Car

Aa

b

c

d

a

ARRA

KAFGMMC

1

M(2tt2bagsRhgc

a

0h

Industrial Crops and Products 51 (2013) 194– 201

Contents lists available at ScienceDirect

Industrial Crops and Products

journa l h om epa ge: www.elsev ier .com/ locate / indcrop

hemometric studies on mineral distribution and microstructurenalysis of freeze-dried Aloe vera L. gel at different harvestingegimens

nirban Raya,∗, S. Dutta Guptaa, Sampad Ghoshb, Shashaank M. Aswathac, Bibek Kabid

Agricultural and Food Engineering Department, Indian Institute of Technology, Kharagpur, Kharagpur 721302, West Bengal, IndiaDepartment of Chemistry, National Institute of Technology, Jamshedpur 831014, Jharkhand, IndiaComputer Science and Engineering Department, Indian Institute of Technology, Kharagpur, Kharagpur 721302, West Bengal, IndiaAdvanced Technology Development Centre, Indian Institute of Technology, Kharagpur, Kharagpur 721302, West Bengal, India

r t i c l e i n f o

rticle history:eceived 13 June 2013eceived in revised form 24 August 2013ccepted 30 August 2013

eywords:loe vera L.reeze-dried gel

a b s t r a c t

This paper presents an analysis of elemental distribution, microstructure at pre- and post-rehydration,and swelling of freeze-dried Aloe vera L. gel, obtained from two, three and four-year-old plant in summer.To explore the seasonal influence, three-year-old plants from rainy and winter season are also consideredfor analysis. Mineral distribution is studied following the estimation of Ca, K, Mg, P, Na, Mn, Se, Al, Fe, Zn,Cu and Cd. A consolidated effort has also been made to study the concentration of total C and N contentof gel at different growth periods. Re-hydration mediated continuous topography in poly-dispersed,unorganized and amorphic freeze-dried gel accounts for the post-rehydration cross-linking among the

rowth periodsineral distributionicrostructure

hemometrics.

component polysaccharides. Chemometric analysis (principal component and cluster analysis) suggeststhat the mineral content of gel change as a function of growth periods of plants. High content of organic Cand Se with maximum water absorbing potential (in the form of swelling activity) are mainly associatedwith three-year-old aloe of summer; the contents of N, Na, Ca, K, Mg, P and Cu are observed to be high intwo-year plants. To the best of our knowledge and belief, this is the first time, Se (a bioactive antioxidant

nt) is

activity promoting eleme

. Introduction

In the present investigation Aloe vera L. (A. vera), native toediterranean region, is a plant from the Xanthorrhoeaceae family

Angiosperm Phylogeny Group III System, 2009; Ray and Aswatha,013), considered as the plant-system to study the mineral dis-ribution at different age groups and harvesting seasons. Variousherapeutic properties of A. vera (Choi and Chung, 2003; Hamman,008; Dutta Gupta, 2010; Ray et al., 2012, 2013) which is attributedy the diverse array of component compounds in gel including

combination of polysaccharides and their derived compounds,lycoproteins, phenolics, enzymes, minerals, amino acids, sterols,aponins, vitamins, etc. (Eshun and He, 2004; Rodriguez et al., 2010;ay and Dutta Gupta, 2013), makes it popular from ancient times of

uman history. To explicate the diverse array of bioactivity of A. verael (AG), Davis (1997) proposed the “Conductor-Orchestra con-ept” that explains the functional reciprocity among more than 200

Abbreviations: A. vera, Aloe vera L.; AG, A. vera gel; FD, reeze-dried.∗ Corresponding author. Tel.: +91 9231695435.

E-mail addresses: anirbanrayiitkgp@gmail.com,nirban@agfe.iitkgp.ernet.in (A. Ray).

926-6690/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.indcrop.2013.08.080

found to be present in aloe gel.© 2013 Elsevier B.V. All rights reserved.

biologically active components present in aloe parenchyma. Pre-sumably, the diverse array of biological activities of AG is expressedby the synergistic action of the component compounds rather thana single component. Different bioactive potentials, moisturizingand excipient activities of AG are exploited for the productionof creams, lotions, soaps, shampoos, facial cleansers, ointments,tablets, capsules and so on, and it appears as a base material ina myriad of cosmeceutical products (Eshun and He, 2004; DuttaGupta, 2010; Rodriguez et al., 2010; Ray and Dutta Gupta, 2013).It is worthy to mention that hypoglycemic potential attributesto AG, of several mineral elements are demonstrated by differ-ent authors (Rajasekaran et al., 2005; Rajasekaran et al., 2006;Rajendran et al., 2007; Hamman, 2008). It has been documentedthat the presence of potassium ions attributes the wound healingactivity to AG (Femenia et al., 1999). Calcium, one of the impor-tant minerals responsible for bone and teeth formation (Ozcan andHaciseferogullari, 2007), is found to be abundant in AG (Yamaguchiet al., 1993). In nutraceutical industry, AG is being used as a mineralsource of different functional foods, and as a supplement in other

food products for the production of various health drinks and bever-ages (Rajendran et al., 2007; Hamman, 2008; Rodriguez et al., 2010).

