Issue 1, March 2001
Biological & Biomedical Sciences
Influence of Substrate Surface Chemistry on the Binding of DNA-RecA
Blaine C. Butler, Gina MacDonald, Brian H. Augustine
Department of Chemistry, James Madison University
A procedure for imaging calf thymus and plasmid DNA using contact
mode atomic force microscopy (AFM) is described. The stable tethering
of double stranded calf thymus and plasmid DNA molecules to a mica
surface was facilitated by surface modification of freshly cleaved
mica with Mg+ ions. Mica surfaces treated with spermidine
are shown to result in aggregation of DNA at high treatment concentrations
(> 5x10-2 mg/mL) and in little binding at low treatment
concentrations (< 5x10-2 mg/mL). Improvements in tips
and scanning conditions are still needed in order to achieve higher
Background and Introduction
The protein recA and its interactions with DNA have long been studied
(Ogawa et al. 1993). This is a result of the fact that recA
is a protein involved in homologous genetic recombination This is
the recombination of DNA between DNA fragments with identical or
very similar sequences (Ogawa et al. 1993; Roca and Cox 1990;
Shan et al. 1996). RecA is also involved in post-replicative
repair of damaged DNA molecules (Ogawa et al. 1993; Roca
and Cox 1990; Shan et al. 1996). RecA can bind to either
single-stranded or double-stranded DNA in the presence of adenosine
triphosphate (ATP), ATPs
(which is the nonhydrolyzable form of ATP) or adenosine diphosphate
(ADP). When ATP is bound to recA, the protein adopts an active conformation
and has a strong affinity for DNA (Ogawa et al. 1993). When
ADP is bound to recA, the protein adopts an inactive conformation
and has a low affinity for DNA. Only when ATP or ATPs
is bound to recA, can the protein self-assemble into nucleoprotein
filaments and induce DNA strand exchange (Ogawa et al. 1993).
An example of this strand exchange is shown in Figure 1 (http://www.bcbp.gu.se/mbg/recomb/).
1: Schematic of homologous genetic recombination, as carried
out by the recA protein. The single-stranded DNA, red, is
switched with the homologous strand, green, in the double-stranded
The purple spheres are the recA proteins, which are polymerized along
the single-stranded DNA, in red. The double-stranded DNA, yellow and
green, then binds to the polymerized protein. Strand exchange then
occurs between the red and green strands. RecA is also an essential
protein in the "SOS" functions of DNA repair (Ogawa et al.
1993; Roca and Cox 1990; Shan et al. 1996). The SOS response
is the coordinated activation of diverse metabolic functions in response
to severe DNA damage (Ogawa et al. 1993; Roca and Cox 1990;
Shan et al. 1996). RecA is able to induce its own synthesis
on damaged DNA. This is done by the cleavage of a repressor protein,
LexA. Once the repressor has been cleave, all of the SOS proteins
are then able to be expressed.
RecA is one of the most intensively studied of the enzymes involved
in homologous recombination for many reasons. One reason is that there
are large quantities of recA in the cell, it accounts for several
percent of total cellular protein after induction (Ogawa et al.
1993). It can also be purified with relative ease, and it is possible
for recA to catalyze a strand-exchange reaction in vitro by
itself (Ogawa et al. 1993; Roca and Cox 1990; Shan et al.
1996). One of the most important reasons for studying recA is that
there is a growing belief that the specific nucleoprotein structure
formed by the recA proteins on the DNA strands may be a universal
structure throughout all of biology (Ogawa et al. 1993). This
suggests that recA is the ideal model protein for how homologous genetic
recombination occurs in humans.
