Eur. J. Biochem. 56, 403-411 (1975)
Studies on the Fate of Homologous DNA Applied to Seedlings of Matthiola incana Vera HEMLEBEN, Nortrude ERMISCH, Dorothee KIMMICH, Birgit LEBER, and Gabriele PETER Department of Biology 11, University of Tiibingen (Received March 13iApril 21, 1975)
Seedlings of Mutthiolu incunu (crucifer) are able to take up exogenous homologous DNA by the roots. DNA homogenously labelled with [3H]adenine and 5-bromodeoxyuridine is incorporated into the plants in a macromolecular form. Intact donor DNA and a fraction with a buoyant density intermediate between that of the donor and the recipient DNA can be recovered. Analysis of this intermediate fraction by ultrasonication and alkali treatment allows the suggestion that homologous DNA is integrated as a double-stranded DNA which becomes covalently linked to the recipient DNA. Control experiments in which seedlings were incubated in a mixture simulating donor DNA degradation products in the presence and absence of unlabelled competitors suggest that these results are not due to the breakdown of donor DNA and reincorporation of the products during DNA synthesis in the recipient plants. When ultrasonicated or thermally denatured DNA is applied to the plants it may be degraded and reused for recipient DNA synthesis but it is not recovered in a macromolecular form. The possibility that the intermediate DNA fraction arises by bacterial contamination of the plants can be excluded by several arguments. Autoradiographic studies show that at least part of the radioactivity of the donor DNA taken up by the plants is associated with the cell nucleus.
There have been many reports that high-molecularweight DNA can be incorporated by plant protoplasts, intact cells and whole plants. These experiments involved the use of heterologous DNA from various bacteria and phages [l - 81. Identification of the foreign DNA occurred mostly by virtue of the different buoyant densities of donor and host DNA. In order to differentiate an exogenous homologous DNA from the recipient DNA by caesium chloride equilibrium centrifugation it is necessary to use donor DNA with an altered buoyant density. This can be obtained by the incorporation of 5-bromodeoxyuridine (BrdUrd) which increases the density of the DNA. If in addition a radioactive label is incorporated, this allows the detection of small amounts of the “heavy” DNA. Abbreviations. BrdUrd, 5-bromodeoxyuridine; DNase, deoxyribonuclease I; RNase, ribonuclease A. Enzymes. Deoxyribonuclease I (EC 3.1.4.5); pronase (EC 3.4.21.4 and 3.4.24.4); ribonuclease A (EC 3.1.4.22).
Eur. J. Biochem. 56 (1975)
Uptake of BrdUrd-labelled homologous DNA in a macromolecular form by animal or human cells in culture has been shown [9- 121. In this kind of experiment it is necessary to distinguish between the uptake of intact exogenous DNA and the reutilisation of labelled or BrdUrd-containing degradation products of the exogenous DNA for recipient DNA synthesis. This is possible by including unlabelled DNA precursors in the incubation medium. Addition of a large amount of unlabelled thymidine should prevent the utilisation of BrdUrd from degraded BrdUrd-labelled DNA. The incorporation of radioactive degradation products can also be prevented by employing the appropriate competitor. In this report we used a [3H]thymidine or [3H]adenine-BrdUrd-labelled DNA from Mutthiolu incunu as donor DNA and unlabelled seedlings of Mutthiolu as recipients. The uptake and fate of the DNA was studied by measuring the amount of radioactivity within the plants, by autoradiography and caesium
404
chloride equilibrium centrifugation of the recipient DNA.
MATERIALS AND METHODS Growth and Labelling o j the Plants
Batches of 200-400 seeds of Mutthiolu incanu (strain 17 or 19, Department of Genetics, University of Tubingen) were sterilised by shaking with Orthocid (Merck, Darmstadt) for 10 min, washed several times with sterile water and grown in glass vials (15 x 65 mm), each containing 5 seeds in 200 p1 distilled water, in continuous light of 500 foot candles (5381 lx) at 20 "C. To obtain radioactively labelled or radioactively and density labelled homologous donor DNA the washed seeds were incubated in solutions of [6-3H]thymidine (15 pCilml; specific activity : 26 Ci/mmol) or [8-3H]adenine (15 pCi/ml; specific activity: 2427 Ci/mmol) and 5-bromodeoxyuridine (500 pg/ml). After 8 days the plants were harvested and the DNA extracted. The specific activity was generally between 2- 10 x lo6 counts min-' mg D N A - ~ . Incubation of (he Seedlings with Donor D N A
7-day-old seedlings were incubated (5 seedlings in 200 p1 solution) with sterile dialysed DNA preparations labelled with [3H]thymidine or [3H]adenine and BrdUrd (100 pgiml; about 5 x lo5 counts min-' ml-'), together with unlabelled DNA precursors (1 mg adenine and thymidine/ml) dissolved in 0.01 M KC1-0.001 M sodiumcitrate buffer pH 6.7 for various lengths of time in continuous light at 20 "C. Only the roots were in contact with the solution. The incubation mixture was plated on nutrient broth agar plates for bacterial counts. All steps were carried out under sterile conditions. The recipient plants were then immediately harvested, washed and treated twice with DNase I (20 pg/ml) for 30 min at 37 "C and 1 h with pronase (50 pgiml) at 37 "C to remove the exogenous DNA from the surface and the DNA was extracted from the seedlings. In some experiments the seedlings were washed with sterile water after incubation with donor DNA and reincubated for a further period of time with distilled water and unlabelled precursors before DNase treatment and DNA extraction. Isolation of' D N A
The seedlings were gently homogenised with a glass tube and pestle with 0.1 M Tris - 0.005 M EDTAHCl buffer pH 7.5 and sodium dodecyl sulfate (1 final concentration). The homogenate was shaken
Fate of Homologous DNA in Seedlings of Mutthiolo
with 1 volume of phenol mixture (88 ml phenol, water saturated, 12 ml m-cresol, 0.1 g 8-hydroxyquinoline) for 20 min. After centrifugation (Minifuge, Christ, 10 min, 5000 rev./min, 4 'C) the aqueous phase was removed by a syringe with a 3-mm dsameter needle. The phenol phase was reextracted with TrisEDTA-HCl buffer, the aqueous phases were combined and again shaken with half the volume of phenol mixture and centrifuged. The final aqueous phase was precipitated with two volumes of ethanol for at least 3 h at - 28 "C. The alcohol precipitate was dissolved in Tris-EDTA-HC1 buffer, treated with RNase A (20 pg/ ml) for 2 h and with pronase (100 pgIml) for 2 h at 37 "C. The DNA was further purified by shaking with 1 volume of chloroform - isoamyl alcohol (10 : 1) for 30 min, followed by centrifugation. This treatment was repeated twice with the supernatant and the DNA was precipitated with 2 volumes of ethanol at - 28 "C. For CsCl gradient centrifugation the DNA was dissolved in 1 ml Tris-EDTA-HC1 buffer. Incubation experiments were carried out with DNA dialysed against 0.01 M KCl-0.001 M sodium citrate buffer pH 6.7 for 24 h under sterile conditions. The A260/A280 ratio was about 1.85- 2.0. The specific activity was measured by counting 10 pl of a DNA sample in 2 ml Bray's solution in a liquid scintillation counter (Packard, Model 3390). Cuesium Chloride Equilibrium Centr ifugat ion
DNA samples dissolved in Tris-EDTA-HC1 buffer were fractionated on neutral or alkaline (pH 12.5) CsCl gradients. The density was adjusted to 1.74 g/ml for neutral gradients and 1.81 gjml for alkaline gradients. Samples (2.5 or 3.0 ml) were centrifuged for 45 h a t 33 000 rev./min in the SW 65 rotor of a Beckman Model L 2 preparative ultracentrifuge at 20 " C . The tubes were punctured from the bottom and 60-70 three-drop fractions were collected per gradient. Every tenth fraction was used for measurements of the refractive index; to every second fraction I ml of Tris-EDTA-HC1 buffer was added for measurement of the absorbance. To every other remaining fraction 1 ml yeast RNA (250 pg/ml) was added as carrier and the nucleic acids were precipitated with 1 ml 10% trichloroacetic acid at 0 "C. The precipitated samples were collected on Sartorius membrane filters (0.45 pm, 24-mm diameter), washed twice with 3 "/, trichloroacetic acid and put in new scintillation vials with 2 ml toluene scintillator solution. The radioactivity was measured in a liquid scintillation counter. Samples were counted for 10-20 min and corrected for background radiation. After separation on CsCl gradients certain fractions were combined, placed in dialysis bags, dialysed Eur. J Biochem. 56 (1975)
V. Hemleben, N . Ermisch, D. Kimmich, B. Leber, and G. Peter
405
Fraction number Fig. 1. Characterisation of [ 3H]adeninelBrdUrd-labellrd homologous Matthiola D N A by CsCI gradient centrifugation. DNA was isolated from seedlings grown in [3H]adenine (15 pCi/ml) and BrdUrd (500 pg/ml) for 8 days. 10 pg labelled D N A were separated together with unlabelled Matthiola D N A as reference (A) on a neutral CsCl gradient, (C) on an alkaline CsCl gradient (pH 12.5). Ultrasonicated labelled DNA was fractionated (B) on a neutral and (D) on an alkaline CsCl gradient. (Rotor SW 65, 33000 rev./min, 45 h, 20 "C.) ( 0 4 ) Absorbance at 260 nm; ( o - - - o ) 3H radioactivity
and concentrated in a solution of 30 % (w/w) Dextran T 70 (Pharmacia, Uppsala). After concentration these fractions were either ultrasonicated for 2 min (50 W ; sonifier Branson) and fractionated again on neutral CsCl gradients or separated on alkaline gradients.
