Mango pectin quality as influenced by cultivar, ripeness, peel particle size, blanching, drying, and irradiation

Food Hydrocolloids 51 (2015) 241e251

Contents lists available at ScienceDirect

Food Hydrocolloids journal homepage: www.elsevier.com/locate/foodhyd

Mango pectin quality as influenced by cultivar, ripeness, peel particle size, blanching, drying, and irradiation Christian Hubert Geerkens a, Andreas Nagel a, Kathrin Meike Just a, Petra Miller-Rostek a, Dietmar Rolf Kammerer b, Ralf Martin Schweiggert a, *, Reinhold Carle a, c a

University of Hohenheim, Institute of Food Science and Biotechnology, Chair Plant Foodstuff Technology and Analysis, Garbenstrasse 25, 70599 Stuttgart, Germany €lden, WALA Heilmittel GmbH, Department of Analytical Development & Research, Section Phytochemical Research, Dorfstrasse 1, 73087 Bad Boll/Eckwa Germany c King Abdulaziz University, Faculty of Science, Biological Science Department, P. O. Box 80257, Jeddah 21589, Saudi Arabia b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 26 January 2015 Received in revised form 20 April 2015 Accepted 2 May 2015 Available online xxx

Industrial recovery and application of valuable mango (Mangifera indica L.) peel constituents, such as dietary fiber and pectins, require the conversion of the yet under-utilized and highly perishable byproduct into a stable commodity. Focusing on efficient pectin recovery, the impact of different cultivars and ripeness degrees as well as various technological procedures on pectin quality by affecting pectin yield, molecular size distribution of pectic polymers, galacturonic acid content, degree of esterification, and content of interfering substances was analyzed. Cultivar and ripeness degree revealed a significant effect on pectin quality. Preservation processes, i.e. oven drying and lyophilization each with and without previous blanching of integral fruits as well as gamma irradiation, notably influenced the quality of the obtained pectin. Blanching prior to drying reduced arabinogalactan and ash impurities, whereas galacturonic acid contents were increased. Most importantly, grinding of dried mango peels to obtain a particle size of ca. 42 mm (d43) significantly enhanced both extraction yield (þ70%) and galacturonic acid content (þ20%) without increasing the contents of the above mentioned impurities as compared to a peel particle size of 10 mm. Mango pectin produced from such peel powders with a small particle size (120 mm) improved breaking and sugar binding capacities as well as gelling units (up to 5476 GU). The production of mango peel pectin and its applications were favored by implementing the proposed procedures into the valorization cascade of mango peels. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Galacturonic acid Neutral sugar Molecular weight distribution Gel Dietary fiber Mangifera indica L.

1. Introduction Due to its savory taste and high nutritive value (Tharanathan, Yashoda, & Prabha, 2006), worldwide mango (Mangifera indica L.) production has continuously increased to 46.7 million mt in 2012 (FAOSTAT, 2014). Besides their sale on fresh markets, value added mango products such as juice, canned pulp, and chutneys are produced at industrial scale. The accruing by-products, i.e. the stones and peels, often generate a major disposal problem for producers. While the fat of the stones can be recovered and utilized as cocoa butter equivalent up to a maximum of 5% in certain EU member states (Directive 2000/36/EC), a valorization process for

* Corresponding author. Tel.: þ49 (0) 711 459 22995; fax: þ49 (0) 711 459 24110. E-mail address: [email protected] (R.M. Schweiggert). http://dx.doi.org/10.1016/j.foodhyd.2015.05.022 0268-005X/© 2015 Elsevier Ltd. All rights reserved.

the peels is still urgently needed. The recovery of valuable phytochemicals, such as flavonol glycosides, monoterpenes, alkylresorcinols, and, most importantly, mango pectin, has been proposed previously (Neidhart, Sirisakulwat, Nagel, Sruamsiri, & Carle, 2009; Schieber, Stintzing, & Carle, 2001; Sirisakulwat, Nagel, Sruamsiri, Carle, & Neidhart, 2008). The high content in such valuable bioactives and techno-functional compounds as well as their excellent digestibility has been shown in numerous studies (Engels et al., 2009; Geerkens et al., 2013; Geerkens, Matejka, Carle, & rez, Borroto, & Saura-Calixto, Schweiggert, 2015; Larrauri, Rupe 1996; Schieber et al., 2001), thus making mango peels a promising target for commercial valorization (Nagel, Neidhart, et al., , Ralet, Bonnin, & Thibault, 2009). However, after 2014; Panouille producing the primary mango product, e.g., juice or puree, the wet peels are highly perishable and prone to microbial spoilage and endogenic enzymatic degradation reactions. In particular,

242

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

endo- (EC 3.2.1.15) and exo-polygalacturonases (EC 3.2.1.67) as well as pectin methyl esterases (EC 3.1.1.11) represent a serious obstacle for pectin recovery, unless these deteriorative enzymes are inactivated (Ali, Chin, & Lazan, 2004). Although Sirisakulwat, Sruamsiri, Carle, and Neidhart (2010) reported only insignificant degradation of the mango peel pectin within 5 h after peeling, middle- and long-term storage of the peels is impeded due to enzymatic decay and microbial spoilage. Hence, post-processing treatments preserving the peels and retaining their valuable constituents were previously recommended (Sirisakulwat et al., 2008). In the EU, the galacturonic acid content of food grade pectins shall be not less than 65% according to EU regulation No 231/2012. While pectins obtained from citrus peel and apple pomace commonly meet this requirement, lower galacturonic acid levels are important quality defects of mango “pectin”. In contrast to food use, there are no regulations for feed use (Regulation 68/2013/EU). However, the application of modified technological procedures has previously been shown to afford galacturonic acid contents greater than 65% (Nagel, Neidhart, et al., 2014). The first objective of the present study was the stabilization of wet peels by blanching the integral fruits and subsequent drying of the manually obtained peels. By these means, inactivation of pectin-degrading enzymes and prevention of microbial decay should be achieved. Furthermore, the influence of peel particle size reduction for pectin extraction on pectin quality and yield should be analyzed. In addition to blanching, drying, and particle size reduction, the effect of gamma irradiation on mango peel pectin was analyzed, since India is obliged to irradiate fresh mango fruits destined for the US market in order to extend their shelf-life and improve their phytosanitary status (Alothman, Bhat, & Karim, 2009). Moreover, gamma irradiation may exert potential effects on chemical and physical product properties (Chung & Liu, 2009). Beyond investigating the influence of technological procedures, pectin quality and yield from peels of three monoembryonic and a polyembryonic cultivar was compared. Peels from two of these cultivars were obtained in unripe and ripe condition in order to elucidate the effect of fruit ripeness on pectin quality and yield. The chemical and techno-functional properties of all pectins obtained were characterized in detail. Carbohydrate composition including contents of galacturonic and glucuronic acid and seven neutral sugars was analyzed. Furthermore, the degree of esterification (DE), molecular weight distribution, and gelling properties were analyzed. Besides pectin, the total dietary fiber content of the peels was examined, since mango peels were previously shown to contain valuable amounts of total dietary fiber (Ajila & Prasada Rao, 2013; Nagel, Neidhart, et al., 2014). 2. Materials and methods 2.1. Raw material and chemicals Mimicking industrial pectin extraction by processing fruits of different ripeness degrees, mangoes (M. indica L.) of the cultivars Tommy Atkins, Kent, Palmer, and Nam Dokmai were purchased from a local market in Stuttgart, Germany having different maturity, and stored at 13  C until processing. Peels of cv. Kaew were obtained from Chiang Mai (Thailand), while peels of cv. Totapuri were from Uttar Pradesh (India) and stored in sealed vacuum bags at the University of Hohenheim until used. The neutral sugar reference standards L-(þ)-arabinose, L()-fucose, D-(þ)-glucose, D-(þ)-galactose, D-(þ)-mannose, L(þ)-rhamnose, and D-(þ)-xylose, and the uronic acids D-(þ)-galacturonic acid and D-(þ)-glucuronic acid were purchased from SigmaeAldrich (Steinheim, Germany). The dietary fiber test kit

