ELECTROANALYTICAL STUDY OF PROFLAVINE INTERCALATION IN
5-METHYL- OR INOSINE-CONTAINING AMPLICONS
Despina K.Alexiadou a, Andrea K.Ioannou a,
Sofia Kouidou – Andreou b, Anastasios N.Voulgaropoulos a, Stella Th.Girousi a *
a
Laboratory of Analytical Chemistry, Department of Chemistry, Aristotle University
of Thessaloniki, 541 24 Thessaloniki, Greece
b
Laboratory of Biochemistry, Department of Medicine, Aristotle University of
Thessaloniki, 541 24 Thessaloniki, Greece
*e-mail: girousi@chem.auth.gr
Abstract
Amplicons corresponding to the GC-rich p53 exon 5 and its analogues, containing 5methyl-cytosine (60%) instead of cytosine or inosine (60%) instead of guanosine as
well as GC-poor p53 exon 6 were synthesized and investigated electrochemically in
the presence and absence of proflavine by differential pulse voltammetry (DPV). The
incorporation of base analogues and the thermal stability of the resulting amplicons
were testified in the presence of a fluorescent probe (Sybr–Green). Peak current at 1V
was lower for the methylated compared to unmethylated PCR amplicons and was
similarly affected by proflavine intercalation. On the contrary, considerable peak
current differences were observed in the presence of proflavine for unmodified exon 5
v.s. exon 6 or inosine-containing amplicons. Thermal analysis verified the expected
Tm shifts due to the base analogue incorporation and GC-content variations. In
conclusion, methylated and unmethylated PCR amplicons could be distinguished in
model DNA systems using differential pulse voltammetry (DPV), while the use of
proflavine could serve as an electrochemical probe for identifying different DNA
conformations.
Keywords: Amplicons, 5-methyl-cytosine, inosine, melting temperature, DPV,
proflavine
1
Introduction
5-methyl-cytosine is a modified pyrimidine base, which frequently substitutes for
cytosine, particularly in CpG sequences [1-4]. Substitution of cytosine by 5-methylcytosine affects the structural and dynamic properties of the double helix and leads to
helical stabilization [5-7]. The epigenetic DNA modification via global cytosine
methylation is a dynamic mechanism adapted in differentiation and genetic
reprogramming. Methylation differences, extensive hyper- and hypomethylation loci
characterize cancerous versus non-cancerous DNA [8]. The challenging issue of
studying DNA methylation has been delayed due to its loss in amplified DNA such as
products of polymerase chain reaction (PCR), cloning etc. The relatively recent
introduction of a chemical method for differentiating between methylated and
unmethylated cytosines by PCR has greatly facilitated the study of this fundamental
biological mechanism [9]. However, a direct approach for studying the extent of
methylation based on the physical-chemical changes introduced by cytosine
modification, would greatly simplify the evaluation of global DNA methylation.
Inosine (I) is a guanosine derivative formed by hypoxanthine and ribose. In
comparison to the chemical structure of guanosine, inosine lacks an amine group but
can still substitute for guanosine in the helix of DNA. Inosine-cytosine base pairs can
form only two hydrogen bonds instead of the three typical of guanosine-cytosine base
pairs [10]. Inosine became a favorite tool in the examination of the crucial role of
guanine’s exocyclic 2-amino group in DNA recognition by various drugs [11].
Differential pulse voltammetry (DPV) is mainly used in bioelectrochemical
studies. The use of carbon paste electrode (CPE) and transfer differential pulse
voltammetry provides signal of guanine’s oxidation [12] and has been used to
investigate DNA damage involving base residues [13]. Lately, genosensors have been
developed for the detection of DNA hybridization and DNA modifications [14-16].
Melting curve analysis of nucleic acids using photometric or fluorometric
methods is also a useful tool for analyzing nucleic acid biophysical and
conformational parameters, which are frequently also detected by gel electrophoresis.
