This article was downloaded by: [North Carolina State University]
On: 17 December 2012, At: 17:29
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Critical Reviews in Food Science and Nutrition
Publication details, including instructions for authors and subscription information:
/>Enzymatic browning reactions in apple and apple
products
Jacques J. Nicolas
a
, Florence C. Richard‐Forget
b
, Pascale M. Goupy
b
, Marie‐Josèphe
Amiot
b
& Serge Y. Aubert
b
a
Chaire de Biochimie Industrielle et Agro‐Alimentaire, Conservatoire National des Arts et
Métiers, 292 Rue Saint‐Martin, PARIS Cedex 03, 75141, France
b
Laboratoire de Biochimie des Dégradations, Station de Technologie des Produits Végétaux,
Institut National de la Recherche Agronomique, Domaine Saint‐Paul, B.P. 91, Montfavet
Cedex, 84143, France
Version of record first published: 29 Sep 2009.
To cite this article: Jacques J. Nicolas , Florence C. Richard‐Forget , Pascale M. Goupy , Marie‐Josèphe Amiot & Serge Y.
Aubert (1994): Enzymatic browning reactions in apple and apple products, Critical Reviews in Food Science and Nutrition,
34:2, 109-157

To link to this article: />PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: />This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
Critical Reviews
in
Food Science and Nutrition, 34(2): 109-157 (1994)
Enzymatic
Browning
Reactions
in
Apple
and
Apple
Products
Jacques
J.
Nicolas
Chaire
de
Biochimie
Industrielle
et
Agro-Alimentaire,
Conservatoire

National
des
Arts
et
Metiers,
292
Rue
Saint-Martin,
75141
PARIS
Cedex 03, France
Florence
C.
Richard-Forget,
Pascale
M.
Goupy,
Marie-Josèphe
Amiot,
and Serge
Y.
Aubert
Laboratoire
de
Biochimie
des Degradations,
Station
de
Technologie
des

Produits
V6getaux,
Institut
National
de
la
Recherche
Agronomique,
Domaine
Saint-Paul, B.P. 91,84143
Montfavet
Cedex, France
ABSTRACT: This review examines
the
parameters
of
enzymatic browning
in
apple
and
apple products that
is,
phenolic compounds, polyphenoloxidases,
and
other factors (ascorbic acid
and
peroxidases), both qualitatively
and quantitatively. Then
the
relationships between intensity

of
browning
and the
browning parameters
are
discussed, including
a
paragraph
on the
methods used
for
browning evaluation. Finally,
the
different methods
for
the control
of
browning
are
presented.
KEY WORDS: apple, browning, polyphenols, enzymes, polyphenoloxidases
I. INTRODUCTION
Based
on
size
of
production,
the
apple
is one

of the main fruit crops
in the
world.
In
1980,
the
world apple production exceeded
35
million
t,
corresponding
to
fourth place after grapes, citrus,
and banana.
356
The
four major producers
are the
U.S.,
France, Italy,
and
China. More than half
of
the apple crop
is
sold
in
the fresh produce market.
Depending
on the

year,
40 to
60%
of
apple pro-
duction
is
used
by the
industry. Juice, either
as
single-strength sweet juice
or as
fermented juice
(cider),
apple sauce, and slices, are the main prod-
ucts
of
processed apples. According
to
Salunkhe
et al.,
357
one
fourth
of the
fruit
and
vegetables
harvested

is
never consumed because
of
spoilage
during postharvest manipulations
or
processing.
Concerning apples,
the
postharvest losses
are
considerably less, because
it has
been estimated
at less than 10%
for
the fresh market
by
Sparks
408
and about
14% in
developing countries
by
Steppe.
410
The main causes
of
losses are physical
injuries, physiological disorders

or
diseases
dur-
ing storage,
96
and
improper conditions
for pro-
1040-8398/94/$.50
© 1994
by
CRC Press,
Inc.
cessing.
In
most,
if not all
cases,
the
observed
symptoms correspond
to a
discoloration
of the
apple
or
the apple products. Thus, bruising, which
is
the
most common defect

of
apples
186
seen
on
the market, results
in a
flattened area
on the
side
of the fruit with
the
flesh browned beneath.
352
-
356
Similarly, bitter pit,
161
superficial
171
and
senes-
cent scalds, internal (senescent
or
low-tempera-
ture) breakdown,
15
-
166
watercore,

239
and
core
flush
223
all
result
in
browning
of
either the skin
or
the internal tissue. Finally, juice, puree, and apple
slices that
are
badly processed brown intensely
and give the final products
a
bad appearance which
is rejected
by
the consumer.
74287
-
317
These general
discoloration phenomena
are
mainly related
to

enzymatic browning. However, browning can also
originate from nonenzymatic reactions such
as
the Maillard reaction,
16
-
106217
which occurs mainly
in heat-processed apple products.
228
-
244328
-
462
Basically, enzymatic browning
can be de-
fined
as an
initial enzymatic oxidation
of
phenols
into slightly colored quinones.
13
-
47
-
101
-
185
-

251
-
350
-
470
These quinones are then subjected
to
further reac-
tions,
enzymically catalyzed
or
not, leading
to the
109
Downloaded by [North Carolina State University] at 17:29 17 December 2012
formation of pigments. The colors of the latter
differ widely in hue and intensity, following the
phenols from which they originate and the
environmental factors of the oxidation reac-
tion
_50.113,184,243,298,308-311,325,351,395,463 Enzymatic
browning is mainly associated with polyphenol
oxidases, which are able to act on phenols in
the presence of oxygen.
253
"
256
-
338
-

350
-
489
-
510
Two
kinds of enzymes are classified under this trivial
name.
The first class, catechol oxidases
(E.C.
1.10.3.1),
catalyze two distinct reactions (Figure 1): the
hydroxylation of monophenols in o-diphenols
(reaction 1) and the oxidation of o-diphenols in
o-quinones (reaction 2). These two enzymatic
reactions consume oxygen and are referred to
as monophenolase (or cresolase) activity and
o-diphenolase (or catecholase) activity, respec-
tively. The former activity is not always present,
and when both activities are present, the ratio of
cresolase/catecholase activities varies widely from
1 to 10 or even
40.
34
-
443
The second class, laccases (E.C.I. 10.3.2),
oxidizes o-diphenols as well as p-diphenols (Fig-
ure 2), forming their corresponding quinones.
Besides other differences in properties,

256
-
475
the
unique ability to oxidize p-diphenols can be used
to distinguish laccase activity from that of the
first class.
The nomenclature of these enzymes
176
is some-
what confusing because besides the two numbers
E.C.I.10.3.1 and E.C.I. 10.3.2, a third one exists,
E.C.I.14.18.1.
It is referred to as monophenol
monooxygenase (tyrosinase) and corresponds to
OH
CRESOLASE
1/2 0,
OH
CATECHOLASE
,0H V2 0
2
H
2
0
(2)
FIGURE
1. Reactions
catalyzed
by

polyphenoloxidases
(E.C.
1.14.18.1
and
E.C.
1.10.3.1).
(1)
Hydroxylation
of
monophenol
to
o-diphenol;
(2) dehydrogena-
tion
of
o-diphenol
to
o-quinone.
FIGURE
2. Reactions
catalyzed
by
laccases
(E.C.
1.10.3.2).
110
Downloaded by [North Carolina State University] at 17:29 17 December 2012
the same enzymes
as
E.C.I.10.3.1, which always

catalyze
the
hydroxylation
of
monophenols.
The peroxidases (E.C.I. 11.1.7)
can
also
be
considered
as
participating
in
enzymatic brown-
ing.
These enzymes, whose primary function
is to
oxidize hydrogen donors
at the
expense
of
perox-
ides,
are
highly specific
for
hydrogen peroxide.
On
the
other hand, they accept

a
wide range
of
hydrogen donors, including polyphenols.
36
-
342
-
343
-
443
The primary products
of
oxidized phenols
are
probably quinones similar
to
those obtained with
polyphenol oxidases. Although peroxidases
are
distributed widely, especially
in
plants, they
gen-
erally appear
to be
little involved
in
enzymatic
browning

of
fruits
and
vegetables following
a
mechanical stress.
The
explanation could
be
that
the peroxidase activity
is
limited
by the
internal
level
of
hydrogen peroxide. However, their
in-
volvement
in a
slow process such
as
internal
browning
is
possible.
423424
Nevertheless,
the di-

rect involvement
of
peroxidase
in
enzymatic
browning
of
apple
and
apple products remains
questionable,
as
does that
of
laccases.
The
latter
enzymes
are
mainly present
in
fungi
and in cer-
tain higher plants
256
and are
almost always absent
in sound fruits
and
vegetables, with

the
exception
of peaches
146
and
apricots.
87
Therefore,
the
bulk
of
the
discussion below concentrates
on the two
main factors
of
apple enzymatic browning, that
is,
phenolic substrates
and
polyphenoloxidase
(E.C.I.10.3.1) activity, with
a
small part devoted
to other factors such
as
peroxidase
and
ascorbic
acid.

The
different ways
to
estimate browning,
the correlations among
the
intensity
of
browning
and
its
causal factors,
and,
finally,
the
control
of
browning
are
then examined successively.
II. PARAMETERS OF
ENZYMATIC
BROWNING
IN APPLE AND APPLE
PRODUCTS
A.
Phenolic Compounds
of
Apple
and

Apple
Products
1.
Total
and
Individual
Phenolics
in
Ripe
Apple
Fruit and
Apple
Juice
A great number
of
works have been devoted
to
the
study
of
phenolics
in
apple
and
apple prod-
ucts.
Among
the
several classes
of

plant pheno-
lics,
six
classes
are
present
in
apple fruit:
hydroxycinnamic derivatives, flavonols, antho-
cyanins, dihydrochalcones, monomeric flavan-3-
ols,
and
tannins.
236
Since
the
first modern
monograph published
by
Harborne
142
-
144
in 1964,
considerable information
has
been compiled
on
the classes
of

these phenolics,
135
-
169
-
238
-
321
-
411
-
454
as
well
as on
methodologies
for
their purification,
isolation, quantitation,
and
structure elucida-
tion i
27
.
143
.
162
^™^.
455
The quantitation

of
total phenolic compounds
is
not
accurate because
of the
extraction method,
which often cannot guarantee
a
total solubiliza-
tion
of all
phenolics,
and to the
assay method,
which
is
unavoidably
a
compromise
on the
reac-
tivity
or
absorption characteristics
of the
different
classes
of
phenols.