Bioquant® was from Merck (Darmstadt, Germany). 2-(N-morpholino)ethanesulfonic acid (MES) was from VWR International (Darmstadt, Germany), and 2-amino-2-hydroxymethyl-propane1,3-diol (TRIS) was obtained from Pharmacia Biotech AG (Uppsala, Sweden). Sodium hydroxide solution (50%, w/w) was from J.T. Baker (Avantor Performance Materials, Griesheim, Germany). All other reagents or solvents (analytical or HPLC grade) were purchased from SigmaeAldrich (Steinheim, Germany) and VWR International (Darmstadt, Germany). Ultrapure water was used throughout. Anhydrous methanolic 2 M hydrochloric acid (HCl) was prepared as described previously (Nagel, Sirisakulwat, Carle, & Neidhart, 2014). 2.2. Mango processing The ripening index (RPI) of mango fruits was analyzed according squez-Caicedo, Heller, Neidhart, & Carle (2006). After manual to Va peeling, convective oven drying of peels was carried out for 8 h at 60  C using a UT 6120 drying cabinet (Hanau, Germany) with and without previous steam blanching of integral fruits for 3 min at 100  C. Lyophilization of mango peels for 90 h was performed after grinding with liquid nitrogen with and without the above mentioned steam blanching prior to peeling using a Lyovac GT 4 (Oerlikon Leybold Vacuum, Cologne, Germany). Gamma irradiation of mango peels of cv. Totapuri was accomplished with 60Co for 3 h, reaching a final dose of 10 kGy as detailed recently (Geerkens, Matejka, et al., 2015). The dried peels of all cultivars were milled and sieved with a ZM 1 grinder (Retsch, Haan, Germany) equipped with a 0.25 mm ring sieve. Regarding peels from cv. Kaew, greater particle sizes were obtained by sieving the milled peels with a 0.5 mm ring sieve and manual cutting (10 mm) of the integral peels with a stainless steel knife, respectively. The Sauter mean diameter (d43) of dried mango peel powders was analyzed using a Mastersizer 2000 (Malvern Instruments, Worcestershire, UK). 2.3. Pectin extraction Hot-acid pectin extraction was performed with 20 g dried and ground or cut peels and 380 g water under continuous stirring at boiling temperature. The slurry obtained was cooled to room temperature in an ice bath, adjusted to pH 1.5 with aqueous sulfuric acid (2 N), and heated at 90  C for 2.5 h followed by recooling to room temperature. The solution was filtered and pressed manually using a nylon cloth. The solid residue retained by the cloth was washed with 200 mL water and pressed. The combined filtrates were added to 3 L of 2-propanol to precipitate alcohol insoluble solids (AIS). The AIS were filtered and pressed using the nylon cloth, washed with 2-propanol, and pressed again. Finally, the AIS were dried at 60  C for 14 h using a convective oven dryer, milled, and sieved with a ZM 1 grinder (Retsch, Haan, Germany) equipped with a 0.25 mm ring sieve. Pectin extraction was conducted in duplicate for each cultivar and ripeness degree, respectively, according to Geerkens, MillerRostek, et al. (2015). 2.4. Quantitation and characterization of dietary fiber For the determination of total (TDF), soluble (SDF), and insoluble dietary fiber (IDF), the enzymatic-gravimetric AOAC Official Methods 985.29 and 991.43 (1997) were conducted in duplicate using the MES/TRIS buffer solution. The swelling (SC), water holding (WHC), and oil holding capacities (OHC) were analyzed in duplicate according to Nagel, Neidhart, et al. (2014).

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

243

out for the determination of significantly different means (P < 0.05).

2.5. Pectin characterization AIS hydrolysis and carbohydrate analysis of the hydrolyzed AIS was carried out by HPAEC-PAD according to Nagel, Sirisakulwat, et al. (2014) in quadruplicate. The degrees of methylation (DMe) and acetylation (DAc) were simultaneously determined as detailed by Sirisakulwat et al. (2008). The starch content of the AIS was analyzed according to the instructions of the enzymaticphotometric starch determination kit (Art. No. 10207748035) by RBiopharm (Darmstadt, Germany). The molecular size distribution of the AIS was analyzed by high-performance size exclusion chromatography (HPSEC) as described previously (Sirisakulwat et al., 2008). 2.6. Characterization of technofunctional pectin properties The breaking capacity of pectin-sucrose gels with total soluble solid contents of 65  Brix at pH 3.0 ± 0.5 was determined using a Herbstreith pectinometer Mark III for two different pectin concentrations (0.25% and 0.30%) according to a previously described method (Neidhart, Hannak, & Gierschner, 2003). Additionally, the gelling units (GU) were determined for evaluating the commercial usability according to Schilling et al. (2008). CIE-L*a*b* color values of the gels (0.30%) were examined by a CR-300 colorimeter (Konica Minolta Sensing, Osaka, Japan), measuring 10 points across the gels' surfaces (gel height approx. 3 cm). A white standard was used for calibration (Konica Minolta Sensing). Breaking capacity, GU, and color values were analyzed from three independently produced gels. 2.7. Statistics Statistical analyses were conducted using SPSS 20 (IBM, Armonk, NY, USA). Analysis of variance (ANOVA) was conducted, and Tukey's HSD (Honestly Significant Difference) test was carried

3. Results and discussion 3.1. Yield of mango peels and stones from different cultivars The polyembryonic fruits (cv. Nam Dokmai) were slightly smaller (284 ± 16 g) than the monoembryonic ones from cvs. Palmer, Kent, and Tommy Atkins (415e581 g). The higher surfaceto-volume ratio of cv. Nam Dokmai significantly resulted in the highest relative yield in peels (13.6 ± 0.4% of total fruit weight), while yields from the above mentioned monoembryonic fruits were slightly inferior (9.1e11.9%). Stone weight accounted for ca. 8e10% of the total fruit weight. The total percentage of by-products (peels and stones) ranged from 17.9 to 23.3% (Table 1). 3.2. Dietary fiber characterization 3.2.1. Influence of cultivar Total dietary fiber (TDF) contents of mango peels from different cultivars were in the range of 29e42 g/100 g DM. The polyembryonic cv. Nam Dokmai contained the highest TDF (42.4 g/100 g DM) content as compared to the monoembryonic cultivars (29e36 g/100 g DM). Comparable results were previously reported by Nagel, Neidhart, et al. (2014), describing TDF values of 27e51 g/ 100 g DM. However, higher TDF contents were reported for lemon (60e68 g/100 g DM) and apple pomace (61e90 g/100 g DM), being the most usual starting materials for pectin extraction (Figuerola, vez, Chiffelle, & Asenjo, 2005). Nevertheless, due to Hurtado, Este their TDF contents, mango peels were previously shown to represent a valuable source of ruminant feed (Geerkens et al., 2013). As shown in Table 2, TDF was subdivided into soluble (SDF) and insoluble dietary fiber (IDF), ranging from 16 to 22 g/100 g DM and

Table 1 Fruit weight and accruing by-products, i.e. peels and stones of different mango cultivars. Cultivar

RPI []

Fruit weight [g]

Peel

Stone

Weight [g] Tommy AtkinsM KentM PalmerM Nam DokmaiP

8.8 7.3 5.4 9.0

415 518 581 284

± ± ± ±

6b 21c 39d 16a

41 62 53 39

± ± ± ±

1a 4b 6b 2a

Mass proportion [% w/w] 9.8 11.9 9.1 13.6

± ± ± ±

0.2a 0.2b 0.6a 0.4c

DM [%] 23 28 29 26

± ± ± ±

2 2 1 1

Total by-products

Weight [g] 34 43 55 28

± ± ± ±

6ab 5b 5c 2a

Mass proportion [% w/w] 8.1 8.3 9.5 9.8

± ± ± ±

1.5a 1.3a 1.4a 0.5a

Mass proportion [% w/w] 17.9 20.2 18.6 23.3

± ± ± ±

1.3a 1.1a 1.2a 0.4b

M Monoembryonic, Ppolyembryonic, RPI ripening index. Statistically significant differences (P < 0.05) between mango cultivars are marked with lower case letters.

Table 2 Dietary fiber and physical properties of mango peels of different cultivars and influenced by processing. Cultivar/treatment

RPI []

Dietary fiber Total

Soluble

WHC [g/g DM] Insoluble

Ratio []

OHC [g/g DM]

SC [mL/g DM]

[g/100 g DM] Influence of cultivar Tommy Atkins Kent Palmer Nam Dokmai

8.8 7.3 5.4 9.0

36.3 36.0 28.7 42.4

19.7 18.4 15.9 21.7

16.6 17.7 12.8 20.7

1:0.84 1:0.96 1:0.80 1:0.96

4.2 4.5 3.8 2.7

± ± ± ±

0.1bc 0.0d 0.1b 0.2a

2.8 3.0 2.2 2.8

± ± ± ±

0.1b 0.2b 0.0a 0.1b

9.6 7.1 10.3 6.4

± ± ± ±

0.1c 0.0b 0.0d 0.0a

Influence of mango processing Palmer 5.4 Oven drying Lyophilization Blanching þ oven drying Blanching þ lyophilization

28.7 29.8 31.2 31.1

15.9 15.6 15.0 14.9

12.8 14.2 16.2 16.2

1:0.80 1:0.91 1:1.08 1:1.09

3.8 3.9 5.2 3.4

± ± ± ±

0.1A 0.1A 0.9B 0.0A

2.2 2.5 2.1 2.9

± ± ± ±

0.0AB 0.1B 0.0A 0.0C

10.3 7.8 8.6 10.2

± ± ± ±

0.0B 0.3A 0.3A 0.3B

RPI ripening index, WHC water holding capacity, OHC oil holding capacity, SC swelling capacity. Statistically significant differences (P < 0.05) between cultivars are marked with lower case letters and between treatments with upper case letters.

