Inulin suitable as reduced-kilojoule carrier for production of microencapsulated spray-dried green Cyclopia subternata (honeybush) extract

Inulin suitable as reduced-kilojoule carrier for production of microencapsulated spray-dried green Cyclopia subternata (honeybush) extract

LWT - Food Science and Technology 75 (2017) 631e639 Contents lists available at ScienceDirect LWT - Food Science and Technology journal homepage: ww...

1MB Sizes 2 Downloads 70 Views

LWT - Food Science and Technology 75 (2017) 631e639

Contents lists available at ScienceDirect

LWT - Food Science and Technology journal homepage: www.elsevier.com/locate/lwt

Inulin suitable as reduced-kilojoule carrier for production of microencapsulated spray-dried green Cyclopia subternata (honeybush) extract Claire Pauck a, Dalene de Beer a, b, *, Marique Aucamp c, Wilna Liebenberg c, Nicole Stieger c, Chantelle Human a, b, Elizabeth Joubert a, b a b c

Department of Food Science, Stellenbosch University, Private Bag X1, Matieland, 7602, Stellenbosch, South Africa Post-Harvest and Wine Technology Division, ARC Infruitec-Nietvoorbij, Private Bag X5026, 7599, Stellenbosch, South Africa Centre of Excellence for Pharmaceutical Sciences, North-West University, Potchefstroom, South Africa

a r t i c l e i n f o

a b s t r a c t

Article history: Received 4 July 2016 Received in revised form 4 October 2016 Accepted 7 October 2016 Available online 11 October 2016

Retention of phenolic compounds during spray-drying of an anti-diabetic C. subternata extract and physicochemical characteristics of spray-dried powders (pure extract and extract-carrier mixtures) were evaluated. Extract-carrier mixtures contained three levels (250, 500 and 750 g/kg) of the microencapsulating agents, namely corn syrup solids, commonly used by the food industry, and inulin, a lowkilojoule alternative. The amorphous spray-dried powders ranged from nearly free-flowing to cohesive. Their moisture content and water activity fell within the range of their monolayer moisture values. The moisture sorption isotherm of the pure extract showed very little hysteresis, contrary to the mixtures containing carriers. Similar values for calculated and experimental heat flow, determined by isothermal microcalorimetry, indicated the carriers to be compatible with the extract, except when used in a mixture containing 750 g/kg corn syrup solids per total solids. Spray-drying had no detrimental effect on the individual phenolic content, in particular the heat labile mangiferin, isomangiferin and 3-b-D-glucopyranosyliriflophenone, and the total antioxidant capacity of the extract. Microencapsulation of C. subternata extract with inulin by spray-drying thus provides a stable low-kilojoule powder, suitable for formulation of single-serve beverage mixtures that can be used by diabetics. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Phenolic compounds Physicochemical properties Stability Thermal analysis Chemical compounds studied in this article: Eriocitrin (PubChem CID: 83489) Hesperidin (PubChem CID:10621) Isomangiferin (PubChem CID: 5318597) Mangiferin (PubChem CID: 5281647)

1. Introduction In the ongoing search for novel sources of bioactive compounds Cyclopia species (honeybush) have been identified for the production of value-added extracts for use in nutraceuticals and functional food products. Of particular interest is the xanthone, mangiferin, found in high concentrations in Cyclopia spp. Its potent antioxidant capacity and other health benefits, which include anti-diabetic properties, are well-documented (Vyas, Syeda, Ahmad, Padhye, & Sarkar, 2012). Cyclopia subternata also contains substantial quantities of isomangiferin with similar glucose-lowering activity in vitro to mangiferin (Schulze et al., 2016). Other compounds of interest are the benzophenone a-glucosidase inhibitors (Beelders

* Corresponding author. Post-Harvest and Wine Technology Division, ARC Infruitec-Nietvoorbij, Private Bag X5026, 7599, Stellenbosch, South Africa. E-mail address: [email protected] (D. de Beer). http://dx.doi.org/10.1016/j.lwt.2016.10.018 0023-6438/© 2016 Elsevier Ltd. All rights reserved.

et al., 2014), contributing to the anti-diabetic activity of Cyclopia extracts through postprandial regulation of blood glucose levels. An aqueous extract of “unfermented” C. subternata has been demonstrated to have glucose lowering (Schulze et al., 2016) and antiobesity properties (Dudhia et al., 2013), further substantiating interest in honeybush for the production of functional products. Other phenolic glycosides present in this extract include flavanones, dihydrochalcones and a flavone (De Beer et al., 2012; Schulze et al., 2016). A common application of tea and herbal tea extracts is the production of ready-to-drink iced teas and “instant” teas, i.e. dry formulated powder mixtures sold in bulk or convenient singleserve format. For these purposes a stable, free-flowing powder is required for further formulation. Spray-drying is used extensively to produce dry powders from natural extracts (Gharsallaoui, Roudaut, Chambin, Voilley, & Saurel, 2007). It is common practice to microencapsulate natural extracts by spray-drying with carriers. Microencapsulation of the extract aids protection of phenolic

632

C. Pauck et al. / LWT - Food Science and Technology 75 (2017) 631e639

Abbreviations aw water activity BET model Brunauer, Emmett and Teller model CS corn syrup solids CS25 C. subternata extract spray-dried with 250 g CS/kg CS50 C. subternata extract spray-dried with 500 g CS/kg CS75 C. subternata extract spray-dried with 750 g CS/kg DE dextrose equivalents DP degree of polymerisation DSC differential scanning calorimetry DTA differential thermal analysis DTG simultaneous TG/DTA HPLC-DAD high-performance liquid chromatography with diode-array detection

