Banana is a nutritious fruit with a pleasant flavor that is widely consumed
throughout the world. It is a commercially important fruit crop in the
world trade. In countries such as Costa Rica and Honduras, bananas account
for more than 25% of the total export (Ortiz et al., 1998). Bananas
are prone to rapid browning during handling, peeling and slicing operations
and even storage, if ripening is not adequately controlled. This phenomenon
lowers the fruit quality and decreases its marketability (Lozano, 2007).
Browning is mainly attributed to the oxidation of phenolic compounds
by polyphenol oxidases (PPO). These enzymes catalyze the oxidation of
different phenols to the corresponding quinines; highly reactive compounds
that finally polymerize to melanins. The relationship between the phenolic
content, PPO activities, pH, temperature, oxygen availability within the
tissue and the browning rate has been examined for various fruits (Marshall
et al., 2000; Yoruk and Marshall, 2003). Griffiths (1961) presented
evidence indicating that the browning of banana fruit resulted from the
enzymatic oxidation of 3,4-dihydroxyphenylethylamine (dopamine). The presence
of a large quantity of dopamine in the different tissues of bananas was
confirmed in later reports (Tono et al., 1999).
In spite of some valuable studies on the features and function of the
banana PPO (Table 1), there are still ambiguities regarding the identity
and, more seriously, the physiological role of the banana PPO (both pulp
and peel tissues). A clear understanding of banana PPO not only leads
to more efficient control of the enzyme, but it also sheds light on its
physiological role. Literature survey reflects the ambiguity encircling
the identity of the banana pulp PPO (BP-PPO) as both the enzyme numbers
of (EC 220.127.116.11) for catechol oxidase and (EC 18.104.22.168) for phenolases
are being used (Yang et al., 2004; Wuyts et al., 2006; Unal,
2007). To address this controversy, the BP-PPO was purified to a single
isozyme level, then, the activities and some important biochemical properties
of the enzyme were investigated. The assay methods and the results are
compared with previous reports and finally discussed in terms of assumptions
regarding the physiological roles of the BP-PPO.
MATERIALS AND METHODS
Chemicals and solutions: Bananas (Musa cavendishii) in
the yellow to light brown spot stage of ripening were purchased from the
local market. Sephadex G-100 was purchased from Pharmacia and diethylaminoethyl
cellulose (DE-52) from Whatman™ (Maidstone, UK).
Bovine serum albumin (BSA), ovalbumin, 3,4-Dihydroxy cinnamic acid (caffeic
acid), 3,4-dihydroxyphenyl ethylamine (Dopamine) and 4-aminoantipyrine
were purchased from Sigma Chemical Company. Triton X-100, insoluble polyvinylpyrrolidone
and ascorbic acid were purchased from Merck Chemical Company. 4-[(4-Methylphenyl)azo]-phenol
(MePAPh) was prepared as described earlier (Haghbeen and Tan, 1998). Mushroom
tyrosinase (MT) was prepared according to Haghbeen et al. (2004).
All the other chemicals and reagents used in this research were taken
from the authentic samples. Double-distilled water was used for preparing
the desired solutions. The applied extraction solution in this research
was a phosphate buffer solution (PBS, 0.02 M, pH 7.0 ± 1) containing
triton X-100 (1%), ascorbic acid (0.01%) and water-insoluble polyvinylpyrrolidone
(2%). PBS 0.01 M at pH 6.8 ± 1 was used for the purification and
assaying the enzyme.
Protein extraction: The diced fresh pulp was frozen in liquid
N2, then, 50 g of that was powdered in a pre-chilled porcelain
mortar. Extra liquid N2 was applied intermittently during grinding
to prevent the tissues from thawing. Next, 100 mL of the extraction medium
was added to the frozen powder and macerated for 20 min. The macerate
was passed through a cotton cloth. The filtrate centrifuged at 4 °C
and 20000xg for 20 min. The supernatant was collected for further processing.
Enzyme purification: The obtained extract was subjected to a two-stage
protein precipitation. Ammonium sulfate powder was added to the collected
supernatant from the former step to make a 30% saturated solution. The
resulting solution was stirred in an ice bucket for 30 min, then, centrifuged
for 30 min at 20000xg and 4 °C. After removing the precipitate at
this stage, the supernatant was saturated to 65% of ammonium sulfate.