The Real Time PCR technology and fluorescence technique has facilitated thermal
melting applications involving the study of DNA biophysical parameters such as
DNA annealing, since fluorescence is greatly influenced by the DNA melting process.
Because the melting temperature of nucleic acids is affected by length, GC content
2
and the presence of base mismatches, among other factors, products can often be
distinguished by their melting characteristics [17-20].
Proflavine is a known intercalator whose electrochemical behavior has been
previously studied [21]. Electrochemical studies on the interaction of proflavine with
calf thymus DNA have shown that proflavine intercalates within the DNA helix and
induces conformational changes [22-24]. According to a previous study its binding
affinity to DNA depends upon the structure of the double-stranded polynucleotides
and their ability to be distorted [25]. In this study the interaction of proflavine with
model DNA systems (PCR products – amplicons) of different composition (GC
content) (exon 5 and exon 6 of gene p53) containing different base analogues, such as
methyl-cytosine and inosine was studied using amplicon biosensors.
3
Experimental
Reagents
Natural bases triphosphate monomers (dNTPs) were purchased from Invitrogen (UK).
5-methyl-deoxycytosine-triphosphate
(5-methyl-dCTP)
and
deoxy-inosine-
triphosphate (dITP) monomers were purchased from Roche (Germany). Proflavine
was purchased from Sigma. Stock solutions of proflavine (10-1 M) were prepared in
doubly distilled water and then diluted accordingly, just before use. The supporting
electrolyte of voltammetric experiments was acetate buffer solution 0.2 M at pH 5.0 +
20 mM NaCl. The fluoresence dye used for melting curves was Sybr Green I and was
purchased from Molecular Probes (USA). All other reagents used were of analytical
grade.
Clinical sample and DNA isolation
All tissues were frozen in liquid nitrogen immediately after excision and removal of a
specimen for pathological examination. All specimens used for this study were
obtained from consenting individuals. The tissues were pulverized in liquid nitrogen
and used for isolation of DNA using organic solvents [26].
Instrumentation
Differential pulse voltammetric measurements were performed using a PalmSens
potentiostat controlled by PalmSensPC software (IVIUM Technologies, The
Netherlands). The working electrode was a carbon paste electrode with 3 mm inner
and 6 mm outer diameter of the PTFE sleeve. The reference electrode was Ag/AgCl/3
M KCl and the counter electrode was a platinum wire. A thermocycler was used for
PCR (MJ Research, USA) and Real Time PCR (DNA Engine Opticon ® 2 System
using a CFD-3220 Opticon™ 2 Detector - MJ Research, USA) was used for the
melting curve analysis. Eppendorf centrifuge 5417R was used for the purification of
the PCR products.
PCR amplification of p53 exon 5 DNA and its analogues
DNA (150 ng) was amplified using 1.25 U of Taq DNA polymerase (Recombinant,
Invitrogen) in a tolal volume of 50mL (1.5mM MgCl2, 10mM Tris—HCl pH 8.3,
0.5mM KCl, 0.2mM dNTPs) containing 100 pmol of each primer. p53 ex5 sense:
5'-TTCCTCTTCCTACAGTAC-3'; p53ex5 antisense: 5'-GCCCCAGCTGCTCACC
4
ATCG-3'. Cycling conditions were: 95 ◦C×5 s; 95 ◦C×45 s; 52 ◦C×45 s; 72 ◦C×1 min,
35 cycles, followed by a step at 72 ◦C for 10 min. Amplicons were analyzed by 1%
agarose gel electrophoresis. Amplicons with inosine were synthesized by substituting
dGTP in the dNTPs mixture with dITP (10-80%). Similarly, amplicons with methylcytosine were produced by substituting dCTP with 5-methyl-dCTP (10-80%).
