Embed Size (px)
Citation preview
Ultrasonics Sonochemistry 21 (2014) 1470–1476
Contents lists available at ScienceDirect
Ultrasonics Sonochemistry
journal homepage: www.elsevier .com/ locate/ul tson
Ultrasonic extraction of steroidal alkaloids from potato peel waste
http://dx.doi.org/10.1016/j.ultsonch.2014.01.0231350-4177/� 2014 Elsevier B.V. All rights reserved.
⇑ Corresponding author.E-mail addresses: [email protected] (M.B. Hossain), nigel.brunton@
ucd.ie (N.P. Brunton).1 Tel.: +353 7162807.
Mohammad B. Hossain a,⇑, Brijesh K. Tiwari a, Nirupama Gangopadhyay a, Colm P. O’Donnell b,Nigel P. Brunton c,1, Dilip K. Rai a
a Teagasc, Ashtown Food Research Centre, Ashtown, Dublin 15, Irelandb School of Biosystems Engineering, University College Dublin, Dublin 4, Irelandc School of Agriculture and Food Science, University College Dublin, Dublin 4, Ireland
a r t i c l e i n f o a b s t r a c t
Article history:Received 2 September 2013Received in revised form 8 January 2014Accepted 25 January 2014Available online 15 February 2014
Keywords:UltrasoundExtractionSteroidal alkaloidsResponse surface methodology
Potato processors produce large volumes of waste in the form of potato peel which is either discarded orsold at a low price. Potato peel waste is a potential source of steroidal alkaloids which are biologicallyactive secondary metabolites which could serve as precursors to agents with apoptotic, chemopreventiveand anti-inflammatory properties. The present study investigated the relative efficacy of ultrasoundassisted extraction (UAE) and solid liquid extraction (SLE) both using methanol, to extract steroidal alka-loids from potato peel waste and identified optimal conditions for UAE of a-solanine, a-chaconine, sola-nidine and demissidine. Using response surface methodology optimal UAE conditions were identified asan amplitude of 61 lm and an extraction time of 17 min which resulted the recovery of 1102 lg steroidalalkaloids/g dried potato peel (DPP). In contrast, SLE yielded 710.51 glycoalkaloid lg/g DPP. Recoveries ofindividual glycoalkoids using UAE yielded 273, 542.7, 231 and 55.3 lg/g DPP for a-solanine, a-chaconine,solanidine and demissidine respectively. Whereas for SLE yields were 180.3, 337.6, 160.2 and 32.4 lg/gDPP for a-solanine, a-chaconine, solanidine and demissidine respectively. The predicted values fromthe developed second order quadratic polynomial equation were in close agreement with the experimen-tal values with low average mean deviation (E < 5%) values. Predicted models were highly significant(p < 0.05) for all parameters studied. This study indicates that UAE has strong potential as an extractionmethod for steroidal alkaloids from potato peel waste.
� 2014 Elsevier B.V. All rights reserved.
1. Introduction
Food processing industries particularly potato-crisps manufac-turing industries generate a huge volume of potato peel as by-product. Industrial processing generates between 70 and 140 thou-sand tons of peels worldwide annually [1]. This by-product is usu-ally discarded causing environmental concern or used as low valueanimal feed. This massive amount of waste offers significant eco-nomic potential for creative uses other than animal feeds or fertil-izers. Moreover, the discarded potato peel represents a severedisposal problem to the potato industry as the wet peels are proneto microbial spoilage [2]. Utilization of potato peel for the extrac-tion of steroidal alkaloids will reduce if not eliminate the disposalproblem while paving the way for a new phyto-pharmaceuticalindustry.
Potato constitutes a very important part in human diet in manycountries of the world. The tuber is a good source of carbohydrates,high quality proteins, antioxidative polyphenols [3], vitamins andminerals [2]. However, it also contains a group of toxic compoundsknown as steroidal alkaloids which are largely concentrated in thepeels [4]. Steroidal alkaloids are secondary metabolites presentmainly in the plants of Solanaceae family that have been associatedwith defence against bacterial [5], fungal [6] and insect attacks [7].Symptoms associated with glycoalkaloid toxicity are colic pain inthe abdomen and stomach, gastroenteritis, diarrhoea, vomiting, fe-ver, rapid pulse, low blood pressure, and neurological disorders[8,9]. These adverse effects on human health could be attributedto the anticholinesterase activity and membrane disruptionproperties of steroidal alkaloids affecting mainly digestive andcentral nervous systems [10–12]. In vitro experiments showed thata-solanine and particularly a-chaconine are potent cytotoxins(IC50 = 4.1 lM), acting rapidly to induce cell lysis [13]. Howeverin addition to their toxic effects, studies in the last decade havedemonstrated that these compounds may possess beneficial prop-erties such as anticancer and anti-inflammatory effects, depending
M.B. Hossain et al. / Ultrasonics Sonochemistry 21 (2014) 1470–1476 1471
on dose and conditions of use [14,15]. Therefore, maintaining oreven enhancing these bioactivities but lowering the cytotoxicityto normal dividing cells by chemical modifications of the steroidalalkaloids could be a useful strategy to use the steroidal alkaloidsfor pharmaceutical applications. For this purpose efficient extrac-tion techniques to recover sufficient quantities of starting materialfor chemical modification are required particularly for the alka-mine solanidine (aglycone) as evidence indicates that many ofthe toxic effects of these compounds are related to the presenceof glycosidic residues.