compounds during spray-drying, as well as improves the physicovy, 2011). chemical properties of the powder (Munin & Edwards-Le A common carrier used by industry in a variety of food products is corn syrup solids, a cheap, versatile and readily available product. However, it has a similar metabolic effect to that of sugar (Gross, Li, Ford, & Liu, 2004) with an energy value of 16.3 kJ/g (Roberfroid, 1999) making it unsuitable for products aimed at healthconscious or diabetic consumers. Inulin, on the other hand, is a low-kilojoule (6.3 kJ/g) and indigestible prebiotic fibre (Barclay, Ginic-Markovic, Cooper, & Petrovsky, 2010; Schaller-Povolny, Smith, & Labuza, 2000) that stimulates the growth of beneficial bifidobacteria species (Kolida, Tuohy, & Gibson, 2002). Inulin is thus a healthy alternative to corn syrup solids for use in functional beverages. The objectives of this study were to determine the effect of spray-drying on the total antioxidant capacity and phenolic composition of green C. subternata extract. Different ratios of the microencapsulating agents, corn syrup solids and inulin, added to the extracts were investigated to improve the physicochemical properties of the powder. Compatibility of the extract with corn syrup solids and inulin was evaluated to indicate the potential storage stability of the extract when microencapsulated with these carriers. Adsorption and desorption moisture isotherms, amorphous/crystalline nature, thermal characteristics and bulk properties of the spray-dried powders were determined to gain insight into their expected stability during storage.

2. Materials and methods 2.1. Chemicals, reagents, extracts and carriers Authentic reference standards (purity > 950 g/kg) were from Sigma-Aldrich (St Louis, USA: mangiferin, hesperidin, maclurin, 3b-D-glucopyranosyliriflophenone), Extrasynthese (Genay, France: luteolin) and Phytolab (Vestenbergsgreuth, Germany: eriocitrin, vicenin-2). Sigma-Aldrich supplied HPLC gradient grade acetonitrile for HPLC analysis. Deionised water, prepared using an Elix® water purification system (Merck-Millipore, Darmstadt, Germany), was further purified to HPLC grade using a Milli-Q™ Reference Aþ System (Merck-Millipore). Analytical grade chemicals were obtained from either Sigma-Aldrich or Merck-Millipore. A vacuum-dried, hot water extract of unfermented C. subternata with proven in vivo glucose-lowering activity (Schulze et al., 2016) was used to prepare the liquid feed solution for spray-drying with or without added carrier. Star Dri® 200 corn syrup solids with

IN IN25 IN50 IN75 M0 MSI RH SEM TAC Tg TG TPC XRPD

inulin C. subternata extract spray-dried with 25 g IN/kg C. subternata extract spray-dried with 500 g IN/kg C. subternata extract spray-dried with 750 g IN/kg monolayer moisture content moisture sorption isotherm relative humidity scanning electron microscope total antioxidant capacity glass transition temperature thermogravimetry total polyphenol content X-ray powder diffraction

20e23 dextrose equivalents (DE) was kindly donated by Tate and Lyle (Cape Town, South Africa). Orafti® HP inulin, a long-chain inulin food additive derived from chicory roots (Cichorium intybus) with 21e26 degrees of polymerisation (DP), was procured from Savannah Fine Chemicals (Pty) Ltd (Gardenview, South Africa). 2.2. Spray-drying treatments The dry extract and carriers were mixed in predefined ratios to obtain a total mass of 40 g solids. The feed solution (100 g/L) was prepared by adding the dry powder mixtures slowly to deionised water (400 mL) at ca 55  C, while stirring on a magnetic stirrer until completely dissolved (ca 10 min). Each carrier was added at three treatment levels, namely at 250, 500 and 750 g/kg of the total solids content. A control sample consisting of pure extract was also spraydried. Each of the treatments was replicated 4 times (8 replicates for control) in a completely randomised design. Spray-drying was conducted using a Büchi B-290 mini spraydryer (Büchi Labortechnik AG; Flawil, Switzerland) equipped with a glass cyclone separator and a 1.5 mm nozzle aperture diameter. The feed solution was sprayed in a co-current direction using air as a drying medium. Following preliminary experiments to assess suitability of operating conditions in terms of powder yield (>600 g/kg) and moisture content, final operating conditions selected were: inlet temperature, 180  C, aspirator rate, 35 m3/h, peristaltic pump speed, 7.5 mL/min, atomisation air flow rate, 667 L/h, and nozzle cleaner, 8 strikes/min. The outlet temperature (ca 90e100  C) varied depending on the composition of the feed solution. The collected powder was weighed in amber vials to determine the powder yield (calculated as g powder obtained per kg solids in the feed solution) and stored at ambient temperature in a desiccator with silica gel until analysis. 2.3. Characterisation of powders 2.3.1. Moisture content, water activity and moisture sorption isotherms The moisture content of the spray-dried powders was determined by drying ca 2 g of sample at 100  C for 60 min using an HR73 Halogen Moisture analyser (Mettler Toledo; Greifensee, Switzerland) and expressed as g/kg of the total mass of the product (wet basis), as well as g/kg of dry mass (dry basis). Water activity (aw) was determined using a Novasina LabMASTER-aw electric hygrometer (Lachen, Switzerland). Moisture sorption analyses of the spray-dried powders were