27
-
28
-
81
-
234
-
396
-
398
-
413
Nevertheless, these measurements still
are
used
for
rough comparisons among samples.
Indeed,
a
wide variability
is
observed
in the
total phenolic contents
of
ripe fruits
and
apple
juices (Table

1).
Obviously, this variation
can be
partly attributed
to the
different methods used
by
the authors
for the
quantitation
of
phenols.
How-
ever, even when
the
same method
is
used, consid-
erable variation
in
results
is
still evident.
69150
-
448
As
in
other fruits,
the

levels
of
phenolic
com-
pounds
in
apple
are
highly dependent
on
many
factors, such
as
variety, stage
of
maturity,
and
environmental factors.
236
-
491
Besides quantitative
variations among
the
classes
of
phenolic
com-
pounds, large variations
are

also apparent
in the
qualitative distribution, which
are
caused
by ge-
netic
and
external factors. Modern methods,
par-
ticularly high-performance liquid chromatography,
considerable progress
in
both
the
separa-
tion
and
quantitation
of
individual phenolics
in
apple fruit.
The
main phenolics identified
in
apple
fruit
and
apple products

are
given
in
Table
2, and
the contents
of the
different classes
are
given
in
Table
3 for the
ripe fruits.
The
tables show that:
1.
In the
cortex, three phenolics accounted
for
more than
90% of the
total phenolic
con-
tent
for
most cultivars:
one
caffeoyl quinic
acid (chlorogenic acid)

and two
flavan-3-
ols ((-)-epicatechin
and
procyanidin
B2).
2.
In the
peel,
the
hydroxycinnamic acid
derivatives
are not as
high
as in the
cortex,
111
Downloaded by [North Carolina State University] at 17:29 17 December 2012
TABLE
1
Total
Phenolic Compounds in
Ripe
Apple
Fruit
and Apple Juice
Samples
7 varieties
8
10 varieties'

1
400 varieties'
1
7 varieties'*
2 varieties'"
8
varieties
8
3
varieties"
Various
varieties
6
8
varieties
6
8
varieties'
12 varieties^
5
varieties
6
6 varieties
0
3
varieties
(x 2 technologies)'
5
varieties
0

5
varieties'
22 varieties
3
10 varieties'
4 varieties'
2
varieties
(x 6 technologies)*
Industrial
concentrate (72°
Brix)
e
Ripe
Fruit
Peel Pulp
8.7-19.2
1.5-3.8
0.9-2.1
0.6-1.2
0.2-1.9
0.5-0.7
Apple
Juice
Total
0.7-2.5
0.6-2.0
0.5-17
0.15-0.21
0.6-0.8

1.0-1.9
0.49-0.84
0.5-11
0.16-0.44
0.15-0.58
0.24-0.62
0.14-0.37
0.21-15.2
0.37-0.62
0.02-0.14
0.3-2.7
1.49
Ref.
69
448
157
516
449
381
288
450
323
193
150
172
464
302
407
407
63

64
74
126
14
Note: Values are
given
in grams per
kilogram
(FW
ripe
fruit)
or in grams per
liter
(juice).
a
Chlorogenic
acid
equivalent.
b
o-Diphenols
(chlorogenic
acid
equivalent).
c
Gatechin
equivalent.
d
Tannins.
8
Not

given.
'
Gallic
acid
equivalent.
a
o-Diphenols
(catechol
equivalent).
h
Tannic
acid
equivalent.
whereas the flavan-3-ol and flavonol
derivatives constitute the major part of the
phenolic compounds.
3.
In the hydroxycinnamic derivatives, quinic
acid is the only hydroxyacid that forms
esters with caffeic
403
and /7-coumaric
41
-
492
acids.
Although not detected in the original
fruit, 5'-feruloyl quinic acid has been
reported in cell suspensions from apple
fruit.

195
In the latter case, it accumulates in
the latter stage of cell culture growth after
the exponential growth phase. Similar
behavior was shown for the sinapoyl
glucose ester, which was observed only in
in vitro cell suspension cultures from apple
parenchyma.
195
-
307
In apple fruit, besides
quinic acid, hydroxycinnamic acids form
esters with sugars and, more precisely, with
glucose. Thus, Macheix
231
-
232
reported that
1-0-p-coumaroyl glucose was one of the major
p-coumaric derivatives in apple (var. CalviUe
blanc) together with several p-coumaroyl
112
Downloaded by [North Carolina State University] at 17:29 17 December 2012
TABLE
2
Main
Phenolics
Identified
in

Apple
Fruit and
Apple
Products
(According
to
Reference
236,
with
Modifications)
Phenolic Acids
4
'
95
'
107
-
112
-
165
'
229
'
232
'
235
'
266
'
279

'
306
'
336
'
403
'
487
'
490
'
492
'
497
5'-caffeoylquinic,
a
4'-caffeoylquinic,
3'-caffeoylquinic,
5'-p-coumaroylquinic, 4'-p-coumaroylquinic,
3'-p-coumaroylquinic, p-coumaroylglucose, caffeoylglucose, feruloylglucose
Flavan-S-Ols
4
'
107
-
112
-
119
'
175

'
203
'
211
'
229
'
266
'
279
'
283
'
336
'
380
'
385
-
392
-
429
-
453
'
480
'
481
'
497

(-)-epicatechin,
(+)-catechin,
B1, B2, B5,
C1, higher
polymeric
forms
Flavonols
85
'
86
-
294
'
306
-
391
>
425
Quercetin-3-O-p-D-galactopyranoside (hyperin)," Quercetin-3-O-p-o-glucopyranoside
(isoquercitrin),
Querce-
tin-3-O-p-D-xyloside
(reynoutrin),
Quercetin-3-O-a-L-rhamnopyranoside
(quercitrin),
Quercetin-3-O-oc-L-arabino-
furanoside
(avicularin),
Quercetin-3-O-rutinoside
(rutin)

Dihydrochalcones
85
-
86
-
95
-
266
-
294
'
497
Phloretin-2'-O-glucoside
(phlorizin),
Phloretin-2'-xyloglucoside
AnthOCyaninS
c
94,152,163,260,412,426,427,428
Cyanidin-3-galactoside (ideain), Cyanidin-3-glucoside
(kuromarin),
Cyanidin-3-xyIoside, Cyanidin-3-arabino-
side;
acylated derivatives
of
Cyanidin-3-galactoside, Cyanidin-3-arabinoside, Cyanidin-3-glucoside,
and
Cyanidin-3-xyloside
Note:
The
major compounds

in
each
class
are
written
in
boldfaced type.
a
The
IUPAC
recommendations
177
were applied
for the
nomenclature
of the
hydroxycinnamic
quinic
esters.
Thus,
5'-caffeoylquinic
is
chlorogenic acid,
4'-caffeoylquinic
acid
is
cryptochlorogenic acid, and
3'-caffeoylquinic
acid
is

neochlorogenic acid.
b
A
structure
of
quercetin-3-O-a-galactoside
was
proposed
by
Teuber
and
Herrmann.
425
0
Present
in
apple
cultivars
with
red-colored skins.
4.
quinic esters. Similarly,
in
their compilation,
Risch
and
Herrmann
336
indicated that
the

glucose esters
in
apple were mainly
p-coumaroyl
and
feruloyl glucose esters.
Glucoside derivatives
in
which
the
phenolic
group
is
engaged
in
bonding with glucose
are
not
present
in
apple fruit, although they
are often encountered
in
plants.
159
Finally,
according
to
Macheix
et

al.,
237
the
overall
balance
of
the three hydroxycinnamic acids
(either
as
quinic esters
or as
glucose esters)
ranges between
75 and
94%
for
caffeic acid,
5 and 20% forp-coumaric acid,
and 1 to 5%
for ferulic acid.
In
the
flavan-3-ol derivatives, procyanidin
B2 (dimer
of two
(—)-epicatechin units
linked
in a
bond between
C4, the

"upper
unit",
and C8, the
"lower unit")
and
(-)-epicatechin
are the
most abundant.
35
-
306
The (+)-catechin isomer
is
also always
present
in
apple fruit, although sometimes
only
as
traces.
336
According
to
Mosel
and
Herrmann,
283
the
mean content
of

(-)-epicatechin
in
apples
is
five times
higher than that
of
(+)-catechin. The latter
authors indicated that (+)-gallocatechin and
(—)-epigallocatechin, which
are
normally
absent
in
apples, have been found
in
some
cultivars
283
-
284
in
relation
to
certain climatic
factors
and
growing conditions.
The
data

concerning
the
polymeric forms
of
flavan-
3-ols
("tannins")
are
scarce, probably owing
to
the
difficulties encountered
in
their total
extraction and precise quantitation. Almost
113
Downloaded by [North Carolina State University] at 17:29 17 December 2012
TABLE
3
Content
in
Chlorogenic
Acid
(CG),
Other
Hydroxycinnamic
Derivatives
(HCd),
Flavan-3-Ols
(F3OI),

Flavonols
(FVOd), and
Dihydrochalcones
(DHC) of
Ripe
Apple
Fruit and
Apple
Juice
Samples
Compilation
7
var.
5
var.
3
var.
16
var.
8
var.
11
var.*
7
var. (pulp)
5
var. (pulp)
8
var. (pulp)
3

var. (pulp)
11
var.
(pulp)*
12
var. (pulp)*
7
var.
(peel)
5
var.
(peel)
8
var.
(peel)
3
var.
(peel)
16
var.
(peel)*
11
var.
(peel)*
12
var.
(peel)*
4
Com.
samples

8
var.
5
var.
5
var.
CG
5-51
2.6-13.6
15-33
6-51
4.3-19
a
18-173
19-55
4.2-32
9-^3
4-6
18-171
18-171
0.1-20
2-15
3-6
5-150
3-89
0.25-0.91
6.6-23.2
0.73-25
0.3-5.9
HCd

F3OI
Ripe
Fruit
0.5-11.2
—
— 25-65
— 4.4-10.6
0.7-5.2
1.9-2.6"
11-51
0-5.5
0.35-^.1
15-50
12-56
0.96-4.9
4-33
0.2-12
3.5-16
90-261
33-124
3-21.5
13-30
5-29
98-214
98-214
230-650
19.4-81.5
45.5-143
25-127
100-960

50-450
120-575
Apple
Juice
0.23-0.45
tr-0.16
3.2-20
0.4-3.3
0.3-46.5
0.1-2.5
FVOd
0-2.9
6-26
4-15
tr-3.8
0.02-O.24
16-52
15.8-142
154-285
78-107
120-550
81-353
0.23-0.73
0-0.6
DHC
0.4-1.9
9-29
0.3-1.6
1.1-2.5
2-4

0.6-65
8.7-33
18-33
8-220
7-211
Ref.
236
69
172
158
336
283
5,7
322
305,
306
266
35
4
12
322
305,
306
85,
266
35
4
436
12
1.4-2.7

108
—
33
0.48-3.5
80
0.1-1.6
407
Note:
All
results
are given
in
milligrams per 100
g FW
basis
or in
milligrams per 100
ml
(juice)
except
those
marked
by an
asterisk
(*), which are given
in
milligrams per 100
g
DM
basis.

a
expressed
as
caffeic
acid
equivalent
(contains
all
the caffeoyl
esters).
b
expressed
as
p-coumaric
and ferulic
acids
(contains
all
the coumaroyl and
feruloyl
esters).
all
the
data
are
concerned with apple juices
and ciders, owing
to the
interest
in the

sensory
and
technological properties
of
these products.
76
-
203
-
208
-
209
-
212
-
280
-
282
-
302
-
407
-
451
-
452
A considerable range existed
in the
total
tannin content

of
apple juices.
80
-
207
-
210
Thus,
large differences were found
in the
phenolics
of
juices obtained from
a
dessert
apple (var. Bramley)
and a
cider apple
(var.
Dabinett).
208
However,
the
largest difference
was associated with
the
tannin fraction
as
the Dabinett juice contained twofold more
phenolic acids, fivefold more phloridzin,

and
tenfold more epicatechin,
but
20-fold more
procyanidin
B2
than
the
Bramley juice.
208
5.
In the
flavonol derivatives,
the
only
characteristic feature
is the
presence
of
quercetin
as
aglycone. Methods have been
proposed
for
the certification
of
apple juices
based
on
their flavonol contents.