Autoradiography
7-day-old seedlings were incubated with [3H]thymidine-labelled donor DNA ( 5 x lo5 counts min-' m1-I) and unlabelled thymidine (500 pg/ml) dissolved in 0.01 M KCl-0.001 M sodium citrate buffer for 24 h under sterile conditions and treated with DNase I and pronase as described above. The root tips of some plants were fixed, Feulgen stained and used for squash preparations for autoradiography. From the remaining seedlings the nuclei were isolated. The isolated nuclei were fixed with methanol/formalin/glacial acetic acid (85 : 15 : 5 ; v/v/v), distributed on slides and stained by Feulgen staining after hydrolysis with 5 N HCI for 1 h at room temperature. Kodak AR 10 stripping film was applied to the slides. The exposure time was between 30- 90 days at 4 "C. The autoradiograms were developed with Kodak D-l9b, fixed with Kodak Unifix and used for microphotographical investigations. Eur. J. Biochem. 56 (1975)
Isolation of Nuclei
Seedlings were gently homogenised in a medium containing 0.05 M Tris, 0.01 M MgCI,, 0.2 M sucrose and 2.5% Ficoll (Pharmacia, Uppsala), pH 7.8. The homogenate was filtered successively through nylon grids (300, 100, 30 and 10 pm pore diameter). The 10-pm filtrate was centrifuged at 1000 x g for 10 min at 2 "C (Minifuge, Christ) and washed once with homogenisation buffer. The sediment containing the intact nuclei and some chloroplasts was used for autoradiographic studies. RESULTS Characterisation of Native and BrdUrd-Labelled Matthiola D N A
Unlabelled DNA isolated from seedlings of Matthiola incana has a buoyant density of 1.698 g/ml when separated on a neutral CsCl gradient (Fig. 1A). If the seedlings are grown in a solution of 5-bromodeoxyuridine (500 pg/ml) and simultaneously labelled with [3H]adenine for 8 days the DNA buoyant density shifts to 1.741 g/ml (Fig. 1A) [13] and the DNA has a specific activity of about 2- 10 x lo6 counts min-' mg-'. Previous experiments had shown that at lower
406
Fate of Homologous D N A in Seedlings of Mutthiolu
,
I
20
I
I
LO
1
I
I
60
Fraction number
Fig. 2. CM’I grudient cenlrifugution y f D N A isolutedfiom Matthiola seedlings incubated with liomologous D N A . (A) Neutral CsCl gradient centrifugation of DNA isolated froin 8-day-old seedlings previously incubated for 20 h with [3H]adenine/BrdUrd-labelled D N A (100 pg ml) and unlabelled adenine and thymidine (1 mg/ml each) as competitors. (B) Neutral CsCl gradient centrifugation of DNA isolated from 8-day-old secdlings grown in BrdUrd (500 pg/ml) and previously incubated with “light” [3H]adeninei[3H]guanosine-labclledDNA and unlabelled precursors as competitors for 20 h. The recipient plants were treated with DNase and pronase before the isolation of the DNA. Centrifugation was as described for Fig.1. In (B) unlabelled ‘‘light’’ Mutthiolu D N A was added as reference. (O---o) Absorbance at 260 nm; (0- - 0 ) ’H radioactivity ~
concentrations of BrdUrd in the growth medium only part of the DNA shifts to a higher density. Using higher doses of BrdUrd (200- 1000 pg/ml) the density shift is nearly constant at about 0.043 g/ml. N o higher incorporation of BrdUrd was observed by attempts to inhibit thymidine kinase with 5-fluordeoxyuridine [14]. Ultrasonicated [’H]adenine/BrdUrd-labelled DNA shows a broad peak with a maximum of radioactivity at the original buoyant density when fractionated on a neutral CsCl gradient (Fig. 1 B). The ultrasonicated DNA fractionated on an alkaline CsCl gradient appears as a broad peak with a maximum at a density of 1.85 glml (Fig. 1 D). Unsonicated DNA samples separated on an alkaline CsCl gradient band at the same density (Fig.1C). These experiments show that the distribution of BrdUrd within the [3HH]adenine!BrdUrd-labelledDNA molecules is homogenous. The criteria required for studies on uptake and identification of a homologous DNA within the recipient, i . r . difference of buoyant density between donor and recipient DNA, homogenity of the BrdUrd distribution and high specific activity of the donor DNA, are therefore fulfilled.
Effect of BrdUrd on Matthiola Seedlings High concentrations of BrdUrd in the growth medium affect the growth and development of the plants. The BrdUrd-labelled plants (500 pg BrdUrd/ ml) are smaller and show no root hairs and the DNA content per seedling is decreased. For uptake experiments where the donor DNA is probably taken up by
the roots and root hairs [3H]adenineiBrdUrd-labelled DNA was mostly used as donor DNA and normally growing “light” seedlings as recipients. Incubation of Matthiola Seedlings tvith Honzologous D N A
Preliminary studies showed that seedlings of Mutthiolu take up exogenous DNA irrespectively of their developmental stage between the 4th and the 8th day after germination. In subsequent experiments incubations were carried out with 7-day-old seedlings with fully developed root hairs. Uptake studies showed radioactivity in the recipient plants after 5 - 10 h of incubation with t3H]adenine/BrdUrdlabelled DNA but greater incorporation was obtained if the incubation period lasted between 16 and 20 h. Between 0.1 and 1 ”/, of the applied DNA was taken up by the seedlings. After longer incubation periods the donor D N A ( M , = 2- 4 x 10‘) had a reduced size of about 2-5 x lo5. Batches of 7-day-old unlabelled seedlings were incubated with [3H]adenine/BrdUrd-labelled-DNA (100 pg/ml; specific activity: 5 x 10’ counts min-’ mg DNA- ’) and unlabelled competitors thymidine and adenine (1 mg/ml each) in KCl- sodium citrate buffer for 20 h. The DNA was isolated from the DNase and pronase-treated plants and fractionated on a neutral CsCl gradient (Fig.2A). Two main peaks of radioactivity were separated with buoyant densities of about 1.741 and 1.717 g/ml. Very small amounts of radioactivity coincided with the “light” recipient DNA The peak with a density of 1.741 gjml corresponds to F u r J Biochem 56 (1975)
V. Hemleben, N. Ermisch, D. Kimmich, B. Leber, and G. Peter
407 80 1
c
E
60
0
v1 C
(D N
c
0.10
8
0 a,
LO
c
e
r .5 ._ c
0
0.05
20
Q
7 cc
0 0
20
LO
60 Fraction number
20
LO
60
Fig. 3. CsCl gradient centrifugation of D N A isolated from Matthiola Jeedlings incubated with ultrasonicated or heat-denatured DNA. Neutral CsCl gradient centrifugation of DNA isolated from 8-day-old unlabelled seedlings previously incubated for 20 h with (A) ultrasonicated and (B) heat-denatured [3H]adenine/BrdUrd-labelledDNA (100 pg/ml) and unlabelled adenine and thymidine as competitors. Centrifugation was as described for Fig. 1. (0--0) Absorbance at 260 nm; (O----+) 3H radioactivity
the intact donor DNA, the second peak shows an intermediate density between donor and recipient DNA. The radioactivity in the DNA of the recipient plants indicates a certain breakdown of donor DNA and reutilisation of the products for host DNA synthesis [15]. Similar results were obtained when “light” [3H]adenine/[3H]guanosine-labelledDNA, together with unlabelled adenine, guanosine and BrdUrd as competitors of donor DNA degradation products, was applied to recipient plants in which the DNA had been previously labelled with BrdUrd (500 pg/ml). DNA isolated from these plants after 20 h of incubation and separated on a neutral CsCl gradient showed three peaks of radioactivity (Fig. 2B), representing applied donor DNA (buoyant density: 1.698 g/ml), an intermediate fraction (1.719 g/ml) and a small amount of radioactivity with the density of the “heavy” recipient DNA indicating very slight reincorporation of degradation products. Recipient plants were incubated with [3H]adenine/ BrdUrd-labelled DNA as described for Fig. 2A except that the plants were removed from the DNA-containing medium after 2 min. After treatment of the plants with DNase and pronase and isolation of the DNA no radioactivity was found in the DNA preparation. This indicates that the exogenous DNA is completely removed from the surface of the seedlings after treatment with DNase. Therefore the peaks of radioactivity with the density of 1.741 g/ml in Fig. 2A and with the density of 1.698 giml in Fig.2B represent intact donor DNA present in the recipient plants but inaccessible to DNase. No peaks of radioactivity with the density of intact donor DNA or intermediate density appeared within the recipient plant DNA if the donor DNA applied Eur. J. Biochem. 56 (1975)
to the seedlings was ultrasonicated ( M , = 5 x lo5) or heat-denatured to single-strand DNA before incubation with the plants. The DNA isolated from seedlings treated in this way showed radioactivity only in the “light” peak (Fig. 3A, B) when separated on neutral CsCl gradients indicating that small fragments of double-stranded DNA or single-stranded DNA are degraded by the recipient nucleases and reutilised in the DNA synthesis. However, in the presence of unlabelled competitors it is not possible to compare the amount of native, ultrasonicated and denatured DNA taken up by the plants. These experiments show that the occurrence of (a) DNase-resistant donor DNA in the plants and (b) the intermediate DNA peak depends on the donor DNA being of high molecular weight and doublestranded. The Incorporation of Radioactivity by Recipient Plants Utilising Donor D N A Precursors Two batches of 7-day-old unlabelled seedlings were incubated for 20 h in a medium simulating the application of a degraded [3H]adenine/BrdUrd-labelled donor DNA. The incubation medium contained [3H]adenine (lo6 counts min-’ ml-l) and BrdUrd (50 pg/ml) corresponding to the [3H]adenine and BrdUrd content of the donor DNA usually applied to the plants (see Fig. 4A) and in one case unlabelled adenine and thymidine (500 pg/ml) was added. The DNA was then isolated and separated on neutral CsCl gradients (Fig.4A and B). In the absence of the BrdUrd competitor thymidine and unlabelled adenine the newly synthesised radioactive DNA is heavy, with a mean density of 1.739 g/ml although some less dense DNA fractions are also detectable (Fig. 4A). In the presence