A number of techniques are available for the extraction of sec-ondary metabolites from plants, including ultrasound-assistedextraction, supercritical fluid extraction, microwave-assistedextraction, and solvent extraction [16–18]. Among these, ultra-sound-assisted extraction (UAE) offers an inexpensive, environ-mently friendly and time efficient alternative to conventionalextraction techniques [19–21]. Similar to other secondary metabo-lites gylcoalkaloids are contained within cells and ultrasound in-duced cavitation can disrupt plant cell wall causing release ofintracellular components thus enhancing extraction [22]. Ultra-sound can also enhance extraction by reducing particle size andincreasing the net hydrophobic character of the extraction medium(when the target molecule is apolar) [23].
Ultrasound also offers a mechanical effect allowing greater pen-etration of solvent into the sample matrix, increasing the contactsurface area between the solid and liquid phase, and as a result,the solute quickly diffuses from the solid phase to the solvent[24,25]. A number of parameters can be manipulated to optimiseextraction of target compounds using ultrasound including extrac-tion time and temperature, ultrasound amplitude and extractingsolvent. Investigations involving the manipulation of all theseparameters to optimise extraction can be time consuming andexpensive. However, response surface methodology (RSM) is a sta-tistical technique which allows the user to identify optimal condi-tions for a selected response while minimising the number ofexperiments required. Therefore in the present study a responsesurface methodology (RSM) approach was undertaken to optimisethe ultrasound assisted extraction parameters such as amplitudelevel and extraction time by employing a central composite designto maximise extraction of steroidal alkaloids from potato peel slur-ry provided a major potato processor in Ireland. The optimal con-ditions were benchmarked against a standard solid liquidextraction approach. The study for the first time has reported theeffect of ultrasound on the extraction of steroidal alkaloids frompotato peel.
2. Material and methods
2.1. Samples and reagents
The steroidal alkaloids a-solanine, a-chaconine, solanidine anddemissidine were purchased from Extrasynthese (Genay Cedex,France). High performance liquid chromatography (HPLC) gradesolvents such as methanol, water, formic acid and acetonitrile werepurchased from Sigma–Aldrich (Wicklow, Ireland).
2.2. Drying of the peel slurry
Potato peel slurry was provided by Largo Foods Limited (Meath,Ireland). Prior to freezing the slurry was pressed by hand to removeexcessive water. Then the peels were spread on aluminium traysand cooled to �40 �C. Freeze-drying was carried out on the frozenpeel in A 6/14 freeze-drier (Frozen in Time Limited York, UK) at atemperature of �54 �C and a pressure of 0.064 mbar for 72 h.Freeze dried samples were immediately powdered, vacuum packed
and kept in �20 �C for extraction within two weeks. The film (75micron thickness) of the vacuum pack pouches (Allfo vakuumver-packungen Hans Bresele KG, Germany) were composed of a mix-ture of polyamide (PA) and polyethylene (PE) with an oxygenand carbon dioxide permeability rate of 60 cm3/m2/24 h/atomand 180 cm3/m2/24 h/atom respectively (23 �C, 75% RH). The watervapour permeability of the film was 2.7 g/m2/24 h at 23 �C and 85%RH.
2.3. Sonication treatment
Dried and powdered potato peel (7 g) were placed in a 100 mLjacketed vessel through which water was circulated at 15 ± 0.5,25 ± 0.5 and 35 ± 0.5 �C with a flow rate of 0.5 L/min. Extractionsolvent (70 mL methanol) was added to the sample and the ultra-sound probe was submerged to a depth of 25 mm in the solvent. A1500 W ultrasonic processor (VC 1500, Sonics and Materials Inc.,Newtown, USA) with a 19 mm diameter probe was used for sonica-tion (Fig. 1). Samples were processed at a constant frequency of20 kHz. The energy input was controlled by setting the amplitudeof the sonicator probe. Extrinsic parameters of amplitude (24.40,30.5, 42.70, 54.9 and 61.0 lm) and processing time (3, 5, 10, 15and 17 min) were varied with pulse durations of 5 s on and 5 soff. Ultrasound intensities of the mentioned amplitudes werecalculated as 9.24, 10.16, 13.28, 17.17 and 22.79 W/cm2 respec-tively. The ultrasonic power (P) was determined as described byTiwari et al. [26] using Eq. (1) where dT/dt is the change in temper-ature over time (�C s�1), Cp is the specifc heat of water(4.18 kJ kg�1 �C�1), and m is the mass (kg).