PCR amplification of p53 exon 6 DNA
DNA (150 ng) was amplified using 1.25 U of Taq DNA polymerase (Recombinant,
Invitrogen) in a tolal volume of 50mL (1.5mM MgCl2, 10mM Tris—HCl pH 8.3,
0.5mM KCl, 0.2mM dNTPs) containing 100 pmol of each primer. p53 ex6 sense:
5'-CACTGATTGCTCTTAGGTCTGGC-3'; p53 ex6 antisense: 5'-AGTTGCAAACC
AGACCTCAGGCG-3'. Cycling conditions were: 95 ◦C×5 s; 95 ◦C×45 s; 52 ◦C×45 s;
72 ◦C×1 min, 35 cycles, followed by a step at 72 ◦C for 10 min.
Melting curves of amplicons
DNA melting curves of the native and modified amplicons were acquired on the DNA
Engine Opticon® 2 System by measuring the fluorescence of Sybr Green I during a
linear temperature transition from 50 °C to 95 °C at 0.1 °C/s. Fluorescence data were
converted into melting peaks by the Opticon Monitor software (Ver. 2.02) to plot the
negative derivative of fluorescence over temperature vs temperature (-dF/dT vs T). 19
μL of amplicon was mixed with 1 μL of a 1:100.000 dilution of Sybr Green I before
melting curve analysis.
DPV measurements
Prior DPV measurements, amplicons were purified with cold ethanol and
CH3COONa. 2.5 volumes of cold absolute ethanol and 0.1 volume of 3 M
CH3COONa were added in order to precipitate the oligonucleotides. The precipitated
oligonucleotides were recovered by centrifugation (15 min at 14000 rpm, at 4 oC) and
removal of the supernatant. An extra wash with 70% ethanol was included and
followed by centrifugation (5 min at 14000 rpm, at 4 oC). After the removal of the
supernatant, amplicons were diluted at 10mM Tris-HCl (pH 7) in order to use them in
the electroanalytical study.
5
Adsorptive transfer stripping differential pulse voltammetry was used for the
analysis and carbon paste (CPE) as working electrode. The carbon paste was prepared
in the usual way by hand-mixing graphite powder and nujol oil in a ratio 75:25. The
resulting paste was packed tightly into a Teflon sleeve. Electrical contact was
established with a stainless steel screw. The surface was polished to a smooth finish
before use. Prior to the accumulation step the electrode was pre-treated by applying a
potential of +1.7 V for 1 min on the working electrode dipped in the supporting
electrolyte. The electrochemical pre-treatment produces a more hydrophilic surface
state and a concomitant removal of organic layers.
After the pre-treatment of the CPE, as previously described, amplicons were
immobilized onto the activated electrode surface by adsorptive accumulation for 5
min at +0.5 V. The transduction was carried out in blank solution (only supporting
electrolyte) with differential pulse voltammetry and the following conditions: Ebegin =
0.1 V, Eend = 1.5 V, Estep = 0.005 V, Epulse = 0.025 V, scan rate = 0.01 V·s-1 and tpulse =
0.07 s.
In order to study the intercalation of proflavine, the electrode with the
immobilized amplicon was transferred into the stirred sample solution (analyte plus
supporting electrolyte) for the optimal interaction time. Then the transduction was
carried out in blank solution with the same conditions as mentioned above. Prior to
each medium exchange the electrode was rinsed carefully with water for 5 s. The
interactions between the different types of immobilized amplicons and increasing
concentrations of proflavine in solution were studied.
6
Results and discussion
Synthesis of modified amplicons
Nucleotide analogues were most efficiently incorporated in amplicons when 60% of
the natural bases substituted by the modified bases used. 80% substitution of the
native bases yielded lower amplification efficiencies (results not shown). In the case
of substitution of cytosine by 5-methyl-cytosine, the amplification efficiency was
higher in the presence of 7% DMSO (Fig.1).
