Internal discoloration of various varieties of Macadamia nuts as influenced by enzymatic browning and Maillard reaction

Scientia Horticulturae 192 (2015) 180–186

Contents lists available at ScienceDirect

Scientia Horticulturae journal homepage: www.elsevier.com/locate/scihorti

Internal discoloration of various varieties of Macadamia nuts as influenced by enzymatic browning and Maillard reaction Warangkana Srichamnong a,b,∗ , George Srzednicki b a b

Institute of Nutrition, Mahidol University, Phuttamonton, Nakhonpathom 73170, Thailand School of Chemical Engineering, Department of Food Science and Technology, UNSW Australia, NSW 2032, Australia

a r t i c l e

i n f o

Article history: Received 9 April 2015 Received in revised form 2 June 2015 Accepted 5 June 2015 Keywords: Browning Enzymatic browning reaction Internal discolouration Macadamia Maillard reaction

a b s t r a c t Brown kernels or internal discolouration (IDC) in various varieties of Macadamia nuts (Macadamia integrifolia and Macadamia tetraphylla) occurred through three different pathways: (i) an enzymatic browning reaction, (ii) Maillard reaction and (iii) infection by microorganisms. The phenolic compounds and polyphenol oxidase (PPO) activities of macadamias were analysed. Brown sections of the same kernel had higher levels of bound phenolics compared to the white sections, indicating the participation of phenolic compounds in the formation of brown kernel. The Maillard reaction was studied by determining the sugar amount using HPLC. The reducing sugars during drying reacted with kernel proteins causing the formation of brown pigments. Among the various varieties studied, ‘Daddow’ variety showed the least degree of hydrolysis. When kernel was infected with Penicillium aurantiogriseum, the kernel turns brown. The internal discoloration in macadamias is the first report that has been explored in our study. The findings of this study have potential to improve the existing postharvest techniques used in the Macadamia processing industry. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Macadamia nuts are packed with numerous health benefiting nutrients, minerals, antioxidants and vitamins that are essential for optimum health and wellness. Macadamia nuts have sweet taste and are rich source of monounsaturated oil like oleic acid (18:1) and palmitoleic acids (16:1) (Murano, 2003). They are widely grown in Australia, South Africa, Guatemala and United States of America, especially, in Hawaii (USDA, 2011). Only two species viz., Macadamia integrifolia and Macadamia tetraphylla and their hybrid varieties are commercially available as compared to other species, which have a bitter taste and unsuitable for consumption. Macadamia nuts contain high content of unsaturated fats, which are beneficial to health by reducing the low density lipoprotein cholesterol levels and improving the markers for oxidative stress, inflammation and clotting tendencies (Griel et al., 2008). From the review literature, high consumption of unsaturated fat could have some health benefits, including prevention of diabetes, control of body weight and prevention of cardiovascular disease (Lovejoy, 2005). Garg et al. (2007) reported that consumption of macadamia

∗ Corresponding author at: Institute of Nutrition, Mahidol University, Phuttamonton, Nakhonpathom 73170, Thailand. Fax: +66 2 441 9344. E-mail address: [email protected] (W. Srichamnong). http://dx.doi.org/10.1016/j.scienta.2015.06.012 0304-4238/© 2015 Elsevier B.V. All rights reserved.

kernels resulted in significant reduction of oxidative stress and inflammation in human body. In addition, the high concentration of oleic acid in macadamias is beneficial for decreasing the coronary heart disease due to the high percentage of unsaturated fatty acids (Sinanoglou et al., 2014) The development of browning is associated with Macadamia nuts, which usually occur during the thermal processing of postharvest treatment. Postharvest treatments are the steps performed after the nuts are no longer on the tree. Proper postharvest procedures are needed to prevent physical and chemical damage which can lead to loss of quality. There are generally six postharvest steps in macadamia nut processing; harvesting, de-husking, thermal processing, cracking, grading and packaging. The internal discolouration in macadamia nuts causes the formation of off-flavours and aroma, which in turn leads to economic loss and decrease the nut quality.The extent of this defect, is largely unquantified at this stage. As a consequence, it costs the industry an estimated loss of several million dollars per year due to kernel downgrading and processing inefficiency. More importantly, the occurrence of brown centers in nuts sold at the retail level has the potential to greatly undermine customer confidence and thus reducing repeat sales. The fundamental factors associated with this defect are still unknown. Since this problem occurs erratically, therefore the basic cause of this defect has been difficult to investigate (Lagadec, 2009). However, the internal discolouration or browning of the kernel cen-

