Food Research International 53 (2013) 900–908

Contents lists available at ScienceDirect

Food Research International

journal homepage: www.elsevier.com/locate/foodres

Antioxidant activity of white, green and black tea obtained from the same tea cultivar
Patricia Carloni a, Luca Tiano b, Lucia Padella b, Tiziana Bacchetti c, Chisomo Customu d,
Alexander Kay d, Elisabetta Damiani c,⁎
a
Dipartimento Scienze Agrarie, Alimentari e Ambientali, Università Politecnica delle Marche, Ancona I‐60131, Italy
b
Dipartimento Scienze Cliniche, Specialistiche e Odontostomatologiche, Università Politecnica delle Marche, Ancona I‐60131, Italy
c
Dipartimento Scienze della Vita e dell'Ambiente, Università Politecnica delle Marche, Ancona I-60131, Italy
d
Satemwa Tea Estates, P.O. Box 6, Thyolo, Malawi

a r t i c l e i n f o a b s t r a c t

Article history: The present study explored what effect manufacturing has on the antioxidant properties of teas coming from
Received 17 February 2012 a single cultivar but processed differently to give a white, two black (Orthodox and CTC) and two green
Received in revised form 24 July 2012 (low-caffeine and non-decaffeinated) teas. Total phenol (TPC), flavonoids (TFC), theaflavins, individual cate-
Accepted 26 July 2012
chins content, and chelating activity were also investigated. Using the ABTS, ORAC and LDL assays the follow-
ing ‘antioxidant profile’ was obtained: green ≥ low-caffeine green > white ≥ black Orthodox > black CTC, with
Keywords:
Tea
statistically significant correlation between ORAC and LDL assays (r 2 = 0.444, p = 0.0067), whereas TPC and
Green/black/white tea processing TFC significantly correlate with the ABTS one (r2 = 0.871, p = 0.000 and r2 = 0.438, p = 0.007, respectively).
Antioxidant activity Metal chelating activity, which was lowest in the green teas, does not correlate with antioxidant activity
Chelating activity but appears to be influenced by theaflavins content. The results contribute to better understand how the
LDL manufacturing process influences the antioxidant activity of tea when variables (geographical region, envi-
Phenolic composition ronmental conditions, cultivar type, plucking techniques) are kept to a minimum. Secondly, we show that
novel African green, white and black Orthodox teas, made from tea varieties typically used in black CTC tea
production, may have potential health benefits comparable with commonly consumed teas.
© 2012 Elsevier Ltd. All rights reserved.

1. Introduction through steaming or heating using different methods (pan frying,

roasting, baking) before the rolling and drying process. Instead, white
Tea is the most popular beverage after water, enjoyed by many tea is the least processed in that it goes only through sunshine withering
people across the globe, and which is receiving continual interest and drying. Furthermore, white teas are characterized by the fact that
due to the many beneficial health effects associated with its regular only buds still covered with fine white hair, and one or two very young
consumption. leaves are used (Balentine, 1992).
All varieties of tea are produced from the young, tender leaves of the Consumers of black tea far outweigh those of green tea world-
Camellia sinensis (L.) (family Theaceae), a perennial, evergreen, leafy crop wide. However, in recent times green tea consumption has increased
that enjoys warm and humid climates with plenty of rainfall. Freshly, especially in the West, due to the numerous studies that indicate a
plucked leaves are then processed to give black tea (fermented), oolong wide variety of health benefits following its regular consumption:
tea (semi-fermented), green and white teas (unfermented). Fermenta- reduced risk of cardiovascular disease and certain types of cancer, in-
tion consists in the enzymatic oxidation by endogenous polyphenol flammatory bowel, liver and neurodegenerative diseases, diabetes,
oxidases and peroxidases of tea polyphenols, namely colorless flavanols, and even weight loss (Dufresne & Farnworth, 2001). These health
that are partially converted into theaflavins and thearubigins, respon- benefits are attributed to its high content of catechins which have
sible for the characteristic aroma and color of black and oolong teas been described as potent antioxidants ameliorating disease states re-
(Obanda, Owuor, & Mang'oka, 2004). This oxidation process is accelerat- lated to reactive oxygen species (ROS) (Rietveld & Wiseman, 2003).
ed by rupturing the withered tea leaves using orthodox rollers or ma- Since tea leaves are processed differently to produce black, green,
chines (CTC: crush-tear-curl), so that the oxidases released can more oolong and white teas, it is of interest to know which tea could poten-
readily react with oxygen. In unfermented teas, fermentation of the tially be more beneficial in terms of antioxidant activity. Many studies
withered leaves is prevented by inactivating the endogenous enzymes compare the activities of these teas but a conclusive answer has yet to
be reached. Generally, antioxidant activity decreases in the order:
green tea > oolong tea > black tea (Richelle, Tavazzi, & Offord, 2001;
⁎ Corresponding author at: Dipartimento Scienze della Vita e dell'Ambiente, Via
Brecce Bianche, Università Politecnica delle Marche, I-60131 Ancona, Italy. Tel.: +39
Roginsky, Barsukova, Hsu, & Kilmartin, 2003; von Gadow, Joubert, &
71 2204135; fax: +39 71 2204398. Hansmann, 1997) although some studies show that black teas
E-mail address: [email protected] (E. Damiani). are better than green ones (Hoff & Singleton, 1977; Khokhar &

0963-9969/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodres.2012.07.057
 P. Carloni et al. / Food Research International 53 (2013) 900–908 901