Melting curve analysis of amplicons
Incorporation of the base analogues was verified by thermal denaturation of the
resulting amplicons. The melting characteristics of amplicons synthesized in the
presence of various concentrations of 5-methyl-cytosine evaluated by fluorometric
analysis also showed that in the presence of 60% 5-methyl-cytosine in the
amplification mixture, incorporation was maximal and strongly influenced the PCR
product stability [27] (Fig.2). On the contrary, fluorometric results showed that
incorporation of inosine was followed by a decrease of Tm (not shown) related to the
decrease of hydrogen bonds in inosine containing DNA [11].
The melting curves of amplicons of exon 5, exon 6 and modified exons 5
(60% 5-methyl-dCTP instead of dCTP or 60% dITP instead of dGTP) were also
compared (Fig.3). As expected the amplicon containing 5-methyl-cytosine was the
most stable (Tm = 94C), while the inosine-containing amplicon was the least stable
(Tm = 83C). Furthermore, exon 6 (Tm = 85C, G:C content 50%) was less stable
than the G:C-rich exon 5 (Tm = 91C, G:C content 61.7%).
Amplicon-biosensors
The optimal concentration of PCR products for the full coverage of the CPE surface
was around 5·10-3 g.L-1. Native double stranded DNA yielded a positive peak at 1.0
V, which corresponded to the oxidation of the guanine residues [12]. Amplicons also
yielded this positive peak at 1.0 V (Table 1).
The peak current intensity at 1.0 V was around 0.4 μA at the amplicons of
exon 5 and exon 6 indicating no influence due to the GC content. This peak was lower
at the amplicons that contain 5-methyl cytosine (0.3 μΑ) than those containing
cytosine, indicating that less guanine residues were exposed to electrode surface. This,
7
most probably, was a result of more compact oligonucleotide structure [6] in case of
methylated cytosine as compared to the natural amplicon. The peak current intensity
at inosine-modified amplicons was only 0.2 μA due to the considerable amount of
substituted guanine molecules by inosine, which is electrochemically inactive at 1.0
V. Thus, useful information derived by applying differential pulse voltammetry to
different amplicons, without using any indicator.
Study of the intercalation of proflavine with amplicons-biosensors
Further investigation of the conformation of amplicons was performed by using
proflavine as a probe. The modified electrode (amplicons immobilized at its surface)
was prepared as previously described, washed and subsequently immersed into
proflavine solutions of concentrations ranging from 10-8 to 10-7 M (in 0.2 M sodium
acetate, pH 5.0) for 300 s.
The influence of proflavine in all amplicons used in the present work was
investigated. A decrease of the peak current intensity at 1.0 V was observed in all of
these PCR products by increasing the concentration of proflavine (Fig.4). The
decrease of the peak current intensity was attributed to the stacking of proflavine in
the nucleic acid backbone. Insertion of proflavine caused a major change in the
phosphodiester group orientation [28]. As a result guanine residues were exposed to a
lower degree to the electrode surface than before the intercalation of proflavine [22].
In order to compare the results, the relative per cent value of current intensity
with reference to the amplicons in absence of proflavine was used (Fig.5). It was
obvious, that at lower than 1.5·10-7 M concentrations of proflavine, there were no
distinct differences between amplicons, but there was a significant difference at
higher concentrations. It was also observed, that in all amplicons at proflavine’s
concentrations higher than 5·10-7 M the current intensity remained almost constant,
revealing that no more proflavine could be intercalated.
The decrease of the peak at 1.0 V by increasing the concentration of PF was
stronger in the inosine-containing exon 5 as compared to the other amplicons. This
indicated that in the presence of inosine more proflavine was intercalated and induced
a more pronounced conformational change compared to the natural amplicons. On the
other hand, the amplicons containing 5-methyl cytosine didn’t favor the intercalation
of proflavine to high extent, since the reduction of the oxidation peak current wasn’t
so high. Finally, the decrease of guanine’s peak current was stronger in exon 6 than in
8
exon 5, indicating that exon 6 favoured the intercalation of proflavine compared to
exon 5.
By taking into consideration all the results (Table 1), it was observed that
higher guanosine content and thermal stability corresponded to lower intercalation.