P ¼mCpðdT=dtÞt¼0 ð1Þ
Ultrasonic intensity (UI) dissipated from an ultrasonic probe tipwith diameter D is given by Eq. (2)
UI ¼ 4P
pD2 ð2Þ
All treatments were carried out in two batches which includedthree replications in each sample.
2.4. Conventional solid/liquid extraction
In order to benchmark optimal UAE condition against a conven-tional extraction technique, solid/liquid extractions were carriedout according to the method of Shan et al. [27] with slight modifi-cations. Briefly, dried and ground samples (7 g) were mixed with70 mL methanol and shaken for 17 min at room temperature(�23 �C). The sample suspension was then immediately filteredthrough Buchner funnel (pore size: 1.0 lm). The extracts were keptat �20 �C until subsequent analysis. Extraction was performed ontwo batches which included three replications in each sample.
2.5. Identification and quantification of steroidal alkaloids in potatopeel by ultra-performance liquid chromatography coupled withtandem mass spectrometry
Steroidal alkaloids were analysed using Waters Acquity (WatersCorporation, MA, USA) ultra-performance liquid chromatographycoupled with tandem mass spectrometry (UPLC-MS/MS). The com-pounds were separated on a Waters Acquity BEH C18 column(50 � 2.1 mm, particle size 1.7 lm) using 0.5% formic acid in water(Solvent A) and a mixture of acetonitrile, 2-propanol and formicacid in the ratio of 94.5:5:0.5 (solvent B). The following stepwisegradient program was carried out: 0–1 min 10% B, 2–6 min 20.5%B, 7–9 min 30% B, 9.5 min 90% B and 10–11 min 10% B at a flow rateof 0.5 mL/min. The injection volume for all the samples was 5 lL.All the standards in the concentration range from 0.1 to 1 lg/mL
(1) Ultrasound transducer
(2) Ultrasonic generator
(3) Ultrasound probe
(4) Data logger
(5) Temperature probe
(6) Jacketed beaker
(7) Computer
(8) Sample and solvent
(h) Depth of probe in to the sample (2.5 cm)
2
3
71
4
6
5
h
8
Fig. 1. Diagrammatical representation of ultrasound assisted extraction system used in the present study.
1472 M.B. Hossain et al. / Ultrasonics Sonochemistry 21 (2014) 1470–1476
for quantification purposes were dissolved in methanol. Samplesfor MS analysis were prepared in (methanol) following the extrac-tion described in the above Section 2.4. MS detection of ions wasperformed in the positive mode using multiple reaction monitoring(MRM). The parameters for the MRM transitions shown in Table 1were determined and optimised using the Waters integrated Intel-listartTM software. The ionization source conditions were as fol-lows: capillary voltage 3 kV, Cone voltage 30 V, extractor voltage3 V, source temperature 120 �C, desolvation temperature 350 �C,desolvation gas flow 800 L/hr, cone gas flow 50 L/hr, and collisiongas flow 0.10 mL/min. The MRM traces were acquired using theWaters MassLynx V4.1™ software while the quantifications ofthe data were carried out using the Waters TargetLynx TM software.
2.6. Statistical analysis
Optimal UAE extraction conditions were determined by RSMwhich was performed using the Design Expert Version 7.1.3 soft-ware (Stat-Ease, Inc., Minneapolis, MN). A central composite design(CCD) was used to investigate the effects of two independent vari-ables (ultrasonication amplitude and extraction time) on thedependent variables (steroidal alkaloids content). The data ob-tained from the CCD design was fitted with a second order polyno-mial equation. The equation was as follows:
Y ¼ b0 þX2
i¼1
biXi þX2
i¼1
biiX2i þ
X
i
X
j¼iþ1
bijXiXj ð3Þ
where Y is the predicted response; b0 is a constant; bi is the linearcoefficient; bii is the quadratic coefficient, bij is the interaction coef-ficient; and Xi and Xj are independent variables. The adequacy of themodel was determined by evaluating the lack of fit, coefficient ofregression (R2) and the Fisher test value (F-value) obtained fromthe analysis of variance (ANOVA). Statistical significance of themodel and model variables was determined at the 5% probability le-vel (p < 0.05). The software uses the quadratic model equationshown above to build response surfaces. Three-dimensionalresponse surface plots were generated by keeping one response
Table 1MS detection parameters for steroidal alkaloids of potato peel.
