Mutagenesis vol.11 no.6 pp.537-546, 1996
A comparison of conventional metaphase analysis of
Giemsa-stained chromosomes with multi-colour fluorescence
in situ hybridization analysis to detect chromosome aberrations
induced by daunomycin
Sian Ellard1-3, Sheila Toper2, GUI Stemp2,
Elizabeth M.Parry1, Phillip Wilcox2 and James M.Parry1
'School of Biological Sciences, University of Wales Swansea, Singleton
Park, Swansea SA2 8PP, "kjlaxo Wellcome Group Research and
Development Ltd, Park Road, Ware, Herts SGI2 ODP, UK
3
To whom correspondence should be addressed at: Royal Devon and Exeter
Hospital NHS Trust, Department of Pathology, Church Lane, Heavitree,
Exeter EX2 5AD, UX
Chromosome aberrations induced by daunomycin, a widely
used positive control compound for in vitro cytogenetics
assays, were identified by multi-colour fluorescence in situ
hybridization with probes for chromosomes 1, 2 and 3.
The frequency and distribution of aberration types were
compared to conventional metaphase analysis of Giemsastained chromosomes from parallel human lymphocyte
cultures. Multi-colour chromosome painting was a more
sensitive method for detecting daunomycin-induced chromosome aberrations compared with conventional metaphase analysis because: (i) a higher level of statistical
significance was achieved at low doses; and (ii) the increases
in aberration frequencies compared with controls were
greater. The majority of exchanges identified by Giemsastaining were unstable and were likely to lead to cell death.
In contrast, those detected by FISH were mostly stable
exchanges which may be transmitted to cell progeny. Multicolour FISH using whole chromosome probes may provide
an elegant solution to the problem of identifying non-lethal,
heritable exchange events. The benefit of this technique
is the quantification of a cytogenetic endpoint directly
associated with carcinogenesis.
Introduction
Quantification of structural chromosome aberrations is an
integral component in the strategy for the assessment of
the in vitro genotoxic potential of new pharmaceuticals and
industrial chemicals. Current protocols rely upon the enumeration of exchanges, deletions and gaps in Giemsa-stained
chromosomes mainly derived from human primary lymphocyte
or Chinese hamster immortal cultures. The predominant categories of exchanges observed are chromatid interchange
figures, dicentric chromosomes and rings. Reciprocal translocations are evident in Giemsa-stained slide preparations only
when the rearranged chromosomes differ significantly from
the normal karyotype (Bender et al, 1988), hence it is not
feasible to make meaningful estimates of such frequencies.
Banding techniques (e.g. G-banding; Sumner, 1981) may reveal
the presence of translocations but such analyses are timeconsuming and requires considerable expertise. Fluorescence in
situ hybridization (FISH) with chromosome-specific composite
DNA probes ('chromosome painting') permits the rapid identification of reciprocal translocations involving 'target' chromosomes. Rearrangements involving the target chromosome and
any other unlabelled chromosome are revealed as bi-coloured
© UK Environmental Mutagen Society/Oxford University Press 1996
structures. Although the analysis of translocations by FISH is
limited to a fraction of the genome, the scoring of chromosome
painting is up to 30 times faster than G-banding (Tucker et al,
1993) and a greater number of cells may be scored.
Cells which contain dicentric, ring chromosomes or deletion
aberrations are generally unstable and such cells are likely to
die within the next few generations. In contrast, reciprocal
translocations are regarded as stable, heritable exchanges
because there is no loss of genetic information. In the in vivo
situation, chromosome painting has been used to demonstrate
the presence of translocations in the peripheral lymphocytes
of Hiroshima atom bomb survivors several decades after
exposure (Lucas etal., 1992a). The persistance of translocations
relative to dicentric chromosomes has been shown in an
individual exposed to tritiated water in 1985. The translocation
frequency measured by chromosome painting 6 years later
was similar to the dicentric frequency measured 39 days postexposure, although the frequency of dicentrics had been
reduced by a factor of about three during the intervening
period (Lucas et al, 1992b).
Chromosomal rearrangements may lead to mutation, transformation, gene amplification, gene activation and thus to
carcinogenesis. The 'master gene' model suggests that transcription factors (the 'master genes' which are responsible for
positive or negative regulation of 'responder genes') are the
main target for chromosome translocation activation in acute
leukaemias and solid tumours (Rabbitts, 1991, 1993). Translocations in leukaemias lead to the activation of transcription
factors either by association with T-cell receptor or immunoglobulin genes, or by gene fusion (reviewed by Rabbitts,
1994). For example, the t(8;14)(q24;q32) translocation seen in
Burkitt's lymphoma results in the activation of the c-MYC
oncogene whereas formation of the Philadelphia chromosome
in chronic myeloid leukaemia by a translocation of chromosomes 9q34 and 22qll results in the expression of a fusion
protein. Investigations of translocations in solid tumours have
demonstrated the production of chimaeric transcription factors
in liposarcoma, Ewing's sarcoma, rhabomyosarcoma and synovial sarcoma (Rabbitts, 1994 and references therein).
The effects of reciprocal translocations in germ cells may
be diverse, depending upon segregation patterns and the
breakpoint locations. An individual carrying a balanced translocation will usually be phenotypically normal. However, such
persons have an increased risk of producing gametes with an
unbalanced chromosome complement which can result in
spontaneous miscarriage, perinatal death or in a child with a
congenital abnormality. It is estimated that approximately one
in 500 newboms carry a balanced translocation (COM, 1989).
The ease of identification of reciprocal translocations by
chromosome painting has been demonstrated after in vitro
exposures to radiation (Lucas et al., 1992a, 1993; Natarajan
et al., 1992; Schmid et al, 1992; Bauchinger et al, 1993;
Nakano et al., 1993; Tucker et al, 1993, 1995; Matsuoka
et al, 1994), restriction endonucleases (Abella Columna et al.
