Journal of Liquid Chromatography & Related

Technologies

ISSN: 1082-6076 (Print) 1520-572X (Online) Journal homepage: http://www.tandfonline.com/loi/ljlc20

Development and Validation of High-Performance

Thin-Layer Chromatographic Method for
Determination of Amygdalin

Aleksandra Radoičić, Radivoj Petronijević, Filip Andrić, Živoslav Tešić &

Dušanka Milojković-Opsenica

To cite this article: Aleksandra Radoičić, Radivoj Petronijević, Filip Andrić, Živoslav Tešić &
Dušanka Milojković-Opsenica (2017): Development and Validation of High-Performance Thin-Layer
Chromatographic Method for Determination of Amygdalin, Journal of Liquid Chromatography &
Related Technologies

To link to this article: http://dx.doi.org/10.1080/10826076.2017.1298032

Accepted author version posted online: 01

Mar 2017.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ljlc20

Download by: [University of Newcastle, Australia] Date: 03 March 2017, At: 07:02
 DEVELOPMENT AND VALIDATION OF HIGH-PERFORMANCE THIN-
LAYER CHROMATOGRAPHIC METHOD FOR DETERMINATION OF
AMYGDALIN

Aleksandra Radoičić1, Radivoj Petronijević2, Filip Andrić1, Živoslav Tešić1, Dušanka

Milojković-Opsenica1
1
Faculty of Chemistry, University of Belgrade, Belgrade, Serbia, 2Institute of Meat
Hygiene and Technology, Belgrade, Serbia

Corresponding author: Dr Dušanka Milojković-Opsenica,Faculty of Chemistry,

University of Belgrade,Studentski trg 12-16, P.O. Box 51,11158 Belgrade, Serbia.E-mail:
dusankam@chem.bg.ac.rs

Abstract

A new method for the extraction and quantitative determination of amygdalin has been

proposed. Accelerated solvent extraction (ASE) was applied for the extraction, and

reversed-phase high-performance thin-layer chromatography (RP-HPTLC) method was

developed, validated and applied for the determination of amygdalin in the extracts of

apricot, plum, almond, and peach kernels. The chromatographic system used was: RP-18

silica, as stationary phase and acetonitrile/water (50:50, v/v), as mobile phase.

Densitometric scanning was performed at 210 nm. The method was validated with

respect to specificity, linearity, precision, and accuracy. The results showed that the peak

area responses were linear within the concentration range of 2.5-50.0 µg / spot (R2 =

0.9984). The limit of quantification was 4.28 µg per spot, and the detection limit 1.28 µg

per spot. The intra-day and inter-day reproducibility, in terms of %RSD, were in the

range of 0.81-1.15 and 1.32-1.89, respectively. The accuracy data were in range from

99.98 to 100.56%. The method is linear, quantitative and reproducible, and could be used

1
as an efficient and economical green chromatographic procedure for the determination of

amygdalin in the fruit kernel.

GRAPHICAL ABSTRACT

KEYWORDS: Amygdalin; Method validation; Method development; Thin-layer

chromatography; Accelerated solvent extraction; Fruit kernels.

INTRODUCTION

Amygdalin (D-mandelonitrile-β-D-gentiobioside, D-mandelonitrile-β-D-glucoside-6-β-

glucoside) is one of the many nitrilosides, which are natural cyanide-containing

substances present in the seeds of prunasin family. It is the one major components of

apricot kernels, bitter almonds and peach, plum, pear and apple seeds and gives the

characteristic bitter taste of almond.[1–4] These plants contain various amounts of

cyanogenic glycosides depending on plant species, climate and environmental

conditions.[1,5–7]

2
Amygdalin has been used for the treatment of asthma, bronchitis, emphysema, leprosy,

and diabetes, as well as for preventing and treating migraine, hypertension, chronic

inflammation and other reaction source diseases. It was reported that amygdalin exhibits

analgesic and anti-inflammatory effects. Also, amygdalin can be used as a cerebral

function improver, effective as a therapeutic agent for cerebrovascular lesions such as

psychogenic symptoms, nerve symptoms, subjective symptoms and daily life activity

disorder.[4,8,9] In addition, it is reported that amygdalin has been used to treat cancer

because it was assumed that this molecule has selective killing effect on cancer cells[4],

but there is no reliable scientific evidence to support claims that amygdalin can treat

cancer. Although there are conflicting opinions about its efficiency in the treatment of

cancer, interest in amygdalin is gradually increasing and its use as secondary cancer

therapy has been encouraged.