408
Fate of Homologous DNA in Seedlings of Mutihioh
Fraction number
Fig. 4. Tllc i n c . o ~ ~ ~ o ~ uof t i oBrdC‘rd ii und [”H/udenine into Matthiola D N A it? the presence and ub.wnce of unluhellrd coiwpetitor.~.7-daj-old seedlings cultivated in distilled water were treated for 20 h with 50 pg BrdUrdiml and [3H]adenine (10’ counts min-’ inl~-’)(A) without and (B) with unlabelled adenine and thymidine (500 pgirnl each). The D N A was isolated and separated on a neutral CsCl gradient as described in the legend to Fig. 1. (0- 0)Absorbance at 260 nm; (o-----o) 3H radioactivity -
of an excess of thymidine and adenine the radioactive peak is completely shifted to the “light” region of the gradient (1.698 giml) without any intermediate fractions (Fig. 4 B). In addition, the presence of unlabelled adenine as competitor of the [3H]adeninedramatically reduces the incorporation of radioactivity into DNA (compare F i g 4 A and B). These results allow us to conclude that the radioactivity appearing in the recipient plants after a certain incubation time with [3H]adenine/BrdUrd-labelled donor DNA (Fig. 2A) results at least partly from donor DNA in an undegraded form.
Density(g/ml)
I
1.713
0.15
$
60
5?0.10
vi
N
c
t 7
I
0
0 U
LO ;
8
c 2 c
0
f
U
0.05
; 0
20
Q
0
Stabilitj~of thc Incorporated Donor D N A
The fate of the exogenous DNA after uptake into the plants was studied in the following way. Seedlings were incubated with [3H]adenine/BrdUrd-labelled DNA as described in the legend to Fig.2A for 20 h with unlabelled adenine and thymidine as competitors, then the DNA was removed and the plants were allowed to grow for a further period of 24 h in distilled water with unlabelled adenine and thymidine. The DNA from these plants was then isolated and fractionated on a neutral CsCl gradient (Fig.5). Under these conditions the fraction with a density of 1.741 g/ml corresponding to donor DNA, disappears (compare with Fig. 2A) whereas the intermediate fraction with a density of about 1.713g/ml appears to be more stable. Characterisation
of
the Intermediate D N A Fraction
The studies of uptake of homologous DNA show that the exogenous DNA enters the plant cells in a
0
20
40 Fraction number
60
Fig. 5. l h e “disuppeurance” ~f homologous [‘H]adenine,’BrrlUrdlabelled D N A f r o m Matthiola seedlings during u chase incuhution. 7-day-old unlabelled seedlings were incubated with [3H]adenine, BrdUrd-labelled D N A and unlabelled competitors for 20 h, washed and incubated in distilled water with adenine and thymidine (1 mg/ml each) for a further 24 h. D N A was extracted and fractionated on a neutral CsCl gradient as described in the legend to Fig.1. (0 0) Absorbance at 260nm; (0 - - ~ - O )‘H radioactivity
macromolecular form. However, DNA of donor density does not persist although a fraction of intermediate density is stable (Fig. 5). It is possible that the intermediate DNA fraction results from the integration of donor DNA into the “light” recipient DNA. The intermediate fraction was therefore analysed by mechanical shearing or alkali treatment following the methods of Ledoux et al. [6]. The dialysed, concentrated fractions of the intermediate fraction formed after incubation of the seedlings with “heavy” donor DNA for 20 h (Fig. 2A, fractions 31 - 39) were ultrasonicated Eur. J. Biochem. 56 (1975)
409
V. Hemleben, N. Ermisch, D. Kimmich, B. Leber, and G. Peter Density( glml)
30
I
. G E
E
6 0.10
w
10 C
c
20
c
8
DNA bands at the expected density of 1.762 g/ml (Fig. 1C). Heavy BrdUrd-labelled DNA would band at a density of 1.853 g/ml if separated in alkaline CsCl (Fig. 1C). This results suggest that donor DNA is covalent bound to light recipient DNA.
0
x
c
s
0.05
c
0
0 10
10 9
n
Localisation of Exogenous D N A after Uptake into the Seedlings
-0
Q
LY
0 20
0
LO Fraction number
B
60
Density(g/rnl)
Fraction number Fig. 6. Characterisation oj the intermediate D N A fraction. (A) DNA from fractions 31-39 of the CsCl gradient shown in Fig.2A was concentrated, ultrasonicated and refractionated on neutral CsC1. Unlahelled Marzhiola DNA was added as reference. (B) DNA from fractions 31 -41 of the CsCl gradient shown in Fig. 5 was concentrated and refractionated without ultrasonication on alkaline CsCl (pH 12.5). Unlahelled Mutthiola DNA was added as reference. +) Absorbance at Centrifugation as described for Fig. 1. (0 -~~ 260 nm; ( C - - O ) 3H radioactivity
and refractionated on a neutral CsCl gradient (Fig. 6 A). This apparently homogenous peak now splits up into at least two fractions of radioactivity, one with the same density as the “heavy” donor DNA (1.741 g/ml) and the other one with an intermediate density (1.721 g/ml). This suggests that the intermediate peak contained relatively high-molecularweight fragments of double-stranded donor DNA which were released by ultrasonication. The corresponding release of “light” DNA from the intermediate peak cannot be directly demonstrated or differentiated from the unlabelled reference DNA because such DNA would not be labelled. The intermediate peak obtained after chase conditions (Fig. 5, fractions 31 -41) was refractionated on an alkaline CsCl gradient with unlabelled Matthiola DNA as reference (Fig. 6B). Under these conditions the radioactive peak is still homogenous but is shifted to a density of 1.781 g/ml while the unlabelled reference Eur. J. Biochem. 56 (1975)
Batches of 7-day-old seedlings were incubated with [3H]thymidine-labelled DNA (5 x lo5 counts min-’ m1-I) for 24 h with a large excess of unlabelled thymidine to prevent the reincorporation of 3Hlabelled degradation products. In control experiments plants were incubated with [3H]thymidine (5 x lo5 counts min-’ m1-I) and unlabelled thymidine. The plants were then treated with DNase and pronase. Some of them were fixed, Feulgen stained and used for root-tip squash preparations. From the remaining plants the nuclei were isolated, fixed and distributed on slides and Feulgen stained. The squashed root tips and the isolated nuclei were analysed by autoradiography. After incubation in [3H]thymidine for 24 h most of the nuclei in the elongating zone of the root were labelled. In contrast much less radioactivity was associated with the roots when it was supplied as [3H]thymidine-labelled DNA. Nevertheless, many of the silver grains were localised over nuclei and there was no evidence for the accumulation of radioactivity at the root epidermis or in the cell walls. Preparations of isolated nuclei show that at least part of the radioactivity becomes associated with nuclei (Fig. 7). The results of the autoradiographic investigations are consistent with the evidence described above which suggests that exogenous homologous DNA may become associated with the recipient genome. DISCUSSION These results show that exogenous homologous DNA is incorporated into seedlings of Matthiola incana. After short incubations all the exogenous DNA can be removed from the plants by treatment with DNase but subsequently DNA taken up by the seedlings becomes resistant to this treatment. After 20-h incubation [3H]adenine/BrdUrd-labelled DNA with the same buoyant density as the original donor DNA can be recovered from the seedlings together with a second DNA fraction with a buoyant density intermediate between that of the donor and the recipient. The donor DNA and the intermediate peak are only detected in recipient plants if the applied DNA is of high molecular weight ( M , > lo6) and double-
410
Fate of Homologous DNA in Seedlings of A4ut//iio/u
Fig. 7. A ~ / i o i . ~ r t / i ~ , ~ i . t o/ r i ~ Isohted /i~. rruciei. 7-day-old seedlings were treated for 24 h with [’Hlthymidine-labelled homologous DNA (100 pg:ml; I 5 x lo5 counts min-’ ml-’) and unlabelled thymidine as competitor. Isolation and preparation ofthe nuclei and autoradiography was dons a s described in Materials and Methods. Three representative nuclei are shown