92
-
93
Although kaempferol derivatives have been
reported
by Van
Buren
450
and
Herrmann,
158
their presence
has
never been confirmed
since then. Similarly,
the
presence
of a
114
Downloaded by [North Carolina State University] at 17:29 17 December 2012
6.
diglucoside quercetin found only by Fisher
1
•'
seems unlikely. In apple fruit, the
diversity is at the sugar level, because six
different quercetin-3-glycosides have
been characterized fully. They are all of
a pyranose form with the exception of
arabinose, which has a furanose form. The

amount of each giycoside derivative
(galactose, xylose, glucose, arabinose,
rhamnose, and rutinose) and their relative
balance are highly dependent on variety
(Table 4). According to Teuber and
Herrmann
425
and Oleszek et al.,
294
hyperin
(the galactoside derivative) is the most
abundant.
In the dihydrochalcone derivatives, two
compounds have been characterized in apple
and apple products. For a long time, it was
stated that phloridzin (phloretin-2'-0-
glucoside), a characteristic compound of the
genusMalus (Rosaceae),
42
-
468
-
494
was present
in leaves, stems, and seeds but absent in
fruits.
493
However, a first description of
phloridzin was reported in the ethyl acetate
7.

extract of the core tissue of Mclntosh apples
95
and the methanol extract of peel of Democrat
apples.
111
Then, using HPLC, this compound
was fully characterized in the different parts
of apple fruit
85
-
305
and in apple juice.
497
In
the same extracts, another phloretin giycoside
was also characterized corresponding to the
xyloglucoside
294
as well as in Golden
Delicious apple juices and jams.
431
-
432
The
phloridzin content varied from 87 to 331 fig/g
of fresh peel in eight cultivars grown in
Canada
266
and from 21 to 105 |Xg/g for five
cultivars grown in Spain.

306
In the cortex,
levels were not as high as in the peel, as they
ranged from 0.1 to 0.25 |Llg/g for the latter
cultivars
306
on a fresh-weight basis and from
10 to 290 (ig/g for 11 French cultivars on a
dry-weight basis.
7
In the anthocyanin derivatives, cyanidin was
the only aglycone found in apple cultivars
with red-colored skin. There is a general
agreement to indicate that ideain (the
cyanidin-3-galactoside) is the major pig-
ment,
412
-
468
because it represented more than
TABLE
4
Quercetin
Giycoside
Concentration
in
Apple
Peel (tng.kg-
1
FW

Basis)
for
Three
Cultivars
Grown
in U.S.,
Eight
Cultivars
Grown
in
Canada,
266
Five
Cultivars
Grown
in
Spain,
306
and One in
Germany
425
Cultivar
Golden
Delicious
Empire
Rl
Greening
Red
Delicious
Spartan

Cortland
Jerseyman
Mclntosh
Golden
Delicious
Gravenstein
Northern
Spy
Starking
Red
Reinette
Golden
Delicious
Verde
Doncella
Granny
Smith
Golden
Delicious
Rhamnoside
230
200
220
104
71
80
128
361
369
440

661
69.7
57.9
124
131
347
73
Rutinoside
—
—
117
57
159
185
117
110
139
169
0.5
1.1
3.3
15
19.2
3
Xyloside
100
110
150
210
244

192
119
254
195
145
297
—
—
—
—
—
34
Galactoside
+
glucoside
360
330
. 500
604
783
735
554
696
869
594
924
91
78
428
613

954
220
a
Arabinosfde
130
140
200
546
662
589
556
777
539
508
801
21.1
21.3
29.1
47.7
105
75
Ref
35
35
35
266
266
266
266
266

266
266
266
306
306
306
306
306
425
a
Only
the galactoside
derivative
was
assayed.
115
Downloaded by [North Carolina State University] at 17:29 17 December 2012
88%
of the anthocyanins in 11 cultivars
grown in England
427
and approximately 40%
in the cultivar Scugog grown in Canada.
260
The presence of a cyanidin-7-arabinoside
was postulated by Sun and Francis.
412
How-
ever, later on, Timberlake and Bridle
427

raised
some doubt as to the presence of derivatives
with sugars linked in the 7 position of
cyanidin. They indicated that all the deriva-
tives are esterified by sugars in the 3 posi-
tion. Moreover, the same authors
427
have
found minor pigments corresponding to
acylated forms of cyanidin-3-monoglyco-
sides,
but the nature of the acids involved in
acylation was not determined.
As already stated, several factors can induce
large variations in both the quantitative and quali-
tative content of phenolic compounds in apple.
2.
Phenolic
Variations
at the
Subcellular
Level
At the subcellular level, the phenolics are
located mainly in the vacuoles. Yamaki
505
indi-
cated that 97% of the total phenolics present in
apple cells accumulate in vacuoles, while 3% are
in free space and none in cytoplasm. According to
the same author,

505
the calculated concentration
of phenols is higher than 0.1 M in vacuoles of
immature fruit flesh compared with the 1 to 10 mAf
usually found in mature apples.
450
3.
Phenolic
Variations
at the Tissue
Level
Although the subcellular distribution of
soluble phenolics appears to be homogenous, the
situation is different at the tissue level. Thus, it
can be seen in Tables 1 and 3 that the epidermal
and subepidermal layers (peel) have a higher con-
tent of phenolics than the internal tissue (cortex).
In different cultivars, the peel/cortex phenolic
content ratio ranged from 3 to
lO.
4
.
35
.
167
-
266
.
305
.

306
-
322
In a more precise study on the Calville Blanc
variety, Macheix
232
-
234
divided apple fruit into four
zones, from the outer to the inner parts of the
fruit, corresponding to (1) the peel, (2) the outer
part of the cortex (the major edible portion of
apple),
(3) the circular zone surrounding the car-
pels,
and (4) the central core, respectively. Zone
1 was the richest in chlorogenic acid, catechins,
and flavonols. The flavonol content was less than
40 mg/kg (FW basis) in zones 2 to 4 compared
with the 178 mg/kg in zone 1. The chlorogenic
acid content was the lowest in zone 2 and in-
creased slightly in zones 3 and 4. Although less
important than in the peel, the catechin content
was higher in zone 3 than in zones 2 and 4.
Concerning chlorogenic acid content in the peel,
this last result was slightly contradictory to the
general comments given in Section
2.1.1.
for
Tables 2 and 3. This probably comes from differ-

ences in the repartitions between peel and cortex
used by the authors. Risch and Herrmann
336
also
reported a greater amount of chlorogenic acid in
the core than in the outer part of the cortex for the
Jonathan cultivar. Similarly, Harel et al.
150
found
a nonuniform concentration of o-diphenols in the
flesh of the Grand Alexander cultivar. They re-
ported that the amount of phenolics was highest
in the peel, lowest in the outer part of cortex, and
gradually increased toward the core. This uneven
spatial distribution of phenolic compounds in apple
fruit flesh is important, as it can induce different
tissue sensitivities to enzymatic browning.
4.
Phenolic
Variations
at the
Cultivar
Level
A considerable variation was observed in the
content of both total and individual phenolics
among different cultivars of apple. Thus, in their
compilations, Van Buren
450
and Herrmann
157

in-
dicated ranges of 1 to 11 and 1 to 34, respectively,
in the total phenolics content among varieties.
Similarly, concerning the main individual
phenolics, a variation of 1 to 10 was given for
chlorogenic acid by Macheix et al.
236
and for
(—)-epicatechin by Risch and Herrmann.
336
Simi-
lar variations (1 to 9) were also found for fla-
vonols in apple peel.
305
Apple cultivars with green-
and yellow-colored skin are pigmented by chloro-
phylls, carotenoids,
23
and quercetin derivatives
503
and are obviously devoid of anthocyanins. Never-
116
Downloaded by [North Carolina State University] at 17:29 17 December 2012
theless,
a large variation (1 to 3) was also ob-
served for ten Delicious apple strains with red-
colored skin grown in the
U.S.
394
Interestingly

enough, the latter authors
394
found a correlation
between the peel luminance L*, measured by a
portable tristimulus colorimeter, and the antho-
cyanin level of deep-red-colored fruits.
5.
Influence
of
Maturity
and
Postharvest
Storage
Although, as already mentioned, the assay
methods for total phenol content are not accu-
rate,
there is general agreement that the con-
centrations of phenolic compounds are very high
in young fruits and then rapidly decrease during
fruit development.
236
Numerous studies carried
out on the phenolic content of developing apples
reflect this general trend.
69
-
150
-
157
-

229
-
466
-
488
-
516
The
phenolic concentrations decline sharply 1 month
after the petal drop, reaching a low level
10 weeks later, and remain approximately
constant thereafter.
150
-
488
On a fruit basis, the
change in total phenols is less pronounced dur-
ing the 4- to 14-week period after the petal
drop.
Because in that period the number of cells
is fixed, the decrease in the concentration of
total phenols is primarily the result of a dilution
of phenolic compounds in the vacuole.
236
After
harvest, the concentration of total phenols
remains essentially constant or decreases
slightly.
35
-

69
-
150
-
172
-
229
-
445
-
448
The changes in content of the different classes
of phenols and of some individual phenols have
also been followed during apple fruit develop-
ment and its subsequent storage after harvest.
Thus,
during apple development and 1 month
after the petal drop, a general decrease in the
hydroxycinnamic (caffeic, p-coumaric, and feru-
lic) acid derivatives was observed.
284
An example
is shown in Figure 3A concerning changes in the
levels of /j-coumaroylquinic acid, p-coumaroyl-
glucose, and chlorogenic acid (by far the most
important).
233
For the latter phenolic compound,
similar results were given by several au-
thors.

35
-
69
-
230
-
466
During cold storage, the variations
in the content of hydroxycinnamic acid deriva-
tives were less important and, depending on the
cultivars, the authors indicated a decrease,
283
-
284
a
constant level,
69
or fluctuations.
172
-
445
On a fruit
basis (Figure 3B), after the initial and rapid rise
in chlorogenic acid (the same was observed for
p-coumaroylquinic acid and p-coumaroyl-
glucose), this phenol steadily accumulated, and it
was only after 10 weeks that its amount decreased
slightly.
233
Depending on the cultivars, this last

decline was not observed
283
-
284
or only during cold
storage.
69
-
172
The variations in catechins were simi-
lar to those obtained with hydroxycinnamic de-
rivatives. The concentrations in catechins rose
sharply during the 30 to 40 d after flowering and
then decreased rapidly to stabilize at a low level
in mature fruit
230
-
283
-
284
and throughout the storage
period.
35
On a fruit basis, a later peak was ob-
served and the decrease was delayed to the end of
the maturation period.
283
Moreover, in two apple
varieties, an increase in the (-)-epicatechin-to-
(+)-catechin ratio was observed during the pro-

gressive growth of the fruit.
284
Only a few works
have been devoted to the variations of quercetin
and phloretin glycosides. The quercetin glyco-
sides per gram of fresh weight remained fairly
constant during the 2 months preceding the com-
mercial harvest (var. Golden Delicious and Grimes
Golden) but, because of fruit growth, increased
on a per fruit basis.
503
Moreover, inside the com-
mercial harvest, the same author
503
indicated that
the flavonol content of green fruit was only half
of that of yellow fruit (var. Golden Delicious). In
that case, no change was observed in the total
flavonol content during storage. However, for
Mclntosh apples, a two- to threefold increase in
the levels of quercetin glycosides was found during
the first 2 months of cold storage.
85
A similar
trend was found for phloridzin.
85
Finally, although
the total quercetin glycosides remained relatively
constant until the climacteric, considerable varia-
tions were observed among the individual

glycosides.
85
6.
Influence
of
External
and
Internal
Factors
during
Cultivation
and Storage
As in other fruits, the regulation of phenolic
metabolism in apple depends greatly on both ex-
ternal and internal factors (light, temperature,
ethylene,
growth regulators, nutrients, pesticides,
117
Downloaded by [North Carolina State University] at 17:29 17 December 2012
120-
a
b
•300
c
•200y
•100
y
<.
———
b