244

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

method, blanching of integral fruits slightly reduced the SDF and simultaneously increased IDF contents, ultimately leading to greater TDF contents (Table 2). Blanching reduced the proportion of pulp adherent to the peel after peeling and, obviously, resulted in leaching of soluble fiber, low-molecular weight carbohydrates, and other water-soluble pulp constituents. Consequently, decreased SDF as well as increased IDF and TDF contents were observed after blanching. Hence, SDF:IDF ratio of 1:<1 for the drying processes without previous blanching was changed to 1:>1 for the drying processes with previous blanching. In agreement, Jagtiani, Chan, & Sakai (1988) described a facilitated debonding of pulp from the peel after blanching, thus increasing pulp yields during industrial depulping of mangoes.

13e21 g/100 g DM, respectively. SDF:IDF ratios varied from 1:0.80 to 1:0.96. Since a balanced SDF:IDF ratio of ca. 1:1 has been recommended for human diet (Larrauri, 1999), mango peels might represent a valuable source of dietary fiber in food supplements. TDF contents decreased with progressing fruit ripeness (Table 2). In agreement, degradation of dietary fiber during ripening was shown to be caused by endogenous enzymes, ultimately contributing to fruit softening (El-Zoghbi, 1994). At the same time, the dietary fiber-degrading activity of polygalacturonase, b-galactosidase, and (1 / 4)-b-glucanase as well as the pectolytic activity of pectinesterase and polygalacturonase increased with advanced maturity (Aina & Oladunjoye, 1993; Ali et al., 2004; Labib, El-Ashwah, Omran, & Askar, 1995; Mitcham & McDonald, 1992). The water holding (WHC) and oil holding capacities (OHC) of dried and ground mango peels from different cultivars were in the range of 2.7e4.5 g/g DM and 2.2e3.0 g/g DM, respectively (Table 2). Larrauri et al. (1996) observed a similar OHC and a significantly greater WHC for peels of cv. Hayden (11.4 g/g). Koubala, Kansci, Garnier, Thibault, & Ralet (2013) reported a WHC of 6.1 g/g DM for peels from cv. Mango. Applying identical analytical procedures, Nagel, Neidhart, et al. (2014) reported a similar OHC (2.1e2.7 g/g), while WHC was increased to 4.2e7.1 g/g by technological processing steps, such as blanching and subsequent pressing of peels. Swelling capacity (SC) was superior for peels from ripe fruits of cv. Palmer (10.3 mL/g DM) and inferior for those from unripe fruits of cv. Nam Dokmai (6.4 mL/g DM, Table 2). A similar SC (9.1 mL/g) was reported by Nagel, Neidhart, et al. (2014). Maximum SC of 20.7 mL/g was previously reported by Koubala et al. (2013), who additionally washed the peels with ethanol studying a peel particle size of 0.5 mm. Ethanol washing notably reduced the content of sucrose, lipids, pigments, and phenolic compounds in the peels and, thus may be responsible for the increased SC. For comparison, WHCs of 2.7, 5.2, and 1.6e1.9 g/g and SCs of 7.8, 13.8, and 6.6e8.3 mL/g were reported for Ambarella (Spondias cytherea Sonn.) peels, lime (Citrus latifolia Tanaka) peels, and apple pomace, respectively (Figuerola et al., 2005; Koubala et al., 2013), being in a similar range like the WHCs and SCs of mango peel powders. Thus, its properties turn mango peel into a potential dietary fiber supplement allowing to enrich the WHC of foods. For instance, mango peels were previously supplemented to wheat flour for dough biscuits production, yielding biscuits with increased TDF and WHC (Ajila, Leelavathi, & Prasada Rao, 2008).

3.3. Pectin characterization 3.3.1. Influence of cultivar and ripeness In order to assess the influence of the cultivar and ripeness on pectin quality, only the results of oven dried peels will be discussed (Table 3). The degree of methylation (DMe) of the pectin from monoembryonic cultivars was in the range of 56e62%, while the DMe of the polyembryonic cvs. ranged from 53 to 79%. Nam Dokmai provided the AIS with the greatest DMe of 79%. In agreement with our findings of consistently high esterified pectins (DMe > 50%) from mango peels, DMe ranging from 70 to 81% for several polyembryonic cultivars and from 56 to 66% for monoembryonic cultivars have been reported previously (Berardini, Fezer, et al., 2005; Nagel et al., 2015; Sirisakulwat et al., 2010). In good agreement with previous results (Berardini, Fezer, et al., ne mou, & Kratchanov, 1991; Sirisakulwat 2005; Kratchanova, Be et al., 2008; Srirangarajan & Shrikhande, 1979), the AIS yields obtained from mango peels varied between 19 and 27 g/100 g DM (Table 3). When comparing AIS yields from different cultivars, their highly variable starch contents (1.1e27.7% w/w DM) were deceiving when evaluating their potential for pectin production. For instance, high AIS yields were determined for both peels from ripe cv. Tommy Atkins (25.2%) and peels from unripe cvs. Palmer (25.4%) and Nam Dokmai (26.4%). However, after correcting the AIS values for their starch levels, only the AIS values of the cvs. Tommy Atkins (24.6%) and Nam Dokmai (24.3%) remained high, while the starchcorrected AIS contents of unripe Palmer peels were found to be the lowest of all peels investigated (18.3%). As starch degradation occurs during ripening, the galacturonic acid content of cv. Nam Dokmai increased from ca. 36 to 48 g/100 g AIS with progressing ripeness, mostly at the expense of glucose in

3.2.2. Influence of drying method and blanching As shown in Table 2, similar dietary fiber yields were achieved by oven drying and lyophilization. Irrespective of the drying

Table 3 Alcohol insoluble solids, starch, ash, and moisture content and degree of esterification of pectin produced from peels of different cultivars including the influence of peel particle size. Cultivar/peel particle size

RPI []

Visual & haptic inspection

Particle size d43 [mm]

Tommy Atkins Kent Palmer

8.8 7.3 10.0 5.4 9.0 6.3

Ripe Ripe Unripe Overripe Unripe Ripe

33 54 36 44 59 47

Nam Dokmai

Kaew 0.25 mm sieve 0.5 mm sieve 10 mm

± ± ± ± ± ±

1a 2 cd 2ab 2bc 6d 4c

AIS [g/100 g DM] 25.2 21.1 25.4 19.2 26.4 21.3

± ± ± ± ± ±

0.2bc 0.2a 1.2c 0.1a 0.2c 0.2ab

AISStarch-corr. [g/100 g DM] 24.6 20.5 18.3 18.9 24.3 20.5

± ± ± ± ± ±

0.8b 0.8a 1.2a 0.5a 1.3b 0.3a

Starch [g/100 g AIS] 2.3 2.7 27.7 1.1 8.1 3.7

± ± ± ± ± ±

0.1b 0.0bc 0.2e 0.0a 0.6d 0.0c

Ash [g/100 g AIS]

Moisture [g/100 g AIS]

6.4 ± 0.0b 5.8 ± 0.1b n.a. 12.5 ± 0.5c 3.4 ± 0.1a n.a.

5.5 ± 5.7 ± n.a. 4.9 ± 6.7 ± n.a.

6.2 ± 0.3 6.1 ± 0.0 5.8 ± 0.1

3.5 ± 0.3A 6.0 ± 0.1B 6.0 ± 0.0B

0.1ab 0.1ab 0.8a 0.0b

DMe [%] 56.3 61.7 n.a. 58.0 79.1 n.a.

± 1.1a ± 1.4b ± 2.6a ± 1.1c

DAc [%] 2.5 ± 2.9 ± n.a. 1.1 ± 3.9 ± n.a.