Most of the characterizations of AG were framed within thephysical and biochemical paradigms during various processing

s and P

teicm2is(ccma(evtp2

hwebefbPn2itAdRAicih

psatega

2

2

teeHMa

2

o

A. Ray et al. / Industrial Crop

echniques (Femenia et al., 2003; Miranda et al., 2009; Guliat al., 2010; Rodriguez-Gonzalez et al., 2011, 2012). The mod-fication of the bioactive substances during processing leads toertain variations in the potential physico-chemical properties andodulates the inherent bio-active potential of AG (Gulia et al.,

010). Among the different dehydration processes, freeze-dryings considered as a potential dehydration method, which con-erves the physico-chemical properties of AG to a greater extentFemenia et al., 2003). Different dehydration methods and pro-essing techniques can modify the micro-structure of gel whichan be delineated by scanning electron microscopy (SEM). Theicro-structure of the AG under different drying temperature

nd processing techniques have been studied by different authorsMiranda et al., 2010; Rodriguez-Gonzalez et al., 2011; Vega-Gálvezt al., 2011); however, the effect of growth periods remains unin-estigated. Apart from the polysaccharides and different processingechniques, mineral concentrations also influence the bioactiveotential and nutritional attributes of A. vera (Femenia et al., 1999,003; Rajasekaran et al., 2005; Miranda et al., 2009).

Different minerals have distinct biological significance in theuman diet, which are categorized as macro and micro nutrients,ith reference to the functional foods. The desired level of min-

rals in the daily diet is necessary to maintain the physiologicalalance and the synchronized functionality of the living cells (Lozakt al., 2002). It has been documented that the presence of differentunctional proteins in A. vera also influences the significance of theioactive potential of gel (Kostalova et al., 2004; Hamman, 2008).rotein content in AG has been estimated as the factor of totalitrogen content at different processing techniques (Miranda et al.,009; Gulia et al., 2010) though the influence of growth periods

s not yet explored. Since the agronomic management informa-ion on A. vera is scarce (Rodriguez-Garcia et al., 2007; Ray andswatha, 2013), and AG has been supplemented in various healthrinks and used as a functional ingredient in food (Hamman, 2008;odriguez et al., 2010), the knowledge of elemental distribution inG in relation to different harvesting regimens would be of help

n selecting the potential growth period of A. vera to optimize theomponents of the value chain of A. vera processing. Therefore its imperative to study the mineral distribution of AG at differentarvesting regimens.

In this context, we present an analysis of the influence of growtheriods of plants on mineral distribution of AG. The chemometrictudies with reference to principal component analysis and clusternalysis have also been employed to delineate the mineral distribu-ion in gel from different growth periods of A. vera. A consolidatedffort has also been made to study the total carbon content, nitro-en content, micro-structure and swelling of freeze-dried (FD) gelt different harvesting regimens.

. Materials and methods

.1. Chemicals

Double deionized water obtained from Milli-Q water purifica-ion system (Millipore) was used in all of the dilutions. The standardlement solutions were prepared by serial dilution of multi-lement standards obtained from Merck (Darmstadt, Germany).NO3, HCl and HClO4, and other reagents were also procured fromerck (Darmstadt, Germany). All reagents and solvents were of

nalytical reagent grade.

.2. Plant materials

The A. vera L. plants were identified by Prof. G.G. Maity, Tax-nomist, University of Kalyani, West Bengal, India, and was grown

roducts 51 (2013) 194– 201 195

in the Agricultural and Food Engineering Departmental farm ofIndian Institute of Technology Kharagpur, India (Ray et al., 2012). A.vera saplings were sown in July–August (2007–2008), under a ran-domized block design with forty replications (Ray and Aswatha,2013). The overall topography of Kharagpur (22.33◦ N, 87.32◦

E) is flat to undulating terrain with lateritic sandy-loamy soil;seasonal changes are evident in Kharagpur throughout the year(Rautaray et al., 2003). On an average, 70–90% relative humidity,about 1800 mm annual rainfall and the velocity of wind ranged at10–25 km/h creates a favorable condition for the growth of differentmedicinal and industrial crops in Kharagpur (Ray et al., 2013).

2.3. A. vera gel harvesting from different growth periods of plants

After 24 ± 1, 36 ± 1 and 48 ± 1 months of growth, the leaves wereharvested during the summer (April–May) and used for extrac-tion of parenchymatous gel. For clarity, the different age group ofplants was hereafter referred to as two, three and four-year-plants.To study the influence of seasons apart from summer, three-year-plants were used to extract the gel during the rainy (July–August)and winter (December–January) season. It is worthy to mentionthat selection of different growth periods was made keeping inview of our previous investigations (Ray and Aswatha, 2013; Rayet al., 2013). Healthy and fresh leaves having a length of 35–45 cmwere harvested from the first whorl position of plants between 8.30a.m. and 9 a.m., and transferred to the laboratory on ice. Harvestedleaves were washed with double distilled water to remove the dirt,and filleted. The translucent foliar parenchyma was blended into asingle composite gel and allowed to freeze overnight at −20 ◦C forsolidification, which was followed by freeze-drying for a consecu-tive 72 h. Pale yellowish FD parenchymatous gel obtained from two,three and four-year-old aloe in summer were assigned as S2, S3and S4, respectively. FD-gel prepared from three-year-old-plantsin rainy and winter season were designated as R3 and W3, respec-tively. Drying by sunlight or by any other convective drying methodhas not been opted to avoid the possible interference by high tem-perature on nutrient mineralization. The FD-AG was used in all ofthe analyses to study different parameters.