C. Pauck et al. / LWT - Food Science and Technology 75 (2017) 631e639

performed, using a VTI-SA vapour sorption analyser (TA Instruments; New Castle, USA). A pre-drying step was performed at 40  C until the mass fluctuated by no more than 0.001 g. The analysis was conducted at 25  C, while the percentage relative humidity (% RH) was ramped from 5 to 75% RH during the adsorption phase, followed by a desorption step (inverse of adsorption) and a second adsorption step. The program criteria were set to 2-min stability of mass gained or lost before the program would continue to the next set parameter. The data were fitted to the classical Brunauer, Emmett and Teller (BET) sorption isotherm suitable for powders with low aw (Labuza & Altunakar, 2008). 2.3.2. Quantification of major phenolic compounds HPLC-DAD analysis, performed on an Agilent 1200 series instrument (Agilent Technologies Inc., Santa Clara, USA), was used to quantify the major phenolic compounds in the feed solutions and the spray-dried powders according to a method validated for C. subternata (De Beer et al., 2012). Briefly, the method entailed separation using a Gemini-NX C18 column (150  4.6 mm; 3 mm; 110 Å; Phenomenex, Santa Clara, USA) with a multilinear gradient of (A) 2% acetic acid and (B) acetonitrile at a flow rate of 1 mL/min: 0e2 min (8% B), 2e27 min (8e38% B), 27e28 min (38e50%), 28e29 min (50% B), 29e30 min (50e8% B), 30e40 min (8% B). Compounds were identified based on retention time and UVeVis spectra (200e450 nm) of the available authentic standards or literature values (Beelders, De Beer, Stander, & Joubert, 2014; De Beer et al., 2012). Due to scarcity of authentic reference standards, isomangiferin, scolymoside, 3-b-D-glucopyranosyl-4-b-Dglucopyranosyloxyiriflophenone and 30 ,50 -di-b-D-glucopyranosylphloretin were quantified using previously determined response factors with compounds present in the calibration mixture. 3Hydroxy-phloretin-30 ,50 -di-C-hexoside was quantified as 30 ,50 -dib-D-glucopyranosylphloretin equivalents due to unavailability of a reference standard. Results were converted to represent g compound/kg pure extract solids (recalculated to dry basis) in order to directly compare treatments, irrespective of carrier level. 2.3.3. Total polyphenol and total antioxidant capacity assays The total polyphenol content (TPC) and total antioxidant capacity (TAC) of the feed solutions and spray-dried powders were determined using the Folin-Ciocalteau and DPPH radical scavenging assays, respectively, adapted for microplate format as described by Arthur, Joubert, De Beer, Malherbe, and Witthuhn (2011). TPC and TAC were expressed as g gallic acid equivalents/ kg pure extract (dry basis) and mmole Trolox equivalents/g pure extract (dry basis), respectively. 2.3.4. Isothermal microcalorimetry A 2277 Thermal Activity Monitor (TAM III) (TA Instruments; New Castle, USA) equipped with an oil bath (stability ± 100 mK over 24 h) was used for compatibility studies. The heat flow was measured for the single components, as well as the spray-dried mixtures at 60  C. The observed calorimetric outputs for the single components were summed to give a theoretical response for the mixture and a heat flow difference calculated, expressed in mW/g. 2.3.5. Simultaneous thermogravimetry/differential thermal analysis (DTG) A Shimadzu DTG-60 instrument (Kyoto, Japan) was used to record differential thermal analysis (DTA) and thermogravimetry (TG) simultaneously. TG measures changes in the material during heating leading to mass loss (mg), e.g. moisture loss and degradation, while DTA measures the temperature difference between an inert reference and a sample. Therefore, the characteristic temperature of any endothermic or exothermic changes in the sample

633

can be detected, e.g. transformations (crystal structure changes), melting, glass transitions (Tg), crystallisation or sublimation. Samples (3e5 mg) were accurately weighed in open aluminium cells and heated from 25 to 300  C at a heating rate of 10  C/min and a nitrogen gas purge rate of 35 mL/min. Spray-dried samples were compared to pure inulin and corn syrup solids. 2.3.6. X-ray powder diffraction (XRPD) X-ray powder diffraction measurements were performed at ambient temperature on the spray-dried samples, using a PANalytical Empyrean X-ray diffractometer with a PIXcel3D detector (Almelo, Netherlands). Samples were evenly distributed on a zero background sample holder. The measurement conditions were: target, Cu; voltage, 40 kV; current, 40 mA; divergence slit, 2 mm; anti-scatter slit, 0.6 mm; detector slit, 0.2 mm; scanning speed, 2 / min (step size, 0.02 ; step time, 1.0 s). 2.3.7. Objective colour measurement CIE L*a*b* objective colour measurements were performed on the spray-dried samples and pure carriers, using a Konica Minolta CM-5 spectrophotometer (Osaka, Japan) with a 152 mm integrating sphere. Measurement conditions were: mode, reflectance; illuminant, D65 with diffuse illumination; measurement area, 30 mm; viewing angle, 8  ; observer angle, 10  . Auto-calibration (built-in standard white calibration plate) and manual zero calibration (black inverted cone cylinder) were performed prior to measurements. Powders were placed in quartz cuvettes and analysed in duplicate. Chroma, hue and colour difference between the control and treatment samples (DE) were calculated. 2.3.8. Determination of powder bulk properties The poured bulk density (Dp) of the pure C. subternata spraydried extract and C. subternata extract microencapsulated with corn syrup solids or inulin was calculated as the ratio between the sample mass and its volume occupied after gently pouring into a 100 mL cylinder. The tapped bulk density (Dt) was calculated in the same way after tapping the cylinder until no measurable change in volume was noticed (ca 100 times). These values were used to calculate Carr's compressibility index (Ic) and the Hausner ratio (HR) (Schüssele & Bauer-Brandl, 2003). Measurements were made in triplicate.

Ic ¼ Dp  Dt



Dp

 HR ¼ Dp Dt

2.3.9. Scanning electron microscope (SEM) imaging SEM imaging of the spray-dried microparticles was accomplished using a Zeiss MERLIN field emission SEM and Zeiss SmartSEM software (Carl Zeiss Microscopy GmbH; Jena, Germany). The samples were mounted on a small stub using carbon tape and then sputter coated with a thin (10 nm) layer of gold, using an Edwards S150A Gold Sputter Coater (Crawley, United Kingdom) to make the sample conductive prior to imaging. Secondary electron images of the sample were captured using a Zeiss inlens (SE1 e Secondary Electron type 1) detector (Carl Zeiss Microscopy GmbH). The beam conditions during the image analysis were 5 kV acceleration voltage (extra-high tension target), 250 pA beam current (Iprobe), 3 mm - 5.5 mm working distance and a high resolution column configuration (column mode). 2.3.10. Turbidity measurement of powder solutions The powders were dissolved at room temperature in deionised

634

C. Pauck et al. / LWT - Food Science and Technology 75 (2017) 631e639

water at a concentration relevant for an iced tea formulation, namely 1.75 g extract/L. The turbidity of the solutions was measured in Nephelometric Turbidity Units (NTU) using a Hach 2100N IS Turbidimeter (Loveland, USA).