The solution was left stirring on ice for 2 h, followed by centrifugation
at 20000xg and 4 °C for 30 min. After discarding the supernatant,
the resulting precipitate was dissolved in PBS and dialyzed against similar
PBS at 4 °C overnight.
The dialyzed protein solution was loaded onto an ion exchange (DE-52)
column, 2.5x45 cm. The proteins were eluted from the column by a gradient
of NaCl solution, 0 to 200 mM, with a flow rate of approximately 1 mL
min-1 controlled by a peristaltic pump. The output of the column
was monitored spectrophotometrically at 280 nm. The collected fractions
were also checked for the catecholase activity in the presence of caffeic
acid at each step of elution. According to the monitoring results, the
collected protein fractions at 150 mM salt eluent were pooled and lyophilized.
The final stage of purification was carried out by dissolving the lyophilized
sample in the desired volume of PBS and loading it onto a Sephadex G-100
column (1.6x100 cm Pharmacia). The column was washed by PBS using a peristaltic
pump at a flow rate of 3 mL min-1. The ensuing fractions were
collected and checked for the presence of the protein and enzyme. Native
polyacrylamide gel electrophoresis (PAGE) of the collected fractions was
performed according to the conventional method introduced by Davis and
the protein bands were visualized through staining with Coomassi brilliant
blue G250. The protein concentration of the fractions was determined by
the Bradford method.
Molecular weight estimation and determination of isoelectric pH (PI):
Molecular weight of the purified enzyme was estimated from PAGE observations
using MT (128 kD), BSA (65 kD), ovalbumin (45 kD) as standards. The isoelectric
pH was determined by isoelectric focusing on polyacrylamide gel in glass
tubes (150x1.2 mm) following the technique introduced primarily by O`Farrell
(Sojo et al., 1998a). The ampholine (LKB, Stockholm, Sweden) was
in a pH range of 3.5-10. The electrodes were connected to a Hoefer power
supply and the current was maintained at 600 V for 12 h then, raised to
1000 V for 1 h. The protein bands were revealed by Coomassie blue R-250
(0.01%). The pH gradient was determined by cutting the unstained gel into
0.5 cm fragments which eluted by distilled water for 2 h. The pH of the
eluent was recorded finally.
Assay of enzyme activities: All the enzymatic reactions were run
in PBS at constant temperature, 25 ± 0.1 °C. The final volume
of all the reaction mixtures was 3 mL filling up three quarters of the
conventional, 1 cm width, UV-Vis cuvette. Freshly prepared BP-PPO solution
(1 mg mL-1) was used for both the cresolase and catecholase
reactions (Fig. 1). Using a CECIL CE9500 spectrophotometer,
the rate of the enzymatic reaction of BP-PPO was monitored as described
previously (Haghbeen and Tan, 2003). Therefore, the cresolase and catecholase
activities were assayed through the depletion of MePAPh (λmax
= 352 nm, ε = 20800 M-1 cm-1) and caffeic acid
(λmax = 311 nm ε = 12000 M-1 cm-1),
respectively. Using this method, the Michaelis-Menten constants for the
desired substrates were obtained from the corresponding kinetic data of
their reactions in the presence of a constant amount of the purified enzyme
(150 μL) which analyzed by the Lineweaver and Burk (1934) method.
All the results presented in this article are the averages of, at least,
triplicate measurements. Catecholase activity of BP-PPO in the presence
of dopamine was monitored at 470 nm
||Tyrosinase reactions and possible modification of the ortho-quinonic
product (R = could be any type of substitute)
using Palmer (1963) method. Peroxidase activity of the desired solutions
was also checked according to the Gallati (1977) method.
Dopamine extraction and analysis: Twenty gram of the diced pulp
(or peel) was frozen in liquid N2 and powdered in a prechilled
porcelain mortar. To this frozen powder, 50 mL of degassed 2-propanol
was added and stirred for 20 min. Then, it was passed through Whatman
paper and centrifuged at 8000xg for 10 min. The filtrate, containing phenolic
compounds, was dried at low pressure. Dopamine was separated form the
dried extract by applying an acidic solution at pH 3.
