Highly sensitive DNA methylation analysis at CpG resolution by surface-enhanced Raman scattering via ligase chain reaction

Article abstract of DOI:10.1039/c5cc03921e

Sensitive and accurate DNA methylation analysis at CpG resolution was demonstrated using surface-enhanced Raman scattering (SERS) via ligase chain reaction (LCR). The method was sensitive to 10% changes in methylation and the accuracy of methylation estimates in cells and serum DNA validated with sequencing. The LCR/SERS approach may have broad applications as an alternative (epi)genetic detection method.

Full text of DOI:10.1039/c5cc03921e

View Article Online
View Journal
ChemComm
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: Y. Wang, E. J. H.
Wee and M. Trau, Chem. Commun., 2015, DOI: 10.1039/C5CC03921E.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/chemcomm

View Article Online
DOI: 10.1039/C5CC03921E
Page 1 of 4
ChemComm
ChemComm
RSCPublishing
COMMUNICATION
Highly Sensitive DNA Methylation Analysis at CpG
Resolution by Surface-enhanced Raman Scattering via
Ligase Chain Reaction
Cite this: DOI: 10.1039/x0xx00000x
Yuling Wang, †§* Eugene J. H. Wee†§ and Matt Trau†‡*
Received 00th January 2012,
Accepted 00th January 2012
DOI: 10.1039/x0xx00000x
www.rsc.org/
bands.10-12 In contrast, the overlapping fluorescence emission spectra
Sensitive and accurate DNA methylation analysis at CpG
resolution was demonstrated with surface-enhanced
Raman scattering (SERS) via ligase chain reaction (LCR).
The method was sensitive to 10% changes in methylation
and the accuracy of methylation estimates in cells and
serum DNA validated with sequencing. The LCR/SERS
approach may have broad applications as an alternative
(epi)genetic detection method.
severely limit multiplexing. In addition, SERS also has comparable
sensitivity yet improved photostability over flurophores.10-12
While detecting single DNA base mismatches by SERS is achievable
via label-free methods coupled to intensive post-analysis,13,14 the
poor spectral reproducibility and the similarity of SERS signals from
single and double stranded DNA limit their applications in routine
medical diagnostics procedures.13,14 A strategy to facilitate single
DNA base detection via SERS could be a DNA ligase-based
approaches. For instance, Erikson, et. al. reported single point
mutation by SERS via a ligation detection reaction (LDR).15-16 By
labeling primers with silver nanoparticles and dyes, single-point
mutation with the detection limit of 10-20 pM was achieved by
SERS. Although highly specific, LDR is a linear amplification hence
inherently limited in sensitivity and speed. The ligase chain reaction
(LCR) however, may enhance sensitivity and speed over LDR due to
its exponential amplification characteristics.17 LCR has been applied
to single DNA base change detection by fluorescence and
electrochemical methods,18-21 however, to the best of our knowledge,
no one has reported multiplexed detection of single base changes via
LCR coupled to a SERS readout nor has its application to DNA
methylation at single CpG resolution been reported.
Single DNA base change such as aberrant DNA methylation
epigenetic changes, where a cytosine (C) to thymine (T) base change
occurs after bisulfite treatment of DNA,1 single nucleotide
polymorphism (SNP) and point mutation are common events in
cancer.2-3 The screening of such single DNA base change is clinically
useful for disease diagnosis and the selection of the appropriate
therapies eg. personalized medicine.1-5 However, routine single DNA
base change analysis relying on PCR alone is challenging and has
many limitations.6-9 Additionally, most single-base change detection
methods still rely on fluorescence and mass spectroscopy, which are
labor-intensive or instrumentally-expensive. Therefore, it is highly
desirable to develop simple, cost effective and multiplexed assays
with high sensitivity and specificity for routine diagnosis.
Herein, we demonstrate for the first time, a proof-of-concept
multiplexed single DNA base change detection platform with SERS
nanotags via LCR. We applied the LCR/SERS method to sensitive
DNA methylation analysis at a single CpG site on synthetic DNA, a
panel of breast cancer cell lines and a serum-derived DNA sample. In
addition, methylation estimates also validated well with Next
Generation Sequencing (NGS) thus demonstrating the high accuracy
of our method.
Surface enhanced Raman Scattering (SERS) is
a physical
phenomenon that happens on metal surfaces and has been
demonstrated to be an attractive alternative to fluorescence-based
approaches.10-12 In particularly, SERS has a unique advantage over
fluorescence in parallel detection of multiple analytes (multiplexing)
due to the narrow and distinct bandwidths of vibrational Raman
†Centre for Personalized Nanomedicine, Australian Institute for
Bioengineering and Nanotechnology (AIBN), Corner College and Cooper
Roads (Bldg 75), ‡School of Chemistry and Molecular Biosciences, The
University of Queensland, Brisbane, QLD 4072, Australia.
The principle of this platform for multiplexed detection of a single
DNA base change and its application to DNA methylation analysis is
illustrated in Scheme 1. In this approach (Scheme 1A), genomic
DNA was first treated with bisulfite, in which methylated C
remained inert while unmethylated C were converted to uracils and
subsequently T after PCR.22-23 Following PCR amplification of a
sequence containing a CpG of interest, LCR was then applied to
identify and exponentially amplify the C/T base change (as
illustrated in Scheme S1, ESI) which, in turn, represented the
E-mail: y.wang27@uq.edu.au; m.trau@uq.edu.au
†Electronic Supplementary Information (ESI) available: The experimental
details, image of DNA agarose gel electrophorese of LCR product, optical
image of gold assay, SEM image of the gold assay with and without target.
See http://dx.doi.org/ 10.1039/b000000x/
§ Authors contributed equally.
This journal is © The Royal Society of Chemistry 2012
Chem. Commun., 2013, 00, 1-3 | 1