Compound MRM transitions
a-Solanine m/z 868.53 ? m/z 398.2 m/z 868.53 ? m/z 722.3a-Chaconine m/z 852.53 ? m/z 398.2 m/z 852.53 ? m/z 706.7Solanidine m/z 398.2 ? m/z 147.42Demissidine m/z 400.36 ? m/z 161.16
variable at its optimal level and plotting that against two factors(independent variables). The complete design consisted of 13experimental points including five replications of the central point.
2.7. Model validation
The predictive performance of the developed models describingthe combined effect amplitude (X1), temperature (X2) and time (X3)on independent variables (a-solanine, a-chaconine, solanidine anddemissidine) of potato peel were validated with optimal extractionconditions as predicted by the design.
The criterion used to characterize the fitting efficiency of thedata to the model was the multiple correlation coefficients (R2)and their average mean deviation (E, Eq. (4)).
Eð%Þ ¼ 1ne
Xn
i¼1
VE � VP
VE
����
����� 100 ð4Þ
where, ne is the number of experimental data, VE is the experimen-tal value and VP is the predicted value.
3. Results and discussion
3.1. Ultrasonic extraction of steroidal alkaloids
Potato peel from the potato processing industry is a relativelylarge waste stream in Ireland and is not currently used as a sourceof high-value bioactive compounds. Compounds of phyto-pharma-ceutical interest such as steroidal alkaloids are mainly located inthe peel, which further emphasises the possible value of this wastestream [2]. Fig. 2 presents the response surface plots showing theeffect of two different parameters of UAE (ultrasound amplitude,and extraction time) on content of two glycoalkoids (a-solanine,a-chaconine) and two aglycones (solanadine and demissidine)in methanolic potato peel extracts. The concentrations of the a-solanine in the ultrasound assisted extracts of potato peels weresignificantly (p < 0.05) increased with increasing amplitude andtime. As is evident from the RSM plots amplitude had stronger
Dwell time (ms) Cone voltage (V) Collision energy (eV)
0.042 98 780.042 94 800.042 62 420.042 74 46
24.4 33.5
42.7 51.8
61.0
5.0 8.0
11.0 14.0
17.0
177
199
221
242
264 A
lpha
-Sol
anin
e (µ
g/g
DW
)
Amplitude (µm) Time (Min) 24.4
33.5 42.7
51.8 61.0
5.0 8.0
11.0 14.0
17.0
340
390
440
490
540
Alp
ha-C
haco
nine
(µg/
g D
W)
Amplitude (µm) Time (Min)
a b
24.4 33.5
42.7 51.8
61.0
5.0 8.0
11.0 14.0
17.0
25.0
32.5
40.0
47.5
55.0
Dem
issi
dine
(µg/
g D
W)
Amplitude (µm) Time (Min) 24.4
33.5 42.7
51.8 61.0
5.0 8.0
11.0 14.0
17.0
156
177
198
218
239
Sol
anid
ine
(µg/
g D
W)
Amplitude (µm) Time (Min)
c d
Fig. 2. Response surface plots of ultrasound assisted extracts of potato peel showing the effect of temperature and methanol concentration on alpha-solanine (a), alpha-chaconine (b), solanidine (c) and demissidine (d) contents.
M.B. Hossain et al. / Ultrasonics Sonochemistry 21 (2014) 1470–1476 1473
effect than extraction time for the alkaloids under investigation.For example an increase in amplitude from 24.4 to 61 lm at10 min extraction time increased the concentration of a-solaninefrom 201.6 lg/g DPP to 263.7 lg/g DPP. To date no other studieshave investigated the effect of UAE on recovery of steroidal alka-loids from potato peel. In general higher amplitudes of ultrasoundinduce more damage to cell walls releasing thus releasing moreintracellular components which can then partition into the extract-ing solvent. The efficacy of ultrasonication amplitude in enhancingextraction of polyphenols from Lamiaceae herbs has been reported[28–30]. The factor time when increased allowed the solutes to bein contact with solvent for longer facilitating higher diffusion of thetarget compounds [31]. The pattern of effect of the factors onextraction of a-solanine was found to be linear. On the other hand,the concentrations of a-chaconine were only affected by amplitudeat linear level. The extraction of steroidal alkaloid aglycone solani-dine also showed significant linear increase only by the effect ofamplitude. The other alkaloid aglycone demissidine is the dihydro-genated form of solanidine was predominantly affected by ampli-tude at both linear and quadratic levels. This indicated a curvedpattern in the response surface plots (Fig. 2). Time also had signif-icant (p < 0.05) interaction effect on extraction of demissidinealthough there was no significant (p > 0.05) interaction betweenamplitude and time while affecting the extractions of a-solanine,a-chaconine and solanidine. Overall, time had little (as in case ofa-solanine and demissidine) or no effect (as in case of a-chaconineand solanidine) on extraction of alkaloids within the experimentalrange (3–17 min) used. An ultrasonication treatment time of 3 minwas probably sufficient to extract the major portion of the potatosteroidal alkaloids. Further increase of time did not have pro-nounced effect on extraction of these compounds (Fig. 2). There-
