Mutation Research, 1i 1 (1983) 185-193
185
Elsevier
MTR 08512
The suitability of the human lymphocyte micronucleus assay
system for biological dosimetry
R. Huber, S. Streng a n d M. B a u c h i n g e r
Abteilung fiir Strahlenbiologie, Cytogenetics, GSF, D-8042 Neuherberg, Post Oberschleissheim (West
Germany)
(Received 17 January 1983)
(Accepted 25 May 1983)
Summary
Human whole blood was irradiated with 220 keV X-rays at doses of 0-4.0 Gy.
After incubation periods of 48, 60, 72, 84 and 96 h, lymphocytes were prepared
without colcemid pretreatment according to 2 different methods, and micronuclei
were scored.
The crucial point of lymphocyte preparation was found to be the osmotic pressure
of the hypotonic solution. Only a method that preserves the cytoplasm of lymphoblasts is suitable for a correct association of micronuclei with the main nucleus.
Similar as for structural chromosome changes, now their intercellular distribution
can be analysed. This is necessary for the derivation of appropriate statistical
weights which have to be used for more reliable regression analyses. For 48 h, the
data can be described by the linear model, for 84 and 96 h, by the linear-quadratic
model. For 60 and 72 h no such definite conclusions can be drawn.
For calibration purposes a standardized culture time cannot be recommended.
Because the background frequency is high, the lymphocyte micronucleus assay
system is not sensitive enough to detect a significant increase in the incidence of
micronuclei after exposure to low doses ( < 0.3 Gy).
The occurrence of micronuclei in plant cells after exposure to chemical substances
and ionizing radiation has been a well-known phenomenon for several decades.
Between 1973 and 1975 the so-called 'micronucleus test' was developed for mammalian polychromatic erythrocytes of the bone marrow (Boiler and Schmid, 1970;
Ledebur and Schmid, 1973; Schmid, 1973; 1975). The method is now widely used
for pre-screening of chromosome breakage in mutagenicity testing.
Countryman and Heddle (1976) and Countryman et al. (1977) published a
method that permitted the demonstration of radiation-induced micronuclei in hu0027-5107/83/$03.00 © 1983 Elsevier Science Publishers B.V.
186
man lymphocyte preparations. Since, according to Iskandar (1979), with this technique the nuclei tend to clump, he presented a method by which the cytoplasm could
be preserved. Nevertheless, owing to the quantitative data of Countryman and
Heddle (1976) and Countryman et al. (1977), the application of their micronucleus
assay for biological dose estimation of radiation exposure seemed to be promising.
In this connection, an automated approach for micronucleus scoring has been
presented (Callisen and Norman, 1983). As compared with conventional chromosome analyses, scoring of micronuclei is an easier and faster procedure. Provided
that, in vitro, reliable calibration curves relating the radiation dose to the frequency
of micronuclei can be derived, the micronucleus assay system might be usefully
applied as a supplementary technique for biological dosimetry. The present study
reports on qualitative and quantitative aspects for such an approach.
Materials and methods
Whole blood of a healthy male donor was irradiated at 37°C with 0-4.0 Gy of
220 keV X-rays at a dose rate of 0.5 Gy per rain, 14 mA and 1 mm Cu filtration.
Lymphocyte cultures were set up in a standard procedure of our laboratory: 4 ml
Ham's F10 medium (Gibco-Biocult, Glasgow), 0.5 ml foetal calf serum, 0.13 ml
PHA (Difco), 0.5 ml whole blood. After incubation periods of 48, 60, 72, 84 or 96 h,
lymphocytes were prepared without previous colcemid treatment according to 2
different methods.
Method 1. For hypotonic treatment a 1 : 3 mixture of Hank's solution and distilled
water was used for 9 min at 37°C, plus an additional 5 min for centrifugation at
1000 rpm. After 30 min fixation in a 3 : 1 mixture of methanol and glacial acetic
acid, shdes were first stained with Feulgen and subsequently with acetic orcein.
Method 2. For hypotonic treatment a 0.1-M KCI solution was used for 3 rain at
room temperature, plus an additional 5 rain for centrifugation at 600 rpm. The cells
were quickly, but carefully, resuspended in methanol:glacial acetic acid, 3:1,
centrifuged and fixed for a further 20 min. The cell suspension of 48-, 60- and 72-h
cultures was dropped onto wet slides and allowed to airdry. Immediately after the
dropping of cells from 84- and 96-h cultures, the rear side of each slide was pressed
onto CO 2 snow. After freezing of the cell suspension, slides were airdried. This
modification was necessary because, with increasing culture times, the cell membranes become more sensitive to disruption.