——•
\
C
\a 1
A
d /
•>/
d
100-
50-
e
*
11
31 10 30 10 30
May
June
July
May
June
FIGURE
3.
Changes
in the
levels
of (a)
p-coumaroylquinic
acid,
(b)
p-coumaroylglucose,
(c)

chlorogenic acid,
and
(d) fresh weight
(in
grams) during apple growth.
(A) milligram
per
100 g FW; (B) milligram per fruit.
(From Macheix,
J. J.,
Physiol. Veg.,
12, 25,
1974.
With
permission.)
etc.),
as
demonstrated
by
numerous studies. Prob-
ably because
of
its
impact
on
the
visual quality
of
the fruit, most studies dealt with anthocyanin
bio-

synthesis
in
red-colored apple skins.
Although
the
accumulation level
of
antho-
cyanin
is
genetically controlled,
the
requirement
of light
for red
pigment biosynthesis
in
apple
skin
has
been emphasized
by
many work-
ers
9.44.154.197.326.393
7^
effect
of
temperature
on

anthocyanin biosynthesis, which
has
been
rec-
ognized
for a
long time,
70
depends largely
on
the
fruit maturity stage.
10
-
103
Thus,
for the
Jonathan
cultivar,
the
optimum temperature increased from
12°C
in
unripe fruit
to 16 to
24°C
in
ripe fruit.
103
118

Downloaded by [North Carolina State University] at 17:29 17 December 2012
For these factors, anthocyanin accumulation ap-
peared to be correlated with phenylalanine am-
monia lyase (PAL) activity.
415416
Application of
ethylene
104
or ethylene-releasing compounds
414
to
unripe apples resulted in anthocyanin accumula-
tion to levels similar to those in ripe apples. The
same held for the PAL activity when unripe apples
were ethylene treated.
104
However, with ripe fruits,
the ethylene treatment had no effect on both fac-
tors,
suggesting that during apple ripening, antho-
cyanin accumulation is under the control of the
PAL activity.
105
In addition, treatment with syn-
thetic auxins such as a-napthalene acetic acid and
2-[2,4,5-trichlorophenoxy] propionic acid in
combination with Ethephon® (an ethylene-releas-
ing compound) and Alar® (a growth retardant)
enhanced the red color of the skin of some U.S.
cultivars.

30
-
97
-
136
Besides anthocyanin, cultural prac-
tices can also affect other phenolics. Thus, a sig-
nificantly lower amount of phenols was found in
apples treated with Melprex 65® (a pesticide prepa-
ration containing dodine).
141
Similarly, ciders
obtained from "fed" (NPK fertilizer) Dabinett
apple trees were less bitter and astringent than
those from "unfed" trees, which was related to an
overall decrease of 17% in fruit phenolic concen-
tration.
210
During cold storage, carbon dioxide levels
greater than 73% in an in-package, modified at-
mosphere severely destabilized anthocyanins of
the skin of Starkrimson apples after 23 weeks at
2°C and 76% humidity.
222
B. Polyphenol
Oxidase
in Apple and
Apple
Products
1.

Assay
of Activity
The correct assay
of
polyphenol oxidase
ac-
tivity
is
obviously
a
need,
and, in
this respect,
two
problems have
to be
overcome.
The
first lies
in
the choice
of
assay method
and the
second
is to
properly differentiate between catecholase
and
laccase
on the one

hand
and
peroxidase
on the
other hand.
As already stated,
the
primary products
of the
oxidation
of
phenols
are
highly unstable
and un-
dergo many secondary reactions with both phenols
and proteins.
5230
*"
310
Therefore,
it is
difficult
in
routine assays
to
measure precisely either product
formation
or
phenol consumption

due to the en-
zymatic catalysis. Undoubtedly,
the use of
spec-
trophotometry
to
follow quinone formation
is the
easiest method.
254
-
271
However,
it is in
some ways
an inaccurate method owing
to
side reactions
(re-
sulting
in
nonlinear color formation with time)
and enzyme inactivation through
a
suicide mecha-
nism.
463
Moreover,
the
extinction coefficients

of
the enzymically produced quinones
are not al-
ways accurately known.
463
To
circumvent these
problems, some authors have proposed coupled
assays
in
which quinone accumulation
is
avoided.
This
is
carried
out
either with ascorbic acid,
as in
the "chronometric method",
78
'
313
or
with
com-
pounds forming stable adducts with different
spectral properties such
as
proline,

354
2-nitro-5-
thiobenzoic acid,
102
and
Besthorn's hydra-
zone.
259
-
312
Nevertheless, according
to
Mayer
et al.,
257
the
polarographic measurement
of oxy-
gen uptake
148
-
249
is the
most convenient
and
accu-
rate method
for
determining polyphenol oxidase
activity. However, although this method

has
replaced
the
manometric method,
467
it
suffers
some drawbacks.
The
ratio
of
oxygen consumed
to phenol oxidized changes with
the
reaction
time
and
depends
on the
phenolic structure
and concentration,
pH, and
buffer solution
used.
53
-
58
-
120
-

243
-
257
-
308
-
333
-
335
-
341
The
reaction velocity
is dependent
on
oxygen concentration.
504
Because
the
Km O
2
of
polyphenol oxidase
is in the
0.1-
to
0.5-mM range,
180
-
254

the
enzymes
are not
saturated
by oxygen during
the
assay
and the
true values
of
Vmax
are
seldom determined. Thus, when using
the polarographic method,
it is
essential
to
care-
fully control
the
assay conditions (temperature,
pH,
type
of
buffer,
and air
saturation)
and to
measure only
the

initial rate
of
oxygen uptake.
The unique ability
to
oxidize p-diphenols
(andp-diphenylene diamine)
is
often used
as a
test
for the
presence
of
laccase activity.
How-
ever,
it is not
sufficient,
at
least
in
crude
ex-
tracts.
In the
latter,
the
occurrence
of

endogenous
o-diphenols could provoke coupled oxidation
of
added p-diphenols
by
o-diphenolase, leading
to
the false conclusion that laccase
is
present.
Therefore, additional tests with specific inhibi-
tors
of
each activity
are
required. Thus, phenyl-
119
Downloaded by [North Carolina State University] at 17:29 17 December 2012
hydrazine,
219
salicylhydroxamic acid,
3
2,3-naptha-
lenediol,
258
and cinnamic acids
475
are specific
inhibitors of o-diphenolase, whereas cetyl-
trimethylammonium and other quaternary

ammonium compounds
469
-
475
inhibit p-dipheno-
lase.
In crude extracts, it is also preferable to check
the possible interference of peroxidase activity.
This is easily performed by adding catalase and
ethanol in order to remove residual peroxides
from the reaction medium
258
or by using
tropolone.
188
The latter compound is a very effec-
tive inhibitor of polyphenol oxidase
189
and can
serve as a substrate for peroxidase in the presence
of hydrogen peroxide.
190
2.
Localization
There is general agreement that polyphenol
oxidase is predominantly a plastid enzyme in
higher plants.
155
-
253

-
254
-
430
-
456
In nonsenescent tis-
sues,
it is mainly located on the thylakoid mem-
brane of chloroplasts and in vesicles or other
bodies in nongreen plastid types.
457
However, some
authors reported that polyphenol oxidase can also
exist in mitochondrial fractions or is readily
soluble (nonmembrane associated) in plant
cells.
38
-
40
-
148
-
256
-
401
In apple fruit, polyphenol oxi-
dase has been located both in organelles (chloro-
plast and mitochondria), where it may be tightly
bound to the membrane, and in the soluble fraction

of the cell.
83
-
148
As in many other fruits,
25
-
146
-
192
-
373
the proportion of readily soluble polyphenol oxi-
dase activity increases during the ripening of apple
fruit.
17
-
148
-
178
It has been proposed that polyphenol
oxidase was solubilized from the plastid and re-
leased to the cytoplasm, where the enzyme re-
mained either soluble or associated with the cell
wall.
17
A similar evolution was found with apple
tissue culture.
460
-

461
3.
Extraction
Two problems
are
encountered
for the
optimization
of the
extraction conditions
of
polyphenol oxidase, full solubilization
of the
membrane-bound activity
and
protection against
phenolic oxidation during
and
after extraction.
As already stated,
the
strength
of
polyphenol
oxi-
dase binding
to
membranes
is
variable. There-

fore,
in
most cases, full extraction
of the
activity
requires
the use of a
detergent such
as
Triton®
X100,
122
-
439
-
501
Triton® XI14,
360
-
361
or
SDS.
8
-
25
For
apple fruit, solubilization
was
achieved
by use of

a detergent, either Triton® X100
83
-
148
-
149
-
181
-
409
-
474
or digitonin,
145
or
after preparation
of an
acetone
powder.
132
-
133
-
289
-
349
-
376
-
387

-
405
-
434
-
509
Although
the lat-
ter method avoids
the use of a
detergent, which
may cause some problems during further purifi-
cation,
it
also undoubtedly results
in
modification
of
the
enzyme properties,
the
extent
of
which
has
not always been taken into account.
The second problem arises from
the
simulta-
neous presence

of
quinones
in
crude extracts
of
the enzyme
and its
endogenous phenolic
sub-
strates.
It is
therefore essential
to
avoid,
or at
least
minimize,
the
formation
of
quinones that then
react with proteins. Such reactions
may
result
in
activity losses
and the
formation
of
"new" artifac-

tual enzymatic forms
399
through covalent, hydro-
gen, hydrophobic,
and
ionic bindings between
proteins
and
phenolics
or
quinones.
224
-
225
-
264
Meth-
ods
to
prevent
the
reaction
of
polyphenol oxidase
with phenolics include
the use of
phenol-binding
agents such
as
polyethylene-glycol,

24
soluble
and
insoluble polyvinyl(poly)pyrrolidone
(PVP and
PVPP),
226
and
hydrophobic
227
and
anion
ex-
change
399
resins. However,
as all
plant materials
have different types
and
various amounts
of phe-
nolics, there
is no
universal method
to
effectively
remove phenols. Therefore,
it is
often advisable

to
add to the
extracting medium compounds that
prevent
the
formation
of
quinones
or
trap
the
quinones
as
soon
as
they
are
formed.
In
this
re-
spect,
the
most frequently used compounds
are
sodium diethyldithiocarbamate
308
(a
powerful
chelator

of
copper)
and
reducing agents such
as
ascorbic acid,
337
mercaptoethanol,
435
and cys-
teine.
19
-
241
-
242
Finally, protease inhibitors were
added
to
prevent
the
formation
of
artifactual
multiple forms
of
polyphenol oxidase
by
endog-
enous proteolytic enzymes.

