International Journal of Food Science and Technology 2010, 45, 1647–1658 1647

Original article
Resistance of industrial mango peel waste to pectin degradation
prior to by-product drying

Suparat Sirisakulwat,1,2 Pittaya Sruamsiri,2 Reinhold Carle1 & Sybille Neidhart1*

1 Institute of Food Science and Biotechnology, Chair of Plant Foodstuff Technology, Hohenheim University, Garbenstrasse 25, 70599 Stuttgart,
Germany
2 Department of Crop Science and Natural Resources, Faculty of Agriculture, Chiang Mai University, Chiang Mai 50200, Thailand
(Received 3 December 2009; Accepted in revised form 19 May 2010)

Summary Susceptibility of industrial mango peel waste to pectin degradation during storage at ambient conditions
(25 C, 63% relative humidity) for up to 5 h before by-product stabilisation by drying was explored.
Depending on the interim storage period in the wet state, pectins were recovered from the dried peels by hot-
acid extraction. Most important, pectin degradation during the temporary storage of the wet peels was
insignificant, as revealed by yields, composition, average molecular properties, and techno-functional
quality. Hardly acetylated (DAc 2.5–4.5%), rapid-set high-methoxyl pectins were obtained at starch-
corrected net yields of 14.1–15.6 g hg)1. Irrelevant de-esterification during peel storage in the wet state was
confirmed by overall uniform setting temperatures. Arabinogalactans, uniformly indicated by high molar
galactose ⁄ rhamnose ratios of 13.8–16.9 mol ⁄ mol and an arabinose percentage of 9.5–14.4 mol hmol)1 of
galactose residues, affected the galacturonide contents, intrinsic viscosities, and gel strengths throughout.
The wet peels, derived from widespread manual peeling in mango canning, tolerated intermediate storage for
5 h, thus facilitating by-product stabilisation on smaller scales.
Keywords By-product drying, enzymatic degradation, fruit processing, intermediate storage, mango (Mangifera indica L.), pectin, waste
utilisation.

domestica Borkh.) juice manufacture and the peels

Introduction
and rags of citrus juice processing, chiefly lemon [Citrus
Owing to large quantities and pectin yields up to 21% limon (L.) Burm. F.] and lime [Citrus aurantiifolia
on a dry weight basis (Rehman et al., 2004), mango (Christm.) Swingle], because their pectin properties are
(Mangifera indica L.) by-products, chiefly peel waste suitable for food applications (Endress et al., 2006).
from juice production, drying, and canning, have been Because grinding of apples causes severe loss of cell
deemed a promising source not yet commercially compartmentalisation, apple pomace requires drying
exploited (Panouillé et al., 2007). Reported pectin yields within a few hours (Voragen et al., 1995), preferably
from mango peels greatly varied, despite insignificant straight after de-juicing of the mash (Endress et al.,
depolymerisation of pectins and galactans in the peel 2006), to inactivate pectin-degrading enzymes and to
during postharvest ripening (Sirisakulwat et al., 2008). prevent microbial decay. Due to techniques restricting
Moreover, the gelling capacities of the rapid-set to ultra- de-juicing to the citrus endocarp, tissue damage of citrus
rapid-set pectins from mango peels may be more or less peels is less serious and pectins may be produced from the
limited by a characteristic almost monodisperse fraction wet residues in the vicinity of a citrus processing plant
that may considerably reduce the average molecular (Endress et al., 2006), especially when the essential
weight depending on its mass percentage (Neidhart washing of citrus peels is combined with their immediate
et al., 2009). blanching after juice extraction (Voragen et al., 1995).
Processing by-products like peels are highly perishable However, shipping, storage, and year-round pectin pro-
in the wet state. At present, pectins are commercially duction independent of raw material availability require
recovered from the dried press cake of apple (Malus instantaneous by-product drying in each case.
Like numerous microbial enzymes, a so far limited
*Correspondent: Tel: +49(0)711 459 22317; Fax: +49(0)711 459 24110; number of plant enzymes are involved in pectin degra-
e-mail: Sybille.Neidhart@uni-hohenheim.de dation (Benen et al., 2003b). Besides depolymerisation

doi:10.1111/j.1365-2621.2010.02314.x
 2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology
1648 Stability of mango peel by-products S. Sirisakulwat et al.

through endo- (EC 3.2.1.15) and exo-acting (EC 3.2. As regards a secure and economic supply of high-
1.67) polygalacturonases (PG), pectin methyl esterases quality mango by-products for further upgrading, this
(PME, EC 3.1.1.11) affecting the degree of esterification study therefore aimed at evaluating the proneness of
are most crucial. Without immediate drying or at least industrial peel waste to pectin degradation during a
blanching of citrus peels, even slight de-esterification temporary dwell time at ambient conditions until
because of high endogenous PME activities may cause drying. The tolerable period between waste production
calcium-sensitive pectins unsuitable as gelling agents (peeling) and drying was to be identified, mimicking
(Voragen et al., 1995). Apart from some soluble forms, different operational situations in mango processing.
PME is mainly associated with cell wall proteins by The contribution of by-product handling to the varia-
ionic interactions (Benen et al., 2003a). Additionally, tions of mango pectins in yield and quality was to be
pectin homogalacturonan acetylesterases (PAE, EC 3.1. assessed.
1.6) occur in orange and a few other plant species
(Benen et al., 2003a). Aside from PG, pectate lyase
Materials and methods
(PAL, EC 4.2.2.2) and endo-rhamnogalacturonan
hydrolase (endoRGH) have been found in plants as
Sample material and experimental design
further pectin backbone depolymerases (Benen et al.,
2003b). Degradation of pectin side chains involves Industrial peel waste, originating from ‘Kaew Khiew’
arabinanases and galactanases, such as a carrot mango fruit, was obtained from a canning factory in
a-l-arabinofuranosidase (EC 3.2.1.55) and a tomato Chiang Mai, Thailand, during the mango processing
exo-b(1 fi 4)-d-galactanase (De Vries & Visser, 2003). campaign in April 2005. Nine samples were consecu-
During fruit ripening, plant b-d-galactosidases (EC tively drawn within 4 h. For each of them, 1800 g of
3.2.1.23), increasing in parallel with tissue softening, fresh peel waste was quickly collected from three manual
cause pectin depolymerisation by side-chain loss and peeling lines by randomly picking the material from
enhanced pectin solubilisation (Brummell, 2006). In various positions of the lines directly after its removal
ripening mango mesocarp, b-d-galactosidase activity from the fruit. To explore the tolerable period between
even rose to the eightfold level from the mature-green peeling and waste drying, the peels of each sample were
to the ripe stage, but PG and b(1 fi 4)-d-glucanase exposed to the ambient conditions of the peeling room
(EC 3.2.1.4) only to the 1.6- and 2.4-fold activities, (25 C; 63% relative humidity, RH) in an open bowl
while PME declined to 41% (Ali et al., 2004). for holding times (HT25 C) of 0 (variant MPW1),
Although the mango cultivars may differ in the 15 (MPW2), 30 (MPW3), 60 (MPW4), 90 (MPW5),
properties of three PG isoforms (Prasanna et al., 120 (MPW6), 180 (MPW7), 240 (MPW8), and 300 min
2006; Singh & Dwivedi, 2008), mango PG turned out (MPW9), respectively. Subsequently, the peels of each
to be mainly exo-acting (Ali et al., 2004). Homogalac- sample were packed as a thin layer into a polyethylene
turonan degradation in mango mesocarp has thus (PE) pouch and frozen in a freezer, which was tempo-
recently been ascribed to a softening-related PAL rarily installed near the peeling lines. So far, the
rather than to PG (Chourasia et al., 2006). Consistent experiment was performed close to the peeling lines
with much lower, rather constant PME and PG when they were in full operation. Because of limited
activities in the peel relative to the mesocarp (Yanru drying capacity, the samples were temporarily stored at
et al., 1995), carbohydrate changes in mango peels )20 C until drying. Without previous external thawing,
because of postharvest ripening were mainly restricted the frozen peel samples were dried at 80 C for approx-
to the degradation of starch and matrix glycan frag- imately 4 h in a fluidized-bed laboratory dryer (Rapid
ments (Sirisakulwat et al., 2008). dryer TG 1, Retsch, Haan, Germany) until a moisture
On the other hand, the mango fruit may predomi- content of 4.5–6.6 g hg)1. After weighing to record the
nantly be processed at higher ambient temperature and yield of the dried mango peel waste (MPW)
humidity than the citrus fruit and apples in view of the (21.6–29.8 g hg)1 of fresh peel waste), the dried peels
climate in the respective growing areas. Therefore, were vacuum-packed into PE pouches and transported
rather adverse conditions during a temporary storage by airfreight to Hohenheim University, Stuttgart,
of wet mango peels between peeling and by-product Germany, where they were manually crushed, vacuum-
drying may occur in practice. Their impact on pectin packed into PE pouches, and stored at ambient temper-
quality must be known with respect to minimum ature (approximately 22–25 C) in a desiccator until
requirements for sustainable by-product stabilisation pectin extraction. By using the methodology detailed
processes when evaluating this potential raw material earlier (Sirisakulwat et al., 2008), the recovered pectins
source, primarily because mango processing represents a were characterised in terms of composition as well as
very multifaceted field with a wide range of production average molecular and techno-functional properties.
scales (Vásquez-Caicedo et al., 2004, 2007; Pott et al., Analytical and empirical methods commonly used by
2005). pectin producers for quality assessment were included.