After 1-2 days, the slides were stained for 10 rain in 3% Giemsa (Merck)
dissolved in Weise buffer, rinsed in distilled water, airdried and mounted with
Eukitt. A detailed description of both methods is given elsewhere (Huber, 1981).
For the identification of micronuclei in preparations of Method 1, published
criteria were applied (Countryman and Heddle, 1976; Heddle et al., 1978). These
criteria were extended or modified for an evaluation of preparations of Method 2:
(1) only cells with well preserved and, as compared with the darkly stained main
nucleus, faintly stained cytoplasm were analysed; (2) in cell groups the individual
boundaries of cytoplasm had to be clearly separated; (3) the main nucleus and
micronuclei had to be situated within the cytoplasm; (4) no connection between
187
main nucleus and micronucleus; (5) maximal diameter of 1/2 the main nucleus;
several larger nuclei of similar size were recorded as multinucleated cells; (6) a
micronucleus could be more darkly stained than the main nucleus but was non-refractive; (7) sufficient staining intensity and round main nuclei and micronuclei; (8)
tiny spheres of hyperchromatic chromatin which sometimes occurred in a frequency
of 10-20 were not classified as micronuclei; otherwise the number of micronuclei per
cell was not restricted.
Slides were scored at 580 x magnification with bright light and a green filter.
2 x 1000 cells were scored from two different slides for 0-0.25 Gy and 2 × 500 cells
for 0.5-4.0 Gy.
TABLE 1
FREQUENCY DISTRIBUTION
T U R E T I M E S ( M E T H O D 2)
Culture
time
Dose
(Gy)
(h)
48
60
Micronuclei
2
3
4-5
Total
number
Mean number
per cell
0
1959
39
2
-
-
43
0.022
1950
935
909
48
59
75
2
6
15
I
-
52
71
108
0.026
0.071
0.108
2.0
845
132
20
3
-
181
0.181
4.0
750
198
36
15
I
319
0.319
1951
1953
44
40
3
7
2
-
-
56
54
0.028
0.027
957
910
815
683
38
82
156
220
4
6
28
75
I
2
1
19
3
49
100
215
439
0.049
0.100
0.215
0.439
1959
1969
38
26
3
5
-
-
44
36
0.022
0.018
966
942
847
707
32
52
131
214
2
6
18
57
4
17
5
36
64
179
400
0.036
0.064
0.179
0.400
0.25
1965
1970
33
27
2
2
1
-
37
34
0.019
0.017
0.5
1.0
2.0
4.0
967
949
875
735
30
43
104
202
3
8
18
49
2
12
1
2
36
59
150
344
0.036
0.059
0.150
0.344
1972
1 974
972
956
899
798
28
25
26
41
85
153
1
2
3
14
38
2
8
3
28
27
30
47
119
265
0.014
0.014
0.030
0.047
0.119
0.265
0
0.25
0
0.5
1.0
2.0
4.0
96
"1
CELLS AT DIFFERENT
0.25
0.5
1.0
0.25
84
AMONG
Number of cells with
' n ' micronuclei
0
0.5
1.0
2.0
4.0
72
OF MICRONUCLEI
0
0
0.25
0.5
I.O
2.0
4.0
CUL-
188
TABLE 2
S T A N D A R D E R R O R (S.E.) A N D 95% C O N F I D E N C E LIMITS F O R T H E E S T I M A T E D P A R A M E T E R S b(0) A N D b(1) O F T H E L I N E A R M O D E L Y = b(0)+ b(1).D
Culture
time
(h)
b(0)+S.E.
95%
confidence
limits
b(I)+S.E.
95%
confidence
limits
48
60
72
84
96
0.0166 + 0.0049
0.0132 + 0.0075
0.0071 5:0.0086
0.0078 5:0.0066
0.0064 5:0.0047
0.0030,
- 0.0078,
-0.0168,
- 0.0105,
- 0.0067,
0.0780 5:0.0077
0.0969 5:0.0115
0.0801 + 0.0145
0.0703 5:0.0110
0.0541 + 0.0084
0.0566,
0.0650,
0.0398,
0.0398,
0.0308,
0.0302
0.0341
0.0309
0.0260
0.0194
0.0994
0.1288
0.1204
0.1007
0.0774
Results
Method 1. The application of this technique yielded mainly lymphoblast nuclei
without recognizable cytoplasm. Micronuclei were more or less separated from these
nuclei. This required an arbitrary association, particularly for several micronuclei.
According to criteria of Countryman and Heddle (1976), for such an association a
location of micronuclei within 3 or 4 nuclear diameters of a nucleus was accepted.