Magnusdottir, 2002; Venditti et al., 2010) while others report the ab- 2.2. Tea samples and preparation
sence of any significant differences (Hodgson et al., 1999; Lin, Juan,
Chen, Liang, & Lin, 1996). The numerous variables that affect tea compo- Five tea samples processed from hand plucked leaves of the superior
nents according to cultivar type, growth conditions (season, climate, soil), cultivar PC108, bred in Malawi for typical black CTC tea production by
horticultural practices (mechanical- or hand-plucking, age of leaves) and the Tea Research Foundation based in Mulanje, and grown on a private es-
the different technologies of the tea companies may account for these tate, were analyzed. The green leaves had been harvested from the same
conflicting reports (Kan, 1980; Wicremasinghe, 1974). Hence, it is not field at three different times: February (typically overcast, wet, warm con-
strictly correct to compare the antioxidant activities of black, white, and ditions), April (bright sunny days and cool nights) and September (hot
green teas unless these variables are minimized. clear days and warm nights) 2008. Fresh leaves were then processed
Therefore, the present study was carried out to explore what effect according to the flow-chart reported in Fig. 1 to give five different teas:
manufacturing has on the antioxidant properties of teas coming from a white, green, low-caffeine green, black CTC and black Orthodox teas.
single cultivar (grown and plucked under the same conditions) but The withering was done in troughs and ambient or heated air was
processed differently in the same factory to give two black teas (Orthodox blown into a plenum chamber and up through the leaf. In order to achieve
and CTC), two green teas (low-caffeine and non-decaffeinated) and one the target moisture content, the withering time varied depending on the
white tea. Total phenol, flavonoids, theaflavins, individual catechins and prevailing weather conditions and conditions of leaf on receipt into the
caffeine content, as well as chelating activity were also investigated. factory. The target moisture content for all teas at the end of the drying
process carried out in a fluid bed drier was 3–4%.
Prior to preparation of hot tea infusions, the teas were ground using a
2. Materials and methods pestle and mortar to obtain a homogeneous fine powder. The infusions
were prepared by pouring 20 ml of mineral water at 90 °C on 0.5 g of
2.1. Chemicals and equipment tea and brewed for 7 min. They were then filtered through Whatman
paper filters 43–48 μm and diluted appropriately with water according
All reagents, standards for HPLC analysis and solvents of highest to each specific assay.
purity available, were purchased from Sigma-Aldrich Chemical Co.
(Milan, Italy) and used as received. Ultrapure water was used through- 2.3. Total phenol content (TPC)
out and obtained from a Milli-Q system from Millipore (Milford, MA),
except for preparation of teas where bottled mineral water purchased Total phenol content in the tea infusions was determined using
from local retail shops was used. the Folin–Ciocalteu reagent (Singleton, Orthofer, Lamuela-Raventos, &
Spectrophotometric measurements were recorded on a UV Kontron Lester, 1999). To 1.975 ml water, 0.125 ml of Folin–Ciocalteu reagent
941 spectrophotometer or on a microplate reader (Synergy HT, Biotek, followed by 0.025 ml of tea previously diluted 5 fold, or appropriately
Winooski, VT, USA). diluted gallic acid standard ethanolic solution, or water as blank, were

Fig. 1. Flow chart showing the different stages in the manufacture of the different teas investigated. LTP = Lawrie tea processor; CTC = crush-tear-curl (see Materials and methods
for details).
902 P. Carloni et al. / Food Research International 53 (2013) 900–908

