Content uploaded by Rao Ks Jagannatha
Author content
All content in this area was uploaded by Rao Ks Jagannatha on Jan 29, 2015
Content may be subject to copyright.
Content uploaded by Rao Ks Jagannatha
Author content
All content in this area was uploaded by Rao Ks Jagannatha on Dec 30, 2013
Content may be subject to copyright.
First Evidence for Helical Transitions in Supercoiled DNA
by Amyloid βPeptide (1–42) and Aluminum
ANew Insight in Understanding Alzheimer’s Disease
Muralidhar L. Hegde,1Suram Anitha,1Kallur S. Latha,2
Mohammed S. Mustak,1Reuven Stein,3Rivka Ravid,4
and K. S. Jagannatha Rao*,1
1Department of Biochemistry and Nutrition, Central Food Technological Research Institute,
Mysore-570013, India; 2Centre for Human Genetics, Institute of Biotechnology, G-05,
Discoverer, ITPL, Whitefield Road, Bangalore, India; 3Department of Neurobiochemistry,
George S. Wise Faculty of Life Sciences, Tel-Aviv University, Ramat Aviv 69978, Israel;
and 4The Netherlands Brain Bank, The Netherlands
Received August 10, 2003; Accepted August 11, 2003
Abstract
Previously, we evidenced a B →Z helical change in Alzheimer’s brain genomic DNA, leading to a hypoth-
esis that Alzheimer’s disease (AD) etiological factors such as aluminum (Al), amyloid β(Aβ) peptide, and Tau
might play a role in modulating DNA topology. In the present study, we investigated the interaction of Al and
Aβwith DNA. Our results show that Aβ(1–42) could induce a B →ψ(Psi) conformational change in pUC 18
supercoiled DNA (scDNA), Aβ(1–16) caused an altered B-form, whereas Al induced a complex B-C-A mixed
conformation. Ethidium bromide binding and agarose gel electrophoresis studies revealed that Al uncoiled the
DNA to a fully relaxed form, whereas Aβ(1–42) and Aβ(1–16) effected a partial uncoiling and also showed dif-
ferential sensitivity toward chloroquine-induced topoisomer separation. Our findings show for the first time
that Aβand Al modulate both helicity and superhelicity in scDNA. A new hypothetical model explaining the
potential toxicity of Aβand Al in terms of their DNAbinding properties leading to DNA conformational alter-
ation is proposed.
Index Entries: Alzheimer’s disease; pUC18 supercoiled DNA; helical transitions; topoisomer separation;
B-DNA; Z-DNA; ψ-DNA.
Journal of Molecular Neuroscience
Copyright © 2004 Humana Press Inc.
All rights of any nature whatsoever reserved.
ISSN0895-8696/04/22:19–31/$25.00
Journal of Molecular Neuroscience 19 Volume 22, 2004
Introduction
For the first time, we recently evidenced a left-
handed Z-DNA conformation in severely affected
Alzheimer’s disease (AD) brain hippocampal cells,
whereas normal hippocampal brain DNAexhibited
B-DNA conformation and moderately affected AD
brain DNA exhibited a B-Z intermediate conforma-
This paper is dedicated to the memory of the late Prof. M. A. Viswamitra, Indian Institute of Science, Bangalore, India.
M. L. Hegde and S. Anitha have contributed equally to this paper.
*Author to whom all correspondence and reprint requests should be addressed. E-mail: kjr4n@yahoo.com
20 Hegde et al.
Journal of Molecular Neuroscience Volume 22, 2004
tion (Anitha et al., 2002). Based on these findings,
we hypothesized that AD-associated molecules like
Al, amyloid β(Aβ) peptide, and Tau might play a
pivotal role in modulating DNA topology in AD
brain (Anitha et al., 2002). The puzzling question,
however, was how the nuclear localization of Aβ
might play a role in bringing about changes in DNA
topology. There were two previous reports on Aβ
immunoreactivity in the nuclear envelopes of P19
cells (Grant et al., 2000) and intranuclear accumula-
tion of Aβin AD brain (Gouras et al., 2000). In the
present study, we provided new evidence for Aβ
immunoreactivity in the vicinity of DNA in hip-
pocampal cells from AD brain. Nuclear localization
of Al has been well-established in AD brain (Crap-
per et al., 1980; Perl and Brody, 1980). Based on this
evidence for nuclear localization of Aβand Al in AD
brain, we hypothesized that Aβand Al might play
a role in modulating DNA topology and possibly
contribute to the B →Z helical transition associated
with AD (Anitha et al., 2002). To investigate the above
hypothesis experimentally, we have studied the
interaction of Al, Aβ(1–42), and Aβ(1–16) with pUC
18 supercoiled DNA (scDNA). It is quite evocative
to study plasmid scDNA as a model system, in view
of the observation that a vast array of small scDNA
packets have been found to be present in animal and
human cells and are known to be involved in gene
expression (Bauer et al., 1980). These superhelical
packets are proposed to be analogous to plasmid
DNA supercoiling. Hence, the results can be corre-
lated or extended to human brain genomic DNA to
provide an insight into explaining the possible role
of Al and Aβin the progression of AD pathology
with reference to DNAtopology. Aβ, a hallmark fea-
ture of the senile plaque, is a proteolytic product of
the transmembrane β-amyloid precursor protein
(APP) (Balakrishnan et al., 1998). Al, Aβ(1–42) pep-
tide and its fragments, Aβ(1–16), and Aβ(1–28) are
reported to play a critical role in inducing the pathol-
ogy seen in AD (Kang et al., 1987; Savory et al., 1999).
The apparent role of Aβ, especially Aβ(1–42), is now
considered as a unifying pathological feature of
diverse forms of AD (Selkoe, 1996). The neurotoxi-
city of insoluble Aβaggregates has been widely
reported (Clements et al., 1996; Huang et al., 2000).
The present study attempts to investigate the poten-
tial role of Aβand Al in terms of DNA helical alter-
ation. Though the role of Al in AD is a highly
debatable topic, we investigated the effect of Al to
understand the biology of Al in terms of its effect on
DNA. The understanding of the complex biology of
Aβand Al with respect to DNA topological transi-
tion might provide an avenue to explore new insight
into AD pathology.