2.4. Swelling capacity

Swelling capacity was measured as increased bed volumeafter equilibration in excess buffer (Femenia et al., 2003). FD-gel(100 mg) was weighed into a graduated conical glass tube withan excess of phosphate buffer and allowed to stand for 12–16 hfor optimum swelling. The suspensions were stirred and allowedto settle down to diminish the void volume between the adja-cent swollen gel layers. After equilibration the swelled volume wasrecorded and expressed as ml/100 mg of FD-gel (ml/100 mg).

2.5. Microstructural analysis with SEM, and detection of elementsby SEM-EDS

FD-Gel samples were positioned on stub prior to gold sputtering.The raster scan profiles of FD-AG were captured by JEOL JSMS 5800Scanning Electron Microscope (Japan) at 300× magnification undera pressure of 40 Pa and accelerating voltage of 20 kV. The character-istic spectrum of energy dispersive X-ray spectroscopy (EDS) wascaptured with Oxford EDS detector coupled with SEM instrumentfor the preliminary detection of the predominant elements presentin FD-AG prior to the quantitative estimation.

2.6. XRD and ATR-FTIR spectroscopy

FD-gel from the potential growth period was analyzed with PW1710 X-ray defractometer (Philips, Holland) with Co target and Fe

1 s and Products 51 (2013) 194– 201

fivn

iAtv

2v

2

debia

2

maepw(tbo

2

wricuv

btAfidsnMfg

3

3

aO6st

psg

96 A. Ray et al. / Industrial Crop

lter, under 20 mA current with 40 kV of operating power source toalidate the amorphicity of gel as evident in SEM analysis. The scan-ing rate was adjusted at 6◦/min with scanning scope of 10–60◦.

ATR-FTIR was conducted to detect the functional group presentn gel following the method of our previous study (Ray andswatha, 2013). The spectrum (4000–450 cm−1) was recorded at

he resolution of 4 cm−1 and plotted percent transmittance (%T)ersus wave-number (cm−1).

.7. Elemental analysis of gel at different growth periods of A.era:

.7.1. Estimation of total carbon (C) and nitrogen (N) contentThe N content of FD-gel was estimated by the modified Kjeldahl

istillation method (Miranda et al., 2009). Organic C content wasstimated using Walkley and Black method by the oxidation of car-on in gel with dichromate ion followed by titration with ferrous

on (Rautaray et al., 2003). Total C and N content were expressed as percentage (%) of FD-gel weight.

.7.2. Determination of mineral contentConcentrations of Ca, Mg, Cu, Zn, Fe, Al, Mn, Cd and Se were esti-

ated by atomic absorption spectroscopy (GBC 932 AA, Australia)fter the digestion of an H2SO4, HNO3, and HClO4 mixture (Mirandat al., 2009). Na and K content were estimated by using flamehotometer (Flamephotometer 128 Systronics, India). P contentas estimated by phospho–vanadium–molybdenum blue method

Swain et al., 2007) at 466 nm (UV-1800 UV–VIS Shimadzu Spec-rophotometer, Japan). Each of the concentrations was calculatedy means of a calibration curve, and expressed as �g mineral per gf FD-gel (�g/g).

.8. Statistical analysis and chemometrics

Three different gel samples from each of the growth periods,hich were analyzed individually in triplicate, and data were

eported as mean ± SD. Analysis of variance and Fisher’s least signif-cant difference (LSD) test at 95% confidence interval (p ≤ 0.05) wereonducted to identify significant differences among the meanssing MSTAT-C software package (ver. 1.41, Michigan State Uni-ersity).

Chemometric studies (multivariate statistical technique) haveeen carried out with mineral concentration data by consideringhe techniques of the principal component and cluster analysis.

principal component analysis (PCA) was performed at 95% con-dence interval to analyze the mineral distribution in gels fromifferent growth periods. In cluster analysis, we consider the AGamples from different growth periods, with reference to theirormalized mineral profiles; normalization was carried out withATLAB (ver. 7.0). Similarity between the gel samples obtained

rom different growth periods was depicted by using the dendo-ram with Euclidean distance as the considered measure.

. Results and discussion

.1. Swelling capacity and microstructure of freeze-dried gel

FD-AG exhibits significant variation in water holding potentials evidenced from the swelling of gel at various growth periods.f the tested gel samples, the swelling varied from 5.67 ± 0.42 to.30 ± 0.26 ml/100 mg and S3 showed significantly higher (p ≤ 0.05)welling than S2 (Fig. 1). Seasons did not have any significant con-ribution toward the swelling activity of AG.