The frequencies of micronuclei at different culture periods are illustrated in Fig.
1. Each curve shows a dose-dependent increase of the number of micronuclei. The
shapes are, however, not uniform and reveal strong variations. A mathematical
description of the curves by a uniform model of dose-response relationship is
impossible.
Method 2. This procedure preserves the cytoplasm of most lymphoblasts, and
)~8&h
0.10.
/
/
I
o
i
/
/
/
/p60 h
//
/f,"//
IISIIJ.IsI/
./ //
0.075
.~
o
c
o=.. 0.05
o
s-
o.2s o.s 1.0
2.b
/~72h
96 h
4.0
Dose [Gy]
Fig. 1. Dose-effect curves for micronuclei at different culture times (Method 1).
0.0161 +
0.0197 +
0.0159 +
0.0 ! 43 +
0.0108 +
48
60
72
84
96
0.0064
0.0072
0.0060
0.0043
0.0034
b(0)+S.E.
Culture
time
(h)
- 0.0043, 0.0365
- 0.0031, 0.0425
- 0.0031, 0.0349
0.0007, 0.0279
0.0002, 0.0215
9570
confidence
limits
0.0814 + 0.0226
0.0614 + 0.0222
0.0317 + 0.0189
0.0314 + 0.0140
0.0266+0.0114
b(I)+S.E.
-
0.0095,
0.0093,
0.0286,
0.0132,
0.0096,
0.1533
0.1322
0.0919
0.0759
0.0628
9570
confidence
limits
- 0.001 + 0.0066
0.0118 + 0.0067
0.0172 + 0.0060
0.0134 + 0.0043
0.0098 +0.0036
b(2)+S.E.
-
0.0222,
0.0096,
0.0018,
0.0004,
0.0017,
0.0200
0.0322
0.0363
0.0272
0.0213
9570
confidence
limits
S T A N D A R D E R R O R (S.E.) A N D 95~ C O N F I D E N C E LIMITS F O R T H E E S T I M A T E D P A R A M E T E R S b(0), b ( I ) A N D b(2) O F T H E L I N E A R Q U A D R A T I C M O D E L Y = b ( 0 ) + b ( l ) - D + b(2).D 2
T A BL E 3
190
micronuclei can be easily associated with the main nucleus. The distributions of
micronuclei among cells at different doses and culture periods are shown in Table 1.
From a dispersion index (variance/mean) greater than 1, overdispersion can be
deduced. According to a dispersion-index test (Savage, 1970; Edwards et al., 1979) a
significant over dispersion was observed for 24 distributions of Table 1.
The analysis of the dose-response relationship was carried out under the assumption that either the linear model Y = b ( 0 ) + b(1)-D or the linear-quadratic model
Y = b ( 0 ) + b ( 1 ) - D + b(2). D 2 is relevant for the description of micronuclei production [b(0), b(1), b(2), estimated parameters for spontaneous and radiation-induced
micronuclei, respectively; D, dose]. The data were fitted to either model by a
weighted least-squares method (Kellerer and Brenot, 1974). For this reason each
observation was weighted by the reciprocal sample mean variance (number of
cells/variance). The results are shown in Tables 2 and 3. For a culture time of 48 h
the data could be described by the linear model, for 84 and 96 h the data fitted the
linear-quadratic model. For 60 and 72 h no such definite conclusions could be
drawn.
Discussion
Micronuclei are mainly regarded as equivalents of acentric chromatin elements
that are not incorporated into daughter nuclei during cell division (Heddle and
Carrano, 1977). When a preparation technique is used that only insufficiently
preserves the cytoplasm of lymphoblasts (Method 1), micronuclei may be more or
less separated from the main nuclei. A mutual association is rather arbitrary and
must lead to an inaccurate quantification. The application of Method 2 reveals that
about 5-6 times more micronuclei could be evaluated as compared with Method 1.
The preservation of the cytoplasm mainly depends on the concentration of the
hypotonic solution. In Method 1 the osmotic pressure of this solution was 77
milliosmoles and similar to that used in Countryman and Heddle's method (1976).
Obviously, with both methods these concentrations are too low for a sufficient
preservation of the cytoplasm. For an association of micronuclei with main nuclei
auxiliary criteria have to be adopted (location of micronuclei within 3 or 4 nuclear
diameters of a nucleus).
The osmotic pressure of the hypotonic solution used by Iskandar (1979) was 285
milliosmoles which is nearly the concentration of an isotonic solution. For the
preservation of the cytoplasm and an additional spreading of the cells in our Method
2, a solution of 200 milliosmoles was used. By this technique an exact association of
micronuclei with main nuclei can be achieved. As for structural chromosome
aberrations the exact intercellular distribution of micronuclei can be determined.