added and mixed. After 10 min, 0.375 ml of 20% Na2CO3 were added, were added and mixed. The samples were left for 2 h in the dark
mixed and samples were left for 2 h at room temperature in the dark. at room temperature, and absorbance was read at 734 nm against
Absorbance was read at 760 nm and the results were expressed as water. Inhibition percentage values were calculated according to
mM Gallic Acid Equivalents (GAE) using the linear regression value the equation:
obtained from the gallic acid calibration curve.
Inhibition of A734 ð% Þ ¼ ð1−Ac =A0 Þ  100
2.4. Total flavonoid content (TFC)
where Ac = absorbance of the samples, A0 = absorbance of the con-
The total flavonoid content in the tea infusions was measured using trol. Antioxidant activity was expressed as mM Trolox Equivalents
a colorimetric assay according to the method of Gursoy, Sarikurkcu, (TE) using the linear regression value obtained from the Trolox cal-
Cengiz, and Solak (2009) with some modifications. Briefly, 0.05 ml of ibration curve.
tea infusion or appropriately diluted (+)-catechin standard ethanolic
solution, or water as blank, were added to 1.35 ml of distilled water. 2.8. ORAC assay
After mixing, 0.05 ml of 5% NaNO2 followed by 10% AlCl3 (0.05 ml)
were added and mixed and samples were left for 10 min at room tem- Antioxidant activity of the different teas was also assessed with
perature in the dark. Absorbance was read at 415 nm and the results the ORAC (Oxygen Radical Absorbance Capacity) assay according to
were expressed as mM Catechin Equivalents (CE) using the linear re- the method of Gillespie, Chae, and Ainsworth (2007). Briefly,
gression value obtained from the catechin calibration curve. in each well of a solid black 96-well microplate, 0.15 ml of 0.08 μM
fluorescein (3′,6′-dihydrosyspiro[isobenzofuran-1[3H],9′[9H]-xanthen]-
2.5. Total theaflavins content (TTC) 3-one) dissolved in 75 mM PBS (phosphate buffered saline) was added
followed by 0.025 ml of tea previously diluted 1000 fold. After 10 min
The total theaflavins content in the tea infusions was determined incubation in the dark at 37 °C, 0.025 ml of 147 mM AAPH [2,2′-
according to the Flavognost method (Hilton, 1972). Tea infusions (3 ml) azobis(2-methylpropionamidine)dihydrochloride] were rapidly added
were mixed with 3 ml of 4-methylpentan-2-one (IBMK), vortexed for to each well and fluorescence recorded from the top every 120 s for
10 min and then allowed to stand until the layers separated. One milliliter 3 h, using an excitation wavelength of 485/20 nm and an emission filter
of the upper layer was pipetted into a test tube, followed by 2 ml ethanol of 528/20 nm. The kinetics showed a classic fluorescence decay due to
and 1 ml Flavognost reagent (2% w/v in ethanol). The contents were decomposition of fluorescein that was delayed in the presence of tea
mixed and left for 15 min to allow for color development. Absorbance samples or of Trolox standard ethanolic solution diluted with PBS. AUC
(A) at 625 nm was read against an IBMK/ethanol (1:1 v/v) blank. The (area under the fluorescence decay curve) was automatically calculated
TTC of the aqueous solution was calculated from Beer's law using the by the analytical software KC4 (Biotek, Winooski, VT, USA) connected to
equation [TTC] (mmol/l)=1.1496×A625 (Spiro & Price, 1986). the Synergy HT reader. The net AUC for each standard/compound was
obtained by subtracting the area of the blank sample (PBS). Antioxidant
2.6. Analysis of phenolic compounds and caffeine activity was expressed as mM Trolox Equivalents using the linear regres-
sion value obtained from the Trolox calibration curve.
HPLC for the analysis of phenolic compounds and caffeine was
conducted on a Gilson liquid chromatograph equipped with a 321 2.9. LDL isolation and oxidation
solvent pump, a 234 Autoinjector and a Dionex AD20 Absorbance Detec-
tor set at 280 nm. A C18 guard column and a Waters μ-Bondapack C18 LDL (low density lipoprotein) particles were isolated from pooled,
pre-packed column (3.9×300 mm) were used for separation. The fresh, human blood obtained from volunteers after overnight fasting
solvent compositions used were water:acetic acid, 93:3 v/v (solvent A) according to the method described by Venditti et al. (2010).
and HPLC grade methanol (solvent B), with a flow rate of 0.8 ml min−1. To determine the effects of tea samples on the inhibition of LDL
The conditions for HPLC analysis were modified from those described oxidation, conjugated dienes (a measure of lipid oxidation) were
by Zuo, Chen, and Deng (2002). Briefly, the elution was performed with monitored in LDL samples incubated in the presence and absence of
a gradient starting at 100% A to reach 80% at 15 min and 100% at different teas in a 96-well microplate reader. The volume of each
40 min. The freshly prepared tea infusions from samples harvested in well was adjusted with deionized water to reach a final volume of
April were filtered through a 0.45 μm membrane filter and 0.02 ml of 0.2 ml. Briefly, to 0.01 ml of LDL (100 μg protein/ml), 0.01 ml of tea
each sample were injected after a 1:10 dilution. Quantification of the previously diluted 400 fold was added. The oxidation reaction was
compounds at 280 nm was based on peak area, obtained with valley- initiated by adding 0.1 ml of 0.01 mM CuSO4 to each well and conju-
to-valley integrations, and retention times, using external standards. A gated dienes were monitored at 234 nm for 6 h at 37 °C with read-
dilution series of each standard (0.01 mg/ml–0.1 mg/ml) had been previ- ings taken every 120 s. The control sample lacked tea. The lag time
ously prepared for quantification. was then determined graphically and reported as such.

2.7. ABTS assay 2.10. Metal chelating assay

For measuring the antioxidant activity of the different teas, the ABTS The ferrous ion chelating activity of the tea samples was measured
assay was performed according to the method of Pellegrini, Ke, Yang, by the decrease in absorbance at 562 nm of the iron (II)-ferrozine
Rice-Evans, and Lester (1999). The radical cation (ABTS+•) was gener- complex (Viollier, Inglett, Hunter, Roychoudhury, & Van Cappellen,
ated by reacting aqueous stock solutions of ABTS [2,2′-azinobis-(3- 2000) with modifications. To 0.1 ml of a 0.8 mM FeSO4 solution in
ethylbenzothiazoline-6-sulfonic acid) diammonium salt] and K2S2O8 0.1 M HCl, 0.2 ml of tea previously diluted 5 fold, or of appropriately
to reach final concentrations of 6.3 mM and 2.45 mM respectively, for diluted EDTA (ethylenediaminetetraacetic acid) standard solution, or
12–16 h in the dark at room temperature. Prior to use, the ABTS+• water as control, were added followed by 0.1 ml of 5 mM ferrozine
stock solution was diluted ~ 80 fold with water in order to obtain an ab- [4,4′-[3-(2-pyridinyl)-1,2,4-triazine-5,6-diyl]bisbenzenesulfonic acid]
sorbance at 734 nm ranging between 0.6 and 0.8. To 2.475 ml of this solution in 0.1 M PBS. Finally, 1.6 ml of 0.1 M PBS was added to neutral-
ABTS +• solution, 0.025 ml of tea previously diluted 40 fold or appro- ize the solution. After standing for 10 min in the dark at room temper-
priately diluted Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2- ature, absorbance was read at 562 nm against a blank containing water
carboxylic acid) standard ethanolic solution, or water as control, instead of tea and lacking FeSO4. Antioxidant activity was expressed as
 P. Carloni et al. / Food Research International 53 (2013) 900–908 903