International Journal of Food Science and Technology 2010  2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology
 Stability of mango peel by-products S. Sirisakulwat et al. 1649

Ripeness of the fruit processed was assessed at Darmstadt, Germany). The ground AIS was subjected
Chiang Mai University, Chiang Mai, Thailand, by to pectin characterisation.
analogy to a previous study (Sirisakulwat et al., 2008)
on the basis of five fruits that were randomly selected
Chemical pectin characterisation
among the fruits provided for peeling. The maturity-
specific properties of the edible part included the pH The AIS composition was analysed as detailed previ-
value (pH 3.85 ± 0.01) and the contents of total ously (Sirisakulwat et al., 2008). In brief, the galacturo-
soluble solids (TSS; 18.7 ± 0.07 Brix) and titratable nide content was titrimetrically quantified in duplicate
acids (TA as citric acid, pH 8.1; 0.58 ± 0.01 g hg)1), from the neutralisation and saponification equiva-
corresponding to a sugar ⁄ acid ratio (TSS ⁄ TA) of lents and expressed as galacturonic acid (Mr =
32.2 ± 0.7. Mesocarp firmness, measured with an 194.1 g mol)1) on an ash- and moisture-free basis
Instron Universal Texture Analyzer type 3365 (Instron, [GalUAtitr, g hg)1 of purified AIS (AISp)] according to
Canton, MA, USA) as specific maximum load (SFKS) the legal pectin specification (Joint FAO ⁄ WHO Expert
by using a Kramer shear cell (Vásquez-Caicedo et al., Committee on Food Additives (JECFA), 2007). Based
2006), ranged at 472 ± 9 N hg)1. Compared to the on the recovery rate after AIS purification in acidified
mango fruit of a related study (Sirisakulwat et al., propan-2-ol (AISp ⁄ AIS, g g)1), GalUAtitr was converted
2008), the mesocarp properties came overall most close into the anhydrogalacturonic acid content of the
to those of the ‘Kaew Khiew’ (KAk) batch, which had original sample (AUAtitr, g hg)1 of AIS; Mr =
previously been examined after major fruit softening 176.13 g mol)1). By analogy, its methylester content
(ripeness stage II). For this batch, a TSS ⁄ TA ratio of (Mr = 31 g mol)1) was calculated (MeOHtitr, g hg)1 of
55.2 ± 0.9 was reported. Its mesocarp firmness of AIS) from the saponification equivalents. The degree of
4.4 ± 0.1 N, recorded with a Warner–Bratzler shear esterification (DE) resulted as the percentage of the
cell, corresponded to a SFKS of 519 ± 19 N hg)1 saponification equivalents relative to the total of neu-
measured concomitantly. For comparison, full-ripe tralisation and saponification equivalents. After acid
fruit (stage III) had then been characterised by a hydrolysis of the AIS using H2SO4 (72% w ⁄ w) at
TSS ⁄ TA ratio of 94.1 ± 2.6 and low firmness records ambient temperature, the anhydrouronic acid content
of 1.8 ± 0.04 N and 373 ± 12 N hg)1, respectively (Mr = 176.13 g mol)1) was additionally quantified col-
(Sirisakulwat et al., 2008). Compared to these two orimetrically (AUAc, g hg)1 of AIS) in triplicate with
batches, the fruit of this study thus had a lower sugar m-hydroxydiphenyl in photometric series analyses. The
content and TSS ⁄ TA ratio, but mesocarp firmness was absorbance was measured at 520 nm against a sample
intermediate. blank with a Cary 100 spectrophotometer (Varian,
Mulgrave, VIC, Australia).
The neutral sugars were determined in duplicate
Pectin extraction from dried mango peel waste
as their alditol acetates by gas chromatography (GC)
Industrial hot-acid extraction was mimicked on a with flame ionisation detection after acid hydrolysis
laboratory scale. Pectin was extracted from the dried (Sirisakulwat et al., 2008). After prehydrolysis at ambi-
raw material (DRM) at pH 1.5, applying a standard ent temperature (60 min) using H2SO4 (72% w ⁄ w), the
protocol (Sirisakulwat et al., 2008) to the dried MPW AIS samples were diluted with N2-saturated, ultra-pure
on the double scale throughout. Per run, 40 g of MPW water and subjected to hot-acid hydrolysis (121 C, 1 h)
was filled up with distilled water to a total of 800 g. The in an electric heating block. The neutral sugars were
suspension was boiled for 15–20 min and recooled. The reduced with sodium borohydride, followed by acetyla-
pH was adjusted to 1.5 with H2SO4 (25% w ⁄ w) prior to tion of the alditols with acetic anhydride in the presence
extraction (90 C, 2.5 h) in an oil bath. Separation of the of 1-methyl imidazole as catalyst. For GC analyses, the
extract as well as precipitation in propan-2-ol (‡ 99.9% alditol acetates were extracted from the aqueous phase
v ⁄ v) and drying (60 C, 1–2 h) of the pectins were with ethyl acetate, produced after the addition of
performed as described previously (Sirisakulwat et al., absolute ethanol. Myo-inositol was used as internal
2008). Nine runs were carried out for each MPW standard.
sample. For each run, the yield of dried crude pectin was Based on the simultaneous HPLC analysis of meth-
calculated as alcohol-insoluble substance (AIS, g hg)1 of anol (MeOH, Mr = 32.04 g mol)1) and acetic acid
DRM). Per sample, the AIS of all runs was pooled and (Mr = 60.05 g mol)1) after saponification in alkaline
ground in a centrifugal mill type ZM 1 (Retsch, Haan, propan-2-ol, the degrees of methylation (DMe) and
Germany) to particle sizes <0.25 mm. The starch- acetylation (DAc) were calculated as the respective
corrected pectin yield (AISS-corr, g hg)1 of DRM) was molar ratios to AUAc (Sirisakulwat et al., 2008). The
calculated (Sirisakulwat et al., 2008) after enzymatic methanol and acetyl contents (g hg)1 of AIS) were
quantification of co-extracted starch (g hg)1 of AIS) by quantified by using the 2695 Separations Module of a
means of the respective test combination (R-Biopharm, Waters HPLC Alliance system (Waters, Milford, MA,

 2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology International Journal of Food Science and Technology 2010
1650 Stability of mango peel by-products S. Sirisakulwat et al.