View Article Online
DOI: 10.1039/C5CC03921E
ChemComm
Page 2 of 4
COMMUNICATION
Journal Name
methylation state of the CpG of interest. The use of a PCR/LCR was chosen for use as SERS nanotags in this study. The size and
system has the potential benefit of improving the specificity, optical properties of our SERS nanotags were characterized using
sensitivity and speed of a methylation assay.24 The LCR products TEM and extinction spectroscopy. The TEM image in Fig. 1A
also had on one end, a universal barcode sequence for capture onto reveals AuNPs with a typical diameter of about 60 nm. The localized
the SERS array, while the other end encodes for a methylation state- surface plasmon resonance (LSPR) peak of the bare AuNPs occurs at
specific barcode to enable methylation-specific SERS nanotag ca. 534 nm (Fig. 1B). Upon formation of AuNPs-DNA-Ra
labeling (Scheme 1B). Detailed experimental procedures and DNA conjugates, the LSPR peak of AuNPs exhibits a red shift to 540 and
oligonucleotides sequences used in this study (Table S1) are 542 nm, respectively, depending on the particular Raman reporter.31
provided in the accompanying ESI.
The red-shift of the LSPR of AuNPs clearly indicated the binding of
oligonucleotides and Raman reporters on AuNPs surface due to the
change of the surrounding refractive index. Surface coverage of
DNA on AuNPs is crucial for the stability of the nanoparticles,
which was estimated using a previously described method.32-33 On
average, there were approximately 2235 oligonucleotides strands on
each 60 nm-AuNPs.
The specificity of the optimized LCR reactions was first evaluated
with gel electrophoresis (Figure S1, ESI). A clear band indicated a
positive LCR reaction while no bands were observed for the no-
template (NoT) and unmatched (UM) primer controls, thus indicating
the high specificity of LCR. To enable a SERS-based evaluation
(Scheme 1B), LCR products were then hybridized to the universal
capture DNA probes pre-immobilized on gold surface through Au-S The molecular structures of the Raman reporters and the
bond. To avoid non-specific physical adsorption of LCR products, corresponding SERS fingerprint are shown in Figure 1C. The
the gold surface was blocked with 6-mercaptohexanol (MCH) prior characteristic Raman peak of 4-mercapto-3-nitro benzoic acid
to use. Finally, to identify the captured LCR products and eventually (MNBA) at 1334 cm-1 was assigned to the symmetric nitro stretching
methylation levels, methylation state-specific SERS nanotags (made vibration,34-35 while the peaks at ca. 1080 and ca. 1580 cm-1 of both
by modifying gold nanoparticles with methylation-state-specific MNBA and 4-mercaptobenzoic acid (MBA) arose from phenyl ring
detection DNA probes and Raman reporters) were further hybridized modes.34-35 The SERS spectrum of the binary mixture from the two
to the captured LCR.
Raman reporters (MNBA and MBA) clearly shows the distinct
fingerprints from the corresponding SERS nanotags, which can be
used for identification of the individual targets.
To realize a quantitative SERS/LCR methylation assay, a 2-plex
LCR reaction (simultaneous C and T-specific LCR reactions for
methylated and unmethylated respectively) was performed, captured
on the gold array, SERS nanotagged and interrogated with a laser.
SERS signals generated for the corresponding LCR products were
used to estimate DNA methylation levels at a specific CpG by
measuring the ratio between the different Raman intensities (I)
associated with the C-reaction (methylated) and T-reaction
(unmethylated), i.e. % Methylation = IC/ (I T+ IC).
A
B
A
C
B
Fig. 1. TEM image of AuNPs (A), extinction spectra of bare AuNPs and
SERS nanotags (B), and molecular structure of Raman reporters with
corresponding SERS spectra (C).
We first evaluated the feasibility of the SERS/LCR assay by
applying individual C or T LCR reactions products derived from
various concentrations of synthetic template inputs onto gold
surface. As evident from the optical image of the gold surface
assay, the binding of SERS nanotags can already be observed by the
naked eye on the gold surface (Fig.S2A, ESI) with good specificity.
A clean and bright surface was observed with no-template control
(NoT), un-matched target (UM) and with false SERS nanotags.
With increasing concentration of targets, successively more SERS
nanotags selectively bound onto the gold surface, which lead to the
formation of “SERS dots” of high nanotags densities, thus generating
high Raman intensities. SEM images further indicated the binding of
SERS nanotags on the gold surface with high specificity as
Scheme 1. Schematic illustration for simultaneous C and T single DNA base
change detection by LCR (A), and DNA methylation analysis via simultaneous
SERS nanotags detection on a gold surface array (B).
Studies have shown that gold nanoparticles (AuNPs) with distinct
optical properties and good stability can enhance Raman signals of
reporter molecules chemisorbed on their surface.25-28 Moreover,
AuNPs with the core size of ~60-80 nm are most efficient for SERS
at red and near-infrared (785 nm) excitation.29-30 Therefore, to enable
a robust and sensitive SERS assay, AuNPs with the size of ~60 nm
2 | Chem. Commun., 2013, 1-3
This journal is © The Royal Society of Chemistry 2012