fore, higher levels of time (>17 min) was expected to have slighteffect on the extraction of steroidal alkaloids. However, this wouldhave increased the cost of extraction and an environmentallyfriendly extraction method requires minimal extraction time[31]. Considering this, in the present study, time range was keptlow. Similar to the individual steroidal alkaloids the extractionyield was also increased with the ultrasound amplitude. Amongthe UAE treatments as generated by RSM, the highest extractionyield was observed at the ultrasonication amplitude of 61 lmand extraction time of 10 min. The results showed that the UAEtreatment conditions of 61 lm (amplitude) and 10 min yielded18.57 mg dried extract/g DPP whereas solid liquid extractionyielded 14.56 mg dried extract/g DPP (Fig. 3). Quantification of ste-roidal alkaloids using UAE treatment at amplitude of 61 lm andextraction time of 10 min showed glycoalkaloid content of 263.7,555.3, 248.8 and 48.8 lg/g DPP for a-solanine, a-chaconine, solani-dine and demissidine respectively (Total glycoalkaloid con-tent = 1116.6 lg/g DPP). Whereas for solid liquid extractionyields were observed to be 180.3, 337.6, 160.2 and 32.41 lg/gDPP for a-solanine, a-chaconine, solanidine and demissidinerespectively (Total glycoalkaloid content = 710.51 lg/g DPP). Thisis an average of 1.5-fold increase of each of the four different ste-roidal alkaloids in the UAE with respect to SLE rendering the pos-sibility of extracting the aglycone solanidine 6 times more thanthe conventional SLE. The total glycoalkaloid contents of UAE andSLE as mentioned above were in the range (585 to 5342 lg/gDPP) as reported by Deusser et al. [32], while analysing the steroi-dal alkaloids in the peels of 16 potato varieties grown in Luxem-burg. The predicted values from the developed second orderquadratic polynomial equation were in close agreement with theexperimental values with low average mean deviation (E%) values.
24.4 33.5
42.7 51.8
61.0
5.0 8.0
11.0 14.0
17.0
690
788
885
983
1080
Tot
al s
tero
idal
alk
aloi
d (µ
g/g
DW
)
Amplitude (µm) Time (Min) 24.4
33.5 42.7
51.8 61.0
5.0 8.0
11.0 14.0
17.0
15
16
17
18
19 Y
ield
(mg/
g D
W)
Amplitude (µm) Time (Min)
a b
Fig. 3. Response surface plots showing the effect of ultrasound assisted extraction parameters such as amplitude and time on extraction yield (a) and total steroidal alkaloidscontent (b) from potato peels.
1474 M.B. Hossain et al. / Ultrasonics Sonochemistry 21 (2014) 1470–1476
The extraction yields of the optimal UAE extracts were significantly(p < 0.05) higher than the solid liquid extracts. Predicted modelswere highly significant (p < 0.05) for all parameters studied. Thisstudy indicates that UAE has strong potential as an extractionmethod for steroidal alkaloids from potato peel waste. Plant sec-ondary metabolites have been reported to be degraded by the ef-fect of ultrasound treatment during extraction [33]. However, nodegradation of steroidal alkaloids due to ultrasound treatmentwas observed in the present study (Fig. 4 and Table 2). A solutioncontaining a-solanine, a-chaconine, solanidine and demissidinein the concentration of 1.0 lg/ml was treated with the highestamplitude of ultrasonication used. This solution was analysedagain after treatment by UPLC-MS/MS. The contents of the four ste-roidal alkaloids in the solution before treatment were similar to
1.00 2.00 3.00 4.00 5.00
%
0
100
1.00 2.00 3.00 4.00 5.00
%
0
100
H
HH
N
HH
HOOHHO
O
OOH
OH
O
HO
OHO
HO
OH
O
OH
H
HH
N
HH
HO
OH
OHO O
OOH
OHOHO
HOO
HOOH
α-Chaconine
α-Solanine
Fig. 4. Total ion current chromatogram of the MRM traces of a-solanine (1), a-chaconinecomparison to the solution before treatment (B). The solutions were diluted by 10 time
those of the solution after treatment indicating no degradation ofthese compounds by ultrasound.
