Vol.
177,
June
28,
No.
BIOCHEMICAL
3, 1991
AND
BIOPHYSICAL
RESEARCH
1991
COMMUNICATIONS Pages 1313-1318
CHANGES IN DNA SUPERHEUCAL DENSITY MONITORED BY POLARIZED UGHT SCATTERING C. NICOLINI , A.DIASPRO,
M.BERTOLOlTO,
P. FACCI AND L. VERGANI
INSTITUTE OF BIOPHYSICS I UNlVERSlTY OF GENOVA , Via Giotto, 2 16153 GENOVA-SESTRI PONENTE , ITALY
Received
May 22,
1991
Linear and circular L-DNA at different ethidium bromide concentrations have been studied by means of polarized light scattering, namely the St 4, S34, S33 and St3 elements of Mueller matrix. While S33 at low angle appears well correlated with the total light scattering evaluated by optical density measurements at 632.8 nm for linear and circular DNA of the same mass, the magnitude and slope of the St4, S34 and St3 signals display significant changes for the circular X-DNA depending on the degree of negative superhelical density as induced by the different ethidium bromide concentrations. At the same time, for linear h-DNA the signal remains invariant, making explicit for the differential scattering of polarized light the possibility to obtain additional information by its angular dependence. Strikingly also the effect of 0.2% glutaraldehyde versus ethanol fixation on the native h-DNA structural properties appears to confirm earlier findings by other well-established probes. Results are discussed in terms of first physical principles and of their potential bearings 0 1991 Academic towards our understanding of the mechanism controlling gene expression. mess, Inc.
MATERIALS AND METHODS Biological
Samples
A-DNA (GIBCO-BRL, Maryland, USA) had a molecular weight of 32x106 daltons (48,502 base pairs). The DNA dissolved at a concentration of 500 mglml in TE (pH 7.5) was heated to 65” C for 10 min. and quickly cooled in ice, in order to restore the linear form. At 4” C, hDNA formed linear aggregates and circular structures (due to the presence of cohesive termini) which have been removed by heating. Ligation buffer 10x and T4 ligase (GIBCOBRL, Maryland, USA) were added to h-DNA solution in order to have 1 enzyme unit per mg of DNA in Ix reaction buffer. After incubation at 37” C for 2 hours, the closed circular h-DNA form has been obtained. Samples of circular h-DNA (16.6 ug/ml in TE pH 8) have been incubated at different concentrations of ethidium bromide to induce different superhelicities (see Table I). A quartz cuvette containing 2 ml of the sample’s suspension has been used to perform light scattering measurements. Fixation has been performed as described in the text in TE (pH 7.5). Multiple
Light Scattering
Differential scattering of polarized light has been measured with our CIDS (Circular Intensity Differential Scattering) spectrometer described in details elsewhere (1). All the system is computer controlled by the MUCIDS on-line and off-line managing software. (2).
Vol.
177, No. 3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Our instrument differs slightly from similar ones developed by other investigators (3,.4). Nevertheless it represents a significant improvement because, unlike the most of other instruments of this type, it produces meaningful and reproducible angular spectra for the characterization of the higher order organization of chromatin-DNA and mononucleosomes (1, 5) X-DNA and microorganism suspensions (manuscript in preparation). As discussed in details elsewhere (I), the S33 signal at 35” would appear more appropriate to normalize the elements than St 1 having the same signal/noise ratio of the element to be normalized while the DC component (related to St 1) is usually detected at a different signal/noise ratio. Infact, the S33 element is the predominant part of the AC signal and it is acquired using the highest signal/noise ratio allowed by Lock-In-Amplifier. Furthermore the optical density of the sample, routinely taken at 632.8 nm (Jasco 7800, Tokyo, Japan), and related to the total light scattered by the sample, correlates well with the value of S33 (1). As a biological standard we have utilized a mononucleosome suspension for which a zero CIDS signal has been theoretically predicted (6, 8) and experimentally obtained (1, 5, 9). INTRODUCTION AND RESULTS Over the last several important
years DNA supercoiling
structural
dynamic
appear
cellular
components
feature
in solution
is however either
hampered
in situ (electron
We have then explored
by the
and
Light scattering
has been widely axial
sizes,
these experiments
ratios,
neglect
measurement
in the degree
limitations
of
of present
or in vitro (sedimentation). based on the differential in terms of classical
A-DNA in presence
used in biological
particle
mobilities
the detailed
angular
of the scattering
information
chromatin
fiber
We have
concentrated
scattering
electrodynamics
of (6,
of increasing
ethidium
about
research
to determine
particle
numbers,
(17).