The latter was in accordance with the intercalation model by which double helix
unwound to form the intercalation site [29]. It could be assumed that proflavine
intercalated preferably at amplicons that could easier unwind or had less rigid
structure such as the inosine-containing amplicons [11].
It could be observed that a difference in Tm of exon 5 and exon 5 containing
methyl-cytosine by 3C corresponded to a difference by 2% in current, while a
difference in Tm of exon 6 and exon 5 containing inosine by 2C corresponded to a
difference by 10% in current, which is considerably higher than the former. The
reason could be a stereochemical factor that renders the molecule of PF (Fig. 6a) more
labile than the molecule of Sybr Green I (Fig. 6b), so that almost the same number of
PF molecules to be able to accommodate within both structures of the
oligonucleotides containing either cytosine or 5-methyl-cytosine despite their
different conformation (amplicon containing methyl-cytosine was more rigid than
natural amplicon [6]).
9
Conclusions
The use of pulse voltammetry can differentiate between model DNA systems
containing inosine as a guanosine analogue and depending on the methylation state of
cytosine. Proflavine intercalation correlates with the guanosine content and the
thermal stability of the synthetic DNA systems, being greater in amplicons of lower
guanosine content and lower thermal stability. A weaker dependence of proflavine
intercalation on the amplicon conformation was also observed. Considering the speed,
the simplicity of instrumentation and the low cost required for this procedure, the
above results are promising for the use of electrochemical analysis for the
identification of epigenetic modification in DNA.
Acknowledgments
This paper is part of the 03ED835 research project, implemented within the
framework of the “Reinforcement Programme of Human Research Manpower”
(PENED) and co-financed by National and Community Funds (25% from the Greek
Ministry of Development-General Secretariat of Research and Technology and 75%
from E.U.-European Social Fund).
10
REFERENCES
1.
Robertson KD, Wolffe AP (2000) Nat Rev Gen 1:11-19
2.
Ehrlich M (2002) Oncog. 21:5400-5413
3.
Esteller M (2005) Annu Rev Pharmacol Toxicol 45:629-656
4.
Watson RE, Goodman JI (2002) Toxicol Sci 67:11-16
5.
Mathur P, Xu J, Dedon PC (1997) Biochem (N.Y.) 36:14868-14873
6.
Banyay M, Graslund A (2002) J Mol Biol 324:667-676
7.
Hodges-Garcia Y, Hagerman PJ (1995) J Biol Chem 270:197-201
8.
Kovalchuk O, Tryndyak VP, Montgomery B, Boyko A, Kutanzi K, Zemp F,
Warbritton AR, Latendresse JR, Kovalchuk I, Beland FA, Pogribny IP (2007) Cell
Cycle 6:2010-2018
9.
Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB (1996) Proc Natl
Acad Sci U S A 93:9821-9826
10.
Chou SH, Chin KH, Chen CW (2001) J Biomol NMR 19:33-48
11.
Bailly C, Waring MJ (2001) Meth Enzymol 340:485-502
12.
Gherghi IC, Girousi ST, Voulgaropoulos A, Tzimou-Tsitouridou R (2004)
Anal Chim Acta 505:135-144
13.
Fojta M (2002) Electroanal 14:1449-1463
11
14.
Ozkan-Ariksoysal D, Tezcanli B, Kosova B, Ozsoz M (2008) Anal Chem
80:588-596
15.
Kara P, Cavdar S, Meric B, Erensoy S, Ozsoz M (2007) Bioelectrochem
71:204-210
16.
Eskiocak U, Ozkan-Ariksoysal D, Ozsoz M, Oktem HA (2007) Anal Chem
79:8807-8811
17.
Reed GH, Kent JO, Wittwer CT (2007) Pharmacogen 8:597-608
18.
Deligezer U, Esin Akisik E, Dalay N (2007) Clin Chem Lab Med 45:867-873
19.
Liu Y, Zhu Q, Zhu N (2007) Immun Investig 36:507
20.