Amygdalin is composed of two molecules of glucose, one of benzaldehyde and one

molecule of hydrocyanic acid (Fig. 1). Hydrolytic degradation of amygdalin, as a

consequence of its metabolism, produces hydrogen cyanide, a potential toxin. Since β-

glucosidase, one of the enzymes that catalyze hydrolysis of amygdalin is present in the

human small intestine the consumption of amygdalin is potentially toxic because of

possible cyanide production in enough high levels.[1]

Considering the mentioned pharmacological and toxic effects of amygdalin on the one

hand, and world-wide consumption of almonds and apricot or peach kernels (e.g. for

3
production of marzipan and persipan) on the other hand, an efficient and simple method

for amygdalin analysis from natural sources is highly desirable.

There are many papers describing methods for the determination of amygdalin in

different fruit kernels. In most publications for quantitative determination of amygdalin

different HPLC methods were used. These HPLC analyses were carried out in reversed-

phase mode, on a C18 stationary phase and the mobile phase consisted of acetonitrile-

water in different proportions. For amygdalin detection photodiode array (PDA), UV

detectors or triple-quadrupole mass spectrometer (QqQ MS/MS) with electrospray

ionization (ESI) were used.[2,5,7,9–13]

In recent years development of new analytical methods that are less dangerous and

harmful to human health and the environment has gained increasing attention, i.e. choice

of these methods is guided by the principles of “green chemistry”. Green chemistry is

defined as “the design of chemical products and chemical processes that reduce or

eliminate the use and generation of hazardous substances”.[14] Nowadays, the procedures

that minimize toxic solvent use, use non-toxic alternatives, and save energy are widely

accepted by scientists and by the industry. Accordingly, the goal of green analytical

chemistry is to design and develop analytical methods that, from sampling, through the

sample preparation and analysis, to the disposal, minimize consumption of hazardous

reagents and solvents, and maximize safety for operators and the environment.[15,16]

4
The aim of this work is green approach to development of a simple, efficient and

economic HPTLC-densitometric method for quantitative determination of amygdalin.

HPTLC was chosen as simple, easy to handle and sensitive method for simultaneous

analysis of various organic and inorganic substances.[17,18] Due to sample amount and low

mobile phase consumption method is rapid and economic. Besides reducing the cost of

analysis, reduction of quantity of mobile phase used, and thus, reduction of the

consumption of organic solvents makes HPTLC method environmentally friendly. In

order to use as less as possible organic solvents the reversed-phase systems with a high

content of water in the mobile phase were chosen.

Soxhlet extraction is the most commonly used technique for amygdalin extraction from

fruit kernels.[5, 7] However, besides the efficiency this method has some disadvantages,

such as long extraction time, lack of automation and high amount of solvent used per

sample. In recent years, new extraction techniques have been introduced in order to

reduce the amount of solvent required, to reduce the operation time, to improve the

precision of analysis, and to reduce the cost of sample preparation.[19] In this study, we

have explored the accelerated solvent extraction (ASE) as a new extraction procedure that

uses organic solvents at high pressures and temperatures above the boiling points of

extraction solvents to accelerate the extraction process.[20] Compared to Soxhlet

extraction, which normally takes hours, ASE method decreases extractions times to less

than 30 minutes per sample and reduces the amount of solvent, and by that decreases the

cost of analysis. The optimal ASE extraction conditions were determined by applying

5
experimental design (DoE) which reduces experimental work, making in that way the

analysis greener.