117
-
501
In
crude enzy-
matic extracts
of
apple, PEG,
60
-
474
PVP,
59
-
132
-
133
-
474
cysteine
45
-
467
-
473
(either alone
or
with mercapto-
benzothiazole
409

),
and
ascorbic acid have been
used.
67
-
83
-
150
-
181
-
258
120
Downloaded by [North Carolina State University] at 17:29 17 December 2012
4. Purification
Although numerous works have been devoted
to the purification of this enzyme, only a few
polyphenol oxidases have been purified to appar-
ent homogeneity, probably owing to the difficul-
ties encountered in the preparation of active and
stable crude extracts. After an initial step, which
most frequently was ammonium sulfate fractional
precipitation, further purification involved the use
of one or several chromatographic steps. Apart
from routine chromatographic methods of separa-
tion (adsorption, gel filtration, and ionic
exchange),
32>121
-

292
'
337
hydrophobic chroma-
tography,
I18
-
182
-
500
isoelectric focusing,
174
con-
canavalin-agarose chromatography,
485
and
immobilized metal chelate chromatography
511
-
513
have been applied successfully to the purification
of polyphenol oxidase from several origins.
Surprisingly, although mushroom polyphenol oxi-
dase was probably the first enzyme to which the
affinity concept for chromatography was ap-
plied,
218
there were only a few applications of this
method for other polyphenol oxidases.
168

-
506
Apple polyphenol oxidase was only partially
purified from peel using DEAE cellulose chroma-
tography
83149
'
474
and calcium phosphate adsorp-
tion chromatography,
409
from cortex using
hydrophobic chromatography,
6
-
178
-
181
-
331
-
405
-
434
or
from the whole fruit using gel filtration.
45132
After
electrophoresis, the number of isoforms varied
from one

132
to two
376
-
474
or three,
67
-
149
-
181
which
Harel and Mayer
145
indicated could be because of
various degrees of subunit aggregation of the same
enzyme. Based on the Triton extracts, the pub-
lished purification factors were 260,
149
230,
409
and
150,
181
with a yield ranging from 55 to 35%. After
hydrophobic chromatography, Richard-Forget
331
further purified polyphenol oxidase from Red
Delicious cortex by immobilized metal chelate
chromatography followed by affinity chromatog-

raphy. The former method gave one major peak
containing 75% of the activity with a threefold
purification. Affinity chromatography using
p-coumaric acid as ligand on agarose (the gel was
synthesized by coupling p-coumaric acid to
hexamethylenediamine agarose
178
via an azo link-
age
66
'
72
-
73
) resulted in one purified fraction repre-
senting 25% of the crude extract activity and a
240-fold purification. Nevertheless, silver nitrate
staining after SDS electrophoresis revealed one
major band (90% of the coloration) still accompa-
nied by three faint, minor bands (10% of the
coloration). Finally, two major (60 and 35%) and
one minor (5%) fraction can be separated by ion
exchange chromatography at pH 6.5 using DEAE
sepharose CL6B.
331
This confirmed a previous
experiment of isoelectric focusing that gave two
major peaks with a pHi at pH 4.5 and pH 4.8 and
a minor one with a pHi close to pH 6.7.
181

Most
polyphenol oxidases from plants are 40- to
45-kDa proteins, as was first suggested by in vitro
translation experiments.
114
"
116
-
201
However, 60- to
68-kDa forms, containing a transit peptide, that
target the protein to the chloroplast have also
been purified from broad bean,
123
-
344
and recently
a gene coding for a 66.3-kDa polyphenol oxidase
was isolated from tomato.
386
Gel filtration resolved
apple enzyme activity into a single peak
132
-
181
with
a molecular weight of
26
132
or 46 kDa,

181
or into
three peaks
83
-
145
corresponding to molecular
weights of 24 to 40,60 to 70, and 120 to 134 kDa,
respectively. Finally, in apple buds, three isoen-
zymes were observed by electrophoresis with
molecular weights of 32, 39, and 77 kDa.
478
5. Polyphenol Oxidase Activity
Variations
in Apple
The level of polyphenol oxidase activity at
harvest and its variation during fruit storage are
obviously of great importance from a technologi-
cal point of view. There is general agreement that
enzyme activity is much higher in young fruits
than in ripe fruits (at commercial harvest) on a per
gram of fresh weight as well as on a fruit ba-
sis.
148,230.255,488
However, there were some discrep-
ancies concerning its evolution in the latter stages
of development and during fresh storage. Thus, if
the amount of freely soluble activity always in-
creased in this period
17

-
83
-
148
-
178
the total polyphe-
nol oxidase activity either decreased
17
-
83
-
148
-
181
-
230
or fluctuated,
89
'
181
-
445
-
448
depending on the cultivar.
Only in one case
516
was total enzyme activity
found to increase during this period. During rip-

ening of the Royal Delicious variety, activity in-
creased in the peel but decreased in the cortex.
324
121
Downloaded by [North Carolina State University] at 17:29 17 December 2012
The application of methyl jasmonate, an inhibitor
of ethylene production in postclimacteric apples,
strongly stimulated the polyphenol oxidase activ-
ity.
75
However, this treatment did not induce the
formation of new isoenzymes.
The level of polyphenol oxidase activity is
also tissue dependent. Some authors reported
that enzyme activity was higher in the peel than
in the cortex,
148
-
409
whereas the reverse was found
by others.
193
-
322
-
323
With the Grand Alexander
variety, the activity decreased from the core to
the outer part of the cortex and then increased in
the peel.

148
Finally, of 12 varieties analyzed at
commercial harvest,
181
the levels in the cortex
compared with that in the peel were higher for
seven varieties, equivalent for four varieties, and
lower for one variety. Table 5 shows that polyphe-
nol oxidase activity is also cultivar dependent,
because the relative levels ranged from 1 to 10 in
both the cortex and the peel.
181
-
193
Similar results
were found for different cultivars by other work-
ers
7.12,28.69,150,323,448^16 Compared with other cul-
tivars,
Red Delicious always had the highest
polyphenol oxidase activity.
69
-
181
-
193
6.
Kinetic
Properties
Enzyme specificity for phenolic compounds

has been the subject of several reports, whereas
only a few studies were devoted to the effect of
the second substrate (i.e., oxygen).
253
-
254
-
443
-
489
-
510
Polyphenol oxidases isolated from higher plants
and fungi are able to oxidize a wide range of
monophenols and o-diphenols. However, whereas
the Km values for oxygen only vary between 0.1
and 0.5 mM, the kinetic parameters, Vm and Km,
of the different phenolics are highly variable.
Moreover, most of the latter constants are only
apparent, as they were not determined at a saturat-
ing concentration in oxygen but most often in air-
saturated solutions of the phenolic substrates.
Several enzymatic extracts of plant
origin are devoid of monophenolase activ-
1^24,131,151,174,200,300,337,347,486,502,515
when
present,
cresolase activity was often partially or totally
lost on purification.
19

-
202
-
290
-
388
Moreover, in the
same species, multiple forms may also vary in
the monophenolase-to-o-diphenolase ratio of
activities.
247
-
301
-
511
a.
Phenolic
Substrate
Numerous studies have been carried out on
the specificity of apple polyphenoloxidase for
phenolic substrates. Large differences have been
found in the apparent Km values for phenolics in
air-saturated solutions (Table 6).
TABLE
5
Relative
Polyphenoloxidase
Activity
for
Different

Apple
Cultivars
at
Commercial
Maturity
Variety
8
Delicious
Jonathan
Golden
Delicious
San
Jacinto
Peasgood
Orleans
Reinette
Rome
Beauty
Ben
Davis
Alex.
Kyriati
G.
Alexander
Gallia-Beauty
Cortex
100
100
92
85

82
56
52
50
42
37
30
24
Variety"
Red
Delicious
Mclntosh
Fuji
Gala
Fiorina
Golden
Delicious
Canada
G.
Smith
Mutsu
Jonagold
Charden
Elstar
Peel
100
46
57
30
42

33
48
43
71
43
31
10
Cortex
100
80
71
48
40
30
88
73
54
43
39
20
Variety
6
Richared
Delicious
Braeburn
Sturmer
Pippin
Splendor
Dougherty
Golden

Delicious
G.
Smith
Kempton
Peel
100
62
47
71
14
26
65
34
Cortex
100
38
31
30
30
28
10
9
Adapted
from
Harel
et
al.;
148
manometric
measurement at pH 5.1.

Adapted
from
Janovitz-Klapp
et
al.;
181
polarographic
measurement at pH 4.5.
Adapted
from
Klein;
193
spectrophotometric
measurement at pH 6.2.
122
Downloaded by [North Carolina State University] at 17:29 17 December 2012
TABLE
6
Km
Values
of
Apple
Polyphenoloxidase
Toward
Different
Phenolic
Compounds
in
Air-Saturated
Solutions (All

Values
in
Millimoles)
Catechol
4-Methylcatechol
Protocatechuic
acid
o-Dihydroxyphenylacetic acid
Dihydrocaffeic acid
Caffeic acid
Chlorogenic
acid
Rosmarinic
acid
o-Dihydroxyphenylalanine
(-)-Epicatechin
(+)-Catechin
Procyanidin B2
Procyanidin C1
Quercetin
Butein
3-Hydroxyphloridzin
p-Cresol
Phloretin
Phloridzin
Assay
method
Quinone
formation
(spectrophotometry)

oDiphenols
140,
59
22,
509
18,
231
S.3
4443
2.1 ,
444a
4.7
331
1.6
331
19
331
2331
0.15
331
3.9,
331
1.6"
4
*
26,331 g444b
5.9
331
58
331

1
2
444a
0.3
331
0.3
331
Monophenols
0.5
33
'
0.5
33
'
Note:
The pH conditions of the
assay
were the
following:
Ref. 59 133 509 132
pH 5.5 6 5 4.5
148
467 180 331
5.1
5 4.5 4.5
Oxygen
uptake
(polarography-manometry)
20.3
331

3.6,
148
6,
148
3.2,
132
5.2,
180
4.9
331
1.5
331
2Q331
1.9
331
0^,
13a
0.6,
467
0.14,'
80
0.15
331
5.9,
132
1J,*
67
4.2,
180
4

331
41331
2.7,
132
25.5
331
2.8,
133
5.7132,331
4.7,
132
1.5,*
61
6.2,
180
5.9
331
2.8133
2.8133
0.27
331
0.32
331
0.2132
5.6,
148
26
148
0.47331
1.5,

132
0.49331
444
3.6a
5.4b
Some discrepancies among
the
results
ob-
tained
by
different authors
are
apparent. These
variations could have different origins. First, they
could originate from differences
in the
assay
methods, although Richard-Forget
331
obtained
similar apparent
Km
values
for 14
phenols when
she compared
the
results given
by the

spectropho-
tometric
and
polarographic methods. Second,
various apple cultivars have been used
as the
enzyme source,
and
small differences have been
found
in the
apparent values
for the
poly-
phenoloxidases extracted from Jonathan
and
Starking cultivars.
444
Third,
in the
same cultivar,
differences were observed between
the
crude
ex-
tract from chloroplasts
and
that from mitochon-
dria.
148

Moreover,
the
same authors
149
showed that
the apparent
Km
values were affected
by
purifica-
tion, since
for
4-methylcatechol,
the Km of the
crude extract from chloroplasts
was 6
mM, whereas
it
was 15.4 and 4.9 mM for two
fractions isolated
by ionic exchange chromatography
on
DEAE
cellulose. However, other authors
178
-
474
using simi-
lar methods
did not

find
any
variation
in the
speci-
ficity
of
apple polyphenoloxidase during
its
purification. Finally,
pH
undoubtedly affected
the
apparent
Km
values.
148
-
444
It was
shown
181
that
the
apparent
Km
values
for
4-methylcatechol, chlo-
rogenic acid,

and
(+)-catechin remained almost
constant between
pH 3.5 and 5, but
increased
as
the
pH
increased above
pH 5.
When oxygen uptake
123
Downloaded by [North Carolina State University] at 17:29 17 December 2012
was compared for different phenolic substrates,
4-methylcatechol was always found to be the most
rapidly oxidized at saturating concentra-
tions 13
2
.
14
8.149,180,181,331,409,474
However,
in
terms
of
efficiency, Vm/Km (this
ratio
is
proportional
to the

enzyme activity
at
substrate concentrations
far
below
Km; the con-
cept
is
reexamined
in
Section II.B.7), many
phe-
nolic compounds appeared
to be
better substrates
than 4-methylcatechol (Table
7).
This
is the
case
for compounds with
the
structure
of, or
close
to
that
of,
caffeoyl derivatives such
as

butein
and
caffeic, hydrocaffeic, rosmarinic,
and
chlorogenic
acids.
The
apparent
Km of the
main phenolics
of
the apple fruit cortex, that
is,
chlorogenic acid,
the
two catechin isomers,
and the
procyanidins
B2
and
Cl,
were
in the 2 to 6 mM
range (Table
6).
These values indicated that
the
apple enzyme
TABLE
7