537
S.EIlard et aL
1993) and bleomycin (Hoffmann et al., 1994; Ellard et al.,
1995).
Primary reciprocal translocations may be observed at the
first metaphase after X-irradiation or restriction endonuclease/
bleomycin treatment of G) chromosomes where the initial
lesions are double strand breaks. Secondary translocations may
arise at the second metaphase after exposure to UV-irradiation
or S-phase dependent chemicals, as the products of symmetrical
chromatid interchanges (seen at the first metaphase) where the
primary lesions are thymidine dimers, 6-4 photoproducts,
DNA: DNA crosslinks or DNA adducts which may be converted
to strand breaks during S phase.
The established role of reciprocal translocations in both
carcinogenesis and birth defects suggests that quantification
of translocations in in vitro genotoxicity studies would be a
particularly relevant endpoint. With the advent of chromosome
painting, the technology is now available to implement such
studies. We elected to commence our studies by investigating
a compound widely used as a positive control for in vitro
chromosome aberration assays.
Daunomycin is a secondary metabolite of Streptomyces
peucetius var. caesius which is used therapeutically in the
treatment of leukaemias. It is an intercalating agent which
induces single- and double-strand breaks and sister chromatid
exchanges (Gilman et al., 1990). Scission of DNA is believed
to be mediated either by the action of topoisomerase II (Tewey
et aL, 1984) or by the generation of free radicals.
In the present study, human lymphocyte cultures were treated
with graded doses of daunomycin according to the standard
protocol used for in vitro cytogenetic assays in the Glaxo
Wellcome laboratory. Conventional metaphase analysis was
conducted by a Glaxo Wellcome cytogeneticist and compared
to multi-colour FISH analysis performed in the Swansea
laboratory.
[Sigma, type III-A; final concentration 100 Ug/ml in 2X sodium chloride/
sodium citrate (SSC)] at 37°C. Human chromosome specific DNA probes were
purchased from Vysis and FISH performed according to the manufacturer's
instructions The hybridization mix comprised 0.5 uj chromosome 1 probe
labelled with Spectrum Orange, 0.5 (il chromosome 1 probe labelled with
Spectrum Green, 1 \L\ chromosome 2 probe labelled with Spectrum Orange,
1 ul chromosome 3 probe labelled with Spectrum Green and 7 uJ hybridization
buffer (or 1 ul chromosome 4 probe labelled with Spectrum Green, 1 nl
chromosome 8 probe labelled with Spectrum Orange, 1 |il dH2O and 7 |il
hybridization buffer). Probe DNA was denatured at 70°C for 5 min and
applied to chromosomal DNA (denatured for 2 min at 70°C in 7 0 * formamide
(Sigma)/2x SSC]. Following overnight hybridization at 37°C, slides were
washed at 45°C in 5fJ% formamide/2X SSC (3X10 min), once in 2X SSC
(10 min) and finally in 2X SSC/0.1% Nonidet-P-40 for 5 min The slides
were mounted in a 1/10 dilution of 4'6-diamidino-2-phenylindole (DAPI)
counlerstain (VysisVantifade (Vector) and stored in the dark at 4°C After
FISH analysis, the slides were stained with Giemsa (20%) and aberrant
metaphases were re-examined using standard light microscopy (Nakano
et al., 1993) in order to confirm the number of centromeres per rearranged
chromosome.
Scoring of aberrations
Conventional metaphase analysis (200 cells per dose) was conducted in the
Glaxo Wellcome laboratory. A Miamed Metaphase Finder (Leitz) was used
to locate metaphase cells and aberrations were classified according to the
scheme of Savage (1976) with modifications described by Scott et al. (1983).
Fluorescence microscopy was performed using an Olympus BH2 microscope
equipped with single band pass filters to visualize Spectrum Green and Orange
(Olympus) and a triple band pass filter (chromatechnology) which allows
simultaneous detection of DAPI, Spectrum Green and Spectrum Orange.
Where possible, 500 metaphases were scored for each dose point.
A Probemaster digital imaging system (PSI Ltd) was used to capture,
digitize, annotate and print multi-colour images All aberrant metaphase cells
were recorded. The Protocol for Aberration Identification and Nomenclature
Terminology (PAINT) system described by Tucker et al. (1995) was employed
to describe aberrations detected by chromosome painting, where t = translocation, die = dicentric/polycentnc chromosome, ins = insertion and r = ring.
Fragmented chromosomes are denoted as e.g. f(C), f(c) or f(ac) in order to
describe both acentric and centric chromosome products of breakage In
contrast to staining by Giemsa, chromosome painting clearly reveals the
deleted chromosome (i.e. centnc with part of an arm missing) in addition to
the acentric fragment. An example is shown in Figure la. [N.B. The PAINT
system includes only acentric fragments, e.g ace(c) or ace(ac)]
Materials and methods
Cylological preparation
Peripheral blood was obtained from a healthy male donor (aged 40 years).
Cultures (10 ml) were initiated in the Glaxo/Wellcome laboratory by the
addition of 1-9 ml Iscove's medium (Gibco BRL) supplemented with penicillin
(100 U/ml), streptomycin (100 |ig/ml) and phytohaemagglutinin [Wellcome
(reagent grade), final concentration 0.6% v/v] and incubated at 37°C for 48 h
before the addition of daunomycin.