EXPERIMENTAL

Reagents And Chemicals

Standard compound, amygdalin (98% purity), was purchased from Sigma Aldrich (St.

Louis, USA). All of the used solvents (Merck, Darmdstat, Germany) were of analytical or

HPLC-grade purity. Ultrapure water (Thermo Fisher TKA, MicroPure water purification

system, 0.055 µS/cm) was used to prepare solutions.

Preparation of standard solutions

A stock solution of amygdalin was prepared by dissolving accurately weighed quantity of

amygdalin in water to get the concentration of 25 µg µL–1. The final working standard

solutions were prepared by diluting of stock solution with water to get the concentration

of 20, 10, 5, 2.5 and 1.25 µg µL–1, respectively.

HPTLC Determination

Aliquots of 2 µL of standard solutions of amygdalin were applied on 10 cm × 10 cm

HPTLC RP-18 F254 (Art. 13124, Merck, Darmdstat, Germany) and CN F254S (Art. 16464,

Merck) plates to obtain concentration of 50, 40, 20, 10, 5 and 2.5 µg per spot.

Standard solutions and samples were applied as 8 mm wide bands, 10 mm from the

bottom and left edge of the plate with a CAMAG microlitre syringe using a CAMAG

6
Linomat 5 applicator (CAMAG, Muttenz, Switzerland). A constant application rate of 50

nL s-1 was employed. Four mL of mobile phase consisted of acetonitrile (ACN)-water

(50:50, v/v) were used for development of chromatograms in a horizontal developing

chamber (10 cm × 10 cm) saturated with vapors of the mobile phase. The optimized

chamber saturation time was 15 min. The development distance was 80 mm. After

development, the HPTLC plates were air dried. All experiments were performed at

ambient temperature (22 ± 2°C).

Densitometric scanning was performed in the absorbance mode at 210 nm on CAMAG

TLC Scanner3. The source of radiation used was deuterium lamp and scanning rate of 20

mm/s was employed. Concentrations of the amygdalin were determined from the

intensity of the diffused light. Evaluation of amygdalin content was performed by linear

regression of peak areas determined by absorption as a function of sample amounts. All

instruments were controlled by WinCATS software (CAMAG).

Method Validation

Aliquots of 2 μL of standard solutions were spotted on HPTLC plate to obtain final

amygdalin concentrations of 2.5-50 μg/spot. Each experiment was repeated six times.

Validation of the HPTLC method for amygdalin determination included the

following parameters: linearity and range, specificity, precision, accuracy, limit of

detection as well as limit of quantification, and was carried out considering

recommendations of the International Conference of Harmonization.[21]

7
Linearity and range were determined by developing the plates using previously described

chromatographic system, and the peak areas were plotted against the corresponding

concentrations to obtain the calibration curves.

Specificity of the method was determined by analyzing standard substance and test

samples. The spot for amygdalin in the samples was confirmed by comparing the RF

value of the amygdalin spot in the samples with that of the standard.

Precision of the method was verified by repeatability and intermediate precision studies.

Repeatability was performed by repeating analysis of the amygdalin six times on the

same day, and the intermediate precision of the method was checked by repeating

analysis on three different days.

Accuracy of the method was determined by applying the method of standard

addition. Standard solution was added to the sample at three different concentration

levels, covering the low, medium and high ranges of the calibration curve. At each level

of the amount, three determinations were performed. Each experiment was performed in

triplicates. Accuracy was expressed as a recovery rate.

Limit of detection (LOD) is defined as the lowest concentration of an analyte in a sample

that can be distinguished from a blank, and limit of quantification (LOQ) is defined as

the lowest concentration of an analyte in a sample that can be determined with acceptable

8
precision and accuracy. LOD and LOQ are expressed as a concentration at a certain

specified signal-to-noise ratio: 3 for LOD and 10 for LOQ.