stranded. Chase incubations show that the donor DNA does not persist in its native form but is converted into the intermediate density DNA. The apparent decrease in buoyant density of the donor DNA (0.034-0.028 giml) could be explained if it becomes attached to DNA of much lower density. This interpretation is supported by the fact that DNA segments of original donor DNA density can be recovered from the intermediate fraction by ultrasonication. The donor DNA appears to be covalently attached in a double-stranded form because no single-stranded DNA of donor density is obtained following the recentrifugation of unsonicated intermediate density DNA in alkaline CsCl. The density of the intermediate DNA fraction after 16- 20 h of incubation of the recipient plants with [3H]adenine/BrdUrd-labelled DNA varies from experiment to experiment from 1.727 to 1.717 g/ml [I51 and the peak is not always as sharp as shown in Fig. 2A. That means that the integrated donor fragments are of difTerent size. A DNA fraction with almost intermediate density between donor and recipient DNA must consist of a relatively similar amount of light recipient DNA (molecular weight of the DNA preparation about 2- 5 x lo6)and fragments of integrated heavy DNA. Two further possibilities of interpretation of the intermediate fraction should be discussed. (a) The donor DNA enters the cell and serves as a template for endogenous DNA polymerases. This would be expected to produce one light strand and one heavy strand. However, the intermediate (or “hybrid”) fraction obtained contains double-stranded fragments of original donor density which is inconsistent with such an interpretation. (b) BrdUrd-labelled donor DNA becomes associated with homologous sites in
the recipient genome in a process of single-strand recombination forming “hybrids”. This possibility can also be excluded by the ultrasonication experiment mentioned above and by the results of alkali treatment of the intermediate fraction after chase conditions : only single strands with an intermediate density are separated on an alkaline CsCl gradient. These results support the interpretation of an integrated doublestranded donor DNA segment with light recipient DNA similar to Ledoux’s observations with heterologous DNA integrated in plant DNA [6]. In this type of study it is very important to rule out the possibility that the DNA with intermediate density is derived from contaminating bacteria. Bendich [16] showed that labelling of contaminated plants results in “false” D N A satellites with intermediate densities. Other DNA satellites with similar buoyant densities to the intermediate fraction described here originate from bacterial contamination [17]. Our experiments were carried out under aseptic conditions and seedlings and DNA samples were routinely monitored for bacterial growth. In addition, in experiments simulating the incubation conditions with completely degraded [3H]adenine/BrdUrd-labelled donor DNA by adding the corresponding amount of [‘HH]adenine and BrdUrd and unlabelled competitors no intermediate fraction occurred (Fig. 4B). After incubation with ultrasonicated or thermally denatured [“Hladenine,’ BrdUrd-labelled DNA and unlabelled competitors only “light” products of recipient DNA synthesis were detected on neutral CsCi gradients (Fig.3A and B). After ultrasonication the intermediate peak disintegrated into two fractions of original donor DNA (1.741 g/ml) and intermediate fractions (1.721 g] ml) whereas products of bacterial DNA synthesis would be expected to be homogenous. The autoradioELK.J Biocheni. Sh (197.5)
41 1
V. Hemleben, N . Ermisch, D. Kimmich, B. Leber, and G. Peter
graphic studies showed labelling preferably in the cell nuclei and no radioactivity on the surface of the plant root tips arising from labelled bacteria. Other investigations of the uptake of homologous or beterologous DNA by human or animal cells in culture, using BrdUrd labelling of DNA have yielded conflicting results [9- 121. Ayad and Fox [9,10] using [3H]BrdUrd-labelled donor DNA and recipient cells with 14C-labelled DNA observed unintegrated undegraded donor DNA, integrated donor DNA with intermediate density and labelled recipient DNA. Kao et al. [12] found no intermediate fraction using [‘“CIBrdUrd-labelled fibroblasts treated with [3H]thymidine-labelled DNA provided that utilisation of [3H]thymidine from degraded donor DNA was excluded by addition of excess unlabelled thymidine. Ledoux et al. [6,18] obtained intermediate DNA fractions by treating barley, Sinapsis and Arabidopsis seeds with heterologous bacterial DNA of different buoyant densities. In our studies on the uptake of T4 phage DNA into seedlings of Matthiola we found an intermediate fraction suggesting an integration product between donor and recipient DNA [8]. Characterisation of these intermediate fractions by ultrasonication and alkali treatment allows the suggestion that doublestranded foreign D N A may be integrated into the recipient genome. Genetic experiments with bacterialDNA-treated mutants of Arabidopsis support this interpretation and a model of an “integrated episome” has been proposed [19]. Transformation studies on Drosophila [20,21], Neuvospova [22], and Petunia [23] and the investigations of Kao et al. [12] confirm the “exosome” model [20]. The predictions of this model are that the incorporated DNA is never integrated but becomes firmly associated with the homologous part of the chromosome while the original information is still present at that site. Our data presented here suggest that homologous DNA reaches the cell nucleus after uptake into the plants and becomes integrated into the recipient genome. The evidence suggests that the observed intermediate DNA fraction is neither a result of bacterial contamination nor a product of DNA synthesis of the recipient plants by semiconservative replication using degradation products of a BrdUrd-labelled donor DNA. Double-stranded segments of homologous
DNA seem to be integrated and covalently linked to recipient DNA molecules. The stability of the integration process and the translocation of the exogenous DNA within the plants are under further investigations. The technical assistance of Mrs Gisela Vetter in parts of the work is gratefully acknowledged. We thank Mr Arnhardt for cultivating the pure strains of Matthiola incana. We also thank Dr D. Grierson, University of Nottingham, for discussion of the results and assistance in the preparation of the manuscript.
REFERENCES 1. Johnson, C. B. & Grierson, D. (1974) Curr. Adv. Plant Sci. 9, 1- 12. 2. Hen, D. (1972) Naturwissenschaften, 59, 348 - 355. 3. Merril, C. R. & Stanbro, H. (1974) 2.Pfanzenphysiol. 72, 371 - 388. 4. Ohyama, K., Gamborg, 0. L. & Miller, R. A. (1972) Can. J . Bot. 50, 2077 - 2080. 5. Hoffmann, F. (1973) 2. Pfkmzenphysiol. 69,249- 261. 6. Ledoux, L., Huart, R. &Jacobs, M. (1971) Eur. J . Biochem. 23, 96- 108. 7. Bendich, A. J. & Filner, P. (1971) Mutation Research, 13, 199 - 214. 8. Rebel, W., Hemleben, V. & Seyffert, W. (1973) Z. Naturforschg. 28c, 473 - 474. 9. Ayad, S . R. &Fox, M. (1968) Nature (Lond.) 220,35-38. 10. Ayad, S. R. & Fox, M. (1971) in Informative Molecules in Biological Systems (Ledoux, L., ed.) pp. 99- 109, North Holland, Amsterdam. 11. Hill, M. & Hillova, J. (1971) in Informative Molecules in Biological Systems (Ledoux, L., ed.) pp. 113- 123, North Holland, Amsterdam. 12. Kao, P. C., Regan, J. D. & Volkin, E. (1973) Biochim. Biophys. Acta, 324, 1 - 13. 13. Ermisch, N. & Hemleben, V. (1974) Newletters Applied Nuclear Methods in Biology and Agriculture. 3, 7- 8. 14. Haut, H. F. &Taylor. J. H. (1967)J. Mol. B i d . 26,389- 401. 15. Hemleben, V. (1975) Symposium on Genetic Manipulations rith Plant Materials (Ledoux, L., ed.) in press. 16. Bendich, A. (1972) Biochim. Biophys. Acta, 272,494- 503. 17. Pearson, G. G. & Ingle, J. (1972) Cell Differentiation, I , 43- 51. 18. Ledoux, L. & Huart, R. (1969) J . Mol. Biol. 43, 243-262. 19. Ledoux, L. & Huart, R. (1974) Nature (Lond.) 249, 17-21. 20. Fox, A. S. & Yoon, S. B. (1970) Proc. Natl Acad. Sci. U.S.A. 67,1608- 1616. 21, Fox, A. S., Yoon, S. B. & Gelbart, W. M. (1971) Proc. Natl Acad. Sci. U S A . 68, 342-346. 22. Mishra, N. C . & Tatum, E. L. (1973) Proc. Natl Acad. Sci. U.S.A. 70,3875-3879. 23 HeB, D. (1970) 2. PJlanzenphysiol. 68,432-440,
V. Hemleben, N. Ermisch, D. Kimmich, B. Leber, and G. Peter, Institut fur Biologie I1 der Eberhard-Karls-Universitit Tubingen, Lehrstuhl fur Genetik, D-7400 Tubingen, Auf der Morgenstelle, Federal Republic of Germany
Eur. J . Biochem. 56 (1975)
Studies on the Fate of Homologous DNA Applied to Seedlings of Matthiola incana Vera HEMLEBEN, Nortrude ERMISCH, Dorothee KIMMICH, Birgit LEBER, and Gabriele PETER Department of Biology 11, University of Tiibingen (Received March 13iApril 21, 1975)
Seedlings of Mutthiolu incunu (crucifer) are able to take up exogenous homologous DNA by the roots. DNA homogenously labelled with [3H]adenine and 5-bromodeoxyuridine is incorporated into the plants in a macromolecular form. Intact donor DNA and a fraction with a buoyant density intermediate between that of the donor and the recipient DNA can be recovered. Analysis of this intermediate fraction by ultrasonication and alkali treatment allows the suggestion that homologous DNA is integrated as a double-stranded DNA which becomes covalently linked to the recipient DNA. Control experiments in which seedlings were incubated in a mixture simulating donor DNA degradation products in the presence and absence of unlabelled competitors suggest that these results are not due to the breakdown of donor DNA and reincorporation of the products during DNA synthesis in the recipient plants. When ultrasonicated or thermally denatured DNA is applied to the plants it may be degraded and reused for recipient DNA synthesis but it is not recovered in a macromolecular form. The possibility that the intermediate DNA fraction arises by bacterial contamination of the plants can be excluded by several arguments. Autoradiographic studies show that at least part of the radioactivity of the donor DNA taken up by the plants is associated with the cell nucleus.
