygalifolium) honey, the content of total phenolic acids
averaged 5.14 mg/100 g honey, with gallic acid (23.6%)
and coumaric acid (22.2%) as the main components
(1.23 and 1.36 mg/100 g honey, respectively) (Table 3
and Fig. 3). In addition, ellagic and chlorogenic acids
were the secondary compounds (14.7 and 12.0%,
respectively). The other phenolic acids present in this
honey are relatively lower in proportion, ranging from
5.7% (ferulic acid) to 10.7% (an unknown phenolic

acid, labelled as Ph1) of total phenolic acids. No similar

phenolic prole was found among the honeys analysed
except for Australian blue top ironbark honey (Yao et
al., in preparation). However, blue top ironbark honey
has a much higher percentage of gallic, coumaric, ellagic
and chlorogenic acids. In addition, the contents of
minor phenolic acids are much lower in blue top ironbark honey than in jelly bush honey.
Table 3
Phenolic acids in unioral jelly bush and manuka honeys
Sample Phenolic acids (mg/100 g honey)a
GA

CA

Ph01 Caf

Cou

JB1052 1.65 0.36 2.14 0.77 2.50

JB1291 0.70 0.32 0.38 0.83 3.44
JB0041 0.38 0.64 0.06 0.17 0.33
JB0539 4.51 1.11 0.84 0.57 0.88
JB5367 0.58 0.05 0.26 0.36 0.13
JB0481 0.89 0.24 0.40 0.49 0.39
JB5364 0.08 0.07 0.06 0.03 0.09
JB0566 1.54
0.59 0.65 3.15
JB2174 2.78 0.58 0.96 0.62 4.74
JB1053 1.18 0.16 0.35 0.49 0.51
JB0140 0.09 0.16 0.08 0.05 0.05
JB0484 1.37 0.85 0.50 0.22 0.54
JB040
0.28 0.43 0.23 0.36 0.96
Mean
1.23 0.41 0.53 0.43 1.36
S.D.
1.24 0.33 0.56 0.26 1.55
%b
23.6 12.0 10.7
9.70 22.2
MIA01 5.74 0.88 2.24 1.84 0.50
MA01
8.36 0.77 0.99 1.01 1.68
Mean
7.05 0.82 1.62 1.43 1.09
S.D.
1.86 0.08 0.88 0.59 0.84
%b
50.0
5.93 11.9 10.4
7.55

Fer

Ell

0.59
0.49
0.10
0.13
0.27
0.54
0.04
0.31
0.53
0.04

1.18
0.94
0.46
0.83
0.32
0.64
0.08
0.33
1.37
0.24
0.17
0.58
0.58
0.59
0.39
14.7
1.54
1.47
1.50
0.05
10.8

0.12
0.17
0.28
0.21
5.69
0.39
0.57
0.48
0.13
3.39

Syr
0.46

0.18

0.33
0.25

0.29
0.30
0.10
6.05

Total
9.20
7.55
2.14
8.87
1.97
3.76
0.44
6.58
11.9
3.22
0.59
4.18
3.31

13.1
14.9

a
GAgallic acid, CAchlorogenic acid, Cafcaeic acid, Cou
coumaric acid, Ferferulic acid, Ellellagic acid, Syrsyringic acid,
Ph1unknown phenolic acid.
b
Percentage of each individual phenolic acid in the total phenolic
acids.

L. Yao et al. / Food Chemistry 81 (2003) 159168

165

Fig. 3. HPLC chromatograms of phenolic acids and abscisic acid in uniformal Leptospermum honeys (290 nm): (a) Australian jelly bush (L. polygalifolium) honey; (b) New Zealand manuka (L. scoparium) honey. Phenolic acids are: (1) gallic acid, (2) chlorogenic acid, (3) caeic acid, (4)
p-coumaric acid, (5) ferulic acid, and (7) ellagic acid. Abscisic acids are: (8) trans,trans- and (9) cis,trans-abscisic acids.

3.2.2. Phenolic acids in New Zealand manuka (L.

scoparium) honey
The content of total phenolic acids in New Zealand
manuka (L. scoparium) honey was up to 14.0 mg/100 g
honey, the highest amount of total phenolic acids measured in the honeys in this study so far (Yao, 2002). Of
these phenolic acids, gallic acid was the highest phenolic
acid detected, ranging 5.748.36 mg/100 g honey, with
an average of 7.05 mg/100 g honey, representing 50.0%
of total phenolic acids in manuka honey (Table 3). This
level is the second highest among the oral types of
honey studied, with only Australian yellow box honey
having higher levels of gallic acid (Yao et al., in preparation). The unknown phenolic acid (Ph1), ellagic
acid and caeic acid were present in much smaller percentages in the phenolic proles (11.9, 10.8, and 10.4%,
respectively), and are considered as secondary phenolic
acids in this honey. This phenolic prole, dominated by
gallic acid, could be used as a oral marker for New
Zealand manuka honey (Fig. 3 and Table 3).