Sample Preparation

Fruit kernels were obtained from the Faculty of Agriculture, University of

Belgrade. After the breaking kernels, 20 g of apricot seeds were ground in coffee mill,

homogenized and soaked in methanol in a 100 mL flask. Methanol was evaporated at 40

°C in the rotary evaporator. Sample prepared in this way was used for optimization of

amygdalin extraction conditions on ASE 100 (Dionex, Sunnyvale, CA, USA).

ASE conditions were as follows: extraction was performed in 10 mL extraction cells,

with the flush volume of 60 % and purge time of 100 s. Water was used as extraction

solvent.

Screening Design (Doe) And Optimization Of Extraction

After determination of the optimal extraction conditions (temperature 110 oC, sample

mass 2.5 g, 3 extraction cycles, and 10 minutes each) all samples of kernels were

prepared under the same procedure, and analyzed by HPTLC method. As matrix, apricot,

plum, almond, and peach kernels were used.

For the optimization of the extraction procedure full factorial DoE was employed.[22]

Factors used for optimization were: temperature (70 and 110 oC), number of extraction

cycles (1 and 3), one extraction cycle duration (2 and 10 minutes), and sample mass (0.5

9
and 2.5 g). The extraction efficiency of the each experiment was evaluated by

determining the concentration of amygdalin by HPLC. The design included 8

experiments with different combinations of factors. Two additional extractions were

performed under the following conditions: temperature 90 oC, two extraction cycles

duration of 6 minutes, sample mass 1.5 g (average factor values, central points of DoE).

Experiments were designed and DoE results were analyzed by JMP 10 (SAS Institute,

USA) statistical software.

HPLC Determination

Rude extracts, obtained by ASE rapid methanol extraction procedure, were filtered

through membrane filter (nylon filter, pore size 0.45 µm, Kinesis, Cambridgeshire, UK)

into 2 mL vial of autosampler. The HPLC experiments were conducted using Alliance

2695 (Waters, Milford, MA, USA) system equipped with PDA detector (2996

Photodiode Array Detector, Waters, Milford, MA, USA). An octadecylsilica column

Luna C18(2) with C18 precolumn (Phenomenex, Torrance, CA, USA) was used.

Isocratic elution with methanol : water 25:75 (% v/v) mobile phase was employed.

Sample aliquots of 100 µl were injected. Under these conditions, the retention time of

amygdalin chromatographic peak was 7 minutes. All experiments were performed at

ambient temperature (22 ± 2oC).

Application Of The Proposed Method To Fruit Kernel Samples

The developed and validated method was applied for determination of amygdalin content

in twelve real samples. For that purpose the kernels of apricot, plum, peach, and almond

10
samples, the kernels were finely grounded, homogenized, and extracted as described in

the section Sample preparation. The obtained extracts were applied on HPTLC plate,

and the development of chromatograms and scanning of the spots were performed

afterwards.

RESULTS AND DISCUSSION

Method Development

In order to establish an appropriate and simple chromatographic system and to obtain

optimal efficiency, two adsorbents (RP-18 and CN-modified silica gel) and several

solvent systems were examined (Table 1).

Selection of optimal chromatographic conditions included several steps. First, taking into

consideration structure of the amygdalin and literature search, the reversed-phase

chromatography was assumed to be a good choice. Commercially available RP-18 and

CN-modified silica gel plates were tested as stationary phases. However, the obtained

results showed that, regardless to the mobile phase composition, CN-modified silica gel,

due to considerable peak tailing, was not a suitable stationary phase for the analysis of

amygdalin. Such behavior is probable caused by strong interactions between π aromatic

parts of amygdalin and CN groups on the sorbent surface. Therefore, the composition of

the mobile phase was optimized by spotting standard solution on the RP-18 plates and by

running in different solvent systems. Various solvent mixtures consisting of water and

acetonitrile or methanol in different proportions (10 - 50%, v/v, with increment of 10%)

were tested. Regarding to the peaks’ shape and symmetry, the RP-18 silica appeared to be

11
a suitable stationary phase, whereby the mobile phases containing ACN gave better

results than those obtained with a methanol as organic modifier. Finally, the solvent

mixture composed of acetonitrile:water (50:50, v/v) was chosen as a suitable mobile

phase because it provided well resolved compact spots, as well as sharp and well defined

peaks (Fig. 2).