SEM micrographs of the FD-gel revealed the irregular, amor-hous structure with various grooves and fissures on the exteriorurface of the gel granules (Fig. 2). Subtle variation was noted in theel microstructure at different growth periods. FD-gel micrographs

Fig. 1. Swelling of freeze-dried (FD) gel at different growth periods of A. vera L. Barswith the same small letter did not share significant differences at p ≤ 0.05 (Fisher’sleast significant difference test).

varied from compact granular to etched-like protrusions. A fibrillarappearance with a number of switchers was noticed in S2, whereasthe micrograph of S3 appeared to be flake like and overall topog-raphy was smooth with undulating depressions. S4 appeared to begranular with wavy surface and ridges. Smooth texture with luster,and friable particles were observed in R3. Small chinks or crannieswere evident in W3. As the knowledge of the structure–functionrelation needs more insight into the physical and chemical proper-ties of AG, SEM assisted topographic study has also been performedto analyze the AG structures after re-hydration at the micron levelfor the conspicuous visualization of the topographic organizationpertained to the reconstitution of gel. After re-hydration of gel, acontinuous texture with elevated topography was noticed. A uni-form and organized topography of re-hydrated gel may account forthe reconstitution mediated inter-molecular cross-linking of thecomponent polysaccharides. The presence of different hydrophilicfunctional groups in aloe hydro-gel viz. phenolic OH (at around3422.05 cm−1), CO (at around 1609.33 cm−1), COO (at around1419.69 cm−1), COC (at around 1080.43 cm−1), as evinced by theFTIR analysis (Fig. 3a), facilitate the intermolecular cross-linking byH-bond formation in aqueous environment. The presence of differ-ent polar functional groups in AG is in agreement with our previousfindings (Ray et al., 2012, 2013; Ray and Aswatha, 2013) and withthe reports of others (Femenia et al., 2003; Rodriguez-Gonzalezet al., 2011).

It has been documented that processing of AG may modify theorganized structure of parenchyma by breaking down the primaryand secondary cell wall structure, and attributes the unorga-nized and poly-dispersed structure to parenchymatous gel matrix(Miranda et al., 2010; Rodriguez-Gonzalez et al., 2011; Vega-Gálvezet al., 2011; Ray and Aswatha, 2013). The amorphicity of the FD-gelhas been validated by XRD analysis, and representative XRD profileof S2 is shown in Fig. 3b. The disappearance of the crystalline peakin XRD profile corroborates the conviction of amorphicity of FD-AG,which is also evident in rest of the growth periods.

3.2. Total organic C and N content

Total C content was influenced by the age of the plants andranged from a high of 28.23 ± 0.37% to a low of 24.92 ± 0.34%(Fig. 4a). S3 possessed significantly higher (p ≤ 0.05) content of

organic C than those of S2 and S4. Total C content varied signifi-cantly (p ≤ 0.05) with the season, and declined level of C contentwas noticed in rainy season. N content varied from 0.50 ± 0.004 to0.67 ± 0.007% (Fig. 4b). Significant variation (p ≤ 0.05) in N content

A. Ray et al. / Industrial Crops and Products 51 (2013) 194– 201 197

ds (S2

waSNw

mAgpbf

Fig. 2. SEM micrographs of FD-AG from different growth perio

as noted among the FD-gel at different growth stages. High Nccumulation was noted in S2 followed by the other age groups.easonal influence was also apparent in N concentration, and high

content was found in rainy season, followed by the summer andinter.

In any carbohydrate molecule, carbon is present as one of theajor possible structural components. Higher carbon content inG at three-year-old plants, as estimated in the present study, is in

ood agreement with the proclamation that, three-year-old A. veralants are rich in polysaccharide content (Hu et al., 2003). It haseen documented that age of the plants and seasons plays a decisiveactor by which the metabolic flux was modulated in such a way, the

, S3, S4, R3 and W3) of plants, and re-hydrated gel of S3 (RH).

rate of photosynthesis, carbon sequestration, carbon partitioning,as well as the rate of utilization of carbon containing metabolitesvaries (Paez et al., 2000; Ray and Aswatha, 2013; Ray et al., 2013).Carbon assimilation in plant tissue depends on the rate of photo-synthesis, carbon accumulation and subsequent influence on masscarbon transport from the plant leaves (Ho, 1975). Presumably, thechange of photosynthetic rate in relation to plant growth periodsmay influence the carbon sequestration in the foliar gel of A. vera.

A different group of workers estimated protein content in AGas a factor of total N (Gulia et al., 2010; Miranda et al., 2010),and present estimation with N content is comparable with pre-vious findings. Different amino acids (Alanine, arginine, aspartic

198 A. Ray et al. / Industrial Crops and Products 51 (2013) 194– 201

profi

altedogaflhtgaasot

3

eaaott

Fs

Fig. 3. Representative ATR-FTIR spectrum (a) and XRD

cid, glutamic acid, glycine, histidine, hydroxyproline, isoleucine,eucine, lysine, methionine, phenylalanine, proline, threonine,yrosine, valine), glycol-proteins (aloctin, alprogen, etc.) and sev-ral enzymes (bradykinase, carboxypeptidase, catalase, superoxideismutase, glutathione peroxidase) are the predominant sourcef protein in AG (Rodriguez et al., 2010). Total N concentration inel has been influenced by the concentrations of inherent aminocids and proteins varying at different harvesting regimens. Theuctuation in the N content in relation to the ages of the plantsas been reported by different authors. It has been observed thathe reduction of the N concentration in plants occurs as the plantset older due to the dilution effect of enhanced shoot growth at

constant level of N availability in soil (Bergmann, 1992; Graeffnd Claupein, 2003). Although, S4 has been demonstrated withignificantly higher (p ≤ 0.05) N content compared to S3, S2 isbserved with significantly highest concentration (p ≤ 0.05) of Nhan the other age groups.