Materials and Methods
Materials
Three normal human brains (age 60–79 yr) (with
no history of long-term illness, dementia, or neuro-
logical disease), and six AD-affected (70–88 yr) brain
hippocampus samples were obtained from The
Netherlands Brain Bank (NBB). In normal and AD
brains, four each were male. Normal brains were
obtained from persons who died of either cardiac
arrest or accident. The average postmortem time on
average is 6 h. pUC 18 scDNA(cesium chloride-puri-
fied, 90% supercoiled structure) was purchased from
Bangalore Genei, India. Aβ(1–42), Aβ(1–16), agarose,
chloroquine, ethidium bromide (EtBr), and HEPES
were purchased from Sigma (USA). Anti-Aβ(1–42)
antibody was obtained from Chemicon International
(USA).
Al-Maltolate Synthesis
Al-maltolate (Al-maltol) was prepared in our lab-
oratory from aluminum chloride hexahydrate and
maltolate (3-hydroxy-2-methyl-4H-pyran-4-one),
following the method of Finneagan et al. (1986). In
the present study, Al-maltol was used to study DNA-
Al interactions. Al-maltol is highly soluble and
hydrolytically stable from pH 2.0 to 12.0. It over-
comes the hydrolytic speciation chemistry problem
of inorganic Al. This complex of Al can deliver a pre-
dicted amount of free aqueous Al at a physiological
pH (Martin, 1991). The thermodynamically pre-
dicted species that dominate at pH 7.0 for Al-maltol
are Al (mal)2+and Al (mal)22+.
Circular Dichroism (CD) Studies
scDNA was titrated against different concentra-
tions of Al-maltol (5 ×10–7 to 1 ×10–3 M), Aβ(1–42)
(0.8 ×10–7 to 0.8 ×10–4M), and Aβ(1–16) (0.8 ×10–7 to
0.8 ×10–4 M). CD spectra (200–330 nm) were recorded
for each concentration in 10–4 MHEPES buffer (pH
7.0). The CD signal of scDNA was also monitored in
the presence of chloroquine to examine the effect of
uncoiling the DNA on CD signals. The spectra at
each concentration represent an average of four
recordings. The CD spectra were recorded on a
JASCO J 715 Spectropolarimeter at 25°C, with 2-mm
cell length. DNAstructures were characterized using
Aβand Al Affect Supercoiled DNA 21
Journal of Molecular Neuroscience Volume 22, 2004
the references of Gray et al. (1978), Hanlon et al.
(1975), and Shin and Eichhorn (1984).
Agarose Gel Studies
DNA samples (scDNA, scDNA+Al-maltol,
scDNA+Aβ(1–42), and scDNA+Aβ(1–16) com-
plexes) were loaded on 1% agarose gels and elec-
trophoresed at 4 V/cm at room temperature. DNA
concentration loaded in all lanes was 1 µg. Topoiso-
mer separation of scDNA and complexes was car-
ried out in the presence of chloroquine. The samples
were stained with EtBr and photographed.
EtBr Binding Fluorescence Studies
EtBr binding patterns to scDNA, scDNA+Al
maltol, scDNA+Aβ(1–42), and scDNA+Aβ(1–16)
complexes were analyzed by mixing 1:1 (w/w)
DNA/EtBr before measuring fluorescence emission.
DNA/EtBr solutions were excited at 535 nm, and
emission was monitored at 600 nm using the Hitachi
F-2000 Fluorescence Spectrophotometer. The
amount of EtBr bound and the average number of
base pairs per bound EtBr molecule were calculated
using Scatchard’s equation.
Melting Profile
The melting profile curves (1°C/min, 25–95°C)
for scDNA, scDNA+Al maltol, scDNA+Aβ(1–42),
and scDNA+Aβ(1–16) complexes, in 10–4 MHEPES
buffer (pH 7.0) were recorded using the Gilford
Response II UV spectrophotometer with a thermo-
stat. The effect of chloroquine on melting profiles of
the above complexes has also been tested by treat-
ing the complexes with chloroquine (1–10 mg/mL
of sample). Melting temperature (Tm) values were
determined graphically from hyperchromicity
versus temperature plots. The precision in Tmvalues
estimated in triplicate was ± 0.5°C.
Clinical and Neuropathological Diagnostic
Criteria Applied in the Rapid Autopsy
System of the NBB
Human brain material is obtained via the rapid
autopsy system of the NBB, which supplies post-
mortem specimens from clinically well-documented
and neuropathologically confirmed cases. Autop-
sies are performed on donors from whom written
informed consent has been obtained either from the
donor or direct next of kin. The demented patients
are clinically assessed, and the diagnosis of “prob-
able Alzheimer’s disease” is based on exclusion of
other possible causes of dementia by history, phys-
ical examination, and laboratory tests.The clinical
diagnosis is performed according to NINCDS-
ADRDA criteria (McKhann et al., 1984), and the
severity of the dementia is estimated according to
the Global Deterioration Scale (Reisberg et al., 1982).
The control subjects have no known history or symp-
toms of neurological or psychiatric disorders.
Once the brain is removed, it is macroscopically
examined and immediately dissected following a
standard protocol. The specimens are frozen rapidly
in liquid nitrogen and stored at –80°C. All cases are
neuropathologically confirmed using conventional-
histopathological stains on formalin-fixed speci-
mens. The diagnosis is based on the presence and
distribution of the classical hallmarks for each of the
disease groups investigated. The NBB uses a scor-
ing system in which the density of senile plaques,
neurofibrillary tangles (NFTs), disrupted interneu-
ronal network (dINN), neuropil threads, congo-
phylic plaques, and vessels are estimated in Bodian
and Congo stains in four neocortical areas: frontal;
temporal; parietal; and occipital. For the staging of
the various pathological hallmarks, a combination
of a grading system and Braak staging is applied to
all specimens (Braak and Braak, 1991; Braak et al.,
1996; Mirra, 1997; Ravid et al., 1998).
In addition, ApoE allele frequency is determined
for each case as a possible risk factor for AD (Nielsen
et al., 2003). All cases in the disease groups and con-
trols are well matched for various factors, both ante-
mortem and postmortem. Antemortem factors
include age, sex, agonal state, seasonal alterations,
circadian variation, clock time of death, and med-
ication. Postmortem factors include postmortem
delay, fixation, and storage time and laterality (Ravid
et al., 1992, 1995; Ravid and Winblad, 1993). Sam-
ples are also controlled for quality by monitoring the
agonal state of the deceased prior to death (Ravid et
al., 1992, 1995).