There have been many reports that high-molecularweight DNA can be incorporated by plant protoplasts, intact cells and whole plants. These experiments involved the use of heterologous DNA from various bacteria and phages [l - 81. Identification of the foreign DNA occurred mostly by virtue of the different buoyant densities of donor and host DNA. In order to differentiate an exogenous homologous DNA from the recipient DNA by caesium chloride equilibrium centrifugation it is necessary to use donor DNA with an altered buoyant density. This can be obtained by the incorporation of 5-bromodeoxyuridine (BrdUrd) which increases the density of the DNA. If in addition a radioactive label is incorporated, this allows the detection of small amounts of the “heavy” DNA. Abbreviations. BrdUrd, 5-bromodeoxyuridine; DNase, deoxyribonuclease I; RNase, ribonuclease A. Enzymes. Deoxyribonuclease I (EC 3.1.4.5); pronase (EC 3.4.21.4 and 3.4.24.4); ribonuclease A (EC 3.1.4.22).
Eur. J. Biochem. 56 (1975)
Uptake of BrdUrd-labelled homologous DNA in a macromolecular form by animal or human cells in culture has been shown [9- 121. In this kind of experiment it is necessary to distinguish between the uptake of intact exogenous DNA and the reutilisation of labelled or BrdUrd-containing degradation products of the exogenous DNA for recipient DNA synthesis. This is possible by including unlabelled DNA precursors in the incubation medium. Addition of a large amount of unlabelled thymidine should prevent the utilisation of BrdUrd from degraded BrdUrd-labelled DNA. The incorporation of radioactive degradation products can also be prevented by employing the appropriate competitor. In this report we used a [3H]thymidine or [3H]adenine-BrdUrd-labelled DNA from Mutthiolu incunu as donor DNA and unlabelled seedlings of Mutthiolu as recipients. The uptake and fate of the DNA was studied by measuring the amount of radioactivity within the plants, by autoradiography and caesium
404
chloride equilibrium centrifugation of the recipient DNA.
MATERIALS AND METHODS Growth and Labelling o j the Plants
Batches of 200-400 seeds of Mutthiolu incanu (strain 17 or 19, Department of Genetics, University of Tubingen) were sterilised by shaking with Orthocid (Merck, Darmstadt) for 10 min, washed several times with sterile water and grown in glass vials (15 x 65 mm), each containing 5 seeds in 200 p1 distilled water, in continuous light of 500 foot candles (5381 lx) at 20 "C. To obtain radioactively labelled or radioactively and density labelled homologous donor DNA the washed seeds were incubated in solutions of [6-3H]thymidine (15 pCilml; specific activity : 26 Ci/mmol) or [8-3H]adenine (15 pCi/ml; specific activity: 2427 Ci/mmol) and 5-bromodeoxyuridine (500 pg/ml). After 8 days the plants were harvested and the DNA extracted. The specific activity was generally between 2- 10 x lo6 counts min-' mg D N A - ~ . Incubation of (he Seedlings with Donor D N A
7-day-old seedlings were incubated (5 seedlings in 200 p1 solution) with sterile dialysed DNA preparations labelled with [3H]thymidine or [3H]adenine and BrdUrd (100 pgiml; about 5 x lo5 counts min-' ml-'), together with unlabelled DNA precursors (1 mg adenine and thymidine/ml) dissolved in 0.01 M KC1-0.001 M sodiumcitrate buffer pH 6.7 for various lengths of time in continuous light at 20 "C. Only the roots were in contact with the solution. The incubation mixture was plated on nutrient broth agar plates for bacterial counts. All steps were carried out under sterile conditions. The recipient plants were then immediately harvested, washed and treated twice with DNase I (20 pg/ml) for 30 min at 37 "C and 1 h with pronase (50 pgiml) at 37 "C to remove the exogenous DNA from the surface and the DNA was extracted from the seedlings. In some experiments the seedlings were washed with sterile water after incubation with donor DNA and reincubated for a further period of time with distilled water and unlabelled precursors before DNase treatment and DNA extraction. Isolation of' D N A
The seedlings were gently homogenised with a glass tube and pestle with 0.1 M Tris - 0.005 M EDTAHCl buffer pH 7.5 and sodium dodecyl sulfate (1 final concentration). The homogenate was shaken
Fate of Homologous DNA in Seedlings of Mutthiolo
with 1 volume of phenol mixture (88 ml phenol, water saturated, 12 ml m-cresol, 0.1 g 8-hydroxyquinoline) for 20 min. After centrifugation (Minifuge, Christ, 10 min, 5000 rev./min, 4 'C) the aqueous phase was removed by a syringe with a 3-mm dsameter needle. The phenol phase was reextracted with TrisEDTA-HCl buffer, the aqueous phases were combined and again shaken with half the volume of phenol mixture and centrifuged. The final aqueous phase was precipitated with two volumes of ethanol for at least 3 h at - 28 "C. The alcohol precipitate was dissolved in Tris-EDTA-HC1 buffer, treated with RNase A (20 pg/ ml) for 2 h and with pronase (100 pgIml) for 2 h at 37 "C. The DNA was further purified by shaking with 1 volume of chloroform - isoamyl alcohol (10 : 1) for 30 min, followed by centrifugation. This treatment was repeated twice with the supernatant and the DNA was precipitated with 2 volumes of ethanol at - 28 "C. For CsCl gradient centrifugation the DNA was dissolved in 1 ml Tris-EDTA-HC1 buffer. Incubation experiments were carried out with DNA dialysed against 0.01 M KCl-0.001 M sodium citrate buffer pH 6.7 for 24 h under sterile conditions. The A260/A280 ratio was about 1.85- 2.0. The specific activity was measured by counting 10 pl of a DNA sample in 2 ml Bray's solution in a liquid scintillation counter (Packard, Model 3390). Cuesium Chloride Equilibrium Centr ifugat ion
DNA samples dissolved in Tris-EDTA-HC1 buffer were fractionated on neutral or alkaline (pH 12.5) CsCl gradients. The density was adjusted to 1.74 g/ml for neutral gradients and 1.81 gjml for alkaline gradients. Samples (2.5 or 3.0 ml) were centrifuged for 45 h a t 33 000 rev./min in the SW 65 rotor of a Beckman Model L 2 preparative ultracentrifuge at 20 " C . The tubes were punctured from the bottom and 60-70 three-drop fractions were collected per gradient. Every tenth fraction was used for measurements of the refractive index; to every second fraction I ml of Tris-EDTA-HC1 buffer was added for measurement of the absorbance. To every other remaining fraction 1 ml yeast RNA (250 pg/ml) was added as carrier and the nucleic acids were precipitated with 1 ml 10% trichloroacetic acid at 0 "C. The precipitated samples were collected on Sartorius membrane filters (0.45 pm, 24-mm diameter), washed twice with 3 "/, trichloroacetic acid and put in new scintillation vials with 2 ml toluene scintillator solution. The radioactivity was measured in a liquid scintillation counter. Samples were counted for 10-20 min and corrected for background radiation. After separation on CsCl gradients certain fractions were combined, placed in dialysis bags, dialysed Eur. J Biochem. 56 (1975)
V. Hemleben, N . Ermisch, D. Kimmich, B. Leber, and G. Peter
405
Fraction number Fig. 1. Characterisation of [ 3H]adeninelBrdUrd-labellrd homologous Matthiola D N A by CsCI gradient centrifugation. DNA was isolated from seedlings grown in [3H]adenine (15 pCi/ml) and BrdUrd (500 pg/ml) for 8 days. 10 pg labelled D N A were separated together with unlabelled Matthiola D N A as reference (A) on a neutral CsCl gradient, (C) on an alkaline CsCl gradient (pH 12.5). Ultrasonicated labelled DNA was fractionated (B) on a neutral and (D) on an alkaline CsCl gradient. (Rotor SW 65, 33000 rev./min, 45 h, 20 "C.) ( 0 4 ) Absorbance at 260 nm; ( o - - - o ) 3H radioactivity
and concentrated in a solution of 30 % (w/w) Dextran T 70 (Pharmacia, Uppsala). After concentration these fractions were either ultrasonicated for 2 min (50 W ; sonifier Branson) and fractionated again on neutral CsCl gradients or separated on alkaline gradients.