Replicate cultures (eight per dose) were then subjected to a continuous
exposure of daunomycin hydrochloride (Sigma) dissolved in sterile water for
irrigation (25, 50, 75 and 100 ng/ml). Cultures were harvested according to
conventional methods (Scott et ai, 1990) either 20.5 or 44.5 h after the start
of treatment. The 20.5 h period represents 1.5 cell cycles (Glaxo Wellcome
in-house data) and the second harvest takes place 24 h later. Metaphase cells
were accumulated by the addition of demecolcine (Sigma; 0.2 ng/ml) 2 h
prior to harvest. Air-dried chromosome spreads from two cultures per dose
(per harvest) were stained with 10* Giemsa. The cell pellets from the
remaining cultures were transferred to the Swansea laboratory and stored at
-20°C in methanol:acetic acid (3:1) prior to in situ hybridization.
In situ hybridization
Cell pellets were resuspended in fresh, pre-chilled fixative and a few drops
of cell suspension dropped onto microscope slides. These slides were aged
overnight (or for up to 2 weeks) at -20°C and pretreated for 30 min in RNase
Results
Examples of metaphase cells painted simultaneously with
probes for chromosomes 1, 2 and 3 or 4 and 8 are shown in
Figure 1. The technique of ratio-mixing was incorporated in
the chromosome painting protocol in order to generate three
colours from two fluorochromes; chromosomes 1, 2 and 3
appear yellow, red and green respectively when viewed through
a triple band pass filter. Spectrum Green-labelled chromosomes
1, 3 and 4 were visible with the blue filter (excitation 490 nm)
and Spectrum Orange labelled chromosomes 1, 2 and 8 with
the green filter (545 nm). Chromosomes 2, 3, 4 and 8
exhibited red or green fluorescence along their total length.
The pericentromeric heterochromatin of chromosome 1 was
not labelled by the probe, but was visible by the intense staining
observed with DAPI which is specific for the predominance of
AT base pairs which characterize heterochromatic DNA. In
addition, the distal parts of the short arms (Ip33 to lpter)
were not labelled by the probe, thus small deletions or
Fig. 1. Metaphase human lymphocyte cells following multi-colour FISH Aberrant chromosomes are described according to the Protocol for Aberration
Identification and Nomenclature Terminology (PAINT) system where A = any unpainted chromosome. B = chromosome 1, C = chromosome 2 and
D = chromosome 3 Capital and lower case letters denote chromosomal portions with and without centromeres respectively. Aberrations illustrated include:
(a) a chromosome 2 break; (b) a reciprocal translocation involving one copy of chromosome 2 and an unpainted chromosome; (c) a reciprocal translocation
involving one copy of chromosome 3 and an unpainted chromosome, (d) a non-reciprocal translocation involving chromosomes 2 and 3: (e) a dicentric
chromosome with acentric fragment generated by an exchange between chromosome 3 and an unpainted chromosome, (f) shows a metaphase with two
normal copies of chromosomes 4 and 8
538
Chromosome painting to detect daunomycln induced aberrations
539
S.EUard et al
Table I. Proportion of aberrant cells detected by Giemsa staining of all chromosomes or mulu-colour
chromosome painting (chromosomes 1, 2 and 3) in duplicate human lymphocyte cultures treated with
daunomycin
Giemsa staining
Harvest
Dose Mitotic
(ng/ml) Index
Cells
Mean
with
%
exchanges (±SD)
Aberrant
cells
(-gaps)
Multi-colour FISH
Mean
Mean
Cells
%
with
%
(±SD) exchanges (±SD)
Aberrant
cells
(-gaps)
Mean
%
(±SD)
First
0
8.8
1/100
0/100
0.5
(0.71)
4/100
0/100
2.0
(2 83)
0/200
1/200
0.25
(0 35)
0/20O
1/200
0.25
(0 35)
(20.5
hours
post
treatment)
25
5.9
0/100
0/100
0
1/100
1/100
1.0
3/250
3/250
1.2"
4/250
4/250
1.6"
50
4.1
2/100
2/100
2.0
10/100
9/100
9.5"
(0.71)
3/250
4/250
1.4"
(0 28)
6/250
11/250
3.4"
(141)
75
4.0
8/100
9/100
8.5"
(071)
19/100
20/100
19.5"
(0 71)
10/250
6/250
3.2"
(1 10)
18/250
16/250
6.8"
(0 57)
100
2.6
15/100
16/100
15.5"
(071)
34/100
29/100
31.5"
(3 54)
7/100
3/100
5.0"
(2 80)
16/100
10/100
13.0"
(4 24)
Second
0
3.2
0/100
1/100
0.5
(0 71)
2/100
1/100
14
(0 71)
1/300
0/300
0.17
(0.24)
1/300
0/300
0.17
(0 24)
(44.5
hours
post
treatment)
25
2.2
2/100
2/100
2.0
4/100
5/100
4.5"
(0.71)
2/250
1/250
0.6
(0.28)
3/250
4/250
1.4"
(0 28)
50
3.9
4/100
3/100
3.5"
(071)
11/100
6/100
8.5"
(3 54)
6/250
7/250
2.6"
(0 28)
10/250
10/250
4.0"
75
2.1
6/68
5/100
6.5"
(198)
11/68
16/100
16.1"
(0.14)
5/100
2/100
3.5"
(2.12)
8/100
5/100
6.5"
(2 12)
*P < 0.05 (one-tajled Fisher's exact test), in comparison with respective controls.
**P < 0.01 in comparison with respective controls.
rearrangements involving the p arm termini are likely to remain
undetected.