However,
and indices distribuition
parameters
biological
of refraction
bromide
of polarization.
as function
scattering
of angle
systems,
provides
as previously
our interest
on the St 4, proportional
other elements
of the Mueller
matrix have been recently
studying
higher-order
structure
the
instead shown
in
(1, 5, 9). to the Circular
Differential Scattering (CIDS), and a few other elements of the polarized matrix, namely Sf3, 534 and S33. CIDS, -St 4/St 1 in the Mueller notation
discriminating
and
or other
(16).
particle
significant
proteins
changes
invasiveness
light (1, 5) which can be now understood
concentrations
as the most
(9, 13). Its formation
with chromosomal
to monitor
of a technique
of circularized
much attention
expression
feasibility
microscopy)
the utilization
8) in the characterization
The
of gene
as DNA interacts
(10, 14, 15). The
supercoiling
polarized
in the control
modulated
technologies,
has been gaining
microorganisms
classical electrodynamics
of large
in solution
shown powerful
biopolymers
(10, 21). Application
(1,
Intensity
light scattering (18) and some
biophysical
5, 19,
tools in
20),
of the general
to helices has shown that plots of CIDS versus scattering
and
in
theory
of
angle give
shapes whose amplitude, form and peak positions are characteristic of the sense, pitch and radius of the helix (6, 8, 10, 18, 20). With all the above in mind, the S33, S34, St 3 and St4 elements for linear
of the Mueller
and circular
matrix have been measured
X-DNA
brought
in suspension
as a function of the scattering at increasing
ethidium
angle
bromide
concentration. Increasing
the ethidium
DNA remain
rather
bromide
invariant,
concentration, while
circularly 1314
the geometric closed
and optical
DNA changes
features
both
of linear
its number
of
Vol. 177, No. 3, 1991
35
BIOCHEMICAL
50
80 Scamng
AND BIOPHYSICAL
I 140
110 angle (E)
35
RESEARCH COMMUNICATIONS
50
a0 scattermg
140
110 angle (E)
Angular dependence of the the S14 (A) element for h-DNA at different ethidium bromide concentrations, respectively 0 ( m ), 50 ( 0 ) and 200 ( 0 ) pug/ml for circular and 0 ( n ) and 200 ( A ) j.rg/ml from the linear one. The error bars are not visible because their amplitude is smaller than that of the symbols adopted and St4 percent intensity (Et) versus scattering angle for circularized I-DNA at increasing ethidium bromide concentrations. In Fig.18 the St 4 element has been normalized for S33 (35”). The optical density (O.D.) at 632.8 nm and the S 33 intensity at 35 ’ for b-DNA are, for linear and circular DNA, respectively 0.001 OD , 0.5 mV and 0.008 OD , 4 mV.
superhelical tUrflS per 10 base pairs behaviour (Fig.lA) The
and the differenfiaf
can be found in the angular elements
A-DNA
as a function
superhelical
For a DNA concentration
dependence
of different
density,
of the S34
ethidium
bromide
s, has been computed
of 0.572 IO - 4
scatfering
of polarized
fight. This
(data not shown)
and Sl4
concentration.
as previously
M we have calculated
reported
in (14, 26).
for s the following
values : -
0.028 ( at 0 EB mg/ml), + 0.075 (at 50 EB mg/ml), +0.090 ( at 200 EB mglml). Similarly the sample exibhit a different behaviour of the S33 element (not shown)for linear, and circular change
I-DNA
in optical
particular
at different
density,
as measured,
the light scattered
in absence
of absorption
ethidium
bromide
outside
by the circular
the absorption
DNA is higher
the total scattered
addition
correctly
of ethidium
to the ability Similarly compare
the effect of changes bromide
striking
is the difference
concentration.(data fixed circularized
not shown)
subtle
which
Finally the angular appears,
3:l V/V, [Fig. 21; indeed
The normalized
L-DNA
instead,
displays
dramatically
1315
where
for S33 (35”), in order data (Figure
in the superhelical
function
dependence
the ratio is constantly
to (1-e-0.D.)