W. Srichamnong, G. Srzednicki / Scientia Horticulturae 192 (2015) 180–186

ter varies from light brown to dark brown depending on the shape of nuts which ranges from oval to unconformity. The discolouration occurs mostly at the center but could spread to any locations in the Macadamia kernel. In general, the colour of Macadamia nuts varies from white to creamy; therefore the discolouration in these nuts could be easily detected. Walton and Wallace (2015) reported that by reducing time period between harvest and dehusking processing Macadamia at low moisture content (10–12%) could improve the kernel quality. Whilst, Macadamia nuts stored at high moisture content and elevated temperature with limited air circulation showed increased occurrence of brown center (Walton et al., 2013). Other study conducted by Phatanayindee et al., (2012) showed that reduction of drying time using heat pump dryer combined with tunnel drying resulted in improved kernel quality and reduced IDC in Macadamia nuts. This indicated that brown kernels could be found during all postharvest treatments from harvesting to packaging stage. In general, the kernel quality is influenced by internal and external factors, processing steps and environmental conditions. Furthermore, these factors could be interrelated, for example the polyphenol oxidase enzyme could be inactivated during drying, leading to reduced susceptibility to enzymatic browning. Moreover, brown kernels could be found among fresh, dried and roasted samples. Therefore, it was hypothesized that browning in Macadamia kernel could be triggered by enzymatic and non-enzymatic browning reactions. These two biochemical reactions could form brown pigment as their end product. In addition, microorganisms may also influence brown kernel development via the production of extracellular enzymes. Therefore, the aim of this study was to determine the possible mechanisms of kernel browning and study various factors that are likely related with quality deterioration based on different cultivars. The information obtained from this study will contribute to understand the internal discolouration (IDC) and could lead to preventive measures for maintaining nut quality. 2. Material and methods 2.1. Materials Macadamia samples were obtained from various plantations in different seasons located in New South Wales (Deenford plantation, Lismore) and Queensland. The samples included were various commonly grown varieties including ‘A 38’, ‘246’, ‘816’, ‘842’, and ‘Daddow’. Samples were sorted, graded and cracked. The nuts obtained were graded into 4 categories (1) nuts with smooth texture and light colour; (2) nuts with colour defects, uneven browning or off-colour; (3) nuts with brown centers; (4) dark shriveled nuts. The nuts were classified based on the definitions of Prichavudhi and Yamamoto (1987). Category 1 nuts were used in all the experiments, whilst category 2 nuts were used in brown kernel experiments only. Due to the insufficient quantity of brown nuts, the brown kernel samples were also obtained from colour sorter at Macadamia processing company (MPC Ltd, Lismore, NSW). The belt type colour sorter machine (Alstonville, Australia) is used during processing step to separate brown colour kernel from white colour kernel. Therefore, brown nuts used in this study were procured from both plantations as well as from processor. 2.2. Chemicals and reagents 1, 1-Diphenyl-2-picrylhydrazyl (DPPH), Folin-Ciocalteu reagent, phenolic compounds (gallic acid, protocatechuic acid, p-hydroxybenzoic acid, chlorogenic acid, vanillic acid, caffeic acid and syringic acid) and fatty acids standard C12-C20 were purchased from Sigma–Aldrich (Sydney, Australia). HPLC grade

181

methanol acetonitrile and other solvents were obtained from Merck and Honeywell (Sydney, Australia). All chemicals and reagents used in the study were of analytical grade. 2.3. Kernel segmentation The segmentation of kernels was performed on fresh brown nuts using sharp knife to mechanically separate the non-brown and brown sections. The concentration and types of phenolic compounds present in both brown and non-brown sections were determined separately. Prior to the analysis, the non-brown and brown section was carefully placed in different tubes (1.5 mL). The brown kernels samples were divided into 7 groups (A–G) randomly. Each group included the average data of 10 brown Macadamia nuts. 2.4. Processing 2.4.1. Drying Drying of nuts were carried out through a series of experiments using an in-house built cabinet dryer, followed by heat pump dryer (Greenhalgh, Australia) and finally with vacuum oven (Croydon, England). Samples were dried in the form of nut in shell (NIS). The drying conditions in the cabinet varied from 30 to 50 ◦ C and RH range was 20–40%. Heat pump dryer was operated at 30 ◦ C and 20% RH. Vacuum oven drying was performed at 35 ◦ C at 3.29 kPa. The drying time of each experiment was 8 days. 2.4.2. Roasting The roasting experiments were carried out in a convection oven (Contherm, Australia) at 30, 60, 90, 120 and 150 ◦ C for 30 min. 2.5. Chemical analysis 2.5.1. Moisture content and pH determination The moisture contents of kernels and shells were determined by a vacuum oven method 934.01 (18) G (AOAC, 1995). Macadamia kernels were ground with a coffee grinder and placed in pre-dried aluminium dishes and dried in vacuum oven at 75 ◦ C for 24 h at 3.29 kPa. The ground Macadamia nut was mixed with MilliQ water at 1:4 ratio and was used for pH determination using a pH meter at 25 ◦ C. 2.5.2. Enzyme extraction and analysis of polyphenol oxidase (PPO) The extraction and analysis method of PPO were performed according to Srichamnong et al., (2012). Macadamia nuts (10 g) were defatted twice with 100 mL of hexane and10 mL of petroleum ether. Defatted sample (5 g) was transferred to a 250 mL conical flask and mixed with 100 mL phosphate buffer (pH 6.8) containing 5% polyvinylpyrrolidone (PVPP). Extraction time was 8 h at 4 ◦ C. Samples were then filtered through Whatman No. 41 filter paper and the filtrate was centrifuged at 4 ◦ C at 20,000 rpm for 30 min. Supernatant was collected and diluted with acetone (4 × volume) to precipitate protein. Any remaining liquid was evaporated by nitrogen flushing. The acetone precipitates were kept at −18 ◦ C for further analysis. Prior to purification, protein powder was mixed with 1 mL of 0.1 M phosphate buffer (pH 6.8) and passed through a syringe filter (0.22 ␮m) and injected into a Bio-Rad (BioLogic LP) chromatography system with LP Data view v1.03 software. Chromatography conditions for purification were programmed with gradient elution; flow rate was ramped up from 1 to 2 mL/min. Mobile phase was a binary system: mobile phase A was phosphate buffer (pH 6.8) and mobile phase B was phosphate buffer (pH 6.8) with 1 M NaCl. Total run time of 1 cycle was 113 min. Fractions were collected at 2.5 min interval and subjected to enzyme activity analysis. Absorbance measurements were recorded with a Spectramax

