Chromosoma (Berl.) 68, 287-302 (1978)
CHROMOSOMA 9 by Springer-Verlag 1978
The Use of Base Pair Specific DNA Binding Agents as Affinity Labels for the Study of Mammalian Chromosomes K.F. Jorgenson 3, J.H. van de Sande 1 and C.C. Lin 1' 2 Divisions of 1Medical Biochemistryand ZPediatrics, Faculty of Medicine and 3Department of Chemistry, The University of Calgary, Calgary, Alberta, Canada T2N 1N4
Abstract. The fluorochromes Hoechst 33258 and olivomycin are base pair specific D N A binding agents. The fluorescence enhancement of Hoechst 33258 and olivomycin in the presence of D N A can be directly related to the A - T and G - C content of the interacting D N A respectively. Cytological observations of metaphase chromosomes treated with these two compounds suggest that the fluorescent banding patterns produced are the reverse of one another.--Non-fluorescent base pair specific D N A binding agents have been used as counterstains in chromosome preparations to enhance the contrast of the banding patterns produced by the base specific fluorochromes. The non-fluorescent G - C specific antibiotic actinomycin-D enhanced the resolution of fluorescent bands produced by the A - T specific fluorochrome Hoechst 33258. Similarly the non-fluorescent A - T specific antibiotic netropsin was found to enhance resolution of the bands produced by the G - C specific fluorochrome olivomycin. Netropsin was also found to increase the differential fluorescent enhancement of complexes of olivomycin with DNAs of various base composition in solution. These findings suggest that counterstaining agents act through a base sequence dependent inhibition of subsequent binding by base pair specific f l u o r o c h r o m e s . - T h e base specific D N A binding agents have been used to differentiate different types of constitutive heterochromatin in mammalian species, and to facilitate chromosome identification in somatic cell hybrids. Introduction
Two groups of D N A binding fluorochromes which have the capacity to discriminate in their binding between base pairs are currently being used for chromosome identification. The first group consists of the A - T specific fluorochromes Hoechst 33258 (2-[2-(4-hydroxyphenyl)-6-benzimidazolyl]-6-(1-methyl-4-piperazyl)-benzimidazole), DBP (2,7-di-t-butyl proflavine), DAPI (4'-6-diamidino-2phenylindole) and DIPI (4'-6-bis(2'imidazolinyl-4'-5'-H)-2-phenylindole) (Hilwig and Gropp, 1972; Disteche and Bontemps, 1974; Schweizer, 1976; Schnedl
0009-5915/78/0068[0287/$03.20
288
K.F. Jorgenson et al.
et al., 1977a). Solution interaction studies of Hoechst 33258 in the presence of DNA have shown that binding of the fluorochrome to DNA is accompanied by a fluorescence enhancement which can be directly related to the A - T content of the DNA (Weisblum and Haenssler, 1974; Latt and Stetten, 1976). The same "Hoechst" type chromosome banding patterns produced by this group of fluorochromes can then be rationalized in terms of the distribution of A - T base pairs along the arms of the chromosomes. The second group of fluorochromes, consisting of olivomycin, chromomycin A3 and mithramycin, was recently found to produce characteristic R-bands on both plant and mammalian chromosomes (Schweizer, 1976; van de Sande et al., 1977a; Schnedl et al., 1977b). This group of fluorochromes exhibits enhanced fluorescence when bound to DNA which can be directly related to the G - C content of the DNA (van de Sande et al., 1977b). The R-banding patterns on chromosome arms produced by these fluorochromes appear to be a reversal of the banding patterns produced by the A - T specific fluorochromes (van de Sande et al., 1977b). The R-bands produced by this group of fluorochromes are assumed to be a reflection of their binding to G - C rich sequences along the chromosomes. Differential staining of chromosome regions by these groups of fluorochromes facilitates the unequivocal identification of individual chromosomes of a species. In addition, these compounds should also facilitate the identification of certain heterochromatic regions of mammalian chromosomes since it is known that in several species these regions contain mainly A - T or G - C rich DNA (Pardue and Gall, 1970; Jones, 1970; John and Corneo, 1971; Kurnit et al., 1973; Schreck et al., 1974; van de Sande et al., 1977b). However, direct application of all these fluorochromes to chromosome preparations has been found to produce rather poorly differentiated banding patterns (Jalal et al., 1975; Schweizer, 1976). This paper reports comparisons of both the specific banding patterns produced by the two groups of fluorochromes and the observed changes in their fluorescence when interacting with synthetic and natural DNAs of varying base composition. In addition, we report the use of the non-fluorescent base specific binding ligands netropsin and actinomycin D as counterstaining agents on chromosome preparations to improve the resolution of olivomycin R-bands and Hoechst bands respectively. Actinomycin D specifically binds to G - C sites on duplex DNA (Reich and Goldberg, 1964; Miiller and Crothers, 1968; Wells and Larsen, 1970), while netropsin binds specifically to A - T sites (Wartell et al., 1974). The effect of netropsin on the fluorescence modification of olivomycin by DNA in solution was also studied.
Materials and Methods Drugs. Olivomycin was kindly supplied by Dr. G.H. Gause, Institute of New Antibiotics, Moscow, USSR. Hoechst 33258 was the kind gift of Dr. H. Loewe, Hoechst Laboratories, Frankfurt, West Germany, Netropsin was kindly supplied by the Lederle Laboratories, actinomycin D and quinacrine dihydrochloride were purchased from Sigma.
Base Specific Fluorochromes for C h r o m o s o m e Identification
289
DNAs. The alternating copolymers poly d ( A - 1 7 ) - p o l y d ( A - T ) and poly d ( G - C ) , poly d ( G - C ) were obtained from P.G. Biochemicals. E. coli D N A , M. luteus D N A and C. perfringens D N A were supplied by Sigma, Chromosome Preparations. H u m a n , bovine and porcine chromosomes were prepared from phytohaemagglutinin stimulated lymphocytes following short term leukocyte cultures. Slides were prepared by a modified air drying technique (Moorhead et al., 1960). Mouse chromosome slides were obtained directly from bone marrow preparations. The animal was given an intraperitoneal injection of colchicine (0.01 ml of 0.5% colchicine/gm of body weight) 5 h prior to being sacrificed. The marrow cells were aspirated from the hip bone, and treated with a hypotonic solution of 0.075M KC1 for 20 rain. After fixation of the cells in methanol:acetic acid (3:1 v/v), two drops of the cell suspension were dropped on chilled slides and the slides air dried. H u m a n - m o u s e somatic hybrid cells (mouse tumor cells x normal h u m a n fibroblast) were kindly provided by Dr. C. Tan. The hybrid cells were grown in 250 ml plastic culture flasks in H a m ' s F-10 medium with 15% calf serum. Colcemid was added to a concentration of 1 gg/ml 2.5 h prior to harvesting. The cells were then detached using a 0.25% trypsin solution for 2-3 rain, and treated with a hypotonic solution (0.075 M K C I ) for 14 min before fixation in methanol/acetic acid (3:1 v/v). Slides were prepared as described above. Fluorescent Banding Techniques. The detailed Q-banding technique for chromosome preparations of different species has been reported previously (Lin and Uchida, 1973). The olivomycim R-band procedure has been described elsewhere (rand de Sande et al., 1977a). A modification of the Hoechst 33258 banding procedure according to Jalal et al. (1975) was used, which excluded the aging of slides prior to fluorescent microscopy. Counter Staining Procedures. (a) N e t r o p s i n - o l i v o m y c i n R-bands. The slides were first treated with a 0.5 m M netropsin solution (in 0.01 M Na-phosphate, pH 6.8) for 20 rain, and then rinsed in two changes of distilled water for a total of 3 rain and finally air dried. Subsequently, the slides were stained with an olivomycin solution (0.5 mg/ml in 0.01 M Na-phosphate, pH 6.8) following the reported procedure (van de Sande et al., 1977a). Micro-photography was carried out with a Zeiss fluorescent microscope. The excitation filter is UG1, the barrier filter was set at No. 50. K o d a k Tri X-pan film ( A S A = 4 0 0 ) was used and the exposure time was 5 s using the 100 • Apochromatic objective. (b) Actinomycin D - - H o e c h s t 33258 bands. The slides were first treated with an actinomycin D solution (0.128 m M in 0.01 M Na-phosphate, pH 6.8) for 20 rain, followed by two rinses with distilled water for a total of 2 rain. The slides were then treated with a Hoechst 33258 solution (0.05 ~tg/ml in H a n k s BSS) for 10 rain in the dark, followed by two washes with distilled water for a total of 3 rain. The slides were finally m o u n t e d in 0.1 M Na-phosphate buffer, p H 5.5. For microphotography, the excitation filter used is a BG3 filter, and the barrier filter was set at 47. Kodak high contrast film (ASA64) was used and the exposure time was 30-50 s for the 100 • objective. Solution Interaction of Fluorochromes with DNA. All fluorescence measurements were recorded with a Turner 430 spectrofluorometer thermostated at 20 ~ C. Oiivomycin-DNA interactions were carried out in 0.01 M Na-phosphate (pH 6.8) containing 0.001 M MgCI2 and 0.1 m M EDTA. Olivomycin was present at 5• I0 -6 M, excitation was at 440 nm, emission at 532 nan. Hoechst 33258 fluorescent measurements were carried out in 0.01 M Na-phosphate (pH 6.8) containing 0.15 M NaC1, Hoechst 33258 was present at 1 • 10 6 M, excitation was at 355 nm, emission at 480 nm. The effect of netropsin on olivomycin fluorescence in the presence of D N A was carried out as follows : Relative fluorescence of olivomycin in the presence of M. luteus D N A and C. perJi*ingens D N A was measured at 440 n m (max. excitation) and at 532 n m (max. emission). Concentration of both D N A s used was 1.78 • 10 -5 mole nucleotide phosphorous. Netropsin was finally added at increasing concentrations from 10 v M to 8 x 10-3 M and the relative olivomycin fluorescence was recorded after each addition of netropsin.