The main dierences between Australian and New

Zealand Leptospermum honeys are the levels and percentage of total phenolic acids and of gallic acid (much
higher in manuka honey) and coumaric acid (lower in
manuka honey); these may be used to dierentiate
Australian jelly bush honey from New Zealand manuka
honey.
Due to the dierences in analytical methods discussed
earlier, the levels of phenolic acids measured in New
Zealand manuka honey by Weston et al. (1999, 2000)
below than 0.01 mg/100 g honey while, in heather honey
there was benzoic acid alone measured at a level of 0.39
mg/100 g honey (Weston et al., 2000). In this study,
much larger amounts of phenolic acids were detected in
Australian jelly bush honey (5.14 mg/100 g honey) and
New Zealand manuka honey (14.0 mg/100 g honey).
Again, the smaller quantities of phenolic acids detected
in New Zealand manuka honey, in the previous studies,
may be due to their incomplete extraction. Thus, further
investigation on more samples of New Zealand manuka

166

L. Yao et al. / Food Chemistry 81 (2003) 159168

honey is required to conrm the use of phenolic proles

as oral markers for this type of honey. Furthermore,
because the elution of gallic acid from the column is
close to the solvent front (in this HPLC analytical
method) more specied studies are also required to
conrm the gallic acid as one of the biochemical
markers for authentication of the Leptospermum honeys.
3.3. Abscisic acid in Leptospermum honeys
Abscisic acid has been found in both Australian and
New Zealand Leptospermum honey species in large
amounts, with much higher concentrations than the
total phenolic acids (Tables 3 and 4). In Australian jelly
bush (L. polygalifolium) honey, the mean level of
trans,trans-abscisic acid is 7.83 mg/100 g honey, while
the level of its isomer, cis,trans-abscisic acid is 3.76 mg/
100 g honey (Table 4). However, in New Zealand manuka (Leptospermum scoparium) honey, the mean level of
trans,trans-abscisic acid is 31.0 mg/100 g honey, with
the cis,trans-abscisic acid at a level of 1.87 mg/100 g
honey (Table 4). Thus, the level of abscisic acid in New
Zealand manuka honey is much higher than that in
Australian jelly bush honey on that found in Portuguese
heather (Erica spp.) honey (16.6 mg/100 g honey)
(Ferreres, Andrade, & Tomas-Barberan, 1996).
In Fig. 3, the HPLC chromatograms of Australian
and New Zealand Leptospermum honeys are dominated
by gallic and abscisic acids. Thus, the high percentages
and concentrations of these two phytochemical components could be used to dierentiate Leptospermum
honeys from the other types of honeys.
Previously, in New Zealand manuka honey, the level
of cis,trans-abscisic acid was found only at a level of ca.
0.02 mg/100 g honey, while the trans,trans-abscisic acid
was 0.07 mg/100 g honey (Lipp, 1990). These levels are
much lower than those found in this study for two
manuka honey samples (Table 4). In New Zealand willow (Salix spp.) honey, abscisic acid isomers were found
to be as high as 14.8 mg/100 g honey, with cis,transabscisic acid being at a level of 10.6 mg/100 g honey and
trans,trans-abscisic acid being at a level of 4.2 mg/100 g
honey (Tan et al., 1990). In Australian leatherwood
(Eucryphia lucida) honey, trans,trans-abscisic acid was

Table 4
Abscisic acid in unioral jelly bush and manuka honeys
Sample

Jelly bush honey

Manuka honey
a

Abscisic acid (mg/100 g honey)a

trans,trans-ABA

cis,trans-ABA

Total

7.83
31.0

3.76
1.87

11.6
32.8

ABAabscisic acid.