Table 2 Constants and R2 values of moisture sorption data (n ¼ 1) obtained at 25  C for pure spray-dried pure Cyclopia subternata extract, as well as mixtures of extract and carriers, inulin (IN) and corn syrup solids (CS) at levels of 250, 500 and 750 g/kg of the total soluble solids (IN25, IN50, and IN75; CS25, CS50 and CS75), using the Brunauer, Emmett and Teller (BET) sorption model, as well as their measured water activity (aw) and moisture content (n ¼ 4).

2.4. Statistical analysis The experimental design was completely random with replicate experiments (n ¼ 4) for each of the 8 treatments, namely pure extract before spray-drying, pure extract after spray-drying, CS25, CS50, CS75, IN25, IN50 and IN75 (section 2.3). Univariate analysis of variance was performed using the General Linear Models procedure of SAS software (Version 9.2; SAS Institute Inc.; Cary, USA). The Shapiro-Wilk test was performed to test for normality and data points removed when their standardised residual deviated with more than three standard deviations from the model value. Fisher's least significant difference was calculated at P ¼ 0.05 to compare treatment means.

Treatments

M0 (g/kg, dry basis)

R2

aw

Pure extract IN25 IN50 IN75 CS25 CS50 CS75

51.0 47.7 56.6 60.2 50.4 49.4 49.8

0.999 0.979 0.997 0.999 0.993 0.995 0.995

0.17 0.18 0.11 0.16 0.21 0.19 0.16

Moisture (g/kg, dry basis) ab ab b b a ab b

34.1 36.3 29.2 37.7 41.6 37.1 34.9

a a a a a a a

Abbreviations: M0, monolayer moisture content; aw, water activity. Means in the same column with the same letter are not significantly different (P  0.05).

Spherical particles with wrinkled or dimpled surfaces were obtained. Images of IN50 showed slightly larger particles with a more wrinkled surface, while spray-drying of the pure extract and CS50 delivered smaller particles with a more dimpled surface. The shape and surface of the particles are mainly affected by the carrier (Toneli, Park, Negreiros, & Murr, 2010) and process parameters (i.e. drying air inlet temperature and feed flow rate) (Paramita, Iida, Yoshii, & Furuta, 2010). Maltodextrin and corn syrup solids have been shown to produce particles with indented surfaces (Sheu & Rosenberg, 1998). Pure C. subternata spray-dried extract was an amorphous powder, indicated by the typical “halo pattern” and absence of distinct diffraction peaks of the XRPD diffractogram (Fig. 2). The other powders gave similar diffractograms (data not shown), confirming their amorphous state. The amorphous nature of the powders may be attributed to a combination of the spray-drying process and the fact that it is a complex natural substance s-Rojas & Oliveira, 2012). In spite of the thermodynamic (Corte instability of amorphous powders, they are usually more soluble than their crystalline counterparts, an important characteristic in iced tea product applications. Furthermore, knowledge of the amorphous nature of the powder is important for understanding its stability and storage requirements.

3. Results and discussion 3.1. Powder yield and general characteristics Spray-drying produced fine powders with varying light brown colour depending on the ratio of extract to white-coloured carriers. At the same extract:carrier ratio corn syrup solids and inulin caused the same total colour change (DE) (Table 1). Powder yield ranged from 610 to 640 g/kg (Table 1), considered acceptable for herbal ~ a, & Bucal plant extracts (Gallo, Ramírez-Rigo, Pin a, 2015). Moisture content of the powders varied between 38.3 and 86.9 g/kg (Table 2). Based on Carr's compressibility index, the flowability of the powders can be classified as mostly poor (0.23e0.35), with CS25 extremely poor (>0.40) (Davies, 2004). When considering the Hausner ratios, the powders could be classified from nearly freeflowing (close to 1.2 for pure extract and CS75) to cohesive (>1.6 for CS25). Addition of the carriers did not improve flowability, since values were similar or higher for microencapsulated powders. Flowability of powders are important during the manufacturing of products containing dry powder mixtures, especially during dosing and mixing operations. The powders gave relatively clear solutions when dissolved in water at a concentration (1.75 g extract/L) relevant to an iced tea formulation with values between 25 and 38 NTU. This falls in the range observed for a large sample set of honeybush tea infusions (Bergh, 2014). SEM images of the spray-dried pure extract, CS50 and IN50 showed size polydispersity with particle diameters generally less than 20 mm (Fig. 1), confirming their microparticulate nature.

3.2. Phase transition temperatures In a glassy amorphous form molecular mobility is higher due to the higher free energy state in which it exists. In order to ensure

Table 1 Colour measurements and yield of powders (g/kg) after spray-drying (mean ± standard deviation; n ¼ 4), as well as flowability indices of powders and turbidity of solutionsa (n ¼ 1). Treatments were composed of Cyclopia subternata extract spray-dried with corn syrup solids (CS) and inulin (IN) at levels of 250, 500 and 750 g/kg of the total soluble solids (IN25, IN50, and IN75; CS25, CS50 and CS75). Treatment

L*

Pure extract CS25 CS50 CS75 IN25 IN50 IN75 CS100d IN100d

68.8 71.8 74.8 80.6 72.0 75.3 80.1 98.9 97.7

f ± 0.2 e ± 0.3 d ± 0.3 a ± 0.3 e ± 0.3 c ± 0.4 b ± 0.2

a*

b*

10.6 a ± 0.2 9.0 b ± 0.2 7.6 c ± 0.3 4.9 d ± 0.2 9.1 b ± 0.3 7.4 c ± 0.4 5.1 d ± 0.1 0.2 0.3

30.0 27.7 25.7 21.5 28.7 25.7 21.1 2.0 4.7

C* a ± 0.3 c ± 0.3 d ± 0.7 e ± 0.5 b ± 0.8 d ± 1.3 e ± 0.6

31.8 29.1 32.5 22.0 30.2 26.8 21.8 2.0 4.7

h a ± 0.4 b ± 0.3 c ± 0.6 d ± 0.5 b ± 0.9 c ± 1.3 d ± 0.7

70.5 72.1 73.5 77.1 72.4 73.9 76.4 95.6 93.8

Means in the same column with the same letter are not significantly different (P  0.05). a Turbidity of a solution equalling 1.75 g extract/L. b Carr's compressibility index. c Hausner ratio. d CS100 and IN100 were measured separately for the purpose of comparison.

