Establishment and validation of a dose-effect curve for y-rays by cytogenetic analysis. Joan Francesc Barquinero a,*, Leonardo Barrios b, Maria Rosa Caballin ...
Fundamental and Molecular Mechanisms of Mutagenesis
ELSEMER
Mutation
Establishment
Research
326 (1995) 65-69
and validation of a dose-effect curve for y-rays by cytogenetic analysis
Joan Francesc Barquinero a,*, Leonardo Barrios b, Maria Rosa Caballin apt, Rosa Mir6 c,d,Montserrat Ribas e, Antoni Subias e, Josep Egozcue b,c a Unitat d’Antropologia, Departament de Biologia Animal, Biobgia Vegetal i Ecologia, Facultat de Cihcies, Universitat Authoma de Barcelona, 08193 Bellaterra, Spain b Unitat de Biologia CeLlular, Departament de Biologia Cel.lular i Fisiologiu, Facultat de Cithcies, Universitat Autboma de Barcelona, 08193 Bellaterra, Spain ’ Institut de Biologia Fondamental ‘Went War Palas?, Universitat Autboma de Barcelona, 08193 Bellaterra, Spain ’ Unitat a’eBiologia, Departament de Biologia Celhlar i Fisiologia, Facultat de Medicina, Universitat Autboma de Barcelona, 08193 Bellaterra, Spain e Servei d’Oncologia, Hospital de la Santa Creu i Sant Pay Universitat Authoma de Barcelona, 08193 Bellaterra, Spain. Received
4 May 1994; revision
received
29 July 1994; accepted
16 August
1994
Abstract
A dose-effect curve obtained by analysis of dicentric chromosomes after irradiation of peripheral blood samples, from one donor, at 11 different doses of y-rays is presented. For the elaboration of this curve, more than 18000 first division metaphases have been analyzed. The results fit very well to the linear-quadratic model. To validate the curve, samples from six individuals (three controls and three occupationally exposed persons) were irradiated at 2 Gy. The results obtained, when compared with the curve, showed that in all cases the 95% confidence interval included the 2 Gy dose, with estimated dose ranges from 1.82 to 2.19 Gy. Keywords:
Dose-effect curve; Gamma-rays; Cytogenetic analysis
1. Introduction
The analysis of dicentric translocations in lymphocytes is considered the most sensitive biological method to quantify exposure to radiation for doses over 0.1 Gy, if an adequate number of
* Corresponding 5811321. 0027-5107/95/%09.50
author.
Tel.
34-3-5812049;
0 1995 Elsevier
SSDI 0027-5107(94)00150-2
Science
Fax
34-3-
metaphases is analyzed. For the estimation of a dose, each laboratory must have its own dose-effeet curve (see Lloyd and Edwards, 1983). Although a certain degree of variability in the response to radiation has been described (Kakati et al., 1986), most laboratories can estimate a given dose with an accuracy acceptably close to the true values (Lloyd et al., 1987). We present a dose-effect curve obtained using 60& y-irradiation, and a study in six individuals, three occupationally exposed persons and three
B.V. All rights reserved
66
J.F. Barquinero et al. /Mutation Research 326 (1995) 65-69
controls, whose peripheral irradiated at 2 Gy.