The extracted dopamine from the banana pulp (or peel) was dissolved in
a mixture of acetonitrile and water (20:80, v/v) and subjected to High
Performance Liquid Chromatography by a Beckman HPLC, 125 Solvent module
with Beckman C18 silica column (4.6x250 mm) and 168 Detector. A mobile
phase of acetonitrile-water (20:80, v/v) at a flow rate of 0.8 mL min-1
was applied. The injection volume was 50 μL and detector was set
at 288 nm.
RESULTS AND DISCUSSION
Extraction of enzyme: Some scientists have reported extraction
of BP-PPO using a buffer system containing no detergent (Padron et
al., 1975; Yang et al., 2000; Unal, 2007). In contrast, some
others applied extraction medium containing insoluble PVP and Triton X-100
(Thomas and Janave, 1986; Jayaraman et al., 1987; Ngalani et
al., 1993). Wuyts et al. (2006) reported the optimized amount
of these substances in the extraction medium for the PPO from the Musa
acuminate root. Results of our experiments on the BP-PPO were generally
in agreement with the later case. Apparently, detergents increase the
extractability of PPO while the removal of phenols in the presence of
PVP is sufficient to avoid browning of the enzyme solution. However, it
seems necessary to dialyze the extract against PBS since it improves the
stability and activity of the enzyme (Table 2). This
effect could be due to the renaturation of PPO in the presence of phosphate
ions and the removal of inhibitory substances and detergents during dialysis
(Galeazzi et al., 1981).
Purification of the BP-PPO: To avoid the possible risk of the
denaturating effect of acetone (Thomas and Janave, 1986), it was preferred
to use ammonium sulfate for precipitating proteins from the pulp extract.
The partially purified enzyme was obtained after removing the heavy proteins
from the pulp extract at 30% ammonium sulfate saturation. Precipitate
collected at this stage showed little activity while the proteins collected
at 65% ammonium sulfate saturation showed both catecholase and cresolase
activities (Fig. 2, Table 2). Interestingly,
the BP-PPO showed the characteristic lag time observed during the cresolase
activity of tyrosinases (Sojo et al., 1998b).
To purify the BP-PPO, the precipitate collected at 65% ammonium sulfate
saturation was first chromatographed on an ion-exchange column (DE-52).
Similar to that observed during the MT purification (Haghbeen et al.,
2004), the enzymatic assays revealed the highest PPO activity in the fraction
eluted by 150 mM NaCl while the other fractions were either inert or showed
low activity (Fig. 3A). The native PAGE of the separated
fractions is shown in Fig. 3B.
||Progress of the cresolase reactions of the A) crude extract and
B) purified isozyme of the BP-PPO in the presence of MePAPh. Refer
to the materials and method section for the experimental details
||The resulting chromatogram of the BP-PPO elution from the DE-52
column (A-bottom). The results of the catecholase assay of each fraction
in the presence of caffeic acid (A-top) and their corresponding PAGE
(B). Refer to the materials and method section for the experimental
||The information collected from some important reports on the BP-PPO.
The numbers in parentheses show the extinction coefficients (cm-1
M-1) of the ortho-quinonic intermediate formed after
the oxidation of a phenolic compound by a PPO. The numbers with astric
show the Km value of caffeic acid. R2 is the
regression coefficient of the kinetic plot. Blank cell and minus sign
are for the “lack of any information and not identified, correspondingly
The collected fraction at 150 mM NaCl was dialyzed, concentrated and
transferred onto a Sephadex G-100 column for final purification. Accordingly,
it seems that two active isoforms of the BP-PPO can be separated applying
this method (Fig. 4A). The result of the native electrophoresis
of the most active isoform of BP-PPO in comparison with MT, ovalbumin
and BSA is shown in Fig. 4B. Considering these results
and shown in Fig. 3A
||The resulting chromatogram of the BP-PPO elution from the Sephadex
G-100 column (A-bottom). The results of the catecholase assay of each
fraction in the presence of caffeic acid (A-top). The PAGE (B) of
the purified isozyme of the BP-PPO with the highest activity (3) in
comparison with ovalbumine (1), BSA (2) and MT (4). Refer to the materials
and method section for the experimental details
as well as the high similarity between the PAGE patterns of the collected
fractions shown in Fig. 3B indicate the existence of
several isoforms for the BP-PPO, however, the fraction eluted by 150 mM
NaCl contained the most active ones. It means that the other isoforms
of the BP-PPO had lost a great deal of their activities either prior or
during the extraction process.
