Amygdalin in Bitter and Sweet Seeds of Apricots

Abstract

Hydrogen cyanide (HCN) poisoning due to amygdalin (AMY) in apricot seeds is one of the public health issues in Turkey. The aim of this study was to investigate the AMY content of 13 different apricot seeds including bitter and sweet ones, and which are either sulfurized or roasted. The AMY content was determined by high performance liquid chromatography (HPLC). Release of HCN was predicted and total amount of seeds which can cause poisoning was calculated. The mean AMY content of bitter seeds was 26 ± 14 mg g−1 and that of sweet seeds was 0.16 ± 0.09 mg g−1. The consumption of small amounts of bitter seeds may cause cyanide poisoning.

Full text

Amygdalin in bitter and sweet seeds of apricots
Nazan Karsavuran
a
, Mohammad Charehsaz
b
, Hayati Celik
c
, Bayram Murat Asma
d
,
Cengiz Yakıncı
a
and Ahmet Aydın
b
*
a
Department of Pediatry, Faculty of Medicine, Inonu University, Malatya, Turkey;
b
Department of
Toxicology, Faculty of Pharmacy, Yeditepe University, Istanbul, Turkey;
c
Department of Analytical
Chemistry, Faculty of Pharmacy, Yeditepe University, Istanbul, Turkey;
d
Apricot Research and
Application Center, Inonu University, Malatya, Turkey
(Received 27 August 2014; accepted 14 March 2015)
Hydrogen cyanide (HCN) poisoning due to amygdalin (AMY) in apricot seeds is one
of the public health issues in Turkey. The aim of this study was to investigate the
AMY content of 13 different apricot seeds including bitter and sweet ones, and which
are either sulfurized or roasted. The AMY content was determined by a high-
performance liquid chromatography. Release of HCN was predicted and total amount
of seeds which can cause poisoning was calculated. The mean AMY content of bitter
seeds was 26 §14 mg g
¡1
and that of sweet seeds was 0.16 §0.09 mg g
¡1
. The
consumption of small amounts of bitter seeds may cause cyanide poisoning.
Keywords: cyanide; HPLC; poisoning; amygdalin; apricot
1. Introduction
Plants which may release hydrogen cyanide (HCN) upon acidic or enzymatic hydrolysis
are known as cyanogenic plants, and 23 different glycosides have been discovered in
2000 cyanogenic plants, the most common being amygdalin (AMY) (Akintonwa and
Tunwashe 1992). Due to high AMY content and relatively more easy HCN release, apri-
cot seeds draw attention of toxicologists among other seeds, such as those of apple and
peach. Cyanogenic glycosides are not toxic on their own but when ingested they are
hydrolyzed by b-glycosidase or emulsion (Newton et al. 1981; Suchard, Wallace, and
Gerkin 1998; Dorr and Paxinos 1978; Haisman and Knight 1967) to yield one molecule
HCN, two molecules glycose, and one molecule benzaldehyde (Herbert 1979). As the
seeds containing AMY are consumed, HCN is released under the acidic conditions of the
stomach (Speijers 2014; Megarbane et al. 2003). The HCN content of apricot seeds is
estimated in the range of 0.14.1 mg g
¡1
(an average of 2.9 mg g
¡1
) (Holzbecher, Moss,
and Ellenberger 1984) and the amount of cyanide in the seeds differ depending on the
kind and the region where they are cultivated. Bitter seeds contain more AMY than sweet
ones and the bitterness gets stronger as the amount of AMY increases (Godfredsen et al.
1978). Cyanide can cause sudden deaths since it is rapidly taken up into cells to block the
mitochondrial electron transport within seconds (Shepherd and Velez 2008). The lowest
fatal oral dose of cyanide for humans was estimated as 0.56 mg kg
¡1
(Gettler and Baine
1938).
Apricot seeds are used in the cosmetics and drug industry and are an economically
important commodity of the Malatya region (Asma and Mısırlı 2007; Dwivedi and Ram
*Corresponding author. Email: ahmet.aydin@yeditepe.edu.tr
Ó2015 Taylor & Francis
Toxicological & Environmental Chemistry, 2015
http://dx.doi.org/10.1080/02772248.2015.1030667
Downloaded by [Yeditepe Universitesi], [Ahmet Aydin] at 07:18 17 April 2015