Scanning probe microscopy (SPM) is a new surface analytical tool with
a potentially enormous impact on the study of biological systems,
due to the near-atomic resolution attainable with these microscopes
(Hansma et al. 1988; Lyubchenko et al. 1992). SPM offers
the unique advantages of high-resolution imaging of DNA, RNA, and
other biological molecules in the absence of stains, shadows, and
labels. Furthermore, they can be operated in either air or liquid
ambients (Lyubchenko et al. 1993). The key to successful imaging
of DNA and other biomolecules such as DNA-protein complexes is the
need to ensure that these molecules are both chemically stable and
sufficiently tethered to the substrate surface. Although the atomic
force microscope (AFM) is able to produce near-atomic resolution,
its potential has been limited in microstructural studies of DNA,
RNA, and other biological materials. The most immediate limitation
to the practical application of AFM in studying biological complexes
is sample preparation. The molecules must be attached to the substrate
surface in order to avoid movement caused by lateral forces generated
by the sweeping probe during scanning. If they are not adequately
attached, the drift experienced by these molecules limits image resolution.
An ideal tethering scheme will be able to both secure a DNA molecule
to the surface, but will also allow the molecule to undergo changes
in conformation without significant steric hindrance, in the presence
of other reactive biomolecules. This is needed in order to study chemical
reactions between proteins and DNA on a surface. A second limitation
to AFM resolution may be adhesion of biological macromolecules to
the scanning tip. This again, is more likely to occur if the molecules
are not stably tethered to the surface. Both of these issues must
be addressed in order to successfully image biological molecules using
One approach to tethering DNA is by chemically modifying a substrate
surface in such a way as to increase the affinity, and therefore binding
of the DNA to the surface. This technique has become the most common
procedure for stably attaching DNA to mica surfaces. Many chemical
modification processes have been attempted with favorable results,
such as MgCl2-treated mica (Bezanilla et al. 1994),
glow-discharged mica (Hansma et al. 1992), spermidine-treated
mica (Tanigawa et al. 1997), and aminopropyltriethoxy silane
(APTES)-treated mica (Lyubchenko et al. 1992). Each of these
processes works by modifying the surface chemistry to be more amenable
to the charge and chemical state of biomolecules.
Previous research on recA and DNA complexes has used Scanning electron
microscopy (SEM), Fourier transfer infrared spectroscopy (FTIR), and
X-ray crystallography to determine the difference between the active
and inactive conformations (Ogawa et al. 1993). Our ultimate
research goal is to use contact mode AFM to study the protein conformation
in the presence of different nucleotides, ATP, ADP, or ATPs.
Before this can be accomplished, we report the necessary surface modification
techniques and sample treatment in order to enable us to pursue more
ambitious protein conformation studies.
More information on scanning probe microscopy and recA can be found
at the following websites:
Materials and Methods
Mg+ Surface Treatment
Pieces of freshly cleaved ruby mica were immersed in solutions of
33 mM magnesium acetate (Aldrich Co.) in deionized/distilled water
(Hansma et al. 1992; Tanigawa et al. 1997), for periods
of 12 hrs - 2.5 weeks. They were then rinsed with deionized/distilled
water and dried under a gentle compressed nitrogen flow at room temperature
for approximately 30 seconds. Mica pieces were stored in a nitrogen
atmosphere until they were used, up to 2 weeks after treatment.
Spermidine Surface Treatment
Freshly cleaved mica pieces were submerged in 5 x 10-2
mg/mL, 2.5 x 10-2 mg/mL, 2.5 x 10-1 mg/mL, and
5 x 10-1 mg/mL spermidine (Aldrich Co.) solutions, as reported
by Tanigawa et al. (1997). They were then allowed to incubate
for 5 minutes. The pieces were subsequently rinsed with deionized/distilled
water and dried under compressed nitrogen. These treated mica pieces
were used immediately.