present at a level of 0.65 mg/100 g honey, i.e. 6 times

more than the level of its isomer cis,trans-abscisic acid
(0.10 mg/100 g honey) (Lipp, 1990; Sun, 1995). In
addition, both isomers of abscisic acid have been found
in various European honeys (Tomas-Barberan et al.,
2001) with total amounts as follows: Portuguese heather
(Erica spp.) honey, 0.41.8 mg/100 g honey; rapeseed
(Brassica spp.) honey, 0.030.25 mg/100 g honey; lime
tree (Tilia europaea) honey, 0.050.5 mg/100 g honey,
and acacia (Robinia pseudoacacia) honey, 0.100.25 mg/
100 g honey. Again, the largest amount of these isomers
were reported, ranging from 2.5 to 16.6 mg/100 g honey
in Portuguese heather honey (Ferreres, Andrade, &
Tomas-Barberan, 1996). These results suggest that the
isomers of abscisic acid are present at variable levels in
dierent types of honey. Therefore, the occurrence (and
the levels) of abscisic acid could be used as a biochemical marker for aiding in the authentication of oral
origin of honey (Tomas-Barberan et al., 2001).
Abscisic acid could be used alone as a oral marker
for New Zealand manuka honey. This is because, in
Australian jelly bush honey, the content of abscisic acid
is less than the amount present in New Zealand manuka
honey (ranging 31.434.3 with an average of 32.8 mg/
100 g honey). This makes their dierentiation possible.
Moreover, abscisic acid has been found in quite large
amounts in heather honey (2.516.6 mg/100 g honey)
(Ferreres, Andrade, & Tomas-Barberan, 1996; TomasBarberan et al., 2001), but it is still much below that
found in New Zealand manuka honey. Thus, abscisic
acid may be suitable for authenticating the oral origin
of the honey species alone. Further studies on more
New Zealand manuka honey samples are necessary to
conrm its use as a oral marker.

4. Conclusion
The Leptospermum honeys, including Australian jelly
bush and New Zealand manuka honeys, show a common avonoid prole, mainly comprising quercetin,
luteolin, quercetin 3-methyl ether, luteolin and an
unknown avonoid F01. This avonoid prole is
characteristic of these oral types of honeys and thus
could be used as a biochemical marker for the botanical
authentication of these honeys. Chrysin in New Zealand
manuka honey, and myricetin, in jelly bush honey,
could be used to distinguish these Leptospermum oral
types from each other, suggesting a species-specic differentiation occurring between these two Leptospermum
honeys. Gallic acid and abscisic acid are dominant
compounds in New Zealand manuka honey, which
could further assist in the authentication of New Zealand manuka honey. These results show that non-volatile HPLC analysis can be an objective tool for honey
oral authentication.

L. Yao et al. / Food Chemistry 81 (2003) 159168

Acknowledgements
We thank Mr. William G. Winner of Capilano Honey
Ltd, Australia and Dr. Peter Molan, University of
Waikato, New Zealand for the supply of honey samples
and Ms. Katherine Raymont, Ms. Brenda Mossel, and
Mr. Gavin Rintoul of the University of Queensland for
their technical supports and sample sourcing. Appreciation is also expressed to the Rural Industries
Research and Development Corporation (RIRDC,
Australia) for nancial support for this project, and
Department of Education, Science and Training
(DEST, formerly DETYA, Australia) for providing an
IPRS (formerly OPRS) fund support. Our gratitude is
extended to Dr. Bruce DArcy for discussion during the
preliminary preparation of this work.