g ± 0.1 f ± 0.2 d ± 0.2 a ± 0.3 e ± 0.1 c ± 0.2 b ± 0.2

DE

Product yield

0.0 d ± 0.0 4.2 c ± 0.5 8.0 b ± 0.4 15.6 a ± 0.3 3.8 c ± 0.4 8.4 b ± 1.0 15.4 a ± 0.8 e e

622 615 614 613 652 635 625 e e

b ± 36 b ± 26 b ± 14 b ± 16 a ± 11 ab ± 7 ab ± 7

Icb

HRc

Turbidity (NTU)

0.23 0.41 0.29 0.23 0.28 0.27 0.30 e e

1.29 1.69 1.41 1.29 1.39 1.37 1.43 e e

27.1 26.9 25.0 31.6 20.8 25.3 37.8 e e

C. Pauck et al. / LWT - Food Science and Technology 75 (2017) 631e639

635

Fig. 2. X-ray powder diffractogram of spray-dried pure Cyclopia subternata extract.

mixtures. No distinct thermal events were observed on the DTA thermogram of pure extract (Fig. 3). This may be attributed to the fact that C. subternata extract is a natural substance containing a complex mixture of different components. It is possible that glass transition reactions took place in some of the compounds across a variety of temperatures, however, the reactions were too small to be detected. A gradual mass loss of C. subternata extract due to moisture loss was observed up to ca 240  C in the TGA thermogram where after a more rapid decrease in mass occurred (Fig. 3). The mass loss for pure extract due to moisture and thermal degradation was 65.8 and 241.7 g/kg, respectively (Table 3). Substantially higher mass losses due to degradation were observed for inulin and corn syrup solids (Table 3), indicating that the extract is more thermostable than these carbohydrate polymers used for microencapsulation. With a mass loss of 520.4 g/kg due to degradation, inulin is more thermally sensitive to degradation than corn syrup solids with a mass loss of 305.6 g/kg due to degradation (Table 3). As the levels of inulin and corn syrup solids increased in the samples, the mass loss due to degradation also increased as expected (Table 3). Degradation, however, took place above 200  C in all cases, which is higher than temperatures experienced within the spray-dryer, further confirming that it is a suitable processing method for C. subternata extract and extract-carrier mixtures. Further insight into the thermal behaviour of pure inulin and corn syrup solids and their extract mixtures were obtained using DTA (Fig. 4). Both pure inulin and corn syrup solids had melting enthotherms at 246.6  C and 242.6  C, respectively. Of the mixtures, only IN75 had a melting endotherm (232.01  C). A Tg for pure inulin between 50.3 and 59.2  C was observed using DSC analysis (Fig. S1; Supplementary Information), but not for the spray-dried extract-inulin mixtures. The Tg of inulin is

Fig. 1. Scanning electron microscope images of (a) spray-dried pure Cyclopia subternata extract, as well as Cyclopia subternata extract microencapsulated with 500 g/kg (b) corn syrup solids or (c) inulin.

product stability over long storage periods this state should not alter with time. Given the confirmed amorphous nature of the spray-dried powders, insight into conditions that may compromise the amorphous state is required. In particular, relative humidity and high temperatures result in phase transitions which lead to caking of such powders (Bhandari & Howes, 1999). Thermal analysis, such as TG and DTA, gives information about changes in material properties as a function of temperature and the results help to explain the behaviour of the components within the

Fig. 3. Differential thermal (DTA) and thermogravimetric (TGA) analysis of pure Cyclopia subternata extract.

636

C. Pauck et al. / LWT - Food Science and Technology 75 (2017) 631e639

Table 3 Mass loss due to loss of moisture and degradation of compounds in spray-dried samples as measured by thermogravimetric analysis (n ¼ 1). Treatments were composed of pure Cyclopia subternata extract, spray-dried inulin (IN100) and corn syrup solids (CS100), as well as C. subternata extract spray-dried with IN or CS at levels of 250, 500 and 750 g/kg of total solids (IN25, IN50, and IN75; CS25, CS50 and CS75). Treatments

Mass loss due to moisture (g/kg, wet basis)

Mass loss due to degradation (g/kg, wet basis)

Total mass loss (g/kg, wet basis)

Pure extract IN25 IN50 IN75 IN100 CS25 CS50 CS75 CS100

65.8 85.5 38.3 70.2 86.3 67.4 86.9 69.0 75.3

241.7 285.3 336.1 443.5 520.4 299.6 364.2 349.7 305.6

309.0 371.1 390.4 520.5 608.3 368.6 455.7 416.9 388.0

similar to that reported by Schaller-Povolny et al. (2000) for Orafti® HP inulin with an average DP of 23 (the same product used in this study) at an aw of 0.33. No distinct Tg was observed for pure corn syrup solids or any of the spray-dried extract-corn syrup solids mixtures. During the analysis of large molecules, it is common for water to evaporate, causing a broad endotherm, which masks the Tg. Glass

Fig. 4. Differential thermal analysis of (a) pure inulin (IN100), as well as Cyclopia subternata extract microencapsulated with inulin at 250, 500 and 750 g/kg of the total solids (IN25, IN50 and IN75) and (b) pure corn syrup solids (CS100), as well as Cyclopia subternata extract microencapsulated with corn syrup solids at 250, 500 and 750 g/kg of the total solids (CS25, CS50 and CS75).