Autoradiography
7-day-old seedlings were incubated with [3H]thymidine-labelled donor DNA ( 5 x lo5 counts min-' m1-I) and unlabelled thymidine (500 pg/ml) dissolved in 0.01 M KCl-0.001 M sodium citrate buffer for 24 h under sterile conditions and treated with DNase I and pronase as described above. The root tips of some plants were fixed, Feulgen stained and used for squash preparations for autoradiography. From the remaining seedlings the nuclei were isolated. The isolated nuclei were fixed with methanol/formalin/glacial acetic acid (85 : 15 : 5 ; v/v/v), distributed on slides and stained by Feulgen staining after hydrolysis with 5 N HCI for 1 h at room temperature. Kodak AR 10 stripping film was applied to the slides. The exposure time was between 30- 90 days at 4 "C. The autoradiograms were developed with Kodak D-l9b, fixed with Kodak Unifix and used for microphotographical investigations. Eur. J. Biochem. 56 (1975)
Isolation of Nuclei
Seedlings were gently homogenised in a medium containing 0.05 M Tris, 0.01 M MgCI,, 0.2 M sucrose and 2.5% Ficoll (Pharmacia, Uppsala), pH 7.8. The homogenate was filtered successively through nylon grids (300, 100, 30 and 10 pm pore diameter). The 10-pm filtrate was centrifuged at 1000 x g for 10 min at 2 "C (Minifuge, Christ) and washed once with homogenisation buffer. The sediment containing the intact nuclei and some chloroplasts was used for autoradiographic studies. RESULTS Characterisation of Native and BrdUrd-Labelled Matthiola D N A
Unlabelled DNA isolated from seedlings of Matthiola incana has a buoyant density of 1.698 g/ml when separated on a neutral CsCl gradient (Fig. 1A). If the seedlings are grown in a solution of 5-bromodeoxyuridine (500 pg/ml) and simultaneously labelled with [3H]adenine for 8 days the DNA buoyant density shifts to 1.741 g/ml (Fig. 1A) [13] and the DNA has a specific activity of about 2- 10 x lo6 counts min-' mg-'. Previous experiments had shown that at lower
406
Fate of Homologous D N A in Seedlings of Mutthiolu
,
I
20
I
I
LO
1
I
I
60
Fraction number
Fig. 2. CM’I grudient cenlrifugution y f D N A isolutedfiom Matthiola seedlings incubated with liomologous D N A . (A) Neutral CsCl gradient centrifugation of DNA isolated froin 8-day-old seedlings previously incubated for 20 h with [3H]adenine/BrdUrd-labelled D N A (100 pg ml) and unlabelled adenine and thymidine (1 mg/ml each) as competitors. (B) Neutral CsCl gradient centrifugation of DNA isolated from 8-day-old secdlings grown in BrdUrd (500 pg/ml) and previously incubated with “light” [3H]adeninei[3H]guanosine-labclledDNA and unlabelled precursors as competitors for 20 h. The recipient plants were treated with DNase and pronase before the isolation of the DNA. Centrifugation was as described for Fig.1. In (B) unlabelled ‘‘light’’ Mutthiolu D N A was added as reference. (O---o) Absorbance at 260 nm; (0- - 0 ) ’H radioactivity ~
concentrations of BrdUrd in the growth medium only part of the DNA shifts to a higher density. Using higher doses of BrdUrd (200- 1000 pg/ml) the density shift is nearly constant at about 0.043 g/ml. N o higher incorporation of BrdUrd was observed by attempts to inhibit thymidine kinase with 5-fluordeoxyuridine [14]. Ultrasonicated [’H]adenine/BrdUrd-labelled DNA shows a broad peak with a maximum of radioactivity at the original buoyant density when fractionated on a neutral CsCl gradient (Fig. 1 B). The ultrasonicated DNA fractionated on an alkaline CsCl gradient appears as a broad peak with a maximum at a density of 1.85 glml (Fig. 1 D). Unsonicated DNA samples separated on an alkaline CsCl gradient band at the same density (Fig.1C). These experiments show that the distribution of BrdUrd within the [3HH]adenine!BrdUrd-labelledDNA molecules is homogenous. The criteria required for studies on uptake and identification of a homologous DNA within the recipient, i . r . difference of buoyant density between donor and recipient DNA, homogenity of the BrdUrd distribution and high specific activity of the donor DNA, are therefore fulfilled.
Effect of BrdUrd on Matthiola Seedlings High concentrations of BrdUrd in the growth medium affect the growth and development of the plants. The BrdUrd-labelled plants (500 pg BrdUrd/ ml) are smaller and show no root hairs and the DNA content per seedling is decreased. For uptake experiments where the donor DNA is probably taken up by
the roots and root hairs [3H]adenineiBrdUrd-labelled DNA was mostly used as donor DNA and normally growing “light” seedlings as recipients. Incubation of Matthiola Seedlings tvith Honzologous D N A
Preliminary studies showed that seedlings of Mutthiolu take up exogenous DNA irrespectively of their developmental stage between the 4th and the 8th day after germination. In subsequent experiments incubations were carried out with 7-day-old seedlings with fully developed root hairs. Uptake studies showed radioactivity in the recipient plants after 5 - 10 h of incubation with t3H]adenine/BrdUrdlabelled DNA but greater incorporation was obtained if the incubation period lasted between 16 and 20 h. Between 0.1 and 1 ”/, of the applied DNA was taken up by the seedlings. After longer incubation periods the donor D N A ( M , = 2- 4 x 10‘) had a reduced size of about 2-5 x lo5. Batches of 7-day-old unlabelled seedlings were incubated with [3H]adenine/BrdUrd-labelled-DNA (100 pg/ml; specific activity: 5 x 10’ counts min-’ mg DNA- ’) and unlabelled competitors thymidine and adenine (1 mg/ml each) in KCl- sodium citrate buffer for 20 h. The DNA was isolated from the DNase and pronase-treated plants and fractionated on a neutral CsCl gradient (Fig.2A). Two main peaks of radioactivity were separated with buoyant densities of about 1.741 and 1.717 g/ml. Very small amounts of radioactivity coincided with the “light” recipient DNA The peak with a density of 1.741 gjml corresponds to F u r J Biochem 56 (1975)
V. Hemleben, N. Ermisch, D. Kimmich, B. Leber, and G. Peter
407 80 1
c
E
60
0
v1 C
(D N
c
0.10
8
0 a,
LO
c
e
r .5 ._ c
0
0.05
20
Q
7 cc
0 0
20
LO
60 Fraction number
20
LO
60
Fig. 3. CsCl gradient centrifugation of D N A isolated from Matthiola Jeedlings incubated with ultrasonicated or heat-denatured DNA. Neutral CsCl gradient centrifugation of DNA isolated from 8-day-old unlabelled seedlings previously incubated for 20 h with (A) ultrasonicated and (B) heat-denatured [3H]adenine/BrdUrd-labelledDNA (100 pg/ml) and unlabelled adenine and thymidine as competitors. Centrifugation was as described for Fig. 1. (0--0) Absorbance at 260 nm; (O----+) 3H radioactivity
the intact donor DNA, the second peak shows an intermediate density between donor and recipient DNA. The radioactivity in the DNA of the recipient plants indicates a certain breakdown of donor DNA and reutilisation of the products for host DNA synthesis [15]. Similar results were obtained when “light” [3H]adenine/[3H]guanosine-labelledDNA, together with unlabelled adenine, guanosine and BrdUrd as competitors of donor DNA degradation products, was applied to recipient plants in which the DNA had been previously labelled with BrdUrd (500 pg/ml). DNA isolated from these plants after 20 h of incubation and separated on a neutral CsCl gradient showed three peaks of radioactivity (Fig. 2B), representing applied donor DNA (buoyant density: 1.698 g/ml), an intermediate fraction (1.719 g/ml) and a small amount of radioactivity with the density of the “heavy” recipient DNA indicating very slight reincorporation of degradation products. Recipient plants were incubated with [3H]adenine/ BrdUrd-labelled DNA as described for Fig. 2A except that the plants were removed from the DNA-containing medium after 2 min. After treatment of the plants with DNase and pronase and isolation of the DNA no radioactivity was found in the DNA preparation. This indicates that the exogenous DNA is completely removed from the surface of the seedlings after treatment with DNase. Therefore the peaks of radioactivity with the density of 1.741 g/ml in Fig. 2A and with the density of 1.698 giml in Fig.2B represent intact donor DNA present in the recipient plants but inaccessible to DNase. No peaks of radioactivity with the density of intact donor DNA or intermediate density appeared within the recipient plant DNA if the donor DNA applied Eur. J. Biochem. 56 (1975)