DNA Attachment on treated mica
Pieces of Mg+ -treated mica were mounted onto magnetic
AFM sample discs and placed in plastic storage boxes. A solution of
2.0 + 0.2 mg/mL calf thymus double stranded DNA (Sigma Co.)
in buffer, containing 5 mM Hepes, 5 mM KCl, and 5mM MgCl2
(Bezanilla et al. 1994) was pipetted onto the surface of the
mica. Plasmid double stranded DNA, M13mp8 RF I DNA (Sigma Co.) was
also used. When plasmid DNA was used, the concentration was 4 - 12
µL of M13 DNA as received in 100 µL of the aforementioned
buffer. This solution was then pipetted onto the mica surface. The
DNA-treated mica pieces were placed in a 4oC refrigerator
for 10-15 minutes. They were then rinsed with distilled/deionized
water and dried on ice under a gentle flow of compressed nitrogen
for approximately 30 seconds.
Spermidine-treated mica was mounted onto magnetic AFM sample discs
and placed into plastic storage boxes. 100 µL of 2 mg/mL solution
of calf thymus DNA in the aforementioned buffer was pipetted onto
the spermidine coated surface. This was allowed to incubate for 5
minutes. These samples were then rinsed with deionized/distilled water
and dried under a gentle flow of compressed nitrogen for approximately
30 seconds. These were then imaged immediately.
Atomic Force Microscopy
All imaging was performed on a Nanoscope III extended Multimode AFM
(Digital Instruments) in air with contact mode AFM using standard
Si2 N4 probes (Digital Instruments, NP average
spring constant k=**N/m) or oxide sharpened probes (Digital Instruments,
NP-STT average spring constant k=**N/m). The scan rates varied from
0.45-1.4 Hz, image size varied from 10 µm to 100 nm, and the
z range also varied, from 8-1 nm; decreasing as the scan size decreased.
Results and Discussion
treated surfaces did not appear optimal for DNA imaging. When 5
x 10-2 mg/mL concentrations of spermidine were used,
no DNA was observed on the surface through multiple trials, as shown
in Figure 2.
The concentration of spermidine was then increased by an order of
magnitude to 5 x 10-1 mg/mL. In this case, there is aggregated
DNA across the entire surface, as shown in Figure 3.
3: AFM contact mode image in air, (10 mm scan) of 5 x
10-1 mg/mL spermidine-treated mica surface after
DNA exposure. DNA is aggregated on the surface with an average
of 3-4 µm in diameter.
The aggregated DNA appears to be 3-4 µm in diameter, and is uniformly
aggregated across the mica surface. We also tried 2.5 x 10-2
mg/mL, in which no distinct molecules were seen on the surface, they
appeared to be displaced by the tip. A spermidine concentration of.2.5
x 10-1 mg/mL was also tried, this time there appeared to
be more molecules on the surface, but once again they were easily
displaced by the sweeping AFM tip. We were never able to observe individual
DNA molecules by reproducing the spermidine treated surfaces as reported
by Tanigawa et al. (1997).
Stable DNA tethering was obtained using the Mg+ treatment
as reported from Hansma(Hansma et al. 1992; Tanigawa et
al. 1997). This sample preparation involved not only treatment
of the surface with Mg+ ions, but it also included rinsing
of the surfaces under a stream of water after DNA treatment to remove
excess DNA molecules. This presumably leaves behind only the stably
tethered molecules most suitable for imaging. The prominent features
in the AFM images of the Mg+ treated substrates are numerous
regions of aggregated DNA. These appear as bright patches approximately
1-2 µm in diameter, as indicated in Figure 4.
4: AFM contact mode image in air, of calf thymus DNA on
Mg+ treated mica surface (2.25 µm scan). Individual
DNA molecules are indicated with arrows.
However, single molecules can also be observed around the edges of
these clusters and dispersed over the entire surface. This is also
indicated in Figure 4 with an arrow. These single molecule DNA have
measured lengths of 100-150 nm which is consistent with previous experimental
data (Lyubchenko et al. 1992). We have measured widths of 8-12
nm. This is greater than the literature value of 2 nm (Lyubchenko
et al. 1992), but is consistent with broadening that can be
attributed to the minimum tip radii of the probe as previously determined
experimentally (Lyubchenko et al. 1992).