USA) with a Waters 2410 refractive index detector set to buffer (0.25 mol kg)1 of gel), the setting temperatures
35 C. An Aminex HPX-87H column (300 · 7.8 mm; were assayed by dynamic rheometric analyses in dupli-
Bio-Rad Laboratories, Hercules, CA, USA) was eluted cate or triplicate, using a Bohlin CVO 120 HR rheom-
at 40 C with 0.01 n H2SO4 (0.6 mL min)1). Samples eter (Malvern Intruments, Herrenberg, Germany) with a
were prepared in duplicate with two injections per vial. Searle cylinder device (C 25) (Sirisakulwat et al., 2008).
Lithium lactate, dissolved in alkaline propan-2-ol, The AIS, premixed with a small portion of the sucrose,
served as internal standard. was carefully dispersed at ambient temperature in the
The moisture content (g hg)1 of AIS) was gravimet- diluted buffer that resulted from 68 g of distilled water
rically determined in duplicate as mass loss after drying and 17 g of a KOAc ⁄ HOLc buffer stock solution
of 1 g of AIS in a porcelain crucible at 105 ± 2 C in a [KOAc: 40.48 g L)1; HOLc (90%): 188.05 g L)1]. When
hot-air cabinet for 2 h (Joint FAO ⁄ WHO Expert the AIS was completely dissolved under heating in
Committee on Food Additives (JECFA), 2007) and the presence of 1–2 drops of defoaming emulsion, the
recooling in a desiccator until constant weight. After remaining part of the sugar was stepwise added to the
slow preincineration of the moisture-free AIS on an boiling mixture. TSS was adjusted by cooking to a
electric heating plate, the ash content (g hg)1 of AIS) precalculated net weight of 170 g under stirring. Subse-
was quantified by complete incineration (550 C, 3 and quently, the sample was poured into the preheated C 25
16 h) in a muffle furnace with intermediate wetting of device (90 C). To avoid evaporation and superficial
the recooled sample [2–4 drops of ultra-pure water and hardening, the free sample surface was covered with hot
4–5 drops of H2O2 (30% w ⁄ w)] and drying (105 ± 2 C, low-viscous liquid paraffin. Gel formation was moni-
1 h) (Sirisakulwat et al., 2008). The mass of the white tored by measuring the storage (G¢) and the loss (G¢¢)
ash was verified after further incineration (550 C, 1 h) moduli as well as the phase angle [d = arctan (G¢¢ ⁄ G¢)]
and recooling (60 min). at a fixed frequency (f = 1 Hz) and strain amplitude
(c = 0.015) as a function of time (t, s) during cooling
from 90 to 20 C at a constant rate (DT = )0.999 ±
Quantification of average macromolecular pectin properties
0.002 K min)1). In the sol–gel transition range, the d-t
The intrinsic viscosity ([g], mL g)1) was assessed as an curves were iteratively approximated by a modified
average parameter influenced by all polymers of an AIS logistic four-parameter model (Neidhart et al., 2003),
sample. Based on the single-point relationship (Solo- using the NLIN procedure (Marquardt method) of the
mon-Ciuta equation), it was computed from the specific Statistical Analysis System (SAS) 9.1 (SAS Institute,
viscosities [gsp ¼ ðgs  go Þ=go ] of different AIS solutions Cary, NC, USA). The setting time (td¼45 , s) and
as a mean of ten records and verified by the Huggins temperature (Td¼45 , C) were deduced from the cross-
plot and the Kraemer plot as detailed previously over-point of the moduli [d(1 Hz) = 45].
(Sirisakulwat et al., 2008). The AIS contents (c) of five The gel properties were characterised by the breaking
solutions per sample were adjusted to the gsp range of strengths (BS) of pectin-sucrose gels (TSS = 65 Brix,
0.2–1 (Morris, 1995) by appropriate dilution of a sample pH 3) with AIS doses of 0.3 and 0.35 g hg)1 in a
stock solution (c = 0.3–0.4 g dL)1) with the solvent. KOAc ⁄ HOLc buffer (0.28 mol kg)1 of gel) on the basis
The dynamic viscosities (g = Kqt, mPa s) of AIS of an empirical tension test established for the stan-
solutions (gs) and solvent (go) were calculated from the dardisation of pectins and the comparative evaluation of
flow time (t, s), measured at 20 C for a test volume of pectin sources (Endress & Dilger, 1990). By using the
2 mL (2 · 3 runs for to and ts, respectively) by using a Herbstreith-Pectinometer type Mark III (Herbstreith &
Micro-Ostwald capillary viscosimeter (capillary con- Fox, Neuenbürg, Germany), the BS was recorded in
stant K = 0.01181 mm2 s)2). The densities (q, g cm)3) device-specific units (Herbstreith-Pectinometer units,
of AIS solutions (qs) and solvent (qo) were measured at HPE). The AIS dose required for a standardised gel
20 C with a density meter DMA 48 (Anton Paar, Graz, (BS = 530 HPE) was calculated by interpolation as the
Austria). The AIS was dissolved under automatic breaking capacity of the AIS (BC530HPE, g hg)1 of gel).
shaking (16 h) before adjusting the volume of the stock By dividing the sugar content of this gel (TSS) by
solution to 100 mL. A 0.28 m buffer (pH 3.0 ± 0.05) of BC530HPE, the gelling capacity of the AIS (SBC530HPE,
potassium acetate (KOAc: 4.57 g L)1) and lactic acid g g)1 of AIS) was obtained as the sugar-binding
[HOLc (90%): 23.59 g L)1] served as solvent, mimicking capacity, i.e., the amount of sugar bound by this AIS
the gel systems used for the gelation studies in terms of dose in a gel of 530 HPE. Multiplication of SBC530HPE
pH and concentration. with the AIS content of the DRM (dried MPW) yielded
the corresponding gelling capacity of the DRM
extracted (SBCDRM, g hg)1 of DRM). With each AIS
Characterisation of techno-functional pectin properties
dose, a gel was prepared by cooking the AIS-containing
For pectin-sucrose gels (TSS = 65 Brix, pH 3) with mixture on the basis of 216.0 g of sugar and 147.0 g of a
AIS doses (cp) of 0.4 and 0.55 g hg)1 in a KOAc ⁄ HOLc 0.65 m KOAc ⁄ HOLc buffer [KOAc: 10.5 g L)1; HOLc

International Journal of Food Science and Technology 2010  2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology
 Stability of mango peel by-products S. Sirisakulwat et al. 1651

(90%): 54.25 g L)1] to a final net weight of 338 g and

Sugar residue/Rha (mol mol–1)

700 40
subsequent hot-filling of 100 ± 1 g into each of three 650 35
test cups (Sirisakulwat et al., 2008). After 2 h of gel 600
curing in a water-bath (20 C), BS was measured for 30

[η] (mL g–1)

550
each test cup. TSS and pH of all gel samples were 500 25

MPW 9

MPW 1

MPW 8
experimentally verified. For their normalisation to

MPW 7
MPW 6

MPW 2

MPW 4

MPW 3
450

MPW 5
20
constant AUA levels, BC530HPE and SBC530HPE were 400
related to the AUAtitr content of the AIS and expressed 15
350
as BC530HPE (AUAtitr) [in g of AUA hg)1 of gel] and 10
300
SBC530HPE (AUAtitr) [in g of sugar g)1 of AUA],
250 5
respectively. 0 60 120 180 240 300
MPW Sampling relative to the first sampling point (min)
Intrinsic viscosity Gal/Rha AUA c /Rha AUA titr /Rha
Results and discussion
Figure 1 Variability of peel quality throughout by-product sampling
Impact of peel storage on pectin yields for variants MPW1-MPW9: Intrinsic viscosities ([g]) as well as molar
ratios of galactose to rhamnose (Gal ⁄ Rha) and colorimetric or
Irrespective of the exposure of the wet peels to the titrimetric anhydrogalacturonic acid to rhamnose (AUAc ⁄ Rha,
ambient conditions of the peeling room (25 C; 63% AUAtitr ⁄ Rha) of the alcohol-insoluble substance extracted from
RH) for up to 5 h, the crude pectin yields (AIS) from dried peel waste (MPW) that accrued from the industrial canning of
hot-acid extraction of the dried by-products amounted ‘Kaew Khiew’ mango fruit after exposure of the wet peels to the
to 14.7–16.2 g hg)1 of DRM, with the median ranging ambient conditions of the peeling room (25 C, 63% relative humidity)
at 15.7 g hg)1 (Table 1). Because of comparatively low for different holding times (variants MPW1-MPW9; cf. Table 1)
starch contents of 2.2–5.6 g hg)1 of AIS, the net pectin prior to waste stabilisation. For each variant, the time of sampling is
yields (AISS-corr) were 14.1–15.6 g hg)1 of DRM. The expressed on the abscissa relative to that of variant MPW9, which
was collected first.
chronological order of peel sampling is exemplarily
shown in Fig. 1. As revealed by respective sorting of the
samples, the highest crude and net pectin yields were Industrial hot-acid extraction in pectin recovery from
uniformly recorded for the first (MPW9) and the last established sources (MacDougall & Ring, 2004) was
sample (MPW3), whereas the largest differences were mimicked by the procedure applied on the laboratory
observed for the three samples collected within the first scale. With a median of 15.2 g hg)1 of DRM, the net
64 min (MPW9, MPW1, and MPW8). Moreover, the pectin yields of the samples roughly matched the
peel sample dried without previous dwell period at average reported for hot-acid extraction of dried mango
ambient temperature (MPW1) was among those show- peels (Neidhart et al., 2009), slightly exceeding the
ing the lowest net pectin yield, whereas MPW9 peaked pectin amounts extracted with the same method from
despite the interim storage of the wet peels for 5 h. Thus, peels of the same cultivar (‘Kaew Khiew’, KAk; 12.9–
the temporary exposure of the residues to the ambient 13.1 g hg)1) when the impact of cultivar and fruit
conditions of the peeling room did not affect the ripeness was studied (Sirisakulwat et al., 2008). Findings
achievable pectin yields. In fact, it was the natural detailed elsewhere for peels of related and other mango
heterogeneity of the peel waste that was reflected by the cultivars (Berardini et al., 2005a,b; Koubala et al., 2008)
observed yield variations. were concordant with the median net pectin yield

Table 1 Yields of alcohol-insoluble substance (AIS) extracted from dried peel waste (MPW) that accrued from the industrial canning of
‘Kaew Khiew’ mango fruit after exposure of the wet peels to the ambient conditions of the peeling room (25 C, 63% relative humidity) for
different holding times (HT25 C) up to 5 h prior to waste stabilisation

Variant MPW1 MPW2 MPW3 MPW4 MPW5 MPW6 MPW7 MPW8 MPW9

HT25 °C (min) 0 15 30 60 90 120 180 240 300

AISa (g hg)1 DRM) 14.9 ± 0.1 15.6 ± 0.2 16.2 ± 0.2 15.7 ± 0.2 15.8 ± 0.2 15.5 ± 0.3 15.7 ± 0.1 14.7 ± 0.3 16.2 ± 0.1
Starchb (g hg)1 AIS) 5.6 ± 0.1 4.0 ± 0.2 3.7 ± 0.02 3.0 ± 0.1 4.2 ± 0.2 2.2 ± 0.03 3.6 ± 0.2 3.9 ± 0.2 3.8 ± 0.002
AISS-corrc (g hg)1 DRM) 14.1 ± 0.1 14.9 ± 0.2 15.6 ± 0.2 15.3 ± 0.2 15.1 ± 0.2 15.2 ± 0.3 15.2 ± 0.1 14.1 ± 0.3 15.6 ± 0.1

MPW, mango peel waste.

a
Total yield of extractable polymers as AIS in g hg)1 of dried raw material (DRM, here: MPW).
b
Starch content of the AIS according to enzymatic quantification.
c
Net pectin yield (AISS-corr) as starch-corrected AIS in g hg)1 of DRM (MPW).