182

W. Srichamnong, G. Srzednicki / Scientia Horticulturae 192 (2015) 180–186

M2 (Molecular device) spectrophotometer with temperature set to 25 ◦ C. 2.5.3. Fat extraction and isolation of fatty acids in kernels The fresh oil was methylated according to the method of Bannon and Craske (1987). Briefly, Macadamia nuts were cracked with a hand cracker and the kernel was solvent extracted in a soxhlet apparatus for 6 h with petroleum ether (bp. 40–60 ◦ C) as extraction solvent. Solvent was evaporated in a rotary evaporator (Buchi, Switzerland) and then the sample was dried in an oven at 75 ◦ C for 30 min to complete solvent evaporation of the petroleum ether at the oil surface. Oil was capped with nitrogen to remove traces of oxygen to minimise lipid oxidation. The fresh oil was methylated according to the method of Bannon and Craske (1987) with some modifications. Macadamia oil extract was transferred to a 50 mL volumetric flask and mixed with 5 mL of 0.25 M sodium methoxide in methanol-diethyl ether (1:1 v/v). The mixture was refluxed for 30 sec and removed from the heat source. A volume of 3 mL of isooctane and 15 mL of saturated sodium chloride were added. The aliquot was vortexed for 20 s and saturated NaCl solution was added to bring the total volume to 50 mL. The mixture was left undisturbed for 1.5 h at room temperature. Approximately 2.5 ␮L from the top layer were collected and injected into a gas chromatograph connected to a flame ionisation detector. 2.5.3.1. Quantification. The theoretical relative response factor (TRF), (Perkin, 1993) was used for conversion of raw peak area to corrected area when analysing fatty acid methyl ester (FAME) by gas chromatography. Software used was GC Solution. 2.5.4. Sugar extraction The method was adapted from Wall and Gentry (2007) with slight modification; 5 g of defatted Macadamia kernel were transferred into a 50 mL test tube and mixed with 80% aqueous ethanol. The mixture was vortexed for 3 min, sonicated for 30 min at room temperature and then heated for 15 min at 70 ◦ C in a water bath. The resulting mixture was evaporated under a stream of nitrogen gas to dryness. Extraction was repeated twice. Spectrophotometric method was performed according to Canizares-Macias, et al., (2001) for quantification. 2.5.4.1. Sugar analysis. Sugar analysis (glucose, fructose and sucrose) was performed by HPLC with a mobile phase of acetonitrile (ACN) and water (80:20 v/v). Several ratios were initially trialed including 75:25, 70:30, and 88:12 v/v. The mobile phase was vacuum filtered through 0.2 ␮m membrane before using at a flow rate of 1.5 mL/min with 10 ␮L sample injection volume. The column used was a Luna 5 ␮ NH2 100A model (Phenomenex, USA) with PDA detector set at 200–800 nm wavelength followed by Refractive index detector (Wall and Gentry, 2007). Quantification was performed by comparing the peak area to known standard curve. Software used was LC Solution. 2.5.5. Phenolic extraction The ground samples were defatted twice with hexane and petroleum ether (bp. 40–60 ◦ C) and defatted samples (5 g) were transferred into a 50 mL centrifuge tube and extracted with solvent. The extraction solvent used for extraction of phenolic compounds was acetone: methanol: water (7:7:6 v/v), with water added to increase the extraction efficiency of phenolics, as most phenolic compounds in Macadamia nuts are hydrophilic. In addition, as phenolic compounds exist in various forms in nature, the use of solvent with different polarity could enhance solubility of phenolics. The supernatant was sonicated for 15 min which was the optimum time for high phenolic recovery and then centrifuged at 4000 rpm for 10 min. The acid hydrolysis and base hydrolysis of the resulting