to the seedlings was ultrasonicated ( M , = 5 x lo5) or heat-denatured to single-strand DNA before incubation with the plants. The DNA isolated from seedlings treated in this way showed radioactivity only in the “light” peak (Fig. 3A, B) when separated on neutral CsCl gradients indicating that small fragments of double-stranded DNA or single-stranded DNA are degraded by the recipient nucleases and reutilised in the DNA synthesis. However, in the presence of unlabelled competitors it is not possible to compare the amount of native, ultrasonicated and denatured DNA taken up by the plants. These experiments show that the occurrence of (a) DNase-resistant donor DNA in the plants and (b) the intermediate DNA peak depends on the donor DNA being of high molecular weight and doublestranded. The Incorporation of Radioactivity by Recipient Plants Utilising Donor D N A Precursors Two batches of 7-day-old unlabelled seedlings were incubated for 20 h in a medium simulating the application of a degraded [3H]adenine/BrdUrd-labelled donor DNA. The incubation medium contained [3H]adenine (lo6 counts min-’ ml-l) and BrdUrd (50 pg/ml) corresponding to the [3H]adenine and BrdUrd content of the donor DNA usually applied to the plants (see Fig. 4A) and in one case unlabelled adenine and thymidine (500 pg/ml) was added. The DNA was then isolated and separated on neutral CsCl gradients (Fig.4A and B). In the absence of the BrdUrd competitor thymidine and unlabelled adenine the newly synthesised radioactive DNA is heavy, with a mean density of 1.739 g/ml although some less dense DNA fractions are also detectable (Fig. 4A). In the presence
408
Fate of Homologous DNA in Seedlings of Mutihioh
Fraction number
Fig. 4. Tllc i n c . o ~ ~ ~ o ~ uof t i oBrdC‘rd ii und [”H/udenine into Matthiola D N A it? the presence and ub.wnce of unluhellrd coiwpetitor.~.7-daj-old seedlings cultivated in distilled water were treated for 20 h with 50 pg BrdUrdiml and [3H]adenine (10’ counts min-’ inl~-’)(A) without and (B) with unlabelled adenine and thymidine (500 pgirnl each). The D N A was isolated and separated on a neutral CsCl gradient as described in the legend to Fig. 1. (0- 0)Absorbance at 260 nm; (o-----o) 3H radioactivity -
of an excess of thymidine and adenine the radioactive peak is completely shifted to the “light” region of the gradient (1.698 giml) without any intermediate fractions (Fig. 4 B). In addition, the presence of unlabelled adenine as competitor of the [3H]adeninedramatically reduces the incorporation of radioactivity into DNA (compare F i g 4 A and B). These results allow us to conclude that the radioactivity appearing in the recipient plants after a certain incubation time with [3H]adenine/BrdUrd-labelled donor DNA (Fig. 2A) results at least partly from donor DNA in an undegraded form.
Density(g/ml)
I
1.713
0.15
$
60
5?0.10
vi
N
c
t 7
I
0
0 U
LO ;
8
c 2 c
0
f
U
0.05
; 0
20
Q
0
Stabilitj~of thc Incorporated Donor D N A
The fate of the exogenous DNA after uptake into the plants was studied in the following way. Seedlings were incubated with [3H]adenine/BrdUrd-labelled DNA as described in the legend to Fig.2A for 20 h with unlabelled adenine and thymidine as competitors, then the DNA was removed and the plants were allowed to grow for a further period of 24 h in distilled water with unlabelled adenine and thymidine. The DNA from these plants was then isolated and fractionated on a neutral CsCl gradient (Fig.5). Under these conditions the fraction with a density of 1.741 g/ml corresponding to donor DNA, disappears (compare with Fig. 2A) whereas the intermediate fraction with a density of about 1.713g/ml appears to be more stable. Characterisation
of
the Intermediate D N A Fraction
The studies of uptake of homologous DNA show that the exogenous DNA enters the plant cells in a
0
20
40 Fraction number
60
Fig. 5. l h e “disuppeurance” ~f homologous [‘H]adenine,’BrrlUrdlabelled D N A f r o m Matthiola seedlings during u chase incuhution. 7-day-old unlabelled seedlings were incubated with [3H]adenine, BrdUrd-labelled D N A and unlabelled competitors for 20 h, washed and incubated in distilled water with adenine and thymidine (1 mg/ml each) for a further 24 h. D N A was extracted and fractionated on a neutral CsCl gradient as described in the legend to Fig.1. (0 0) Absorbance at 260nm; (0 - - ~ - O )‘H radioactivity
macromolecular form. However, DNA of donor density does not persist although a fraction of intermediate density is stable (Fig. 5). It is possible that the intermediate DNA fraction results from the integration of donor DNA into the “light” recipient DNA. The intermediate fraction was therefore analysed by mechanical shearing or alkali treatment following the methods of Ledoux et al. [6]. The dialysed, concentrated fractions of the intermediate fraction formed after incubation of the seedlings with “heavy” donor DNA for 20 h (Fig. 2A, fractions 31 - 39) were ultrasonicated Eur. J. Biochem. 56 (1975)
409
V. Hemleben, N. Ermisch, D. Kimmich, B. Leber, and G. Peter Density( glml)
30
I
. G E
E
6 0.10
w
10 C
c
20
c
8
DNA bands at the expected density of 1.762 g/ml (Fig. 1C). Heavy BrdUrd-labelled DNA would band at a density of 1.853 g/ml if separated in alkaline CsCl (Fig. 1C). This results suggest that donor DNA is covalent bound to light recipient DNA.
0
x
c
s
0.05
c
0
0 10
10 9
n
Localisation of Exogenous D N A after Uptake into the Seedlings
-0
Q
LY
0 20
0
LO Fraction number
B
60
Density(g/rnl)
Fraction number Fig. 6. Characterisation oj the intermediate D N A fraction. (A) DNA from fractions 31-39 of the CsCl gradient shown in Fig.2A was concentrated, ultrasonicated and refractionated on neutral CsC1. Unlahelled Marzhiola DNA was added as reference. (B) DNA from fractions 31 -41 of the CsCl gradient shown in Fig. 5 was concentrated and refractionated without ultrasonication on alkaline CsCl (pH 12.5). Unlahelled Mutthiola DNA was added as reference. +) Absorbance at Centrifugation as described for Fig. 1. (0 -~~ 260 nm; ( C - - O ) 3H radioactivity
and refractionated on a neutral CsCl gradient (Fig. 6 A). This apparently homogenous peak now splits up into at least two fractions of radioactivity, one with the same density as the “heavy” donor DNA (1.741 g/ml) and the other one with an intermediate density (1.721 g/ml). This suggests that the intermediate peak contained relatively high-molecularweight fragments of double-stranded donor DNA which were released by ultrasonication. The corresponding release of “light” DNA from the intermediate peak cannot be directly demonstrated or differentiated from the unlabelled reference DNA because such DNA would not be labelled. The intermediate peak obtained after chase conditions (Fig. 5, fractions 31 -41) was refractionated on an alkaline CsCl gradient with unlabelled Matthiola DNA as reference (Fig. 6B). Under these conditions the radioactive peak is still homogenous but is shifted to a density of 1.781 g/ml while the unlabelled reference Eur. J. Biochem. 56 (1975)
Batches of 7-day-old seedlings were incubated with [3H]thymidine-labelled DNA (5 x lo5 counts min-’ m1-I) for 24 h with a large excess of unlabelled thymidine to prevent the reincorporation of 3Hlabelled degradation products. In control experiments plants were incubated with [3H]thymidine (5 x lo5 counts min-’ m1-I) and unlabelled thymidine. The plants were then treated with DNase and pronase. Some of them were fixed, Feulgen stained and used for root-tip squash preparations. From the remaining plants the nuclei were isolated, fixed and distributed on slides and Feulgen stained. The squashed root tips and the isolated nuclei were analysed by autoradiography. After incubation in [3H]thymidine for 24 h most of the nuclei in the elongating zone of the root were labelled. In contrast much less radioactivity was associated with the roots when it was supplied as [3H]thymidine-labelled DNA. Nevertheless, many of the silver grains were localised over nuclei and there was no evidence for the accumulation of radioactivity at the root epidermis or in the cell walls. Preparations of isolated nuclei show that at least part of the radioactivity becomes associated with nuclei (Fig. 7). The results of the autoradiographic investigations are consistent with the evidence described above which suggests that exogenous homologous DNA may become associated with the recipient genome. DISCUSSION These results show that exogenous homologous DNA is incorporated into seedlings of Matthiola incana. After short incubations all the exogenous DNA can be removed from the plants by treatment with DNase but subsequently DNA taken up by the seedlings becomes resistant to this treatment. After 20-h incubation [3H]adenine/BrdUrd-labelled DNA with the same buoyant density as the original donor DNA can be recovered from the seedlings together with a second DNA fraction with a buoyant density intermediate between that of the donor and the recipient. The donor DNA and the intermediate peak are only detected in recipient plants if the applied DNA is of high molecular weight ( M , > lo6) and double-
410
Fate of Homologous DNA in Seedlings of A4ut//iio/u
