Download Estimation of Circular DNA Size Using r

177,110-114
ANALYTICALBIOCHEMISTRY
(1989)
Estimation of Circular DNA Size Using r-Irradiation
Pulsed-Field Gel Electrophoresis
M. Beverley
Stephen
Department
Received
and
of Biological
September
Chemistry
and Molecular
Pharmacology,
Harvard
Medical
School, Boston, Massachusetts
02115
7,1988
A method is described
for estimating
the size of large
circular
DNAs
found
within
complex
chromosomal
DNA preparations.
DNAs are treated
with low levels
of y-irradiation,
sufficient
to introduce
a single doublestranded
break per circle, and the resulting
linear DNA
is sized by pulsed-field
electrophoresis
and blot hybridization.
The method is fast, reproducible,
and very conveniently
applied
to the agarose-enclosed
chromosomal
DNA
preparations
commonly
used in pulsed
field
electrophoresis.
0 1989 Academic
Press, Inc.
lish this relationship in the buffers employed in this
work. Moreover, the effect of enclosing the DNA within
agarose sample plugs must also be determined, as this
method is commonly employed in the preparation of
large chromosomal DNA suitable for pulsed-field electrophoresis. These variables were examined and the results allow the estimation of the y-irradiation dose appropriate for examining circular DNAs of differing sizes.
METHODS
DNAs
Large circular DNAs are constituents of many genomes, including those of viruses, prokaryotes, and eukaryotes (1). Certain DNA amplifications in drug-resistant Leishmania
((2-5); T. E. Ellenberger and S. M. Beverley, in preparation) and cultured mammalian cells (69) also exist as circular DNAs. We have recently employed a new method for rapidly determining the size
of large circular DNAs in complex chromosomal DNA
preparations (5). In this approach, y-irradiation is used
to introduce limited breakage into chromosomal DNA
prepared within agarose sample plugs, which converts
supercoiled circular DNAs successively into open circular and then linear molecules. This approach is based on
the classic method employing limited DNAse cleavage of
circular DNAs. The size of the linearized DNA is then
determined by separation with pulsed-field electrophoresis, followed by blot hybridization
and comparison
with appropriate linear DNA size markers.
In this report several parameters important in the application of this method are examined. The effects of yirradiation on DNA have been extensively studied (summarized in (10)). y-Irradiation
introduces a variety of
lesions into DNA, including both single- and doublestranded breaks. The dose-dependence of the introduction of strand breaks into DNA is known to vary with
different buffers (lo), and thus it was necessary to estab110
X DNA (cI857Sam7) was purchased from New England Biolabs. The 33-kb circular plasmid pK300 has
been described (11). pK300 contains the entire 30-kb circular R region (2), which includes the bifunctional dihydrofolate reductase-thymidylate
synthase gene (14,X)
which is amplified in certain methotrexate-resistant
lines of Leishmania
major (2). Agarose samples containing DNA of Escherichia
coli cells bearing pK300 were
prepared as described (12). Chromosomes of the RlOOO11 line were prepared in agarose plugs as described and
stored at 4°C in 200 mM Tris, 100 mM EDTA, pH 8.0
(storage buffer; (11)).
y-Irradiation
y-Irradiation was performed using an ICN GR9 irradiator, which employs a 6oCosource. DNAs in solution
or agarose sample plugs were placed in a standard 1.5ml polypropylene microcentrifuge tube and then irradiated at ambient temperatures for the time necessary to
obtain the desired radiation doses. Samples remained at
room temperature for at least 30 min prior to electrophoresis.
Puked-Field
Electrophoresis
Pulsed-field electrophoresis was performed using the
method of Chu et al. (13), which is referred to as a con0003-2697/89
$3.00
Copyright
0 1989 by Academic
Press, Inc.
All rights of reproduction
in any form reserved.
SIZING
OF
CIRCULAR
DNA
BY
PULSED-FIELD
111
ELECTROPHORESIS
tour-clamped
homogeneous electric field (CHEF)l
electrophoresis.