Method validation

Results of validation studies on the developed RP-HPTLC method for quantitative

determination of amygdalin are given below (Table 2).

Linearity and range

Linearity of the analytical method is its ability, within a given range, to obtain test results

that are directly, or through a mathematical transformation, proportional to concentration

of the analyte. Calibration plots of the peak areas against the concentrations were linear

in the range of 2.5-50.0 µg /spot. Linear regression equation was found to be Y = 903.6

(±243.7) + 439.2 (±8.8)X. For this equation, the correlation coefficient, R2, was 0.9984.

Precision

Precision was determined as intra-day repeatability, and intermediate precision was

determined as inter-day method variability at three concentration levels. Results of the

experiments were expressed in terms of %RSD, and are shown in Table 3.

12
As the RSD values for repeatability and intermediate precision studies were < 2%, the

proposed method was found to provides acceptable both intra-day and inter-day

precision. Moreover, the results obtained at lowest concentration level (0.81 % RSD for

for intra-day and 1.89 % RSD for for inter-day), i.e. the least precise results, were

statistically compared with intra-day and inter-day precision obtained by HPLC method

(0.83, i.e. 0.31 % RSD). Since the calculated F-values (F = 1.04 for intra-day and F =

36.90 for inter-day precision) were below the critical values (Fcr = 39 for both intra-day

and inter-day precision) it can be concluded that there is no significant difference

between precision of two methods.

Accuracy

Accuracy was determined by applying standard addition method. The amounts of

standard were added at three concentration levels. The experiment was repeated in

triplicates. Recovery was calculated as a ratio of amount found and added, and the results

are given in Table 4. Accuracy was expressed as recovery percentage.

As it can be seen from Table 4, good recoveries in the range from 99.98 to 100.56%, with

low standard deviation values, were obtained for different added concentrations. These

results, i.e. recovery values close to 100% and their low standard deviation values (< 1%

RSD), show that the HPTLC method is highly accurate and, in terms of accuracy,

comparable to the recently proposed HPLC method. [12]

Limit Of Detection (LOD) And Limit Of Quantification (LOQ)

13
Signal-to-noise ratios of 3:1 and 10:1 were obtained for LOD and LOQ, respectively. The

LOD and LOQ were found to be 1.28 µg and 4.28 µg per spot, respectively.

Specificity

Specificity of the method was determined by comparing RF value of the peak of the

standard with the corresponding peak of amygdalin in the sample. The two peaks had the

same RF values.

In summary, the developed HPTLC method is specific, precise and accurate. The

results of statistical analysis demonstrate that the values of all validation parameters are

acceptable and, therefore, the method is suitable for the quantitative determination of

amygdalin.

Extraction

In this study we have explored an alternative to Soxhlet extraction - accelerated solvent

extraction (ASE).

Comparative studies of accelerated solvent extraction, Soxhlet extraction, and ultrasonic-

assisted extraction were carried out and reported in literature earlier.[23] Although in some

cases the highest efficiency was obtained by Soxhlet extraction, ASE exhibits better

reproducibility, it is much faster and reduces solvent consumption, so appears to be a

promising alternative to classical extraction methods. ASE method has not been reported

in the literature as a method for the extraction of amygdalin. Based on literature data

14
dealing with application of this method for the extraction of other compounds from plant

material and also taking into account all the advantages of this method, ASE was chosen

as a quite suitable method for quantitative extraction of amygdalin.

Optimization Of Extraction

Optimal extraction conditions were determined by the design of experiment (DoE). Table

5 presents the experimental conditions used in optimizing the extraction of amygdalin.

The extraction efficiency in each experiment is expressed as the concentration of

amygdalin in µg/g. Advantage of DoE is in extracting as much information as possible

from the limited set of experiments. Significance of factors was estimated, and the

obtained results are summarized in Table 6.