mM EDTA Equivalents (EE) using the inhibition percentage value which was treated to reduce caffeine levels (low-caffeine green tea)
obtained from the EDTA calibration curve as described for the ABTS had only slightly lower, non-significant levels of caffeine than the
assay. green tea. This unexpected result may be due to the fact that the tea
leaves had been treated in boiling water for only 15 s (Fig. 1) which
2.11. Data analysis may have been too short a time for sufficient extraction of water-
soluble caffeine. In fact, although a common technique of discarding a
Appropriate controls were carried out throughout all the experiments short steep (30–60 s) is believed to reduce caffeine content in a subse-
described above. The data reported represent average values from at least quent brew by 80–90%, research suggests that a 5 min steep yields up to
three independent experiments each performed in duplicate. Statistical 70% of the caffeine, and a second steep has a third of the caffeine of the
significance of correlations was evaluated using a two-tailed probability first (about 23% of the total caffeine in the leaves) (Hicks, Hseih, & Bell,
test. Statistical significance was performed using one-way ANOVA and 1996).
Tukey's multiple comparison test. A value of pb 0.05 was considered
statistically significant. 3.3. Antioxidant activity

3. Results The antioxidant activity of the tea infusions was assessed using three
independent assays, ABTS, ORAC, and inhibition of LDL oxidation. The
3.1. Levels of total tea polyphenols, flavonoids and theaflavins results obtained from the ABTS assay reported in the box-plot of
Fig. 4A show that the green tea has significantly higher antioxidant ac-
The distribution of the data of total phenols (TPC) and flavonoids tivity (55.0 ±12.7 mM TE) compared to all the rest, while similar activ-
(TFC) content measured in the tea infusions is reported in Fig. 2A ities are found for the low-caffeine green (44.0± 10.2 mM TE), white
and B. TPC was found to be higher in green and low-caffeine green (38.4 ± 9.9 mM TE) and black Orthodox (34.0 ± 8.4 mM TE) teas; the
teas (23.6 ± 5.4 and 22.6 ± 5.0 mM GAE) followed by the white antioxidant activity of the black CTC tea is significantly lower than all
(18.1 ± 5.5 mM GAE) and black CTC and Orthodox teas (10.7 ± 4.0 the others (25.5 ± 7.5 mM TE). Furthermore, a statistically significant
and 14.9 ± 5.9 mM GAE). No significant differences were observed correlation was found between the results obtained with the ABTS
amongst the green teas, whilst differences can be observed amongst assay and TPC and TFC reported in Fig. 2A and B (r2 = 0.871, p = 0.000
the two black teas processed differently: the black Orthodox tea has and r2 = 0.438, p = 0.007, respectively).
statistically significant (p b 0.05) higher levels of polyphenols than The results reported in Fig. 4B from the ORAC assay, follow in gen-
the black CTC tea. The trend of TFC is similar to that of TPC, except eral a similar trend to the ABTS one, although not in total accordance
for white tea (4.75 ± 0.65 mM CE) whose value is lower than black with it. In fact, the high antioxidant activity of the green teas is re-
Orthodox tea (6.60 ± 1.52 mM CE). Total theaflavins content (TTC) versed: the low-caffeine tea has a higher (70 ± 9 mM TE) but not sig-
was also determined, since it is considered to be an important criteri- nificant antioxidant activity than the non-decaffeinated one (61 ±
on of black tea quality. The TTC (Fig. 2C) was very low in the green 10 mM TE). Furthermore, there appears to be a remarkable antioxi-
teas as expected (b 0.01 mM theaflavins), while higher levels were dant activity in the black Orthodox tea (62 ± 8 mM TE) which is sig-
found in the remaining teas, with black CTC exhibiting the highest levels nificantly greater than the white tea (47 ± 8 mM TE) and parallels
(0.172 ± 0.019 mM theaflavins) being the most oxidized tea. White tea that of the green teas. Once again, the black CTC tea shows the lowest
which is not usually considered a fermented tea, also showed a signifi- antioxidant activity (32 ± 4 mM TE).
cant amount of theaflavins (0.093± 0.018 mM theaflavins). The lag times reported in Fig. 4C concerning the ability of the teas to
suppress copper-induced LDL oxidation, are in significant agreement
3.2. Individual phenolics and caffeine content with the trend of antioxidant activity using the ORAC assay reported
in Fig. 4B (r2 = 0.444, p = 0.0067). Furthermore, the black CTC tea
Individual phenolic compounds and caffeine content in tea infusions (123±9 min) did not appreciably prolong the lag time compared to
were determined in the teas harvested in April only and are reported in the control without tea (111± 19 min).
Table 1. Epigallocatechin (EGC) is the major flavanol in all the tested in-
fusions, especially in the two green teas, whereas in black CTC, catechin 3.4. Metal chelating activity
(C) is the major one. The levels of epigallocatechin-3-gallate (EGCG) are
similar for the two green teas and black Orthodox tea, lower in the The chelating activity of the tea infusions measures how effective
white tea and very low in the black CTC one. Epicatechin-3-gallate the compounds in them can compete with ferrozine for ferrous ions.
(ECG) levels are instead similar in all teas except in the black CTC Fig. 5 shows that the two green teas have the lowest chelating activity
which has the lowest level of this flavanol. The levels of C were found (0.178 ± 0.090 and 0.147 ± 0.066 mM EE) with no differences among
to be low in the black Orthodox and in the white tea while gallic acid each other, while the white tea displays the highest (0.477± 0.136 mM
(GA) was found to be highest in the black Orthodox tea. The distribution EE). Among the black teas, the Orthodox one shows a significantly
of the data of the total catechins content is reported in the box-plot of higher chelating power (0.449 ± 0.110 mM EE) compared to the CTC
Fig. 3. Total catechin levels are significantly higher in the low-caffeine one (0.249 ±0.056 mM EE).
green tea than in the black CTC and white teas but not with respect to
the green and black Orthodox teas. Furthermore, black CTC has signifi- 4. Discussion
cantly lower total catechin levels than the non-decaffeinated green
tea. By analyzing the total catechins content reported in Table 1 with The present study was carried out to determine how antioxidant
TPC and TFC data from the April harvest (data not shown), a statistically activity varies in infusions of tea leaves plucked from the same culti-
significant correlation was observed in both cases (r2 = 0.8091, p = var, grown in the same conditions, and processed differently in the
0.0377; r2 = 0.9018, p = 0.0135, respectively). same factory to make white, green and black tea. Since no single mea-
The caffeine levels were similar in all the teas, except for the black surement of antioxidant status is sufficient, a “battery” of different
CTC which had the lowest. This result is not surprising since fermen- assays, which work on different principles, was employed (Prior &
tation results in a slight loss of extractable caffeine (Cloughley, 1983) Cao, 1999; Roginsky et al., 2003). The ABTS assay is an indirect meth-
and the black CTC tea, which undergoes severe leaf disruption leading od that simply measures the capacity of the ABTS radical cation to ab-
to a greater surface area for enzymatic oxidation of the catechins, is stract an hydrogen atom or an electron from the compound/s under
the most fermented one. However, surprising is the fact that the tea study (Prior, Wu, & Schaich, 2005). The simplicity of the method has
904 P. Carloni et al. / Food Research International 53 (2013) 900–908