290
K.F. Jorgenson et al.
Results
Comparison of "Hoechst "-Bands and Olivomycin R-Bands A comparison was made between the " H o e c h s t " - - a n d olivomycin-bands on metaphase chromosomes from human, mouse and bovine. Chromosomes treated with Hoechst 33258 exhibited characteristic " Q - l i k e " banding patterns on the chromosome arms. The overall fluorescence intensity of the bands on chromosomes produced by Hoechst 33258 appeared to be brighter than those stained by quinacrine in the species examined, but the differentiation of banding patterns along the chromosome arms is less clear than with quinacrine. However, as previously observed, specific chromosomal regions, the paracentromeric area of human chromosomes 1, 9 and 16 (Fig. la) and the centromeric regions of mouse chromosomes (except the Y chromosome, Fig. 1 c) show bright Hoechst fluorescence (Hilwig and Gropp, 1972; Raposa and Natarajan, 1974). All these regions stain negative with quinacrine. Olivomycin produces fluorescent banding patterns on human, mouse and bovine chromosomes which are the reverse to those produced by Hoechst 33258. The paracentric regions of human chromosomes 1, 9 and 16 (Fig. 1 b) and the centromeric regions of mouse chromosomes (Fig. ld) stain negative with olivomycin which is the same as observed for quinacrine, but the opposite to the bright fluorescence produced in these regions by Hoechst 33258. In addition, the Hoechst (and quinacrine) fluorescent negative centromeric regions of bovine chromosomes (fig. 1 e) showed bright olivomycin fluorescence with the exception of the sex chromosomes' centromeres (Fig. 1 f). These observations clearly indicate the complementarity between the Hoechst 33258 and olivomycin bright and dull regions on chromosomes of the three species examined.
Interaction of DNA with Olivomycin and Hoechst 33258 The fluorescence intensity of olivomycin in the presence of several DNAs was measured as is shown in Figure 2. Small spectral shifts are observed in the presence of D N A for both the excitation (425 to 440 nm) and the emission (525 to 532 nm) spectum. The presence of Mg +§ is an absolute requirement for the interaction between olivomycin and DNA. The olivomycin fluorescence enhancement was found to be dependent on the G - C content of the DNA. In contrast, the fluorochrome Hoechst 33258 exhibits an increase in fluorescence depending on the A - T content of the DNA. Again, the two fluorochromes, olivomycin and Hoechst 33258, act in a complementary fashion when interacting with D N A in solution.
Effect of Netropsin on Olivomycin Interaction with DNA Olivomycin fluorescence was shown to be enhanced by D N A depending upon the G - C content of the DNA. When netropsin was added to the olivomycin D N A complex, the extent of enhancement of olivomycin fluorescence due to
Base Specific Fluorochromes for C h r o m o s o m e Identification
291
Fig. la-d. Partial karyotypes and metaphase spreads of human, mouse and bovine. Shown are h u m a n chromosomes Nos. 1, 9 and 16 from (a) a metaphase spread stained with Hoechst 33258 (after counterstained with actinomycin D) and (b) from a metaphase spread stained with olivomycin (after counterstained with netropsin). Paracentric regions of these chromosomes are indicated by bars. c A mouse metaphase spread stained with Hoechst 33258 (Y-chromosome, indicated), d A mouse metaphase spread stained with olivomycin, e A metaphase spread of a lymphocyte from a cow stained with Hoechst 33258 (centromeric regions of three autosomes and the X-chromosomes are indicated), f A metaphase spread from a bull stained with olivomycin (centromeric regions of three autosomes and the sex c h r o m o s o m e s are indicated)
292
K.F. Jorgenson et al.
Olivomycin 5x 10-6M -
9. . . .
o
Hoechst 33258 I x 10-6M
/
M. t u t e u s (72 % G+C}
I
~"~
40
E. c o i l
o (50O/oG+C)
-A ...... 9 C . p e r f r i n g e n s
f ! ,g?
(31~ G+C)
//
30
l~
~7-
O
/
>,5-
20
,i
9"
n- 3 -
10
I_ . ~ A''j I 10-7 10-6 10-5
/o
i-
9"Z 1
/z
e/e
I 10-4
10-8 10-7 10-6 Nucteotide concentration (M)
//
10-5
10-4
Fig. 2. Relative fluorescence of olivomycJn and Hoechst 33258 in the presence of varying concentrations of DNA. The olivomycin (5 x 10 -6 M) D N A binding was carried out in 0.01 M Na phosphate (pH 6.8) containing 10 -3 M MgCI2 and 10 4 M EDTA. The Hoechst 33258 (1 x 10 -6 M) D N A binding was carried out in 0.01 M Na phosphate (pH 6.8) containing 0.15 M NaC1
Olivomycin 5 x 10-6k4 6
18
M. l u t e u s ( 7 2 % G + C ) ) ~ o , , ~ o ~
5
o
= >
3
o+ O
15
-
iii;ii-i,
"~.o,~. "-o~. ~ o - - .
o--.--o
12 "q O + 9 o
~--.~-4= z=.--
% G+C)
O
6~ ,,q
n-
D '
~
I
10-7
10-6
~l[~*~"o~''O~
L
-- ~ 0 . _ _ o _ _ . O ~
I
10-5 10-4 [ Netropsin ]
_ _0_ __ o
I
10-3
"0
I
10-2
Fig. 3. The effect of netropsin on the enhancement of olivomycin fluorescence by M. luteus -9 and C. perfringens - e - e - DNAs. Observed fluorescence intensity is recorded relative to that of the free dye (RF = 1) as increasing a m o u n t s of netropsin are added to equivalent a m o u n t s of the two D N A s which are saturated in 5 x 10 -6 M olivomycin. As the concentration of netropsin is increased the ratio of the extent of fluorescence enhancement (from R F = I) for the two D N A s [i.e. {AF(72% G + C ) / , d F (31% G + C ) } ] is shown to increase - o - n - o -
Base Specific Fluorochromes for Chromosome Identification
293
the presence of D N A was found to decrease with increasing netropsin concentration (Fig. 3). This reduction of the enhanced fluorescence was observed to level off at a ratio of netropsin/olivomycin of approximately 200. Saturation of C. perJ?ingens D N A by netropsin is reached at a lower netropsin concentration than is saturation of M. luteus DNA. Figure 3 shows that although the total decrease in the enhanced fluorescence is similar for both M. luteus (72% G + C) D N A and C. perfringens (31% G + C ) D N A , there is a considerable increase in the ratio of enhanced fluorescence (AF. M. luteus/AF. C. petfringens) in the presence of netropsin as compared to the same ratio in the absence of netropsin. This ratio was found to increase from 3.6 in the absence of netropsin to 17 in the presence of saturating concentrations of this antibiotic. This five-fold increase in the ratio of olivomycin fluorescence enhancement by the two D N A s would indicate that better differentiation between regions along the chromosome of slightly varying G - C content might be observed with olivomycin in the presence of netropsin.
Netropsin-Olivomycin R-Bands Banding patterns on chromosomes of h u m a n metaphase spreads were more distinct on preparations pretreated with netropsin and then followed by olivomycin staining as compared to preparations stained directly with olivomycin (Fig. 4). Although individual chromosomes can be identified by the R-banding patterns in the direct olivomycin stained preparation (Fig. 4a), clear R-bands can only be observed after the counterstaining procedure (Fig. 4b).
Fig. 4 a and b. a A metaphase spread of a human lymphocyte stained with olivomycin only. b A human metaphase spread from a preparation pretreated with netropsin and then stained with olivomycin
294
K.F. Jorgenson et al.