2008). On the other hand, poisoning of children by apricot seeds is common (Akyildiz
et al. 2010; Cheok 1978). Hence, further investigation on harm/benefit comparison and
quantitative determination of AMY in apricot seeds is necessary. Therefore, the aim of
this study was to determine the AMY level in bitter and sweet seeds of 13 different spe-
cies of apricots, which are cultivated in the Malatya region of Turkey. Based on the
results, the toxicological risks to be associated with apricot seeds have been quantified.
2. Materials and methods
2.1. Sample collection and preparation
The apricot seeds that are investigated in our study have been supplied from the Apricot
Research Center at
_
In€
on€
u University, Malatya, Turkey. The seeds have been collected by
hand-picking at the harvest season. After the fruits were picked, their seeds were removed
and they were left to dry for two days at room temperature. The species commonly farmed
and chosen for this study in nonsulfurized form were Hacıhalilo
glu, Hasanbey, Kabaa¸sı,
¸C€
olo
glu, ¸Catalo
glu, So
gancı, ¸Sekerpare, Turfanda Eskimalatya, Alyanak, Paviot, Hungar-
ian Best, Ninfa, and Roksana; those that were commonly consumed in roasted, sulfurized,
or roasted and sulfurized forms were Hacıhalilo
glu and Kabaa¸s ı species.
Sulfurization was carried out in an isolated chamber (15 m
3
), built in apricot orchard.
Apricots stuffed into crates were placed in this chamber. About 1.5 kg of sulfur powder
(98%99% purity) for a ton of apricots was used for sulfurization. For this purpose, the
calculated amount of sulfur powder was melted on a stainless steel plate in the chamber
following burning to release SO
2
gas. Then the door of the chamber was closed and apri-
cots were exposed for 9 hours to the gas. Finally, the apricots were dried in open air for
three days and apricot seeds were removed at the end of this process.
The roasted sample was prepared in Petri dishes at 150 C for 15 minutes and
allowed to cool down to room temperature within 3 hours. This process is similar to the
industrial one.
2.2. Chemicals
AMY ([(6-O-b-D-glucopyranosyl-b-D-glucopyranosyl)oxy] (phenyl) aceto nitrile) was sup-
plied from Alfa Aesar GmbH (Karlsruhe, Germany). Methanol, ethanol, phosphoric acid,
and monosodium hydrogen phosphate monohydrate (NaH
2
PO
4
.H
2
O) were purchased from
Sigma-Aldrich (St. Louis, MO, USA). Disodium hydrogen phosphate heptahydrate
(Na
2
HPO
4
.7H
2
O) was supplied from Merck (Darmstadt, Germany). All chemicals used
were of analytical or high-performance liquid chromatography (HPLC) grade. Ultra-pure
distilled water was collected from millipore simplicity (Molsheim, France).
2.3. Apparatus and separation conditions
A liquid chromatographic instrument (HP 1100 Series, Agilent, Waldbronn, Germany)
equipped with autosampler and diode array detector (DAD) (G-1365B, Agilent) was used.
Separation was carried out with an Inertsil ODS-3 C18 column (250 £4.6 mm i.d., 5
mm). Isocratic elution was performed with a mixture of phosphate solution (pH D2.8;
0.2 mmol L
¡1
H
3
PO
4
and 1 mmol L
¡1
NaH
2
PO
4
): methanol (75:25 v/v) at a flow rate of
1.00 mL min
¡1
. The sample injection volume was 20 mL and the temperature was set to
30 C during the experiments.
2N. Karsavuran et al.
Downloaded by [Yeditepe Universitesi], [Ahmet Aydin] at 07:18 17 April 2015