Electrophoresis
was performed
in 0.5X
TBE buffer (45 mM Tris, 45 mM boric acid, 1 mM EDTA,
pH 8.3) at 4” using a voltage gradient of 6 V/cm. The
pulse times and lengths of electrophoresis
were chosen
to be appropriate for the DNA size range to be separated.
Following
electrophoresis
sample plugs were sealed
within the well, and the gel was stained with ethidium
bromide prior to photography
and processing for blot hybridization
as described (2,11).
200 mM Tris, 100 mM EDTA
RESULTS
y-Irradiation
of Circular and Linear DNA Standards
Treatment of the 33 kb supercoiled circular plasmid
pK300 in 200 mM Tris, 100 mM EDTA, pH 7.4, with increasing doses of y-irradiation results in the successive
generation of open circular followed by linearized DNA
(Fig 1A). A dose of 5-7 krad converts about 50% of the
supercoiled circle to open circular forms (Fig 1, lanes 4
and 5), while 25-50 krad results in complete conversion
to open circular molecules (Fig. lA, lanes 8 and 9). In
contrast, generation of significant levels of the 33-kb linear form of pK300 begins to occur at 25-50 krad and occurs maximally at about 300 krad (Fig. lA, lanes 13-15).
At 200 krad and above, DNA less than 33 kb in size appears, indicative of the introduction of multiple doublestrand breaks (Fig. lA, lanes 12-15). A similar quantitative dependence of the relative introduction of singleand double-stranded breaks in DNA in different buffers
has been observed in previous studies employing y-irradiation (10).
Quantitation of y-Irradiation-Induced
used: CHEF,
contour-clamped
100 mM EDTA
Breakage
y-Irradiation of phage X DNA was utilized to measure
the dose-dependence of double-stranded breakage. Introduction of single-stranded breaks into this DNA does
not alter the mobility of the phage DNA, whereas double-stranded breaks will generate fragments smaller
than the intact phage, thereby decreasing the amount of
intact DNA. Figure 1B shows that loo-150 krad irradiation reduces the level of intact X DNA by about 50%.
Assuming that the introduction of breakage is (i) random, (ii) follows a linear dose-response curve, and (iii)
follows a Poisson distribution, the rate of introduction
of double-stranded breaks was determined to be approximately 7 X 10-5/kb/krad under these conditions. A similar rate was determined from the pK300 data shown in
Fig. 1A.
i Abbreviation
field.
200 mMTris,
homogeneous
electric
FIG.
1. Dose dependence
of y-irradiation-induced
breakage
of
DNA. (A) The 33-kb plasmid
pK300 (11) was placed in 200 mM Tris,
100 mM EDTA,
pH 7.4 (6 pg/ml),
and subjected
to the indicated
doses
of y-irradiatiomin
Krad). X DNA (0.25 rg) cut withBamH1
was added
to each sample, which was then heated at 67°C for 10 min and immediately electrophoresed
in 0.8% agarose gels at 1.1 V/cm for 48 h in 40
mM Tris, 20 mM acetate,
2 mM EDTA
(pH 8.1) buffer. The ethidium
bromide-stained
gel is shown. oc, open circular
DNA; SC, supercoiled
circular
DNA; 1, linear DNA. The brackets
on the right delineate
the
positions
of the h/BamHI
fragments.
The migration
positions
of open
circular
and linear pK300 DNAs were confirmed
by analysis of DNAs
treated
with topoisomerase
or appropriate
restriction
enzymes,
respectively.
(B) h DNA (~1587 Sam7,6 fig/ml)
in 200 mM Tris, 100 mM
EDTA,
pH 7.4, was subjected
to irradiation
and electrophoresis,
as
described
in (A). (Cl pK300 DNA (6 fig/ml was placed in 10 mM Tris,
1 mM EDTA, pH 7.4, and subjected
to irradiation
and electrophoresis
as described
in (A).
112
STEPHEN
M.
Buffer Effects
Irradiation
of pK300 in 10 mM Tris, 1 mM EDTA, pH
7.4, results in the expected conversion to open circular
and then linearized DNA forms (Fig. 1C). However, the
conversion to open circles is 50% complete after only 0.6
-1 krad, and linearized DNA appears maximal at 40
krad. These dosages are about eightfold lower than required in the 200 mM Tris, 100 mM EDTA buffer (Fig.