Skow A, Mangold KA, Tajuddin M, Huntington A, Fritz B, Thomson RB,
Kaul KL (2005) J Clin Microbiol 43:2876-2880
21.
Girousi ST, Alexiadou DK, Ioannou AK (2008) Microch Acta 160:435-439
22.
Alexiadou DK, Ioannou AK, Girousi ST (2008) Anal Let 41:in press
23.
Aslanoglu M (2006) Anal Sci 22:439-443
24.
Vanícková M, Labuda J, Bucková M, Surugiu I, Mecklenburg M, Danielsson
B (2000) Collect Czech Chem Commun 65:1055-1066
25.
Deubel V, Leng M (1974) Biochimie 56:641-648
26.
Sambrook J, Fritsch E, Maniatis FT (1989) Molecular Cloning. A Laboratory
Manual, 2nd ed Cold Spring Harbor Laboratory Press New York
27.
Guldberg P, Worm J, Grønbæk K (2002) Methods, 27:121-127
12
28.
Tang P, Juang CL, Harbison GS (1990) Science 249:70-72
29.
Rosu F, Pirotte S, De Pauw E, Gabelica V (2006) Int J Mass Spectrom
253:156-171
13
Figure 1: PCR amplification of exon-5 by using 1) natural bases, 2) natural bases and 7%
DMSO, 3) 60% dITP instead of dGTP, 4) 60% 5-methyl-dCTP instead of dCTP and 5) 60% 5methyl-dCTP instead of dCTP and 7% DMSO. L) 100 base pairs DNA ladder.
Figure 2: Melting curves (-dF/dT) of amplicons produced in the presence of Sybr Green by using:
(a) natural bases, (b) 30% 5-methyl-dCTP instead of dCTP and (c) 60% of 5-methyl-dCTP
instead of dCTP.
14
d
c
b
a
Figure 3: Melting curves (-dF/dT) of the following amplicons; (a) exon 5 synthesized in the
presence of 60% 5-methyl-dCTP instead of dCTP, (b) exon 5, (c) exon 6, (d) exon 5 synthesized in
the presence of 60% dITP instead of dGTP.
15
i
i
ii
ii
(a)
(c)
i
i
ii
ii
(b)
(d)
Figure 4: The peak current intensity at 1.0 V of amplicons i) without proflavine and ii) with
proflavine (C = 8·10-7 M). (a) exon 5 synthesized in the presence of 60% 5-methyl-dCTP instead
of dCTP, (b) exon 5, (c) exon 6, (d) exon 5 synthesized in the presence of 60% dITP instead of
dGTP. Voltammetric conditions as mentioned in the experimental part.
16
Relative current (%)
100
80
60
a
b
c
d
40
20
0
0
5
10
-7
Cproflavine (10 M)
Figure 5: The relative variation of the peak current intensity at 1.0 V of amplicons by increasing
proflavine’s concentration. Amplicons of (a) exon 5 synthesized in the presence of 60% 5-methyldCTP instead of dCTP, (b) exon 5, (c) exon 6, (d) exon 5 synthesized in the presence of 60% dITP
instead of dGTP. Voltammetric conditions as mentioned in the experimental part.
Table 1: The peak current intensity of the synthesized amplicons at 1.0 V without proflavine,
their Tm and their relative peak current intensity at 1.0 V when proflavine (C=8·10 -7 M) is
added. Voltammetric conditions as mentioned in the experimental part.
Amplicons
I (μΑ)
Tm (C)
I (%)a
exon 5 synthesized in the
presence of 60% 5-methyldCTP instead of dCTP
0.3
94
46
exon 5
0.4
91
44
exon 6
0.4
85
30
exon 5 synthesized in the
presence of 60% dITP instead
0.2
83
20
of dGTP
a
100% is the peak current intensity at 1.0 V of each amplicon without proflavine
17
(a)
(b)
Figure 6: Structures of (a) proflavine and (b) Sybr Green I.
18