Kinetic
Parameters of
Purified
Apple
Polyphenoloxidase
Toward
Different
Phenolic
Compounds in
Air-Saturated
Solutions at
30°C
and pH 4.5
4-Methylcatechol
Catechol
Chlorogenic
acid
Caffeic
acid
Rosmarinic
acid
Butein
Dihydrocaffeic
acid
o-Dihydroxyphenylacetic
acid
o-Dihydroxyphenylalanine
(-)-Epicatechin
(+)-Catechin
Protocatechuic

acid
Quercetin
Phloretin
Phloridzin
Km
4.9
20.3
4
0.15
4.1
0.32
1.9
20
25.5
5.7
5.9
1.5
0.27
0.47
0.49
Vm
100
80
99
5.8
125
20
67
94
18

52
60
0.8
4.1
6.1
0.7
Vm/Km
20.4
3.9
24.8
38.7
30.5
62.5
35.3
4.7
0.71
9.1
10.2
0.53
15.2
13
1.4
Note:
Km values are in
millimoles
and Vm in percent
of
4-methylcatechol.
Adapted
from

Richard-Forget,
F. M., Recherches sur le
brunissement
enzymatique.
Etudes sur
I'oxydation
de
phenols et sur
I'inhibition
de la
polyphenoloxydase
isolee
de
la Pomme
(Malus
sylvestris,
var. Red
Delicious),
Ph.D. thesis,
University
of Paris, 1992, 7.
With
permis-
sion.
affinity
for
these natural substrates
was not
very
high,

as was
usually found
for
other poly-
phenoloxidases from other origins.
254
-
510
More-
over,
if
chlorogenic acid
and the two
catechins
were rapidly oxidized
by
apple polyphenoloxidase,
the procyanidins
B2 and Cl
were very slowly
degraded. Thus, their maximum rates
of
oxygen
uptake represented only
20 and 7%,
respectively,
of that obtained with
the
monomeric flavan-3-ol
(i.e.,

(-)-epicatechin).
133
The
monophenolase
activity
was
either absent
59
-
467
or,
when present,
much lower compared with
the
o-diphenolase
activity obtained with
the
best phenolic substrate.
The monophenolic substrates were p-cresol,
148
p-chlorophenol,
148
p-coumaric acid,
132
phloretin,
431331
and phloridzin.
43
'
331

When tested, tyrosine
was
never found
to be a
substrate
of the
apple
en-
zyme.
43
-
59
-
132
-
148
-
289
-
467
Concerning
the
apple
fla-
vonols,
the
glycosylated derivatives
of
quercetin
were

not
substrates,
331
-
387
'
467
although
the
enzyme
was able
to
slowly oxidize
the
aglycone.
331
-
409
However, this oxygen uptake
did not
result
in the
formation
of
colored compounds.
331
-
387
Finally,
neither cyanidin

nor its
galactosyl derivative
(ideain)
was a
substrate
of
apple polyphenol-
oxidase.
331
Similar results were obtained
for the
oxidizability
of the
procyanidins
B2 and Cl, the
phloretin
and
quercetin derivatives,
and
antho-
cyanins
by
other polyphenoloxidases
of
different
origins.
18
-
295
-

298
-
304
-
314
-
327
b.
Oxygen
Only two reports dealt with the effect of oxy-
gen on apple polyphenol oxidase activity. Using
4-methylcatechol (7.5 mM) as substrate, Harel et
al.
148
indicated an apparent Km of 25.8% O
2
for a
crude enzymatic extract from apple chloroplasts.
In a more complete study using three phenols
(chlorogenic acid, 4-methylcatechol, and (+)-cat-
echin) at different concentrations, Janovitz-Klapp
et al.
180
found an equilibrium constant of 0.29 mM
(independent of the phenolic substrate) between
oxygen and a purified preparation of the apple
polyphenol oxidase, at pH 4.5 and 30°C. This
value, close to the oxygen concentration of air-
saturated aqueous solutions at this temperature,
499

confirms that during the routine activity assay,
polyphenol oxidase is only approximately
half-
124
Downloaded by [North Carolina State University] at 17:29 17 December 2012
saturated by oxygen (as already stated in Section
I.B.I). In the same study,
180
the Lineweaver-Burk
representations of the concentration effect of oxy-
gen and phenolic compounds gave a series of
intersecting lines. This was characteristic of an
ordered mechanism where one of the substrates is
required to bind to the enzyme before the second
substrate can bind.
384
The inhibition mode of
benzoic acid, which was competitive with
4-methylcatechol and uncompetitive with oxy-
gen, permitted the conclusion that oxygen is the
first substrate to be bound by the apple enzyme.
180
This is in agreement with results obtained with
polyphenol oxidases extracted from other
sources
173
-
248337
and with the reaction mechanism
proposed for fungal tyrosinases,

216
-
498
if the deoxy
form of the enzyme is arbitrarily chosen as the
first step of the catalysis cycle.
510
In contrast,
some reports indicated a ping-pong mechanism, a
random mechanism, or an ordered one where
oxygen does not bind first.
90
-
137
'
140
-
220
There are
several explanations for such discrepancies. First,
in numerous enzymatic extracts such as those
from fungi,
90
-
140
grape,
220
and potato,
248
the

monophenolase activity was not negligible. In
this case, the mechanism is further complicated,
because o-diphenol has to be considered both a
product of and a substrate for the enzyme. Sec-
ond, some authors
90
have used a spectrophoto-
metric method to obtain kinetic data, and the part
played by secondary products in the absorption,
which is very difficult to determine, was not taken
into account. Third, progress curves of changes in
oxygen concentration obtained in polarographic
assays have also been used to determine the in-
stantaneous velocity at different oxygen concen-
trations.
173
-
220
-
248
This can lead to erroneous results,
because the initial velocity was not measured for
each oxygen concentration. It is well known that
secondary products can play an important role in
oxygen consumption by a nonenzymatic process
333
and that polyphenol oxidase undergoes inactiva-
tion during its catalysis.
129130
c. Kinetics

with
Two Phenolics—Coupled
Oxidations
Most of the kinetic studies on polyphenol
oxidase were carried out with a reaction medium
containing one phenol. However, in natural prod-
ucts such as fruits and vegetables, several pheno-
lics are present and can be used as a substrate by
the enzyme. Few studies have been devoted to the
enzymatic oxidation of phenolic mixtures in model
solutions, and they were mainly concerned with
grape polyphenol oxidase.
51
-
53
"
55
-
58
-
296
-
334
With apple polyphenol oxidase, Janovitz-
Klapp et al.
180
conducted kinetic studies on model
systems containing two phenolic compounds. They
developed an equation
v. =

(Vm, *S,/Km
1
) + (Vm
2
*S
2
/Km
2
)
(1)
that allows one to predict the initial oxygen up-
take v
t
from the relative concentrations (S, and
S
2
) and kinetic parameters (Km^ Vrnj and Km
2
»
Vm
2
) of each individual phenolic substrate present.
Equation 1 can be extended to a more complex
solution containing more than two phenolics with-
out any special difficulty, provided the concentra-
tions and kinetic parameters of the additional
compounds are known. However, when the enzy-
matic reaction was followed by spectrophotom-
etry, the kinetics obtained did not agree with this
equation. There are different explanations for this

fact. (1) The absorption coefficients of the differ-
ent quinones are only partially known
351
-
463
and,
a fortiori, their respective contribution to the final
absorption. (2) The quinones are highly unstable
and very rapidly undergo nonenzymatic reac-
tions.
251
-
333
(3) Some enzymically produced quino-
nes are likely to react nonenzymically with
phenols.
53
-
269
-
395
These coupled oxidations can be
very rapid and are dependent on the respective
redox potential of the different quinone/phenol
couples present
53
-
281
The existence of such coupled
oxidations was demonstrated in caftaric (caffeoyl

tartaric) acid/flavan-3-ol mixtures with grape
polyphenol oxidase,
51
-
54
and chlorogenic acid/cat-
echin mixtures with mushroom
296
and apple
331
polyphenol oxidases. In all cases, compared with
the oxidation rate of the phenol alone, the flavan-
3-ol degradation was faster, while that of the
caffeoyl derivative was slower. The mechanism
proposed
53
involved enzymatic oxidation of the
two phenols, followed by chemical oxidation of
125
Downloaded by [North Carolina State University] at 17:29 17 December 2012
the flavan-3-ol by the caffeoyl quinone with re-
generation of the caffeoyl derivative. Moreover,
the o-quinones formed by enzymatic or coupled
oxidation can also react with another phenolic
molecule to yield condensation products
53
-
395
in
the form of oligomers or copolymers. A number

of these secondary products (which can also be
obtained by autooxidation) recently have been
characterized mainly by HPLC using a diode array
detector.
52
-
61
-
62
-
296
-
331
Besides flavan-3-ols, o-quin-
ones of caffeoyl derivatives were able to cooxidize
several other phenols such as thiol adducts,
56
-
332
anthocyanins,
314
-
327
'
331
4-methylcatechol,
333
fla-
vonols,
331

and dihydrochalcones.
298
-
331
According
to Cheynier et al.,
53
the redox potential order of
the grape phenolics appeared to be the following:
caftaric acid > epicatechin, catechin > glutathionyl
adduct of caftaric acid > procyanidins >
epicatechin gallate. After oxidation studies on
several pairs of phenolics, Richard-Forget
331
proposed a similar order: chlorogenic acid >
4-methylcatechol > epicatechin, catechin > cysteinyl
adducts of the four preceding phenols, anthocya-
nins,
flavonols, and dihydrochalcones derivatives.
Thus,
in the two cases, o-quinones of the caffeoyl
derivatives (caftaric and chlorogenic acids) were
the most efficient carrier for coupled oxidation
and seemed to play a determinant role in the
degradation of the other phenolics, including those
that are not a substrate of polyphenol oxidase.
d. pH and Temperature
Effects
Although a pH optimum around 7 was found
for the polyphenol oxidase activity extracted from

an apple mitochondrial fraction,
148
'
149
most stud-
ies indicated that the bulk of apple polyphenol
oxidase had a pH optimum of activity between
4.5 and 5.5.59,132,181.289.434,467.509 Moreover, the en-
zyme seemed relatively tolerant to acid pH, be-
cause the activity at pH 3 still represented 40% of
the maximum activity.
178
Thus, control of enzy-
matic browning by acidification only is difficult
unless a very low pH is obtained.
The effect of temperature on both enzyme
activity and stability has been investigated by
numerous workers.
13
'
246
-
251
'
272
-
317
'
443
Most assays of

the activity of apple polyphenol oxidase are car-
ried out between 25 and 35°C, and it is only for
temperatures higher than 40°C that a significant
thermal inactivation occurs during the assay.
Compared with other enzymes responsible for
food-quality degradation, polyphenol oxidase is
not a very heat-stable enzyme.
1
'
13
-
443
Therefore,
this activity is rarely used as an index of blanch-
ing treatment.
495
A partially purified extract of
apple polyphenol oxidase had a half-life of 12
min at 65°C and was destroyed at 80°C.
467
In
apple juices treated at the same temperature, 30%
of the activity remained after 20 s with the
Gravenstein variety
88
and 1 min with the Reine
des Reinettes variety.
82
Both reports stressed the
importance of the pH of the juice for thermosta-

bility, because for the same treatment, the re-
sidual activity at pH 3 was less than 10% of that
obtained at pH 6 (maximum stability). Likewise,
acidification of apple juice to pH 2 by HC1 for 45
min followed by a return to the original pH with
NaOH resulted in an 88% decrease of polyphenol
oxidase activity.
514
7.
Inhibitors
of Polyphenol Oxidase
This section addresses only inhibitors acting
on the enzyme. The other approaches used to
control enzymatic browning, that is, physical
methods and chemical agents acting on either
substrates or products, are examined in Section
IV.A and B. Different modes of action have been
proposed for the inhibitors acting directly on
polyphenol oxidase. The classification of Mayer
and Harel
254
contained two groups. Compounds
interacting with copper are in the first group and
those affecting the active site for the phenolic
substrate are in the second.
In the first group, inhibition by metal ion
chelators such as azide,
153
cyanide,
90

carbon mon-
oxide,
2
-
252
tropolone,
189
-
438
and sodium diethyl-
dithiocarbamate,
121
-
511
which are more or less
specific for copper, is well documented for
polyphenol oxidases of different sources.
Similarly, inhibition by halide ions has long
been recognized
196
for polyphenol oxidases
from apple
46
'
84
'
179
'
348
-