The percentages of aberrant cells detected by Giemsa
staining or multi-colour FISH following treatment with daonomycin are shown in Table I. A statistically significant increase
(P < 0.01) in the proportion of aberrant cells and the number
of cells containing exchange aberrations was recorded in
cultures exposed to doses of 50 ng/ml (second harvest only),
75 or 100 ng/ml (both harvest times) and analysed by Giemsa
staining or multi-colour FISH. However, differences between
the two methods were observed at the lower doses. For the
first harvest, after treatment with 25 ng/ml daunomycin, no
increase in aberrant cells was seen by Giemsa staining, although
a significant increase in aberrant cells (P < 0.01) and cells
with exchanges (P < 0.05) was detected by multi-colour
painting with probes for chromosomes 1. 2 and 3. Similarly,
at a dose of 50 ng/ml, the number of cells with exchange
aberrations identified by Giemsa staining was not significantly
540
elevated compared with the control, whereas for FISH, significance at the P < 0.01 level was achieved.
Table II shows the frequencies of aberrations detected by
conventional metaphase analysis and multi-colour chromosome
painting. Greater fold increases compared to the control were
observed in aberration frequencies enumerated by FISH compared to Giemsa staining.
The aim of this study was to compare the performance of
multi-colour chromosome painting with that of conventional
metaphase analysis for the detection of chromosomal rearrangements. We wished to compare the estimated genomic aberration
frequencies obtained from the chromosome painting analysis
with the data generated by Giemsa staining. The A-group
chromosomes 1. 2 and 3 comprise 23.9% of the genome
(Mayall et al., 1984). Using the approach suggested by Lucas
et al. (1989a,b. 1992a, 1993), the proportion of the total
exchanges detected was calculated as 40.2% (Ellard et al.,
1995).
Chromosome painting to detect daunomycin induced aberrations
Table II. Frequency of aberrations and fold increases compared to controls detected by Giemsa
staining or multi-colour chromosome painting (1,2 and 3) in human lymphocyte cultures treated
with daunomycin. Genomic exchange frequencies for chromosome painting were calculated by
multiplying the observed frequencies by 2.5. This calculation is explained in the Results section
Exchanges per 100 cells
Harvest
Aberrations per 100 cells
Dose
(ng/ml)
FISH
Giemsa
Observed
Genomic
x2.5
Giemsa
FISH
First
0
0.5
0.25
0.63
2.5
0.25
(20.5
hours
post
treatment)
25
0
1.2
(4.8x)
3
1
1.6
(6.4x)
50
2
(4x)
1.4
(5.6x)
3.5
11.5
(4.6x)
3.4
(13.6x)
75
9
(18x)
3.8
(15.2x)
9.5
23.5
(9.4x)
8.2
(32.8x)
100
25
(50x)
8
(32x)
20
55.5
(22.2x)
16.5
(66x)
Second
0
0.5
0.17
0.43
1.5
0 17
(44.5
hours
post
treatment)
25
3.5
(7x)
0.8
(4.7x)
2
6.5
(4.3x)
1.6
(9.4x)
50
4
(8x)
2.8
(16.5x)
7.5
14.5
(9.4x)
4.6
(27x)
75
7.8
(15.6x)
3.8
(22.4x)
10
29.2
(19.5x)
6.8
(40x)
The estimated genomic frequencies calculated from the
chromosome painting data are included in Table n. These
genomic frequencies were calculated by multiplying the
observed frequency of exchanges by 2.5 (i.e. 100/40.2) and
are similar to the frequencies observed in Giemsa-stained
preparations.
The results from the analysis of Giemsa-stained chromosomes are presented in Table m. A dose-dependent response
was observed for the induction of both chromosome and
chromatid-type deletions and exchanges. At the first harvest,
the proportion of induced chromatidxhromosome aberrations
was almost equal. Conversion of chromatid aberrations to chromosome-type aberrations during subsequent cell divisionfs)
resulted in a predominance (89%) of chromosome-type aberrations at the second harvest, 24 h later.
Table IV details the frequency and types of aberrations
detected by multi-colour FISH. These included deletions (see
Figure la), reciprocal and non-reciprocal translocations (Figure
lb, c and d), dicentrics (Figure le), centric rings and nonreciprocal exchanges involving multiple mutual sites
(NEMMS). With only two exceptions, these aberrations were
of the chromosome-type.
The distribution of exchange aberrations identified by
Giemsa staining and FISH is illustrated by Figure 2. The
majority of exchanges (53 and 73% at the first and second
harvest respectively) detected by FISH arc likely to be stable
(i.e. translocations, symmetric quadriradials and some inter/
intra arm exchanges). In contrast, exchange aberrations identified by conventional metaphase analysis were mostly unstable
(19 and 28% of exchanges from the first and second harvest
respectively were assigned to the categories listed above).
In view of the smail number of exchanges involving the
A-group chromosomes identified by Giemsa-staining of cultures harvested after 44.5 h exposure to daunomycin (12%
541
S.EIlard et at
Table III. Frequency of chromosome and chromatid aberrations detected by Giemsa staining in human lymphocytes exposed to daunomycin
Harvest
Chromosome aberrations / 100 cells
Chromatid aberrations / 100 cells
Dose
(ng/ml)
Cells
scored
Gaps
/100
cells
DEL
0.5
First
0
200
2.0
(20.5
hours
post
treatment)
25
200
5.0
50
200
16.5
3.0
75
200
14.0
6.0
100
200
28.0
130
Second
0
200
3.5
0.5
(44.5
hours
post
treatment)
25
200
2.0
1.0
50
200
5.5
1.5
75
168
6.0
1.2
SQ
AQ
TRI
IAE
DEL
TRANS
1 5
DIC
RING
DM
0.5
1.0
0.5
20
65
05
1.0
1.5
3.0
0.5
8.5
0.5
25
0.5
30
7.0
0.5
17.5
1.5
85
2.0
0.5
0.5
0.6
1 2
0.5
0.5
2.0
1.5
1.5
9.0
1.0
2.0
20.2
0.6
5.4
0.5
DEL = deletion; SQ = symmetrical quadriradial; AQ = asymmetric quadriradial; TRI= triradial, IAE = inter or intra-arm interchange; TRANS =
translocation; DIC = dicentnc; RING = centric nng; DM = double min. The number of exchanges involving A-group chromosomes was 22/62 (36%) and
3/25 (12%) at the first and second harvest times respectively.
compared with a predicted 40.2%), we investigated possible
chromosome bias by employing probes for chromosomes 4
(B-group) and 8 (C-group) for further analysis of untreated
cells and those exposed to a concentration of 50 ng/ml
daunomycin for 44.5 h. These results are included in Table
IV. The estimated genomic exchange frequency after treatment
with 50 ng/ml daunomycin was calculated as 4.6/100 cells.