.3. Mineral analysis

In the present study the concentration of different macro-lements (Ca, Mg, K and P), micro-elements (Mn, Fe, Zn and Cu)nd non-essential elements (Na, Se, Al, and Cd) were determined

t different growth periods of A. vera (Fig. 5a–c). The presencef different elements in AG was validated by SEM-EDS prior tohe quantitative analysis. Representative SEM-EDS profile withhe detection of aforementioned minerals is shown in Fig. 6.

ig. 4. (a) Total organic carbon content of FD-gel at different growth periods of A. vera L. (ame small letter did not share significant differences at p ≤ 0.05 (Fisher’s least significan

le (b) of FD-gel at two-year-old A. vera L. in summer.

Comparative differences were evident in all of the elementsby quantitative estimation method, which are influenced bythe pertained age groups. Apart from the age of the plants, gelharvesting season has also the significant (p ≤ 0.05) contributiontoward the distribution of minerals in AG. The concentrationsof Ca, K, Mg, P and Na were found to be higher as comparedto the rest of the elements; K was found to be highest in S2(36133.33 ± 251.66 �g/g). The concentration of Ca, K, Mg andP ranged from 24163.33 ± 159.48 to 14103.33 ± 115.04 �g/g,36133.33 ± 251.66 to 4796.67 ± 25.17 �g/g, 28402.67 ± 200.05to 22166.67 ± 208.17 �g/g and 23036.67 ± 49.33 to5440.00 ± 40.00 �g/g, respectively. Among the different growthperiods, the content of Se varied from a high of 752.83 ± 30.40 �g/gto a low of 613.33 ± 20.21 �g/g, and found to be high in S3 followedby S4, R3, W3 and S2. To the best of our knowledge and belief, thisis the first time information of Se found to be present in A. vera.The concentration of Mn, Al and Fe varied from 1248.67 ± 20.03to 794.77 ± 6.24 �g/g, 1284.80 ± 16.95 to 717.00 ± 12.64 �g/g,338.60 ± 18.10 to 233.53 ± 3.14 �g/g, respectively. Cd contentranged from 1.37 ± 0.21 �g/g to 3.37 ± 0.25 �g/g, was the leastwith reference to the different elements analyzed. A variation inthe concentration of different elements at different harvesting reg-imens is in good agreement with our previous semi-quantitative

analysis of minerals at different growth periods of A. vera (Ray andAswatha, 2013).

In the present quantitative estimation, mineral fingerprintingrequired chemometric description of the complexity, resulting

b) nitrogen content in FD-AG from different growth periods of plants. Bars with thet difference test).

A. Ray et al. / Industrial Crops and Products 51 (2013) 194– 201 199

Fig. 5. Mineral contents (�g/g) of FD-gel at different growth periods of A. vera L.(a) macro-elements (Ca, K, Mg and P) and Na; (b) microelements (Mn, Fe, Zn andCu); and (c) Se, Al and Cd. Values (mean ± SD) are average of three different samplesof each growth periods, analyzed individually in triplicate (n = 1 × 3 × 3). Differentletters on the bars with same mineral indicate the significant differences at p ≤ 0.05(Fisher’s least significant difference test).

Fig. 6. EDS spectrum of FD-gel from two-year-old A. vera L. in summer season (S2).

from the simultaneous acquisition of mineral analysis data at differ-ent harvesting seasons and age groups. The elemental distributionin gel from different periods of plant growth was illustrated withthe help of PCA (as one of the chemometric discrimination method).The eigenvalues of first nine principal components and the percent-age of variance explained by them for mineral analysis with gelsamples are presented in Table 1. From the cumulative variance, itis observed that first four PCs could explain 98.68% of the variance.Since much of the information is retained in the first four PCs, thesePCs are enough to analyze the overall variations in the considereddata samples. In PCA, each of the data samples was projected overPC1–PC2 axes and PC3–PC4 axes, and their corresponding bi-plots(shown in Fig. 7a and b, respectively) were analyzed to describe themineral distribution. The gel samples representing various growthperiods are presented in the bi-plots, which shows the overall dis-tribution of minerals present in AG at different growth periods ofaloe. A correlation was observed between various minerals, withthe correlation values of the first and second principal components(PC1 and PC2) being 42.29 and 27.30%, respectively. The score ofthe principal components described 69.59% of the variance in AGmineral distribution. In the bi-plot, PC1 correlated positively withNa, Mn, Zn, Se and Cd, whereas PC2 positively correlated with Ca,Mg, K and Cu and negatively with P and Fe. Gel samples from S2were clustered around the elements namely, K, Mg, Ca and Cu. Thedistribution of Zn, Se and Cd was more predominant in S3. The gelsamples assigned as S4 clustered around Al, and the samples pre-pared from three-year old aloe in winter (W3 samples) were inthe close proximity with Mn and Fe. Compared to PC1–PC2 projec-tions, a lower percentage of variance of various minerals could beexplained by the third and fourth PCs, which is found to be 18.50%and 10.59%, respectively. Therefore, PC3–PC4 combination couldexplain 29.09% of the total variance in the pertained mineral dis-