Fig. 5. A fluorescent R-banding karyotype of a boar prepared from a metaphase spread which was stained with netropsin and olivomycin. All centromeric regions of the bi-arm chromosomes (Nos. 1-12) show bright olivomycin fluorescence (except the Y-chromosome) whereas the centromeric regions of the Mocentric chromosomes (Nos. 13-I8) are fluorescence negative
Better contrast in R-bands of porcine chromosomes was also observed by the netropsin-olivomycin procedure. The centromeric region of all the bi-arm chromosomes (except the Y chromosome) exhibit bright olivomycin fluorescence (positive R-band) whereas all centromeric regions of the telocentric chromosomes exhibit negative olivomycin fluorescence (Fig. 5). Actinomycin D - Hoechst 33258 Staining
A significant improvement in the differentiation of fluorescent bands produced by Hoechst 33258 was obtained after a pretreatment with actinomycin D (Fig. 6 b). The Hoechst type bands obtained after counterstaining were in general very similar to the Q-bands, including the variable fluorescent regions on the chromosomes. Close examination of the counterstained Hoechst-banded chromosomes in comparison with Q-bands, shows that the paracentric regions of chromosomes 1, 9 and 16 show bright fluorescence, which is not observed in the Q-bands (Fig. 1 a). Using the counterstaining technique, bands on chromosomes appear more distinctly and in addition the actinomycin D - H o e c h s t fluorescent bands on chromosomes fade much slower than the quinacrine fluorescent bands, thus making photomicrography much easier. A similar effect of actinomycin D counterstaining was found to improve Hoechst 33258 type
Fig. 6a and b. Metaphase spreads of human lymphocytes, a Stained with Hoechst 33258. b Pretreated with actinomycin D and then stained with Hoechst 33258
,~
9
g
o
l::r
g
o
Fig. 7. Metaphase spread of a somatic human-mouse hybrid cell stained with the actinomycin D-Hoechst technique. Mouse chromosomes show bright Hoechst fluorescence in the centromeric regions. Characteristic banding patterns on human chromosomes allow for their individual identification (indicated by numbers)
7~
Base Specific Fluorochromes for Chromosome Identification
297
bands on chromosome arms of mouse and porcine. The centromeric regions of mouse chromosomes (except the Y chromosome) remain Hoechst-fluorescent positive and the centromeric regions of all porcine chromosomes are Hoechstfluorescent negative by this procedure (data not shown).
Chromosome Identification in Somatic Cell Hydrids The actinomycin D - Hoechst technique was also used for chromosome identification in human • mouse somatic hybrid cells. Human chromosomes and mouse chromosomes in a hybrid cell are clearly differentiated since the centromeric regions of mouse chromosomes stain brightly, but the human centromeres stain dull (Fig. 7). In addition, the banding patterns on human chromosomes are well-defined which allow unequivocal identification of specific human chromosomes in the hybrid cells.
Discussion
The benzimidazole derivative Hoechst 33258 has been demonstrated to produce specific banding patterns on metaphase chromosomes similar to those produced by quinacrine (Raposa and Natarajan, 1974). The following observations indicate that this D N A binding fluorochrome may act as an affinity label for A - T basepairs in chromosomal DNA: 1. Hoechst 33258 binds preferentially to A - T rich DNA (Latt and Wohlleb, 1975). 2. Hoechst 33258 exhibits a fluorescence enhancement upon interaction with DNA. The extent of fluorescence enhancement can be directly related to the A - T content of the D N A (Weisblum and Haenssler, 1974). 3. Mouse centromeric regions, the location of A - T rich satellite DNA, stain brightly with Hoechst 33258 (Pardue and Gall, 1970; Jones, 1970; Hilwig and Gropp, 1972). The paracentric regions on the long arms of human chromosomes Nos. 1, 9 and 16, which reportedly contain A - T rich satellite II D N A (Jones and Corneo, 1971; Schreck et al., 1974) also stain brightly with Hoechst 33258 (Raposa and Natarajan, 1974). 4. In addition, several other A - T specific fluorochromes, 2,7-di-t-butylproflavine, DAPI and DIPI, have been shown to produce Hoechst 33258-1ike banding patterns on chromosomes (Dist6che and Bontemps, 1974; Schweizer, 1976; Schnedl et al., 1977 a). A second group of fluorochromes, the chromomycinone antibiotics olivomycin, chromomycin A3 and mithramycin were found to produce characteristic reverse bands (R-bands) on chromosomes (Schweizer, 1976; van de Sande et al., 1977a). Several observations suggest that this group of D N A binding antibiotics acts as affinity labels for G - C base pairs in chromosomal DNA: 1. Chromomycin A 3 inhibits D N A dependent RNA synthesis when the D N A contains G - C basepairs (Ward et al., 1965).
298
K.F. Jorgenson et al.
2. The number of binding sites on D N A for the chromomycinone dyes increases with the G - C content of the D N A up to 66% G - C (Behr et al., 1969). 3. Olivomycin fluorescence is enhanced upon interaction with D N A which can be directly related to the G - C content of the DNA (van de Sande et al., 1977a). 4. The centromeric region of bovine chromosomes, which contain G - C rich satellite DNA, stain brightly with olivomycin (Kurnit et al., 1973; van de Sande et al., 1977a). The non-fluorescent G - C specific antibiotic, actinomycin-D, enhances the contrast in the fluorescent bands on chromosome preparations produced by the A - T specific fluorochrome Hoechst 33258. The use of actinomycin-D as a counterstaining agent to enhance the resolution of fluorescent bands produced by the A - T specific fluorochrome DAPI has been previously reported (Schweizer, 1976). Similarly, the non-fluorescent A - T specific D N A binding drug, netropsin, serves as an effective counterstaining agent thus increasing the resolution of chromosome bands produced by the G - C specific fluorochrome olivomycin. Netropsin also increases the differential fluorescence enhancement of olivomycin interacting with DNAs of very different base composition in solution. Consequently, the mechanism responsible for the effect of netropsin on olivomycin fluorescence in solution in the presence of DNA, could also account for the counterstaining effect of netropsin on olivomycin stained metaphase chromosomes. The compatibility of binding of different base pair specific ligands to DNA has been previously reported (Wartell et al., 1975). It was shown that the A - T specific ligand netropsin and the G - C specific drug actinomycin D can bind in close proximity. However, the binding of actinomycin D to D N A is decreased in the presence of netropsin. This effect is dependent upon the G - C content of DNA. A similar finding is reported here for the effect of netropsin on olivomycin binding to D N A as expressed by decreases in olivomycin fluorescence enhancement. The antibiotic netropsin covers three base pairs when bound to D N A (Wartell et al., 1974) and thus could eliminate potential olivomycin binding sites. This effect would be most pronounced for G - - C sites interspersed in an A - T rich region. Prior binding of netropsin in such an A - T rich region could mask olivomycin binding sites resulting in a reduction in of the " w e a k " olivomycin fluorescence present in the absence of the counterstaining agent (Fig. 8). The overall effect of the pretreatment with netropsin would be the accentuation of differential olivomycin fluorescence between A - T and G - C rich regions. This postulate would explain the observed increase in definition between bright and dull regions in counterstained metaphase chromosomes. A different counterstaining procedure was recently reported by Schweizer et al. (1978). The non-fluorescent A - T specific antibiotic distamycin A was found to reduce the overall fluorescence intensity of chromosomes banded with DAPI. The remaining bright regions in the chromosomes after this sequential staining procedure contain repetitive DNA, mainly A - T rich satellites. The molecular explanation of this finding remains open. A difference in the binding
Base Specific Fluorochromes for Chromosome Identification
Olivomycin-Netropsin
WN
A
g'N
k./
k_/
k,J
299
DNA complex
B I ill[llll[llll(ll)ll(ll)ll(ll)lllll)ll(ll)ll(l")llllllllllllllllll~ll~l k,#
C I
I
I
I II
~
~
I1{11)11{11)11(11)11 IIIlil
I ' ~ - O[ivomycin( sG+C pecific)
I
I
kJ
II
kJ
II
0 - Netropsin ( A+T specific)
Fig. 8. Model for the effect of netropsin binding to DNA on subsequent binding by olivomycin. The olivomycin binding sites are shown in A, the netropsin binding sites in B. The exclusion of olivomycin binding following prior binding by netropsin is indicated in C
of distamycin A to heterochromatic or euchromatic regions has been suggested (Schweizer et al., 1978). Since the effect of distamycin A as a D A P I counterstaining agent is concentration dependent, an alternate explanation may be that under the conditions used, all D A P I binding sites on chromosome arms are blocked by distamycin A, but there are still D A P I binding sites available in some specific constitutive heterochromatic regions. The observations that A - T and G - C specific fluorochromes produce banding patterns on chromosomes which are complementary to one another indicate that sequence arrangement in chromosomal D N A is a primary determinant in the production of fluorescent bands on metaphase chromosomes. The finding that counterstaining with non-fluorescent base specific D N A binding ligands of complementary specificity enhances the resolution of fluorescent bands on chromosomes produced by both A - T and G - C specific fluorochromes further supports this hypothesis. Previously, the observation was made that the resolution of Hoechst 33258 fluorescent bands on metaphase chromosomes improves with aging of the stained c h r o m o s o m e preparation (Jalal et al., 1975). The use of actinomycin D and netropsin as counterstaining agents for Hoechst 33258 and olivomycin respectively, provides a simple, rapid means by which an even stronger enhancement in the differentiation of fluorescent bands can be achieved. With Hoechst 33258, the counterstaining procedure results in the production of clearly defined banding patterns which are more stable than those produced by staining with quinacrine. This makes the actinomycin D - H o e c h s t 33258 procedure a useful tool
300
K.F. Jorgenson et al.
