Talanta 53 (2000) 69 74

www.elsevier.com/locate/talanta

The use of simulation in the EPR spin probe technique for

detection of irradiated seeds
M. Maral Sunnetcioglu a,*, Dilek Dadayl b
a

Department of Physics Engineering, Faculty of Engineering, Hacettepe Uni6ersity, 06532 Beytepe, Ankara, Turkey
Department of Physics, Faculty of Art and Science, Zonguldak Karaelmas Uni6ersity, 67100 Zonguldak, Turkey

Received 28 September 1999; received in revised form 8 February 2000; accepted 14 February 2000

Abstract
An electron paramagnetic resonance (EPR) spin probe study of irradiated wheat seeds was performed depending
on irradiation dose. The structural changes in the membrane integrity were followed using aqueous solutions of
4-hydroxy-TEMPO (TANOL) spin probe and a line broadening material. In the studies dry seed embryos were kept
in these solutions for 150 min. The spectra were recorded at various times of air drying process. The simulation of
these spectra indicated a decrease in the water content of the embryos depending on the increasing irradiation dose.
This indicates the increase in the permeability of the membranes as a result of the radiation damage. From the decay
curves it is possible to determine about irradiation dose, however, this approach is not very successful at close
irradiation doses. 2000 Elsevier Science B.V. All rights reserved.
Keywords: Electron paramagnetic resonance; Nitroxide spin probes; Irradiated wheat; Simulation of electron paramagnetic
resonance spectra

1. Introduction
Irradiation of food for preservation is a common technique accepted by many countries. This
can be utilized for various purposes such as killing
bacteria, viruses, insects or to delay the ripening
of some fruits. The permitted limit of irradiation
dose in foods was stated as 10 kGy [1]. Therefore,
it is necessary to develop convenient methods
capable to distinguish between irradiated food
and unirradiated one and also to establish the
* Corresponding author. Tel.: + 90-312-2977224; fax: +90312-2352550/2354341.

amount of irradiation dose [2]. Frequently used

techniques are thermoluminescence and electron
paramagnetic resonance (EPR). After the first
applications in 1970s [35], EPR is now a leading
technique in the search of irradiated foods [610].
This technique is based on the presence of a
substantial amount of free radicals in the interested food. The radicals produced, as a result of
g-irradiation, usually have a limited lifetime and
in storage the numbers of radicals decay to their
background values. The lifetime of free radicals in
seeds is approximately 30 days [1114]. EPR spin
probe technique is an important tool to investigate samples, which have insufficient amount of

0039-9140/00/$ - see front matter 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 0 3 9 - 9 1 4 0 ( 0 0 ) 0 0 3 7 3 - 8

70

M.M. Sunnetcioglu, D. Dadayl / Talanta 53 (2000) 6974

free radicals for direct EPR applications [15,16].

Paramagnetic probes are introduced into the sample and they transfer valuable information about
the dynamic and structural changes in their environments without perturbing this environment.
This technique has a wide application in biological systems such as model membranes, blood cells,
proteins [17 21]. Recently, spin probe technique
was applied to the investigation of plant seeds
[22]. After this first application the technique was
utilized to test the viability of seeds [2326]. It
was first applied to the investigation of irradiated
seeds in 1997 [27] and further studies followed
[28,29].
The aim of the current study is to develop a
more practical way for detection of irradiated
seeds using spin probe technique together with
simulation methods and to test the ability of the
spin probe method for long storage times.

2. Experimental
Irradiated and control Durum wheat samples
(c.v. Kunduru 1149), harvested in 1994, were used
in the studies. Prior studies, germination tests
were carried out and 98% germination was obtained. Irradiation was performed at 1.0, 2.5, 5.0,
10.0, 20.0 kGy doses using g-irradiation from
calibrated 60Co source at Saraykoy Nuclear Research Institute, Ankara. The absorbed dose was
checked by Fricke dosimetry. For spin probe
studies, after irradiation at May 1996, samples
were stored in a steel cabinet at room temperature
for approximately 2 years. Experiments were performed between September 1998 and July 1999.
Dry embryos of wheat seeds were used in the
studies. Sample preparation was performed as
stated in a previous study [25]. Aqueous solutions
(10 mM) of 4-hydroxy-TEMPO (TANOL) were
prepared including the line broadening material
potassium ferricyanide (K3Fe(CN)6) at 250 mM
concentration. The presence of potassium ferricyanide provides the disappearance of the signal
from extracellular regions as a result of linebroadening. For damaged cells following the leakage it
causes an increase in the linewidth of the signal
from intracellular regions. In our previous studies,