View Article Online
DOI: 10.1039/C5CC03921E
Page 3 of 4
Journal Name
ChemComm
COMMUNICATION
demonstrated in Figure S2 (ESI). No nanoparticles were observed on targets were mixed in ratios 1:0; 3:1; 1:1. 1:3; 1:9 and 0:1. These
the gold surface with UM control (ESI, Fig. S2B). In contrast, an titrations represented samples at 100%, 75%, 50%, 25%, 10% and
abundance of AuNPs appeared on the gold surface at even very low 0% methylation. With an increasing ratios of C to T, i.e,
(5 pM) initial input concentration (ESI, Fig. S2C). These results not %methylation, the SERS intensity of MNBA (SERS nanotags to C-
only demonstrated the potential for good assay sensitivity, it also reaction) at 1334 cm-1 increased. Since DNA methylation is based on
suggested a possible naked eye evaluation assay for LCR/SERS- C to T base change after bisulfite treatment and owing to the
based applications. Fig. 2A shows the typical SERS spectra obtained specificity of LCR, the SERS intensity (I) ratio of IC-base/IT-base+IC-base
from different concentrations of input synthetic targets and negative may be used to estimate the degree DNA methylation at a single
controls. The distinct SERS peak located at 1334 cm-1 was observed CpG. As indicated in Figure 3B, the signal from the C-reaction
corresponding to MNBA for the C-reaction thus indicating the increased with increasing methylation and as low as 10% differences
presence of the LCR product. Figure 2B clearly shows the positive in DNA methylation could be detected (RSD of 3.99 % over 5
relationship between SERS intensity and target concentration. In independent tests). Overall, these results indicated that LCR/SERS
contrast, control experiments performed with no template input assay was a feasible multiplex detection method for single DNA base
(NoT), unmatched target (UM) and false SERS nanotags (non- changes and one potential application was DNA methylation analysis
specific hybridization control), no particles and therefore only at single CpG resolution.
background Raman signals (Fig. S3, ESI and Figure 2A) were
A
B
observed. The detection limit was determined to be 0.5 pM based on
the signal-to-noise ratio being three-times higher than the
background (No template control, NoT) with relative standard
deviation (RSD) of 6.4% over 5 independent tests (20 SERS spectra
were typically collected for each of the 5 independent tests; Fig. S3,
ESI), demonstrating good assay reproducibility. This level of
sensitivity was 10-fold better than previous LDR/SERS approaches
and comparable to fluorescence and electrochemical-based LCR
methods.15-16, 18-21
.
Fig.3. Typical SERS spectra for duplex point mutation detection (A) and
SERS intensity ratio response to methylation level (B). Error bars represent
±SD, n = 5.
Likewise, with the T-reaction, which uses MBA-modified SERS
nanotags instead, had similar assay performance to the C-reaction
(Fig 2C and 2D). As expected, the characteristic Raman peaks at
1080 and 1580 cm-1 arose from phenyl ring mode of MBA. As low
as 0.5 pM T-base target could be detected with very high specificity
and reproducibility (RSD=1.8% with n=5; Figure 2D), Together,
these data underscored the high specificity and sensitivity of the
LCR/SERS approach and suggests its viability as an alternative to
fluorescence and electrochemical-based methods.
Next, to demonstrate a complex biological application, we applied
the LCR/SERS assay to detect the methylation state at CpG12 of the
miR200b P2 promoter, a potential biomarker of breast cancer
metastasis.36-38 LCR/SERS assay was first employed to measure the
level of the DNA methylation in a panel of breast cancer cell lines.
As indicated in Figure 4A, typical SERS spectra obtained from the
no-template control were of only low background signals, while the
signals from the three cell lines were of the distinct SERS nanotags
signals profiles depending on their methylation states. The intensity
ratio between the peak at 1334 cm-1 (C-reaction; methylated) and
1080 cm-1 (T-reaction; unmethylated) was used to estimate the
degree of DNA methylation in the three cell lines (Figure 4B, blue
bars). The estimated DNA methylation level were 76% in MDA-MB
231 cell lines, 0.4% in MDA-MB 468 cell lines and 47% with HCC
1937 cell lines. These methylation estimates were then validated with
Next Generation Sequencing (Figure 4B; black bars) and were also
consistent with the literature19, 38 thus indicating that LCR/SERS had
potential as an accurate DNA methylation analysis platform with
single CpG resolution. To further demonstrate clinical viability, a
clinically normal serum DNA sample was also successfully tested
(Figure S4). DNA agarose gel electrophoresis indicated the presence
of the target while SERS measurments estimated methylation to be
approximately 58.74%.
B
A
C
D
SERS
NGS
B
A
80
60
40
20
0
Fig.2. Concentration-dependent raw SERS spectra (A, C) with non-cognate
probe controls and (B, D) the corresponding concentration-dependent
response derived from SERS peaks in (A) and (C) for C and T-reactions
respectively. Error bars represent ±SD, n = 5.
Due to the unique multiplexing capability of SERS, we next turned
our attention to the simultaneous detection of the C and T-reactions
for quantitative methylation analysis. As indicated in Scheme 1B,
multiplexed LCR products may be immobilized onto a gold surface
followed by simultaneous SERS nanotags interrogation. As a-proof
of-concept, Figure 3A shows the typical SERS spectra (normalized
to the peak at 1080 cm-1) for a 2-plex LCR detection in which C:T
MDAMB231
MDAMB468
HCC
Fig.4. Typical SERS spectra for a panel of cell lines detection (A) and
comparison of estimated methylation in three breast cancer cell lines with
SERS-LCR assay (B). Error bars represent ±SD, n = 5.
This journal is © The Royal Society of Chemistry 2012
Chem. Commun., 2013, 00, 1-3 | 3