for routine chromosome analysis. In addition to the increased stability of the fluorescent bands, this procedure also demonstrates the existence of polymorphisms in the secondary constriction regions of human chromosomes 1, 9 and 16 as previously reported (Raposa and Natarajan, 1974). The variations which occur in these regions in normal polulations are not readily visualized by the Q-banding technique and can generally be identified only by the C-banding technique. The actinomycin D - H o e c h s t 33258 staining has also proved useful in the identification of individual chromosomes in somatic cell hybrids. Unequivocal identification of chromosomes in somatic cell hybrids complement is essential if this technique is to be used for gene mapping on chromosomes (McKusick and Ruddle, 1977). Hoechst 33258 fluorescent staining has previously been used to discriminate between human and mouse chromosomes in man-mouse somatic cell hybrids, by the differential staining intensity in the centromeric regions of the "parental chromosomes" (Lin et al., 1974). However, these studies required that the hybrid cells undergo one cycle of D N A replication in the presence of BUdR in order to enhance the difference in the Hoechst 33258 stained human and mouse centromeres. The present study shows that the A - T specific fluorochrome Hoechst 33258 after counterstaining with actinomycin D can be directly used in the unequivocal identification of parental chromosomes in mail-mouse somatic cell hybrids without prior treatment of the somatic cells with BUdR. The differential staining of centromeric heterochromatin in some mammalian species appears to be the result of highly repetitive satellite DNAs located in these regions. In the mouse essentially one kind of A - T rich satellite D N A is distributed to almost every centromeric region of the mouse complement (except the Y-chromosome), whereas G - C rich satellite DNAs are located in the centromeric regions of the bovine chromosome complement (except the sex chromosomes). These two cases represent an extreme contrast in terms of the base composition of their centromeric heterochromatins. The distribution of centromeric heterochromatin in the porcine chromosome complement appears different than in either mouse or cattle. In this species, only the bi-arm chromosomes stain olivomycin bright and contain G - C rich centromeric heterochromatin. Whether or not this G - C rich centromeric heterochromatin represents a repetitive satellite D N A of the procine genome remains unclear at this time. These studies indicate that three different responses can be obtained in the staining of centromeric heterochromatin with the base specific fluorochromes Hoechst 33258 and olivomycin. The centromeres which show Hoechst 33258 positive, olivomycin negative fluorescence contain A - T rich sequences, while the centromeres which show olivomycin positive, Hoechst 33258 negative fluorescence represent G - C rich sequences. The sequence arrangement in the centromeric regions which stain negative with both Hoechst 33258 and olivomycin, e.g. the centromeric regions of porcine telocentric chromosomes and human centromeric regions, remains a question at this time. The observations that the base pair specific fluorochromes of opposing specificity act in a complementary fashion both with D N A in solution and with metaphase chromosomes, and that the bands produced on cytological prepara-
Base Specific Fluorochromes for Chromosome Identification tions can be effectively counterstained using nonfluorescent DNA
301 binding agents
of opposite specificity, strongly support the hypothesis that base sequence arr a n g e m e n t w i t h i n c h r o m o s o m e s is a p r i m a r y d e t e r m i n a n t i n t h e p r o d u c t i o n o f f l u o r e s c e n t b a n d s . T h e e n h a n c e d d e f i n i t i o n in f l u o r e s c e n t b a n d s , a t t a i n e d using counterstaining, aids both chromosome identification and the investigation of certain heterochromatic regions.
Acknowledgements.We would like to thank Mrs. Heidi K. Jamro for her excellent technical assistance. This work was supported by the Medical Research Council of Canada grant Nos. MA-4655 and MA-4885 and The National Foundation-March of Dimes grant No. 1-544.
References Behr, W., Honikel, K., Hartmann, G. : Interaction of the RNA polymerase inhibitor chromomycin with DNA. Europ. J. Biochem. 9, 82-92 (1969) Distbche, Bontemps, J. : Chromosome regions containing DNAs of known base composition, specifically evidenced by 2,7-di-t-butyl proflavine. Chromosoma (Berl.) 47, 263 281 (1974) Hilwig, I., Gropp, A.: Stains of constitutive heterochromatin in mammalian chromosome with a new fluorochrome. Exp. Cell Res. 75, 122-126 (1972) Jalal, S.M., Markvong, A., Hsu, T.C.: Differential chromosomal fluorescence with 33258 Hoechst. Exp. Cell Res. 90, 443-444 (1975) Jones, K.W. : Chromosomal and nuclear location of mouse satellite DNA in individual cells. Nature (Lond.) 225, 912 915 (1970) Jones, K.W., Corneo, G.: Location of satellite and homogeneous DNA sequences on human chromosomes. Nature (Lond.) New Biol. 233, 268 271 (1971) Kurnit, D.M., Shafit, B.R., Maio, J.J.: Multiple satellite deoxyribonucleic acids in the calf and their relation to the sex chromosomes. J. molec. Biol. 81, 274 284 (1973) Latt, S.A., Stetten, G.: Spectral studies on 33258 Hoechst and related bisbenzimidazole dyes useful for fluorescent detection of deoxyribonucleic acid synthesis. J. Histochem. Cytochem. 24, 24-33 (1976) Latt, S.A., Wohlleb, J. : Optical studies of the interaction of 33258 Hoechst with DNA, chromatin and metaphase chromosomes. Chromosoma (Berl.) 52, 297 316 (1975) Lin, C.C., Uchida, I.A. : Fluorescent banding of chromosomes (Q-bands). In: Methods and applications of tissue culture (P.F. Kruse and M,K. Patterson, eds.), pp. 778 781. New York, N.Y.: Academic Press, 1973 Lin, M.S., Latt, S.A., Davidson, R.L. : Identification of human and mouse chromosomes in humanmouse hybrids by centromere fluorescence. Exp. Cell Res. 87, 429-433 (1974) McKusick, V.A., Ruddle, F.H.: The status of the gene map of human chromosomes. Science 196, 390-405 (1977) Moorhead, P.S., Nowell, P.C., Mellman, W.J., Batipps, D.M., Hungerford, D.A.: Chromosome preparations of leukocytes cultured from human peripheral blood. Exp. Cell Res. 20, 613 616 (1960) Mfiller, W., Crothers, D.M.: Studies of the binding of actinomycin and related compounds to DNA. J. molec. Biol. 35, 251 290 (1968) Pardue, M.L., Gall, J. : Chromosomal localization of mouse satellite DNA. Science 168, 1356-1358 (1970) Raposa, T., Natarajan, A.T.: Fluorescence banding pattern of human and mouse chromosomes with a benzimidazole derivative (Hoechst 33258). Humangenetik 21, 221 226 (1974) Reich, E., Goldberg, I.H.: Actinomycin and nucleic acid function. Progr. Nucleic Acid Res. molec. Biol. 3, 183 234 (1964) Sande, J.H., van de, Lin, C.C., Jorgenson, K.F.: Reverse banding on chromosomes produced by a guanosine-cytosine specific DNA binding antibiotic: olivomycin. Science 195, 400~402 (1977a)
302
K.F. Jorgenson et al.