Quantification of Apoptosis
in Hippocampal Neurons
Fifteen-micron brain sections were cut from the
hippocampal region of three normal and six AD-
affected human brains. Brain sections were first
stained for apoptosis using the fluorescence TUNEL
(terminal deoxynucleotidyl transferase [TdT]-
mediated dUTP nick-end labeling) technique
(Promega apoptosis detection kit, Madison, WI)
(Rao et al., 1998), costained for Aβ(1–42) using
22 Hegde et al.
Journal of Molecular Neuroscience Volume 22, 2004
monoclonal antibody, and developed by
3,3’-diaminobenzidine (DAB) reagent (Sigma,
Aldrich). Studies were also conducted for colocal-
ization of Aβin nonapoptotic cells. The apoptotic
detection system utilizes fluorescein, which mea-
sures the fragmented DNA of the apoptotic cells by
the catalytic incorporation of fluorescein-12-dUTP
at the 3’-OH DNA ends with the enzyme TdT. The
latter forms a polymeric tail by applying the TUNEL
assay. The fluorescein-12-dUTP-labeled DNAis then
visualized directly by fluorescence microscopy. Sec-
tions were photographed at 520 nm under ×400 mag-
nification using a fluorescence microscope. The slides
were scanned using Adobe Photoshop 3.0 and a
Nikon scanner (Nikon LS-3510 AF, 365-mm film scan-
ner, no. 6070192). The captured images were used
for quantification. The cells, which were stained with
flourescein (green), were counted as positive for
apoptosis based on the following criteria as reported
earlier by Rao et al. (1998): (1) DNA fragmentation
with no indication of chromatin margination or other
morphologic changes in the nucleus; (2) markedly
condensed nuclei with a nuclear diameter of 2.5 µm
or more; (3) swollen nuclei containing cytoplasmic
fragments of DNA; and (4) intracellular or extracel-
lular chromatin fragments. Aminimum of 18 fields
were photographed at an initial magnification of
×400 from hippocampal (CA1) regions of each brain
section. Six sections from each brain were evaluated.
Atotal area of 1 mm2for each section was evaluated.
The number of apoptotic cells was quantified by two
independent observers, and the interobserver vari-
ability approximated 2%. The number of apoptotic
cells was represented as cells/mm2.
Colocalization of Apoptosis
and AβImmunoreactivity
The hippocampus sections were first stained for
apoptosis using the TUNEL procedure, followed by
costaining for Aβimmunoreactivity, using anti-Aβ
monoclonal antibody, and reactivity was developed
using the ABC kit. In brief, the sections, prestained
for apoptosis, were incubated overnight at 4°C
with anti-Aβ(1–42) antibody. The bound immuno-
globulins were detected using the avidin-biotin-
peroxidase complex method (Vectastain ABC kit,
Vector Laboratories, Burlingame, CA) and visual-
ized by incubation with 0.05% DAB. The colocal-
ization of apoptosis and Aβin single hippocampal
neuron in CA1 region was imaged by a laser-fitted
confocal microscope.
Results
CD Studies
In an attempt to examine the potential role of Aβ
and Al in modulating DNAtopology with relevance
to the Z-DNA conformation found in AD brain, we
employed the CD method. scDNA was treated with
Al-maltol, and changes in DNA helicity were mon-
itored by analysis of CD spectra. As shown in Fig.1a,
CD spectra of scDNA showed B-DNAconformation
with a characteristic positive peak at 275 nm and
negative peak at 245 nm.Addition of increasing
concentrations of Al-maltol (from 5 ×10–7 to
1 ×10–3 M) to the DNAcaused a decrease in the pos-
itive band at 275 nm with no concomitant change in
the magnitude of the negative peak at 245 nm (Fig.
1b–d). In addition, there was a severalfold increase
in the magnitude of the negative peak at 210 nm.
This characteristic decrease in the positive peak at
275 nm, without change in the magnitude of the neg-
ative peak at 245 nm, reveals a C-DNA form; how-
ever, according to Maestre and Wang (1971), the
lowering of superhelical density of the scDNA also
involved a similar spectral change. To exclude this
possibility, we studied CD spectra of scDNA in the
presence of chloroquine (10 µg/mL), which uncoils
the DNA, like topoisomerase I. No change was
observed in the positive or negative peak on com-
plete uncoiling of the scDNA with chloroquine, and
the spectra were similar to that of control scDNACD
spectra (data not shown). It is clear from the above
experiments that the relaxation of supercoils is not
associated with decrease in positive peak intensity
at 275 nm; hence, we conclude that the changes
observed in the above spectra are not the result of a
simple uncoiling phenomenon but a conformational
change caused by Al-maltol. This conformation of
DNAbelongs to C family; however, the deep, narrow
negative peak at 210 nm is associated with A-DNA
(Gray et al., 1978). In view of this, we favor the inter-
pretation that the spectrum reflects the properties of
a B-C-A mixed conformation. A possible simultane-
ous helical transition of B →C and B →Acould have
been formed leading to a complex conformation.
Hanlon et al. (1975) have reported a similar pattern
of transition in calf thymus DNA.
Next we examined the effect of increasing con-
centrations of Aβ(1–42) on the ellipticity values of
scDNA. As shown in Fig. 2, substantial changes
of the CD spectra were observed upon interaction of
scDNA with Aβ(1–42) (0.8 ×10–7 to 0.8 ×10–4 M). At
Aβand Al Affect Supercoiled DNA 23
Journal of Molecular Neuroscience Volume 22, 2004
Aβ(1–42)/DNA molar ratios lower than 0.1, a DNA
secondary structural transition from the native B-
DNAto the C motif was observed. The spectral changes
involved decrease in a positive band at 275 nm with
no concomitant change in the magnitude of the neg-
ative peak at 245 nm (Fig. 2A). The spectral modifi-
cations of the positive peak were continuous with
the increasing Aβ(1–42) concentrations. The limit C-
DNA motif is characterized by a small positive CD
band at 275 nm and a long negative signal at 245 nm.