1A). A similar effect was observed using X DNA (not
shown). The effect of buffers was examined on pK300
and X DNA following irradiation, using the following
buffers: 10 mM Tris, 100 mM EDTA; 200 mM Tris, 1 mM
EDTA; 10 mM Tris, 1 mM EDTA, 500 mM NaCl; and 10
mM Tris, 1 mM EDTA, 4 mg/ml sorbitol, all at pH 7.4.
The 500 mM NaCl-containing buffer showed a quantitative dependence of breakage very similar to that observed in 10 mM Tris, 1 mM EDTA (not shown), while
the other three buffers exhibited sensitivities identical
to that observed in 200 mM Tris, 100 mM EDTA (not
shown).
Irradiation
in Agarose Sample Plugs
The buffer effects described above, especially that observed with sorbitol-containing buffers, suggested that
the sensitivity of DNAs to y-irradiation within agarose
plugs could differ from that observed in free solution. E.
coli cells containing pK300 were cast in agarose plugs
and processed to yield chromosomal DNA. This protocol
yields intact DNAs enclosed by a cell-sized space within
the agarose matrix (as opposed to simply casting free
DNA in agarose), as employed in typical eukaryotic
chromosomal preparations for pulsed-field electrophoresis. These sample plugs were washed extensively in either 10 mM Tris, 1 mM EDTA or 200 mM Tris, 100 mM
EDTA, and irradiated with 5,20, or 100 krad. The DNAs
were electrophoresed, and the pK300 DNA visualized by
blot hybridization (Fig. 2). The buffer dependence of irradiation sensitivity observed in solution is clearly evident in the agarose-enclosed DNAs; 100 krad in 10 mM
Tris/l
mM EDTA yields mostly linear or degraded
pK300 (Fig. 2A, lane 4), whereas 100 krad irradiation of
samples in 200 mM Tris/lOO mM EDTA results in mostly
open circular and linear species (Fig. 2B, lane 4). Moreover, comparison of the results of Fig. 2 with those in
Figs. 1A and 1C indicate that quantitatively similar levels of breakage occur in the two buffers, regardless of
whether pK300 is free in solution or contained within E.
coli-sized spaces within the agarose matrix. These data
indicate that the irradiation-dependence of DNA breakage is comparable in free and agarose-enclosed DNAs.
Example: Sizing of Circular DNAs in MethotrexateResistant Leishmania
As an example of the use of this technique, I show the
analysis of the extrachromosomal circular H region
BEVERLEY
Krad
A.
0
5 20 (00
i0mM Tris
1mM EDTA
8: 0
Krad
20 100
5
200 mM Tris
100 mM EDTA
FIG. 2.
y-Irradiation-induced
breakage
of DNA enclosed in agarose
sample plugs. E. coli containing
pK300 were cast in agarose and the
DNA prepared
as described
(12). Individual
sample plugs from the
same preparation
(containing
the same number
of cells) were irradiated with the indicated
dose of y-rays
and electrophoresed
as described in Fig. 1. The gel was then blotted to Gene Screen Plus membranes and hybridized
with a radiolabeled
plasmid
containing
a Lei.shmania fragment
derived
from pK300
(the 4.5-kb Hind111
fragment;
(2)), the results of which are shown. oc, Open circular
DNA, SC, supercoiled circular
DNA, 1, linear DNA.
present within the RlOOO-11 line of L. major (2). This
highly methotrexate-resistant
line also bears a stable
amplification of the R region, encoding the bifunctional
dihydrofolate reductase-thymidylate synthase (2,14,15).
The biochemical mechanism of resistance encoded by H
region is unknown, although current data indicate that
it bears a functional drug resistance element which does
not mediate decreased methotrexate uptake ((16); Ellenberger and Beverley, in preparation). Similar results
have been presented for the amplified H region DNA
present in certain laboratory stocks of Leishmania tarentolae (5).