Fig. 2. Box-plot diagrams of the levels of total tea polyphenols, flavonoids and theaflavins in the tested teas. The cross represents the median value, 50% of the data are contained
within the box and the lines indicate the upper and lower quartiles, each one corresponding to 25% of the data. Panel A: Total polyphenols content (expressed as mM Gallic Acid
Equivalents) determined using the Folin–Ciocalteu reagent; Panel B: Total flavonoid content (expressed as mM Catechin Equivalents) determined using the aluminum chloride
colorimetric method; Panel C: Total theaflavin content (expressed as mM theaflavines) determined using the Flavognost reagent (see Materials and methods for details). Statistical
significance of one-way ANOVA was p b 0.0001. p for Tukey's least significant difference is summarized below each box-plot; ns = non-significant.
 P. Carloni et al. / Food Research International 53 (2013) 900–908 905

Table 1
Content of phenolic compounds and caffeine in tea infusions expressed in mg/ml.

GA C EGC EGCG ECG Total Cathechins content CAFF

White 0.018 ± 0.004 0.078 ± 0.022 0.331 ± 0.071 0.213 ± 0.052 0.120 ± 0.096 0.80 ± 0.36 0.547 ± 0.044
Green 0.007 ± 0.008 0.283 ± 0.064 1.214 ± 0.151 0.544 ± 0.091 0.178 ± 0.052 1.94 ± 0.57 0.604 ± 0.043
Green low-caffeine 0.017 ± 0.003 0.183 ± 0.110 1.333 ± 0.163 0.506 ± 0.145 0.135 ± 0.116 2.17 ± 0.55 0.514 ± 0.039
Black CTC 0.045 ± 0.004 0.163 ± 0.026 0.029 ± 0.007 0.011 ± 0.009 0.038 ± 0.010 0.24 ± 0.06 0.361 ± 0.047
Black Orthodox 0.081 ± 0.006 0.0209 ± 0.051 0.848 ± 0.118 0.461 ± 0.103 0.230 ± 0.066 1.75 ± 0.37 0.617 ± 0.070

GA = gallic acid, C = (+)-catechin, EGC = (−)-epigallocatechin, EGCG = (−)-epigallocatechingallate, ECG = (−)‐epicatechingallate, CAFF = caffeine. Values are mean of duplicate
determinations (n = 6) ± standard deviation.