This frequency does not appear significantly different to that
calculated for chromosome painting with probes for 1,2 and
3, although the number of exchanges observed was small.
Discussion
Multi-colour chromosome painting is a relatively new development in the advancement of FISH technology. Until recently,
the maximum number of colours reported by ratio-mixing was
12 (Dauwerse et al, 1992). The ultimate goal is to generate a
molecular karyotype where each chromosome is recognized
by its unique ratio of fluorochromes. This has now been
achieved by Speicher et al. (1996). However, until image
analysis systems with the appropriate computer software
become widely available for semi-automated analysis of chromosome aberrations, this approach is not practical for the
analysis of large numbers of cells. In the meantime, a compromise is sought between using multiple chromosome probes to
obtain maximum information per metaphase cell without
sacrificing the speed of analysis which is one of the main
advantages of the technique.
In this study, exchanges involving A-group chromosomes
were monitored by FISH. In order to compensate for the
reduced proportion of exchanges detected (relative to Giemsastaining where the whole genome is examined), the number
of cells scored for all except the top doses (100 ng/ml first
harvest and 75 ng/ml second harvest) was increased to 500,
542
i.e. 200 cell equivalents. It was possible to score 500 painted
cells within 5 h (including time for image capture, annotation
etc.). This is comparable with the time required to analyse
200 Giemsa-stained metaphases and yields a greater amount
of information per cell.
Analysis of 2000 metaphase cells from untreated lymphocyte
cultures by chromosome painting revealed three reciprocal
translocations (two involving chromosome 2 and an unpainted
chromosome; one involving chromosomes 4 and 8). This low
background frequency of aberrations (the genomic exchange
frequency is 0.5%) undoubtedly contributes to the increased
statistical sensitivity of the FISH method compared to conventional metaphase analysis (see Table I) and to the greater fold
increases observed compared with the controls (Table II).
At the first harvest (20.5 h from the start of treatment) 49%
of the induced aberrations identified by Giemsa staining
were of the chromatid type. The majority of chromatidtype aberrations appear to go undetected by FISH analysis,
presumably due to inferior chromosome morphology/reduced
optical resolution obtained with painted preparations. However,
as predicted, the frequency of translocations identified by FISH
exceeded that determined by conventional analysis, with the
result that for the first harvest, the estimated genomic exchange
frequencies for the two methods were similar. At the second
harvest, the majority of aberrations identified in Giemsa stained
preparations were of the chromosome-type (89%). We therefore
expected to obtain higher genomic exchange frequencies by
chromosome painting, since both dicentrics and translocations
would be readily visible. This was indeed the case for doses
of 50 and 75 ng/ml, but not for 25 ng/ml.
Conventional metaphase analysis of these second harvest
cultures by Giemsa-staining revealed the involvement of the
A-group chromosomes in only 3 of the 25 exchange aberrations
Chromosonte painting to detect d a u n o m y d n induced aberrations
Table IV. Frequency of chromosome aberrations detected by multi-colour FISH ( 1 , 2 and 3) in human lymphocytes exposed to
daunomycin
Harvest
First
(20.5
hours
post
treatment)
Dose
(ng/ml)
Cells
scored
0
400
0
50
75
100
(44.5
hours
post
treatment)
DEL
RT
2
3
OJ25
Total
0.25
NRT
DIC
RING
NEMMS
4
8
Toad
25
Second
500
Aberrations/ 100 cells
Paint
0
0
25
50
2
3
0.4
02
0J2
Tout
04
0.4
1
2
3
08
10
02
04
0.6
0.2
Total
20
1.0
02
02
1.0
24
1.0
02
1.0
02
07"
0.8
0.4
01
02
2
3
Total
4.4
14
1.4
0.8
02
1
2
3
30*
2.5
3.0
10
0.5
0.5
0.5
05
05
0.5
10
1.0
2.0-
Total
85
10
10
1.0
1.0
4.0
500
500
500
200
600
500
500
500
50
1000
75
200
1
2
3
0.17
Total
0.17
4
8
0.1
0.1
Total
02
0J
0.5
08
02
03
0.4
1
2
3
0.6
02
02
02
02
02
Total
0.8
0.4
0.4
1
2
3
0.6
0.6
06
0.6
0.4
0.4
02
02
Total
1.8
14
0.4
0.6
4
S
09
0.5
02
0.1
0.4
0.1
Total
1.4
03
1
2
3
2.0
0.5
0.5
1J25
0 75
1.25
Total
30
2.0
2.0
0.4
02
02
02
0.4
0J2
0.6
0.1
0.75
DEL = deletion, RT = reciprocal translocation, NRT = non-reciprocal translocation, DIC = dicentric, RING = centric ring,
NEMMS = non-reciprocal exchange involving multiple mutual sites.
"Includes one insertion.
''Includes one chromatid deletion.
Includes one triradial.