 2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology International Journal of Food Science and Technology 2010
1652 Stability of mango peel by-products S. Sirisakulwat et al.

mentioned above. At last, the mango peels were quite experimental full-ripe fruit batch of the same cultivar
similar to apple pomace in terms of AIS yields (Endress (specified as cv. ‘Kaew Khiew’ at ripeness stage III;
et al., 2006). Sirisakulwat et al., 2008). Similarities between this
However, in wet mango peels, the pectic substances sample and the nine pectins from industrial peel waste
obviously degraded less readily compared to wet apple were furthermore observed in terms of the total neutral
pomace, although the mango fruit may predominantly sugars (ANS, 35.9 ± 0.1 vs. 31–36 g hg)1) as well as the
be processed at higher ambient temperature and humid- titrimetrically quantified anhydrogalacturonic acid
ity. Unlike apple pomace, MPW usually accrues with far (AUAtitr, 34.7 ± 0.1 vs. 38–42 g hg)1) and methylester
less tissue disintegration. In manual peeling, which is contents (MeOHtitr, 4.38 ± 0.02 vs. 4.7–5.2 g hg)1).
still usual practice in many canning and drying factories, The slightly higher values of the latter nine samples
thick peel strips accumulate. Therefore, cell de-com- (Table 2) might largely be ascribed to improved purity,
partmentalisation is restricted to the thin cutting edges which became evident in lower ash contents (16.2 ± 0.2
and the mesocarp-facing cut surface. Consistently, cold vs. 8–10 g hg)1) at comparable moisture levels (4.3 ±
storage (2 ± 1 C) of fresh and blanched MPW (mois- 0.1 vs. 4.9–5.7 g hg)1). However, it should be noted
ture content: 85–89%) for 3 and 6 days, respectively, that, unlike the previous study (Sirisakulwat et al.,
was considered tolerable without loss of pectin yield 2008), mass balances of 90–93% [total (2), Table 2],
(Sudhakar & Maini, 2000), but pectin quality was not resulting from the total neutral sugars, both titrimetri-
evaluated in this context. cally analysed parameters, ash, and moisture, indicated
slight underestimation of the composition for the nine
AIS samples. Proteins reported for acid-extracted
Impact of peel storage on pectin structure
mango peel pectins might contribute further 1.1–3.3%
Like the yield, also the overall composition of the (Kratchanova et al., 1991). Whereas the AUAtitr and
pectins was apparently unaffected by the exposure of MeOHtitr contents were expected to be confirmed by
the wet mango peels to the ambient conditions of the further methods applied (Sirisakulwat et al., 2008), the
peeling room for up to 5 h (Table 2). With 4.8 ± colorimetric anhydrogalacturonic acid contents of
0.2 g hg)1 of AIS, a low starch content within the limits the nine samples (AUAc, 49–58 g hg)1) systematically
observed for the nine AIS samples (Table 1) was exceeded the titrimetric values (AUAtitr) by approxi-
previously reported for the pectin, which had been mately 12 g hg)1 on average, the methoxyl contents
extracted in the same way (pH 1.5) from the peels of an resulting from methanol analysis by HPLC (MeOH,

Table 2 Composition of the alcohol-insoluble substance (AIS) extracted from dried peel waste (MPW) that accrued from the industrial canning of
‘Kaew Khiew’ mango fruit after exposure of the wet peels to the ambient conditions of the peeling room (25 C, 63 % relative humidity) for
different holding times (HT25 C) up to 5 h prior to waste stabilisation: Mass balancea

Variant MPW1 MPW2 MPW3 MPW4 MPW5 MPW6 MPW7 MPW8 MPW9

HT25 °C (min) 0 15 30 60 90 120 180 240 300

Moisture 5.7 ± 0.1 4.9 ± 0.1 5.1 ± 0.1 5.3 ± 0.03 5.5 ± 0.01 5.5 ± 0.04 5.3 ± 0.1 5.5 ± 0.2 5.2 ± 0.1
Ash 9.0 ± 0.1 8.0 ± 0.1 8.4 ± 0.004 8.2 ± 0.002 8.4 ± 0.001 8.6 ± 0.03 8.7 ± 0.1 8.9 ± 0.02 10.1 ± 0.1
ANSb 35.7 ± 0.5 33.4 ± 0.6 32.9 ± 0.2 32.2 ± 0.2 33.8 ± 0.2 31.1 ± 0.5 33.3 ± 0.3 32.3 ± 0.1 31.7 ± 1.1
AUAc 48.5 ± 1.2 53.1 ± 0.7 50.7 ± 0.7 55.2 ± 1.2 49.9 ± 0.8 54.9 ± 0.8 50.3 ± 1.0 58.0 ± 0.5 51.0 ± 1.2
MeOHc 6.09 ± 0.23 7.27 ± 0.10 7.42 ± 0.11 7.30 ± 0.10 6.80 ± 0.10 7.16 ± 0.08 6.40 ± 0.24 6.74 ± 0.03 6.76 ± 0.08
Acc 0.55 0.71 0.78 0.72 0.76 0.67 0.69 0.50 0.63

Total (1) 105.4 107.3 105.2 108.9 105.1 107.9 104.7 112.0 105.4

AUAtitr 38.2 ± 0.9 41.2 ± 0.2 40.2 ± 0.3 42.2 ± 0.6 37.5 ± 0.2 40.8 ± 0.1 40.5 ± 0.4 40.4 ± 0.9 38.9 ± 0.5
MeOHtitr 4.69 ± 0.14 5.05 ± 0.03 4.93 ± 0.05 5.23 ± 0.10 4.68 ± 0.04 5.00 ± 0.02 5.06 ± 0.07 4.89 ± 0.16 4.82 ± 0.04

Total (2) 93.1 92.5 91.5 93.1 89.9 91.0 92.9 92.0 90.7

ANS, total content of neutral sugars, calculated as their anhydro forms; AUAc, colorimetric anhydrogalacturonic acid content; MeOH, methoxyl content
(as methanol); Ac, acetyl content (as acetic acid); AUAtitr, titrimetric anhydrogalacturonic acid content; MeOHtitr, titrimetric methylester content; Total
(1), mass balance based on the contents of moisture, ash, ANS, AUAc, MeOH, and Ac; Total (2), mass balance based on the contents of moisture, ash,
ANS, AUAtitr, and MeOHtitr.
a
Mean content ± standard error (SE) in g hg)1 of AIS (Ac: SE £ 0.01 g hg)1).
b
Including the glucose content of starch (cf. Table 1) and ribose (£ 0.24 g hg)1).
c
Quantified by HPLC.

International Journal of Food Science and Technology 2010  2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology
 Stability of mango peel by-products S. Sirisakulwat et al. 1653

a
Table 3 Esterification indices of the alcohol-insoluble substance (AIS) extracted from dried peel waste (MPW) that accrued from the
industrial canning of ‘Kaew Khiew’ mango fruit after exposure of the wet peels to the ambient conditions of the peeling room (25 C, 63%
relative humidity) for different holding times (HT25 C) up to 5 h prior to waste stabilisation

Variant MPW1 MPW2 MPW3 MPW4 MPW5 MPW6 MPW7 MPW8 MPW9

HT25 °C (min) 0 15 30 60 90 120 180 240 300

DE (%) 69.7 ± 0.5 69.7 ± 0.1 69.6 ± 0.2 70.4 ± 0.4 70.8 ± 0.1 69.6 ± 0.1 71.0 ± 0.3 68.8 ± 0.7 70.4 ± 0.4
DMe (%) 69.0 ± 3.0 75.2 ± 1.4 80.5 ± 1.6 72.7 ± 1.9 74.9 ± 1.6 71.7 ± 1.3 69.9 ± 3.0 63.9 ± 0.7 72.8 ± 1.9
DAc (%) 3.3 ± 0.1 3.9 ± 0.1 4.5 ± 0.1 3.8 ± 0.1 4.5 ± 0.1 3.6 ± 0.1 4.1 ± 0.1 2.5 ± 0.02 3.6 ± 0.1

DE, degree of esterification (as molar MeOHtitr ⁄ AUAtitr ratio, cf. Table 2); DMe, degree of methylation (as molar MeOH ⁄ AUAc ratio, cf. Table 2); DAc,
degree of acetylation (as molar Ac ⁄ AUAc ratio, cf. Table 2).
a
Mean ± standard error.