0.4b 0.4c 0.3a 0.5d

n.a. 42 ± 3A 117 ± 1B 10 mmCa

26.6 ± 0.2C 20.7 ± 0.5B 15.6 ± 0.2A

24.1 ± 0.0C 18.8 ± 0.6B 14.3 ± 0.1A

8.9 ± 0.4 9.2 ± 0.2 9.4 ± 0.2

66.4 ± 1.6C 53.3 ± 2.9A 59.8 ± 0.9B

RPI ripening index, AIS alcohol insoluble solids, DMe degree of methylation, DAc degree of acetylation. n.a. not analyzed. Statistically significant differences (P < 0.05) between cultivars are marked with lower case letters and between peel particle sizes with upper case letters. a Particle size as obtained after manual cutting.

3.6 ± 0.2B 3.0 ± 0.0A 3.8 ± 0.0C

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

the AIS (Table 4). The same relationship was observed for cv. Palmer, showing 41 g galacturonic acid in 100 g AIS from unripe and 54 g/100 g AIS from ripe fruits. When calculating on an ash-free and dry matter basis, the galacturonic acid content for cv. Palmer increased to 64.9 g/100 g AIS. These findings are of major importance for pectin production, since a galacturonic acid level of 65 g/ 100 g AIS on an ash-free and dry matter basis (JECFA, 2007) is a prerequisite for acceptable gelling properties and, most importantly, to meet legal pectin specifications. It has to be noted that the high galacturonic acid content in AIS from cv. Palmer was only observed for overripe fruits of this cultivar. Generally, the ashcorrected galacturonic acid contents of other cultivars and ripeness degrees ranged from 34.2 to 54.6 g/100 g AIS. Nevertheless, a careful selection of cultivars and the use of peels from ripe fruits are of utmost importance for improving the galacturonic acid content in the obtained AIS. However, the use of peels from ripe fruits might be hampered, if ripening-induced endogenic pectinases degraded the polymeric pectin chains. Therefore, the molecular weight distribution of the AIS was analyzed (Fig. 1A1 and A2). The high-molecular pectin fraction (integration zone I) of oven dried mango peels of different cultivars accounted for 48% (cv. Nam Dokmai) to 64% (cv. Palmer) of the total peak area. The shape and peak maximum of the pectin peak (integration zone I) presumably depended on both the cultivar and ripeness, thus affecting its molecular weight (Table 4). As illustrated by Fig. 1A1 and Table 4, the peak molecular weight MP of

245

the high-molecular weight fraction of unripe cv. Nam Dokmai (MP 735k) was considerably greater than that of ripe cv. Palmer (MP 235k). Accordingly, cv. Palmer has previously been described as a fruit containing soluble polysaccharides of a lower molecular , Lozano, & Brillouet, 1996). weight than other cultivars (Olle Furthermore, the average molecular weight of the high-molecular weight fraction was reduced in the AIS of more ripened cultivars (1049ke1841k) compared to the AIS of the unripe cv. Nam Dokmai (1904k), indicating cultivar specific molecular weight differences and/or enzymatic pectin degradation during fruit ripening. Pectin degradation in mango fruits was previously ascribed to the depolymerization of the homogalacturonan backbone by greater polygalacturonase activity at advanced ripeness (Ali et al., 2004; ElZoghbi, 1994). Besides starch and low-molecular pectin degradation products, an arabinogalactan has previously been shown to represent an often overlooked impurity of mango pectin, reducing the galacturonic acid content (Nagel et al., 2015). The characteristic arabinogalactan peak (integration zone II) had a molecular weight MP of 14e19k at the peak maximum (Table 4), previously reported to be composed of a molar ratio of 82/10/4/4 of its monomers galactose/arabinose/rhamnose/glucuronic acid (Nagel et al., 2015). The respective arabinogalactan peak was found in all mono- and polyembryonic cultivars analyzed. In contrast to the variable size distribution of the high-molecular weight fraction, the molecular weight distribution of the arabinogalactan (Fig. 1A1 and A2 and

Table 4 Neutral sugars, uronic acids, and molecular weight properties of mango pectin as influenced by the peel particle size, cultivar, and ripeness. Cultivar

Kaew

Tommy Atkins

Peel particle size

Ripening index

d43 42 mm

Kent

Palmer

Nam Dokmai

d43 117 mm

10 mm

8.8 Ripe

7.3 Ripe

10.0 Unripe

5.4 Overripe

9.0 Unripe

6.3 Ripe

[g/100 g AIS] Fuc Rha Ara Gal Glu Xyl Man

0.09 ± 0.00 1.33 ± 0.05 1.95 ± 0.08 15.53 ± 0.52 17.37 ± 0.77 0.91 ± 0.04 0.48 ± 0.01

0.04 ± 0.00 1.33 ± 0.07 0.99 ± 0.03 19.85 ± 4.40 14.63 ± 4.81 0.39 ± 0.04 0.23 ± 0.01

0.06 ± 0.00 1.17 ± 0.04 1.26 ± 0.01 18.23 ± 0.45 18.34 ± 0.23 0.66 ± 0.04 0.31 ± 0.01

0.09 ± 0.01 1.13 ± 0.01 1.63 ± 0.00 14.90 ± 0.23 3.95 ± 0.23 0.83 ± 0.04 0.28 ± 0.00

0.08 ± 0.01 1.64 ± 0.06 1.21 ± 0.02 14.08 ± 0.72 4.66 ± 0.11 0.76 ± 0.06 0.37 ± 0.02

0.01 ± 0.00 0.87 ± 0.02 0.38 ± 0.03 9.00 ± 0.20 25.39 ± 2.47 0.30 ± 0.10 0.00 ± 0.00

0.05 ± 0.00 0.81 ± 0.05 1.29 ± 0.06 14.14 ± 0.12 2.40 ± 0.07 0.22 ± 0.02 0.17 ± 0.01

0.06 ± 0.00 1.25 ± 0.02 1.27 ± 0.02 21.54 ± 0.47 8.56 ± 0.18 0.63 ± 0.04 0.36 ± 0.01

0.02 ± 0.00 1.29 ± 0.02 0.64 ± 0.00 22.21 ± 0.62 4.69 ± 0.06 0.24 ± 0.02 0.00 ± 0.00

ANS

37.65 ± 1.15

37.48 ± 2.27

40.03 ± 0.39

22.81 ± 0.32

22.79 ± 0.90

35.96 ± 2.56

19.09 ± 0.30

33.66 ± 0.69

29.08 ± 0.65

GalUA GlcUA UA

36.25 ± 1.27 0.89 ± 0.03 37.14 ± 1.30

38.14 ± 2.57 2.28 ± 0.08 40.43 ± 2.57

30.20 ± 1.52 2.76 ± 0.09 32.96 ± 1.59

48.06 ± 1.67 2.33 ± 0.03 50.39 ± 1.70

44.74 ± 0.81 1.43 ± 0.05 46.17 ± 0.86

41.37 ± 2.34 0.46 ± 0.02 41.83 ± 2.36

53.58 ± 3.38 0.77 ± 0.04 54.35 ± 3.41

35.84 ± 0.39 2.35 ± 0.04 38.19 ± 0.40

47.86 ± 0.96 1.18 ± 0.04 49.04 ± 0.99

ANS þ UA

74.79 ± 2.45

77.91 ± 4.84

72.99 ± 1.98

73.20 ± 2.02

68.96 ± 1.76

77.78 ± 3.58

73.44 ± 3.71

71.85 ± 1.03

78.11 ± 1.63

GalUAa

40.14

43.39

34.24

54.55

50.55

n.a.

64.87

39.87

n.a.

Molecular weight [1000] High-molecular polymers fraction [1] Mn 312 259 MW 2167 1435 MP 747 475 D 6.94 5.54

268 1679 490 6.26

292 1629 551 5.58

337 1841 618 5.46

n.a.

166 1049 235 6.33

351 1904 735 5.40

n.a.

Middle-molecular polymers fraction [2] Mn 14 13 MW 21 19 MP 15 14 D 1.43 1.48

14 20 15 1.44

14 20 17 1.37

16 22 19 1.32

14 18 17 1.27

14 20 16 1.48

Fuc fucose, Rha rhamnose, Ara arabinose, Gal galactose, Glu glucose, Xyl xylose, Man mannose, ANS anhydro neutral sugars, GalUA galacturonic acid, GlcUA glucuronic acid, UA uronic acids, n.a. not available, Mn number-average molecular weight, MW weight-average molecular weight, MP peak molecular weight, D polydispersity. For cvs. Tommy Atkins, Kent, Palmer, and Nam Dokmai, a peel particle size of d43 < 60 mm (cf. Table 3) and oven drying was used. a Estimated on ash-free and dry matter basis.