543
S.EIIard et al
FIRST
HARVEST
W RECIPROCAL TRANSLOCATION (FISH)
22j TRANSLOCATION (GIEMSA)
^ ] SYMMETRICAL QUADRIRADIAL
GIEMSA
FISH
^
DICENTRIC CHROMOSOME
Q
ASYMMETRICAL QUADRIRADIAL
[ \ - NON-RECIPROCAL TRANSLOCATION
^2
TRIRADIAL INTERCHANGE
|
CHROMOSOME INTER / INTRA ARM EXCHANGE
SECOND
|
HARVEST
|
GIEMSA
CHROMATID INTER / INTRA ARM EXCHANGE
| NON-RECIPROCAL EXCHANGE INVOLVING
MULTIPLE MUTUAL SITES (NEMMS)
FISH
Fig. 2. Distribution of exchanges induced by daunomycin
(12%). To explore the possibility that chromosomes 1, 2 and
3 are underepresented in exchange aberrations at the second
harvest, we probed slide preparations from cultures exposed
to 50 ng/ml daunomycin with paints for chromosomes 4 and
8. The estimated genomic frequencies were not significantly
different to those calculated from FISH with probes for
chromosomes 1, 2 and 3. Further investigation by testing other
chromosome combinations is required to establish whether
chromosome bias is a confounding factor.
G-banding analysis of radiation induced translocation
breakpoints has revealed over- or under-representation of
certain chromosomes in human fibroblast and lymphocyte
cultures irradiated in vivo and in vitro (San Roman and Bobrow,
1973; Bauchinger and Gotz, 1979; Dutrillaux et al., 1981; Lee
and Kamra, 1981; Tanaka et al., 1983; Buckton, 1983; Kano
and Little, 1986; Barrios et al., 1989; Lucas et al., 1992a).
The ease of identification of translocations by chromosome
painting presents an opportunity to test this hypothesis by
comparing the relative propensity for participation in exchanges
of different chromosome pairs or combinations of chromosomes. However, the results obtained to date show some
discord. The frequency of exchanges induced in irradiated
human lymphocytes (Pandita et al, 1994) and fibroblasts
(Kovacs et al., 1994) was proportional to DNA content of the
hybridized chromosomes, whereas Knehr et al. (1994) observed
some deviations from the predicted frequencies for chromosomes 2, 9, 10 and 12 in human lymphocytes. The issue of
chromosome bias is an important question when deciding
upon the target chromosomes for hybridization. Until more
information is available regarding both the spontaneous distribution of translocation breakpoints and the spectrum induced
by genotoxins, this issue will remain unresolved.
The failure to detect chromatid interchange figures is a
severe limitation of the chromosome painting method; no
quadriradials and only one triradial were observed in this study.
Presumably such rearrangements were erroneously identified as
chromosomes lying in close proximity to each other. This has
implications for the application of chromosome painting for
the analysis of S phase dependent chemicals which induce
chromatid-type aberrations at the first metaphase following
exposure. However, we have recently utilized a sequential
544
hybridization protocol to label the centromeres with a pancentromeric DNA probe (Cambio) after painting chromosomes
1, 2 and 3. We have successfully applied this methodology
to the identification of chromatid interchanges induced by
camptothecin (unpublished data).
Multi-colour chromosome painting demonstrated the induction of both reciprocal and non-reciprocal translocations by
daunomycin. These non-reciprocal translocations may be true
terminal translocations or reciprocal translocations where one
of the breakpoints is located close to the telomere. Nonreciprocal translocations comprised 43% (24/56) of translocations which suggests that daunomycin may preferentially
target the telomeric/subtelomeric region of chromosomes. Nonreciprocal translocations have been identified in peripheral
lymphocytes from unexposed persons, victims of the Chernobyl
accident and lymphocytes irradiated in vitro (Tucker and
Senft, 1994).
A greater number of reciprocal translocations compared to
dicentric chromosomes were observed by chromosome painting
(Table IV), particularly at the second harvest. If some of the
chromosome exchanges are secondary aberrations derived from
chromatid-type exchanges, this may be due to the loss of
cells carrying asymmetric chromatid exchanges (triradials
and asymmetric quadriradials) and propagation of cells with
symmetric quadriradials which are duplicated to form translocations (both reciprocal and non-reciprocal).
FISH with chromosome-specific probes is a powerful tool
for the identification of structural chromosome rearrangements.
Analysis by chromosome painting is fast and more metaphases
are analysable compared with conventional methods since the
requirement for perfectly spread metaphase chromosomes is
obviated. In the present study, ~90% of metaphases at the top
dose were judged to be scorable by FISH (based upon the
criteria that the painted chromosomes were brightly painted
and clearly identifiable as normal or aberrant) whereas only
-50% were scored on the equivalent Giemsa-stained slides
where the criteria were more stringent; metaphase spreads
with 45 or 46 clearly defined centromeres and well spread
chromatid arms. This dose of daunomycin (100 ng/ml) caused
some aberrant metaphases which, due to the nature and
complexity of the aberrations together with inferior chromo-
Chromosome painting to detect daunomycin induced aberrations
some morphology, did not fulfil the criteria for analysis. Since
a greater proportion of metaphase cells are scored by FISH, a
more accurate estimate of the percentage of aberrant cells may
be obtained because bias introduced by the cell selection
process is avoided. For example, if a chemical has a detrimental
effect upon chromosome morphology, the metaphase population selected for analysis may be biased towards undamaged
cells with the result that the aberration frequency is underestimated.