349
'
389
-
400
'
419
and other ori-
gins.
25
'
240
'
303
-
339
In the latter case, the same authors
showed that inhibition by halides was strongly
pH dependent, increasing when the pH was de-
creased. It was proposed that halide and copper
126
Downloaded by [North Carolina State University] at 17:29 17 December 2012
formed a complex after displacement of a proto-
nated histidine from copper.
25
-
303
A mechanism
was also postulated whereby halide binds to the
enzyme in competition with the phenolic sub-

strate only when the enzyme molecule is in the
protonated form.
303
Moreover, in a comparative
study of polyphenol oxidases from different
sources, it was suggested that the degree of en-
zyme inhibition by halide ions was the result of
the balance between the stability of the copper-
halide complex responsible for inhibition and the
accessibility of the copper in the active site to the
halide ion.
339
For the apple enzyme, the chloride
ion was shown to be noncompetitive,
179
-
389
while
the other halide ions were competitive.
179
The
order was F~ » Cl" > Br\I-,
175>389
because at pH
4.5 the apparent Ki values were 0.07,20,106, and
HOmM, respectively.
179
When the pH was var-
ied, it was postulated that the undissociated form
HF was responsible for inhibition by fluoride (in

agreement with results obtained for broad bean
polyphenol oxidase
339
) with a Ki close to 4 ^M.
179
In addition, the apparent Ki for chloride was shown
to vary according to the equation
Ki,app = Ki(Ka/[H
+
]
(2)
with pKa = 3.65 and Ki = 2.4 mil/.
179
Thus, for a
pH below pKa, the limiting value of Ki,app is
2.4 mM, whereas for a pH greater than pKa, an
increase of one pH unit corresponds roughly to a
tenfold increase of the apparent Ki for chloride.
179
Among inhibitors belonging to the second
group,
aromatic carboxylic acids of the benzoic
and cinnamic series have been studied widely
since the first works of Kuttner and Wagreich
199
and Krueger.
196
Although most authors found that
these compounds were competitive inhibitors of
polyphenol oxidase because of their structural

similarities with phenolic substrates, some au-
thors indicated that the type of inhibition was
dependent on the substrate used for the assay and
was either competitive, noncompetitive, or
mixed.
90
-
139
-
265
-
268
-
293
-
337
-
402
For apple, following the
substrate and the enzyme preparation used, Walker
and Wilson
477
found different inhibition patterns
(mainly competitive) for substituted cinnamic
acids.
With 4-methylcatechol as a substrate,
Janovitz-Klapp et al.
179
indicated that the 11 aro-
matic carboxylic acids tested in the benzoic, cin-

namic, phenylacetic, and phenylpropionic series
were all pure competitive inhibitors (Table 8). It
appeared that cinnamic acids were more potent
inhibitors than their benzoic homologs, as the Ki
values were 2 to 30 times lower than in the ben-
zoic series. Moreover, in both series, the p-hy-
droxy substitution slightly enhanced the inhibitory
properties, whereas adding one or two methoxy
groups in the meta position greatly decreased
inhibitor affinity for the enzyme. In addition,
when the carboxyl group was separated from the
benzene cycle by a methylene group as in
phenylacetic acid, the inhibition was greatly
reduced. However, it was partially restored by an
additional methylene group, as in phenylpropionic
acid, and again enhanced by a p-hydroxy substi-
tution (p-hydroxyphenylpropionic acid). There-
fore,
it is apparent that the substitution pattern for
the aromatic carboxylic acids leads to a similar
effect for the affinity of the inhibitors and the
corresponding substrates, because the Ki values
of cinnamic, benzoic, phenylpropionic, and
phenylacetic acids (Table 8) are in the same order
(although lower) as the Km values of caffeic,
protocatechuic (o-dihydroxybenzoic), hydro-
caffeic (o-dihydroxyphenylpropionic), and o-dihy-
droxyphenylacetic acids (Table 7). Finally, sorbic
acid is almost as efficient a competitive inhibitor
as benzoic acid.

179
Thus, the presence of the ben-
zene nucleus is not an absolute requirement for
the inhibitory effect, because it can be replaced by
conjugated double bonds. The latter result was
also found for apricot polyphenol oxidase.
402
The degree of inhibition by carboxylic acids
is also pH dependent, increasing as the pH de-
creases. Following experiments with broad bean
polyphenol oxidase,
339
it was proposed that the
undissociated form of the acid is responsible for
inhibition by forming a complex with copper at
the active center. Similar results were obtained
with a purified apple polyphenol oxidase.
179
One
can speculate that the binding site of the latter
enzyme recognized conjugated double bonds that
are contained in the benzene cycle or in an unsat-
urated alkyl chain. When a carboxyl group was
present, either directly bound to the benzene cycle
(benzoic series) or to the conjugated double bonds
(cinnamic series or sorbic acid), it could form a
complex with the copper at the active site. When
127
Downloaded by [North Carolina State University] at 17:29 17 December 2012
TABLE

8
Inhibition
Effects of Carboxylic Acids on
Purified
Apple
Polyphenol Oxidase at pH 4.5 in Relation to
Their
Structures
Acid
Benzoic
p-Hydroxybenzoic
Vanillic
Syringic
Cinnamic
p-Coumaric
Ferulic
Sinapic
Phenylacetic,
Substitution
P
Benzoic
OH
OH
OH
Cinnamic
OH
OH
OH
m m'
Series

OCH
3
OCH
3
OCH3
Series
OCH3
OCH3 OCH3
Phenylpropionic,
and
Sorbic
Phenylacetic
Phenylpropionic
p-Hydroxyphenylpropionic
OH
Sorbic
App Ki (mAf)
0.64
0.57
10
34.5
0.092
0.04
0.29
15
Acids
13
1.4
1.1
0.51

Adapted
from
Janovitz-Klapp,
A. H.,
Richard,
F. C,
Goupy,
P. M., and
Nicolas,
J. J., J.
Agric.
Food
Chem., 38, 926, 1990.
With
permission.
such a structure is present in the same molecule
together with an o-diphenolic structure, the inter-
action with the o-diphenolic would be greatly
reduced, as shown by the low Vra values obtained
for caffeic and protocatechuic acids (Table 7). In
addition, when the caffeic and chlorogenic acids
are compared, the esterification of the carboxyl
group by quinic acid reduced the affinity of the
caffeoyl moiety for the apple enzyme (increase of
Km).
However, it prevented formation of the com-
plex between copper and the vicinal undissoci-
ated carboxyl group, leaving the metal free for the
catalysis of o-diphenol oxidation, as illustrated by
the large increase of Vm (Table 7).

Use of cinnamic acids to control enzymatic
browning in fruit juices has been suggested.
471
Cinnamic acid added to Granny Smith juice at
concentrations greater than 0.5 mM resulted in an
inhibition of browning for over 7 h.
471
However,
apple plugs dipped in 10 mAf sodium cinnamate
were protected for several hours but then exhib-
ited a severe browning over extended storage
times.
369
It has been proposed that cinnamic acid
may undergo gradual conversion at the cut sur-
face to a polyphenol oxidase substrate by
cinnamate hydroxylase and other enzymes in-
volved in the biosynthesis of phenols.
369
Similar
results obtained with sodium benzoate led the
authors to recommend not using these aromatic
carboxylic acids as components of antibrowning
formulations.
369
The substituted resorcinols, which are also
structurally related to phenolic substrates, were
recently recognized as polyphenol oxidase inhibi-
tors.
262

-
263
In a structure-activity study on a series
of 4-substituted resorcinols, the highest inhibition
was obtained for hydrophobic substituents in the
4-position such as 4-hexyl, 4-dodecyl, and
4-cyclohexyl resorcinols.
263
The same authors in-
dicated that 4-hexyl resorcinol can be used for the
browning control of fresh and hot-air dried apple
slices as well as of apple juice.
263
In addition, they
128
Downloaded by [North Carolina State University] at 17:29 17 December 2012
claimed that 4-hexyl resorcinol has several ad-
vantages over sulfites for the prevention of shrimp
melanosis and for use in foods in general.
Kojic acid [5-hydroxy-2-(hydroxymethyl)-4-
pyrone] has been shown to inhibit fungal tyrosi-
nase.
375
Recently, it was found that it had an
inhibitory effect on plant and crustacean poiyphe-
noi oxidases.
49
It is a competitive inhibitor for the
oxidation of chlorogenic acid and catechol by
apple poiyphenoi oxidase.

49
In addition, this com-
pound can reduce pigments or pigment precursors
to colorless compounds.
48
Oxalate has been shown to inhibit poiyphenoi
oxidase in two ways: (1) upon incubation, it slowly
inactivated the enzyme following a first-order
kinetic in the enzyme and (2) it acted as a com-
petitive inhibitor of o-diphenol oxidation by fun-
gal
507
and apple
331
poiyphenoi oxidases. The
addition of copper can restore the activity lost in
the first process, as was shown for the fungal
507
and spinach
377
enzymes.
Finally, a mixed type of inhibition by several
natural aliphatic alcohols was reported for grape
poiyphenoi oxidase.
440
Inhibition seemed mainly
related to the hydrophobic chain rather than to the
alcohol function.
440
To our knowledge, no study

was carried out on these compounds with the
apple enzyme.
8. Secondary Reactions
Affecting
o-Quinones
The primary products of enzymatic oxidation
are o-quinones. These molecules have different
spectral properties, and their color mainly de-
pends on the pH and phenol from which they
originate.
421
Thus, after oxidation, catechin is
bright yellow with a maximum absorbance at 380
nm, chlorogenic acid is a dull orange-yellow with
a maximum at 420 nm, and DOPA (o-dihy-
droxyphenylalanine) is pink with a maximum close
to 480 nm. The molar extinction coefficients at
the maximum wavelengths are given in Table 9
and show a wide range of variation.
Moreover, the o-quinones are reactive com-
pounds,
110
-
269
-
350
as illustrated in Figures 4 and 5.
o-Quinones can react with another molecule of
phenol, resulting in dimers of the original phe-
no

126i,395,508 (reaction 1, Figure 4). These dimers,
having an o-diphenolic structure, can be subject
to reoxidation
52
-
395
either enzymically or by an-
other o-quinone, resulting in the formation of
larger oligomers with different color intensities.
The o-quinones can also react with a different
phenol molecule, either leading to a copolymer or
regenerating the original phenol and giving a dif-
ferent o-quinone by coupled oxidation (reactions
2 and 3, Figure
4).
51
-
54
-
296
-
331
Reactions can also occur with nonphenolic
compounds (Figure 5). Thus, another coupled
oxidation was observed with ascorbic acid (reac-
tion 1), giving the regeneration of the phenol with
formation of dehydroascorbic acid.
420
With sulfites
(reaction 2), colorless addition compounds are

formed
99
-
378
-
395
-
479
together with regenerated phe-
nol,
317
although the latter possibility remained
questionable. The o-quinones also form addition
compounds with thiol groups by nucleophilic
TABLE
9
Molar
Extinction
Coefficient
of Quinones
from
Different
o-Diphenolic Substrates of Poiyphenoi Oxidase
(Wavelength
at
Maximum
of Absorbance)
o-Diphenolic
Substrate
Pyrocatecho!