6.1–7.4 g hg)1) surpassed the titrimetric methylester displaying a median of 72.7% (Table 3). Evidently, the
contents (MeOHtitr) by approximately 2 g hg)1 two extreme DMe levels of 64% (MPW8) and 81%
(Table 2). As the respective mass balance averaged out (MPW3) represented outliers ascribed to overestimated
at 106.8% [total (1), Table 2], especially the colorimetric maximum contents of AUAc and MeOH, respectively,
anhydrogalacturonic acid contents were obviously over- but were not causally related to the storage of the wet
estimated throughout this study, most notably in case of peels at 25 C. Likewise, the DAc was unaffected by the
sample MPW8. Despite the deficiencies of the associated storage of the wet peels, as shown by its slight variation
mass balance [total (2)] in this study, the precision of the around a median of 3.8% (Table 3). In conformity with
titrimetric results (AUAtitr and MeOHtitr) of the nine the nine AIS samples from industrial peel waste, the
samples overall appeared to be better than their AUAc ‘Kaew Khiew’ peel pectin mentioned earlier for com-
and MeOH contents. parison was characterised by a DE of 71.8 ± 0.1%, a
Because the titrimetrically quantified DE of the DMe of 77.8 ± 2.0%, and a DAc of 3.6 ± 0.1%
galacturonans closely varied around a median of (Sirisakulwat et al., 2008).
69.7% (Table 3), de-esterification by PME during peel As detailed above, all nine AIS samples revealed
storage at 25 C was irrelevant. A DE of 70% was almost equal contents of anhydrogalacturonic acid and
uniformly observed when the wet peels had been dried total neutral sugars (Table 2). Among the latter, galac-
without a dwell time at 25 C (MPW1) and after storage tose (Gal) residues were by far prevailing (Table 4) with
for 5 h at this temperature (MPW9). Despite the a median of 145 mmol hg)1 of AISS-corr, corresponding
problems faced in this study with the colorimetric to 23–25 g hg)1 of AISS-corr and 59–66 mol hmol)1 of
AUAc and the chromatographic MeOH analysis, the AUAtitr. Although the galactose contents seemed to
findings in terms of DE were confirmed by the DMe be maximum and minimum within that range after

a
Table 4 Molar sugar composition of the alcohol-insoluble substance on a starch-free basis (AISS-corr) after recovery from dried peel waste
(MPW) that accrued from the industrial canning of ‘Kaew Khiew’ mango fruit after exposure of the wet peels to the ambient conditions of the
peeling room (25 C, 63% relative humidity) for different holding times (HT25 C) up to 5 h prior to waste stabilisation

Variant MPW1 MPW2 MPW3 MPW4 MPW5 MPW6 MPW7 MPW8 MPW9

HT25 °C (min) 0 15 30 60 90 120 180 240 300

AUAc 292 ± 7 314 ± 7 299 ± 7 323 ± 9 296 ± 8 319 ± 9 296 ± 7 343 ± 11 301 ± 8
AUAtitr 230 ± 6 244 ± 4 237 ± 5 247 ± 5 223 ± 5 237 ± 6 238 ± 3 239 ± 9 230 ± 4
Gal 151 ± 3 146 ± 5 145 ± 3 145 ± 3 142 ± 3 144 ± 5 145 ± 2 144 ± 5 139 ± 7
Ara 16 ± 0 15 ± 0 17 ± 1 16 ± 0 21 ± 1 14 ± 0 19 ± 1 14 ± 0 17 ± 1
Glcb 19 ± 2 15 ± 1 14 ± 1 14 ± 1 17 ± 1 12 ± 1 15 ± 1 15 ± 1 13 ± 2
Rha 8.9 ± 0.1 10.6 ± 0.6 10.2 ± 0.3 9.9 ± 0.2 9.9 ± 0.2 9.8 ± 0.3 10.0 ± 0.3 9.0 ± 0.3 8.8 ± 0.7
Man 1.8 ± 0.1 2.0 ± 0.04 2.1 ± 0.1 2.2 ± 0.1 2.4 ± 0.1 2.3 ± 0.1 2.2 ± 0.03 1.8 ± 0.1 1.9 ± 0.04
Xyl 1.7 ± 0.03 2.0 ± 0.03 2.2 ± 0.04 1.8 ± 0.1 2.2 ± 0.1 2.0 ± 0.1 2.0 ± 0.1 1.5 ± 0.1 1.6 ± 0.03
Fuc 0.4 ± 0.03 0.4 ± 0.01 0.4 ± 0.02 0.4 ± 0.01 0.5 ± 0.01 0.4 ± 0.01 0.5 ± 0.03 0.4 ± 0.01 0.4 ± 0.03

AUAc, galacturonic acid (colorimetric); AUAtitr, galacturonic acid (titrimetric); Gal, galactose; Ara, arabinose; Glc, glucose; Rha, rhamnose; Man,
mannose; Xyl, xylose; Fuc, fucose.
a
Mean ± standard error in mmol hg)1 of AISS-corr.
b
Without glucose included in starch.

 2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology International Journal of Food Science and Technology 2010
1654 Stability of mango peel by-products S. Sirisakulwat et al.

exposure of the wet peels to 25 C for 0 (MPW1) and purification in acidic alcohol amounted to 56–59 g hg)1
5 h (MPW9), respectively, constant levels were observed of AISp (Table 5), the nine samples did not meet the
for the other seven AIS samples. On the whole, the nine JECFA minimum requirement as to the galacturonide
pectins revealed, regardless of the storage of the wet peel content (65 g hg)1 of AISp; Joint FAO ⁄ WHO Expert
waste, an almost identical neutral sugar composition Committee on Food Additives (JECFA), 2007). How-
(Table 4) as the peel pectin of the experimental ‘Kaew ever, the portions of homogalacturonans (HG) and rham-
Khiew’ fruit batch cited earlier for comparison did nogalacturonans (RG-I) were obviously not affected by
(Sirisakulwat et al., 2008). Their arabinose (Ara) con- the wet peel storage, because largely constant molar
tents slightly varied around a median of 16 mmol hg)1 AUAtitr ⁄ Rha ratios of 23–26 occurred throughout, as
of AISS-corr (i.e. 2.1 g hg)1), followed by median con- indicated by the respective sugar amounts displayed
tents of 15, 9.9, 2.1, and 2.0 mmol hg)1 of AISS-corr in Table 4. Also the backbone indices (Table 5), which
for glucose (Glc) not contained in starch, rhamnose were estimated from the molar AUAc and rhamnose
(Rha), mannose (Man), and xylose (Xyl), respectively contents on the basis of a simplified backbone model
(Table 4). Strongest variation among the nine samples consisting of HG and RG-I (Sirisakulwat et al., 2008),
was observed for arabinose and the nonstarch glucose did not point to any significant backbone degradation
contents. However, no impact of wet peel storage was due to the wet peel storage, because largely constant
noticed. maximum percentages of the pectin backbone resulted
As shown previously (Neidhart et al., 2009), galactose for homogalactacuronans (BBPHG, 93.4–94.9%) and
contents of mango peel pectins may be considerably high, rhamnogalacturonans type I (BBPRG-I, 5.1–6.6%). If all
but they seemed to be specific to the cultivar and were galactose residues were ascribed to galactans with one
neither affected by fruit ripeness nor by the extraction chain branching off from each rhamnose unit of rham-
mode. Thus, even in the presence of minor natural starch nogalacturonans type I, long and ⁄ or highly ramified
contents, the galacturonide content of mango peel AIS galactan side chains of 14–17 residues may be assumed
may markedly be limited. This feature also applied to the according to the molar Gal ⁄ Rha ratios (Table 5). This
nine AIS samples of this study. Because their galact- striking peculiarity and the AUAtitr ⁄ Rha ratio, observed
uronic acid contents (GalUAtitr, Mr = 194.1 g mol)1) here for AIS from industrial peel waste, confirmed
on an ash- and moisture-free base (AISp) after pectin the former report (Sirisakulwat et al., 2008) for the

a
Table 5 Structural indices of the alcohol-insoluble substance (AIS) extracted from dried peel waste (MPW) that accrued from the industrial
canning of ‘Kaew Khiew’ mango fruit after exposure of the wet peels to the ambient conditions of the peeling room (25 C, 63 % relative humidity)
for different holding times (HT25 C) up to 5 h prior to waste stabilisation