Sande, J.H. van de, Lin, C.C., Johnston, F.P., Jorgenson, K.F.: Differential fluorescent labelling of chromosomes and DNA with base pair specific DNA binding antibiotics. In : Human cytogenetics: ICN-UCLA Symp. molec, cell. Biol. Vol. VII, pp. 127-138. New York, N.Y. : Academic Press Schnedl, W., Breitenbach, M., Mikelsaar, A.-V., Stranziner, G.: Mithramycin and DIPI: A pair of fluorochromes specific for G - C and A - T rich DNA respectively. Hum. Genet. 36, 299-305 (1977b) Schnedl, W., Mikelsaar, A.-V., Breitenbach, M., Dann, O. : DIPI and DAPI: fluorescence banding with only negligible fading. Hum. Genet. 36, 167-172 (1977a) Schreck, R.R., Erlanger, B.F., Miller, O.J.: The use of antinucteoside antibodies to probe the organization of chromosomes denatured by ultraviolet irradiation. Exp. Cell Res. 88, 31 39 (1974) Schweizer, D.: Reverse fluorescent chromosome banding with chromomycin and DAPI. Chromosoma (Berl.) 58, 307-324 (1976) Schweizer, D., Ambros, P., Andrle, M.: Modification of DAPI banding on human chromosomes by prestaining with a DNA-binding oligopeptide antibiotic, Distamycin A. Exp. Cell Res. 111, 327-332 (1978) Ward, D.C., Reich, E., Goldberg, I.H.: Base specificity in the interaction of polynucleotides with antibiotic drugs. Science 149, 1259 1263 (1965) Wartell, R.M., Larson, J.E., Wells, R.D. : Netropsin, a specific probe for A - T regions of duplex deoxyribonucleic acid. J. biol. Chem. 249, 6719-6731 (1974) Wartell, R.M., Larson, J.E., Wells, R.D.: The compatibility of netropsin and actinomycin binding to nature deoxyribonucleic acid. J. molec. Biol. 250, 2698 2702 (1975) Weisblum, B., Haenssler, E.: Fluorometric properties of the bibenzimidazole derivative Hoechst 33258, a fluorescent probe specific for A - T concentration in chromosomal DNA. Chromosoma (Berl.) 46, 255-260 (1974) Wells, R.D., Larson, J.E.: Studies on the binding of actinomycin D to DNA and DNA model polymers. J. molec. Biol. 49, 319 342 (1970)
Received May 9 - J u n e 13, 1978 / Accepted June 13, I978 by W. Beermann Ready for press June 18, 1978
CHROMOSOMA 9 by Springer-Verlag 1978
The Use of Base Pair Specific DNA Binding Agents as Affinity Labels for the Study of Mammalian Chromosomes K.F. Jorgenson 3, J.H. van de Sande 1 and C.C. Lin 1' 2 Divisions of 1Medical Biochemistryand ZPediatrics, Faculty of Medicine and 3Department of Chemistry, The University of Calgary, Calgary, Alberta, Canada T2N 1N4
Abstract. The fluorochromes Hoechst 33258 and olivomycin are base pair specific D N A binding agents. The fluorescence enhancement of Hoechst 33258 and olivomycin in the presence of D N A can be directly related to the A - T and G - C content of the interacting D N A respectively. Cytological observations of metaphase chromosomes treated with these two compounds suggest that the fluorescent banding patterns produced are the reverse of one another.--Non-fluorescent base pair specific D N A binding agents have been used as counterstains in chromosome preparations to enhance the contrast of the banding patterns produced by the base specific fluorochromes. The non-fluorescent G - C specific antibiotic actinomycin-D enhanced the resolution of fluorescent bands produced by the A - T specific fluorochrome Hoechst 33258. Similarly the non-fluorescent A - T specific antibiotic netropsin was found to enhance resolution of the bands produced by the G - C specific fluorochrome olivomycin. Netropsin was also found to increase the differential fluorescent enhancement of complexes of olivomycin with DNAs of various base composition in solution. These findings suggest that counterstaining agents act through a base sequence dependent inhibition of subsequent binding by base pair specific f l u o r o c h r o m e s . - T h e base specific D N A binding agents have been used to differentiate different types of constitutive heterochromatin in mammalian species, and to facilitate chromosome identification in somatic cell hybrids. Introduction
Two groups of D N A binding fluorochromes which have the capacity to discriminate in their binding between base pairs are currently being used for chromosome identification. The first group consists of the A - T specific fluorochromes Hoechst 33258 (2-[2-(4-hydroxyphenyl)-6-benzimidazolyl]-6-(1-methyl-4-piperazyl)-benzimidazole), DBP (2,7-di-t-butyl proflavine), DAPI (4'-6-diamidino-2phenylindole) and DIPI (4'-6-bis(2'imidazolinyl-4'-5'-H)-2-phenylindole) (Hilwig and Gropp, 1972; Disteche and Bontemps, 1974; Schweizer, 1976; Schnedl
0009-5915/78/0068[0287/$03.20
288
K.F. Jorgenson et al.
et al., 1977a). Solution interaction studies of Hoechst 33258 in the presence of DNA have shown that binding of the fluorochrome to DNA is accompanied by a fluorescence enhancement which can be directly related to the A - T content of the DNA (Weisblum and Haenssler, 1974; Latt and Stetten, 1976). The same "Hoechst" type chromosome banding patterns produced by this group of fluorochromes can then be rationalized in terms of the distribution of A - T base pairs along the arms of the chromosomes. The second group of fluorochromes, consisting of olivomycin, chromomycin A3 and mithramycin, was recently found to produce characteristic R-bands on both plant and mammalian chromosomes (Schweizer, 1976; van de Sande et al., 1977a; Schnedl et al., 1977b). This group of fluorochromes exhibits enhanced fluorescence when bound to DNA which can be directly related to the G - C content of the DNA (van de Sande et al., 1977b). The R-banding patterns on chromosome arms produced by these fluorochromes appear to be a reversal of the banding patterns produced by the A - T specific fluorochromes (van de Sande et al., 1977b). The R-bands produced by this group of fluorochromes are assumed to be a reflection of their binding to G - C rich sequences along the chromosomes. Differential staining of chromosome regions by these groups of fluorochromes facilitates the unequivocal identification of individual chromosomes of a species. In addition, these compounds should also facilitate the identification of certain heterochromatic regions of mammalian chromosomes since it is known that in several species these regions contain mainly A - T or G - C rich DNA (Pardue and Gall, 1970; Jones, 1970; John and Corneo, 1971; Kurnit et al., 1973; Schreck et al., 1974; van de Sande et al., 1977b). However, direct application of all these fluorochromes to chromosome preparations has been found to produce rather poorly differentiated banding patterns (Jalal et al., 1975; Schweizer, 1976). This paper reports comparisons of both the specific banding patterns produced by the two groups of fluorochromes and the observed changes in their fluorescence when interacting with synthetic and natural DNAs of varying base composition. In addition, we report the use of the non-fluorescent base specific binding ligands netropsin and actinomycin D as counterstaining agents on chromosome preparations to improve the resolution of olivomycin R-bands and Hoechst bands respectively. Actinomycin D specifically binds to G - C sites on duplex DNA (Reich and Goldberg, 1964; Miiller and Crothers, 1968; Wells and Larsen, 1970), while netropsin binds specifically to A - T sites (Wartell et al., 1974). The effect of netropsin on the fluorescence modification of olivomycin by DNA in solution was also studied.
Materials and Methods Drugs. Olivomycin was kindly supplied by Dr. G.H. Gause, Institute of New Antibiotics, Moscow, USSR. Hoechst 33258 was the kind gift of Dr. H. Loewe, Hoechst Laboratories, Frankfurt, West Germany, Netropsin was kindly supplied by the Lederle Laboratories, actinomycin D and quinacrine dihydrochloride were purchased from Sigma.
Base Specific Fluorochromes for C h r o m o s o m e Identification
289
DNAs. The alternating copolymers poly d ( A - 1 7 ) - p o l y d ( A - T ) and poly d ( G - C ) , poly d ( G - C ) were obtained from P.G. Biochemicals. E. coli D N A , M. luteus D N A and C. perfringens D N A were supplied by Sigma, Chromosome Preparations. H u m a n , bovine and porcine chromosomes were prepared from phytohaemagglutinin stimulated lymphocytes following short term leukocyte cultures. Slides were prepared by a modified air drying technique (Moorhead et al., 1960). Mouse chromosome slides were obtained directly from bone marrow preparations. The animal was given an intraperitoneal injection of colchicine (0.01 ml of 0.5% colchicine/gm of body weight) 5 h prior to being sacrificed. The marrow cells were aspirated from the hip bone, and treated with a hypotonic solution of 0.075M KC1 for 20 rain. After fixation of the cells in methanol:acetic acid (3:1 v/v), two drops of the cell suspension were dropped on chilled slides and the slides air dried. H u m a n - m o u s e somatic hybrid cells (mouse tumor cells x normal h u m a n fibroblast) were kindly provided by Dr. C. Tan. The hybrid cells were grown in 250 ml plastic culture flasks in H a m ' s F-10 medium with 15% calf serum. Colcemid was added to a concentration of 1 gg/ml 2.5 h prior to harvesting. The cells were then detached using a 0.25% trypsin solution for 2-3 rain, and treated with a hypotonic solution (0.075 M K C I ) for 14 min before fixation in methanol/acetic acid (3:1 v/v). Slides were prepared as described above. Fluorescent Banding Techniques. The detailed Q-banding technique for chromosome preparations of different species has been reported previously (Lin and Uchida, 1973). The olivomycim R-band procedure has been described elsewhere (rand de Sande et al., 1977a). A modification of the Hoechst 33258 banding procedure according to Jalal et al. (1975) was used, which excluded the aging of slides prior to fluorescent microscopy. Counter Staining Procedures. (a) N e t r o p s i n - o l i v o m y c i n R-bands. The slides were first treated with a 0.5 m M netropsin solution (in 0.01 M Na-phosphate, pH 6.8) for 20 rain, and then rinsed in two changes of distilled water for a total of 3 rain and finally air dried. Subsequently, the slides were stained with an olivomycin solution (0.5 mg/ml in 0.01 M Na-phosphate, pH 6.8) following the reported procedure (van de Sande et al., 1977a). Micro-photography was carried out with a Zeiss fluorescent microscope. The excitation filter is UG1, the barrier filter was set at No. 50. K o d a k Tri X-pan film ( A S A = 4 0 0 ) was used and the exposure time was 5 s using the 100 • Apochromatic objective. (b) Actinomycin D - - H o e c h s t 33258 bands. The slides were first treated with an actinomycin D solution (0.128 m M in 0.01 M Na-phosphate, pH 6.8) for 20 rain, followed by two rinses with distilled water for a total of 2 rain. The slides were then treated with a Hoechst 33258 solution (0.05 ~tg/ml in H a n k s BSS) for 10 rain in the dark, followed by two washes with distilled water for a total of 3 rain. The slides were finally m o u n t e d in 0.1 M Na-phosphate buffer, p H 5.5. For microphotography, the excitation filter used is a BG3 filter, and the barrier filter was set at 47. Kodak high contrast film (ASA64) was used and the exposure time was 30-50 s for the 100 • objective. Solution Interaction of Fluorochromes with DNA. All fluorescence measurements were recorded with a Turner 430 spectrofluorometer thermostated at 20 ~ C. Oiivomycin-DNA interactions were carried out in 0.01 M Na-phosphate (pH 6.8) containing 0.001 M MgCI2 and 0.1 m M EDTA. Olivomycin was present at 5• I0 -6 M, excitation was at 440 nm, emission at 532 nan. Hoechst 33258 fluorescent measurements were carried out in 0.01 M Na-phosphate (pH 6.8) containing 0.15 M NaC1, Hoechst 33258 was present at 1 • 10 6 M, excitation was at 355 nm, emission at 480 nm. The effect of netropsin on olivomycin fluorescence in the presence of D N A was carried out as follows : Relative fluorescence of olivomycin in the presence of M. luteus D N A and C. perJi*ingens D N A was measured at 440 n m (max. excitation) and at 532 n m (max. emission). Concentration of both D N A s used was 1.78 • 10 -5 mole nucleotide phosphorous. Netropsin was finally added at increasing concentrations from 10 v M to 8 x 10-3 M and the relative olivomycin fluorescence was recorded after each addition of netropsin.