tribution. As shown in Fig. 7b, PC3 correlates positively with Ca, Se,Mg, K, P and Zn, whereas PC4 correlates with all of the mineralstested, except Zn and Cu. With reference to PC3–PC4 bi-plot, FD gelsamples from S2 and R3 are clustered around the elements Zn and

Table 1Eigenvalues of first nine principal components and the percentage of varianceexplained by them for mineral analysis with A. vera gel samples.

PCs Eigenvalues Percentage of variance Cumulative (%)

1 5.0746 42.29 42.292 3.27659 27.30 69.593 2.21963 18.50 88.094 1.27133 10.59 98.685 0.07994 0.67 99.356 0.05741 0.48 99.837 0.0098 0.08 99.918 0.00617 0.05 99.969 0.00401 0.03 100.00

200 A. Ray et al. / Industrial Crops and Products 51 (2013) 194– 201

F on elep

Cat

fdwcsatdgcocsiiafs

vhe

Ff

ig. 7. Principal component loading bi-plots of PC1–PC2 (a) and PC3–PC4 (b) based

eriod of A. vera L.

u, respectively. The gel samples of S4 are found to be clusteredround Al. Also, the prominence of Se in aloe samples is justified inhe bi-plot of PC3–PC4.

Cluster analysis was performed with the AG samples obtainedrom different growth periods with reference to their mineralistribution. Complete linkage cluster analysis was performedith normalized dataset and the Euclidean distance measure was

onsidered to analyze the underlying similarities among the gelamples pertained to different growth periods. A hierarchicalgglomerative based data clustering method was used to formhe clusters. The results obtained have been depicted as a den-ogram in Fig. 8. Four main clusters corresponding to each of therowth periods has been found in the present analysis. The firstluster was composed of the gel samples from S3 and S4. The sec-nd cluster was composed of gel samples from R3. Third one wasomposed of the second cluster along with W3, and last cluster con-tituted the gel samples from S2. The main clusters were dividednto subgroups, and each of the subgroups was connected with theirnherent growth periods. The distance among the gel samples fromny particular growth period were marginal as compared to dif-erent growth periods, accounting to close mineral distribution atimilar growth periods.

A considerable number of works on elemental analysis of A.era and bioactive potential of AG in relation to mineral contentave been reported in literature (Femenia et al., 1999; Rajasekarant al., 2005; Miranda et al., 2009; Rodriguez et al., 2010). A high

ig. 8. Dendogram of cluster analysis based on of the mineral distribution in FD-AGrom different growth periods of plants.

mental analysis data showing the mineral distribution in FD-gel at different growth

concentration of Ca in AG as determined in the present investi-gation is in good agreement with the proclamation that A. verabelongs to a group of calcium-rich vegetables (Yamaguchi et al.,1993). Presence of Se in AG has hardly been discussed in liter-ature. Se has been described as an indispensable element withunique anti-oxidative potential required to sustain the antioxi-dant defense mechanism in the human body. Deficiency of Sein regular diet may affect the formation of antioxidant enzymes,leading to an imbalance in the antioxidant-defense mechanismin the human body system (Burk, 2002). The present work anal-ysis reports the presence of Se in AG which makes it effective fornutraceutical applications. It is worthy to mention that the increas-ing relative abundance of divalent Ca ions decreases the negativecharge density of the polyanionic cell wall matrix to a higher extent,compared to the effect of the monovalent cations; diffuse doublelayer cell-wall matrix in divalent ion system exists in a compressedform, resulting in the decrease of the swelling pressure in cell-wall matrix (Shomer et al., 1991). It has been reported that calciumions have the potential to restrict the cell wall swelling by form-ing the bridges between galacturonic acid and adjacent units ofpectic chains (Miranda et al., 2009). Apart from the concentra-tion of polysaccharides in AG, the higher level of Ca2+ content atyounger age group of plants (S2) also restrict the swelling activityof the FD-AG. Investigation of the influence of in situ and ex situmineral concentration on aloe parenchyma with reference to theswelling and reconstitution of dehydrated AG can be a scope offuture work.

The influence of growth periods on the concentration of theelements estimated in the present investigation by chemometricssuggests the selective accumulation of minerals change in func-tion of the growth periods. It is evident from the present studythat the content of essential elements in AG is conditional and theconcentration can be modulated by the physiological ability of theplants to selectively accumulate the elements; similar convictionhas also been proposed by others (Mirosławski et al., 1995; Lozaket al., 2002). Presumably, the variation in the elemental concen-tration at different harvesting regimens indicates the differentialrequirement of elements necessary for plant growth, which shouldbe taken in account at the outset of AG processing.