Variant MPW1 MPW2 MPW3 MPW4 MPW5 MPW6 MPW7 MPW8 MPW9

HT25 °C (min) 0 15 30 60 90 120 180 240 300

)1
GalUAtitr (g hg of AISp) 55.5 ± 1.3 58.2 ± 0.3 56.7 ± 0.4 59.0 ± 0.8 58.5 ± 0.3 58.5 ± 0.2 59.2 ± 0.6 58.8 ± 1.4 58.1 ± 0.8
Molecular parameters
[g] (mL g)1) 282 ± 2 317 ± 2 341 ± 2 329 ± 2 334 ± 2 308 ± 2 325 ± 3 276 ± 2 271 ± 2
Side chain indices
Ara ⁄ Rha (mol mol)1) 1.8 ± 0.02 1.4 ± 0.1 1.7 ± 0.1 1.6 ± 0.04 2.1 ± 0.04 1.5 ± 0.03 1.9 ± 0.1 1.5 ± 0.03 1.9 ± 0.2
Gal ⁄ Rha (mol mol)1) 16.9 ± 0.3 13.8 ± 0.8 14.2 ± 0.4 14.6 ± 0.1 14.2 ± 0.1 14.7 ± 0.4 14.5 ± 0.4 16.0 ± 0.3 15.7 ± 1.4
Ara ⁄ Gal (mol hmol)1) 10.7 ± 0.2 10.4 ± 0.3 11.7 ± 0.5 11.1 ± 0.3 14.4 ± 0.3 10.0 ± 0.3 12.9 ± 0.3 9.5 ± 0.1 12.2 ± 0.7
Backbone indices (of AISS-corr)
HGmaxb (mmol hg)1) 283 ± 7 304 ± 7 288 ± 7 313 ± 9 286 ± 8 309 ± 9 286 ± 7 334 ± 11 293 ± 8
RG-Imaxc (mmol hg)1) 17.9 ± 0.2 21.1 ± 1.1 20.4 ± 0.7 19.8 ± 0.3 19.9 ± 0.4 19.6 ± 0.5 20.1 ± 0.5 17.9 ± 0.7 17.6 ± 1.3
BBPHGd (%) 94.1 93.5 93.4 94.1 93.5 94.0 93.4 94.9 94.3
BBPRG-Ie (%) 5.9 6.5 6.6 5.9 6.5 6.0 6.6 5.1 5.7

GalUAtitr, titrimetric galacturonic acid content (g hg)1 of AISp, i.e., of nondestarched AIS after acidic purification before titration); [g], intrinsic viscosity
(AIS dissolved in 0.28 M KOAc ⁄ HOLc buffer, pH 3.0); Ara, arabinose (mmol hg)1 of AIS); Rha, rhamnose (mmol hg)1 of AIS); Gal, galactose (mmol hg)1
of AIS).
a
Mean ± standard error.
b
Estimated maximum units of the total homogalacturonan backbone (HG) calculated from the AUAc and Rha contents in mmol hg)1 of AISS-corr
(HGmax = AUAc-Rha).
c
Estimated maximum units of the total rhamnogalacturonan-I backbone (RG-I) calculated from the Rha content in mmol hg)1 of AISS-corr
(RG-Imax = 2 Rha).
d
Maximum homogalacturonan (HG) percentage of the pectin backbone [BBPHG = 100(AUAc ) Rha)/(AUAc + Rha)].
e
Maximum rhamnogalacturonan-I (RG-I) percentage of the pectin backbone [BBPRG)I = 100(2ÆRha)/(AUAc + Rha)].

International Journal of Food Science and Technology 2010  2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology
 Stability of mango peel by-products S. Sirisakulwat et al. 1655

analogously extracted peel AIS from the experimental Table 5 clearly exceeded the intrinsic viscosity of the
‘Kaew Khiew’ fruit batch. But galactan degradation reported ‘Kaew Khiew’ peel pectin (234 mL g)1), which
during storage of the wet peels by b-d-galactosidase had displayed a large arabinogalactan fraction [(Ara +
activities was obviously negligible under the conditions Gal) ⁄ Rha = 19.2] similar to the MPW1 pectin, but only
applied. Because of the acid-lability of arabinofuranosyl molar AUA ⁄ Rha ratios of 23 and 24 based on AUAc
linkages, arabinose residues are more or less split off and AUAtitr, respectively (Sirisakulwat et al., 2008). As
during hot-acid extraction, depending on the pH condi- previously shown for this pectin by high-performance
tions applied (Sirisakulwat et al., 2008; Neidhart et al., size exclusion chromatography, its comparatively low
2009). However, like the Gal ⁄ Rha ratios, also molar intrinsic viscosity resulted from the co-existence of two
Ara ⁄ Rha ratios of 1.4–2.1 around a median of 1.7 mol ⁄ - distinct major fractions. Apart from a high-molecular
mol and Ara ⁄ Gal ratios of 10–14 mol hmol)1 (Table 5) weight fraction also found in commercial apple pectin,
pointed to insignificant side-chain degradation during the an additional, almost monodisperse fraction with a peak
storage of the wet peels. molecular weight of 17 000 relative to dextran consid-
The clearest evidence for the preservation of pectin erably reduced the average molecular weight of this
backbones and side chains during the storage of the wet pectin (Sirisakulwat et al., 2008). Depending on the
peels at 25 C was finally provided by the intrinsic amounts of this characteristic fraction, the gelling and
viscosities ([g]) of the samples (Table 5). Within a range thickening properties of mango peel pectins may be
of 271–341 mL g)1, [g] closely varied around a median more or less limited, as demonstrated by numerous
of 317 mL g)1, indicating good consistency among the samples (Neidhart et al., 2009).
nine pectins in terms of their average molecular dimen-
sions (Morris, 1995). The lowest intrinsic viscosities
Impact of peel storage on techno-functional pectin
(271–282 mL g)1) were observed for the AIS group
properties
from the three peel waste samples collected within the
first 64 min of sampling (MPW1, MPW8, MPW9, From the similarities in the average composition
Fig. 1), including the extreme situations, i.e. immediate (Tables 2-4) and the intrinsic viscosity reflecting the mean
stabilisation and exposure to 25 C for 5 h prior to molecular weight (Table 5), no differences in the techno-
drying, respectively. Thus, the slight differences among functional properties were expected among the nine
the nine pectins were definitely ascribed to the inherent pectins. On the other hand, weak activities of the
heterogeneity of the fruits and the reproducibility of respective esterases and depolymerising hydrolyses and
pectin recovery (Sirisakulwat et al., 2008) rather than to lyases may nevertheless cause enhanced variation among
the storage of the wet peels at ambient temperature. The the molecules in terms of molecular size, side-chain
rhamnose contents were slightly lower in the AIS from lengths, and the degrees of methylation and acetylation.
the peel samples MPW1, MPW8, and MPW9 (approx- Likewise, the intramolecular distribution of methyl-
imately 9 mmol hg)1 of AISS-corr, Table 4) compared to esterified units might be affected, especially because of
the remaining pectins. Owing to the minor variation of the blockwise mode of action of plant PME (Benen et al.,
galactose contents, these three pectins were further 2003a). Owing to such differences in their fine structure
characterised by maximum Gal ⁄ Rha ratios (Fig. 1). (Guillotin et al., 2007), pectins with similar average
Thus, their arabinogalactan side chains were possibly characteristics may display different gelling behaviour
more extended than in the other cases, as equally (Neidhart et al., 2003).
indicated by their molar (Ara+Gal) ⁄ Rha ratios of Consistent with their low galacturonide contents
18–19 relative to 15–16 for the other pectins (Table 5). (Table 2) and intrinsic viscosities (Table 5), the nine
The intrinsic viscosity would consistently decline along samples produced only soft gels. Low breaking strengths
with a relative increase in branched neutral sugar side of 369–497 HPE resulted even at an AIS dose of
chains due to overall more condensed coil dimensions 0.3 g hg)1 of gel (BS0.3%; Table 6) and 0.35 g hg)1 were
(Morris, 1995). In the present case, this seemed to be necessary to achieve BSs of 452–594 HPE (BS0.35%). As
partly compensated by more expanded homogalacturo- displayed by the AIS amounts required for 100 g of a
nan parts, because especially the AIS from MPW8 and standard gel with a BS of 530 HPE, the nine samples
MPW9 besides MPW1 also displayed elevated molar were quite uniform in terms of their breaking capacities
AUA ⁄ Rha ratios (both on the basis of the AUAtitr and (BC530HPE; Table 6). With a BC530HPE range of 0.33–
the AUAc contents; Fig. 1). Nevertheless, their lower 0.4 g hg)1 of gel, the breaking capacity of the ‘Kaew
intrinsic viscosities were most probably owing to larger Khiew’ peel AIS reported as a reference was confirmed;
arabinogalactan parts, because the MPW1 AIS came but compared to a commercial, unstandardised apple
quite close to the MPW4 and MPW6 samples in terms pectin that was also included in the previous study
of the AUA ⁄ Rha ratios despite the significantly higher (Sirisakulwat et al., 2008), 1.9–2.4 times more AIS was
intrinsic viscosities of the latter group ([g] ‡ 308 mL g)1; needed to produce this standard gel. Consequently, the
Fig. 1). Both the maximum and minimum [g] levels in sugar amount bound in a sugar-acid standard gel by 1 g

 2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology International Journal of Food Science and Technology 2010
1656 Stability of mango peel by-products S. Sirisakulwat et al.