2.4. Standard solutions and sample preparation
Stock solution of AMY was prepared in the mobile phase as the final concentration at
1gL
¡1
. This solution was used to prepare the standard solutions for the calibration curve.
The concentrations range was between 20 and 400 mg L
¡1
(20, 50, 100, 200, 300, and
400 mg L
¡1
). All of the subsequent dilutions for the standard solutions were prepared by
using the mobile phase.
Ten seeds from each kind of apricots have been powdered in a mortar and 0.5 g from
each powdered sample was weighed and transferred into a round-bottom flask containing
20 mL of ethanol. The sample was extracted for 2 hours with an extractor (Heidolph,
Schwabach, Germany) and then was placed into an ultrasonic bath for 30 minutes. Then,
the ethanol layer was removed and 20 mL of fresh ethanol was added to the residue. This
process was repeated three times. Then, the combined ethanolic extracts were transferred
into a round-bottom flask and ethanol was evaporated at 35 C with the rotary evaporator,
and the dried residue containing AMY was dissolved in 2 mL mobile phase. This solution
was filtered with 0.2 mm single-use filter (Sartorius, minisart RC 25, Goettingen,
Germany), and was placed into the autosampler of the HPLC instrument.
2.5. Method validation and statistical analysis
The analytical method validation was performed in accordance to International Confer-
ence on Harmonisation guidelines (ICH 2005). Assay validation parameters were linear-
ity, limit of detection (LOD), limit of quantification (LOQ), intra-day and inter-day
variation (Celik et al. 2014). Recovery of AMY extraction was determined by the method
of Zhou et al. (2007).
All statistical analyses were performed using SPSS software version 16.0. Data are
presented as mean §RSD. In terms of seed flavor, MannWhitney U test was applied
for the AMY content comparisons. A p<0.05 was considered statistically significant.
3. Results and discussion
A chromatogram of standard AMY and sample is shown in Figure 1. The LOD and LOQ
of HPLC method was 1.2 and 4.0 mg L
¡1
. Intra-day and inter-day variations were 0.8%
and 3.8%, respectively. Recovery was 91 §10%.
The AMY content of bitter apricot seeds is significantly higher than the sweet seeds
(p<0.05). The average AMY content of bitter seeds was calculated as 26 §14 mg g
¡1
(13.9644.41 mg g
¡1
), whereas for sweet seeds this amount was 0.16 §0.09 mg g
¡1
(<0.080.40) (Table 1).
Figure 1. Representative chromatogram of (a) standard AMY (100 mg L
¡1
,R
t
D11.90 min),
(b) an apricot seed (R
t
D11.80 min).
Toxicological & Environmental Chemistry 3
Downloaded by [Yeditepe Universitesi], [Ahmet Aydin] at 07:18 17 April 2015