290
K.F. Jorgenson et al.
Results
Comparison of "Hoechst "-Bands and Olivomycin R-Bands A comparison was made between the " H o e c h s t " - - a n d olivomycin-bands on metaphase chromosomes from human, mouse and bovine. Chromosomes treated with Hoechst 33258 exhibited characteristic " Q - l i k e " banding patterns on the chromosome arms. The overall fluorescence intensity of the bands on chromosomes produced by Hoechst 33258 appeared to be brighter than those stained by quinacrine in the species examined, but the differentiation of banding patterns along the chromosome arms is less clear than with quinacrine. However, as previously observed, specific chromosomal regions, the paracentromeric area of human chromosomes 1, 9 and 16 (Fig. la) and the centromeric regions of mouse chromosomes (except the Y chromosome, Fig. 1 c) show bright Hoechst fluorescence (Hilwig and Gropp, 1972; Raposa and Natarajan, 1974). All these regions stain negative with quinacrine. Olivomycin produces fluorescent banding patterns on human, mouse and bovine chromosomes which are the reverse to those produced by Hoechst 33258. The paracentric regions of human chromosomes 1, 9 and 16 (Fig. 1 b) and the centromeric regions of mouse chromosomes (Fig. ld) stain negative with olivomycin which is the same as observed for quinacrine, but the opposite to the bright fluorescence produced in these regions by Hoechst 33258. In addition, the Hoechst (and quinacrine) fluorescent negative centromeric regions of bovine chromosomes (fig. 1 e) showed bright olivomycin fluorescence with the exception of the sex chromosomes' centromeres (Fig. 1 f). These observations clearly indicate the complementarity between the Hoechst 33258 and olivomycin bright and dull regions on chromosomes of the three species examined.
Interaction of DNA with Olivomycin and Hoechst 33258 The fluorescence intensity of olivomycin in the presence of several DNAs was measured as is shown in Figure 2. Small spectral shifts are observed in the presence of D N A for both the excitation (425 to 440 nm) and the emission (525 to 532 nm) spectum. The presence of Mg +§ is an absolute requirement for the interaction between olivomycin and DNA. The olivomycin fluorescence enhancement was found to be dependent on the G - C content of the DNA. In contrast, the fluorochrome Hoechst 33258 exhibits an increase in fluorescence depending on the A - T content of the DNA. Again, the two fluorochromes, olivomycin and Hoechst 33258, act in a complementary fashion when interacting with D N A in solution.
Effect of Netropsin on Olivomycin Interaction with DNA Olivomycin fluorescence was shown to be enhanced by D N A depending upon the G - C content of the DNA. When netropsin was added to the olivomycin D N A complex, the extent of enhancement of olivomycin fluorescence due to
Base Specific Fluorochromes for C h r o m o s o m e Identification
291
Fig. la-d. Partial karyotypes and metaphase spreads of human, mouse and bovine. Shown are h u m a n chromosomes Nos. 1, 9 and 16 from (a) a metaphase spread stained with Hoechst 33258 (after counterstained with actinomycin D) and (b) from a metaphase spread stained with olivomycin (after counterstained with netropsin). Paracentric regions of these chromosomes are indicated by bars. c A mouse metaphase spread stained with Hoechst 33258 (Y-chromosome, indicated), d A mouse metaphase spread stained with olivomycin, e A metaphase spread of a lymphocyte from a cow stained with Hoechst 33258 (centromeric regions of three autosomes and the X-chromosomes are indicated), f A metaphase spread from a bull stained with olivomycin (centromeric regions of three autosomes and the sex c h r o m o s o m e s are indicated)
292
K.F. Jorgenson et al.
Olivomycin 5x 10-6M -
9. . . .
o
Hoechst 33258 I x 10-6M
/
M. t u t e u s (72 % G+C}
I
~"~
40
E. c o i l
o (50O/oG+C)
-A ...... 9 C . p e r f r i n g e n s
f ! ,g?
(31~ G+C)
//
30
l~
~7-
O
/
>,5-
20
,i
9"
n- 3 -
10
I_ . ~ A''j I 10-7 10-6 10-5
/o
i-
9"Z 1
/z
e/e
I 10-4
10-8 10-7 10-6 Nucteotide concentration (M)
//
10-5
10-4
Fig. 2. Relative fluorescence of olivomycJn and Hoechst 33258 in the presence of varying concentrations of DNA. The olivomycin (5 x 10 -6 M) D N A binding was carried out in 0.01 M Na phosphate (pH 6.8) containing 10 -3 M MgCI2 and 10 4 M EDTA. The Hoechst 33258 (1 x 10 -6 M) D N A binding was carried out in 0.01 M Na phosphate (pH 6.8) containing 0.15 M NaC1
Olivomycin 5 x 10-6k4 6
18
M. l u t e u s ( 7 2 % G + C ) ) ~ o , , ~ o ~
5
o
= >
3
o+ O
15
-
iii;ii-i,
"~.o,~. "-o~. ~ o - - .
o--.--o
12 "q O + 9 o
~--.~-4= z=.--
% G+C)
O
6~ ,,q
n-
D '
~
I
10-7
10-6
~l[~*~"o~''O~
L
-- ~ 0 . _ _ o _ _ . O ~
I
10-5 10-4 [ Netropsin ]
_ _0_ __ o
I
10-3
"0
I
10-2
Fig. 3. The effect of netropsin on the enhancement of olivomycin fluorescence by M. luteus -9 and C. perfringens - e - e - DNAs. Observed fluorescence intensity is recorded relative to that of the free dye (RF = 1) as increasing a m o u n t s of netropsin are added to equivalent a m o u n t s of the two D N A s which are saturated in 5 x 10 -6 M olivomycin. As the concentration of netropsin is increased the ratio of the extent of fluorescence enhancement (from R F = I) for the two D N A s [i.e. {AF(72% G + C ) / , d F (31% G + C ) } ] is shown to increase - o - n - o -
Base Specific Fluorochromes for Chromosome Identification
293
the presence of D N A was found to decrease with increasing netropsin concentration (Fig. 3). This reduction of the enhanced fluorescence was observed to level off at a ratio of netropsin/olivomycin of approximately 200. Saturation of C. perJ?ingens D N A by netropsin is reached at a lower netropsin concentration than is saturation of M. luteus DNA. Figure 3 shows that although the total decrease in the enhanced fluorescence is similar for both M. luteus (72% G + C) D N A and C. perfringens (31% G + C ) D N A , there is a considerable increase in the ratio of enhanced fluorescence (AF. M. luteus/AF. C. petfringens) in the presence of netropsin as compared to the same ratio in the absence of netropsin. This ratio was found to increase from 3.6 in the absence of netropsin to 17 in the presence of saturating concentrations of this antibiotic. This five-fold increase in the ratio of olivomycin fluorescence enhancement by the two D N A s would indicate that better differentiation between regions along the chromosome of slightly varying G - C content might be observed with olivomycin in the presence of netropsin.
Netropsin-Olivomycin R-Bands Banding patterns on chromosomes of h u m a n metaphase spreads were more distinct on preparations pretreated with netropsin and then followed by olivomycin staining as compared to preparations stained directly with olivomycin (Fig. 4). Although individual chromosomes can be identified by the R-banding patterns in the direct olivomycin stained preparation (Fig. 4a), clear R-bands can only be observed after the counterstaining procedure (Fig. 4b).
Fig. 4 a and b. a A metaphase spread of a human lymphocyte stained with olivomycin only. b A human metaphase spread from a preparation pretreated with netropsin and then stained with olivomycin
294
K.F. Jorgenson et al.