Table 6 Gelling capacity of the alcohol-insoluble substance (AIS) extracted from dried peel waste (MPW) that accrued from the industrial
canning of ‘Kaew Khiew’ mango fruit after exposure of the wet peels to the ambient conditions of the peeling room (25 C, 63 % relative
humidity) for different holding times (HT25 C) up to 5 h prior to waste stabilisation

Variant MPW1 MPW2 MPW3 MPW4 MPW5 MPW6 MPW7 MPW8 MPW9

HT25 °C (min) 0 15 30 60 90 120 180 240 300

Gelling capacity of the AIS and the dried raw material (DRM), respectively
BS0.3% (HPE) 458 ± 3a 478 ± 2 369 ± 3 399 ± 5 463 ± 9 497 ± 8 463 ± 6 416 ± 4 394 ± 1
BS0.35% (HPE) 586 ± 5a 564 ± 13 452 ± 8 594 ± 5 570 ± 6 563 ± 4 587 ± 14 528 ± 6 509 ± 2
BC530HPE (g hg)1) 0.33 0.33 0.40 0.33 0.33 0.33 0.33 0.35 0.36
SBC530HPE (g g)1 of AIS) 198 197 164 195 196 200 199 185 181
SBCDRM (g hg)1) 2951 3062 2644 3066 3092 3108 3124 2717 2935
Normalised gelling capacity of the AIS relative to the AUAtitr content of the standardised gel with 530 HPE
BC530HPE(AUAtitr) (g hg)1) 0.13 0.14 0.16 0.14 0.12 0.13 0.13 0.14 0.14
SBC530HPE(AUAtitr) (g g)1 of AUA) 519 478 407 462 523 490 491 459 465

a
Mean ± standard error. BS0.3% and BS0.35%, breaking strength of gels with AIS doses of 0.3 and 0.35% (w ⁄ w), respectively, in Herbstreith-Pectinometer
units (HPE); BC530HPE, breaking capacity of the AIS as gram of AIS required for 100 g of a gel with 530 HPE; SBC530HPE, sugar-binding capacity of
the AIS as gram of sugar bound per gram of AIS in a gel of 530 HPE; SBCDRM, gelling units (GU) of the dried raw material (DRM; here: MPW) as its
sugar-binding capacity in gram of sugar bound by 100 g of DRM in a gel of 530 HPE; BC530HPE(AUAtitr), normalised breaking capacity of the AIS
(g of AUA hg)1 of a gel with 530 HPE); SBC530HPE(AUAtitr), normalised sugar-binding capacity of the AIS (g of sugar g)1 of AUA for a gel of 530 HPE).

of this apple pectin exceeded the sugar-binding capacities ison to pectins with improved techno-functionality from
(SBC530HPE) of the nine mango AIS samples (164– other mango peels (Neidhart et al., 2009).
200 g g)1; Table 6) by the same factors. Because of their Like the gel properties, the setting behaviour of the
uniform AIS yields (AISS-corr; Table 1) that slightly AIS was also not verifiably affected by the intermediate
exceeded that of this former ‘Kaew Khiew’ peel sample exposure of the wet peels to ambient temperature.
(13.1 ± 0.2 g hg)1; Sirisakulwat et al., 2008), the dried Despite some variation of the setting temperature that
peels MPW1-MPW9 had the capacity to bind more sugar was more pronounced at the upper AIS dose, equally
in a standard gel by the recovered AIS than the latter high setting temperatures were recorded for AIS that
(SBCDRM; 2644–3124 vs. 2250 g hg)1 of dried peel; was obtained from peels after their wet storage at 25 C
Table 6). Thus, the industrial peel waste yielded slightly for 0, 15, and 300 min (Fig. 2).
more AIS than the cited peel sample, but of comparable
quality. The three samples displaying minimum intrin-
sic viscosities of the extractable AIS (MPW1, MPW8, 45
MPW9; Fig. 1) were among the four samples with a 40
SBCDRM below the median. However, together with the
T (δ1.0 Hz = 45°) (°C)

35
other AIS samples from industrial peel waste, their AIS
30
was almost identical to that of the cited ‘Kaew Khiew’
peel sample (0.14 g hg)1; Sirisakulwat et al., 2008) and 25

even came close to this apple pectin (0.11 g hg)1; 20

Sirisakulwat et al., 2008) in terms of the normalised 15 cP (g hg–1 of gel):

gelling capacity [BC530HPE(AUAtitr); Table 6], when the 10 0.4

0.55
breaking capacities (BC530HPE) were expressed as the 5
respective AUAtitr doses required for a standard gel of 0
530 HPE. Thus, the functional weakness of mango peel 0 50 100 150 200 250 300 350
AIS shown by the non-normalised parameters in Table 6 HT25 °C (min)
was neither caused by the use of industrial peel waste
as an AIS source nor by the extended exposure of the Figure 2 Gelation of alcohol-insoluble substance (AIS) samples
extracted from dried peel waste (MPW) that accrued from the
wet peels to ambient temperature. Also under the given
industrial canning of ‘Kaew Khiew’ mango fruit after exposure of
conditions, deficits seemed to be rather because of the wet peels to the ambient conditions of the peeling room (25 C,
galacturonide contents, which were markedly lowered 63% relative humidity) for different holding times (HT25 C) up to
in favour of arabinogalactan portions, and the presence 5 h prior to waste stabilisation: Setting temperatures (Td=45) of
of a characteristic fraction that considerably reduced the standard gels (64.7 ± 0.3 Brix and pH 3.25 ± 0.02 according to
average molecular weight, as demonstrated earlier for the concomitant refractometer readings and pH analyses of the cured gels;
cited AIS sample (Sirisakulwat et al., 2008) in compar- cp, AIS dose). Error bars represent the single standard deviations.

International Journal of Food Science and Technology 2010  2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology
 Stability of mango peel by-products S. Sirisakulwat et al. 1657

Conclusion Acknowledgments
The findings reported for mango peel AIS from exper- This research was funded by Deutsche Forschungs-
imental fruit batches (Sirisakulwat et al., 2008; Neidhart gemeinschaft, Bonn, Germany: Project SFB 564-E2.2. It
et al., 2009) were overall confirmed by industrial is part of the Special Research Program ‘Research for
by-products from the fruit of a further crop year. Most Sustainable Land Use and Rural Development in
important, temporary exposure of wet mango peels to Mountainous Regions of Southeast Asia’ (The Uplands
ambient temperature was innocuous in terms of yield, Program).
composition, and techno-functionality of the extract- The authors thank Herbstreith & Fox KG, Neu-
able AIS. The interim storage of the wet peel waste in enbürg, Germany, in particular Hans-Ulrich Endress
the peeling room prior to drying can thus be tolerated and Christine Rentschler, for providing laboratory
even up to 5 h without notable pectin degradation. facilities for pectin extraction as well as Northern Food
Limited tissue damage during manual peeling and low Co., Ltd., Chiang Mai, Thailand for supplying MPW.
activities of degrading endogenous enzymes in the peels Klaus Mix and Martin Leitenberger are acknowledged
(Yanru et al., 1995) might support this storage stability. for their technical assistance with the experiments at
On a relatively large part of the peel waste surface, wax Hohenheim University.
layers and the thickness of the cuticle initially still act as
physical barriers. Additionally, minor antimicrobial
References
effects might still emanate from polyphenolics like
gallotannins and alk(en)ylresorcinols present in mango Ali, Z.M., Chin, L.-H. & Lazan, H. (2004). A comparative study on
peels even at consumption ripeness of the fruit (Berar- wall degrading enzymes, pectin modifications and softening during
ripening of selected tropical fruits. Plant Science, 167, 317–327.
dini et al., 2004; Engels et al., 2009; Knödler et al., Benen, J.A.E., van Alebeek, G.-J.W.M., Voragen, A.G.J. & Visser, J.
2009). Compared to twelve other mango cultivars, peels (2003a). Pectic esterases. In: Handbook of Food Enzymology (edited
of ‘Kaew’ fruit were reported to be comparatively rich by J.R. Whitaker, A.G.J. Voragen & D.W.S. Wong). Pp. 849–856.
in alk(en)ylresorcinols (1.4 g kg)1 of peel dry matter; New York, NY, USA: Marcel Dekker.
Benen, J.A.E., Voragen, A.G.J. & Visser, J.. (2003b). Pectic enzymes.
Knödler et al., 2009), which play a role in plant disease In: Handbook of Food Enzymology (edited by J.R. Whitaker, A.G.J.
resistance (Hassan et al., 2007). Because of the en- Voragen & D.W.S. Wong). Pp. 845–848. New York, NY, USA:
hanced flexibility, the observed tolerance of intermedi- Marcel Dekker.
ate storage is important for a future exploitation of peel Berardini, N., Carle, R. & Schieber, A. (2004). Characterization of
waste from mango processing on smaller industrial gallotannins and benzophenone derivatives from mango (Mangifera
indica L. cv. ‘Tommy Atkins’) peels, pulp and kernels by high-
scales, where immediate peel drying may impose con- performance liquid chromatography ⁄ electrospray ionization mass
siderably higher logistic and economic investment than spectrometry. Rapid Communications in Mass Spectrometry, 18,
on large scale with completely continuous processing 2208–2216.
and waste stabilisation. As mango residues result from Berardini, N., Fezer, R., Conrad, J., Beifuss, U., Carle, R. & Schieber,
A. (2005a). Screening of mango (Mangifera indica L.) cultivars for
multifaceted processing on different scales, the precon- their contents of flavonol O- and xanthone C-glycosides, anthocya-
ditions for economic by-product drying are rather nins, and pectin. Journal of Agricultural and Food Chemistry, 53,
individual. Knowledge about the susceptibility of indus- 1563–1570.
trial mango peels to pectin degradation prior to drying Berardini, N., Knödler, M., Schieber, A. & Carle, R. (2005b).
is of utmost importance for the selection of adjusted Utilization of mango peels as a source of pectin and polyphenolics.
Innovative Food Science and Emerging Technologies, 6, 442–452.
drying technologies and economic concepts for the Brummell, D.A. (2006). Cell wall disassembly in ripening fruit.
overall processes. Functional Plant Biology, 33, 103–119.
The interim storage of the wet mango peels neither Chourasia, A., Sane, V.A. & Nath, P. (2006). Differential expression of
caused perceptible variations in pectin quality nor the pectate lyase during ethylene-induced postharvest softening of
mango (Mangifera indica var. Dashehari). Physiologia Plantarum,
comparably inferior galacturonide contents and gelling 128, 546–555.
capacities of the pectins, which were observed in spite of De Vries, R.P. & Visser, J. (2003). Enzymes releasing L-arabinose and
hot-acid extraction that mimicked industrial procedures D-galactose from the side chains of pectin. In: Handbook of Food
for pectin recovery from established sources. Despite Enzymology (edited by J.R. Whitaker, A.G.J. Voragen & D.W.S.
individual promising findings based on laboratory Wong). Pp. 867–877. New York, NY, USA: Marcel Dekker.
Endress, H.-U. & Dilger, S. (1990). Checking pectin jelly strength with
extraction (Rehman et al., 2004; Koubala et al., 2009; the Pektinometer. Food Technology International, 1, 279–282.
Neidhart et al., 2009), quality deficiencies of mango peel Endress, H.-U., Mattes, F. & Norz, K. (2006). Pectins. In: Handbook
pectin, which have repeatedly been reported (Pedroza- of Food Science, Technology, and Engineering (edited by Y.H. Hui),
Islas et al., 1994) as in the present case, require further Vol. 3. Pp. 140 ⁄ 1–140 ⁄ 35. Boca Raton, FL, USA: CRC Press,
Taylor & Francis Group.
studies. Research into sustainable recovery of purified Engels, C., Knödler, M., Zhao, Y.-Y., Carle, R., Gänzle, M.G. &
pectins from mango peels like those used in the present Schieber, A. (2009). Antimicrobial activity of gallotannins isolated
and previous trials (Sirisakulwat et al., 2008) are from mango (Mangifera indica L.) kernels. Journal of Agricultural
currently under way. and Food Chemistry, 57, 7712–7718.