3.2. Model fitting
The analysis of variance showed that the R-squared statistic ofall the parameters was in the range of 0.717–0.969 indicating highrepresentation of the variability of the parameters by the models(Table 3). The quadratic polynomial models generated were highlysignificant with p-value ranging from 0.0121 to 0.0001. The lack offit statistics of all the parameters were not significant (p > 0.05)and high degree of F-value (range 6.90–43.21) further strength-ened the reliability of the models. The predicted values obtainedby the quadratic polynomial equations showed strong correlation
Time6.00 7.00 8.00 9.00
6.00 7.00 8.00 9.00
400.4
852.8868.5
398.2
400.4
868.5 852.8
398.2
H
HH
N
HH
HHO
H
HH
N
HH
HHO
1
2
3
4
1 2
3
4
A
B
TIC3.87 e4
TIC3.76 e4
(Min)
Solanidine
Demissidine
(2), solanidine (3) and demissidine (4) in solutions after ultrasound treatment (A) ins before analyses.
Table 2Quantity of a-solanine, a-chaconine, solanidine and demissidine in solution afterultrasound treatment in comparison to the solution before treatment.
Ultrasoundtreatment
Name of thecompound
Before treatment(lg/ml)
After treatment(lg/mla)
61 lm and 17 min a-Solanine 1.0 0.98 ± 0.02a-Chaconine 1.0 1.07 ± 0.04Solanidine 1.0 1.06 ± 0.03Demissidine 1.0 0.99 ± 0.01
a Data was expressed as mean ± SD (n = 3).
M.B. Hossain et al. / Ultrasonics Sonochemistry 21 (2014) 1470–1476 1475
with actual experimental values with Pearsons correlation coeffi-cients (r) from 0.86 to 97.
3.3. Optimisation and model validation
The RSM guided optimisation demonstrated that the optimumtreatment for maximising the extraction of steroidal alkaloids frompotato peel was 61 lm (amplitude), and 17 min (time). The pre-dicted values of a-solanine, a-chaconine, solanidine and demissi-dine at optimal UAE conditions were 263.0 lg/g DW, 534.7 lg/gDW, 239.0 lg/g DW and 54.1 lg/g DW respectively. The predictedvalues were in close agreement with experimental values (Table 4)and were found to be not significantly different at p > 0.05 using apaired t-test. In addition variations between the predicted andexperimental values obtained for all the steroidal alkaloids studiedwere within acceptable error range as depicted by average mean
Table 3Analysis of the variance of the regression coefficients of the fitted polynomial quadrapeel.
Coefficients a-Solanine a-Chaconine Solan
b0 (Intercept) +125.921 +223.674 +104Linearb1 (Amplitude) +1.740c +4.331b +2.0b2 (Time) +1.829a +2.777ns +0.7Quadraticb11 – – –b22 – – –Cross productb12 – – –R2 0.783 0.717 0.79CV 5.24 7.43 6.07p 0.0005 0.0015 0.00Lack of fit 0.8043 0.0621 0.35F-value 18.03 12.70 19.1
ns Not significant.a Significant at p 6 0.05.b Significant at p 6 0.01.c Significant at p 6 0.001.d Significant at p 6 0.0001.
Table 4Predicted and experimental values of the parameters tested at optimal UAE conddeviation between predicted and experimental values of optimal UAE.
Parameter Optimum UAE condition forall the parameters combined
Prediat op
a-Solanine (lg/g DW) 61 lm and 17 min 263a-Chaconine (lg/g DW) 534Solanidine (lg/g DW) 239Demissidine (lg/g DW) 54Total steroidal alkaloid (lg/g DW) 1090Yield (mg/g DW) 19
tic eq
idine
.278
02d
37ns
1
04504
ition i
cted vtimal
.0
.7
.0
.1
.8
.0
deviation (E%), Table 4); therefore, the predictive performance ofthe established model may be considered acceptable. The experi-mental extraction values of the alkaloids at optimal conditionswere significantly (p < 0.05) higher than those of the solid liquidextracts. In fact, demissidine content of the optimal UAE extractwas 70.62% higher than that of the solid–liquid extract. The otheralkaloids also showed an increase of 44.28–60.75% in optimalUAE than those of solid/liquid extracts. Ultrasound treatment is apowerful tool with a number of positive and negative effects dur-ing extraction of phytochemicals. Vilkhu et al. [23] reviewed thepotential use of UAE for food industries and highlighted a numberof advantages in this technology including increased yield ofextraction, increased rate of extraction, increased bioavailabilitythrough chemical modifications, reduction in extraction time, sol-vents over conventional extraction techniques such as maceration.The major drawbacks of ultrasound treatment are generation offree radicals such as hydrogen peroxide and increase of tempera-ture which can lead to degradation of the extracted compounds.However, the positive effects of ultrasound assisted extraction out-weigh its negative effects. Therefore, use of ultrasound for extrac-tion is gradually increasing. Although the effect of ultrasound onextraction of plant metabolites is promising, scale-up remains asignificant technological challenge to industrial adoption. Utilisa-tion of power ultrasound at industrial scale extraction requires en-ergy-efficient ultrasound processors. For food applications, thedesign of the probe is very important. Non contact or coatedtransducers where the construction materials are non-reactive,and erosion resistant are required.