Fig. 7. A ~ / i o i . ~ r t / i ~ , ~ i . t o/ r i ~ Isohted /i~. rruciei. 7-day-old seedlings were treated for 24 h with [’Hlthymidine-labelled homologous DNA (100 pg:ml; I 5 x lo5 counts min-’ ml-’) and unlabelled thymidine as competitor. Isolation and preparation ofthe nuclei and autoradiography was dons a s described in Materials and Methods. Three representative nuclei are shown
stranded. Chase incubations show that the donor DNA does not persist in its native form but is converted into the intermediate density DNA. The apparent decrease in buoyant density of the donor DNA (0.034-0.028 giml) could be explained if it becomes attached to DNA of much lower density. This interpretation is supported by the fact that DNA segments of original donor DNA density can be recovered from the intermediate fraction by ultrasonication. The donor DNA appears to be covalently attached in a double-stranded form because no single-stranded DNA of donor density is obtained following the recentrifugation of unsonicated intermediate density DNA in alkaline CsCl. The density of the intermediate DNA fraction after 16- 20 h of incubation of the recipient plants with [3H]adenine/BrdUrd-labelled DNA varies from experiment to experiment from 1.727 to 1.717 g/ml [I51 and the peak is not always as sharp as shown in Fig. 2A. That means that the integrated donor fragments are of difTerent size. A DNA fraction with almost intermediate density between donor and recipient DNA must consist of a relatively similar amount of light recipient DNA (molecular weight of the DNA preparation about 2- 5 x lo6)and fragments of integrated heavy DNA. Two further possibilities of interpretation of the intermediate fraction should be discussed. (a) The donor DNA enters the cell and serves as a template for endogenous DNA polymerases. This would be expected to produce one light strand and one heavy strand. However, the intermediate (or “hybrid”) fraction obtained contains double-stranded fragments of original donor density which is inconsistent with such an interpretation. (b) BrdUrd-labelled donor DNA becomes associated with homologous sites in
the recipient genome in a process of single-strand recombination forming “hybrids”. This possibility can also be excluded by the ultrasonication experiment mentioned above and by the results of alkali treatment of the intermediate fraction after chase conditions : only single strands with an intermediate density are separated on an alkaline CsCl gradient. These results support the interpretation of an integrated doublestranded donor DNA segment with light recipient DNA similar to Ledoux’s observations with heterologous DNA integrated in plant DNA [6]. In this type of study it is very important to rule out the possibility that the DNA with intermediate density is derived from contaminating bacteria. Bendich [16] showed that labelling of contaminated plants results in “false” D N A satellites with intermediate densities. Other DNA satellites with similar buoyant densities to the intermediate fraction described here originate from bacterial contamination [17]. Our experiments were carried out under aseptic conditions and seedlings and DNA samples were routinely monitored for bacterial growth. In addition, in experiments simulating the incubation conditions with completely degraded [3H]adenine/BrdUrd-labelled donor DNA by adding the corresponding amount of [‘HH]adenine and BrdUrd and unlabelled competitors no intermediate fraction occurred (Fig. 4B). After incubation with ultrasonicated or thermally denatured [“Hladenine,’ BrdUrd-labelled DNA and unlabelled competitors only “light” products of recipient DNA synthesis were detected on neutral CsCi gradients (Fig.3A and B). After ultrasonication the intermediate peak disintegrated into two fractions of original donor DNA (1.741 g/ml) and intermediate fractions (1.721 g] ml) whereas products of bacterial DNA synthesis would be expected to be homogenous. The autoradioELK.J Biocheni. Sh (197.5)
41 1
V. Hemleben, N . Ermisch, D. Kimmich, B. Leber, and G. Peter
graphic studies showed labelling preferably in the cell nuclei and no radioactivity on the surface of the plant root tips arising from labelled bacteria. Other investigations of the uptake of homologous or beterologous DNA by human or animal cells in culture, using BrdUrd labelling of DNA have yielded conflicting results [9- 121. Ayad and Fox [9,10] using [3H]BrdUrd-labelled donor DNA and recipient cells with 14C-labelled DNA observed unintegrated undegraded donor DNA, integrated donor DNA with intermediate density and labelled recipient DNA. Kao et al. [12] found no intermediate fraction using [‘“CIBrdUrd-labelled fibroblasts treated with [3H]thymidine-labelled DNA provided that utilisation of [3H]thymidine from degraded donor DNA was excluded by addition of excess unlabelled thymidine. Ledoux et al. [6,18] obtained intermediate DNA fractions by treating barley, Sinapsis and Arabidopsis seeds with heterologous bacterial DNA of different buoyant densities. In our studies on the uptake of T4 phage DNA into seedlings of Matthiola we found an intermediate fraction suggesting an integration product between donor and recipient DNA [8]. Characterisation of these intermediate fractions by ultrasonication and alkali treatment allows the suggestion that doublestranded foreign D N A may be integrated into the recipient genome. Genetic experiments with bacterialDNA-treated mutants of Arabidopsis support this interpretation and a model of an “integrated episome” has been proposed [19]. Transformation studies on Drosophila [20,21], Neuvospova [22], and Petunia [23] and the investigations of Kao et al. [12] confirm the “exosome” model [20]. The predictions of this model are that the incorporated DNA is never integrated but becomes firmly associated with the homologous part of the chromosome while the original information is still present at that site. Our data presented here suggest that homologous DNA reaches the cell nucleus after uptake into the plants and becomes integrated into the recipient genome. The evidence suggests that the observed intermediate DNA fraction is neither a result of bacterial contamination nor a product of DNA synthesis of the recipient plants by semiconservative replication using degradation products of a BrdUrd-labelled donor DNA. Double-stranded segments of homologous
DNA seem to be integrated and covalently linked to recipient DNA molecules. The stability of the integration process and the translocation of the exogenous DNA within the plants are under further investigations. The technical assistance of Mrs Gisela Vetter in parts of the work is gratefully acknowledged. We thank Mr Arnhardt for cultivating the pure strains of Matthiola incana. We also thank Dr D. Grierson, University of Nottingham, for discussion of the results and assistance in the preparation of the manuscript.
REFERENCES 1. Johnson, C. B. & Grierson, D. (1974) Curr. Adv. Plant Sci. 9, 1- 12. 2. Hen, D. (1972) Naturwissenschaften, 59, 348 - 355. 3. Merril, C. R. & Stanbro, H. (1974) 2.Pfanzenphysiol. 72, 371 - 388. 4. Ohyama, K., Gamborg, 0. L. & Miller, R. A. (1972) Can. J . Bot. 50, 2077 - 2080. 5. Hoffmann, F. (1973) 2. Pfkmzenphysiol. 69,249- 261. 6. Ledoux, L., Huart, R. &Jacobs, M. (1971) Eur. J . Biochem. 23, 96- 108. 7. Bendich, A. J. & Filner, P. (1971) Mutation Research, 13, 199 - 214. 8. Rebel, W., Hemleben, V. & Seyffert, W. (1973) Z. Naturforschg. 28c, 473 - 474. 9. Ayad, S . R. &Fox, M. (1968) Nature (Lond.) 220,35-38. 10. Ayad, S. R. & Fox, M. (1971) in Informative Molecules in Biological Systems (Ledoux, L., ed.) pp. 99- 109, North Holland, Amsterdam. 11. Hill, M. & Hillova, J. (1971) in Informative Molecules in Biological Systems (Ledoux, L., ed.) pp. 113- 123, North Holland, Amsterdam. 12. Kao, P. C., Regan, J. D. & Volkin, E. (1973) Biochim. Biophys. Acta, 324, 1 - 13. 13. Ermisch, N. & Hemleben, V. (1974) Newletters Applied Nuclear Methods in Biology and Agriculture. 3, 7- 8. 14. Haut, H. F. &Taylor. J. H. (1967)J. Mol. B i d . 26,389- 401. 15. Hemleben, V. (1975) Symposium on Genetic Manipulations rith Plant Materials (Ledoux, L., ed.) in press. 16. Bendich, A. (1972) Biochim. Biophys. Acta, 272,494- 503. 17. Pearson, G. G. & Ingle, J. (1972) Cell Differentiation, I , 43- 51. 18. Ledoux, L. & Huart, R. (1969) J . Mol. Biol. 43, 243-262. 19. Ledoux, L. & Huart, R. (1974) Nature (Lond.) 249, 17-21. 20. Fox, A. S. & Yoon, S. B. (1970) Proc. Natl Acad. Sci. U.S.A. 67,1608- 1616. 21, Fox, A. S., Yoon, S. B. & Gelbart, W. M. (1971) Proc. Natl Acad. Sci. U S A . 68, 342-346. 22. Mishra, N. C . & Tatum, E. L. (1973) Proc. Natl Acad. Sci. U.S.A. 70,3875-3879. 23 HeB, D. (1970) 2. PJlanzenphysiol. 68,432-440,
V. Hemleben, N. Ermisch, D. Kimmich, B. Leber, and G. Peter, Institut fur Biologie I1 der Eberhard-Karls-Universitit Tubingen, Lehrstuhl fur Genetik, D-7400 Tubingen, Auf der Morgenstelle, Federal Republic of Germany
Eur. J . Biochem. 56 (1975)