transition temperatures of 141 and 111.85  C were reported for dry maltodextrin with 20 (Roos & Karel, 1991) and 19 DE (Avaltroni, Bouquerand, & Normand, 2004), respectively. The presence of water affects the Tg of maltodextrin and other biopolymers with Tg decreasing with increasing water content (Avaltroni et al., 2004; Roos & Karel, 1991; Schaller-Povolny et al., 2000). Molecular weight also plays a role, i.e. in a homologous series of polymers an increase in DE (¼ decrease in molecular weight) causes a decrease in Tg (Avaltroni et al., 2004). It is unlikely that samples will be exposed to temperatures as high as those of the Tg of their carriers during storage. However, should these samples be exposed to high RH conditions it could lead to a decrease in their Tg. Storage of the powders at low relative humidity is therefore recommended to prevent phase transition that would lead to caking. 3.3. Relationship between moisture content and water activity Moisture content, aw and the relationship between the two are some of the most significant factors affecting the stability of powdered extracts. The moisture sorption isotherm (MSI) of pure C. subternata extract after spray-drying displayed a distinct sigmoidal curve (Fig. 5a), characteristic of type II isotherms (Labuza & Altunakar, 2008). Samples irreversibly deliquesced at RH levels above 75%. The BET sorption model, applied to the raw and transformed MSI data (Fig. 5a and b), showed R2 values above 0.97 for each of the curves (Table 2). The BET model fitted the MSI data well, particularly at aw < 0.4 (Fig. 5a) as reported for other low aw foods (Labuza & Altunakar, 2008). The calculated monolayer moisture content (M0) values (47.7e60.2 g/kg solids; Table 2) fall within the range of 0.2e0.3 aw (Fig. 5a). The samples containing inulin at 500 and 750 g/kg per total solids had higher M0 values than the corresponding samples containing corn syrup solids. The M0 value refers to the maximum amount of water which can be strongly absorbed to specific sites on the surface of the material and above this value water is more available for chemical reactions and degradation (Labuza & Altunakar, 2008). It thus represents the maximum water content of a food during storage to ensure stability (Labuza & Altunakar, 2008) and depends on the temperature, i.e. decreasing with an increase in temperature. The aw of the samples did not exceed 0.25, with samples containing inulin having lower aw than corresponding samples containing corn syrup solids (Table 2). The moisture content of samples (dry basis) was lower than their M0 values (Table 2) and powders should thus be shelf-stable if stored under the correct conditions. With aw < 0.25, it is recommended that spray-dried honeybush powders be stored below 30% RH. RH conditions higher than this will result in the absorption of moisture, increasing aw levels to the range where physical state changes will start to occur. Hysteresis, i.e. the phenomenon when the adsorption and desorption curves do not follow the same path, was evident to varying degrees. Pure C. subternata extract showed very little indication of hysteresis, while it was particularly noticeable when inulin was present at the highest level (750 g/kg) (Fig. 6). Similarly hysteresis was also observed for samples containing corn syrup solids (Fig. S2; Supplementary Information). The hysteresis loops for the samples containing the same levels of carrier in the mixture were more or less of the same size for the two carriers, indicating that their extent of water binding is more or less the same. The presence of hysteresis in the microencapsulated extract samples is likely an indication that the samples transitioned from a glassy to a rubbery state with the absorption of water. Transition occurs because water acts as a plasticiser which drives the process of glass transition forward by decreasing the Tg. Plasticisation of the polysaccharides occurs when their surface comes into contact with high RH conditions resulting in agglomeration and caking of the

C. Pauck et al. / LWT - Food Science and Technology 75 (2017) 631e639

637

correct conditions to prevent changes in the structure of the powder which cannot be reversed by subsequent drying. 3.4. Compatibility of C. subternata extract with carriers Microcalorimetry was used as a tool to detect incompatibility between the C. subternata extract and the carriers. The heat flow of the mixtures and individual components of the mixtures was measured. From these results a theoretical curve for the mixture was calculated and compared to the measured heat flow of the actual mixture. If a heat flow difference is observed, the components are considered to be potentially incompatible with each other. The interaction curves (difference between theoretical and measured) for IN25, IN50, IN75, CS25 and CS50 were very close to zero, indicating compatibility between the extract and carrier (Fig. S3e8; Supplementary Information). Treatment CS75, however, showed significant interaction with a heat flow difference of 922 mW/g (Fig. S5; Supplementary Information), indicating possible incompatibility between the extract and corn syrup solids when the latter is present at such a high level (750 g/kg). CS75 is, therefore, considered unsuitable as an option for further development into an instant iced tea due to the detected incompatibility as problems could arise during product storage. 3.5. Retention of phenolic compounds

Fig. 5. (a) Moisture sorption isotherm and (b) transformed moisture sorption isotherm data of spray-dried pure Cyclopia subternata extract (filled circles) fitted with the Brunauer, Emmett and Teller (BET) sorption model (dashed line).

, 2003). In most powdered foods this powders (Mathlouthi & Roge occurs in the range of 0.35e0.45 aw (Labuza & Altunakar, 2008). In practical terms it is essential that the powders are stored under the

One of the concerns during the spray-drying process is the degradation of bioactive phenolic compounds (Bott, Labuza, & Oliveira, 2010; Fang & Bhandari, 2011). In the present study the conditions used for spray-drying did not detrimentally affect total polyphenol content, total antioxidant capacity and individual polyphenol content, including the unstable benzophenone, 3-b-Dglucopyranosyliriflophenone (Table 4). Beelders (2016) demonstrated a higher thermal degradation rate constant for this compound, compared to those of the xanthones, mangiferin and isomangiferin. The slight increase in the individual polyphenol content may be attributed to heterogeneous dispersion of phenolic

Fig. 6. Moisture sorption isotherms (n ¼ 1) for spray-dried (a) pure Cyclopia subternata extract, as well as Cyclopia subternata extract microencapsulated with inulin at (b) 250, (c) 500 and (d) 750 g/kg of the total solids (IN25, IN50 and IN75) showing first adsorption (circles), desorption (triangles) and second adsorption (squares) curves at 25  C (RH ¼ relative humidity).