246

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

Fig. 1. Molecular size distribution and relative peak area of mango pectin. A1 and A2: Influence of the mango cultivars Palmer (RPI 5.4, ripe), Kent (RPI 7.3, ripe), Tommy Atkins (RPI 8.8, ripe), Nam Dokmai (RPI 9.0, unripe), and an apple pectin. B1 and B2: Influence of peel particle size, C1 and C2: Influence of mango processing (cv. Nam Dokmai). Molecular weight of integration zone I  51k, II  4k, III  0.9k, IV  0.2k. B blanching, L lyophilization, OD oven drying.

Table 4) was more uniform among all samples analyzed, indicating its conserved structure being independent of cultivar and ripeness degrees. Arabinogalactans of mango fruits mainly occur in the sap of peel resin ducts (Nagel et al., 2015), where their content may considerably depend on the different desapping procedures optionally conducted after harvest. A detailed description of desapping procedures was compiled by Johnson & Hofman (2009). Since the arabinogalactan-rich sap content per fruit was reported to range from 0.09 to 1.63 g/fruit depending on cultivar and desapping procedures (Hassan, Irving, Dann, Coates, & Hofman, 2009; Johnson & Hofman, 2009), the contents of galactose, arabinose, and rhamnose in the produced pectins is consequently highly variable. When comparing the cultivars used in our study, the arabinogalactan peak of cv. Nam Dokmai was the most pronounced (integration zone II, Fig. 1A1 and A2). As expected, the greatest proportions of galactose, arabinose, and rhamnose of up to 24 g/ 100 g AIS were determined for cv. Nam Dokmai (Table 4). Concomitantly high amounts of glucuronic acid, being a constituent of the arabinogalactan (Nagel et al., 2015), were also determined (Table 4). Consequently, the high amounts of the arabinogalactan impurity contributed to the lowest galacturonic acid content (36 g/ 100 g AIS, Table 4) in the AIS from cv. Nam Dokmai. In brief summary, mango pectin analyses by HPSEC allowed rough prediction of its qualitative composition. A higher share of

integration zone I (high-molecular weight fraction) implied higher galacturonic acid contents, while a lower share of integration zone II (middle-molecular weight arabinogalactan fraction) indicated both lower glucuronic acid and galactose contents and vice versa. As outlined below, the arabinogalactan content may be drastically affected by peel processing. In contrast, the molecular weight of the arabinogalactan (MW ca. 18e22k, cf. integration zone II) was similar for all cultivars and ripeness degrees, and was independent of processing. In Fig. 1A1 and A2, peak areas in integration zone IV (MW  650) showed a high correlation with the ash contents (3e13 g/100 g AIS, Table 3) determined. While the AIS from unripe fruits of cv. Nam Dokmai had the lowest ash content (3 g/100 g AIS), that from overripe cv. Palmer was more than 3-fold higher (13 g/100 g AIS). Accordingly, the peak in integration zone IV of cv. Nam Dokmai was the smallest, whereas the peak of cv. Palmer was the highest (Fig. 1A1 and A2). When compared to peels of unripe fruits, maximum ash contents have been previously reported in peels of ripened fruits (Sirisakulwat et al., 2008). In brief summary, arabinogalactan, starch, and ash are recognized as most important interfering substances affecting pectin quality and, thus, should be removed. Recently, the importance of specified peel processing was demonstrated by a process achieving galacturonic acid contents of up to 83% on an ash-free and dried basis (Nagel, Neidhart, et al., 2014).

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

3.3.2. Influence of peel particle size Due to its potential effect on pectin yield, the influence of milling and sieving the peels to obtain varying peel particle sizes prior to pectin recovery was investigated. The particle size was shown to significantly (P < 0.001) affect AIS yields of cv. Kaew. The AIS yield increased remarkably by 71% from 16 to 27 g/100 g DM (starchcorrected AIS yield increased from 14 to 24 g/100 g DM) after peel particle size reduction from 10 mm to 42 mm (d43, Table 3). Unexpectedly, the enhanced AIS yield was associated with increased galacturonic acid contents (þ20%) from 30 to 36 g/100 g AIS (Table 4). On an ash-free and dry matter basis, the galacturonic acid contents increased from 34 to 40 g/100 g AIS. The smaller particle size possibly facilitated the release of protopectin from the cell wall as previously hypothesized for the enhanced pectin yield from more finely ground apple pomace (Canteri-Schemin, Ramos Fertonani, Waszczynskyj, & Wosiacki, 2005). By analogy to the enhanced AIS extraction, the absolute starch contents increased due to an improved accessibility of starch from smaller peel particles, although differences in relative starch contents were insignificant (Table 3). Beyond higher AIS yields and constant relative starch levels, the particle size also had a substantial impact on the extent of the above mentioned arabinogalactan impurity. According to Nagel et al. (2015), the ratio of pectin to arabinogalactan can be estimated based on the ratio of galacturonic acid to glucuronic acid. The molar ratio successively increased with a smaller particle size from 10.9 (10 mm) to 16.7 (d43 117 mm), and ultimately to 40.7 mol mol1 (d43 42 mm). Accordingly, a smaller peel particle size resulted in lower relative arabinogalactan contents and, thus, improved pectin purity. As highlighted by Fig. 1B1 and B2, the smaller peel particle size improved extractability of the targeted high-molecular pectin fraction (integration zone I) rising its proportion from 46% (10 mm) to 55% (d43 42 mm) of the total peak area. At the same time, the relative amount of the undesired arabinogalactan fraction (integration zone II) was diminished from 49% (10 mm) to 39% (d43 42 mm). In contrast to the cell wall-associated pectin, the absolute extractability of non-cell wall constituents, such as the arabinogalactan, may not be strongly enhanced by reducing the peel

247

particle size. In good agreement with this hypothesis, higher relative galacturonic acid contents (Table 4) confirmed our findings of an increased proportion of the high-molecular weight pectin fraction (Fig. 1B2) in the AIS obtained from peels of smaller particle size. Moreover, the number-average Mn (312k), weight-average MW (2,167k), and peak molecular weight MP (747k) reached their maximum values for AIS isolated from peels having the smallest particle size. Since higher molecular weights were associated with enhanced pectin gel quality parameters, using a smaller peel particle size for AIS isolation significantly improved pectin quality in terms of the galacturonic acid content and molecular weight. Our results were in agreement with previous studies, analyzing the influence of the peel particle size on pectin quality parameters after the extraction from lemon peels (Lerotholi, Carsky, & IkhuOmoregbe, 2011) and watermelon rind (Rasheed, 2008). In addition, mango pectin recovery from finely ground pomace by pressing was shown to be feasible. Remarkably, while particle size reduction was shown to improve mango pectin extraction, this measure has been found to be detrimental to pectin extraction from apple pomace. During processing the latter, small particles rapidly lead to an impermeable press cake during pressing (Heiss, 2004). In contrast, a peel particle size reduction is highly recommended for the mango pectin extraction as outlined above. 3.3.3. Influence of drying method, blanching, and irradiation Irrespective of the cultivar, the extraction of lyophilized peel powders increased AIS yields when compared to oven dried samples (Table 5 and S1). Despite applying the same grinding procedure, the mean particle size of the lyophilized powders was found to be smaller (d43 23 mm) than that of oven dried peel powders (d43 59 mm). In agreement with Section 3.3.2, extraction yield was enhanced by smaller particle sizes. Enzymatic degradation apparently played a minor role, since blanching insignificantly affected AIS yields, independent of the drying method (Table 5). In agreement, Sudhakar & Maini (2000) reported AIS yields to be unaffected by blanching. By analogy to AIS yields, slightly higher starch levels were observed after pectin recovery from lyophilized powders compared to oven dried powders (Table S1). Particularly, starch

Table 5 Alcohol insoluble solids, starch, neutral sugars, and uronic acid contents of pectin of cv. Nam Dokmai (RPI 9.0) influenced by mango processing. cv. Nam Dokmai Oven drying Particle size d43 [mm] AIS [g/100 g DM] Starch [g/100 g AIS]

59 ± 6 26.4 ± 0.2 8.1 ± 0.6

Blanching þ oven drying 56 ± 1 25.4 ± 0.1 8.1 ± 0.2

Lyophilization 23 ± 1 30.1 ± 0.0 9.5 ± 0.5

Blanching þ lyophilization 26 ± 1 32.8 ± 0.1 9.5 ± 0.6

[g/100 g AIS] Fuc Rha Ara Gal Glu Xyl Man

0.06 1.25 1.27 21.54 8.56 0.63 0.36

ANS

33.66 ± 0.69

22.07 ± 0.88

36.47 ± 1.28

25.00 ± 0.66

GalUA GlcUA UA

35.84 ± 0.39 2.35 ± 0.04 38.19 ± 0.40

38.29 ± 1.80 0.97 ± 0.05 39.27 ± 1.83

28.27 ± 0.42 2.01 ± 0.02 30.28 ± 0.44

33.84 ± 1.22 1.04 ± 0.07 34.88 ± 1.27

ANS þ UA

71.85 ± 1.03

61.34 ± 2.66

66.74 ± 1.50

59.88 ± 1.75

± ± ± ± ± ± ±

0.00 0.02 0.02 0.47 0.18 0.04 0.01

0.09 1.04 1.28 9.91 8.47 0.81 0.48

± ± ± ± ± ± ±

0.01 0.07 0.07 0.37 0.30 0.05 0.03

0.07 1.09 1.24 17.51 15.38 0.76 0.42

± ± ± ± ± ± ±

0.01 0.04 0.02 0.58 0.65 0.05 0.02

0.14 1.00 1.34 10.12 10.17 1.45 0.78

± ± ± ± ± ± ±

0.01 0.02 0.03 0.25 0.31 0.08 0.05

AIS alcohol insoluble solids, Fuc fucose, Rha rhamnose, Ara arabinose, Gal galactose, Glu glucose, Xyl xylose, Man mannose, ANS anhydro neutral sugars, GalUA galacturonic acid, GlcUA glucuronic acid, UA uronic acids.