supernatant were conducted to reduce the signal to noise ratio of the chromatogram and release hydrolysable phenolic compounds. Acid hydrolysis was conducted by addition of 2 N trifluoroacetic acid (TFA) while addition of 4 N NaOH was used for base hydrolysis. The mixture was heated in a water bath at 70 ◦ C for 2 h. The residues were re-extracted again and the supernatants were combined to obtain the free phenolic extract. Bound phenolics were extracted similarly to free phenolics except that the solvent used was ethyl acetate: diethyl ether (1:1 v/v) (Naczk and Shahidi, 2006). Pool supernatant was evaporated to dryness with rotary evaporator and re-dissolved with methanol. These extracts were used for HPLC and antioxidant analysis. Prior to inject into HPLC, the extract was filtered through PTFE 0.22 ␮m syringe filter. 2.5.5.1. Phenolic content determination. Total phenolic content (TP) was measured with the Folin–Ciocalteu reagent and gallic acid as a standard in a 96 well plate. The amount of solution and reagent used were optimised and modified from Tsantili et al., (2010). Phenolic extract (50 ␮L) was pipetted into a well followed by MilliQ water (90 ␮L). Folin–Ciocalteau (2 N) was diluted to 1 mL in 9 mL aqueous solution and added into the same well allowing the reaction to occur for 8 min. Finally 7.5% sodium carbonate solution (100 ␮L) was added. The plate was incubated at 25 ◦ C for 2 h before reading the absorbance at 765 (end product) and 280 (protein content) nm wavelengths. Protein content was monitored due to some proteins which have phenol ring structure will give positive result in TP assay hence wavelength at 280 was monitored and absorbance was maintained as low as possible. Total free and bound phenolics content of brown and white sections within brown kernel samples were also analysed by this method.

2.5.5.2. Analysis of phenolics by HPLC. Binary mobile phase system was used in this experiment. The mobile phase A was 0.3% formic acid aqueous and B was 100% methanol. Gradient mode was as follows: 0–40 min B concentration 37%, 40–50 min B 100%, 50–70 min B 100 %, 70–85 min B 6%, and 85–110 min B 37%. Column temperature was 40 ◦ C to reduce mobile phase viscosity and facilitate the flow of mobile phase which was 0.3 mL/min with 1 ␮L injection volume of the sample. The column type used was reverse phase C18 (Gemini, Phenomenex) with an internal diameter of 3 ␮m and 15 cm in length. Quantification and identification were conducted by comparison to known external standard curve.

2.5.6. DPPH antioxidant activity and melanoidin analysis The test was performed according to the method described by Xu et al., (2007) with a slight modification by using methanol as a solution and phosphate buffer instead of Tris-HCl buffer with similar pH. The activity was given as absorbance at 517 nm.

2.5.7. Microbiological analysis One gram of either mould nut in shell (NIS) or mould kernels sample was homogenised with 9 mL of peptone water. An aliquot of 1 mL was then mixed with 9 mL peptone water in test tube for serial dilution and vortexed for 1 min. After well mixing, 1 mL of each serial dilution was pipetted out and dispersed on Dichloran Rose Bengal Chlortetracycline (DRBC) agar using spread plate method. After inoculation, petri dishes were incubated at 25 ◦ C for 3 days for the growth of colonies. The young colonies were used as wet mount sample. The matured colonies after 6 days of growth were used for the determination of visible colony morphology. The structure of the young spores was observed under a microscope (Leica DM 2500) at 100×. The Leica Application Suite was used for data analysis.

W. Srichamnong, G. Srzednicki / Scientia Horticulturae 192 (2015) 180–186

183

Corretec absorbance at 413 nm

0.5

0.4

A38 246 816 842 Dad

0.3

0.2

0.1

0.0

0

2

4

6

8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

Fig. 1. Polyphenol oxidase activity of different Macadamia kernel of various varieties.

2.6. Statistical analysis Samples were analysed in triplicate and results were subjected to statistical analysis, which included determination of the mean data set of sample and standard deviation as well as analysis of variation (ANOVA) at the significance level p ≤ 0.05. The differences were tested using Duncan Multiple Range Test (DMRT) at the significance level of p ≤ 0.05. 3. Results and discussion The three hypotheses which were the basic cause of browning including: enzymatic browning, non-enzymatic browning and an extracellular enzyme mediated browning caused by microorganisms. Enzymatic browning occurs principally due to PPO and was tested in all Macadamia varieties. Non-enzymatic browning reactions are intensified by heating processes (eg. drying and roasting). Microorganisms also played a role in kernel discoloration as indicated by the experimental results shown below. With exception of the ‘Daddow’ variety, internal browning was observed in all of the varieties studied. In the ‘Daddow’ variety, the kernel colour was slightly yellowish, but not brown. 3.1. Polyphenol oxidase activity in Macadamia kernel PPO was studied in order to evaluate its potential to catalyse enzymatic browning reactions in Macadamia kernels. No significant difference in PPO activity was observed in different varieties (Fig. 1). This indicated that all varieties studied have the same level of browning susceptibility due to enzymatic activity. However, as shown in Fig. 2A, the ‘Daddow’ variety had lowest phenolic concentration compared to other varieties. This indicated that enzyme activity with the limitation of substrate could result in minimum or less activity. (Auden and Dawson, 1931) and thus could be correlated with other varieties which were more prone to ICD. This was postulated that ‘Daddow’ variety has the lowest susceptibility to form brown kernels. These results were in agreement with the previous study conducted by Walton et al., (2013) which stated that brown kernels are formed through high metabolic activity and most of the postharvest browning resulted from enzymatic process. 3.2. Phenolic compounds and antioxidant activity The ‘A38’ and ‘842’ varieties had the highest phenolic content in fresh kernels in the following order ‘A38’> ‘842’> ‘246’> ‘816’> ‘Daddow’ (Fig. 2A). The phenolic content of individual brown kernels was studied. Since, the total phenolic content is represented by both