of the Mueller
as
the
nm (1). In
that have been induced
changes
in the S13 element linear
X-DNA with 0.2% glutaraldeheyde
the DNA superhelix, fixation
versus
circularized
proportional
in superhelicity
to monitor
parallels
band, at 632.8
the St4 element
to the initial solutions.
of this parameter
which
than that by the linear one, when
light becomes
O.D. is the optical density measured at 632.8 nm. Consequently it appears appropriate to normalize to monitor
concentration,
of
matrix ethidium
by the 1B) point
structures. when
we
bromide
of the ratio of unfixed versus the lack of any perturbation altered
with ethanol-acetic
close to that with glutaraldehyde,
on acid for
Vol.
177, No. 3, 1991
35
40
45
BIOCHEMICAL
50
55 60 65 70 Scattemg Angle
75
80
85
AND BIOPHYSICAL
90
35
40
RESEARCH COMMUNICATIONS
45
50
55 60 65 IO Scattemg Angle
75
80
65 90
Ratio of unfixed versus fixed, with either 0.2% glutaraldehyde (XX:‘) or 3:i V/V ethanol acetic-acid (YE); S14 (left, A) and S34 (right, B) element for circularized h-DNA. The bands represent the upper and lower value within one standard deviation.
both St4
and S34 confirming
pitch) are unaffected Mueller
that size, shape
and geometric
features of the supehelix
by this fixative. On the contrary the ratios for the same elements
matrix range between
5 and 10 in presence
of ethanol-acetic
(i.e., of the
acid.
DlSCUSSlON The S33 element
monitors
information
optical density measurements outside particular S33 changes in intensity, effect may be explained reasonable
to be for circular
Furthermore, in addition The
studies
to the primary
supercoil,
differential
scattering
information
detecting,
in a non-invasive
average
size of the circularized
that there is an increased
gyration
DNA almost from 3 to 6 times greater
experimental
additional
to the total light scattering
on circular
superhelical
light
makes
dependence
way, subtle changes
L-DNA. radius
by
It seems
that may be
than for linear
DNA (22).
DNA (23, 24) suggested
a higher order of supercoiling
of polarized
by its angular
as confirmed
the absorption bands odf DNA (see Fig.1 legend). In but not in shape, for linear and circular DNA. This
with the different
from DNA knots theory
estimated
related
of the DNA is present.
explicit
(Fig. l A-8) in superhelical
that,
the possibility and exibithes density.
to obtain an ability
Namely
while
in for
linear A-DNA the signal remains quite invariant, for the circular one significant changes in both the magnitude and slope of the St4,534 and St3 signals are observed depending on the degree
of
concentrations.
negative superhelical density induced by the different ethidium In particular from the St3 data (not shown) appears conservative
that the apparent the degree detailed
difference
of supercoiling
characterization
the scattering
in the polarized
light scattering
profiles
and not to the formation of ethidium of the relationship
signal appears
between
inherent
to the sample
significant
changes
in the magnitude
preparation.
All above
and slope of the polarized
1316
aggregates.
of superhelical
at this stage difficult and unwarranted,
variability
is related to changes
bromide
the degree
bromide to deduce in
A further
density
and
mainly considering
the
experiments
strongly
light scattering
point
to
signal, namely
Vol.
177,
No.
BIOCHEMICAL
3, 1991
of the S34 and St4, superhelix
induced
Furthermore
which
appear
from ethidium
when
the sample
of the St4 and S34 signals of circular
h-DNA,
studies
thereby
fixation with 0.2% glutaraldehyde In conclusion
it should
level, is suitable superhelix
formation,
expression
in chromatin-DNA
This differential
that is very sensitive the degree
preserves
to detect with
does not change
the structural fixation
with
shown
in the (not
the above
the sensitivity
also evident
(Fig.
and optical
properties
degree
of negative
shown).