This is necessary for the application of a weighted least squares method for
estimating regression parameters of dose-response functions.
The present study shows that clear dose-response relations can be described for
the frequency of micronuclei at different culture times. Nevertheless, after 48 h the
data are fitted by a linear model and after 84 and 96 h, by the linear-quadratic
191
model. The 60- and 72-h data cannot be interpreted unequivocally (Tables 2, 3).
Applying an unweighted least squares method, Countryman and Heddle (1976)
described a dose exponent, n = 1.2, in the power law model, for the production of
micronuclei in 96-h cultures.
In contrast to these authors who observed the maximal frequency of radiation-induced micronuclei at 84 or 96 h, we found the highest numbers at 60 h.
Owing to the influence of cell proliferation, quantitative analyses of dose-response data based on radiation-induced structural chromosome aberrations should
be carried out by scoring exclusively first-division cells (Crossen and Morgan, 1977;
Scott and Lyons, 1979; Bauchinger, 1983). This can be attained by the FPG-straining technique after BrdU treatment of lymphocyte cultures. In the micronucleus
assay system, where whole cells are analysed, it cannot be determined from an
interphase nucleus whether a cell is in its first or a later cycle. This is a further
drawback of this system. Based on data from human lymphocytes established with
conventional staining (Sasaki and Norman, 1967) it was estimated that only about
20% of total acentrics are lost during first mitosis (Carrano, 1972; Carrano and
Heddle, 1973). Their equivalents can be observed as micronuclei in the cytoplasm
associated with a 2nd cycle interphase main nucleus. Acentrics that had been
incorporated into a daughter nucleus will be duplicated during a 2nd cell cycle and
may appear as micronuclei, associated with a 3rd-cycle interphase main nucleus.
Consequently, a maximal response for micronuclei does not necessarily reflect the
actual amount of expected (20% of total) acentrics. From a determination of
harlequinized metaphases (Perry and Wolff, 1974; Apelt et al., 1981; Kolin-Gerresheim and Bauchinger, 1981) in parallel cultures, (2 × 300 metaphases per point), a
pronounced increase of metaphases beyond first division was observed, especially
between 48 and 60 h. This was true even for 4.0 Gy, where a radiation-induced
mitotic delay has to be considered:
48 h: 0 Gy, 60.5%; 2.0 Gy, 46.0%; 4.0 Gy, 13.9%.
60 h: 0 Gy, 94.8%; 2.0 Gy, 87.2%; 4.0 Gy, 69.7%.
This implies that an appropriate quantitative interpretation of the micronucleus data
is difficult because a standardized culture time cannot be derived.
A peculiarity of the system is the high background frequency of micronuclei
which reduces the ability of the test to demonstrate a significant increase in the
incidence of micronuclei after exposure to low levels of dose (i.e. below 0.3 Gy). The
control frequency for structural chromosome aberrations in 24000 first-division cells
of 47 unexposed individuals, analysed in our laboratory, was 30 excess acentrics and
5 dicentrics plus centric rings per 10 4 cells (Bauchinger, 1983; Wagner et al., 1983).
Together with chromatid breaks, a value of about 50 acentric chromatin elements
per 104 ceils results. Even if we assume that all of them would be lost during cell
division and remain in the cytoplasm as micronuclei (generally only 20% are
considered), we still find about 5 times more at 0 Gy of 48-72-h culture time. With
increasing culture time there is a tendency towards lower numbers. The revealing
differences indicate that micronuclei may not only arise from acentric chromosomal
elements but also from whole chromosomes (at least from the smaller ones) or, e.g.
from chromatin particles removed from disintegrating nuclei.
192
Conclusions
Our data do not indicate that at present, the human lymphocyte micronucteus
assay system can replace the chromosome analysis for a biological estimation of
radiation doses. The main reasons are: uncertainties in the interpretation of dose-response relationship; standardization of culture time to obtain actual maximal
response; the insensitivity of the method at low levels of dose.
The micronucleus assay, however, might provide a means of rapid prescreening of
acute exposure to high doses. Additionally, micronucleus scoring could be performed
to detect potential radiosensitive individuals, e.g. in pre-employment examinations
of radiation workers. For this reason, irradiation can be confined to high levels of
dose. After exposure of blood samples in vitro, micronuclei can be rapidly scored
and an enhanced yield could give an indication for an increased chromosome
sensitivity. This could be verified by a separate analysis of the dose-response of
chromosomal aberrations in first-division cells.
Acknowledgement
We thank Mrs. G. Rocznik for her kind help in preparation of the manuscript.