blood samples were
2. Materials and methods Irradiation conditions
Samples were irradiated using a cobalt source (Theratron-780) located at the Hospital de la Santa Creu i Sant Pau (Barcelona). Dose determinations were made by the Unit of Radiophysics and Radioprotection of the hospital after calibration of the teletherapy unit with an electrometer (Farmer 2570). Dose rates ranged from 117.5 cGy/min to 107 cGy/min due to the decay of the cobalt source. For the elaboration of the curve, doses of 0, 0.1, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4 and 5 Gy were used. IAEA recomendations (Beninson et al., 1986) have been followed for the irradiation. To validate the curve, peripheral blood samples from six individuals were irradiated at 2 Gy in the same conditions. Three of them had been occupationally exposed to corrected doses that ranged from 5.29 to 25.21 mSv. The corrected doses were calculated according to Bauchinger et al. (1984). Culture conditions
For the elaboration of the curve, a donor with no history of exposure to mutagenic agents in-
Table 1 Cytogenetic
results
obtained
after irradiation
at different
cluding radiation was chosen. Blood samples, irradiated at the doses indicated above, were cultured for 48 h in RPM1 1640 medium supplemented with 20% fetal calf serum, antibiotics and phytohemagglutinin. Colcemid was added 2 h before harvesting. To select first division metaphases, 12 pgg/ml bromodeoxyuridine was added to the cultures. Two- to three-day-old slides were stained with the Giemsa counterstain of the fluorescence plus Giemsa stain technique. The same culture conditions were used in the other six individuals. Cytogenetic analysis
Chromosome analysis was carried out exclusively in first division metaphases containing 46 or more centromeres. To provide an adequate calibration curve, a minimum of 5000 metaphases were scored for doses of 0 and 0.1 Gy; 2000 for doses of 0.25 and 0.5 Gy; and as many metaphases as needed to count at least 100 dicentrics for the remaining doses. All metaphases with chromosome abnormalities were independently analyzed by three investigators. Chromosome abnormalities were classified as follows: dicentric chromosomes were only considered when the acentric fragment was present. A translocation was recorded only when the morphology of the derivative chromosomes was clearly indicative of this kind of rearrangement, although its frequency was underestimated be-
doses of y-rays
Cells analyzed
die
rings
ace
other
ctb
gaps
NA
(Gy) 0.00 0.10 0.25 0.50 0.75
5000 5002 2008 2002 1832
8 14 22 55 100
1 _
34 30 16 44 30
3 3 _ 2 8
29 22 10 8 6
26 25 11 14 2
3 3 2 1 _
1.00 1.50 2.00 3.00
1168 562 332 193
109 100 103 108
4 7 6 6
79 56 68 65
5 11 10 9
18 2 6 5
19 3 4 3
1 _ _ _
4.00 5.00
103 59
103 107
11 9
72 65
12 13
_ _
4 _
_
Dose
die, dicentric; ace, acentric; chromosome and chromatid;
1
other, includes translocations, NA, numerical abnormalities.
inversions
etc.; ctb, chromatid
breaks
including
tetraradials;
gaps, both
.I.F. Barquinero et al. /Mutation Research 326 (I 995) 65-69
67
Table 2 Dicentric distribution within cells, Papworth U and dispersion index of the dose-effect curve (GY)
Dose
Cells analyzed
Total die
Cells with 0 die
Cells with 1 die
0.00 0.10 0.25 0.50 0.75 1.00 1.50 2.00 3.00 4.00 5.00
5000 5002 2008 2002 1832 1168 562 332 193 103 59
8 14 22 55 100 109 100 103 108 103 107
4992 4988 1987 1947 1736 1064 474 251 104 35 11
8 14 20 55 92 99 76 62 72 41 19
Cells with 2 die
1 4 5 12 16 15 21 11
Cells with 3 die
Cells with 5 die
-
3 2 4 9
_
_ 2 6
3
U
a’/~
- 0.07 -0.13 2.61 -0.86 0.79 - 0.02 1.08 2.31 - 1.64 - 0.84 0.81
1.00 1.00 1.08 0.97 1.03 1.00 1.06 1.18 0.83 0.88 1.13
analysis
To check if the distribution of dicentrics for each dose followed a Poisson distribution, the dispersion index D = a2/y and the normalized unit of this index W> were used (Papworth, 19751. Curve fitting was done using the method of iteratively reweighted least squares.