uation for steroidal alkaloids (lg/g DW), and extraction yield (mg/g DW) from potato
Demissidine Total steroidal alkaloid Yield
+89.384 +460.769 +21.812
�2.725d +8.511c �0.324b
�1.947ns +5.705ns �0.142ns
+0.029d – +4.115b
+0.013ns – +4.213ns
+0.054b – +3.698ns
0.969 0.760 0.8314.25 6.48 3.210.0001 0.0008 0.01210.3261 0.1704 0.627043.21 15.87 6.90
n comparison to the conventional solid/liquid extraction values and average mean
aluesUAE
Desirability Experimental valuesat optimal UAE
(E%) Solid/liquidextraction values
0.934 273 ± 3.20 3.80 180.3 ± 2.11542.7 ± 5.12 1.49 337.6 ± 4.15231 ± 4.32 3.34 160.2 ± 3.0955.3 ± 1.56 2.21 32.41 ± 1.111102 ± 1.02 2.32 710.5 ± 6.0419.6 ± 0.56 3.15 14.56 ± 0.78
1476 M.B. Hossain et al. / Ultrasonics Sonochemistry 21 (2014) 1470–1476
4. Conclusion
The extraction efficiency of UAE of steroidal alkaloids wasmainly influenced by amplitude although time had significant ef-fect on extraction of a-solanine and demissidine. The effect ofamplitude on extraction was higher than that of time. The optimalUAE had considerably higher alkaloid content than the conven-tional solid–liquid extraction. UAE is thus a low cost and potentialenvironment friendly extraction choice for GCA from potato peelswhich can be scaled up to industrial level. RSM guided optimisa-tion tool was useful to identify the extraction parameters to max-imise the extraction of steroidal alkaloids.
References
[1] K.C. Chang, Polyphenol antioxidants from potato peels: extraction,optimization and application to stabilizing lipid oxidation in foods, in:Proceedings of the National Conference on Undergraduate Research (NCUR),Ithaca College, New York, USA, 2011.
[2] A. Schieber, M.D.A. Saldana, Potato peels: a source of nutritionally andpharmacologically interesting compounds-A review, Food 2 (2009) 23–29.
[3] H.H. Wijngaard, M. Ballay, N. Brunton, The optimisation of extraction ofantioxidants from potato peel by pressurised liquids, Food Chem. 133 (2012)1123–1130.
[4] M. Friedman, Potato glycoalkaloids and metabolites: roles in the plant and inthe diet, J. Agric. Food Chem. 54 (2006) 8655–8681.
[5] S. Austin, E. Lojkowska, M.K. Ehlenfeldt, A. Kelman, J.P. Helgeson, Fertileinterspecific somatic hybrids of Solanum: a novel source of resistance toErwinia soft rot, Phytopathology 78 (1988) 1216–1220.
[6] A.M. Fewell, J.G. Roddick, Interactive antifungal activity of the glycoalkaloidsalpha-solanine and alpha-chaconine, Phytochemistry 33 (1993) 323–328.
[7] L.L. Sanford, R.S. Kobayashi, K.L. Deahl, S.L. Sinden, Diploid and tetraploidSolanum chacoense genotypes that synthesize leptine glycoalkaloids and deterfeeding by Colorado potato beetle, Am. Potato J. 74 (1997) 15–21.
[8] S.C. Morris, T.H. Lee, The toxicity and teratogenicity of Solanaceaeglycoalkaloids, particularly those of the potato (Solanum tuberosum): areview, Food Technol. Aust. 36 (1984) 118–124.
[9] P. Slanina, Solanine (glycoalkaloids) in potatoes: toxicological evaluation, FoodChem. Toxicol. 28 (1990) 759–761.
[10] S.J. Jadhav, R.P. Sharma, D.K. Salunkhe, Naturally occurring toxic alkaloids infoods, CRC Crit. Rev. Toxicol. 9 (1981) 21–104.
[11] J.A. Maga, Potato glycoalkaloids, CRC Crit. Rev. Food Sci. Nutr. 12 (1980) 371–405.
[12] J.G. Roddick, The acetylcholinesterase-inhibitory activity of steroidalglycoalkaloids and their aglycones, Phytochemistry 28 (1989) 2631–2634.
[13] B.J. Phillips, J.A. Hughes, J.C. Phillips, D.G. Walters, D. Anderson, C.S.M.Tahourdin, A study of the toxic hazard that might be associated with theconsumption of green potato tops, Food Chem. Toxicol. 34 (1996) 439–448.
[14] M. Friedman, Potato glycoalkaloids and metabolites: roles in the plant and inthe diet, J. Agric. Food Chem. 54 (2006) 8655–8681.