Examination of the distribution of exchange aberration types
revealed that the two methods of analysis detect different types
of exchanges which will have different implications for the
ultimate fate of the affected cell. Conventional metaphase
analysis detects mainly unstable exchanges and deletions which
will lead to cell death, whereas chromosome painting allows
the identification of non-lethal, stable exchanges which are
transmissible. The major advantage of the chromosome painting technique is the detection of these reciprocal translocations,
both primary and secondary. Reciprocal translocations are
associated with proto-oncogene activation and congenital
abnormalities, hence this approach provides an opportunity to
monitor the induction of an endpoint which is directly associated with carcinogenesis and birth defects.
Acknowledgements
This work was supported by a grant from GlaxoAVellcome Group Research.
The expert technical assistance of Karen Sykes was appreciated.
References
Abella Columna,E., Giaccia.AJ., EvansJ.W., Yates.B.L. and Morgan, W.F.
(1993) Analysis of restriction enzyme-induced chromosomal aberrations by
fluorescence in situ hybridization. Environ. Mol. Mutagenesis, 22, 26-33.
Barrios.L, Miro.R., CaballiruM.R., Fuster.C, Guedea,F., Subias.A. and
EgozcueJ. (1989) Cytogenetic effects of radiotherapy breakpoint
distribution in induced chromosome aberrations. Cancer Genetics and
Cytogenetics, 41, 61-70.
Bauchinger.M. and Gotz,G. (1979) Distribution of radiation induced lesions
in human chromosomes and dose-effect relation analysed with G-banding.
Rad. Env. Biophysics, 16, 355-366.
Bauchinger.M., Schmid,E., Zitzelsberger.H., Braselmann.H. and Nahrstedt,U.
(1993) Radiation-induced chromosome aberrations analysed by two-colour
fluorescence in situ hybridization with composite whole chromosomespecific DNA probes and a pancentromeric DNA probe. Int. J. Rad. Biol.,
64, 179-184.
Bcnder.M.A., AwaAA., Brooks.A.L., Evans.H.J., Groer.P.G., Littlefield.L.G.,
Pereira.C, Preston.RJ. and Wachholz,B.W. (1988) Current status of
cytogenetic procedures to detect and quantify previous exposures to
radiation. Mutat. Res., 196, 103-159.
Buckton.K.E. (1983): Chromosome aberrations in patients treated with Xirradiation for ankylosing spondylitis. In Radiation-Induced Chromosome
Damage in Man. Liss, New York, pp. 491-511.
Committee on Mutagenicity of Chemicals in Food, Consumer Products and
the Environment (1989) Department of Health Report on Health and Social
Subjects. 35. Guidelines for the Testing of Chemicals for Mutagenicity. 1 st
cdn. HMSO, London, pp. 74-86.
DauwerseJ.G., WieganU., RaapAK., Breuning.M.H. and van Ommen.GJ.B.
(1992) Multiple colors by fluorescence in situ hybridization using ratiolabelled DNA probes create a molecular karyotype. Hum. Mol. Genet., 1,
593-598.
Dutrillaux.B., Viegas-PequinotE.. Prod'homme.M.. Sportes.M. and Prieur.M.
(1981; Tentative estimate of the risk of chromosomal disease due to
radiation induced translocations in man. Mutat. Res., 82, 191-200.
Ellard.S., Parry.E.M. and ParryJ.M. (1995) Use of multi-colour chromosome
painting to identify chromosomal rearrangements in human lymphocytes
exposed to bleomycin: a comparison with conventional cytogenetic analysis
of Giemsa-stained chromosomes. Environ. Mol. Mutagenesis, 26, 44—54.
Gilman,A.G., Rall.T.W.. Nies.A.S. and Taylor.P. feds) U990) The
Pharmacological Basis of Therapeutics. 8th edn. Pergammon, pp. 12411243.
Hoffmann.G.R.. Sayer.A.M. and Littlefield.L.G. (1994) Potentiation of
bleomycin by the aminothiol WR-1065 in assays for chromosomal damage
in G o human lymphocytes. Mutat. Res., 307, 273-283.
Kano.Y. and LittleJ.B. (1986) Site-specific chromosomal rearrangement
induced in human diploid cells by X-irradiation. Cytogenet. Cell Genet.,
41, 22-29.
Knehr.S., Zitzelsberger.H., Braselmann.H. and Bauchinger.M. (1994) Analysis
for DNA-propomonal distribution of radiation-induced chromosome
aberrations in various triple combinations of human chromosomes using
fluorescence in situ hybridization. Int. J. Rad. Bio!., 65, 683-690.
Kovacs.M.S., EvansJ.W.. Johnstaone.I.M. and BrownJ.M. (1994) Radiationinduced damage, repair and exchange formation in different chromosomes
of human fibroblasts determined by fluorescence in situ hybridization. Rad.
Res., 137, 34-13.
Lee.C.L.Y. and Kamra.O.P. (1981) The pattern of radiation-induced
transmissible aberrations in a human cell culture. Hum. Genet., 57, 380-384.
LucasJ.N., Tenjin.T., Straume.T., Pinkel.D., Moore II.D., Lil^M. and GrayJ.W.
(1989a) Rapid human chromosome aberration analysis using fluorescence
in situ hybridization. Int. J. Rod. Biol., 56, 35-44.
LucasJ.N., Tenjin.T., Straume.T., Pinkel.D., Moore II.D., Lilt,M. and GrayJ.W.
(1989b) Erratum. Int. J. Rod. Biol., 56, 201.
LucasJ.N., Awa.A., Straume.T., Poggensee.M., Kodama,Y., Nakano.O,
Ohtaki.K., Weier,H.-U., Pinkel.D., GrayJ. and Littlefield.G. (1992a) Rapid
translocation frequency analysis in humans decades after exposure to
ionizing radiation. Int. J. Rad. Biol., 62, 53-63.