made it very popular and suitable for assessing the ability of compounds tea in disease states related to reactive oxygen species (ROS), devel-
to act as hydrogen/electron donors and to evaluate their antioxidant ac- opment of cancer, cardiovascular diseases and other pathologies
tivity (Re et al., 1999). The ORAC assay is instead a direct method that (Dufresne & Farnworth, 2001; Rietveld & Wiseman, 2003). From our
measures antioxidant inhibition of peroxyl radical-induced oxidations results, the green teas have the highest content of catechins, especial-
and thus reflects classical, radical chain-breaking antioxidant activity ly EGC and EGCG, and they display the highest antioxidant activity.
by hydrogen atom transfer; its popularity is due to the fact that it can The remarkable antioxidant effect of these two catechins has been at-
measure both antioxidant capacity (duration of antioxidative action) tributed to the higher degree of hydroxylation of the B ring and/or hy-
and reactivity (Prior et al., 2005). The suppression of the oxidation of droxyl group at C-3 position of the basic catechin structure (Amic,
LDL, i.e. a test that directly analyzes the effect of a food containing antiox- Davidovic-Amic, Beslo, & Trinajstic, 2003) which increases its radical
idants on the oxidative degradation of a model system, was also evaluat- scavenging ability. The high antioxidant activity of catechins is fur-
ed (Sanchez-Moreno, Jimenez-Escrig, & Saura-Calixto, 2000). From the ther demonstrated when comparing the two black teas manufactured
results obtained using the different tests, a significant correlation was with the Orthodox method (for large leaf grades) and with the CTC
found between the ORAC and LDL assays which are both based on the method (small particles suitable for tea bags), where significant dif-
same principle, i.e. inhibition of an oxidative process in which several ferences were always observed. The CTC processing method involves
oxidative species are involved; hence a similar trend in results should a more rapid and severe leaf disruption producing smaller fragments
be expected. Instead, the ABTS assay gives an indication of the scaveng- and a greater surface area for enzymatic oxidation and this leads to a
ing efficiency of the different teas towards only one radical species, higher degree of fermentation which translates into lower catechin
i.e. ABTS radical cation, which reacts readily with hydrogen/electron levels, as confirmed by the HPLC analysis (Table 1). Astill et al. also
donors such as polyphenols. As expected, a strong correlation was found lower total catechin contents in CTC-manufactured black teas
found between the results obtained from the ABTS and TPC assays, but compared with Orthodox-manufactured ones determined on a batch of
both do not correlate with those of LDL and ORAC ones. leaves from the same area of an Assam estate (Astill, Birch, Dacombe,
Although some small differences in the results using the ABTS, Humphrey, & Martin, 2001). Orthodox teas have an enhanced yellow
ORAC and LDL assays were found, a general trend in antioxidant ac- color and higher aroma than the poorer flavor of CTC teas (Saikia &
tivity can be delineated, i.e. green tea ≥ low-caffeine green tea > white Mahanta, 2002), and now from the present study one can add that
tea ≥ black Orthodox tea > black CTC tea, which clearly demonstrates they also have a more potent antioxidant activity. From Fig. 2A and B,
that the manufacturing process influences the properties of tea. It is one can also observe that TPC and TFC generally decrease from green
well known that green tea is higher in total catechins than black to white to black tea, in good agreement with the antioxidant activities
tea, since the fermentation process reduces these levels in the latter observed, and also with the TPC trend generally reported in the literature
one as they are converted to theaflavins and thearubigins. It is also (Lin, Tsai, Tsay, & Lin, 2003; Roginsky et al., 2003).
known that high catechins levels positively correlate with antioxidant The trend in metal ion chelating activity observed in Fig. 5 deserves a
activity (Gramza-Michalowska & Korczak, 2007; Karori, Wachira, separate discussion. This activity is among one of the several factors
Wanyoko, & Ngure, 2007), which contribute to the protective role of which account for the antioxidant activity of polyphenols that includes

Fig. 3. Box-plot diagram of the total catechins content (mg/ml) in hot tea infusions from tea leaves harvested in April. The cross represents the median value, 50% of the data are contained
within the box and the lines indicate the upper and lower quartiles, each one corresponding to 25% of the data. The graph was constructed from the total catechins content reported in
Table 1. Statistical significance of one-way ANOVA was pb 0.0001. p for Tukey's least significant difference is summarized below each box-plot; ns=non-significant.
906 P. Carloni et al. / Food Research International 53 (2013) 900–908

Fig. 4. Box-plot diagrams of antioxidant activity of the tested teas. The cross represents the median value, 50% of the data are contained within the box and the lines indicate the
upper and lower quartiles, each one corresponding to 25% of the data. Panel A: ABTS assay results expressed as mM Trolox Equivalents; Panel B: ORAC assay expressed as mM Trolox
Equivalents; Panel C: LDL assay shown as the lag time for formation of conjugated dienes. Human LDL (100 μg protein/ml) oxidation was initiated by addition of 5 μM Cu2+ and
formation of conjugated dienes at 234 nm was determined in the presence of tea infusions. The control sample lacked tea. The lag time was calculated from the sigmoidal kinetic
curves of LDL oxidation obtained for each sample. See Materials and methods for details. Statistical significance of one-way ANOVA was p b 0.0001. p for Tukey's least significant
difference is summarized below each box-plot; ns = non-significant.
 P. Carloni et al. / Food Research International 53 (2013) 900–908 907

Fig. 5. Box-plot diagram of metal chelating activity (expressed as mM EDTA Equivalents) of the tested teas determined using the Ferrozine assay (see Materials and methods for
details). The cross represents the median value, 50% of the data are contained within the box and the lines indicate the upper and lower quartiles, each one corresponding to
25% of the data. Statistical significance of one-way ANOVA was p b 0.0001. p for Tukey's least significant difference is summarized below each box-plot; ns = non-significant.