 2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology International Journal of Food Science and Technology 2010
1658 Stability of mango peel by-products S. Sirisakulwat et al.

Guillotin, S.E., Bakx, E.J., Boulenguer, P., Schols, H.A. & Voragen, Panouillé, M., Ralet, M.-C., Bonnin, E. & Thibault, J.-F. (2007).
A.G.J. (2007). Determination of the degree of substitution, degree of Recovery and reuse of trimmings and pulps from fruit and vegetable
amidation and degree of blockiness of commercial pectins by using processing. In: Handbook of Waste Management and Co-product
capillary electrophoresis. Food Hydrocolloids, 21(3), 444–451. Recovery in Food Processing (edited by K. Waldron), Vol. 1. Pp.
Hassan, M.K., Dann, E.K., Irving, D.E. & Coates, L.M. (2007). 417–447. Cambridge, UK: Woodhead.
Concentrations of constitutive alk(en)ylresorcinols in peel of com- Pedroza-Islas, R., Aquilar-Esperanza, E. & Vernon-Carter, E.J.
mercial mango varieties and resistance to postharvest anthracnose. (1994). Obtaining pectins from solids wastes derived from mango
Physiological and Molecular Plant Pathology, 71, 158–165. (Mangifera indica) processing. AIChE Symposium Series, 300, 36–41.
Joint FAO ⁄ WHO Expert Committee on Food Additives (JECFA) Pott, I., Neidhart, S., Mühlbauer, W. & Carle, R. (2005). Quality
(2007). Compendium of Food Additive Specifications. FAO JECFA improvement of non-sulphited mango slices by drying at high
Monographs 4. Rome, Italy: Food and Agriculture Organization temperatures. Innovative Food Science & Emerging Technologies,
of the United Nations. http://www.fao.org/docrep/010/a1447e/ 6(4), 412–419.
a1447e00.htm. Prasanna, V., Prabha, T.N. & Tharanathan, R.N. (2006). Multiple
Knödler, M., Reisenhauer, K., Schieber, A. & Carle, R. (2009). forms of polygalacturonase from mango (Mangifera indica L. cv.
Quantitative determination of allergenic 5-alk(en)yl-resorcinols in Alphonso) fruit. Food Chemistry, 95, 30–36.
mango (Mangifera indica L.) peel, pulp, and fruit products by high- Rehman, Z.U., Salariya, A.M., Habib, F. & Shah, W.H. (2004).
performance liquid chromatography. Journal of Agricultural and Utilization of mango peels as a source of pectin. Journal of the
Food Chemistry, 57, 3639–3644. Chemical Society of Pakistan, 26(1), 73–76.
Koubala, B.B., Kansci, G., Mbome, L.I., Crépeau, M.-J., Thibault, Singh, P. & Dwivedi, U.N. (2008). Purification and characterization of
J.-F. & Ralet, M.-C. (2008). Effect of extraction conditions on some multiple forms of polygalacturonase from mango (Mangifera indica
physicochemical characteristics of pectins from ‘‘Améliorée’’ and cv. Dashehari) fruit. Food Chemistry, 111, 345–349.
‘‘Mango’’ mango peels. Food Hydrocolloids, 22(7), 1345–1351. Sirisakulwat, S., Nagel, A., Sruamsiri, P., Carle, R. & Neidhart, S.
Koubala, B.B., Kansci, G., Garnier, C., Mbome, I.L., Durand, S., (2008). Yield and quality of pectins extractable from the peels of
Thibault, J.-F. & Ralet, M.-C. (2009). Rheological and high gelling Thai mango cultivars depending on fruit ripeness. Journal of
properties of mango (Mangifera indica) and ambarella (Spondias Agricultural and Food Chemistry, 56(22), 10727–10738.
cythera) peel pectins. International Journal of Food Science and Sudhakar, D.V. & Maini, S.B. (2000). Isolation and characterization
Technology, 44, 1809–1817. of mango peel pectins. Journal of Food Processing and Preservation,
Kratchanova, M., Bénémou, C. & Kratchanov, C. (1991). On the 24(3), 209–227.
pectic substances of mango fruits. Carbohydrate Polymers, 15, Vásquez-Caicedo, A.L., Neidhart, S. & Carle, R. (2004). Postharvest
271–282. ripening behavior of nine Thai mango cultivars and their
MacDougall, A.J. & Ring, S.G. (2004). Pectic polysaccharides. In: suitability for industrial applications. Acta Horticulturae, 645,
Chemical and Functional Properties of Food Saccharides (edited by 617–625.
P. Tomasik). Pp. 181–195. Boca Raton, FL, USA: CRC Press. Vásquez-Caicedo, A.L., Heller, A., Neidhart, S. & Carle, R. (2006).
Morris, E.R. (1995). Polysaccharide rheology and in-mouth percep- Chromoplast morphology and b-carotene accumulation during
tion. In: Food Polysaccharides and Their Applications (edited by postharvest ripening of mango cv. ‘Tommy Atkins’. Journal of
A.M. Stephen). Pp. 517–546. New York, NY, USA: Marcel Dekker. Agricultural and Food Chemistry, 54, 5769–5776.
Neidhart, S., Hannak, C. & Gierschner, K. (2003). Sol-gel transition Vásquez-Caicedo, A.L., Schilling, S., Carle, R. & Neidhart, S. (2007).
of high-esterified pectins and their molecular structure. In: Advances Effects of thermal processing and fruit matrix on b-carotene stability
in Pectin and Pectinase Research (edited by A.G.J. Voragen, H.A. and enzyme inactivation during transformation of mangoes into
Schols & R. Visser). Pp. 431–448. Dordrecht, The Netherlands: purée and nectar. Food Chemistry, 102(4), 1172–1186.
Kluwer Academic Publishers. Voragen, A.G.J., Pilnik, W., Thibault, J.-F., Axelos, M.A.V. &
Neidhart, S., Sirisakulwat, S., Nagel, A., Sruamsiri, S. & Carle, R. Renard, C.M.G.C.. (1995). Pectins. In: Food Polysaccharides and
(2009). Which mango processing residues are suitable for pectin Their Applications (edited by A.M. Stephen). Pp. 287–339. New
recovery in terms of yield, molecular and techno-functional York, NY, USA: Marcel Dekker.
properties of extractable pectins? In: Pectins and Pectinases Yanru, Z., Pandey, M., Prasad, N.K. & Srivastava, G.C. (1995).
(edited by H.A. Schols, R.G.F. Visser & A.G.J. Voragen). Pp. Ripening associated changes in enzymes and respiratory activities in
177–195. Wageningen, The Netherlands: Wageningen Academic three varieties of mango (Mangifera indica L.). Indian Journal of
Publishers. Plant Physiology, 38(1), 73–76.

International Journal of Food Science and Technology 2010  2010 The Authors. Journal compilation  2010 Institute of Food Science and Technology