However, on the addition of a higher concentration
of Aβ(1–42) [Aβ(1–42)/DNA ratio > 0.1], the nega-
tive CD band extended in the nonabsorbing region
in the form of CD tails, with a large CD magnitude
compared to the intrinsic CD of scDNA, which is
intriguing. This spectral change presumably reflects
the asymmetric compaction of scDNA by Aβ(1–42)
to form ψ(+) DNA. This pattern of CD signal (Fig. 2B)
is a typical characteristic of the (+)ψform of DNA
(Shin and Eichhorn et al., 1984). All of the spectra
showed an isodichroic point at the cutting point of
215 nm. Examination of DNA ellipticities, as shown
in Fig. 2C,D, indicated the transformation of native
B-DNA to C-DNAto ψ-DNA in scDNA induced by
Aβ(1–42). Thus, Aβ(1–42) caused B →C →ψ
conformation in scDNA.
Next we examined the effect of the different con-
centrations (0.8 ×10–7 to 0.8 ×10–4 M) of the shorter
Aβ(1–16) peptides on scDNA conformation. Higher
concentration (0.8 ×10–4 M) of Aβ(1–16) modi-
fied the scDNA to an unusual altered B-form
(Fig. 3). Addition of Aβ(1–16) caused a decrease in
the intensity of the positive peak centered around
275 nm, with a concomitant reduction of the nega-
tive peak around 245 nm. However, there was a sig-
nificant increase in the negative band intensity at
205 nm, indicating a modification in the usual B-sec-
ondary structure. The cutting point has been shifted
from 220 to 224 nm as the concentration of Aβ(1–16)
is increased. All of the spectra have two isodichroic
points at 228 and 252 nm.
Agarose Gel Electrophoresis Studies
These studies were conducted to monitor the
uncoiling process of scDNA. scDNA (1 µg) was
incubated with Al-maltol (1 ×10–3 M), Aβ(1–42)
(0.8 ×10–4 M), and Aβ(1–16) (0.8 ×10–4 M) and sub-
jected to agarose gel electrophoresis (Fig. 4).
Al-maltol uncoiled the scDNA nearer to the fully
relaxed form (lane B). Aβ(1–42) and Aβ(1–16) par-
tially uncoiled the DNA(lanes C and D, respectively).
To study the sensitivity of the complexes to chloro-
quine-induced topoisomer separation and to obtain
information on the stability of the complexes, gel elec-
trophoresis was carried out in the presence of chloro-
quine (Fig. 5). A fair degree of sensitivity was
Fig. 1. CD spectra of Al-maltol interaction with scDNA. Effect of Al-maltol binding on conformation of pUC 18 scDNA
in 10–4 MHEPES buffer (pH 7.0). scDNA was titrated against increasing concentrations of Al-maltol, and the changes in
molar ellipticities were monitored at 250°C in a 2-mm path length cell. The scDNA alone showed B-DNA conforma-
tion (a) On addition of increasing concentrations of Al-maltol (1 ×10–6, b; 1 ×10–4, c; 5 ×10–3, d MAl), the native
B-DNA conformation was transformed into a complex B-C-A mixed conformation.
24 Hegde et al.
Journal of Molecular Neuroscience Volume 22, 2004
observed toward chloroquine for DNA+Aβcom-
plexes (lanes C and D, respectively) compared to
DNA alone (lane A). The topoisomer separation was
observed for scDNA at 1 µg/mL of chloroquine,
whereas Aβ(1–42) and Aβ(1–16) DNA complexes
were relaxed to topoisomers at much lower amounts
of chloroquine (0.8 µg/mL). Chloroquine had no
effect on the DNA+Al complex, as Al-maltol had
totally relaxed the scDNA (lane B).
EtBr Binding Studies
The quantification of the uncoiling pattern of
scDNA induced by Al-maltol, Aβ(1–42), and
Aβ(1–16) was studied by measuring EtBr fluores-
cence intensity at 1:1 (w/w) DNA/EtBr concentra-
tions (Fig. 6). EtBr fluorescence of scDNAwas 13.74
(Fig. 7a), whereas it was 5.65 for the DNA+Al-maltol
complex (Fig. 6B). The significant decrease in the
fluorescence intensity in the case of the DNA+
Al-maltol complex might be the result of the relax-
ation of the scDNA. The fluorescence intensity for
the scDNA+Aβ(1–42) complex was 8.81 (Fig. 6C), and
this might be attributed to ψ-DNAconformation. The
fluorescence intensity for the scDNA+Aβ(1–16) com-
plex was 17.81 (Fig. 6D), and this might be the result
of the modified B-form of the scDNA. The Scatchard’s
plot analysis of the values of the EtBr binding to
the scDNA indicated that the average number of
base pairs per bound EtBr molecule was 6.85 for
scDNA, 86.6 for the scDNA+Al complex, 15.7 for
the scDNA+Aβ(1–42) complex, and 4.6 for the
scDNA+Aβ(1–16) complex.
Fig. 2. The interaction of Aβ(1–42) with scDNA. The changes in the ellipticities at 275 () and 220 nm () were
expressed as percentage of its value for control (native B-DNA) and plotted against the molar ratio of Aβ(1–42)/scDNA
(C,D). The native B-DNA was transformed to a limit C-motif at a Aβ(1–42)/scDNA molar ratio lower than 0.1 (A). At a
higher Aβ(1–42)/scDNA ratio (<0.1), the C-DNA was further converted into an asymmetrically condensed ψ-DNA (B).
Notably, the significant increase in the magnitude of negative CD signal (B,D) of scDNAat Aβ(1–42)/scDNA (<0.1) indi-
cates the gradual transformation of C-DNA to ψ-DNA.
Aβand Al Affect Supercoiled DNA 25
Journal of Molecular Neuroscience Volume 22, 2004
Fig. 3. The interaction of Aβ(1–16) with scDNA (25 ×10–6 g). Aβ(1–16) alters the B-DNA conformation of scDNA to
modified B-DNA. Experimental procedures for CD were as mentioned earlier. (a) scDNA; (b) 5 ×10 –7 MAβ(1–16); (c)
1 ×10–5 MAβ(1–16); (d) 0.8 ×10–4 MAβ(1–16).
Fig. 4. Effect of Al-maltol, Aβ(1–42), and Aβ(1–16) bind-
ing on scDNA mobility in 1% agarose gel. scDNA was incu-
bated in the absence and presence of Al-maltol and
Aβpeptides for 12 h and then was separated by 1% agarose
gel electrophoresis. scDNA alone (cesium chloride
purified) showed >90% superhelicity (lane A). Al-maltol
(5 ×10– 3 M) uncoiled the scDNA nearer to fully relaxed
form (lane B), whereas Aβ(1–42) (0.8 ×10–4 M) and Aβ(1–16)
(0.8 ×10–4 M) partially uncoiled the scDNA (lanes C,D).