CS75

5.06 c ± 0.02 3.28 d ± 0.01 4.14 d ± 0.02 10.93 de ± 0.10 8.93 c ± 0.02

6.11 c ± 0.02 7.66 cd ± 0.09 12.34 bc ± 0.05 6.21 c ± 0.02 30.6 ab ± 0.03 2378 a ± 79 315.2 b ± 5.4

CS50

5.35 a ± 0.04 3.46 a ± 0.03 4.36 a ± 0.04 11.06 ab ± 0.09 9.12 ab ± 0.15

6.39 a ± 0.03 7.81 ab ± 0.10 12.61 a ± 0.12 6.52 a ± 0.05 3.53 a ± 0.07 2277 a ± 234 309.5 ab ± 13.9

C. Pauck et al. / LWT - Food Science and Technology 75 (2017) 631e639

compounds within the batch of vacuum dried extract used for preparation of the feed solution, as well as minor experimental variation, incurred at different stages of sample preparation. The absence of degradation of the phenolic compounds can be ascribed to the short passage time through the spray-dryer (in the range of a few seconds). The addition of a carrier therefore does not play a crucial role in protecting the phenolic compounds during processing, because the phenolic compounds in the pure extract did not undergo degradation in the first place. It can therefore be concluded that the spray-drying process is suitable for producing C. subternata powders.

6.34 ab ± 0.08 7.85 ab ± 0.07 12.55 a ± 0.14 6.48 a ± 0.10 3.39 b ± 0.11 2447 a ± 52 295.4 ab ± 26.2

Green C. subternata honeybush extract was successfully spraydried producing adequate yields of powders with moisture content below their monolayer moisture content and low water activity, which would ensure stability under the correct storage conditions. Furthermore, the heating process had no detrimental effect on the major phenolic compounds present in the extract, not even the thermally labile xanthones and benzophenones, in the presence or absence of carriers. Extract microencapsulated with inulin had similar powder characteristics to the corn syrup solidsmicroencapsulated extract as assessed using solid-state chemistry methods. The extract and carriers, except at the highest ratio of corn syrups solids in the mixture, were compatible based on the absence of a difference between the calculated and experimental heat flow, implying good storage stability. The prebiotic properties and relatively low energy value of inulin make it a suitable substitute for the production of low energy health products, such as a single-serve “instant” honeybush iced tea. To our knowledge, the stability of xanthones and benzophenones in a plant extract during spray-drying is reported for the first time. Acknowledgements

Means in the same row with the same letter are not significantly different (P  0.05). a Total antioxidants expressed as mmole Trolox/g extract. b Total polyphenols expressed as g gallic acid/100 g extract.

6.39 a ± 0.12 7.75 bc ± 0.06 12.44 ab ± 0.09 6.44 ab ± 0.04 3.44 ab ± 0.02 2422 a ± 85 292.2 a ± 16.4 6.23 bc ± 0.13 7.82 ab ± 0.09 12.54 a ± 0.16 6.39 ab ± 0.18 3.39 b ± 0.1 2387 a ± 209 309.4 ab ± 12.7 5.94 d ± 0.1 7.32 e ± 0.14 11.73 d ± 0.18 6.02 d ± 0.1 3.22 c ± 0.05 2342 a ± 38.139 297.2 ab ± 5.2

6.29 ab ± 0.07 7.91a ± 0.11 12.61 a ± 0.15 6.51 a ± 0.06 3.50 a ± 0.03 2435 a ± 158 296.0 ab ± 9.7

6.26 ab ± 0.00 7.62 d ± 0.06 12.23 c ± 0.07 6.31 bc ± 0.04 3.28 c ± 0.03 2452 a ± 201 309.4 ab ± 4.7

5.28 ab ± 0.07 3.43 ab ± 0.04 4.33 ab ± 0.06 11.23 a ± 0.10 9.11 a ± 0.08 5.19 b ± 0.04 3.33 cd ± 0.02 4.21 cd ± 0.03 10.65 cd ± 0.07 8.82 d ± 0.05 5.28 ab ± 0.04 3.39 abc ± 0.04 4.28 abc ± 0.04 11.50 c ± 0.17 8.95 bc ± 0.09 5.33 a ± 0.05 3.43 ab ± 0.04 4.32 ab ± 0.05 11.68 bc ± 0.15 9.08 ab ± 0.11 5.18 bc ± 0.15 3.37 bc ± 0.09 4.26 bc ± 0.11 10.93 cd ± 0.45 9.12 a ± 0.04

Mangiferin Isomangiferin Vicenin-2 Scolymoside 3-b-D-glucopyranosyl-4-b-Dglucopyranosyloxy-iflophenone Eriocitrin 3-Hydroxy-phloretin-30 ,50 -di-C-hexoside 30 ,50 -di-b-D-glucopyranosylphloretin Hesperidin 3-b-D-glucopyranosyl-iriflophenone Total antioxidant a Total polyphenols b

4.93 d ± 0.08 3.17 e ± 0.05 4.01 e ± 0.07 10.41 e ± 0.19 8.41 e ± 0.03

IN50 Pure extract Before SD

IN25

IN75

CS25

4. Conclusions

Compound

Table 4 Concentration of phenolic compounds (g/kg extract, dry basis; mean ± standard deviation; n ¼ 4) in the powders. Treatments were composed of Cyclopia subternata extract before and after spray-drying (SD), as well as microencapsulated with inulin (IN) or corn syrup solids (CS) at levels of 250, 500 and 750 g/kg of the total soluble solids (IN25, IN50, and IN75; CS25, CS50 and CS75).

638

Financial support for the study was from the Economic Competitive Support Package for Agroprocessing to the ARC by the South African Government, while a student bursary to C.E.P. was provided by the National Research Foundation of South Africa (NRF; grant nr 89584). The NRF grant holder (C.E.P.) acknowledge that opinions, findings, and conclusions or recommendations expressed in any publication generated by the NRF-supported research are those of the authors and that the NRF accepts no liability whatsoever in this regard. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.lwt.2016.10.018. References Arthur, H., Joubert, E., De Beer, D., Malherbe, C. J., & Witthuhn, R. C. (2011). Phenylethanoid glycosides as major antioxidants in Lippia multiflora herbal infusion and their stability during steam pasteurisation of plant material. Food Chemistry, 127, 581e588. Avaltroni, F., Bouquerand, P. E., & Normand, V. (2004). Maltodextrin molecular weight distribution influence on the glass transition temperature and viscosity in aqueous solutions. Carbohydrate Polymers, 58, 323e334. Barclay, T., Ginic-Markovic, M., Cooper, P., & Petrovsky, N. (2010). Inulin - A versatile polysaccharide with multiple pharmaceutical and food chemical uses. Journal of Excipients and Food Chemicals, 1(3), 27e50. Beelders, T. (2016). Xanthones and benzophenones from Cyclopia genistoides (honeybush): Chemical characterisation and assessment of thermal stability. PhD in Food Science dissertation, Stellenbosch, South Africa: Stellenbosch University. Beelders, T., Brand, D. J., De Beer, D., Malherbe, C. J., Mazibuko, S. E., Muller, C. J. F., et al. (2014). Benzophenone C- and O-glucosides from Cyclopia genistoides