References
Amoit, M. J., Aubert, S., Gonnet, M., & Tacchini, M. (1989). The
phenolic compounds in honeys: preliminary study upon identication and family quantication. Apidologie, 20(2), 115125.
Andrade, P., Ferreres, F., & Amaral, M. T. (1997). Analysis of honey
phenolic acids by HPLC, its application to honey botanical characterization. Journal of Liquid Chromatography & Related Technologies, 20(14), 22812288.
Andrade, P., Ferreres, F., Gil, M. I., & Tomas-Barberan, F. A. (1997).
Determination of phenolic compounds in honeys with dierent
oral origin by capillary zone electrophoresis. Food Chemistry,
60(1), 7984.
Bonaga, G., Giumanini, A. G., & Gliozzi, G. (1986). Chemical composition of chestnut honey: analysis of the hydrocarbon fraction.
Journal of Agricultural and Food Chemistry, 34(2), 319326.
Bosi, G., & Battaglini, M. (1978). Gas chromatographic analysis of
free and protein amino acids in some unioral honeys. Journal of
Apiculture Research, 17(3), 152166.
Cherchi, A., Spanedda, L., Tuberoso, C., & Cabras, P. (1994). Solidphase extraction and high-performance liquid chromatographic
determination of organic acids in honey. Journal of Chromatography
A, 669, 5964.
DArcy, B. R., Rintoul, G. B., Rowland, C. Y., & Blackman, A. J.
(1997). Composition of Australian honey extractives. 1. Norisoprenoids, monoterpenes, and other natural volatiles from blue gum
(Eucalyptus leucoxylon) and yellow box (Eucalyptus melliodora) honeys. Journal of Agricultural and Food Chemistry, 45(5), 18341843.
Davies, A. M. C. (1976). The application of amino acid analysis to the
determination of the geographical origin of honey. Journal of Food
Technology, 11(5), 515523.
Davies, A. M. C., & Harris, R. G. (1982). Free amino acid analysis of
honeys from England and Wales: application to the determination
of the geographical origin of honeys. Journal of Apiculture Research,
21(3), 168173.
Ferreres, F., Andrade, P., Gil, M. I., & Tomas-Barberan, F. A. (1996).
Floral nectar phenolics as biochemical markers for the botanical
origin of heather honey. Zeitschrift fuer Lebensmittel Untersuchung
und Forschung, 202(1), 4044.
Ferreres, F., Andrade, P., & Tomas-Barberan, F. A. (1994). Flavonoids from Portuguese heather honey. Zeitschrift fuer Lebensmittel
Untersuchung und Forschung, 199, 3237.
Ferreres, F., Andrade, P., & Tomas-Barberan, F. A. (1996). Natural
occurrence of abscisic acid in heather honey and oral nectar.
Journal of Agricultural and Food Chemistry, 44(8), 20532056.

167

Ferreres, F., Blazquez, M. A., Gil, M. I., & Tomas-Barberan, F. A.

(1994). Separation of honey avonoids by micellar electrokinetic
capillary chromatography. Journal of Chromatography A, 669(1/2),
268274.
Ferreres, F., Garcia-Viguera, C., Tomas-Lorente, F., & Tomas-Barberan, F. A. (1993). Hesperetin: a marker of the oral origin of
citrus honey. Journal of the Science of Food and Agriculture, 61,
121123.
Ferreres, F., Giner, J. M., & Tomas-Barberan, F. A. (1994). A comparative study of hesperetin and methyl anthranilate as markers of
the oral origin of citrus honey. Journal of the Science of Food and
Agriculture, 65, 371372.
Ferreres, F., Juan, T., Perez-Arquillue, C., Herrera-Marteache, A.,
Garcia-Viguera, C., & Tomas-Barberan, F. A. (1998). Evaluation of
pollen as a source of kaempferol in rosemary honey. Journal of the
Science of Food and Agriculture, 77, 506510.
Ferreres, F., Ortiz, A., Silva, C., Garcia-Viguera, C., Tomas-Barberan,
F. A., & Tomas-Lorente, F. (1992). Flavonoids of La Alcarria
honeya study of their botanical origin. Zeitschrift-fuer Lebensmittel Untersuchung und Forschung, 194, 139143.
Ferreres, F., Tomas-Barberan, F. A., Gil, M. I., & Tomas-Lorente, F.
(1991). An HPLC technique for avonoid analysis in honey. Journal
of the Science of Food and Agriculture, 56, 4956.
Ferreres, F., Tomas-Barberan, F. A., Soler, C., Garcia-Viguera, C.,
Ortiz, A., & Tomas-Lorente, F. (1994). A simple extractive technique for honey avonoid HPLC analysis. Apidologie, 25, 2130.
Hausler, M., & Montag, A. (1989). Isolation, identication and
quantitative determination of the norisoprenoid (S)-(+)-dehydrovomifoliol in honey. Zeitschrift fuer Lebensmittel Untersuchung
und Forschung, 189, 113115.
Hausler, M., Montag, A.(1990).Minor constituents of honey with
aroma-relevance. III. Nectar ow-specic distribution of aromatic
aldehydes, & presence of nitrogen, sulphur heterocyclic compounds
(1990). Deutsche Lebensmittel Rundschau, 86(6), 171174.
Hausler, M., & Montag, A. (1991). Minor constituents of honey with
aroma relevance. IV. Presence and nectar ow-specic distribution
of the aroma compound precursor (S)-dehydrovomifoliol. Deutsche
Lebensmittel Rundschau, 87(2), 3536.
Lipp, J. (1990). Detection and origin of abscisic acid and proline in
honey. Apidologie, 21(3), 249260.
Martos, I., Cossentini, M., Ferreres, F., & Tomas-Barberan, F. A.
(1997). Flavonoid composition of Tunisian honeys and propolis.
Journal of Agricultural and Food Chemistry, 45(8), 28242829.
Martos, I., Ferreres, F., & Tomas-Barberan, F. A. (2000). Identication of avonoid markers for the botanical origin of Eucalyptus honey. Journal of Agricultural and Food Chemistry, 48,
14981502.
Martos, I., Ferreres, F., Yao, L. H., DArcy, B. R., Can, N., &
Tomas-Barberan, F. A. (2000). Flavonoids in monospecic Eucalyptus honeys from Australia. Journal of Agricultural and Food
Chemistry, 48, 47444748.
Sabatier, S., Amoit, M. J., Tacchini, M., & Aubert, S. (1992). Identication of avonoids in sunower honey. Journal of Food Science,
57(3), 733734.
Speer, K., & Montag, A. (1987). Phenylalanine decomposition products as avour compounds in honey. Deutsche Lebensmittel Rundschau, 83(4), 103107.
Steeg, E., & Montag, A. (1988a). Minor constituents of honey, related
to aroma. 1. Aromatic carboxylic acids and esters. Deutsche
Lebensmittel Rundschau, 84(4), 103108.
Steeg, E., & Montag, A. (1988b). Minor constituents of honey, related
to aroma. 2. Decomposition products of aromatic carboxylic acids,
and glycoside-bound aromatic compounds that contribute to
aroma. Deutsche Lebensmittel Rundschau, 84(5), 147150.
Steeg, E., & Montag, A. (1988c). Quantitative determination of aromatic carboxylic acids in honey. Zeitschrift fuer Lebensmittel
Untersuchung und Forschung, 187, 115120.