4. Conclusion

The present study shows that A. vera gel is a potential source ofdifferent mineral components. The contents of different minerals

of A. vera L. gel change as a function of age of the plants and climaticseasons. Unorganized, dispersed and amorphous microstructure offreeze-dried gel accounts for the degradation of compact parenchy-matous cell-wall structures by sublimation at freeze-drying.

s and P

Pcsdyafdtitw

A

cIPIa

R

B

BC

DD

E

F

F

G

G

H

H

H

K

L

M

A. Ray et al. / Industrial Crop

ost-rehydration cross-linking between the component polysac-harides render a uniform topography in the microstructure ofwelled-gel. A maximum accumulation of Ca, K, Mg and P wereetected in the younger age group (here two-year-old aloe). Three-ear-old A. vera possessed higher content of total organic C and Se,nd exhibited maximum swelling as well, whereas, N content wasound to be high at two-year-old aloe. The knowledge of mineralistribution in aloe gel at different periods of growth can be utilizedo optimize the component of the value chain on A. vera process-ng with reference to the pertained usage of the foliar aloe gel ashe functional ingredient in food and in cosmeceutical applicationsith desired level of necessary mineral components.

cknowledgments

This research was partly supported by Indian Council of Agri-ultural Research–National Agricultural Innovation Project (NAIP),ndia (Name of the scheme: Value chain on Aloe vera processing). Dr..B.S. Bhadoria, Head of Agricultural and Food Engineering Dept.,ndian Institute of Technology, Kharagpur, West Bengal, India iscknowledged for providing the necessary facilities for publication.

eferences

ergmann, W., 1992. Nutritional Disorders of Plants—Development, Visual and Ana-lytical Diagnosis. Gustav Fischer Verlag, New York.

urk, R.F., 2002. Selenium, an antioxidant nutrient. Nutr. Clin. Care 5 (2), 75–79.hoi, S., Chung, M.H., 2003. A review on the relationship between Aloe vera compo-

nents and their biologic effects. Semin. Integr. Med. 1 (1), 53–62.avis, R.H., 1997. Aloe vera: A Scientific Approach. Vantage Press, New York.utta Gupta, S., 2010. Cosmetic attributes of Aloe vera L. gel. Med. Arom. Plant Sci.

Biotechnol. 4, 10–18.shun, K., He, Q., 2004. Aloe vera: a valuable ingredient for the food, pharmaceutical

and cosmetic industries – a review. Crit. Rev. Food Sci. Nutr. 44, 91–96.emenia, A., Garcıa-Pascual, P., Simala, S., Rossello, C., 2003. Effects of heat treatment

and dehydration on bioactive polysaccharide acemannan and cell wall polymersfrom Aloe barbadensis Miller. Carbohyd. Polym. 51, 397–405.

emenia, A., Sanchez, E.S., Simala, S., Rosello, C., 1999. Compositional features ofpolysaccharides, from Aloe vera (Aloe barbadensis Miller) plant tissues. Carbohyd.Polym. 39, 109–117.

raeff, S., Claupein, W., 2003. Quantifying nitrogen status of corn (Zea mays L.) inthe field by reflectance measurements. Eur. J. Agron. 19 (4), 611–618.

ulia, A., Sharma, H.K., Sarkar, B.C., Upadhyay, A., Shitandi, A., 2010. Changes inphysico-chemical and functional properties during convective drying of Aloevera (Aloe barbadensis) leaves. Food. Bioprod. Process 88, 161–164.

amman, J.H., 2008. Composition and applications of Aloe vera leaf gel. Molecules13, 1599–1616.

o, L.C., 1975. Variation in the carbon/dry matter ratio in plant material. Ann. Bot.40, 163–165.

u, Y., Xu, J., Hu, Q., 2003. Evaluation of antioxidant potential of Aloe vera (Aloebarbadensis Miller) extracts. J. Agric. Food Chem. 51, 7788–7791.

ostalova, D., Bezakova, L., Oblozinsky, M., Kardosova, A., 2004. Isolation and char-acterization of active compounds from Aloe vera with possible role in skinprotection. Ceska Slov. Farm. 53 (5), 248–251.

ozak, A., Soltyk, K., Ostapczuk, P., 2002. Determination of selected trace elementsin herbs and their infusions. Sci. Total Environ. 289, 33–40.

iranda, M., Maureira, H., Rodriguez, K., Vega-Galvez, A., 2009. Influence of tem-perature on the drying kinetics, physicochemical properties, and antioxidantcapacity of Aloe vera (Aloe barbadensis Miller) gel. J. Food Eng. 91, 297–304.

roducts 51 (2013) 194– 201 201

Miranda, M., Vega-Gálvez, A., García, P., Di Scala, K., Shi, J., Xue, S., Uribe, E., 2010.Effect of temperature on structural properties of Aloe vera (Aloe barbadensisMiller) gel and Weibull distribution for modelling drying process. Food Bioprod.Process. 88 (2), 138–144.