DNA concentration loaded in all lanes was 1 µg, and elec-
trophoresis was carried out at 4 V/cm at room temperature.
The samples were stained with EtBr.
Figure 5. Sensitivity for chloroquine-induced topoiso-
mer separation. scDNA was incubated with chloroquine in
the absence and presence of Al-maltol and Aβpeptides.
The effective topoisomer separation was observed for scDNA
at 1 µg/mL chloroquine (lane A), whereas Aβ(1–42) and
Aβ(1–16) DNA complexes were relaxed to topoisomers at
much lower chloroquine levels (0.8 µg/mL of sample volume)
(lanes C,D). Chloroquine had no effect on the DNA-Al com-
plex (lane B).
26 Hegde et al.
Journal of Molecular Neuroscience Volume 22, 2004
Melting Profile
The melting temperature study provides insight
on the stability of DNA. Thus, in the next set of exper-
iments, we examined the melting profile of scDNA
in the absence or presence of Al-maltol or Aβ. The
Tmvalue for scDNA was 59.50°C (±0.5), and its melt-
ing profile revealed an unusual monophasic pattern
(Fig. 7a). Al-maltol (at concentrations 10–5 to 10–4 M)
enhanced the Tmto 66.70°C (±0.5) and at higher
(5 ×10 –3 M) concentration did not alter the Tm
(Fig. 7b). These kinds of changes in Tmare typical
for B →Ahelical transitions (Champion et al., 1998).
The Tmfor the scDNA+Aβ(1–42) complex showed a
typical biphasic melting pattern with two Tmvalues
(59°C and 88°C) (Fig. 7c). This is the first report study-
ing the melting pattern of ψ(+) DNA induced by
Aβ(1–42). Aβ(1–16) did not alter the Tmat lower con-
centrations (10–6 M), whereas it enhanced the Tmto
71.4°C (±0.5) at higher (10–4 M) concentrations. The
scDNA+Aβ(1–16) complex showed a customary
monophasic melting profile (Fig. 7d). To further
explore the nature of the unusual monophasic melt-
ing profile of scDNA and to see the effect of uncoil-
ing the scDNA on its Tm, we studied the melting
pattern of the scDNA in the presence of chloro-
quine. There was no change in the Tm(59.5°C ±0.5)
in scDNA treated with chloroquine at 1 µg/mL
(topoisomer separation concentration as confirmed
by agarose gel) and 10 µg/mL (complete relaxation
of superhelicity to linear DNA) concentration (data
not shown). These results indicate that the unusual
monophasic melting profile observed in the case of
scDNA might not be the result of supercoiling,
because the same melting pattern was obtained with
linearized scDNA as well.
Gallium Chloride Does Not Affect
scDNA Conformation
To examine whether the effect of Al on scDNA
conformation is specific or a more general effect of
metals that have a similar ionic radius and belong
to the same group (group 13), we studied the effect
of gallium, another group 13 metal having essen-
tially a similar chemical behavior in aqueous solu-
tion as Al. The ionic radii of Al and gallium are
0.57 Å and 0.62 Å, respectively. Thus, we compared
the effect of gallium chloride on scDNA to those of
Al-maltol. These experiments were carried out in a
similar manner (methods used and concentration
range) as those described above for Al-maltol. The
results showed that in contrast to Al-maltol, gallium
chloride did not alter either the conformation or the
superhelicity of scDNA. In addition, gallium chlo-
ride did not change the Tmof the scDNA (data not
shown). At higher concentrations, gallium caused
aggregation of scDNAas observed by the zeroing or
flattening of the CD signal. Taken together, these
results suggest that the effect of Al on DNAconfor-
mation does not represent a general effect of group
13 metals but rather is a specific feature of Al.
Localization of Aβin the Nuclei of CA1 Neurons
in Hippocampal Region of Brain
In previous studies we showed that hippocampal
DNA obtained from AD brain samples has a prefer-
entially left-handed Z-DNA conformation (Anitha
et al., 2002), and our present results show that Aβ
induces similar (?) conformational changes in vitro.
Moreover, it was shown that Aβinduces apoptosis
in cell lines and rabbit brains (Clements et al., 1996;
Selkoe, 1996; Huang et al., 2000). Thus, we hypoth-
esize that Aβis presented in the neuronal nuclei of
affected AD brain areas, where it might induce DNA
conformational changes. These changes might in
turn lead to apoptotic neuronal death. In an attempt
to examine whether Aβ(1–42) is present in neuronal
nuclei in AD brains and explore its relationship to
apoptosis, the presence of Aβ(1–42) immunoreac-
tivity and apoptotic nuclei (TUNEL positive) was
examined in neurons of the hippocampal CA1 region
Fig. 6. EtBr binding pattern. Equal concentrations (w/w)
of DNA and EtBr were used as models to study the effect of
EtBr intensity on Al-maltol and Aβpeptides. EtBr intensity
also provides information on DNA conformation. The
uncoiling of scDNA was quantified by measuring the EtBr
fluorescence intensity of 1:1 (w/w) DNA/EtBr solutions. The
solutions were excited at 535 nm and emission monitored
at 600 nm. The emission intensity values for scDNA were
13.74 (A); scDNA+Al, 5.65 (B); scDNA+Aβ(1–42), 8.81 (C);
and scDNA+Aβ(1–16), 17.81 (D).
Aβand Al Affect Supercoiled DNA 27
Journal of Molecular Neuroscience Volume 22, 2004
of normal and AD brains. The number of Aβand/or
TUNEL-positive cells counted per square millime-
ter of brain section is given in Table 1. Quantifica-
tion of data revealed that in normal brain
hippocampal sections, 25% of the cells were apop-
totic, whereas in the case of AD brain, about 75% of
the cells were apoptotic. When normal brain hip-
pocampal sections were tested for colocalization of
apoptosis (TUNEL positive) and Aβimmunoreac-
tivity, we could not find a TUNEL-positive cell that
was also Aβpositive. But in the AD brain sections,
out of 100 TUNEL-positive cells counted, 50% of cells
were found to be positive also for Aβ. These results
suggest that Aβis deposited in the vicinity of DNA
in the nuclear region of AD cells. In the case of normal
brain sections, Aβ(1–42) immunoreactivity was not
observed in either apoptotic or nonapoptotic hip-
pocampal neurons. Figure 8A presents a confocal
image of a single representative neuron, showing
absence of Aβdeposition in an apoptotic nucleus
from the hippocampal section of normal brain. Figure
8B shows the localization of Aβimmunoreactivity
in the apoptotic nucleus of hippocampal neuron in
the CA1 region.