A molecular weight of 58.2 kD can be estimated from Fig.
4B for the purified BP-PPO. Yang et al. (2000) and Sojo et
al. (1998a) had also attained a single band PAGE for the Musa sapientum
L. and Musa acuminate PPO, respectively. The former had estimated
a molecular weight of 41-42 kD for the separated PPO while the latter
reported no molecular weight. But the MW(s) reported for the Musa Cavendishii
PPO in previous works, 60 kD (Padron et al., 1975) and 65 kD (Galeazzi
et al., 1981) are very close to the MW estimated in this research.
The isoelectic focusing of the purified isoform with the highest activity
was performed in polyacrylamide gel containing ampholines in the range
of pH 3.5-10. The corresponding zymogram (Fig. 5) indicates
the existence of only one band with a pI value of 5.28. This result clearly
supports the success of the applied method for purification of a single
isoform of the Musa cavendishii PPO. Galeazzi et al. (1981)
had reported a similar pI for the Musa cavendishii PPO and Thomas
and Janave (1986) had reported a range of 4 to 5.5 for the pI values of
a group of Musa cavendishii PPO isozymes.
BP-PPO is a tyrosinase: Determining the presence of both monophenolase
and catecholase activities for an
||(Left) The isoelectic focusing of the purified isoform of BP-PPO
on the polyacrylamide gel containing ampholines and (Right) the corresponding
zymogram. The arrows refer to the stained enzyme and its pI on the
left and right figures, respectively
active PPO present in a plant material is a definitive way to categorize
that enzyme as a tyrosinase (Sa´nchez-Ferrer et al., 1988). Tyrosinases
usually show a more pronounced catecholase activity in comparison with
the cresolase activity. Besides, catecholase is much faster than the cresolase.
This is why most researchers prefer to follow the PPO through its ortho-diphenolase
activity. However, ignoring the monophenolase activity of the desired
PPO might bring about confusion regarding the true identity of the PPO.
The identity of BP-PPO was a matter of disagreement until Sojo et al.
||Results of the catecholase assay of BP-PPO in the presence of caffeic
acid (50 μM) at 25 °C in PBS (0.01 M) at each stage of the
purification procedure introduced in this work. The number in parenthesis
shows the specific unit of activity [(μM min-1)/mg]
presented the details of a study on the phenolase activity of Musa
acuminate pulp PPO. Yang et al. (2000) also reported the phenolase
activity of Musa sapientum L. pulp PPO in the presence of p-cresol.
Now, results of this research confirm the cresolase activity of the purified
isozyme of BP-PPO from Musa cavendishii in the presence of a synthetic
phenolic substrate. The cresolase and catecholase activities of the BP-PPO
showed an 11.67 and 13.3 fold recovery, respectively, upon applying the
introduced purification procedure in this paper (Fig. 2,
The extract and the isolated isozyme were checked for the peroxidase
activity too. Results were negative. The outcome of this research and
similar reports on different varieties of banana indicate that BP-PPO
is a tyrosinase. However, the cresolase activity of the enzyme is very
susceptible to the conditions of the purification procedure. In fact,
the lability of the banana pulp tyrosinase decreases the reproducibility
of the observed results which is a setback observed in similar works (Sanchez-Ferrer
et al., 1993).
Kinetics parameters: The kinetic parameters of BP-PPO are usually
determined in the presence of dopamine. As Table 1 shows,
except one report, a spectrophotometric method based on the formation
of the ortho-quinonic intermediate during the dopamine oxidation
by the BP-PPO is commonly used for assaying the enzyme activity (Yoruk
and Marshall, 2003). The UV-Vis spectrum of the intermediate overlaps
with the absorption spectrum of the phenolic substrate at wavelengths
lower than 300 nm. Therefore, the formation of this intermediate is followed
at its other λmax (s) about 420 or 470 nm. Because of
the broad absorption band of the intermediate at 470 nm region, different
λmax (s) of 462, 465, 470 and 480 nm have been selected
(Table 1). There are also two extinction coefficients
for the ortho-quinonic intermediate at this region in the literature.