radical scavenging, stability of the resulting phenoxyl radical, solubility the antioxidant properties of teas compare black, green and white teas
in the lipophilic phase, redox potential (Cao, Sofic, & Prior, 1997; Morel, from different geographical regions. Those that have studied teas from
Lescoat, Cillard, & Cillard, 1994), hence the reason for investigating this the same geographical region, have either used teas processed in differ-
activity. In this study, the green teas showed the lowest metal chelating ent tea factories and from different cultivars (Karori et al., 2007), or
activity while the white tea and the black Orthodox displayed the compared teas processed in the same way, i.e. just green teas (Saito et
highest. This trend is unusual as one would expect that teas with higher, al., 2007) or compared different cultivars (Baruah & Mahanta, 2003).
overall catechins content and hence higher antioxidant activity would Secondly, since there are very few non-CTC tea producers in Africa,
also display higher metal chelating activity. However, this does not ap- this study shows that black Orthodox, white and green teas, can be
pear to be the case, as also observed by Lin, Liu, and Mau (2008). It made from tea varieties bred as quick fermenting, astringent types, typ-
seems that components other than catechins might contribute most to ically used in black CTC tea production. This initial study on the antiox-
the chelating ability; the presence in a compound of a specific group idant properties of the novel African green, white and black Orthodox
bearing adjacent hydroxyl and carbonyl functions has been suggested teas, indicate that they may have comparable or superior potential
to play a role in metal chelation (Lin et al., 2008). This group for example, health benefits that could rival commonly consumed teas present on
can be found in theaflavins which are present in fermented teas and the market.
interestingly, in our study, the teas that show the highest chelating activ-
ity (white and black teas) are also those that have the highest content Acknowledgments
of theaflavins (Fig. 2C). White tea is commonly considered a non-
fermented tea although this is not strictly correct: in fact, white teas do The authors thank the Polytechnic University of the Marche for
not undergo the inactivation of enzymes before withering, and in this financial support. The authors declare that they have no conflict of
stage, polyphenol oxidases and peroxidases, essential for color develop- interest.
ment of this tea, remain active (Jiang, 2009). Therefore white tea poly-
phenols do undergo slight and slow oxidation, as confirmed by the
References
presence of significant amounts of theaflavins found in the white tea in-
vestigated (Fig. 2C). The slight oxidation of white tea polyphenols might Amic, D., Davidovic-Amic, D., Beslo, D., & Trinajstic, N. (2003). Structure–radical scav-
also explain why the antioxidant activity observed for this tea is always enging activity relationships of flavonoids. Croatica Chemica Acta, 76, 55–61.
Astill, C., Birch, M. R., Dacombe, C., Humphrey, P. G., & Martin, P. T. (2001). Factors affect-
slightly lower than the green teas and similar to the black Orthodox one. ing the caffeine and polyphenol contents of black and green tea infusions. Journal of
Tea products are manufactured with tea leaves harvested at differ- Agricultural and Food Chemistry, 49(11), 5340–5347.
ent seasons which is known to influence the flavor and aroma of tea Balentine, D. (1992). Manufacturing and chemistry of tea. In C. -T. Ho, T. Osawa, M. -T.
Huang, & R. T. Rosen (Eds.), Phenolic compounds in food and their effects on health
products (Kan, 1980). In our study, seasonal variations were observed (pp. 102–117). Washington, DC: American Chemical Society.
but it was not possible to predict which season produced teas with Baruah, A. M., & Mahanta, P. K. (2003). Fermentation characteristics of some assamica
higher antioxidant activities since there were no remarkable differ- clones and process optimization of black tea manufacturing. Journal of Agricultural
and Food Chemistry, 51(22), 6578–6588.
ences and an overall trend could not be deduced. Hence the data Cao, G., Sofic, E., & Prior, R. L. (1997). Antioxidant and prooxidant behavior of flavo-
obtained from the three seasons were all pooled and analyzed togeth- noids: Structure–activity relationships. Free Radical Biology & Medicine, 22(5),
er. The only exception was the HPLC analysis of individual catechin 749–760.
Cloughley, J. B. (1983). Factors influencing the caffeine content of black tea: Part 2 —
and caffeine contents reported in Table 1 and Fig. 3 which were rela-
The effect of production variables. Food Chemistry, 10, 25–34.
tive to the April harvest, since the average data of all the other analy- Dufresne, C. J., & Farnworth, E. R. (2001). A review of latest research findings on the
ses were similar to those of this season. health promotion properties of tea. The Journal of Nutritional Biochemistry, 12(7),
404–421.
In conclusion, this paper contains key information on how the
Gillespie, K. M., Chae, J. M., & Ainsworth, E. A. (2007). Rapid measurement of total
manufacturing process of tea affects antioxidant activity, TPC, TFC, TTC antioxidant capacity in plants. Nature Protocols, 2(4), 867–870.
and individual catechins content when variables are kept to a mini- Gramza-Michalowska, A., & Korczak, J. (2007). Polyphenols—Potential food improve-
mum: tea from a single specialty cultivar grown under the same condi- ment factor. American Journal of Food Technology, 2, 662–670.
Gursoy, N., Sarikurkcu, C., Cengiz, M., & Solak, M. H. (2009). Antioxidant activities,
tions and processed in the same factory as a green, white and black tea. metal contents, total phenolics and flavonoids of seven Morchella species. Food
This is the first study of its kind as most literature reports investigating and Chemical Toxicology, 47(9), 2381–2388.
908 P. Carloni et al. / Food Research International 53 (2013) 900–908