248

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

levels in AIS from peels of cv. Palmer revealed higher starch contents after lyophilization (6 g/100 g AIS) than oven dried samples (2 g/100 g AIS). The moisture content of AIS was not affected by the different processes (5e7 g/100 g AIS), being below the maximum legal requirement of 12% (JECFA, 2007). Ash contents were diminished by 5e32% after blanching, most likely due to leaching, e.g., of abundant and easily eluting potassium € et al., 2003). (Puupponen-Pimia Pectin purity, in terms of galacturonic acid content, was substantially influenced by peel processing. Considering economic aspects, oven drying may be favored over lyophilization. Galacturonic acid contents in the AIS of oven dried peels (36e54 g/100 g AIS) were higher than in lyophilized peels (28e44 g/100 g AIS, data not shown) of all cultivars used. Neutral sugar and uronic acid contents of blanched peels will be discussed for cv. Nam Dokmai (Table 5). The galacturonic acid content of AIS from blanched peels (34e38 g/100 g AIS) exceeded that from unblanched peels (28e36 g/100 g AIS). A boost of the peak area of the pectin (integration zone I) was accompanied by a reduction of the arabinogalactan peak area (integration zone II) when comparing the AIS from blanched (MP  877k) and unblanched (MP  775k) peels. The effect of blanching was independent of the applied drying method. Blanching lowered total neutral sugar contents by up to 35%, most likely due to leaching of water soluble arabinogalactan as indicated by the reduced peak area in integration zone II (Fig. 1C1 and C2). Confirming this assumption, reduced contents of the arabinogalactan constituents, i.e. galactose, rhamnose, and glucuronic acid were determined (Table 5), finally leading to purer pectin. As a consequence, the relative amount of hemicellulose constituents such as fucose, mannose, and xylose was increased at the same time. The significant reduction (P < 0.001) of glucuronic acid due to blanching was independent of the drying method. As mentioned above, the uronic acid ratio of galacturonic to glucuronic acid may be used for the determination of the ratio of pectin to arabinogalactans, and, thus, for assessing the pectin purity (Nagel et al., 2015). Processing without blanching led to an uronic acid ratio of 15e16, which was 33e40 after blanching, thus verifying the improved purity. In brief summary, blanching represents a crucial step for the recovery of premium quality mango pectin. After gamma irradiation of mango peels (cv. Totapuri), AIS yield (21.1e21.4 g/100 g DM), starch content (2.1e2.6 g/100 g AIS), and neutral sugar content (21.8e22.8 g/100 g AIS) remained unaffected. However, the galacturonic acid content of irradiated peels (50.5 g/ 100 g AIS) was increased by ~10% compared to unirradiated peels (46.0 g/100 g AIS). Similar findings have previously been reported (Dennison & Ahmed, 1967; Zhao, Moy, & Paull, 1996), studying the effect of gamma irradiation on pectin from the pulp of mango and papaya. Since irradiation may cause a delay in mango ripening

(Dennison & Ahmed, 1967), cell wall degrading activities might be delayed, possibly improving pectin stability. 3.4. Functional properties 3.4.1. Influence of cultivar The gelling properties of pectic hydrocolloids are chiefly influenced by their molecular weight, degree of esterification (DE ¼ DMe þ DAc), and galacturonic acid content (Endress, 2004; Harris & Smith, 2006). The highest molecular weight (MW 1904k), highest DE (83%), and lowest galacturonic acid content (36 g/100 AIS) (Tables 3 and 4) were obtained from peels of cv. Nam Dokmai (RPI 9.0). The resulting AIS revealed a breaking capacity of 0.31 g AIS/100 g gel (Table 6). In contrast, AIS from cv. Palmer (RPI 5.4) had the lowest molecular weight (MW 1049k), a considerably lower DE (59%), and the highest galacturonic acid content (54 g/100 g AIS), resulting in an inferior breaking capacity (0.40 g AIS/100 g gel) when compared to AIS of cv. Nam Dokmai. AIS of cv. Kent (RPI 7.3) produced the strongest mango gels with a breaking capacity of 0.28 g AIS/100 g gel. The gel made from the reference apple pectin (galacturonic acid content 76.9 g/100 g AIS, MW 1563k, DE 59.8%) was 1.8-fold stronger (0.16 g AIS/100 g gel) than that from cv. Kent. The substantially different breaking capacities of apple and mango pectin may be attributed to the non-purified mango pectins, still containing hindering components such as starch, arabinogalactan, and ash. In addition to breaking capacity, the gelling units (GU) allow the assessment of gelling properties and commercial applicability. Maximum GU values were obtained for gels from cv. Nam Dokmai (5476 GU), although possessing the lowest galacturonic acid content (36 g/100 g AIS). While pectin from cv. Palmer contained the highest galacturonic acid content (54 g/100 g AIS), GU were substantially lower (3094 GU) than that of cv. Nam Dokmai. Considering all samples investigated, galacturonic acid contents neither correlated with the GU values (R2linear ¼ 0.004) nor the breaking capacities (R2linear ¼ 0.081). When multiplying the galacturonic acid content with the molecular weight of the pectin fraction (MW) and the DE, two known factors influencing gelling properties, we were able to establish a correlation to GU (R2linear ¼ 0.816, R2polynomial ¼ 0.917) and breaking capacity (R2linear ¼ 0.602, R2polynomial ¼ 0.746), as depicted in Fig. 2. Clearly, the gelling properties of the mango pectins obtained depended on their galacturonic acid contents, molecular weights, and DEs. The GU value of mango pectin (3094e5476 GU) is remarkable when compared to pectin extracted from apple pomace (3635e3814 GU) as reported by Schilling et al. (2008), thus demonstrating the potential of mango pectin when produced from finely ground peels (d43 < 120 mm).

Table 6 Gelling and color properties of mango pectin of different cultivars and influenced by the peel particle size. Cultivar/peel particle size

BC530HPE [g AIS/100 g gel]

SBC530HPE [g/g AIS]

Gelling units [g/100 g DM]

L*

a*

Tommy Atkins (RPI 8.8) Kent (RPI 7.3) Palmer (RPI 5.4) Nam Dokmai (RPI 9.0)

0.36 0.28 0.40 0.31

181 229 161 207

4543 4817 3094 5476

45.9 45.4 50.5 44.6

Kaew d43 42 mm d43 117 mm 10 mm

0.32 0.37 0.52

205 177 126

5446 3667 1953

50.2 ± 4.5 56.5 ± 7.6 61.4 ± 8.2

1.8 ± 0.3 1.1 ± 0.5 0.9 ± 0.4

7.9 ± 1.7 5.2 ± 0.7 4.6 ± 0.8

8.1 ± 1.6 5.2 ± 0.7 4.6 ± 0.8

76.2 ± 4.5 77.5 ± 5.3 79.0 ± 3.5

Apple pectin (reference)

0.16

412

n.a.

52.9 ± 7.8

0.7 ± 0.4

5.6 ± 0.9

5.6 ± 0.9

83.1 ± 3.7

± ± ± ±

4.7 5.0 4.0 5.5

2.4 2.1 1.3 2.0

b* ± ± ± ±

0.4 0.4 0.4 0.4

10.8 5.8 4.1 5.7

C* ± ± ± ±

1.2 0.9 0.5 0.9

11.0 6.2 4.3 6.1

h ± ± ± ±

1.1 0.8 0.6 0.8

77.4 70.0 72.5 70.0

± ± ± ±

2.5 6.2 5.2 4.9

BC530HPE breaking capacity as AIS required for a gel of 530 Herbstreith-Pectinometer units (HPE), SBC530HPE sugar binding capacity of AIS in a gel of 530 HPE, gelling units (GU) of mango peel dry matter (MPDM) as its sugar binding capacity in gram of sugar bound per 100 g MPDM in a gel of 530 HPE, n.a. not analyzed, RPI ripening index.