It was preferred to use the extinction coefficient used by Palmer (1963),
2512 M-1 •cm-1, in this research because it
is very close to that obtained by Waite (1976). Consequently, the Km
and Vmax values of 0.94 mM and 14.8 mM•min-1
were extracted from the fitted equation (y = 6.355x + 0.6779, R2
= 0.995) for the Lineweaver-Burk plot of dopamine shown in Fig.
||The Lineweaver-Burk plots of the kinetics results of dopamine (♦)
and caffeic acid (■) oxidation by the purified isozyme of BP-PPO
in this work. Refer to the materials and methods for the experimental
Caffeic acid is also a natural phenolic acid usually found in plants.
The Michaelis-Menton constants of this substrate in the presence of the
BP-PPO have been reported in some papers using the spectrophotometric
method for following the quinonic intermediate formation (Table
1). However, Haghbeen and Tan (2003) showed that it is possible to
follow the enzymatic oxidation of caffeic acid at its λmax
of 311 nm since the absorption spectra of the substrate and its ortho-quinonic
intermediate do not show large overlap at this wavelength. Therefore,
it is possible to study the kinetics of the enzymatic reaction directly
through the depletion of the substrate. Using this method, the Km
and Vmax values of 18.6 μM and 2.8 μM min-1
were extracted from the fitted equation (y = 6.6387x + 356880, R2
= 0.96) for the Lineweaver-Burk plot of caffeic acid shown in Fig.
6. It is clear that these data are outstandingly smaller than the
previously reported kinetic parameters for caffeic acid or even dopamine
mainly because of the setbacks associated with the assay method at 420
or 470 nm which has been discussed earlier (Haghbeen and Tan, 2003).
BP-PPO and MT: Haghbeen and Tan (2003) obtained the kinetic parameters
of caffeic acid in the presence of MT under conditions similar to those
applied in this work (Km = 27.5 μM and Vmax
= 2.2 μM min-1). Theses parameters are very close to those
obtained for caffeic acid in this report. On the other hand, if no stabilizer
is added to the purely lyophilized MT, its single-band PAGE pattern (Haghbeen
et al., 2004) changes upon storage at 4 °C and shows extra
bands (Fig. 4). It is assumed that these extra bands
belong to the dissociated mers of the tetrameric MT. Interestingly, one
of these dissociations stands right beside the purified BP-PPO isozyme
||HPLC results of (A) the commercial dopamine and (B) the 2-propanol
extract of the Musa cavandishii peel and their corresponding
UV-Vis spectra (C and D), respectively
The kinetic and PAGE evidence suggests significant similarity between
BP-PPO and MT. It seems the most stable quaternary structure of MT is
a tetramer composed from two dimmers while the banana tyrosinase is a
dimmer similar to that suggested by Galeazzi et al. (1981).
Dopamine and physiological role of the banana tyrosinase: As mentioned
earlier in this paper, dopamine presence in banana and its participation
in the browning phenomenon have already been shown (Griffiths, 1961).
The analysis of the 2-propanol extracts of the pulp and peel of the used
banana in this work also confirms the presence of high quantity of dopamine
in these tissues, however, it seems that it is dominant in peel (Fig.
7). It is known that different compartmentalization of the enzyme
and substrate in cells of healthy fruits avoids the occurrence of browning.
But, if direct contact does not occur between these two, what is the real
physiological role of banana tyrosinase in healthy fruit and skin.
All the proposed answers to this question are based on the chemistry
of the final product of BP-PPO reactions. ortho-quinones are highly
reactive substances which can easily polymerized to melanins. They are
prone to nucleophilic attacks and hence, are readily associated with amino
acids or proteins. Besides, ortho-quinones can take part in interamolecular
cyclization. All these options have been taken into account to describe
the role of PPO in the protection of plants against diseases and invading
pathogens or biosynthesis of natural products like betalains (Yoruk and
Marshall, 2003). But, ortho-quinones can also be reduced back to
ortho-dihydroxy compounds which can be used in redox systems or
further enzymatic reaction such as the one proposed in Fig.
1. Considering the PPO location in plant cells, some suggested that
PPO is important mostly for dark processes in thylakoid lumen (Sheptovitsky
and Brudvig 1996), but there is little evidence for participation of this
enzyme in biosynthesis of guaicolic substances.
This project was financially supported by the National Institute for
Genetic Engineering and Biotechnology of Iran (NIGEB, Project 269). This
research was totally conducted in NIGEB. I also appreciate the sincere
collaboration of Miss Jennifer Moll in improving the English of the text.