Once we were able to obtain reproducible AFM images of calf thymus
DNA, we then attempted the same procedure on plasmid, M13mp8 RF I,
DNA. Plasmid DNA was used, since plasmid DNA and single stranded poly
dT, are the types of DNA that will subsequently be used in the experiments
involving DNA-recA protein interactions. Single plasmid DNA molecules
were also observed on the mica surface, as shown in Figure 5.
5: AFM contact mode image in air of plasmid, M13mp8 RF
I, DNA on Mg+ treated surface (2.5 µm scan).
Individual DNA molecules are indicated with arrows.
The DNA was sufficiently bound to the surface, and able to withstand
multiple higher resolution scans in contact mode. The plasmid DNA
molecules can easily be seen, as the circular features, approximately
80-90 nm in diameter as indicated in Figure 6.
6: AFM contact mode image, in air (800 nm scan) of individual
plasmid DNA on the Mg+ -treated mica surface. Diameter
of plasmids is about 80-90 nm.
The width of these features is approximately 10-15 nm. Again, this
is wider than expected, but may be attributed to the relatively large
radius of curvature of even the oxide-sharpened tips. Higher resolution
images could be obtained by using tips that have smaller radii of
curvature such as electron beam deposited tips (Guthold et al.
1999) or nanotube probes (Wong et al. 1998).
Images of DNA have been obtained at relatively high resolutions
using AFM on chemically modified mica surfaces. These molecules
have been successfully tethered to the surface by treating the substrate
with Mg+ ions. DNA is stable for repeated scanning, and
minimal drift occurred in the imaging of these molecules. It appears
that both the surface modification and the sharpness of the tips
were the limiting factors in image resolution, and through this
work a surface modifying procedure has been found that is amenable
to our research. The surface modification and sample procedure still
require testing for the DNA-recA complexes, but we believe that
since this procedure works for DNA, then a DNA-protein complex should
have similar surface-molecule interactions. It is also obvious that
in order to achieve higher resolution images, other scanning techniques
will need to be applied. Some of the techniques to be used in the
future will include scanning in liquid; and sharper tips that have
a smaller radius of curvature and experience less tip contamination
by surface molecules.
The authors would like to thank the Materials Research Society Undergraduate
Materials Research Initiative (MRS-UMRI), the National Science Foundation
Research Experiences for Undergraduates (Grant# 97-31912), and the
National Science Foundation (Grant# MCB 9733566) for support of this
M., Manne, S., Laney, D., Lyubchenko, Y., Hansma, H. Langmuir 1994,
Guthold, M., Zhu, X., Yang, G., Thomson, N., Kasas, S., Hansma, H.,
Smith, B., Hansma, P., Bustamante, C. Biophysics Journal 1999, 77,
Hansma, P.K., Elings, V.B., Mauli, O., Bruker, C.E. (1988) Science
Hansma, H., Vesenka, J., Siegerist, C., Kelderman, G., Morrett, H.,
Sinsheimer, R., Elings, V., Bustamante, C., Hansma, P. Science 1992,
Lyubchenko Y., Gall, A. Shlyakhtenko, L., Harrington, R., Jacobs,
B., Oden, P., Lindsay, S. Journal of Biomolecular Structure and Dynamics
1992, 10, 589-606.
Lyubchenko, Y., Oden, P., Lindsay, S., Dunker, K. Nucleic Acids Research
1993, 21, 1117-1123.
Ogawa, T. Yu, X. Shinohara, A., Egelman, E. H. (1993) Science 259,
Roca, Al, Cox, M. (1990) Critical reviews in biochemistry and molecular
biology 25, 415-456
Shan, Qun, Cox, Michael, Inman, Ross (1996) Journal of Biological
Chemistry 271, 5 5712- 5274
Tanigawa, M.; Machida, M.; Okada, T. CCAB 1997.
Wong, S., Harper, J., Lansbury, P., Lieber, C. Journal American Chemical
Society 1998, 120, 603-606.
Journal of Young
Investigators. 2001. Volume Three.
Copyright © 2001 by Blaine C. Butler and JYI. All rights reserved.