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

249

4. Conclusions The economic production of pectin from mango peels appeared to be most promising when ripe fruits were used. In addition to the potential production of high-quality mango pulp, peels from ripe fruits were shown to provide AIS of higher galacturonic acid contents than those from unripe fruits. The implementation of a blanching step is recommended, since it allows the complete removal of pulp from the peels, thus increasing pulp yield and pectin purity. Particularly, blanching resulted in leaching of interfering arabinogalactans and minerals in the AIS, consequently increasing its galacturonic acid content. Since AIS yields and galacturonic acid contents were substantially improved (þ70% and þ20%, respectively) when finely ground peels (particle size 42 mm) were used for pectin extraction, we highly recommend to assess the feasibility of extracting such finely ground peels at pilot plant and industrial scale. By implementing peel blanching and particle size reduction, the galacturonic acid content of a sample of cv. Palmer was increased to 65%, reaching the minimum legal requirement of pectin for food use. Noteworthy, the galacturonic acid contents of identically obtained AIS from other cultivars ranged from 34 to 55%. Finally, up to 5476 gelling units were accomplished by utilization of mango peels for pectin recovery and gel production. The mentioned blanching and grinding steps may be investigated by further systematic studies to optimize their influence on pectin quality. Considering the technological progress presented both in our study and that previously reported by Nagel, Neidhart, et al. (2014), mango peels may become a highly promising source of pectin, representing an alternative to apple pectin, the latter being threatened by enzymatic liquefaction of apple mash (Kammerer, Kammerer, Valet, & Carle, 2014). Abbreviations

Fig. 2. Correlation of the product of the galacturonic acid content (GalUA), molecular weight (MW), and degree of esterification (DE) with gelling units (A) and breaking capacity (B), respectively. BC530HPE breaking capacity as AIS required for a gel of 530 Herbstreith-Pectinometer units (HPE), gelling units of mango peel dry matter (MPDM) as its sugar binding capacity in gram of sugar bound per 100 g MPDM in a gel of 530 HPE.

3.4.2. Influence of peel particle size As mentioned above, peel particle size reduction substantially improved pectin yield and quality. Furthermore, enhanced breaking and sugar binding capacities of the pectins were achieved by particle size reduction (Table 6). Additionally, GU values were remarkably increased from 1953 GU (10 mm) to 5446 GU (d43 42 mm) at the same time, even surpassing GU of apple pectin (3635e3814 GU, Schilling et al., 2008). Previously, the GU of mango pectin were increased from 1864 to 4131 GU by pressing and blanching of the peels (Nagel, Neidhart, et al., 2014). Therefore, a combination of these measures together with peel particle size reduction may further improve the techno-functionality of mango pectins. However, gels produced from peel powders of larger particle sizes had a whiter color (L* ¼ 61) than those made from small particles (L* ¼ 50, Table 6). The latter gels were also more brownish as reflected by higher redness (þa*) and yellowness (þb*) values. The color may have been affected by co-extracted and, subse€dler, Schieber, & quently, oxidized polyphenols (Berardini, Kno Carle, 2005), which were more efficiently extracted from smaller particles.

AIS alcohol insoluble solids cv. cultivar DAc degree of acetylation DE degree of esterification DM dry matter DMe degree of methylation GU gelling units HPAEC-PAD high-performance anion exchange chromatography with pulsed amperometric detection HPSEC high-performance size exclusion chromatography IDF insoluble dietary fiber MES 2-(N-morpholino)ethanesulfonic acid OHC oil holding capacity RPI ripening index SC swelling capacity SDF soluble dietary fiber TDF total dietary fiber TRIS 2-amino-2-hydroxymethyl-propane-1,3-diol WHC water holding capacity Acknowledgment We thank Bhabha Atomic Research Centre (Mumbai, India) for the irradiation of mango peels, and Dr. Sanjay Nene (National Chemical Laboratory, Pune, India) for the logistics. This project was supported by funds of the Federal Ministry of Food and Agriculture (2813803810) (BMEL) based on a decision of the Parliament of the Federal Republic of Germany via the Federal Office for Agriculture and Food (BLE) under the innovation support program.

250

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.foodhyd.2015.05.022.

References Aina, J. O., & Oladunjoye, O. O. (1993). Respiration, pectolytic activity and textural changes in ripening African mango (Irvingia gabonensis) fruits. Journal of the Science of Food and Agriculture, 63, 451e454. Ajila, C. M., Leelavathi, K., & Prasada Rao, U. J. S. (2008). Improvement of dietary fiber content and antioxidant properties in soft dough biscuits with the incorporation of mango peel powder. Journal of Cereal Science, 48, 319e326. Ajila, C. M., & Prasada Rao, U. J. S. (2013). Mango peel dietary fibre: composition and associated bound phenolics. Journal of Functional Foods, 5, 444e450. Ali, Z. M., Chin, L.-H., & Lazan, H. (2004). A comparative study on wall degrading enzymes, pectin modifications and softening during ripening of selected tropical fruits. Plant Science, 167, 317e327. Alothman, M., Bhat, R., & Karim, A. A. (2009). Effects of radiation processing on phytochemicals and antioxidants in plant produce. Trends in Food Science & Technology, 20, 201e212. AOAC Official Method 985.29. (1997). Total dietary fiber in food. Arlington, VA, USA. AOAC Official Method 991.43. (1997). Total, soluble, and insoluble dietary fiber in foods. Arlington, VA, USA. Berardini, N., Fezer, R., Conrad, J., Beifuss, U., Carle, R., & Schieber, A. (2005). Screening of mango (Mangifera indica L.) cultivars for their contents of flavonol O- and xanthone C-glycosides, anthocyanins, and pectin. Journal of Agricultural and Food Chemistry, 53, 1563e1570. €dler, M., Schieber, A., & Carle, R. (2005). Utilization of mango peels Berardini, N., Kno as a source of pectin and polyphenolics. Innovative Food Science & Emerging Technologies, 6, 442e452. Canteri-Schemin, M. H., Ramos Fertonani, H. C., Waszczynskyj, N., & Wosiacki, G. (2005). Extraction of pectin form apple pomace. Brazilian Archives of Biology and Technology, 48, 259e266. Chung, H.-J., & Liu, Q. (2009). Effect of gamma irradiation on molecular structure and physicochemical properties of corn starch. Journal of Food Science, 74, C353eC361. Dennison, R. A., & Ahmed, E. M. (1967). Irradiation effects on the ripening of Kent mangoes. Journal of Food Science, 32, 702e705. Directive 2000/36/EC. (23 June 2000). Cocoa and chocolate products intended for human consumption. Official Journal of the European Communities. European Parliament and the Council of the European Union, L 197/19eL 197/25. El-Zoghbi, M. (1994). Biochemical changes in some tropical fruits during ripening. Food Chemistry, 49, 33e37. Endress, H.-U. (2004). Pektine. In R. Heiss (Ed.), Lebensmitteltechnologie. Biotechnologische, chemische, mechanische und thermische Verfahren der Lebensmittelverarbeitung (6th ed.). (pp. 285e293). Berlin: Springer. €dler, M., Zhao, Y.-Y., Carle, R., Ga €nzle, M. G., & Schieber, A. (2009). Engels, C., Kno Antimicrobial activity of gallotannins isolated from mango (Mangifera indica L.) kernels. Journal of Agricultural and Food Chemistry, 57, 7712e7718. FAOSTAT Database. (2014). Production quantity of mangoes, mangosteens, and guavas. Retrieved from http://faostat3.fao.org/faostat-gateway/go/to/download/Q/QC/E. vez, A. M., Chiffelle, I., & Asenjo, F. (2005). Fibre Figuerola, F., Hurtado, M. L., Este concentrates from apple pomace and citrus peel as potential fibre sources for food enrichment. Food Chemistry, 91, 395e401. Geerkens, C. H., Matejka, A. E., Carle, R., & Schweiggert, R. M. (2015). Development and validation of an HPLC method for the determination of alk(en)ylresorcinols using rapid ultrasound-assisted extraction of mango peels and rye grains. Food Chemistry, 169, 261e269. Geerkens, C. H., Miller-Rostek, P., Matejka, A. E., Nene, S., Carle, R., & Schweiggert, R. M. (2015). Influence of cultivar, ripeness, blanching, drying, irradiation, and pectin recovery on alk(en)ylresorcinols in mango peels. European Food Research and Technology. Geerkens, C. H., Schweiggert, R. M., Steingass, H., Boguhn, J., Rodehutscord, M., & Carle, R. (2013). Influence of apple and citrus pectins, processed mango peels, a phenolic mango peel extract, and gallic acid as potential feed supplements on in vitro total gas production and rumen methanogenesis. Journal of Agricultural and Food Chemistry, 61, 5727e5737. Harris, P. J., & Smith, B. G. (2006). Plant cell walls and cell-wall polysaccharides: structures, properties and uses in food products. International Journal of Food Science and Technology, 41, 129e143. Hassan, M. K., Irving, D. E., Dann, E. K., Coates, L. M., & Hofman, P. J. (2009). Sap properties and alk(en)ylresorcinol concentrations in Australian-grown mangoes. Annals of Applied Biology, 154, 419e427. Heiss, R. (2004). In Lebensmitteltechnologie: Biotechnologische, chemische, mechanische und thermische Verfahren der Lebensmittelverarbeitung (6th ed.). Berlin: Springer. Jagtiani, J., Chan, H. T., & Sakai, W. S. (1988). Tropical fruit processing. Food science and technology. San Diego, Calif.: Academic Press.