3. Results Table 1 shows the different chromosome abnormalities observed for each dose. The most Table 3 Dose estimations Cells analyzed
Die
Die/cell
Estimated dose (Gy)
95% Confidence dose interval
410 335 387
102 103 100
0.249 0.307 0.258
1.82 2.04 1.86
1.62/2.01 1.82/2X 1.65/2.05
380 313 285
108 95 100
0.284 0.304 0.351
1.96 2.03 2.19
1.75/2.15 1.80/2.24 1.95/2.41
Occupation&y exposed (corrected dose) Case l(25.21 mSv) Case 2 (5.29 mSv) Case 3 (20.01 m.Sv)
Non-exposed Case 4 Case 5 Case 6
*SE
f f f It f f f f + + f
0.02 0.02 0.03 0.03 0.03 0.04 0.06 0.08 0.10 0.14 0.18
evident dose-effect relationship corresponds to dicentrics. Table 2 shows the cell distribution of dicentrics for each dose. As expected, the number of cells with more than one dicentric increases with dose. The distribution of dicentrics follows a Poisson distribution for almost all doses. Only for 0.25 and 2 Gy doses the Papworth U value exceeded f 1.96. In Fig. 1 can be seen the dose-effect curve, the observed frequencies of dicentrics and the fitted values of the C, (Y and /3 coefficients of the linear-quadratic function. The chi-square value of the curve was 6.6 (p = 0.57981, indicating a good fit. The frequencies of dicentrics observed in the six individuals studied and the dose estimations using the coefficients of our dose-effect curve can be seen in Table 3. The frequency of dicentrics per cell in the non-exposed group was higher
cause unbanded preparations were used. Acentries and chromosome breaks (csbl were recorded together. Other abnormalities such as chromatid breaks and gaps were also taken into account. Numerical abnormalities were only considered when hyperdiploidy was observed. Statideal
Cells with 4 die
68
J.F. Barquinero et al. /Mutation
1.6
= %
$ 1.2
!I ._ ; 0.8 a 0.4 0
0
1
3 2 Dose (GY)
4
5
Fig. 1. Dose-effect curve for y-rays. The fitted values of the coefficients of the linear quadratic function Y = C + (YD + pD2 are: C =(0.13+0.05)x 10m2; a =(2.10+0.52)~10-‘; p = (6.31 f 0.40) X 10d2. The 95% confidence intervals are represented by dashes, the obsewed frequencies of dicentrics per cell by black circles and the adjusted frequencies by x
than in the occupationally exposed one (0.310 versus 0.269), but the difference was not significant. Although a certain interindividual variability was observed, in all cases the 95% confidence intervals included the 2 Gy dose.
4. Discussion The main objective of a dose-effect curve is its use in radioprotection, increasing its accuracy with the number of cells analyzed. The distribution of dicentrics for each dose in the present study agrees with a Poisson distribution and, as reported by other authors, does not show overdispersion (Edwards et al., 1979; Schmid et al., 1984; Fabry et al., 1985). When a comparison is made between our dose-effect curve and those reported by Edwards et al. (19791, Schmid et al. (1984) and Fabry et al. (1985), it can be seen that there is an agreement in the fitted coefficients. The slight differences in the frequencies of dicentrics per cell could be due to the number of cells analyzed and to the type of radiation and dose rates employed.
Research 326 (1995) 65-69
On the other hand, a collaborative study including 12 laboratories that used the same material showed that for a given dose of 2.34 Gy of Xand -y-rays, some deviations were observed in the estimated dose that ranged from 0.94 to 3.12 Gy (Lloyd et al., 19871, although in general, most participants in this study consistently obtained results acceptably close to the real value. These results indicate that every laboratory must have its own curve because some differences depending on factors like dose rates, number of cells analyzed, scoring criteria etc., can be produced. The estimated doses, using our dose-effect curve, in six individuals whose blood was irradiated at 2 Gy, varied from 1.82 to 2.19 Gy. In all cases the confidence interval included the 2 Gy dose (Table 3). When the results of non-exposed individuals were compared with those from exposed ones, a higher frequency of dicentrics was observed in the non-exposed group, although the difference was not significant. Within the exposed group it is interesting to point out that case 2, who received the lowest occupational dose, showed the highest frequency of dicentrics after blood irradiation at 2 Gy. These differences could be due to an interindividual variability (Kakati et al., 19861, because a test of homogeneity of the six donors with a chi-square did not show significant differences. However, the existence of some factor(s) that can reduce the sensitivity in the exposed population cannot be discarded (Sankaranarayanan et al., 1989; Bosi and Olivieri, 1989; Bauchinger et al., 1989; Wang et al., 1991; Kesley et al., 1991; Hain et al., 1992). Further studies are needed to clarity this question. Taking into account the above mentioned factors, and as recommended by IAEA (Beninson et al., 19861, the estimation of doses in accidental exposures should be given as 95% confidence intervals.