Fig. 5. A fluorescent R-banding karyotype of a boar prepared from a metaphase spread which was stained with netropsin and olivomycin. All centromeric regions of the bi-arm chromosomes (Nos. 1-12) show bright olivomycin fluorescence (except the Y-chromosome) whereas the centromeric regions of the Mocentric chromosomes (Nos. 13-I8) are fluorescence negative
Better contrast in R-bands of porcine chromosomes was also observed by the netropsin-olivomycin procedure. The centromeric region of all the bi-arm chromosomes (except the Y chromosome) exhibit bright olivomycin fluorescence (positive R-band) whereas all centromeric regions of the telocentric chromosomes exhibit negative olivomycin fluorescence (Fig. 5). Actinomycin D - Hoechst 33258 Staining
A significant improvement in the differentiation of fluorescent bands produced by Hoechst 33258 was obtained after a pretreatment with actinomycin D (Fig. 6 b). The Hoechst type bands obtained after counterstaining were in general very similar to the Q-bands, including the variable fluorescent regions on the chromosomes. Close examination of the counterstained Hoechst-banded chromosomes in comparison with Q-bands, shows that the paracentric regions of chromosomes 1, 9 and 16 show bright fluorescence, which is not observed in the Q-bands (Fig. 1 a). Using the counterstaining technique, bands on chromosomes appear more distinctly and in addition the actinomycin D - H o e c h s t fluorescent bands on chromosomes fade much slower than the quinacrine fluorescent bands, thus making photomicrography much easier. A similar effect of actinomycin D counterstaining was found to improve Hoechst 33258 type
Fig. 6a and b. Metaphase spreads of human lymphocytes, a Stained with Hoechst 33258. b Pretreated with actinomycin D and then stained with Hoechst 33258
,~
9
g
o
l::r
g
o
Fig. 7. Metaphase spread of a somatic human-mouse hybrid cell stained with the actinomycin D-Hoechst technique. Mouse chromosomes show bright Hoechst fluorescence in the centromeric regions. Characteristic banding patterns on human chromosomes allow for their individual identification (indicated by numbers)
7~
Base Specific Fluorochromes for Chromosome Identification
297
bands on chromosome arms of mouse and porcine. The centromeric regions of mouse chromosomes (except the Y chromosome) remain Hoechst-fluorescent positive and the centromeric regions of all porcine chromosomes are Hoechstfluorescent negative by this procedure (data not shown).
Chromosome Identification in Somatic Cell Hydrids The actinomycin D - Hoechst technique was also used for chromosome identification in human • mouse somatic hybrid cells. Human chromosomes and mouse chromosomes in a hybrid cell are clearly differentiated since the centromeric regions of mouse chromosomes stain brightly, but the human centromeres stain dull (Fig. 7). In addition, the banding patterns on human chromosomes are well-defined which allow unequivocal identification of specific human chromosomes in the hybrid cells.
Discussion
The benzimidazole derivative Hoechst 33258 has been demonstrated to produce specific banding patterns on metaphase chromosomes similar to those produced by quinacrine (Raposa and Natarajan, 1974). The following observations indicate that this D N A binding fluorochrome may act as an affinity label for A - T basepairs in chromosomal DNA: 1. Hoechst 33258 binds preferentially to A - T rich DNA (Latt and Wohlleb, 1975). 2. Hoechst 33258 exhibits a fluorescence enhancement upon interaction with DNA. The extent of fluorescence enhancement can be directly related to the A - T content of the D N A (Weisblum and Haenssler, 1974). 3. Mouse centromeric regions, the location of A - T rich satellite DNA, stain brightly with Hoechst 33258 (Pardue and Gall, 1970; Jones, 1970; Hilwig and Gropp, 1972). The paracentric regions on the long arms of human chromosomes Nos. 1, 9 and 16, which reportedly contain A - T rich satellite II D N A (Jones and Corneo, 1971; Schreck et al., 1974) also stain brightly with Hoechst 33258 (Raposa and Natarajan, 1974). 4. In addition, several other A - T specific fluorochromes, 2,7-di-t-butylproflavine, DAPI and DIPI, have been shown to produce Hoechst 33258-1ike banding patterns on chromosomes (Dist6che and Bontemps, 1974; Schweizer, 1976; Schnedl et al., 1977 a). A second group of fluorochromes, the chromomycinone antibiotics olivomycin, chromomycin A3 and mithramycin were found to produce characteristic reverse bands (R-bands) on chromosomes (Schweizer, 1976; van de Sande et al., 1977a). Several observations suggest that this group of D N A binding antibiotics acts as affinity labels for G - C base pairs in chromosomal DNA: 1. Chromomycin A 3 inhibits D N A dependent RNA synthesis when the D N A contains G - C basepairs (Ward et al., 1965).
298
K.F. Jorgenson et al.
2. The number of binding sites on D N A for the chromomycinone dyes increases with the G - C content of the D N A up to 66% G - C (Behr et al., 1969). 3. Olivomycin fluorescence is enhanced upon interaction with D N A which can be directly related to the G - C content of the DNA (van de Sande et al., 1977a). 4. The centromeric region of bovine chromosomes, which contain G - C rich satellite DNA, stain brightly with olivomycin (Kurnit et al., 1973; van de Sande et al., 1977a). The non-fluorescent G - C specific antibiotic, actinomycin-D, enhances the contrast in the fluorescent bands on chromosome preparations produced by the A - T specific fluorochrome Hoechst 33258. The use of actinomycin-D as a counterstaining agent to enhance the resolution of fluorescent bands produced by the A - T specific fluorochrome DAPI has been previously reported (Schweizer, 1976). Similarly, the non-fluorescent A - T specific D N A binding drug, netropsin, serves as an effective counterstaining agent thus increasing the resolution of chromosome bands produced by the G - C specific fluorochrome olivomycin. Netropsin also increases the differential fluorescence enhancement of olivomycin interacting with DNAs of very different base composition in solution. Consequently, the mechanism responsible for the effect of netropsin on olivomycin fluorescence in solution in the presence of DNA, could also account for the counterstaining effect of netropsin on olivomycin stained metaphase chromosomes. The compatibility of binding of different base pair specific ligands to DNA has been previously reported (Wartell et al., 1975). It was shown that the A - T specific ligand netropsin and the G - C specific drug actinomycin D can bind in close proximity. However, the binding of actinomycin D to D N A is decreased in the presence of netropsin. This effect is dependent upon the G - C content of DNA. A similar finding is reported here for the effect of netropsin on olivomycin binding to D N A as expressed by decreases in olivomycin fluorescence enhancement. The antibiotic netropsin covers three base pairs when bound to D N A (Wartell et al., 1974) and thus could eliminate potential olivomycin binding sites. This effect would be most pronounced for G - - C sites interspersed in an A - T rich region. Prior binding of netropsin in such an A - T rich region could mask olivomycin binding sites resulting in a reduction in of the " w e a k " olivomycin fluorescence present in the absence of the counterstaining agent (Fig. 8). The overall effect of the pretreatment with netropsin would be the accentuation of differential olivomycin fluorescence between A - T and G - C rich regions. This postulate would explain the observed increase in definition between bright and dull regions in counterstained metaphase chromosomes. A different counterstaining procedure was recently reported by Schweizer et al. (1978). The non-fluorescent A - T specific antibiotic distamycin A was found to reduce the overall fluorescence intensity of chromosomes banded with DAPI. The remaining bright regions in the chromosomes after this sequential staining procedure contain repetitive DNA, mainly A - T rich satellites. The molecular explanation of this finding remains open. A difference in the binding
Base Specific Fluorochromes for Chromosome Identification
Olivomycin-Netropsin
WN
A
g'N
k./
k_/
k,J
299
DNA complex
B I ill[llll[llll(ll)ll(ll)ll(ll)lllll)ll(ll)ll(l")llllllllllllllllll~ll~l k,#
C I
I
I
I II
~
~
I1{11)11{11)11(11)11 IIIlil
I ' ~ - O[ivomycin( sG+C pecific)
I
I
kJ
II
kJ
II
0 - Netropsin ( A+T specific)
Fig. 8. Model for the effect of netropsin binding to DNA on subsequent binding by olivomycin. The olivomycin binding sites are shown in A, the netropsin binding sites in B. The exclusion of olivomycin binding following prior binding by netropsin is indicated in C
of distamycin A to heterochromatic or euchromatic regions has been suggested (Schweizer et al., 1978). Since the effect of distamycin A as a D A P I counterstaining agent is concentration dependent, an alternate explanation may be that under the conditions used, all D A P I binding sites on chromosome arms are blocked by distamycin A, but there are still D A P I binding sites available in some specific constitutive heterochromatic regions. The observations that A - T and G - C specific fluorochromes produce banding patterns on chromosomes which are complementary to one another indicate that sequence arrangement in chromosomal D N A is a primary determinant in the production of fluorescent bands on metaphase chromosomes. The finding that counterstaining with non-fluorescent base specific D N A binding ligands of complementary specificity enhances the resolution of fluorescent bands on chromosomes produced by both A - T and G - C specific fluorochromes further supports this hypothesis. Previously, the observation was made that the resolution of Hoechst 33258 fluorescent bands on metaphase chromosomes improves with aging of the stained c h r o m o s o m e preparation (Jalal et al., 1975). The use of actinomycin D and netropsin as counterstaining agents for Hoechst 33258 and olivomycin respectively, provides a simple, rapid means by which an even stronger enhancement in the differentiation of fluorescent bands can be achieved. With Hoechst 33258, the counterstaining procedure results in the production of clearly defined banding patterns which are more stable than those produced by staining with quinacrine. This makes the actinomycin D - H o e c h s t 33258 procedure a useful tool
300
K.F. Jorgenson et al.