By analysis of DoE data, the optimal factor values were established. Figure 3 shows

prediction profiles, desirability function and optimal values of factors obtained for

maximal desirability.

Analysis of the applied DoE data show that extraction temperature and sample weight

were significant factors for ASE extraction of amygdalin (Table 6). Optimal amygdalin

extraction conditions, determined by the design analysis, were: temperature 110 oC,

sample mass 2.5 g, three extraction cycles of 10 minutes each. Under the optimal

conditions the total volumes of extracts were between 30 and 40 mL.

Extraction Of Plant Material

15
After establishing the optimal extraction conditions, ASE method was applied to the

extraction of amygdalin from fruit kernel samples.

After the ASE, the extract was used without purification and concentration, opposite to

the Soxhlet extraction method, where this is not possible because of the large amount of

the solvent used. In that way ASE simplifies sample preparation and reduces time of

analysis as well as the exposure to organic solvents. Therefore, the use of ASE in the

daily laboratory practice might be recommended because it is simple, rapid and

environmentally friendly extraction method.

Application Of Proposed Method To Fruit Kernel Samples

The applicability of the developed and validated HPTLC method was tested by

determination of amygdalin content in twelve samples of fruit kernels.

The obtained results are given in Table 7.

It was established that the concentration of amygdalin in kernel samples was between

0.9988-5.3881 g/100 g. The highest values were found in apricots and the lowest in

almond kernels.

Experimental results demonstrate that the proposed HPTLC method can be successfully

applied for the rapid and simple determination of amygdalin content after extraction by

ASE method.

16
 CONCLUSION

In this study, a new method for the extraction (ASE) and determination (HPTLC) of

amygdalin in fruit kernel samples was developed and validated. The methods for ASE

extraction and HPTLC quantification of amygdalin were not reported earlier in the

literature. The presented HPTLC method can be used for quantitative determination of

amygdalin in various samples. The results proves that HPTLC method is specific,

precise, accurate, and repeatable, so it could be said that amygdalin might be determined

by HPTLC with a high level of confidence. Furthermore, compared to HPLC, HPTLC

seems to have some advantages: it is simpler, faster, less expensive, more flexible, the

consumption of organic solvents as well as the analysis time is lower, and it has higher

throughtoutput. ASE extraction technique contributed to the proposed HPTLC method to

be faster and simpler, and to significant reduction in solvents, as well, what makes this

method more acceptable in terms of principles of “green chemistry” than other methods

reported earlier in the literature.

ACKNOWLEDGMENTS

This work was supported by the Ministry of Education and Science of the Republic of

Serbia (Grant No. 172017). The authors wish to thank to Dr Milica Fotirić Aškić (Faculty

of Agriculture, University of Belgrade) for providing selected fruit kernel samples.

REFERENCES

17
[1] Abdel-Rahman, M. K. 2011. Can apricot kernels fatty acids delay the atrophied

hepatocytes from progression to fibrosis in dimethylnitrosamine (DMN)-induced liver

injury in rats? Lipids in Health and Disease 10:114.

[2] Viorica-Mirela, G., Socaciu, C., Jianu, I., Florica, R., Florinela, F. 2006. Identification

and quantitative evaluation of amygdalin from apricot, plum andpeach oils and kernels.

Buletin USAMV-CN 62:246-253.

[3] Yu, L., Ye, H., Zheng, L., Chen, L., Chu, K., Liu, X., et al. 2011. Determination of

the epimerization rate constant of amygdalin by microemulsion electrokinetic

chromatography. Electrophoresis 32:218–222.

[4] Chang, H. K., Shin, M. S., Yang, H.Y., Lee, J. W., Kim, Y.S., Lee, M. H., et al. 2006.

Amygdalin Induces Apoptosis through Regulation of Bax and Bcl-2 Expressions in

Human DU145 and LNCaP Prostate Cancer Cells. Biological & Pharmaceutical Bulleti.