Acknowledgement
This work received financial support from the Consejo de Seguridad Nuclear.
J.F. Barquinero et al. /Mutation Research 326 (1995) 65-69
References Bauchinger, M.H., H. Eckerl, G. Drexler, S. Streng and E. Schmid (1984) Chromosome dosimetry and occupational radiation exposure, Radiat. Protect. Dosimetry, 2, 93-97. Bauchinger, M., E. Schmid, H. Braselmann and U. Nahrstedt (1989) Absence of adaptive response to low-level irradiation from tritiated thymidine and X-rays in lymphocytes of two individuals examined in serial experiments, Mutation Res., 227, 103-107. Beninson, D., D.C. Lloyd, A.T. Natarajan, G. Obe, R.J. Preston and M.S. Sasaki (1986) Biological dosimetry: Chromosomal aberration analysis for dose assessment, IAEA Technical Reports Series, No. 260, Vienna. Bosi, A. and G. Olivieri (1989) Variability of the adaptive response to ionizing radiations in humans, Mutation Res., 211,13-17. Edwards, A.A., D.C. Lloyd and R.J. Purrot (1979) Radiation induced chromosome aberrations and the Poisson distribution, Radiat. Environ. Biophys., 16, 89-100. Fabry, L., A. Leonard and A. Wambersie (1985) Induction of chromosome aberrations in Gc human lymphocytes by low doses of ionizing radiations of different quality, Radiat. Res., 103, 122-134. Hain, J., R. Jaussi and W. Burkart (1992) Lack of adaptive response to low doses of ionizing radiation in human lymphocytes from five different donors, Mutation Res., 283, 137-144. Kakati, S., J.R. Kowalczyk, Z. Gibas and A.A. Sandberg (19861 Use of radiation induced chromosomal damage in human lymphocytes as a biological dosimeter is questionable, Cancer Genet. Cytogenet., 122, 137-141.
69
Kesley, K.T., A. Memisoglu, D. Frenkel and H.L. Liber (1991) Human lymphocytes exposed to low doses of X-rays are less susceptible to radiation-induced mutagenesis, Mutation Res., 263, 197-201. Lloyd, D.C. and AA. Edwards (1983) Chromosome aberrations in human lymphocytes: effect of radiation quality, dose and dose rate, in: T. Ishihara and M. Sasaki (Eds.), Radiation-Induced Chromosome Damage in Man, Liss, New York, pp. 23-49. Lloyd, D.C., A.A. Edwards, J.S. Prosser, N. Barjaktarovic, J.K. Brown, D. Horvat, S.R. Ismail, G.J. Kiiteles, Z. Almassy, A. Krepinsky, M. Kucerova, L.G. Littlefield, U. Mukherjee, A.T. Natarajan and M.S. Sasaki (1987) A collaborative exercise on cytogenetic dosimetry for simulated whole and partial body accidental irradiation, Mutation Res., 179, 197-208. Papworth, D. (1975) Curve fitting by maximum likelihood, Appendix to paper by J.R.K. Savage: Radiation-induced chromosomal aberrations in the plant Tradescantia. Doseresponse curves, Radiat. Bot., 15, 87. Sankaranarayanan, K., A. van Duyn, M.J. Loos and A.T. Natarajan (1989) Adaptive response of human lymphocytes to low-level radiation from radioisotopes or X-rays, Mutation Res., 211, 7-12. Schmid, E., M. Bauchinger, S. Streng and U. Nahrstedt (1984) The effect of 220 kVp X-rays with different spectra on the dose response of chromosome aberrations in human lymphocytes, Radiat. Environ. Biophys., 23, 305-309. Wang, Z.-Q., S. Saigusa and M.S. Sasaki (1991) Adaptive response to chromosome damage in cultured human lymphocytes primed with low doses of X-rays, Mutation Res., 246, 179-186.