the irradiation dose dependence was followed via

the rehydration curves [27,29]. In this study, two
different methods were tried to shorten the spectrometer times and, instead of recording a number
of spectra, for each sample only one spectrum was
recorded.
In the first method (method 1) certain amount
of prepared solution (: 0.5 cm high in the capillary) was taken into a 1 mm i.d. capillary and
single dry embryos were kept in the solution for
150 min. At the end of the waiting time, the
solution in the capillary was absorbed by a paper
and the spectra of the samples recorded
immediately.
In the second method (method II) five dry
embryos were soaked for 150 min in the solution
in a glass tube. After this time embryos were
taken onto glass plates using a paintbrush and left
to air-dry for 15 min. Then the spectra were
recorded.
All of the experiments were repeated at least
three times. The spectra were recorded at room
temperature by the use of a Varian E-9 and a
Bruker EMX X-band spectrometers using the following spectral conditions; modulation amplitude,
0.05 mT; microwave power, 2 mW.

3. Results and discussion

The recorded spectra for control and irradiated
samples using both method I and method II indicate a change in the spectrum shape depending on
irradiation dose. This change is especially effective
at m1 = 1 high field line, which splits into lipid
and aqueous parts [24,25]. The reason of these
relative changes in the spectra is due to changes in
the membrane permeability as a result of irradiation. In healthy cells the plasma membrane is
impermeable to potassium ferricyanide. Since the
signal from the solution is completely removed,
recorded spectra consist of only lipid and polar
signals from intracellular regions, however, since
some of the cells are damaged within embryo,
polar signal consists of two parts. Therefore, the
experimental spectra can be simulated using the
sum of the signals from three main regions.

M.M. Sunnetcioglu, D. Dadayl / Talanta 53 (2000) 6974

1. TANOL/water in healthy cells (water

percentage).
2. TANOL/water +line broadening material in
damaged cells.
3. Lipid parts both in healthy and damaged cells
(lipid percentage).

3.1. Simulation of spectra

In the calculation of the spectra, each line was
represented as a Lorentzian Gaussian sum. In
obtaining the first derivative line, the Gaussian
percentages represent the doubly integrated intensities [30]. Simulation of the spectra was completed in three steps. First, the beginning values of
the parameters were determined from a sixparameter fit of 10 mM TANOL/water and 10
mM TANOL/soyabean oil samples. The fitted
parameters were derivative line widths, g values,
nitrogen hyperfine coupling constant (A) and the
percentage of the Gaussian line. In the second
step, the signals from embryo were fitted by keeping g, A values and Gaussian percentages for
water and lipid parts constant. A 12-parameter fit
was performed for nine line widths, B0 main field
value and for percentages of TANOL in aqueous
and lipid environments. The remaining part represents TANOL in aqueous regions including the

71

line broadening material. These calculations were

done to fine tune the line widths of TANOL in
the intracellular aqueous and lipid regions. Only
control and 20 kGy samples were used since they
represent high and low water limits, respectively.
After obtaining satisfactory results from the
above calculations, in the last step, all spectra
were simulated by keeping the line widths of
TANOL signals from undamaged intracellular
aqueous parts and from lipid parts constant. The
water, lipid percentages and the line widths of
TANOL signal in the aqueous regions including
the line broadening material were left as variables.
Calculations were performed using a FORTRAN
programme including Simplex method as a subroutine [31].
The simulation results for method I and
method II are given in Figs. 1 and 2, respectively.
The average standard deviation in the simulation
of the spectra was smean = 0.03. The line widths of
the prepared solution including the line broadening agent were found for + 1, 0, 1, lines as
1.52, 1.53, 1.579 0.03 mT, respectively. As this
solution run into the intracellular regions, these
line widths decrease. The line widths from the
regions including the line broadening material
distributed about a mean value for each nitrogen
hyperfine line. There is a method dependent dif-

Fig. 1. The spectra recorded with the samples prepared according to method I. Solid line, experimental; dotted line, calculated.