Fig. 2. Total content of phenolic compounds in fresh and roasted nuts (A) and free and esterified phenolics in white and brown portions of the same kernels (B). The letters A–G represent each kernel. Capital letters alone indicate white portion while small letters followed by (b) indicate brown portion of the same kernel (darker colour).

free phenolic content and esterified phenolic compounds. Thus, free and esterified phenolic compounds were analysed separately because free phenolic alone will not represent total phenolic due to it can esterified with either protein or polysaccharide as nature phenol are reactive. The proportion of free and esterified phenolic compounds was determined in the brown and white sections of the same kernels. The number of kernels studied was divided into 7 groups (A–G) randomly. Each group included the average data of 10 Macadamia nuts (Fig. 1B). The results showed a random distribution in an average processed batch as the brown kernel samples contained various varieties including ‘A38’, ‘Daddow’, ‘246’, ‘816’, ‘842’and others. The total phenolics content of the brown kernels varied between 5290 and10,380 mg gallic acid equivalent (GAE)/kg defatted sample. Most of the brown kernels had a higher content of total phenolics in their brown section than in their white section except in groups E, F and G. The total phenolic content of E, F and G in white

184

W. Srichamnong, G. Srzednicki / Scientia Horticulturae 192 (2015) 180–186

Table 1 Sugar concentration in fresh and dried Macadamia kernels. ‘Variety’drying method

Sample (mg/g DW) Sucrose

‘A38¢ ‘A38¢CD ‘A38¢HP ‘246¢ ‘246¢CD ‘246¢HP ‘816¢ ‘816¢CD ‘816¢HP ‘842¢ ‘842¢CD ‘842¢HP ‘Dad’ ‘Dad’CD ‘Dad’HP

Table 2 Sugar level in Macadamia kernels roasted of variety ‘246’ at 60, 90, 120 and 150 ◦ C for 30 min.

41.0 37.0 37.0 23.0 28.0 27.0 33.0 30.0 30.0 44.7 41.3 40.0 50.0 60.0 60.0

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.06a 0.09a 0.03a 0.30b 0.10b 0.60b 0.10b 0.30b 0.30b 0.40a 0.10a 0.60a 0.80c 0.60c 0.60c

Glucose 3.0 10.0 10.0 2.0 13.0 13.0 3.0 15.0 13.0 2.0 7.0 7.0 4.0 4.0 3.0

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.01a 0.01b 0.30c 0.01a 0.01c 0.00c 0.01a 0.10c 0.003c 0.01a 0.01b 0.01b 0.01a 0.02a 0.01a

Fructose nd 5.0 ± 0.20b 5.0 ± 0.30b nd 7.0 ± 0.30b 7.0 ± 0.20b nd 9.0 ± 0.10b 9.0 ± 0.20b nd 5.0 ± 0.50b 5.0 ± 0.50b 2.6 ± 0.30a 2.6 ± 0.20a 2.6 ± 0.20a

nd = not detected. ns = not-significantly different. CD = cabinet dryer. a Values are means ± SD. Mean separation within columns based on significant difference (p ≤ 0.05). b Each of the three replications consisted of an independent extraction from the same batch. c Statistical analysis was done separately in each drying method since the raw materials were not the same.

and brown centers were in the range of 6951–10,327, 9919–10,324 and 7476–9,385 mg gallic acid equivalent (GAE)/kg, respectively. Approximately 70% of the total brown kernels had higher esterified phenolics content in the brown than in the white sections (Fig. 2B (b) column). In addition, 57.1% of the total brown kernels studied had a lower free phenol content compared to the white section (Fig. 2B). The content of esterified phenolic compounds was higher than that of free phenolic compounds in the same brown kernel (85.8%). Furthermore, brown kernels had a higher total phenolic content compared to non-brown kernels (Fig. 2B). This indicates that phenolics content plays an important role in the enzymatic browning mechanism. Srichamnong et al., (2013) reported that esterification of phenolic compounds with associated protein substrates could be observed by staining Macadamia cells with periodic acid. Further study of protein content and esterified phenolic structure could be determined by using NMR which will give more detail of chemical structure orientation. The scrape method could be improved to give more accuracy by better separation of the brown from white section. There were two major phenolic compounds identified in this study. All Macadamia varieties studied contained chlorogenic acid (14–25 ␮g/g) and phydroxybenzoic acid (3.6–10 ␮g/g) as major phenolic compounds. The p-hydroxybenzoic concentration was lower than that (24 ␮g/g) measured by Quinn and Tang (1996). In addition, hydroxycinnamic was not detected in this study. However, there were some unknown chromatogram peaks that need to be further identified with mass spectrometry. 3.3. Sugar determination The purpose of this investigation was to determine the sugar concentration in the fresh and processed kernel, which can act as a substrate for the Maillard reaction. Fresh kernels contain two different sugars namely sucrose and glucose and the concentrations varied according to the variety (Table. 1). The measured sucrose concentrations were in agreement with literature values (Wall and Gentry, 2007) in which sucrose concentrations of fresh sample ranged between 29.5 and 69.1 mg/g while that of reducing sugars