C. Pauck et al. / LWT - Food Science and Technology 75 (2017) 631e639 (honeybush) inhibit mammalian a-glucosidase. Journal of Natural Products, 77, 2694e2699. Beelders, T., De Beer, D., Stander, M., & Joubert, E. (2014). Comprehensive phenolic profiling of Cyclopia genistoides (L.) Vent. by LC-DAD-MS and -MS/MS reveals novel xanthone and benzophenone constituents. Molecules, 19, 11760e11790. Bergh, A. J. (2014). Characterisation of the sensory profile of cyclopia intermedia and optimisation of fermentation parameters for improved product quality. MSc (Food Science) thesis, Stellenbosch, South Africa: Stellenbosch University. Bhandari, B. R., & Howes, T. (1999). Implication of glass transition for the drying and stability of dried foods. Journal of Food Engineering, 40, 71e79. Bott, R. F., Labuza, T. P., & Oliveira, W. P. (2010). Stability testing of spray- and spouted bededried extracts of Passiflora alata. Drying Technology, 28, 1255e1265. s-Rojas, D. F., & Oliveira, W. P. (2012). Physicochemical properties of phytoCorte pharmaceutical preparations as affected by drying methods and carriers. Drying Technology, 30, 921e934. Davies, P. (2004). Oral dosage forms. In M. Gibson (Ed.), Pharmaceutical preformulation and formulation: A practical guide from candidate drug selection to commercial dosage form (second ed., pp. 367e430). Boca Raton, FL: Interpharm/ CRC. De Beer, D., Schulze, A., Joubert, E., De Villiers, A., Malherbe, C., & Stander, M. (2012). Food ingredient extracts of Cyclopia subternata (honeybush): Variation in phenolic composition and antioxidant capacity. Molecules, 17, 14602e14624. Dudhia, Z., Louw, J., Muller, C., Joubert, E., De Beer, D., Kinnear, C., et al. (2013). Cyclopia maculata and Cyclopia subternata (honeybush tea) inhibits adipogenesis in 3T3-L1 pre-adipocytes. Phytomedicine, 20, 401e408. Fang, Z., & Bhandari, B. (2011). Effect of spray drying and storage on the stability of bayberry polyphenols. Food Chemistry, 129, 1139e1147. ~ a, J., & Bucala , V. (2015). A comparative study of Gallo, L., Ramírez-Rigo, M. V., Pin spray-dried medicinal plant aqueous extracts. Drying performance and product quality. Chemical Engineering Research and Design, 4, 681e694. Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley, A., & Saurel, R. (2007). Applications of spray-drying in microencapsulation of food ingredients: An overview. Food Research International, 40, 1107e1121. Gross, L. S., Li, L., Ford, E. S., & Liu, S. (2004). Increased consumption of refined carbohydrates and the epidemic of type 2 diabetes in the United States: An

639

ecologic assessment. American Journal of Clinical Nutrition, 79, 774e779. Kolida, S., Tuohy, K., & Gibson, G. R. (2002). Prebiotic effects of inulin and oligofructose. The British Journal of Nutrition, 87(S2), S193eS197. Labuza, T. P., & Altunakar, B. (2008). Water activity prediction and moisture sorption novas, A. J. Fontana, S. J. Schmidt, & T. P. Labuza isotherms. In G. V. Barbosa-Ca (Eds.), Water activity in Foods: Fundamentals and applications, Volume 13 of Institute of Food Technologists Series (pp. 109e154). John Wiley & Sons. , B. (2003). Water vapour sorption isotherms and the caking Mathlouthi, M., & Roge of food powders. Food Chemistry, 82, 61e71. vy, F. (2011). Encapsulation of natural polyphenolic comMunin, A., & Edwards-Le pounds; a review. Pharmaceutics, 3, 793e829. Paramita, V., Iida, K., Yoshii, H., & Furuta, T. (2010). Effect of additives on the morphology of spray-dried powder. Drying Technology, 28, 323e329. Roberfroid, M. B. (1999). Caloric value of inulin and oligofructose. The Journal of Nutrition, 129, 1436Se1437S. Roos, Y., & Karel, M. (1991). Water and molecular weight effects on glass transitions in amorphous carbohydrates and carbohydrate solutions. Journal of Food Science, 56, 1676e1681. Schaller-Povolny, L. A., Smith, D. E., & Labuza, T. P. (2000). Effect of water content and molecular weight on the moisture isotherms and glass transition properties of inulin. International Journal of Food Properties, 3, 173e192. Schulze, A. E., De Beer, D., Mazibuko, S. E., Muller, C. J. F., Roux, C., Willenburg, E. L., … Joubert, E. (2016). Assessing similarity analysis of chromatographic fingerprints of Cyclopia subternata extracts as potential screening tool for in vitro glucose utilisation. Analytical and Bioanalytical Chemistry, 408, 639e649. Schüssele, A., & Bauer-Brandl, A. (2003). Note on the measurement of flowability according to the European Pharmacopoeia. International Journal of Pharmaceutics, 257, 301e304. Sheu, T.-Y., & Rosenberg, M. (1998). Microstructure of microcapsules consisting of whey proteins and carbohydrates. Journal of Food Science, 63, 491e494. Toneli, J., Park, K., Negreiros, A., & Murr, F. (2010). Spray-drying process optimization of chicory root inulin. Drying Technology, 28, 369e379. Vyas, A., Syeda, K., Ahmad, A., Padhye, S., & Sarkar, F. H. (2012). Perspectives on medicinal properties of mangiferin. Mini Reviews in Medicinal Chemistry, 12, 412e425.