Discussion
AD is associated with several complex neu-
ropathological events like deposition of Aβ, abnor-
mal phosphorylation of Tau, aggregation of these
proteins into NFTs, oxidative stress, and DNA
damage (Iqbal et al., 1994; Lyras et al., 1997). It is of
interest to mention that Aβ(Gouras et al., 2000; Grant
et al., 2000; present study) and Al (Crapper et al.,
1980; Perl and Brody, 1980) were found to be local-
ized in the chromatin region of nuclei. Recently, our
team first evidenced the presence of left-handed,
rigid Z-DNA in severely affected AD brain and a
B-Z intermediate DNA conformation in moderate
AD. In contrast, normal young and aged brains have
usual/right-handed Watson-Crick DNA conforma-
tion (Anitha et al., 2002). It has also been hypothe-
sized that the prime etiological factors of AD, such
as Aβ, Tau, and Al (a highly debatable etiological
factor), might be playing a role in right- to left-handed
helical change associated with AD. Based on this
DNA conformational change, we hypothesized an
explanation for the unusual phenomenon-like nucle-
osome misassembly, G*-specific DNA oxidation,
Fig. 7. Effect of Al and Aβon melting profile of scDNA. The UV absorbance at 260 nm was recorded for scDNA and
scDNA complexes with Al-maltol and Aβpeptides at different temperatures. The melting curves (10°C/min, 25–95°C)
for scDNA and other complexes in 10–4 MHEPES, pH 7.0, were recorded in a UV spectrophotometer with a thermo-
stat. Tmvalues were determined from hyperchromicity vs temperature plots. Tmfor scDNA was 59.5°C (a).
Al-maltol initially enhanced the Tm(not shown in the graph) but did not alter at higher concentrations (b). scDNA+Aβ(1–42)
showed a biphasic melting pattern (c); Aβ(1–16) enhanced the Tmof scDNA (d). The results are representative of three
independent experiments.
28 Hegde et al.
Journal of Molecular Neuroscience Volume 22, 2004
terminal differentiation, and altered gene expression
associated with AD brain (Anitha et al., 2001, 2002).
There were few reports on the binding ability of Al
to DNA (Karlik et al., 1980; Rao et al., 1993; Rao and
Divaker, 1993); and our team first evidenced that Al
could strongly bind to AT*-specific oligomers (Rajan
et al., 1996), whereas Al could bring structural tran-
sition from Z-Aconformation in GC*-rich oligomers
(Champion et al., 1998). However, there is no report
to date on Aβbinding to DNA. In the present study,
we first evidenced the interaction of Aβand Al with
scDNA. We found that Aβ(1–42) not only binds to
scDNA but also is able to alter the conformation of
DNA. An initial B→C transition was observed, which
gradually transformed into ψ-DNA, presumably
reflecting a partial DNA collapse into a ψ-phase
(Zuidam et al., 1999). In ψ-DNA, the DNA molecules
are tightly packed into a toroidal superhelical bundle
whose chiral sense is defined by intrinsic DNAhand-
edness. Specifically, right-handed secondary con-
formations, such as B and C motifs, stabilize a
left-handed tertiary conformation (Reich et al., 1994).
Such a tightly packed left-handed DNA organiza-
tion exhibits negative CD signals whose magnitude
is larger than that characterizing dispersed DNA
molecules, which lack a tertiary structure (Fig. 3b).
The ψ-DNA conformation induced by Aβ(1–42) is
structurally closer to Z-DNA, which was observed
in severely affected AD brain (Anitha et al., 2002).
Studies by Thomas and Thomas (1989) showed
clearly that ψ-DNA, an ordered, twisted, tight-pack-
ing arrangement of the double helix, is structurally
and immunologically closely related to the Z-DNA
family. It is left-handed in conformation like Z-DNA.
This evidently indicates that DNA topological
changes induced by Aβare similar to the changes
seen in AD brain DNA. Al, another strong and highly
debatable etiological factor, could play a key role in
AD pathology by contributing to complex DNAheli-
cal transitions (B →A; B →C, or B-C-A) of scDNA,
and this complex DNAconformation is energetically
weak and is likely to go into Z-DNA conformation
as reported in AD brain (Anitha et al., 2002). Vast
arrays of small scDNA packets have been found to
be present in animal and human cells and are known
to be involved in gene expression (Bauer et al., 1980).
Al at a physiological concentration (10–6 M) was
reported to irreversibly unwind the scDNA (Rao et
Table 1
Number of Apoptotic and Aβ-Positive Cells in Normal and AD-Affected Human Brain Samples
Normal Ad
Brain region TUNEL AβTUNEL + AβTUNEL AβTUNEL + Aβ
Hippocampus 15 ± 1.0 0 0 100 ± 5.0 75 ± 2.0 50 ±4.0
TUNEL and Aβ-positive cells in hippocampal sections were counted microscopically in square millimeters. The results
presented were obtained from sections of each of three brains for normal and six sections of each of six brains for AD. Values
are expressed as mean ± S.D.
Fig. 8. Localization of Aβin human normal and AD brain samples. Frozen sections of 15 µm were cut from hip-
pocampal brain regions and stained for apoptosis using the fluorescence TUNEL technique, costained for Aβ(1–42) with
a monoclonal antibody, and developed with a DAB reagent. Colocalization of apoptosis and Aβimmunoreactivity in
the vicinity of DNA in a nucleus of hippocampal neuron in the CA1 region was imaged by confocal microscopy. (A)
Normal brain apoptotic nuclei showing absence of Aβimmunoreactivity. (B) AD brain apoptotic nuclei from
hippocampal region showing Aβimmunoreactivity. The →indicates Aβimmunoreactivity.