72

M.M. Sunnetcioglu, D. Dadayl / Talanta 53 (2000) 6974

Fig. 2. The spectra recorded with the samples prepared according to method II. Solid line, experimental; dotted line, calculated.

ference in the mean values. For + 1, 0, 1 lines

the line widths were obtained as 0.75, 0.66, 0.879
0.06 mT for method I and 0.52, 0.45, 0.649 0.06
mT for method II, respectively. The dose and
method dependent changes in the water percentage can be followed from the spectra. Since in
method I the spectra were recorded immediately
after the soaking in Fig. 1, the water signal is
superior in the spectra of 0 5 kGy irradiation
dose range and even at 10 and 20 kGy. In Fig. 2,
however, the water percentages are lower. The
results of simulations justify these observations.
The spectra of the samples in method II were
recorded after 1 (Fig. 3) and 2.5 h. It was still
possible to follow the dose dependence but a
consideration of the spectra indicated the presence
of an additional signal of the probe from slow
motional region. To clarify the additional signal
the difference spectrum of 24 h air-dried sample
and lipid signal was drawn (Fig. 4). The line
width of the lipid signal was calculated from step
2 (Table 1).
The decay of water percentage and lipid percentage with increasing irradiation dose was
shown in Fig. 5A and B for method I and method
II, respectively. In both methods there is little
change in the lipid percentage with increasing
irradiation dose, however, there is a percentage
difference for lipid parts in these two methods.
The source of this might be migration of some

Fig. 3. The recorded spectra of the samples prepared as in

method II but air dried for 1 h.

M.M. Sunnetcioglu, D. Dadayl / Talanta 53 (2000) 6974

73

Fig. 4. The sample prepared as in method II, air dried for 24

h. (a) The recorded spectrum of the 5 kGy sample prepared as
in method II but air dried for 24 h (solid line), the lipid signal
calculated from step 2 (dotted line). (b) The difference spectrum. (c) Smoothed difference spectrum.
Fig. 5. The change in the water (squares) and lipid (triangles)
percentages depending on irradiation dose. (A) Method I, (B)
method II.

spin probes in aqueous environments towards

lipid parts as a result of dehydration.
The decay of water percentage is exponential in
both methods and from the exponential fit the
following results were obtained.
method I

y =5.5 + 37.1 exp

 
x
5.2

method II

y= 3.4+ 37.2 exp

 
x
4.2

Here, y is the water percentage (%) and x is the

irradiation dose (kGy). The decay in method II is
rapid relative to method I. From these curves it
might be possible to obtain the irradiation dose of
an unknown sample, however, as it can be seen

Table 1
The fitted values of experimental parameters from step 1 and step 2
Sample

Line widths (mT) (step 2)

m1

+1

TANOL/water (10 mM)

TANOL/lipid (10 mM)

0.160
0.110

0.140
0.115

0.175
0.140

g (Step 1)

A (mT) (step 1)

2.0055
2.0060

1.689
1.543

74

M.M. Sunnetcioglu, D. Dadayl / Talanta 53 (2000) 6974

from Figs. 1 3 air drying time is highly critical.

Any time until 1 h might be used but in relative
studies the waiting times of the samples must be
the same.
In conclusion, the above results indicated the
applicability of our spin probe technique to the
cereals even after long storage times past after
irradiation. The suggested new methods are
promising in shortening the spectrometer time,
however, at close irradiation doses these methods
are not satisfactory. Currently studies on the seeds
have been carried out to widen the applicability of
the spin probe technique to various seeds.

Acknowledgements
We thank Hacettepe University for the financial
support of the project 97 01 602 005. We also
wish to thank Professor Bekir Aktas from Gebze
Yuksek Teknoloji Enstitusu for the use of their
laboratories for some control measurements.