4-f-Butylcatechol
L-DOPA
3,4-Dihydrophenylacetic
acid
4-Methylcatechol
Hydrocaffeic
Chlorogenic
acid
(+)-Catechin
Wavelength
(nm)
390
400
480
390
400
412
420
380
Extinction
coefficient
1417
1150
3388
1311
1400
1124
2000
1200
Ref

463
463
463
463
463
463
351
351
129
Downloaded by [North Carolina State University] at 17:29 17 December 2012
OH
0
POLYPHENOL
H
2
O
2
OXIDASE
Brown
Potymsre
R
o-qulnone
OH
FIGURE
4.
Reactions
of
o-quinones
with
phenolic

compounds.
(All
reactions
are
nonenzymatic
except
those
with
polyphenol
oxidases;
reactions
2
and
3
are
able
to
regenerate
the
original
phenol.)
Products
with
different
color
intensities
are
indicated
by
asterisks.

(From
Rouet-Mayer,
M. A.,
Philippon,
J., and
Nicolas,
J.,
Encyclopaedia
in
Food
Science,
Food
Technology
and
Nutrition,
McRae,
R.,
Robinson,
R.
K.,
and
Sadler,
M.
J., Eds.,
Vol.
1,
Academic
Press,
London,
1993, 499.

With
permission.)
additions (reactions
3
and 4).
Cysteine, either
free
91.243,275,308,341,359
Qr
bound
in
smaU
pep
.
tides
57
-
156
-
309
or
large proteins,
310
gives colorless
compounds. Although these compounds have
an
o-diphenolic structure, they are
not
a
substrate

of
polyphenol oxidase
58
-
329
-
359
'
397
but can
either
be
oxidized
by
laccase
355
or
react with
an
excess
of
o-quinones
by
a
coupled oxidation mechanism
332
and form intensely colored products. Nucleophilic
additions also occur with amino groups
of
amino

acids or peptides.
243
-
309
Both primary (e.g., serine
39
)
and secondary (e.g., proline
437
) amines form addi-
tion compounds with o-quinones (reactions
5
and
6).
Moreover, further substitutions with thiol (re-
action
4) or
amine (reaction
7)
groups contained
in proteins may occur, leading
to
the formation of
intra-
or
intermolecular cross-links.
250
-
264
-

311
Fi-
nally, water
can be
added slowly
to
the
o-quin-
ones
79
*
333
to
form triphenols that
are
readily oxi-
dized
by
polyphenol oxidase
or by
excess
o-quinones, leading
to
the
p-quinones (reaction
8).
The
last pathway
was
favored

by
acid
pJJ
124,125,183,333
The reactivity (or
in
other terms, the stability)
of the o-quinones
in
these different pathways
is
highly variable. The nature
of
the phenol oxidized
as well
as
the oxidation conditions (medium, pH,
temperature, etc.)
are
determinant. Thus,
it
was
shown that
in
the
same conditions, o-quinones
from 4-methylcatechol were more stable than those
from chlorogenic acid, which
in
turn were more

stable than those from catechins.
333
Obviously,
the presence
of
reactive molecules, with amino
or
thiol groups,
in
the medium can greatly affect the
130
Downloaded by [North Carolina State University] at 17:29 17 December 2012
FIGURE 5.
Reactions
of
o-quinones
with
nonphenolic
compounds.
(All
reactions are
nonenzymatic
except those
with
laccase; reactions
1 to 3, 6, and 8 are
able
to
regenerate
the

original
phenol.)
Products
with
different
color
intensities
are
indicated
by
asterisks.
Ox,
further
oxidation
reactions
by
oxygen
or
o-quinone;
Pr-SH
and
Pr-NH
2
,
proteins;
Pro-NH,
Proline;
AA-NH
2
,

amino
acids; Asc
A,
ascorbic
acid;
DHA,
dehydroascorbic
acid;
FTSH,
small
thiol
compounds (e.g., cysteine
or
glutathione).
(From
Rouet-Mayer,
M. A.,
Philippon,
J.,
and
Nicolas,
J.,
Encyclopaedia
of
Food
Science,
Food
Technology
and
Nutrition,

McRae,
R.,
Robinson,
R. K., and
Sadler,
M. J.,
Eds.,
Vol. 1,
Academic
Press,
London,
1993, 499.
With
permission.)
reactivity
of
o-quinones. The secondary products
formed
may or may not be
good substrates
for
polyphenol oxidase
and
may exhibit differences
in their reactivities with o-quinones.
C. Other Factors
1.
Ascorbic Acid
in
Apple

and
Apple
Products
Ascorbic acid
is a
powerful reducing agent
and
in its
presence
the
enzymically formed
o-quinones
are
converted back
to
their precursor
diphenols, preventing the formation of pigments.
472
Therefore, this compound could play
a
role
in
the
extent
of
tissue browning after bruising.
236
Com-
pared with many other fruits, apples
are not

very
rich in vitamin
C.
Recent surveys gave
a
mean
content
of 50
mg/kg (FW basis).
214496
However,
large differences were found among samples,
as
values varied between
30 and
370 mg/kg. Many
factors can be responsible
for
these variations,
for
example, variety
214
and
maturity stage before
488
and after harvest.
496
The peel contains more ascor-
bic acid than the pulp
as

well
as the
side exposed
to
the sun
compared with
the
shaded side.
100
Fi-
nally, because
of
oxidation losses during process-
ing,
the
ascorbic acid content
in
apple juice
is
almost zero,
37
-
208
although some retention
was
131
Downloaded by [North Carolina State University] at 17:29 17 December 2012
found
in
freeze-concentrate enriched juice.

417
In
apple puree,
a
rapid decrease
of
ascorbic acid
content
was
shown after blending, with
a con-
comitant increase
in
dehydroascorbic acid, result-
ing
in
almost no loss
in
total vitamin
C
content.
274
Nevertheless,
it
seemed that endogen ascorbic
acid levels
did not
play
a
major role

in
limiting
the extent
of
browning.
230
-
488
Recently,
a
HPLC
procedure was developed
for
the determination
of
ascorbic acid, dehydroascorbic acid, and ascorbic
acid-2-phosphate
in raw
apple samples supple-
mented with ascorbic acid
or
ascorbic acid-2-
phosphate
to
prevent browning.
365
2. Peroxidase
in
Apple
and

Apple
Products
Peroxidases
are
ubiquitous enzymes
in
fruits
and vegetables that
are
able
to
oxidize
a
large
number
of
molecules.
342
'
343
-
443
However, few stud-
ies have been devoted
to
the peroxidase system
of
apple. Peroxidase activity was proposed
as a
pos-

sible parameter
of
ripening
and
senescence
of
Golden Delicious apples,
134
although
the
isoen-
zyme patterns did not change during this period.
31
Activity was found
to
vary significantly with cul-
tivar, picking date,
and
storage period,
but
with-
out any regularity
for
the three cultivars tested.
447
Peroxidase appeared
to be
more concentrated
in
the peel than

in the
pulp.
285
-
330
For eight cultivars
picked
at
commercial maturity, the lowest peroxi-
dase activity (Mclntosh var.)
in
peel represented
22%
of the
highest (Red Delicious var.).
330
After partial purification,
at
least three isoen-
zymes were characterized from either peel
330
or
pulp.
286
The
major isoenzyme,
the
less heat
stable,
286

was cationic, with
a
pHi close
to
9.
286330
Although separated
by ion
exchange chroma-
tography,
all
isoenzymes eluted
in a
single peak
in
gel
filtration corresponding
to a
molecular
weight
of 40
kDa.
330
The
optimum
for
activity
was close
to pH 5.8, a
value that

is in the
range
given
for
peroxidase activity extracted from other
fruits.
272
-
443
Moreover, apple peroxidase was able
to oxidize
not
only chlorogenic acid
and
(+)-catechin,
but
also several quercetin glyco-
sides
by a
ping-pong, bireactant mechanism.
330
Thus,
the
glycosylated flavonols, which are
not a
substrate
for
polyphenol oxidase, can be degraded
by apple peroxidase
in the

presence
of
hydrogen
peroxide. Note that horseradish peroxidase
is
also
able
to
oxidize kaempferol,
273
quercetin,
383
and
other flavonoids.
77
-
382
However,
to
our knowledge,
no direct correlation
has
been found between
peroxidase activity and the browning susceptibil-
ity
of
apple varieties.
Other works were not directly concerned with
the fruit,
but,

rather,
the
embryo,
353
root-
stocks,
267
-
458
-
459
-
482
-
483
or
cell cultures.
26
All
these
studies used electrophoretic methods
in
order
to
characterize the peroxidase polymorphism
for
the
identification
of
apple cultivars.

III. RELATIONSHIP BETWEEN
INTENSITY
OF BROWNING AND
BROWNING PARAMETERS
A.
Methods
for
Evaluation
of
Browning
Accurate methods
are
required
for the mea-
surement
of
browning
in
tissue slices
and ex-
tracts.
This need is obvious when different cultivars
are compared
for
susceptibility
to
browning
or for
evaluation
of

experimental treatments designed
to control enzymatic browning.
12
Basically,
two
kinds
of
methods
are
avail-
able.
236
The first uses absorbance measurements,
usually
in the
400-nm region,
on
solutions after
extraction
and
purification
of the
brown
pig-
ments.
488
The
second uses reflectometry
or
tristimulus colorimetry

and can be
applied
di-
rectly
to
cut surfaces
or
fruit puree.
364
-
442
Although
both methods
are
easy
and
rapid, they
do
have
serious disadvantages.
Absorbance measurements evaluate only
the
soluble pigments.
It is
well known that,
as the
reaction proceeds, polymerization occurs
and the
solubility
of a

large part
of the
brown pigments
decreases.
395
The insoluble entities are eliminated
during
the
filtration
and
centrifugation steps
in
the purification process. Moreover, depending
on
the kind
of
pigments, which
in
turn depend
on
the
original phenols
and
their relative proportions,
the wavelength
of
maximum absorbance ranges
between 360
and
500 nm. Therefore, absorbance

measurements
at a
single wavelength correlate
poorly with visual evaluation
of
browning.
In order
to
obtain information
on the
relative
importance
of
browning parameters,
it has
been
132
Downloaded by [North Carolina State University] at 17:29 17 December 2012