[15] O.M. Kenny, C.M. McCarthy, N.P. Brunton, M.B. Hossain, D.K. Rai, S.G. Collins,P.W. Jones, A.R. Maguire, N.M. O’Brien, Anti-inflammatory properties of potato
glycoalkaloids in stimulated Jurkat and Raw 264.7 mouse macrophages, LifeSci. 92 (2013) 775–782.
[16] S.B. Glisic, M. Ristic, D.U. Skala, The combined extraction of sage (Salviaofficinalis L.): Ultrasound followed by supercritical CO2 extraction, Ultrason.Sonochem. 18 (2011) 318–326.
[17] M. Vinatoru, An overview of the ultrasonically assisted extraction of bioactiveprinciples from herbs, Ultrason. Sonochem. 8 (2001) 303–313.
[18] H. Wijngaard, M.B. Hossain, D.K. Rai, N. Brunton, Techniques to extractbioactive compounds from food by-products of plant origin, Food Res. Int. 46(2012) 505–513.
[19] Y. Ma, X. Ye, Y. Hao, G. Xu, G. Xu, D. Liu, Ultrasound-assisted extraction ofhesperidin from Penggan (Citrus reticulata) peel, Ultrason. Sonochem. 15(2008) 227–232.
[20] D.T. Velickovic, D.M. Milenovic, M.S. Ristic, V.B. Veljkovic, Ultrasonic extractionof waste solid residues from the Salvia sp. essential oil hydrodistillation,Biochem. Eng. J. 42 (2008) (2008) 97–104.
[21] M. Virot, V. Tomao, C. Le Bourvellec, C.M.C.G. Renard, F. Chemat, Towards theindustrial production of antioxidants from food processing by-products withultrasound-assisted extraction, Ultrason. Sonochem. 17 (2010) (2010) 1066–1074.
[22] S. Balachandran, E. Kentish, R. Mawson, M. Ashokkumar, Ultrasonicenhancement of the supercritical extraction from ginger, Ultrason.Sonochem. 13 (2006) 471–479.
[23] K. Vilkhu, R. Mawson, L. Simons, D. Bates, Applications and opportunities forultrasound assisted extraction in the food industry – a review, Innovative FoodSci. Emerg. Technol. 9 (2008) 161–169.
[24] M. Toma, M. Vinatoru, L. Paniwnyk, T.J. Mason, Investigation of the effects ofultrasound on vegetal tissues during solvent extraction, Ultrason. Sonochem. 8(2001) 137–142.
[25] T.J. Mason, Sonochemistry and sonoprocessing: the link, the trends and(probably) the future, Ultrason. Sonochem. 10 (2003) 175–179.
[26] B.K. Tiwari, K. Muthukumarappan, C.P. O’Donnell, P.J. Cullen, Effects ofsonication on the kinetics of orange juice quality parameters, J. Agric. FoodChem. 56 (2008) 2423–2428.
[27] B. Shan, Y.Z. Cai, M. Sun, H. Corke, Antioxidant capacity of 26 spice extracts andcharacterization of their phenolic constituents, J. Agric. Food Chem. 53 (2005)7749–7759.
[28] S. Albu, E. Joyce, L. Paniwnyk, J.P. Lorimer, T.J. Mason, Potential for the use ofultrasound in the extraction of antioxidants from Rosmarinus officinalis for thefood and pharmaceutical industry, Ultrason. Sonochem. 11 (2204) 261–265.
[29] L. Paniwnyk, H. Cai, S. Albu, T.J. Mason, R. Cole, The enhancement and scale upof the extraction of anti-oxidants from Rosmarinus officinalis using ultrasound,Ultrason. Sonochem. 16 (2009) 287–292.
[30] M.B. Hossain, N.P. Brunton, A. Patras, B. Tiwari, C.P. O’Donnell, A.B. Martin-Diana, C. Barry-Ryan, Optimization of ultrasound assisted extraction ofantioxidant compounds from marjoram (Origanum majorana L.) usingresponse surface methodology, Ultrason. Sonochem. 19 (2012) 582–590.
[31] K. Ghafoor, Y.H. Choi, J.Y. Jeon, I.H. Jo, Optimization of ultrasound-assistedextraction of phenolic compounds, antioxidants, and anthocyanins from grape(Vitis vinifera) seeds, J. Agric. Food Chem. 57 (2009) 4988–4994.
[32] H. Deußer, C. Guignard, L. Hoffmann, D. Evers, Polyphenol and glycoalkaloidcontents in potato cultivars grown in Luxembourg, Food Chem. 135 (2012)2814–2824.
[33] D. Pingret, A. Fabiano-Tixier, F. Chemat, Degradation during application ofultrasound in food processing: a review, Food Control 31 (2013) 593–606.