for routine chromosome analysis. In addition to the increased stability of the fluorescent bands, this procedure also demonstrates the existence of polymorphisms in the secondary constriction regions of human chromosomes 1, 9 and 16 as previously reported (Raposa and Natarajan, 1974). The variations which occur in these regions in normal polulations are not readily visualized by the Q-banding technique and can generally be identified only by the C-banding technique. The actinomycin D - H o e c h s t 33258 staining has also proved useful in the identification of individual chromosomes in somatic cell hybrids. Unequivocal identification of chromosomes in somatic cell hybrids complement is essential if this technique is to be used for gene mapping on chromosomes (McKusick and Ruddle, 1977). Hoechst 33258 fluorescent staining has previously been used to discriminate between human and mouse chromosomes in man-mouse somatic cell hybrids, by the differential staining intensity in the centromeric regions of the "parental chromosomes" (Lin et al., 1974). However, these studies required that the hybrid cells undergo one cycle of D N A replication in the presence of BUdR in order to enhance the difference in the Hoechst 33258 stained human and mouse centromeres. The present study shows that the A - T specific fluorochrome Hoechst 33258 after counterstaining with actinomycin D can be directly used in the unequivocal identification of parental chromosomes in mail-mouse somatic cell hybrids without prior treatment of the somatic cells with BUdR. The differential staining of centromeric heterochromatin in some mammalian species appears to be the result of highly repetitive satellite DNAs located in these regions. In the mouse essentially one kind of A - T rich satellite D N A is distributed to almost every centromeric region of the mouse complement (except the Y-chromosome), whereas G - C rich satellite DNAs are located in the centromeric regions of the bovine chromosome complement (except the sex chromosomes). These two cases represent an extreme contrast in terms of the base composition of their centromeric heterochromatins. The distribution of centromeric heterochromatin in the porcine chromosome complement appears different than in either mouse or cattle. In this species, only the bi-arm chromosomes stain olivomycin bright and contain G - C rich centromeric heterochromatin. Whether or not this G - C rich centromeric heterochromatin represents a repetitive satellite D N A of the procine genome remains unclear at this time. These studies indicate that three different responses can be obtained in the staining of centromeric heterochromatin with the base specific fluorochromes Hoechst 33258 and olivomycin. The centromeres which show Hoechst 33258 positive, olivomycin negative fluorescence contain A - T rich sequences, while the centromeres which show olivomycin positive, Hoechst 33258 negative fluorescence represent G - C rich sequences. The sequence arrangement in the centromeric regions which stain negative with both Hoechst 33258 and olivomycin, e.g. the centromeric regions of porcine telocentric chromosomes and human centromeric regions, remains a question at this time. The observations that the base pair specific fluorochromes of opposing specificity act in a complementary fashion both with D N A in solution and with metaphase chromosomes, and that the bands produced on cytological prepara-
Base Specific Fluorochromes for Chromosome Identification tions can be effectively counterstained using nonfluorescent DNA
301 binding agents
of opposite specificity, strongly support the hypothesis that base sequence arr a n g e m e n t w i t h i n c h r o m o s o m e s is a p r i m a r y d e t e r m i n a n t i n t h e p r o d u c t i o n o f f l u o r e s c e n t b a n d s . T h e e n h a n c e d d e f i n i t i o n in f l u o r e s c e n t b a n d s , a t t a i n e d using counterstaining, aids both chromosome identification and the investigation of certain heterochromatic regions.
Acknowledgements.We would like to thank Mrs. Heidi K. Jamro for her excellent technical assistance. This work was supported by the Medical Research Council of Canada grant Nos. MA-4655 and MA-4885 and The National Foundation-March of Dimes grant No. 1-544.
References Behr, W., Honikel, K., Hartmann, G. : Interaction of the RNA polymerase inhibitor chromomycin with DNA. Europ. J. Biochem. 9, 82-92 (1969) Distbche, Bontemps, J. : Chromosome regions containing DNAs of known base composition, specifically evidenced by 2,7-di-t-butyl proflavine. Chromosoma (Berl.) 47, 263 281 (1974) Hilwig, I., Gropp, A.: Stains of constitutive heterochromatin in mammalian chromosome with a new fluorochrome. Exp. Cell Res. 75, 122-126 (1972) Jalal, S.M., Markvong, A., Hsu, T.C.: Differential chromosomal fluorescence with 33258 Hoechst. Exp. Cell Res. 90, 443-444 (1975) Jones, K.W. : Chromosomal and nuclear location of mouse satellite DNA in individual cells. Nature (Lond.) 225, 912 915 (1970) Jones, K.W., Corneo, G.: Location of satellite and homogeneous DNA sequences on human chromosomes. Nature (Lond.) New Biol. 233, 268 271 (1971) Kurnit, D.M., Shafit, B.R., Maio, J.J.: Multiple satellite deoxyribonucleic acids in the calf and their relation to the sex chromosomes. J. molec. Biol. 81, 274 284 (1973) Latt, S.A., Stetten, G.: Spectral studies on 33258 Hoechst and related bisbenzimidazole dyes useful for fluorescent detection of deoxyribonucleic acid synthesis. J. Histochem. Cytochem. 24, 24-33 (1976) Latt, S.A., Wohlleb, J. : Optical studies of the interaction of 33258 Hoechst with DNA, chromatin and metaphase chromosomes. Chromosoma (Berl.) 52, 297 316 (1975) Lin, C.C., Uchida, I.A. : Fluorescent banding of chromosomes (Q-bands). In: Methods and applications of tissue culture (P.F. Kruse and M,K. Patterson, eds.), pp. 778 781. New York, N.Y.: Academic Press, 1973 Lin, M.S., Latt, S.A., Davidson, R.L. : Identification of human and mouse chromosomes in humanmouse hybrids by centromere fluorescence. Exp. Cell Res. 87, 429-433 (1974) McKusick, V.A., Ruddle, F.H.: The status of the gene map of human chromosomes. Science 196, 390-405 (1977) Moorhead, P.S., Nowell, P.C., Mellman, W.J., Batipps, D.M., Hungerford, D.A.: Chromosome preparations of leukocytes cultured from human peripheral blood. Exp. Cell Res. 20, 613 616 (1960) Mfiller, W., Crothers, D.M.: Studies of the binding of actinomycin and related compounds to DNA. J. molec. Biol. 35, 251 290 (1968) Pardue, M.L., Gall, J. : Chromosomal localization of mouse satellite DNA. Science 168, 1356-1358 (1970) Raposa, T., Natarajan, A.T.: Fluorescence banding pattern of human and mouse chromosomes with a benzimidazole derivative (Hoechst 33258). Humangenetik 21, 221 226 (1974) Reich, E., Goldberg, I.H.: Actinomycin and nucleic acid function. Progr. Nucleic Acid Res. molec. Biol. 3, 183 234 (1964) Sande, J.H., van de, Lin, C.C., Jorgenson, K.F.: Reverse banding on chromosomes produced by a guanosine-cytosine specific DNA binding antibiotic: olivomycin. Science 195, 400~402 (1977a)
302
K.F. Jorgenson et al.
Sande, J.H. van de, Lin, C.C., Johnston, F.P., Jorgenson, K.F.: Differential fluorescent labelling of chromosomes and DNA with base pair specific DNA binding antibiotics. In : Human cytogenetics: ICN-UCLA Symp. molec, cell. Biol. Vol. VII, pp. 127-138. New York, N.Y. : Academic Press Schnedl, W., Breitenbach, M., Mikelsaar, A.-V., Stranziner, G.: Mithramycin and DIPI: A pair of fluorochromes specific for G - C and A - T rich DNA respectively. Hum. Genet. 36, 299-305 (1977b) Schnedl, W., Mikelsaar, A.-V., Breitenbach, M., Dann, O. : DIPI and DAPI: fluorescence banding with only negligible fading. Hum. Genet. 36, 167-172 (1977a) Schreck, R.R., Erlanger, B.F., Miller, O.J.: The use of antinucteoside antibodies to probe the organization of chromosomes denatured by ultraviolet irradiation. Exp. Cell Res. 88, 31 39 (1974) Schweizer, D.: Reverse fluorescent chromosome banding with chromomycin and DAPI. Chromosoma (Berl.) 58, 307-324 (1976) Schweizer, D., Ambros, P., Andrle, M.: Modification of DAPI banding on human chromosomes by prestaining with a DNA-binding oligopeptide antibiotic, Distamycin A. Exp. Cell Res. 111, 327-332 (1978) Ward, D.C., Reich, E., Goldberg, I.H.: Base specificity in the interaction of polynucleotides with antibiotic drugs. Science 149, 1259 1263 (1965) Wartell, R.M., Larson, J.E., Wells, R.D. : Netropsin, a specific probe for A - T regions of duplex deoxyribonucleic acid. J. biol. Chem. 249, 6719-6731 (1974) Wartell, R.M., Larson, J.E., Wells, R.D.: The compatibility of netropsin and actinomycin binding to nature deoxyribonucleic acid. J. molec. Biol. 250, 2698 2702 (1975) Weisblum, B., Haenssler, E.: Fluorometric properties of the bibenzimidazole derivative Hoechst 33258, a fluorescent probe specific for A - T concentration in chromosomal DNA. Chromosoma (Berl.) 46, 255-260 (1974) Wells, R.D., Larson, J.E.: Studies on the binding of actinomycin D to DNA and DNA model polymers. J. molec. Biol. 49, 319 342 (1970)
Received May 9 - J u n e 13, 1978 / Accepted June 13, I978 by W. Beermann Ready for press June 18, 1978