Between 13 species, Paviot was the first place with 44 mg g
¡1
AMY content. The
amount of HCN that would be released from this amount of AMY is 2.5 mg. To assess
the toxicological risk of this seed, the following calculation was done. Considering the
fatal oral dose of cyanide as 0.56 mg kg
¡1
, a child weighing 20 kg can be poisoned seri-
ously with 11.2 mg of cyanide. This toxic cyanide level can be achieved by eating 4.5 g
of total seed of Paviot (Table 1).
The highest AMY content among the sweet seeds was found in Hasanbey kind as
0.4 mg g
¡1
. This apricot is the most commonly produced and consumed in Malatya and
overall Turkey. With this AMY level, 0.022 mg HCN is released from per gram of seed.
Approximately 510 g seed would have to be ingested to cause poisoning of a child weigh-
ing 20 kg.
The effect of sulfurization and roasting on the AMY content of seeds was also evalu-
ated for Hacıhalilo
glu and Kabaa¸sı kind. While the raw form of Hacıhalilo
glu was found
to contain 0.25 mg g
¡1
AMY, roasted form had 0.30 mg g
¡1
. If the raw seed was sulfu-
rized, the content was 0.20 mg g
¡1
. On the other hand, sulfurized and roasted form con-
tained the lowest AMY as 0.15 mg g
¡1
(Table 2). For Kabaa¸sı kind, the amount also
decreased considerably as the seeds were both sulfurized and roasted (Table 2). These
results suggest that processed seeds are less toxic and hence more suitable for consump-
tion in diet and for use in areas such as cosmetics and drugs.
It is obvious that sulfurization process significantly decreases the amount of AMY in
the seed. However, the effect of roasting on the AMY content needs to be clarified with
further studies.
AMY result of our study is consistent with that of Gomez et al. (1998) and Femenia
et al. (1995), but the results of Misirli, Sefer, and G€
ulcan (2006) were somehow lower
(8.2 mg g
¡1
) for the bitter seeds and higher (3.44 mg g
¡1
) for the sweet seeds (Table 3).
Table 1. Amount of AMY in raw form of sweet and bitter apricot seeds.
Kind of apricot
Taste of
seed
Amount of
AMY (mg g
¡1
)
Predicted HCN
release (mg g
¡1
)
Amount which
cause poisoning
(g of seed)

Hasanbey Sweet 0.40 0.022 509
Hacıhalilo
glu Sweet 0.25 0.014 800
Hungarian best Sweet 0.23 0.013 862
So
gancı Sweet 0.13 0.007 1600
Kabaa¸sı Sweet 0.10 0.006 1867
Roksana Sweet 0.10 0.006 1867
¸C€
olo
glu Sweet <0.08 <0.005 
¸Catalo
glu Sweet <0.08 <0.005 
¸Sekerpare Sweet <0.08 <0.005 
Paviot Bitter 44.41 2.50 4.48
Alyanak Bitter 31.30 1.78 6.29
Ninfa Bitter 15.42 0.87 12.87
Primeum Eskimalatya Bitter 13.96 0.79 14.17

To assess the toxicological risk of seeds, following calculation was done. With accepting the knowledge stating
that 457 g of AMY releases 26 g of cyanide, cyanide release from per gram of seed was calculated. Considering
the fatal oral dose of cyanide as 0.56 mg kg
¡1
, amount of seed which cause poisoning in a child weighing 20 kg
was calculated.
4N. Karsavuran et al.
Downloaded by [Yeditepe Universitesi], [Ahmet Aydin] at 07:18 17 April 2015

In spite of the difference between the contents found in these two studies, our results were
still comparable and the differences were due to the investigation of different kinds of
apricots.
Another study carried out by Yıldırım and A¸skın (2010) reported a significant differ-
ence between the AMY amounts in the bitter and sweet seeds. When these results were
compared with ours, AMY in bitter seeds were not different (pD0.07), but that of sweet
ones were statistically different (pD0.001) (Table 3). The differences between studies
may be explained by the increased number and kind of apricots investigated as well as
the differences in the treatment of the soil, in the degree of drought, watering, harvest
time, and the method used for preparing the seeds.
4. Conclusion
In light of this research, new food safety regulations and applications should be developed
for apricot seeds. Bitter seeds of apricot are not suitable for Turkish Food Codex and,
Table 2. Amount of AMY in raw, sulfurized, nonsulfurized roasted, and sulfurized roasted form of
seeds from Hacıhalilo
glu and Kabaa¸sı species.
Species Type
Taste
of seed
AMY
(mg g
¡1
)
Predicted
HCN release
(mg g
¡1
)
Amount which
cause poisoning
(g of seed)