29(8):1597-1602

[5] Yildirim, F. A., & Askin, M. A. 2010. Variability of amygdalin content in seeds of

sweet and bitter apricot cultivars in Turkey. African Journal of Biotechnolog 9(39):6522-

6524.

[6] Femenia, A., Rossell, C., Mulet, A., & Caiiellas, J. 1995. Chemical Composition of

Bitter and Sweet Apricot Kernels. Journal of Agricultural and Food Chemistry 43:356-

361.

[7] Gomez, E., Burgos, L., Soriano, C., & Marn, J. 1998. Amygdalin Content in the

Seeds of Several Apricot Cultivars. Journal of the Science of Food and Agriculture

77:184-186.

18
[8] Jiagang, D., Li, C., Wang, H., Hao, E., Du, Z., Bao, C., et al. 2011. Amygdalin

mediates relieved atherosclerosis in apolipoprotein E deficient mice through the induction

of regulatory T cells. Biochemical and Biophysical Research Communications 411:523–

529.

[9] Yan, J., Tong, S., & Li, J. 2006. Preparative Isolation and Purification of Amygdalin

from Prunus armeniaca L. with High Recovery by High-Speed Countercurrent

Chromatography. Journal of Liquid Chromatography & Related Technologies 29:1271-

1279.

[10] Kwon, H. J., Jeong, J. S., Lee, Y. M., Hong, S. P. 2008. A reversed-phase high-

performance liquid chromatography method with pulsed amperometric detection for the

determination of glycosides. Journal of Chromatography A 1185:251-257.

[11] Lv, Q., Yi, L. Z., Yi, H. Y., & Liang,Y. Z. 2012. Chromatographic fingerprint

of Semen Armeniacae Amarae based on high-performance liquid chromatogram and

chemometric methods. Anaytical Methods 4:299–308.

[12] Savić, I., Nikolić, V., Savić, I., Nikolić, Lj., & Stanković, M. 2012. Development

and validation of HPLC method for the determination of amygdalin in the plant extract of

plum kernel. Research Journal of Chemistry and Environment 16(4):80-86.

[13] Lee, J., Zhang, G., Wood, E., Castillo, C. R., & Mitchell, A. E. 2013. Quantification

of Amygdalin in Nonbitter, Semibitter, and Bitter Almonds (Prunus dulcis) by UHPLC-

(ESI)QqQ MS/MS. Journal of Agricultural and Food Chemistry 61:7754−7759.

[14] Anastas, P. T. 1999. Green Chemistry and the Role of Analytical Methodology

Development. Critical Reviews in Analytical Chemistry 29(3):167-175.

19
[15] Welch, C. J., Wu, N., Biba, M., Hartman, R., Brkovic, T., Gong, X., et al. 2010.

Greening analytical chromatography. Trends in Analytical Chemistry 29(7):667-680.

[16] Milojković-Opsenica, D., Andrić, F. 2014. High Performance Thin-Layer

Chromatography. In: Green Chromatographic Techniques ed. A. M. Inamuddin, 81-101.

Springer.

[17] Morlock, G. E., & Schwack, W. 2007. The Contribution of Planar Chromatography

to Food Analysis. Journal of Planar Chromatography 20(6):399-406.

[18] Tao, Z., Sato, M., Han,Y., Tan, Z., Yamaguchi, T., Nakano, T. 2011. A simple and

rapid method for histamine analysis in fish and fishery productsby TLC determination.

Food Control 22:1154-1157

[19] Liao, Q. G., Li, W. H., Luo, L. G. 2012. Applicability of accelerated solvent

extraction for synthetic colorants analysis in meat products with ultrahigh performance

liquid chromatography–photodiode array detection. Analytica Chimica Acta 716:128-

132.

[20] Spirić, A., Trbović, D., Vranić, D., Djinović, J., Petronijević, R., Matekalo-Sverak,

V. 2010. Statistical evaluation of fatty acid profile and cholesterol content in fish

(common carp) lipids obtained by different sample preparation procedures. Analytica

Chimica Acta 672:66–71.