Recent
findings,
indicating
the fiber diameter h-DNA
of the
h-DNA.
or with ethanol
our notice that this method,
has been
alterations
effects appear
differences
alchool
COMMUNICATIONS
into the circularized
with glutaraldehyde
the kynetic of circular
which
polarized
intercalation
are compatible
not escape
to monitor
RESEARCH
to the geometric
fixative-dependent
disappear fibers
BIOPHYSICAL
sensitive
fixation
allowing
which
of chromatin-DNA
bromide
is treated
that only glutaraldehyde density
highly
to monitor
2), namely superhelical
AND
X-ray that
and pitch.
even at a semi-quantitative
structural
to be very critical
changes
accompaining
in the control
of gene
(5). light scattering
to small
but significant
of DNA supercoiling
(wherever
technique biological
provide changes
a fast non-destructive such as those occurring
a cell alters its metabolic
activity)
probe in
that are not
seen by any other optical technique. ACKNOWLEDGMENTS This work has Bioinstrumentation”, and 90.00072.PF70)
been supported lstituto Superiore and M.P.I.-40%.
by CNR Target Project di Sanita (A.I.D.S. projet),
“Biotechnology and C.N.R. (89.00191.70
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
Diaspro, A., Bertolotto, M., Vergani, L. and Nicolini, C., (1991), IEEE Trans. Biom. Eng., in press. Diaspro, A., Scelza, P., and Nicolini, C., (1990), Computer Application in Biosciences, 6, 3: 229. Nickel et al. (1976) Proc. Natl. Acad. Sci. USA, 73, 2, 486-490. Hunt, A. J., Huffman, D. R., (1973) Rev. Sci. Instr., 44: 1753-1762. Nicolini, C., Vergani, L., Diaspro, A. and Scelza, P., (1988) Biochem. Biophys. Res. Corn. 155:1396. Belmont,A., Zietz, S. and Nicolini, C., (1985) Biopolymers, 24, 1301-1321. Zietz, S., Belmont, A, and Nicolini, C. (1983) Cell Biophysics, 5, 4:163-i 87. Diaspro, A., Nicolini, C., (1987) Cell Biophysics, 10, 45-60. Nicolini, C. and Kendall, F., (1977) Phys. Chem. and Phys. 9, 265. Nicolini, C., (1986) Biophysics and Cancer. Plenum Publishing New York pp. 267. 268 and 382-404. Crick, F.H., (1976) Proc. Natl. Acad. Sci.-USA 73, 2693. Wang, J.C., (1982) Cell. 29, 724-726. Hirose, S. and Suzuki, Y., (1988) Proc. Nall. Acad. Sci-USA 85, 718-722. Vinograd, J., Lebowitz, J. and Watson, R., (1968) J. Mol. Biol. 33, 173. Lilley, D. M., (1987) inStructure and Dynamics of Biopolymers (C. Nicolini, ed.), Martinus Nijhoff Publ. NATO-AS1 El33 pp. 112-136. Bauer, W. and Vinograd, J. (1968) J. Mol. Biol. 33, 141-l 71. Bohren, C. F. , Huffman, D. R., (1983) Absorption and Scattering of Light by Small Particles. Wiley New York. Bustamante, C., Maestre, M. F. and Tinoco, I. Jr. (1980) J. Chem. Phys. 73, 4273. Maestre, M. F., Bustamante, C., Hayes, T. L., Subirana, J. A. and Tinoco, I., (1982) Nature 292, 773-774. Bustamante, C., Maestre, M. F., Keller, D. and Tinoco, I. Jr. (1984) J. Chem. Phys. 80, 4817-4823. 1317
Vol. 177,
21. 22. 23. 24. 25. 26.
No.
3, 1991
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Saltzman G.C., Grace, W. K., McGregor, D. M., Gregg, C. T., (1984) Biotechniques, 3, 243. Sumner, D. W., (1987) in Geometry and Topology, Mainfolds, varieties and knots, (I. MC Croy and T. Shifrin eds), Marcel Dekker Inc. Publ., New York and Basel, 297318. Brady, G.W., Fein, D.B., Brumberger, H., (1976) Nature, 26, 231-234. Benham, C.J., Brady, G.W., Fein, D.B., (1980) Biophys. J., 29: 351-366. Athley, B.D., Smith, M.F., Rankert, D.A., Williams, S.P. and Langmore, J.P. (1990) J. Cell Biol., 111, 795806. Botchan, P., Wang, J.C. and Echols, H. (1973) Proc. Natl. Acad. Sci. USA, 70(11), 3077-3081.
1318