Chromosomes of the RlOOO-11 line were irradiated
with increasing levels of y-radiation and the DNAs were
separated by pulsed-field electrophoresis. Hybridization
of an H region-specific probe to a blot of this gel is shown
in Fig. 3. Without irradiation, strong hybridization is observed in the sample well, and weaker hybridization to a
DNA labeled “SC” (Fig. 3, lane 1). The “SC” DNA exhibits mobility properties characteristic of supercoiled circular DNAs, such as pulse-time dependent mobility (relative to linear DNA size standards) and altered migration direction, and is resistant to exonuclease III
treatment ((11,17); data not shown). The material remaining within the well probably corresponds to nicked
circular H DNAs, as large open circular DNAs (>30 kb)
do not enter the gel under the high voltage gradients utilized (6 V/cm; (11)). Following irradiation, progressively
reduced levels of the supercoiled DNA are found, and a
new DNA species appears with an apparent molecular
SIZING
OF
CIRCULAR
DNA
BY
PULSED-FIELD
113
ELECTROPHORESIS
doses the hybridization
signal is completely removed
from the sample well, indicating that a complete chase
of the well DNA can be obtained. The dose required for
this chase is also consistent with calculations based
upon the size of the H circular DNA and the dose-dependence of strand breakage following irradiation of samples stored in 200 mM Tris, 100 mM EDTA buffer.
SC- 0 *”
i-,r
6xa4h-
-
DISCUSSION
-
FIG. 3.
y-Irradiation
analysis of amplified
H region DNA
mania major. Agarose sample plugs containing
chromosomal
the RlOOO-11
line of methotrexate-resistant
Leishmania
were irradiated
with the indicated
doses of y-rays
and then
by CHEF
electrophoresis.
The pulse time was 25 s, and the
electrophoresis
was 40 h. The gel was blotted to Gene Screen
hybridized
with an H region-specific
probe (a 1.9-kb Hind111
pHM-3;
(2)), the results of which are shown. SC, Supercoiled
H DNA; 1, linearized
H DNA. Size markers
were provided
by
of X DNA, prepared
as described
(12).
in LeishDNA of
major (2)
separated
length of
Plus and
fragment,
circular
oligomers
weight of about 85 kb (the nicked circular species remains in the well, as discussed above). The new 85-kb
DNA is linear, as revealed by mobility characteristics
(lack of altered migration path or pulse-time dependence) and sensitivity to exonuclease (not shown).
These data show that the amplified circular H region is
85 kb in length in the RlOOO-11 line, which is also its
length within the unstable RlOOO-3 parent of this line
(2). A similar estimate for the size of the intact H circular DNA has been obtained by electron microscopy (18).
Interestingly, the hybridization to the linear 85kb DNA
following 50-300 krad is more intense than that observed to the supercoiled circular species, indicating that
at least some of this linear DNA must have arisen from
the material retained in the sample well in the absence
of irradiation. This provides evidence that some portion
of the material hybridizing to the well corresponds to the
85-kb circular DNA. Further discussion concerning the
structure of the amplified R and H region DNAs within
the RlOOO-11 line will appear elsewhere.
The experiment shown in Fig. 3 provides additional
controls for the y-irradiation method. First, the relative
dose-dependence for the appearance of the linearized
forms of the amplified H region DNA of Leishmania is
consistent with the values determined earlier with control DNAs in free solution. Secondly, at high irradiation
With limited y-irradiation it is possible to introduce
a controlled number of double-stranded breaks within
DNA, allowing the generation of linearized forms from
circular DNAs whose size may be readily estimated by
electrophoretic techniques. The y-irradiation approach
is especially convenient for introducing breakage into
circular DNAs found within the complex chromosomal
DNA preparations commonly employed in pulsed-field
electrophoresis, as one simply irradiates the DNA directly in the agarose sample plug with the desired dose.
In comparison, enzymatic cleavage methods require
diffusion of the enzyme into the sample plug, a process
which is difficult to control sufficiently to obtain limited
and reproducible cleavage (Beverley, unpublished observations). A similar approach employing irradiation from
a radioactive Cs source has recently been presented (9).