168

L. Yao et al. / Food Chemistry 81 (2003) 159168

Sun, Y. (1995). A chemical investigation of some New Zealand native honeys. MPhil thesis, University of Waikato, Hamilton, New Zealand.
Tan, S. T., Wilkins, A. L., Holland, P. T., & McGhie, T. K. (1989).
Extractives from New Zealand unioral honeys. 2. Degraded carotenoids and other substances from heather honey. Journal of Agricultural and Food Chemistry, 37(5), 12171221.
Tan, S. T., Wilkins, A. L., Holland, P. T., & McGhie, T. K. (1990).
Extractives from New Zealand honeys: 3. Unioral thyme and willow honey constituents. Journal of Agricultural and Food Chemistry,
38(9), 18331838.
Tan, S. T., Wilkins, A. L., Molan, P. T., Holland, P. T., & Reid, M. A.
(1989). A chemical approach to the determination of oral sources
of New Zealand honeys. Journal of Apiculture Research, 28, 212
222.
Tomas-Barberan, F. A., Blazquez, M. A., Garcia-Viguera, C., Ferreres, F., & Tomas-Lorent, F. (1992). A comparative study of different amberlite XAD resins in avonoid analysis. Phytochemistry
Analysis, 3, 178181.
Tomas-Barberan, F. A., Ferreres, F., Garcia-Viguera, C., & TomasLorente, F. (1993). Flavonoids in honey of dierent geographical
origin. Zeitschrift fuer Lebensmittel Untersuchung und Forschung,
196, 3844.

Tomas-Barberan, F. A., Garcia-Viguera, C., Vit-Olivier, P., Ferreres,

F., & Tomas-Lorente, F. (1993). Photochemical evidence for the
botanical origin of tropical propolis from Venezuela. Photochemistry, 34(1), 191196.
Tomas-Barberan, F. A., Martos, I., Ferreres, F., Radovic, B. S., &
Anklam, E. (2001). HPLC avonoid proles as markers for the
botanical origin of European unioral honeys. Journal of the
Science of Food and Agriculture, 81(5), 485496.
Weston, R. J., Mitchell, K. R., & Allen, K. L. (1999). Antibacterial
phenolic components of New Zealand manuka honey. Food Chemistry, 64(3), 295300.
Weston, R. J., Brocklebank, L. K., & Lu, Y. R. (2000). Identication
and quantitative levels of antibacterial components of some New
Zealand honeys. Food Chemistry, 70, 427435.
Wilkins, A. L., Lu, Y. R., & Molan, P. C. (1993). Extractable organic
substances from New Zealand unioral manuka (Leptospermum
scoparium) honeys. Journal of Apicultural Research, 32(1), 39.
Yao, L. H. (2002). Flavonoids and other polyphenols in Australian tea
and honey (unpublished data).
Yao, L. H., Datta, N., Tomas-Barberan, F. A., Wansri, R., Ferreres,
F., & Martos, I. An HPLC analysis of avonoids in Australian
Eucalyptus honey (in preparation).