Aβand Al Affect Supercoiled DNA 29
Journal of Molecular Neuroscience Volume 22, 2004
al., 1993). Our study also shows that Al uncoils the
scDNA to a fully relaxed form, whereas Aβ(1–42)
and Aβ(1–16) relax the DNApartially. Thus, the pre-
sent study hearsay for the first time that Al and Aβ
uncoil scDNA besides bringing about helicity
changes. Another interesting feature observed is the
sensitivity of DNA+Aβ(1–42) and DNA+ Aβ(1–16)
complexes to chloroquine, a drug that mimics topoi-
somerase I in action. This observed sensitivity indi-
cates a possible alteration in DNA replication and
gene expression in the cells.
The possible complex role of Al and Aβin mod-
ulating DNAhelicity with relevance to AD is hypoth-
esized in Fig. 9. We propose that neuropathological
factors like Al and Aβmight modulate DNA topol-
ogy in AD brain. In stage I, the complex conforma-
tional changes (ψ-DNA, B →A, B →C, or B-C-A,
altered B) observed experimentally in scDNA,
induced by Al, Aβ(1–42), and Aβ(1–16), are pre-
sumably the early events in AD pathology. In mod-
erately affected AD brain, the DNA has a B-Z
intermediary conformation and other intermediary
complex conformations might exist (Anitha et al.,
2001, 2002). In stage II, secondary factors such as
oxidative stress, cell shrinkage, ionic imbalance, and
polyamines are likely to play a role in converting
these intermediary complex conformations to rigid,
left-handed Z-DNA. Accordingly, we reported the
presence of left-handed Z-DNAin severely affected
AD brain (Anitha et al., 2002). We propose that DNA
topological changes play a role in AD progression.
These novel observations and further work in this
direction may have important implications for aiding
in our understanding of toxicity of Aβand Al in terms
of their direct role in altering DNAconformation and
its relevance in neurodegeneration occurring in AD.
Acknowledgments
The authors wish to thank Dr. V. Prakash, Direc-
tor, CFTRI (Mysore) for all of his support and encour-
agement. The authors also thank Prof. S. Sharath
Chandra, Centre for Human Genetics (Bangalore)
for his support. M. L. H. thanks the Council for Sci-
entific and Industrial Research (CSIR) for awarding
the Junior Research Fellowship; S. A., K. S. L., and
Fig. 9. Hypothesis. Possible role of Al, Aβ(1–42), and Aβ(1–16) in modulating scDNA topology with relevance to AD.
30 Hegde et al.
Journal of Molecular Neuroscience Volume 22, 2004
M. S. M. are grateful to CSIR for awarding the Senior
Research Fellowships. This work was supported by
a grant from the Ministry of Science, Culture, and
Sports (Israel) and the Department of Biotechnology
(India). We profoundly thank Prof. Surolia, Indian
Institute of Science, Bangalore, India, for allowing
us to use CD facility and The Netherlands Brain Bank
for providing the human brain specimens.
References
Anitha S., Rao K. S. J., Latha K. S., and Viswamitra M. A.
(2002) First evidence to show the topological change
of DNA from B-DNA to Z-DNA conformation in the
hippocampus of Alzheimer’s brain. J. Neuromol. Med.
2, 287–295.
Anitha S., Reuvin S., Rao K. S. J., Latha K. S., and Viswami-
tra M. A. (2001) Alink between apoptotic DNA damage
and DNA topology in Alzheimer’s disease brain: a
hypothesis. Alz. Rep. 4, 121–131.
Balakrishnan R. Parthasarathy R., and Sulkowski E. (1998)
Alzheimer’s β-amyloid peptide: affinity for metal chela-
tors. J. Peptide Res. 51,91–95.
Bauer W. R. Crick F. H. C., and White H. (1980) Super-
coiled DNA. Sci. Am. 16, 243–245.
Braak H. and Braak E. (1991) Neuropathological stageing
of Alzheimer’s-related changes. Acta Neuropathol. 82,
239–259.
Braak H. Braak E., Bohl J., and Reintjes R. (1996) Age, neu-
rofibrillary changes, A beta-amyloid and the onset of
Alzheimer’s disease. Neurosci. Lett. 210, 87–90.
Champion C. S., Kumar D., Rajan M. T., Rao K. S. J.,
and Viswamitra M. A. (1998) Interaction of Co, Mn,
Mg, and Al with d(GCCCATGGC) and d(CCGGGCC
CGG): a spectroscopic study. Cell. Mol. Life Sci. 54, 488–496.
Clements A., Allsop D., Walsh D. M., and Williams C. H.
(1996) Aggregation and metal-binding properties of
mutant forms of the amyloid beta peptide of
Alzheimer’s disease. J. Neurochem. 66, 740–747.
Crapper D. R., Quittkat S., Krisnan S. S., Dalton A. J., and
DeBoni U. (1980) Intranuclear aluminum content in
Alzheimer’s disease, dialysis encephalopathy, and
experimental aluminum encephalopathy. Acta Neu-
ropathol. 50, 19–24.
Finneagan M. M., Retigg S. J., and Orvig C. (1986) Aneu-
tral water soluble aluminum complex of neurological
interest. J. Am. Chem. Soc. 108, 5033–5035.
Gouras G. K., Tsai J., Naslund J., Vincent B., Edgar M.,
Checher F., et al. (2000) Intraneuronal Aβ42 accumu-
lation in human brain. Am. J. Pathol. 156, 15–20.
Grant S. M., Ducatenzeiler A., Szyl M., and Cuello A. C.
(2000) Aβimmunoreactive material is present in all
intracellular compartments in transfected neuronally
differentiated, P19 expressing the human amyloid
β-protein precursor. J. Alzheimer’s Dis. 2, 207–222.
Gray D. M., Taylor T. N., and Lang D. (1978) Dehydrated
circular DNA: circular dichroism of molecules in
ethanol solutions. Biopolymers 17, 1–45.
Hanlon S., Brudno S., Wu T. T., and Wolf B. (1975) Struc-
tural transitions of deoxyribonucleic acid in aqueous
electrolyte solutions: reference spectra of conforma-
tional limits. Biochemistry 14, 1648–1660.
Huang H. M., Ou H. C., and Hsieh S. J. (2000) Antioxi-
dants prevent amyloid peptide-induced apoptosis and
alteration of calcium homeostasis in cultured cortical
neurons. Life Sci. 66, 1879–1892.