Roasting temp. (◦ C)

Sucrose

Glucose (Sample mg/g DW)

Fructose

60 90 120 150

21.4 ± 0.10b 19.6 ± 0.20b 19.5 ± 0.10b 1.55 ± 0.00a

5.53 ± 0.30b 5.03 ± 0.20b 4.48 ± 0.10b 1.55 ± 0.20a

4.8 ± 0.10b 4.8 ± 0.10b 4.7 ± 0.00b 1.0 ± 0.20a

a Values are means ± SD. Mean separation within columns based on significant difference (p ≤ 0.05). b Each of the three replication consisted of an independent extraction.

was 3.6–5.3 mg/g (dry sample) in comparison with 2.0–4.0 mg/g in fresh sample. When NIS was dried either in cabinet dryer or heat pump dryer, the sucrose was hydrolysed into glucose and fructose (Table 1). In comparison, after thermal processing, sucrose was not hydrolysed in the ‘Daddow’ variety but naturally present in raw nuts, while sucrose was hydrolysed in other varieties (Table 1). In general, sucrose could be readily hydrolysed with acid and the reaction is accelerated at high temperature. Goldberg et al., (1989) stated that the hydrolysis involving non-electrolytes will have a slight dependence on pH, which depends on ionisation capability of the reactant or product. ‘Daddow’ variety had a higher pH compared to the other varieties, perhaps explaining its slower sucrose hydrolysis. Reducing sugar content of the roasted sample was decreased to half due to roasting temperature. It then reacted with protein in the nuts resulted in colour development through Maillard reaction. The change of reducing sugar content at high temperature is shown in Table 2. A detectable increase in sweetness noticed in the Macadamia nut after roasting may be due to caramelisation associated with browning development. Fructose was not detected in fresh samples except the ‘Daddow’ variety, but was present in all dried samples. In addition, glucose concentrations in dried samples increased except in dried ‘Daddow’ kernels. Sucrose, glucose and fructose were still present in the samples roasted at 60 ◦ C (Table 2). When the temperature was increased to 90 ◦ C, the sucrose concentration decreased, while glucose and fructose concentration increased (Table 2). At 90 ◦ C, the Macadamia kernels showed a slightly more intense golden colour compared to 60 ◦ C. This indicated that the hydrolysis of sucrose into glucose and fructose along with Maillard reactions had occurred. However, after roasting at 120 ◦ C, the concentration of glucose and fructose decreased compared to those at 90 ◦ C (Table 2). This indicated that these two compounds were involved in other chemical reactions. Furthermore, a more intense brown colour appeared at 120 ◦ C. Finally, minute quantities of sucrose, fructose or glucose were detected after roasting at 150 ◦ C. Absorbance data revealed that total phenolic decreased (Fig. 3B) and the formation of melanoidin compounds increased as the temperature and heating time increased (Fig. 3B). In general, the colour of the Maillard reaction by-products (MRP) solution was a darker brown at 120 ◦ C compared to 90 ◦ C. However, heating at high temperature (150 ◦ C) for an increased time period (>30 min) degraded the antioxidant-MRP formed in the early stage of the reaction. This change was reflected by a reduction of absorbance (Fig. 3B). 3.4. Fatty acid profiles The fatty acid concentrations of the brown and non-brown kernels of variety ‘246’ are shown in Table 3. Majority of fatty acids were oleic acid (300 and 562 mg/kg kernel) palmitoleic acid (86 and121 mg/kg kernel) and palmitic acid (43.6 and 63.2 mg/kg kernel) from non-brown and brown respectively. There was no significant difference noted among fatty acid profiles. However,

(A) 1.6

(B) 1.6

Absorbance at 765 nm

Absorbance at 517 nm

W. Srichamnong, G. Srzednicki / Scientia Horticulturae 192 (2015) 180–186

1.5 1.4 1.3 1.2

30 C

60 C

90 C

120 C

150 C

Temperature

185

1.5 1.4 1.3 1.2

30 C

60 C

90 C

120 C

150 C

Temperature

Fig. 3. Absorbance indicating total phenolics content (A) and melanoidin production (B) of Macadamia kernels roasted for 30 min at 60, 90, 120 and 150 ◦ C.