If the approximate size of the circular DNA to be measured is known, one can use the values presented earlier
to estimate the appropriate radiation dose necessary for
linearization. With circular DNAs of unknown length, it
is necessary to employ radiation dosesover a wide range.
Otherwise, dosages appropriate for smaller circular
DNAs will introduce multiple breaks in larger DNAs,
and their presence will be overlooked.
ACKNOWLEDGMENTS
I thank members
of my laboratory
for advice, discussions,
labeled probes, D. E. Dobson, T. E. Ellenberger,
and G. M.
comments
on this manuscript,
D. D. Rogers for technical
and the Radiobiology
Department
of the Harvard
School
Health for use of their y-irradiator.
This work was supported
Grant AI 21903. SMB is a Burroughs-Wellcome
scholar in
Parasitology.
and radioKapler
for
assistance,
of Public
by NIH
Molecular
REFERENCES
1. Rush,
M. G., and Misra,
2. Beverley,
S. M., Coderre,
(1984) Cell 38,431-439.
3. Garvey,
E. P., and Santi,
R. (1985).
D. V. (1986)
4. Detke,
S., Chaudhuri,
G., King,
Biol. Chem. 263,3418-3424.
5. Petri110 Peixoto,
5188-5199.
Plusmid
J. A., Santi,
M., and Beverley,
14,177-1985.
D. V., and Schimke,
Science
233,535540.
J. A., and Chang,
S. M. (1988)
R. T.
K-P.
Mol.
(1987)
J.
Cell. Biol.
8,
114
STEPHEN
6. Hamkalo,
B. A., Farnham,
P. J., Johnston,
R., and Schimke,
R.T. (1985).Proc. Nati. Acad. Sci. USA 82,1126-1130.
7. Carroll,
S., Gaudray,
P., DeRose,
M. L., Emery,
J. F., Meinkoth,
J. L., Nakkim,
E., Subler, M., Von Hoff, D. D., and Wahl, G. M.
(1987) Mol. Cell. Biol. 7, 1740-1750.
8. von Hoff, D. D., Needham-VanDevanter,
D. R., Yucel, J., Windle,
B. E., and Wahl, G. M. (1988) Proc. Natl. Acad. Sci. USA 85,
4804-4808.
11. Beverley,
BEVERLEY
13. Chu, G., Vollrath,
F. (1985)
Progr.
C. R., and Borst,
P. (1988)
Nucleic
Acids
Res. 32,115-154.
S. M. (1988) Nucleic Acids Res. l&925-938.
12. Smith, C. L., Warburton,
P. E., Gaal, A., and Cantor,
C. R. (1986)
in Genetic
Engineering
(Setlow,
J. K., and Hollaender,
A., Eds.),
Vol. 8, pp. 45-70, Plenum,
New York.
R. W. (1986)
Science
234,1582-
14. Beverley,
S. M., Ellenberger,
T. E., and Cordingley,
Proc. Natl. Acad. Sci. USA 83,2584-2588.
3. S. (1986)
15. Grumont,
R., Washtien,
W. L., Caput,
Proc. Natl. Acad. Sci. USA 83,5387-5391.
D. V. (1986)
T. E., and Beverley,
D., and Santi,
S. M. (1987)
J. Biol.
C/mm
262,
13501-13506.
17. Hightower,
Nucleic
D., and Davis,
1585.
16. Ellenberger,
9. van der Bliek, A. M., Lincke,
Acids Res. 16,4841-4851.
10. Hutchinson,
M.
R. C., Metge,
D. W., and Santi, D. V. (1987) Nucleic
AcidsRes.
15,8387-8398.
18. Beverley,
S. M., Ellenberger,
T. E., Iovannisci,
D. M., Kapler,
G. M., Petrillo-Peixoto,
M., and Sina, B. J. (1988) in Biology
of
Parasitism
(Englund,
P. T., and Sher, A., Eds.), MBL Lectures
in
Biology,
Vol. 9, pp. 431-448,
Alan Liss, N.Y.