Iqbal K., Alonso A. C., Gong C. X., Khatoon S., Singh T. J.,
and Grundke-Iqbal I. (1994) Mechanism of neuro-
fibrillary degeneration in Alzheimer’s disease. Mol.
Neurobiol. 9, 119–123.
Kang J., Lemaire H. G., Unterbeck A., Salbaum J. M., Mas-
ters C. L., Grzeschik K. H., et al. (1987) The precursor
of Alzheimer’s disease amyloid A4 protein resembles
a cell-surface receptor. Nature 325, 733–736.
Karlik S. J., Eichhorn G. L., Lewis P. N., and Crapper D.
R. (1980) Interaction of aluminum species with deoxyri-
bonucleic acid. Biochemistry 19, 5991–6001.
Lyras L., Cairns N. J., Jenner A., Jenner P., and Halliwell
B. (1997) An assessment of oxidative damage to pro-
teins, lipids, and DNA in brain from patients with
Alzheimer’s disease. J. Neurochem. 68, 2061-2069.
Maestre M. F., and Wang J. C. (1971) Circular dichroism
of superhelical DNA. Biopolymers 10, 1021.
Martin R. B. (1991) Aluminum in biological systems, in
Aluminum in Chemistry, Biology, and Medicine, Nicolini,
P., Zatta, P. F., and Corain, B. eds., Raven, New York,
pp. 3–20.
McKhann G., Drachman D., Folstein M., Katzman R., Price
D., and Stadlan E. (1984) Clinical diagnosis of
Alzheimer’s disease: report of the NINCDS-ADRDA
work group under the auspices of Department of
Health and Human Services Task Force on Alzheimer’s
disease. Neurology 34, 939–944.
Mirra S. S. (1997) The CERAD neuropathology protocol
and consensus recommendations for the postmortem
diagnosis of Alzheimer’s disease: a commentary. Neu-
robiol. Aging 18 (Suppl.), S91–94.
Nielsen S. A., Ravid R., Kamphorst W., and Orgensen
O. S. (2003) Apoliprotein E e4 in an autopsy series of
various dementing disorders. J. Alzheimer’s Dis. 5,
119–125.
Perl D. P., and Brody A. R. (1980) Alzheimer’s disease:
X-ray spectrometric evidence of aluminium accumu-
lation in neuro-fibrillary tangle bearing neurons. Sci-
ence 208, 297.
Rajan M. T., Champion C. S., Kumar D., Vishnuvardhan
D., Rao K. S. J., and Viswamitra M. A. (1996) Interac-
tion of Co, Mn, Mg and Al with d(GCGTACGC): a spec-
troscopic study. Mol. Biol. Rep. 22, 47–52.
Rao K. S. J. and Divaker S. (1993) Spectroscopic studies of
Al-DNA interactions Bull Environ. Contam. Toxicol. 50,
92–99.
Rao K. S. J., Anitha S., and Latha K. S. (2000) Aluminium
induced neurodegeneration in hippocampus of aged
rabbits mimics Alzheimer’s disease. Alz. Rep. 3, 83–88.
Rao K. S. J., Letada P., Haverstick D. M., Herman M. M.,
and Savory J. (1998) Modifications to the in situ TUNEL
Aβand Al Affect Supercoiled DNA 31
Journal of Molecular Neuroscience Volume 22, 2004
method for detection of apoptosis in paraffin-embedded
tissue section. Ann. Clin. Lab. Sci. 28, 131–137.
Rao K. S. J., Rao B. S., Vishnuvardhan D., and Prasad K.
V. S. (1993) Alteration in superhelical state of DNAby
aluminium. Biochem. Biophys. Acta 1172, 17–20.
Ravid R. and Winblad B. (1993) Brain banking in
Alzheimer’s disease: factors to match for, pitfalls and
potentials, in Alzheimer’s Disease: Advances in Clinical
and Basic Research, Corain, B., Iqbal, K., Nicolini, M.,
Winblad, B., Wisniewsky, H., and Zatta, P., eds., John
Wiley and Sons, NY, pp. 213–218.
Ravid R., Van Zwieten E. J., and. Swaab D. F. (1992) Brain
banking and the human hypothalamus—factors to match
for, pitfalls and potentials. Prog. Brain Res. 93, 83–95.
Ravid R., Swaab D. F., Van Zwieten E. J., and Salehi A.
(1995) Controls are what makes a brain bank go round,
in Neuropathological Diagnostic Criteria for Brain Bank-
ing, Biomedical and Health Research, Vol. 10, Cruz-
Sanchez, F. F. Cuzner , M. L., and Ravid, R., eds., IOS
Press, Amsterdam, The Netherlands, pp. 4–13.
Ravid R., Swaab D. F., Kamphorst W., and Salehi A. (1998)
Brain banking in aging and dementia research—the
Amsterdam experience, in Progress in Alzheimer’s and
Parkinson’s disease, Fisher, A., Yorshida, M., and Hanin,
I., eds., Plenum Press, New York, pp. 277–286.
Reich Z., Levin Z. S., Gutman S. B., Arad T., and Minsky
A. (1994) Supercoiling-regulated liquid-crystalline
packaging of topologically-constrained, nucleodsome-
free DNA molecules. Biochemistry 33, 14177–14184.
Reisberg B., Ferris S. H., Del Leon M. J., and Crook T.
(1982) The global deterioration scale for assessment of
primary degenerative dementia. Am. J. Psychol. 139,
1136–1139.
Savory J., Rao K. S. J., Huang Y., Letada P. R., and Herman
M. M. (1999) Age related changes in Bcl-2 and Bax ratio,
oxidative stress, redox active iron, and apoptosis asso-
ciated with aluminium induced neurodegeneration:
increased susceptibility with aging, Neurotoxicology 20,
805–818.
Selkoe D. J. (1996) Amyloid beta-protein and the genetics
of Alzheimer’s disease J. Biol. Chem. 271, 18295–18298.
Shin Y. A. and Eichhorn G. L. (1984) Formation of psi(+)
and psi (–) DNA. Biopolymers 23, 325–335.
Thomas T. J. and Thomas T. (1989) Direct evidence for the
presence of left handed conformation in a supramole-
cular assembly of polypeptides. Nucleic Acids Res. 17,
3795–3810.
Zuidam N. J., Barenholz Y., and Minsky A. (1999) Chiral
DNApackaging in DNA-cationic liposome assemblies.
FEBS Lett. 457, 419–422.