Johnson, G. I., & Hofman, P. J. (2009). Postharvest technology and quarantine treatments. In R. E. Litz (Ed.), The mango. Botany, production and uses (2nd ed.). (pp. 529e605). Wallingford: CABI. Joint FAO/WHO Expert Committee on Food Additives. (2007). Compendium of food additive specifications. In 68th Meeting 2007; Joint FAO/WHO Expert Committee on Food Additives (Vol. 4). Rome: FAO. Kammerer, D. R., Kammerer, J., Valet, R., & Carle, R. (2014). Recovery of polyphenols from the by-products of plant food processing and application as valuable food ingredients. Food Research International, 65, 2e12. Konica Minolta Sensing. Exakte Farbkommunikation. Retrieved from http://www. lepen.de/assets/files/efk_konica.pdf. Koubala, B. B., Kansci, G., Garnier, C., Thibault, J.-F., & Ralet, M.-C. (2013). Physicochemical properties of dietary fibres prepared from ambarella (Spondias cytherea) and mango (Mangifera indica) peels. Food and Bioprocess Technology, 6, 591e597. ne mou, C., & Kratchanov, C. (1991). On the pectic substances of Kratchanova, M., Be mango fruits. Carbohydrate Polymers, 15, 271e282. Labib, A. A. S., El-Ashwah, F. A., Omran, H. T., & Askar, A. (1995). Heat-inactivation of mango pectinesterase and polygalacturonase. Food Chemistry, 53, 137e142. Larrauri, J. A. (1999). New approaches in the preparation of high dietary fibre powders from fruit by-products. Trends in Food Science & Technology, 10, 3e8. rez, P., Borroto, B., & Saura-Calixto, F. (1996). Mango peels as a Larrauri, J. A., Rupe new tropical fibre: preparation and characterization. LWT e Food Science and Technology, 29, 729e733. Lerotholi, L., Carsky, M., & Ikhu-Omoregbe, D. I. O. (2011). A study of the extraction of pectin from dried lemon peels. Advanced Materials Research, 367, 311e318. Mitcham, E. J., & McDonald, R. E. (1992). Cell wall modification during ripening of ‘Keitt’ and ‘Tommy Atkins’ mango fruit. Journal of the American Society for Horticultural Science, 117, 919e924. Nagel, A., Mix, K., Kuebler, S., Bogner, H., Kienzle, S., Elstner, P., et al. (2015). The arabinogalactan of dried mango exudate and its co-extraction during pectin recovery from mango peel. Food Hydrocolloids, 46, 134e143. Nagel, A., Neidhart, S., Anders, T., Elstner, P., Korhummel, S., Sulzer, T., et al. (2014). Improved processes for the conversion of mango peel into storable starting material for the recovery of functional co-products. Industrial Crops and Products, 61, 92e105. Nagel, A., Sirisakulwat, S., Carle, R., & Neidhart, S. (2014). An acetate-hydroxide gradient for the quantitation of the neutral sugar and uronic acid profile of pectins by HPAEC-PAD without postcolumn pH adjustment. Journal of Agricultural and Food Chemistry, 62, 2037e2048. Neidhart, S., Hannak, C., & Gierschner, K. (2003). Solegel transitions of highesterified pectins and their molecular structure. In F. Voragen, H. Schols, & R. Visser (Eds.), Advances in pectin and pectinase research (pp. 431e448). Dordrecht, Netherlands: Springer. Neidhart, S., Sirisakulwat, S., Nagel, A., Sruamsiri, P., & Carle, R. (2009). Which mango processing residues are suitable for pectin recovery in terms of yield, molecular and techno-functional properties of extractable pectins? In H. A. Schols, R. G. F. Visser, & A. G. J. Voragen (Eds.), Pectins and pectinases (pp. 177e195) Wageningen: Wageningen Academic Publishers. , D., Lozano, Y. F., & Brillouet, J.-M. (1996). Isolation and characterization of Olle soluble polysaccharides and insoluble cell wall material of the pulp from four mango (Mangifera indica L.) cultivars. Journal of Agricultural and Food Chemistry, 44, 2658e2662. , M., Ralet, M.-C., Bonnin, E., & Thibault, J.-F. (2009). Recovery and reuse of Panouille trimmings and pulps from fruit and vegetable processing. In K. Waldron (Ed.), Handbook of waste management and co-product recovery in food processing (pp. 417e447). Cambridge: Woodhead Pub. €, R., H€ Puupponen-Pimia akkinen, S. T., Aarni, M., Suortti, T., Lampi, A.-M., Eurola, M., et al. (2003). Blanching and long-term freezing affect various bioactive compounds of vegetables in different ways. Journal of the Science of Food and Agriculture, 83, 1389e1402. Rasheed, A. M. (2008). Effect of different acids, heating time and particle size on pectin extraction from watermelon rinds. Journal of Kerbala University, 6, 234e243. Regulation (EU) No 231/2012 laying down specifications for food additives listed in Annexes II and III to Regulation (EC) No 1333/2008 of the European Parliament and of the Council 1, European Commission 9 March 2012. Regulation (EU) No 68/2013 on the Catalogue of feed materials 1, European Commission 16 January 2013. Schieber, A., Stintzing, F. C., & Carle, R. (2001). By-products of plant food processing as a source of functional compoundsdrecent developments. Trends in Food Science & Technology, 12, 401e413. Schilling, S., Toepfl, S., Ludwig, M., Dietrich, H., Knorr, D., Neidhart, S., et al. (2008). Comparative study of juice production by pulsed electric field treatment and enzymatic maceration of apple mash. European Food Research and Technology, 226, 1389e1398. Sirisakulwat, S., Nagel, A., Sruamsiri, P., Carle, R., & Neidhart, S. (2008). Yield and quality of pectins extractable from the peels of Thai mango cultivars depending on fruit ripeness. Journal of Agricultural and Food Chemistry, 56, 10727e10738. Sirisakulwat, S., Sruamsiri, P., Carle, R., & Neidhart, S. (2010). Resistance of industrial mango peel waste to pectin degradation prior to by-product drying. International Journal of Food Science & Technology, 45, 1647e1658.

C.H. Geerkens et al. / Food Hydrocolloids 51 (2015) 241e251 Srirangarajan, A. N., & Shrikhande, A. J. (1979). Technical note: comparative aspects of pectin extracted from the peels of different varieties of mango. International Journal of Food Science & Technology, 14, 539e541. Sudhakar, D. V., & Maini, S. B. (2000). Isolation and characterization of mango peel pectins. Journal of Food Processing and Preservation, 24, 209e227. Tharanathan, R. N., Yashoda, H. M., & Prabha, T. N. (2006). Mango (Mangifera indica L.), “the king of fruits”dan overview. Food Reviews International, 22, 95e123.

251

squez-Caicedo, A. L., Heller, A., Neidhart, S., & Carle, R. (2006). Chromoplast Va morphology and b-carotene accumulation during postharvest ripening of mango cv. ‘Tommy Atkins’. Journal of Agricultural and Food Chemistry, 54, 5769e5776. Zhao, M., Moy, J., & Paull, R. E. (1996). Effect of gamma-irradiation on ripening papaya pectin. Postharvest Biology and Technology, 8, 209e222.