Mirosławski, J., Wiechula, D., Kwapulinski, J., Rochel, R., Loska, K., Ciba, J., 1995. Theoccurrence of Pb, Cd, Cu, Mn, Ni, Co and Cr in selected species of medicinal plantsin Poland. Bromat. Chem. Toksykol. 28, 363–372.

Ozcan, M.M., Haciseferogullari, H., 2007. The strawberry (Arbutus unedo L.) fruits:chemical composition, physical properties and mineral contents. J. Food Eng. 78,1022–1028.

Paez, A., Gebre, G.M., Gonzalez, M.E., Tschaplinski, T.J., 2000. Growth, soluble carbo-hydrates, and aloin concentration of Aloe vera plants exposed to three irradiancelevels. Environ. Exp. Bot. 44 (2), 133–139.

Rajasekaran, S., Sivagnanam, K., Subramanian, S., 2005. Mineral contents of Aloe veraleaf gel and their role on streptozotocin-induced diabetic rats. Biol. Trace Elem.Res. 108, 185–195.

Rajasekaran, S., Ravi, K., Sivagnanam, K., Subramanian, S., 2006. Beneficial effectsof Aloe vera leaf gel extract on lipid profile status in rats with streptozotocindiabetes. Clin. Exp. Pharmacol. Physiol. 33, 232–237.

Rajendran, A., Narayanan, V., Gnanavel, I., 2007. Study on the analysis of trace ele-ments in Aloe vera and its biological importance. J. Appl. Sci. 3 (11), 1476–1478.

Rautaray, S.K., Ghosh, B.C., Mittra, B.N., 2003. Effect of fly ash, organic wastes andchemical fertilizers on yield, nutrient uptake, heavy metal content and residualfertility in a rice–mustard cropping sequence under acid lateritic soils. Bioresour.Technol. 90, 275–283.

Ray, A., Aswatha, S.M., 2013. An analysis of the influence of growth periods on phys-ical appearance, and acemannan and elemental distribution of Aloe vera L. gel.Ind. Crop. Prod. 48, 36–42.

Ray, A., Dutta Gupta, S., 2013. A panoptic study of antioxidant potential offoliar gel at different harvesting regimens of Aloe vera L. Ind. Crop. Prod.,http://dx.doi.org/10.1016/j.indcrop.2013.09.003.

Ray, A., Dutta Gupta, S., Ghosh, S., 2012. Isolation and characterization ofpotent bioactive fraction with antioxidant and UV absorbing activ-ity from Aloe barbadensis Miller gel. J. Plant Biochem. Biotechnol.,http://dx.doi.org/10.1007/s13562-012-0178-2.

Ray, A., Dutta Gupta, S., Ghosh, S., 2013. Evaluation of Anti-oxidative activity andUV absorption potential of the extracts of Aloe vera L. gel from different growthperiods of plants. Ind. Crop. Prod. 49, 712–719.

Rodriguez, E.R., Martín, J.D., Romero, C.D., 2010. Aloe vera as a functional ingredientin foods. Crit. Rev. Food 50 (4), 305–326.

Rodriguez-Garcia, R., Rodriguez, D.J., Gil-Marin, J.A., Angulo-Sanchez, J.L., Lira-Saldivar, R.H., 2007. Growth, stomatal resistance, and transpiration ofAloe vera under different soil water potentials. Ind. Crop. Prod. 25 (2),123–128.

Rodriguez-Gonzalez, V.M., Femenia, A., Gonzalez-Laredo, R.F., Rocha-Guzman,N.E., Gallegos-Infante, J.A., Candelas-Cadillo, M.G., Ramírez-Baca, P., Simal, S.,Rossello, C., 2011. Effects of pasteurization on bioactive polysaccharide aceman-nan and cell-wall polymers from Aloe barbadensis Miller. Carbohyd. Polym. 86,1675–2168.

Rodriguez-Gonzalez, V.M., Femenia, A., Minjares-Fuentes, R., Gonzalez-Laredo, R.F.,2012. Functional properties of pasteurized samples of Aloe barbadensis Miller:optimization using response surface methodology. LWT – Food Sci. Technol. 47(2), 225–232.

Shomer, I., Frenkel, H., Polinger, C., 1991. The existence of a diffuse electric layer atcellulose fibril surfaces and its role in the swelling mechanism of parenchymaplant cell-walls. Carbohyd. Polym. 16, 199–210.

Swain, D.K., Rautaray, S.K., Ghosh, B.C., 2007. Alkaline coal fly ash amendments arerecommended for improving rice-peanut crops. Acta Agric. Scand. B: Soil PlantSci. 57 (3), 201–211.

Vega-Gálvez, A., Uribe, E., Perez, M., Tabilo-Munizaga, G., Vergara, J., Garcia-Segovia,P., Lara, E., Scala, D.K., 2011. Effect of high hydrostatic pressure pretreat-

ment on drying kinetics, antioxidant activity, firmness and microstructureof Aloe vera (Aloe barbadensis Miller) gel. LWT – Food Sci. Technol. 44 (2),384–391.

Yamaguchi, I., Mega, N., Sanada, H., 1993. Components of the gel of Aloe vera (L.)burm. f. Biosci Biotechnol Biochem. 57 (8), 1350–1352.