Raw Sweet 0.250 0.0140 800
Sulfurized Sweet 0.200 0.0110 1018
Hacıhalilo
glu Nonsulfurized roasted Sweet 0.300 0.0170 659
Sulfurized roasted Sweet 0.150 0.0085 1318
Raw Sweet 0.098 0.0055 2036
Sulfurized Sweet <0.080 <0.0045 
Kabaa¸sı Nonsulfurized roasted Sweet 0.090 0.0051 2196
Sulfurized roasted Sweet 0.015 0.0085 1318

To assess the toxicological risk of seeds, following calculation was done. With accepting the knowledge stating
that 457 g of amygdalin (AMY) releases 26 g of hydrogen cyanide (HCN), HCN release from per gram of seed
was calculated. Considering the fatal oral dose of cyanide as 0.56 mg kg
1
, amount of seed which cause poison-
ing in a child weighing 20 kg was calculated.
Table 3. Amount of AMY according to the taste of apricot seeds in present and other studies.
Study Taste of seed Average AMY amount (mg g
¡1
)
Present study Sweet 0.16 (<0.080.40)

Mısırlı,Sefer and G€
ulcan (2006) Sweet 3.44
Yıldırım and A¸skın (2010) Sweet 8.61

Femenia et al. (1995) Sweet 0.00
Present study Bitter 26.3 (13.9644.41)
Mısırlı, Sefer and G€
ulcan (2006) Bitter 8.22
Yıldırım and A¸skın (2010) Bitter 55.59
Femenia et al. (1995) Bitter 55.00

p<0.05 versus result in bitter seed of present study.

p<0.001 versus result in sweet seed of present study.
AMY: amygdalin.
Toxicological & Environmental Chemistry 5
Downloaded by [Yeditepe Universitesi], [Ahmet Aydin] at 07:18 17 April 2015

hence, could not be sold with a license in Turkey. However, during the removal and stor-
age by the farmers or separation of the seeds and fruits in the factories, children can easily
access these seeds. Therefore, new and effective regulations on this subject are needed,
and the farmers and producers should be informed about the proper methods of storing,
selling, and selection of the seeds.
Acknowledgments
This work was supported by In€
on€
u University, Scientific Research Projects Coordination Unit. Also
authors want to say their special thanks to Prof. Dr Ali Adnan Hayaloglu for supplying some kind
of apricot seeds.
Disclosure statement
No potential conflict of interest was reported by the authors.
References
Akintonwa, A., and O.L. Tunwashe. 1992. “Fatal Cyanide Poisoning from Cassava-Based Meal.”
Human and Experimental Toxicology 11 (1): 4749.
Akyildiz, B.N., S. Kurtoglu, M. Kondolot, and A. Tun¸c. 2010. “Cyanide Poisoning Caused by
Ingestion of Apricot Seeds.” Annals of Tropical Paediatrics 30 (1): 3943.
Asma, B.M., and A. Mısırlı. 2007. “Kayısı ¸Cekirde
gi.” Hasad 261: 5558.
Celik, H., E. Ariburnu, M.S. Baymak, and E. Yesilada. 2014. “A Rapid Validated HPLC Method for
Determination of Sulforaphane and Glucoraphanin in Broccoli and Red Cabbage Prepared by
Various Cooking Techniques.” Analytical Methods 6: 45594566.
Cheok, S.S. 1978. “Acute Cassava Poisoning in Children in Sarawak.” Tropical Doctor 8 (3):
99101.
Dorr, R.T., and J. Paxinos. 1978. “The Current Status of Laetrile.” Annals of Internal Medicine 89
(3): 389397.
Dwivedi, D.H., and R.B. Ram. 2008. “Chemical Composition of Bitter Apricot Kernels from Ladakh,
India.” Acta Horticulturae 765: 335338.
FAO/WHO. 2011. Joint FAO/WHO Expert Committee on Food Additives Seventy-fourth Meeting.
Rome, Summary and Conclusions [Internet]. Accessed October 12, 2014. http://www.who.int/
foodsafety/chem/jecfa/summaries/Summary74.pdf.
Femenia, A., C. Rosello, A. Mulet, and J. Canellas. 1995. “Chemical Composition of Bitter and
Sweet Apricot Kernels.” Journal of Agricultural and Food Chemistry 43 (2): 356361.
Gettler, A.O., and J.O. Baine. 1938. “The Toxicology of Cyanide.” The American Journal of the
Medical Sciences 195 (2): 182198.
Godfredsen, S.E., A. Kjaer, J.D. Madsen, and M. Sponholtz. 1978. “Bitterness in Aqueous Extracts
of Apricot Kernels.” Acta Chemica Scandinavica B 32: 588592.
Gomez, E., L. Burgos, C. Soriano1, and J. Marin. 1998. “Amygdalin Content in the Seeds of Several
Apricot Cultivars.” Journal of the Science of Food and Agriculture 77 (2):184186.
Haisman, D.R., and D.J. Knight. 1967. “The Enzymatic Hydrolysis of Amygdalin.” Biochemical
Journal 103 (2): 528534.
Herbert, V. 1979. “Laetrile: The Cult of Cyanide: Promoting Poison for Profit.” The American Jour-
nal of Clinical Nutrition 32 (5): 11211158.
Holzbecher, M.D., M.A. Moss, and H.A. Ellenberger. 1984. “The Cyanide Content of Laetrile Prep-
arations, Apricot, Peach and Apple Seeds.” Journal of Toxicology Clinical Toxicology 22 (4):
341347.
ICH. 2005. “(International Conference on Harmonisation) of Technical Requirements for Registra-
tion of Pharmacuticals for Human Use, Topic Q2 (R1): Validation of Analytical Procedures:
Text and Methodology.” Accessed Agust 2014. www.ich.org.
Megarbane, B., A. Delahaye, D. Goldgran-Toledano, and F.J. Baud. 2003. “Antidotal Treatment of
Cyanide Poisoning.” Journal of the Chinese Medical Association 66 (4): 193203.
6N. Karsavuran et al.
Downloaded by [Yeditepe Universitesi], [Ahmet Aydin] at 07:18 17 April 2015