[21] International Conference on Harmonization of Technical Requirements for

Registration of Pharmaceuticals for Human Use (ICH). 2005. Validation of Analytical

Procedures: Text and Methodology Q2 (R1).

20
[22] Lundstedt,T., Seifert, E., Abramo, L., Thelin, B., Nyström, A., Pettersen, J., et al.

1998. Experimental design and optimization Chemometrics and Intelligent Laboratory

Systems 42(1-2):3-40.

[23] Shen, J., Shao, X. 2005. A comparison of accelerated solvent extraction, Soxhlet

extraction, and ultrasonic-assisted extraction for analysis of terpenoids and sterols in

tobacco. Analytical and Bioanalytical Chemistry 383:1003–1008.

21
Table 1 The chromatographic systems used.

No. Stationary phase Mobile phase composition

1 RP-18 10-50% ACN in water

2 RP-18 10-50% methanol in water

3 CN-silica gel 10-50% ACN in water

4 CN-silica gel 10-50% methanol in water

22
Table 2 Summary of validation data.

Parameters Amygdalin
Linearity range (µg / spot) 2.5-50.0
Correlation coefficient 0.9984
Limit of detection (µg / spot) 1.28

Lmit of quantification (µg 4.28

/spot)
Intra-day (% RSD) 0.81-1.15
Inter-day (% RSD) 1.32-1.89
Accuracy (% RSD) 0.31-0.83
Specificity Specific

23
Table 3 Precision of the method.

Compound Conc. (µg/spot) Intra-day Inter-day

Amount (%)a % RSD Amount (%)b % RSD

Amygdalin 40 99.02 1.15 99.32 1.32

20 100.34 1.09 103.44 1.76

10 99.71 0.81 103.7 1.89
a
n=6
b
n=3

24
Table 4 Recovery study.

Compound Amount of Recovery (%)a % RSD

amygdalin added

(µg/spot)

Amygdalin 10 100.56 0.83

20 99.98 0.31

40 100.04 0.75
a
n = 3 at each level

25
Table 5 The results of DoE.

Temperature Number of One cycle Sample mass Amygdalin

(oC) cycles duration (g) concentration

(min) (µg/g)*

70 1 10 2.5 666.55

70 3 2 2.5 610

70 1 2 0.5 423.15

70 3 10 0.5 614.78

90 2 6 1.5 827.7

90 2 6 1.5 806.8

110 1 10 0.5 534.33

110 3 2 0.5 754.43

110 1 2 2.5 875.41

110 3 10 2.5 1114.87

*
Determined by the HPLC method

26
Table 6 Estimation of DoE factors.

Factors Estimate Std Error t Ratio Prob>|t|

Temperature (oC)(70, 120.57 41.53605 2.90 0.0337*

110)

Sample weight 117.5175 41.53605 2.83 0.0367*

(g)(0.5, 2.5)

Number of cycles (1, 74.33 41.53605 1.79 0.1335

3)

One cycle duration 33.4425 41.53605 0.81 0.4573

(min)(2, 10)

27
Table 7 Content of amygdalin in fruit kernels.

Sample Amygdalin content

(g/100g)
Apricot 1 (Bela di imola) 4.5444
Apricot 2 (Palumela) 5.3881
Apricot 3 (Agat) 4.5851
Peach 1 (Flaminia spadoni) 2.8530
Peach 2 (Flaminia otomglo) 2.7129
Peach 3 (Flaminia samerset) 2.8597
Plum 1 (Lepotica dz) 3.9941
Plum 2 (Stenli) 2.6743
Plum 3 (Lepotica st. jul) 4.0831
Almond 1 1.0802
Almond 2 0.9988
Almond 3 1.0092

28
Figure 1. The structure of amygdalin.

29
Figure 2. Chromatogram of amigdalin standard solution.

30
Figure 3. Prediction Profiler of DoE for the optimization of the accelerated solvent

extraction of amygdalin.

31