Table 3 Comparison of fatty acid composition of brown and non-brown kernel of Macadamia variety ‘246’. Fatty acid

Palmitic (C16:0) Palmitoleic (C16:1)ns Stearic (C18:0)ns Oleic (C18:1) Linoleic (C18:2)ns Arachidic (C20:0) Eicosenic (C20:1)ns Behenic (C22:0)

Concentration (mg/kg kernel) Brown kernel

Non-brown kernel

62.3 ± 12.2 121.7 ± 25.0 25.2 ± 20.8 561.7 ± 12.4b 9.1 ± 2.7 30.5 ± 0.6b 21.5 ± 7.1 13.0 ± 6.1b

40.3 ± 9.0a 86.4 ± 10.4 15.0 ± 4.6 300 ± 74.0a 10.7 ± 2.4 13.1 ± 3.3a 14.0 ± 2.5 4.5 ± 0.1a

b

ns = Not-significantly different. a Values are means ± SD. Mean separation within lines based on significant difference (p ≤ 0.05). b Each of the three replications consisted of an independent extraction.

the quantity of fatty acids in brown kernels was almost double as compared to non-brown kernels (Table 3). Thus, fatty acid concentrations could be involved in kernel browning. This result was in agreement with the previous study of Rui et al., (2010), which stated that fatty acid accumulation resulted in membrane disintegration, hence internal browning could occur. However, the actual mechanism of the factors which lead to fatty acid accumulation remains unclear. 3.5. Microorganisms and kernel discoloration Browning due to microorganisms such as mould was also evidenced in this study. Interestingly, not all mould infected samples turned brown. The ‘816’ and ‘A38’ varieties were stored under similar conditions as that of ‘842’ and were infected by moulds (greenish area with filaments). However, they did not produce a brown kernel. Samples that were kept at an elevated temperature (35 ◦ C) and high moisture content (22% wet basis) showed mould growth and the kernels turned from white to brown. The moulds present on NIS and kernels were identified as Penicillium aurantiogriseum, when the colony morphology and spore structure were compared with a previous study involving pistachio, hazelnut, pecan and peanuts (Pitt and Hocking, 1997). Based on microscopic examination, microorganisms possess biverticillate type conidiophore, and its shape was slender and in ampulliform (Fig. 4A). Hyphae were septate with branched conidiophores (Fig 4B). In addition, its conidia wall structure was smooth and in spherical shape (Fig 4C). These microorganisms also produce an inverted orange pigment. The picture of conidiophores, hyphae and conidia were showed as Fig 4. Thus; we identified similar mould species in NIS and kernels. In addition, P. aurantiogriseum has high lipolytic activity; and thus infected kernels could experience significant lipid degradation which is undesirable.

Fig. 4. Mycelium (A), (B) and hypae structure (C) conidia under 100X in colonies on DRBC agar, 3 days, 25 ◦ C isolated from brown and mouldy Macadamia kernel.

Uncontrolled storage conditions resulted in browning due to microorganism infection even in unprocessed Macadamia nuts. Therefore, brown colour development in kernels prior to drying could be due to mould infection through cracks or the faulty formation of the nut micropyle. The nuts in the present study were

186

W. Srichamnong, G. Srzednicki / Scientia Horticulturae 192 (2015) 180–186

noticed with the small hole at the micropyle position which could promote the microbial infection and subsequently browning in kernels. 4. Conclusions Browning of kernels occurred through three different pathways (i.e., (i) microorganisms, (ii) Maillard reaction, and (iii) enzymatic browning) based on the postharvest treatments used. Soon after the nuts achieve maturity and falls on the ground, it is susceptible to browning by microorganisms. Microorganism contamination could occur through the open micropyle or faulty shell formation or cracks. It is therefore recommended to harvest the nut as soon as it falls on the ground and use good agricultural practices to minimise unwanted microbial attack. In addition, the wet husk attached to the NIS could be a source of microorganisms. Prompt removal of the husk needs to be strictly followed. Sucrose is hydrolysed into glucose and fructose. These reducing sugars together with a protein component can lead to browning induced by the Maillard reaction during drying. This was evidenced by microscopy work in which high concentration of proteins were distributed in the brown kernel sample. Thus kernels with high protein content are more susceptible to browning. Regardless of sucrose concentration, the amount of glucose and fructose hydrolysed during drying are the key determinants in the degree of browning. The third possible mechanism, enzymatic browning is related to phenol content as the brown sections of the kernels had higher bound phenolic concentrations than the white sections of the same kernel. The brown sections result from esterification of the phenolic compounds with associated protein substrates. The ‘Daddow’ variety had the highest activity of polyphenol oxidase compared to other varieties studied but it contains lowest phenolic content. Thus, that phenolic content as well as polyphenol oxidase activity play a role in kernel discoloration. Acknowledgements The authors would like to thank Mr. Cliff James, Mr. Steven Lee, Brice Kaddatz and Kim Jones for their technical support and Macadamia Processing Company (MPC) Lismore, Australia for supplying the raw materials. The author would also like to thanks Prof William Price at University of Western Sydney, Australia for this guideline and final look for the manuscript. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.scienta.2015. 06.012 References AOAC, 1995. Official methods of analysis, Association of Official Analytical Chemists, 15th ed. AOAC International, Washington D.C.