References
[1] Anonymous, 1981. Wholesomeness of Irradiated Food,
Report of Joint FAO/IAEA/WHO Expert Committee,
World Health Organization, Geneva. Technical Report
Series 659.
[2] K.W. Bogl, Appl. Radiat. Isot. 40 (1989) 1203.
[3] J.A.P. Boshard, D.E. Holmes, L.H. Piette, Int. J. Appl.
Radiat. Isot. 22 (1971) 316.
[4] N.J.F. Dodd, A.J. Swallow, F.J., Ley. Radiat. Phys.
Chem. 26 (1985) 51.
[5] M.F. Desrodiers, Appl. Radiat. Isot. 47 (1996) 1621.
[6] C.R. Hunter, D.R. Hunter, G.J. Troup, Search 19 (1988)
198.
[7] A.R. Forrester, D.B. McKay, Pure Appl. Chem. 65 (2)
(1990) 307.
[8] B.J. Tabner, V.A. Tabner, Radiat. Phys. Chem. 41 (3)
(1993) 545.

[9] J. Raffi, C. Hasbany, G. Lesgards, D. Ochin, Appl.

Radiat. Isot. 47 (11/12) (1996) 1633.
[10] E.M. Steward, R. Gray, Appl. Radiat. Isot. 47 (11/12)
(1996) 1629.
[11] G.J. Troup, J.R. Pilbrow, D.R. Hutton, C.R. Hunter,
G.L. Wilson, Appl. Radiat. Isot. 40 (10 12) (1989) 1223.
[12] J. Raffi, J.-P. Agnel, S.R. Kassis, Sci. Aliments 7 (4)
(1987) 657.
[13] E.P. Munoz, E. Adem, G. Burillo, R.V. Gleason, H.S.
Murrieta, Radiat. Phys. Chem. 43 (4) (1994) 311.
[14] H.S. Murrieta, E.P. Munoz, E. Adem, G. Burillo, M.
Vazquez, E.B. Cabrera, Appl. Radiat. Isot. 47 (11/12)
(1996) 1657.
[15] L.J. Berliner, Spin Labeling; Theory and Applications,
Academic Press, New York, 1976, pp. 1 4.
[16] N. Kocherginsky, H.M. Swartz, Nitroxide Spin Labels,
Reactions in Biology and Chemistry, CRC Press, Boca
Raton, FL, 1995, pp. 1 25.
[17] M.P. Gierula, W.K. Subczynski, A. Kusumi, Biochemie
73 (1991) 1311.
[18] K. Strazalka, W.I. Gruszecki, Biochim. Biophys. Acta
1194 (1994) 138.
[19] S.R. Wassall, L. Wang, R.C.Y. McCabe, W.D. Ehringer,
W. Stillwell, Chem. Phys. Lipids 60 (1991) 29.
[20] P. Muller, A. Zachowski, Y. Beuzard, P.F. Devaux,
Biochim. Biophys. Acta 1151 (1993) 7.
[21] P. Cooper, J. Kudynska, H.A. Buckmaster, R. Kudynski,
Biochim. Biophys. Acta 1139 (1992) 70.
[22] J. Kutcher, T. Hanke, B. Ebert, J. Hellebrant, Stud.
Biophys. 111 (2 3) (1986) 105.
[23] A.I. Smirnov, H.A. Golovina, O.E. Yakimchenko, S.I.
Aksyonov, Y.S. Lebedev, J. Plant Physiol. 140 (1992) 447.
[24] E.A. Golovina, A.N. Tikhonov, Biochim. Biophys. Acta
1190 (1994) 385.
[25] M.M. Sunnetcioglu, D. Dadayl, R. Sungur, G. Bingol, J.
Plant. Physiol. 151 (1997) 196.
[26] E.A. Golovina, A.N. Tikhonov, F.A. Hoekstra, Plant
Physiol. 114 (1997) 383.
[27] D. Dadayl, M.M. Sunnetcioglu, H. Koksel, S. C
elik,
Cereal Chem. 74 (4) (1997) 375.
[28] M.M. Sunnetcioglu, D. Dadayl, S. C
elik, H. Koksel,
Cereal Chem. 75 (6) (1998) 875.
[29] M.M. Sunnetcioglu, D. Dadayl, S. C
elik, H. Koksel,
Appl. Radiat. Isot. 50 (1999) 557.
[30] L.J. Berliner, J. Reuben, Biological Magnetic Resonance,
vol. 8, Plenum Press, New York, 1989, pp. 77 102.
[31] J.A. Nelder, R. Mead, Comput. J. 7 (1965) 308.