References
Apelt, F., I. Kolin-Gerresheim and M. Bauchinger (1981) Azathioprine, a clastogen in human somatic
cells? Analysis of chromosome damage and SCE in lymphocytes after exposure in vivo and in vitro,
Mutation Res., 88, 61-72.
Bauchinger, M. (1983) Cytogenetic effects in human lymphocytes as a dosimetry system, in: W.G. Eisert
and M.L. Mendelsohn (Eds.), Biological Dosimetry: Cytometric Approaches to Mammalian Systems,
Springer, Berlin, in press.
Boiler, K., and W. Schmid (1970) Chemische Mutagenes beim S~uger; Das Knochenmark des Chinesischen Hamsters als in vivo-Testsystem; Hitmatologische Befunde nach Behandlung mit Trenimon,
Humangenetik, I 1, 35-54.
Callisen, H., and A. Norman (1983) Micronuclei of human lymphocytes as a biological dosimeter, in:
W.G. Eisert and M.L. Mendelsohn (Eds.), Biological Dosimetry: Cytometric Approaches to Mammalian Systems, Springer, Berlin, in press.
Carrano, A. (1972) A formula to predict the transmission frequency of acentric fragments, Genetics, 72,
777-782.
Carrano, A., and J. Heddle (1973) The fate of chromosome aberrations, J. Theor. Biol., 38, 289-304.
Countryman, P., and J. Heddle (1976) The production of micronuclei from chromosome aberrations in
irradiated cultures of human lymphocytes, Mutation Res., 41,321-331.
Countryman, P., J. Heddle and E. Crawford (1977) The repair of X-ray-induced chromosomal damage in
trisomy 21 and normal diploid lymphocytes, Cancer Res., 37, 52-58.
Crossen, P.E., and W.F. Morgan (1977) Analysis of human lymphocyte cell cycle time in culture
measured by sister chromatid differential staining, Exp. Cell Res., 104, 453-457.
Edwards, A.A., D.C. Lloyd and R.J. Purrott (1979) Radiation-induced chromosome aberrations and the
Poisson distribution, Radiation Environ. Biophys., 16, 89-100.
Heddle, J., and A. Carrano (1977) The DNA content of micronuclei induced in mouse bone marrow by
y-irradiation; Evidence that micronuclei arise from acentric chromosomal fragments, Mutation Res.,
44, 63-69.
193
Heddle, J., R. Benz and P. Countryman (1978) Measurement of chromosomal breakage in cultured cells
by the micronucleus technique, in: H.J. Evans and D.C. Lloyd (Eds.), Mutagen-lnduced Chromosome
Damage in Man, University Press, Edinburgh, pp. 191-200.
Huber, R. ( 1981) Quantitative Analyse strahleninduzierter Chromosomenaberrationen und Kleinkerne in
menschlichen Lymphozyten, Thesis, Fakult~lt fiir Biologic, Ludwig-Maximilians-Universit~t, Miinchen.
Iskandar, O. (1979) An improved method for the detection of micronuclei in human lymphocytes, Stain
Technol., 54, 221-223.
Kellerer, A.M., and J. Brenot (1974) On the statistical evaluation of dose-response function, Radiation
Environ. Biophys., 11, 1-13.
Kolin-Gerresheim, I., and M. Banchinger (1981) Dependence of the frequency of harlequin-stained cells
on BrdU concentration in human lymphocyte cultures, Mutation Res., 91,251-254.
Ledebur, M. yon, and W. Schmid (1973) The micronucleus test; methodological aspects, Mutation Res.,
19, 109-117.
Perry, P., and S. Wolff (1974) New Giemsa method for the differential staining of sister chromatids,
Nature (London), 251, 156-158.
Sasaki, M., and A. Norman (1967) Selection against chromosome aberrations in human lymphocytes,
Nature (London), 214, 502-503.
Savage, J.R.K. (1970) Sites of radiation chromosome exchanges, Current Topics Radiation Res., 6,
129-194.
Schmid, W. (1973) Chemical mutagen testing on in vivo somatic mammalian cells, Agents and Actions, 3,
77-85.
Schmid, W. (1975) The micronucleus test, Mutation Res., 31, 9-15.
Scott, D., and C.Y. Lyons (1979) Homogenous sensitivity of human peripheral blood lymphocytes to
radiation-induced chromosome damage, Nature (London), 278, 756-758.
Wagner, R., E. Schmid and M. Bauchinger (1983) Application of conventional and FPG-staining for the
analysis of chromosome aberrations induced by low levels of dose in human lymphocytes, Mutation
Res., 109, 65-71.