Hicks, M. B., Hseih, P., & Bell, L. N. (1996). Tea preparation and its influence on methyl- Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., & Rice-Evans, C. (1999).
xanthine concentration. Food Research International, 29(325–330). Antioxidant activity applying an improved ABTS radical cation decolorization
Hilton, P. J. (1972). In vitro oxidation of flavanols from tea leaf. Phytochemistry, 18, assay. Free Radical Biology & Medicine, 26, 1231–1237.
1243–1248. Richelle, M., Tavazzi, I., & Offord, E. (2001). Comparison of the antioxidant activity of
Hodgson, J. M., Proudfoot, J. M., Croft, K. D., Puddey, I. B., Mori, T. A., & Beilin, L. J. (1999). commonly consumed polyphenolic beverages (coffee, cocoa, and tea) prepared
Comparison of the effects of black and green tea on in vitro lipoprotein oxidation in per cup serving. Journal of Agricultural and Food Chemistry, 49, 3438–3442.
human serum. Journal of the Science of Food and Agriculture, 79, 561–566. Rietveld, A., & Wiseman, S. (2003). Antioxidant effects of tea: Evidence from human
Hoff, J. F., & Singleton, K. I. (1977). A method for determination of tannin in foods by clinical trials. Journal of Nutrition, 133(10), 3285S–3292S.
means of immobilized enzymes. Journal of Food Science, 42, 1566–1568. Roginsky, V., Barsukova, T., Hsu, C. F., & Kilmartin, P. A. (2003). Chain-breaking anti-
Jiang, H. Y. (2009). White tea. Its manufacture, chemistry and health effects. In C. T. Ho, oxidant activity and cyclic voltammetry characterization of polyphenols in a
J. K. Lin, & F. Shahidi (Eds.), Tea and tea products. Chemistry and health-promoting range of green, oolong, and black teas. Journal of Agricultural and Food Chemistry,
properties. (pp. 17–30) Boca Raton: CRC Press. 51, 5798–5802.
Kan, N. (1980). The aroma constituents of various teas. Food Industry, 12, 19–22. Saikia, P., & Mahanta, P. K. (2002). Specific fluctuations in the composition of
Karori, S. M., Wachira, F. N., Wanyoko, J. K., & Ngure, R. M. (2007). Antioxidant capacity of lipoxygenase- and glycosidase-generated flavors in some cultivated teas of Assam.
different types of tea products. African Journal of Biotechnology, 6(19), 2287–2296. Journal of Agricultural and Food Chemistry, 50, 7691–7699.
Khokhar, S., & Magnusdottir, S. G. (2002). Total phenol, catechin, and caffeine contents Saito, S. T., Gosmann, G., Saffi, J., Presser, M., Richter, M. F., & Bergold, A. M. (2007).
of teas commonly consumed in the United kingdom. Journal of Agricultural and Characterization of the constituents and antioxidant activity of Brazilian green
Food Chemistry, 50(3), 565–570. tea (Camellia sinensis var. assamica IAC-259 cultivar) extracts. Journal of Agricultural
Lin, Y. -L., Juan, I. -M., Chen, Y. -L., Liang, Y. -C., & Lin, J. -K. (1996). Composition of poly- and Food Chemistry, 55, 9409–9414.
phenols in fresh tea leaves and associations of their oxygen-radical-absorbing ca- Sanchez-Moreno, C., Jimenez-Escrig, A., & Saura-Calixto, F. (2000). Study of
pacity with antiproliferative actions in fibroblast cells. Journal of Agricultural and low-density lipoprotein oxidizability indexes to measure the antioxidant activity
Food Chemistry, 44, 1387–1394. of dietary polyphenols. Nutrition Research, 20, 941–953.
Lin, S. -D., Liu, E. -H., & Mau, J. -L. (2008). Effect of brewing methods on antioxidant Singleton, V. L., Orthofer, R., Lamuela-Raventos, R. M., & Lester, P. (1999). Analysis of
properties of steaming green tea. LWT—Food Science and Technology, 41, total phenols and other oxidation substrates and antioxidants by means of Folin–
1616–1623. Ciocalteu reagent. Methods in Enzymology, 299, 152–178.
Lin, Y. S., Tsai, Y. J., Tsay, J. S., & Lin, J. K. (2003). Factors affecting the levels of tea poly- Spiro, M., & Price, W. E. (1986). Determination of theaflavins in tea solution using the
phenols and caffeine in tea leaves. Journal of Agricultural and Food Chemistry, 51, Flavognost complexation method. Analyst, 111, 331–333.
1864–1873. Venditti, E., Bacchetti, T., Tiano, L., Carloni, P., Greci, L., & Damiani, E. (2010). Hot vs.
Morel, I., Lescoat, G., Cillard, P., & Cillard, J. (1994). Role of flavonoids and iron chelation cold water steeping of different teas: Do they affect antioxidant activity? Food
in antioxidant action. Methods in Enzymology, 234, 437–443. Chemistry, 119, 1597–1604.
Obanda, M., Owuor, P. O., & Mang'oka, R. (2004). Changes in thearubigen fractions and Viollier, E., Inglett, P. W., Hunter, K., Roychoudhury, A. N., & Van Cappellen, P. (2000).
theaflavins levels due to variations in processing conditions and their effects on The ferrozine method revisited: Fe(II)/Fe(III) determination in natural waters.
black tea liquor brightness and total colour. Food Chemistry, 85, 163–173. Applied Geochemistry, 15, 785–790.
Pellegrini, N., Ke, R., Yang, M., Rice-Evans, C., & Lester, P. (1999). Screening of dietary von Gadow, A., Joubert, B., & Hansmann, C. F. (1997). Comparison of the antioxidant ac-
carotenoids and carotenoid-rich fruit extracts for antioxidant activities applying tivity of rooibos tea (Aspalathus linearis) with green, oolong, and black teas. Food
2,2′-azinobis(3-ethylenebenzothiazoline-6-sulfonic acid) radical cation decolori- Chemistry, 60, 73–77.
zation assay. Methods in Enzymology, 299, 379–389. Wicremasinghe, R. L. (1974). The mechanism of operation of climatic factors in the bio-
Prior, R. L., & Cao, G. (1999). In vivo total antioxidant capacity: Comparison of different genesis of tea flavor. Phytochemistry, 13, 2057–2061.
analytical methods. Free Radical Biology & Medicine, 27, 1173–1181. Zuo, Y., Chen, H., & Deng, Y. (2002). Simultaneous determination of catechins, caffeine
Prior, R. L., Wu, X., & Schaich, K. (2005). Standardized methods for the determination of and gallic acids in green, Oolong, black and pu-erh teas using HPLC with a photo-
antioxidant capacity and phenolics in foods and dietary supplements. Journal of diode array detector. Talanta, 57, 307–316.
Agricultural and Food Chemistry, 53, 4290–4302.