Misirli, A., F. Sefer, and R. G€
ulcan. 2006. “A Research on Phenolic and Cyanogenic Compounds in
Sweet and Bitter Kernelled Apricot Varieties.” Acta Horticulturae 701: 167169.
Newton, G.W., E.S. Schmidt, J.P. Lewis, E. Conn, and R. Lawrence. 1981. “Amygdalin Toxicity
Studies in Rats Predict Chronic Cyanide Poisoning in Humans.” The Western Journal of Medicine
134 (2): 97103.
Shepherd, G., and L. Velez. 2008. “Role of Hydroxocobalamin in Acute Cyanide Poisoning.”
Annals of Pharmacotherapy 42 (5): 661669.
Speijers, G. 2014. “Cyanogenic Glycosides.” In International Programme on Chemical Safety (IPCS).
Accessed August 1, 2014. http://www.inchem.org/documents/jecfa/jecmono/v30je18.htm.
Suchard, J.R., K.L. Wallace, and R.D. Gerkin. 1998. “Acute Cyanide Toxicity Caused by Apricot
Kernel Ingestion.” Annals of Emergency Medicine 32: 742744.
Yıldırım, F.A., and M.A. A¸skın. 2010. “Variability of Amygdalin Content in Seeds of Sweet and
Bitter Apricot Cultivars in Turkey.” African Journal of Biotechnology 9 (39): 65226524.
Zhou, C., K. Chen, C. Sun, Q. Chen, W. Zhang, and X. Li. 2007. “Determination of Oleanoli Acid,
Ursolic Acid and Amygdalin in the Flower of Eriobotrya japonica Lindl by HPLC.” Biomedical
Chromatography 21 (7): 755761.
Toxicological & Environmental Chemistry 7
Downloaded by [Yeditepe Universitesi], [Ahmet Aydin] at 07:18 17 April 2015