Auden, A.H., Dawson, R.E., 1931. The hydrolysis of concentrated sugar solution by invertase. J. Biol. Chem. 25 (6), 1909–1916. Bannon, C.D., Craske, J.D., 1987. Gas liquid chromatography analysis of the fatty acid composition of fats and oils: a total system for high accuracy. J. Am. Oil Chem. Soc. 64, 1413–1417. Garg, M.L., Blake, R., Wills, J., 2007. Macadamia nut consumption modulated favourably risk factor for coronary artery disease in hypercholesterlemic subjects. Food Lipid 24, 583–587. Goldberg, R.N., Tewarit, Y.B., Ahluwalia, J.C., 1989. Thermodynamics of the hydrolysis of sucrose. J. Biol. Chem. 246 (17), 9901–9904. Griel, A.E., Cao, Y., Bagshaw, D.D., Cifelli, A.M., Holub, B., 2008. A Macadamia nut-rich diet reduced total and LDL-Cholesterol in mildly hypercholesterolemic men and women. Am. Soc. Nutr. J. 138 (4), 761–767. Lagadec, D., 2009. Kernel brown centers in macadamia: a review. Crop Pasture Sci. 60, 1117–1123. Lovejoy, A., 2005. The impact of nuts on diabetes and diabetes risk. Curr. Diab. Rep. 5, 379–384. Murano, P., 2003. Understanding Food Science and Technology, 1 ed. Thomson Learning, USA, pp. 10–15 (Chapter 1). Naczk, N., Shahidi, F., 2006. Phenolics in cereals, fruits and vegetables: occurrence, extraction and analysis. J. Pharmacol. Biol. Anal. 41, 1523–1542. Phatanayindee, S., Borompichaichartkul, C., Srzednicki, G., Craske Wootton, J.M., 2012. Changes of chemical and physical quality attributes of Macadamia nuts during hybrid drying and processing. Dry. Technol. 30 (16), 1870–1880. Perkin, G., 1993. Analysis of Fats, Oils and Derivatives. AOCs Press, America (Chapter 16). Pitt, J.I., Hocking, A.D., 1997. Fungi and Food Spoilage, 2nd ed. Blackie Academic & Professional, Melbourne. Prichavudhi, K., Yamamoto, H. Y., 1987. Effect of drying temperature on chemical composition and quality of Macadamia nuts. C.M.S. book. 98–104. Quinn, L.A., Tang, H.H., 1996. Antioxidant properties of phenolic compounds in Macadamia nuts. JAOCS 73 (11), 1585–1588. Rui, H., Cao, S., Shang, H., Jin, P., Wang, K., Zheng, Y., 2010. Effects of heat treatment on internal browning and membrane fatty acid in loquat fruit in response to chilling stress. J. Sci. Food Agri. 90, 1557–1561. Sinanoglou, V.J., Kokkotou, K., Fotakis, C., Strati, I., 2014. Monitoring the quality of gamma-irradiated Macadamia nuts based on lipid profile analysis and Chemometrics. Traceability models of irradiated samples. Food Res. Int. 60, 38–47, Special Issue. Srichamnong, W., Wootton, M., Srzednicki, G., 2012. Lipoxygenase and Peroxidase Activity of Macadamia Kernel after Thermal Processing. ISHS Acta Horticulturae 943. Asia Pacific Symposium on Postharvest Research, Education and Extension. Srichamnong, W., Price, B., Gardner, T., Dean, R., Plougonven, E., Léonard, A., Srzednicki, G., 2013. Studies of microstructure of kernels of Macadamia integrifolia and its hybrids through MRI, X-ray tomography and confocal microscopy. J. Food Sci. Eng. 3, 503–516. USDA, 2011. United State Department of Agriculture: Tree nut production in selected countries. USDA national agricultural statistics services. Accessed on 31 March 2010. Wall, M.M., Gentry, T.S., 2007. Carbohydrate composition and color development during drying and roasting of Macadamia nuts (Macadamia integrifolia). Lwt-Food Sci. Technol. 40, 587–593. Walton, D.A., Wallace, H.M., 2015. The effect of mechanical dehuskers on the quality of Macadamia kernels when dehusking Macadamia fruit at differing harvest moisture contents. Sci. Horticult. 182, 119–123. Walton, D.A., Randall, B.W., Le Lagadec, M.D., 2013. Maintaining high moisture content of Macadamia nuts-in-shell during storage induces brown centers in raw kernels. J. Sci. Food Agri. 93 (12), 2953–2958. Xu, Q., Tao, W., Ao, Z., 2007. Antioxidant of vinegar melanoidins. Food Chem. 102, 841–849.