LucasJ.N., Poggensee.M. and Straume.T. (1992b) The persistence of
chromosome translocations in a radiation worker accidentally exposed to
tritium. Cytogenet. Cell Genet., 60, 255-256.
LucasJ.N., Poggensee.M. and Straume.T. (1993) Translocations between
two specific human chromosomes detected by three-colour 'chromosome
painting'. Cytogenet. Cell Genet., 62, 11-12.
Matsuoka,A., TuckerJ.D., Hayashi.M., Yamazaki.N. and Sofuni.T. (1994)
Chromosome painting analysis in human lymphocytes in vitro. Mutagenesis,
9, 151-155.
Mayall.B.H., Carrano.A V., Moore n.D.H., Ashworth.L., Bennett,D.E. and
Mendelsohn.M. (1984) The DNA-based human karyotype. Cytometry, 5,
376-385.
Nakano.M., Nakashima,E., Pawel.DJ., Kodama,Y. and Awa,A. (1993)
Frequency of reciprocal translocations and dicentrics induced in human
blood lymphocytes by X-irradiation as determined by fluorescence in situ
hybridization. Int. J. Rad. Biol., 64, 565-569.
Natarajan.A.T, Vyas.R.C, Darroudi.F. and Vermeulen.S. (1992) Frequencies
of X-ray induced chromosome translocations in human peripheral
lymphocytes as detected by in situ hybridization using chromosome specific
DNA libraries. Int. J. Rad. Biol., 61, 199-203.
PanditaJ.K., Gregoire.V., Dhingra,K. and Hittelman.W.N. (1994) Effect of
chromosome size on aberration levels caused by gamma radiation as
detected by fluorescence in situ hybridization. Cytogenet. Cell Genet., 67,
94-101.
Rabbitts.T.H. (1991) Translocations, master genes and differences between
the origins of acute and chronic leukaemias. Cell, 67, 641-644.
Rabbitts.T.H. (1993) Chromosomal abnormalities in cancer cells: identification
of molecules important for tumour development. Biochem. Soc. Trans., 21,
809-815.
Rabbitts.T.H. (1994) Chromosomal translocations in human cancer. Nature,
372, 143-149.
San Roman.C. and Bobrow.M. (1973) The sites of radiation-induced breakage
in human lymphocyte chromosomes, determined by quinacrine fluorescence.
Mutat. Res., 18, 325-331.
SavageJ.R.K. (1976) Annotation: classification and relationships of induced
chromosomal structural changes. J. Med. Genet., 13, 103-122.
Schmid.E., Zitzelsberger.H., Braselmann.H., GrayJ.W. and Bauchinger.M.
(1992) Radiation-induced chromosome aberrations analysed by fluorescence
in situ hybridization with a triple combination of composite whole
chromosome-specific DNA probes. Int. J. Rad. Biol., 62, 673-678.
ScottD., Danford.N., Dean.B., Kirkland.DJ. and Richardson.C. (1983) In
vitm chromosome aberration assays. In Dean.B J. (ed.). Report nfU.K.E.M.S.
Sub-Committee on 'Guidelines for Mutanenkity Testing' Part I. 1st edn.
pp. 41-64.
Scott,D., Danford.N., Dean.B. and Kirkland.DJ. (1983) Metaphase
chromosome aberration assays in vitro. In KirklaniDJ. (ed.), 'Basic
Mutaqenicity Tests; U.K.E.M.S. Recommended Procedures' Pan I, Revised.
pp. 62-86. '
Speicher.M.R.. Gwyn BallarcLS. and WarcLD.C. r!996) Ka-yotypirjj human
chromosomes by combinatorial multi-fluor FISH. Nature Genet., 12, 368375.
SumnerAT., Evans.HJ. and Buckland.R.A. (1981) New techniques for
545
S.EIIard et al
distinguishing between human chromosomes. Nature (New Biology), 232,
31-32.
Tanaka,K., Kamad,N., Ohkjta.T. and Kuramoto,A. (1983) Nonrandom
distribution of chromosome breaks in lymphocytes of atomic bomb
survivors. J. Rad Res., 24, 291-304.
Tewey,K.M., Chen,G.L., Nelson.E M. and Liu.L.F. (1984) Intercalative
antitumour drugs interfere with the breakage-reunion reaction of mammalian
topoisomerase II. J. Biol. Chem., 259, 9182-9187.
Tucker J.D. and SenftJR. (1994) Analysis of naturally occurring and radiationinduced breakpoint locations in human chromosomes 1, 2 and 4. Radiation
Research 140, 31-36
TuckerJ.D., Ramsey.MJ., Lee.D.A. and MinklerJ.L. (1993) Validation of
chromosome painting as a biodosimeter in human penpheral lymphocytes
following acute exposure to ionizing radiation in vitro Int. J. Rad. Biol.,
64, 27-37.
TuckerJ.D., Lee.D.A., Ramsey.MJ., BrinerJ., Olsen,L. and Moore II.D.H.
(1994) On the frequency of chromosome exchanges in a control population
measured by chromosome painting. Mutat. Res., 313, 193-202.
TuckerJ.D., Lee.D.A. and Moore II.D.H. (1995a) Validation of chromosome
painting. II. A detailed analysis of aberrations following high doses of
ionizing radiation in vitro. Int. J. Rad. Biol., 67, 19-28.
TuckerJ.D.,
Morgan.W.F.,
Awa,A.A.,
Bauchinger.M.,
Blakey.D.,
Comforth.M.N., Littlefield.L.G., Natarajan.A.T. and Shasserre.C. (1995b) A
proposed system for sconng structural aberrations detected by chromosome
painting. Cytogenet. Cell Genet., 68, 211-221.
Received on October 30, 1995; accepted on July 3, 1996
546