View Article Online
DOI: 10.1039/C5CC03921E
ChemComm
Page 4 of 4
COMMUNICATION
Journal Name
In summary, a simple platform for rapid multiplexed detection of
single base changes was developed with SERS via LCR, which could
significantly increase detection sensitivity (0.5 pM) while
maintaining high specificity. More importantly, the developed assay
could be used for DNA methylation analysis by simultaneously
detecting the relative Raman intensities between the C and T
(methylated and unmethylated respectively) DNA base changes
amplified by LCR. As low as 10% differences in methylation could
be detected, demonstrating the potential of SERS for sensitive
multiplexed (epi)genetic monitoring. The LCR/SERS assay was also
successfully applied to breast cancer cell lines and a serum-derived
DNA sample to demonstrate its feasibility on complex biological
samples. Lastly, assay results validated with Next Generation
Sequencing, underscoring its potential for accurate genetic biomarker
detection.
15 Y. S. Huh, A. J. Lowe, A. D. Strickland, C. A. Batt, D. Erickson, J. Am.
16 Y. S. Huh, A. J. Lowe, A. D. Strickland, C. A. Batt, D. Erickson, Anal.
Chem. Soc. 2009, 131, 2208-2213
Chem, 2010, 82, 5810-5814.
17 F. Barany, PCR Methods Appl., 1991, 1, 5.
18 E. J. H. Wee, S. Rauf, K. M. Koo, M. J. A. Shiddiky, M. Trau, Lab on
chip, 2013, 13, 4385.
19 K. M. Koo, E. J. H. Wee, S. Rauf. M. J. A. Shiddiky, Biosens Bioelectron.
2014, 56, 278.
20 E. J. H. Wee, M. J. A. Shiddiky, M. Trau, Chem.Commun. 2012, 48,
12014
21 E. J. H. Wee, S. Rauf, M. J. Shiddiky, A. Dobrovic, M. Trau, Clin.Chem.
2015, 61, 163-171
22 M. Frommer, L. E. Mcdonald, D. S. Millar, Proc. Natl. Acad. Sci. USA,
1992, 89. 1827
23 S. J. Clark, J. Harrison, C. L. Paul, M. Frommer, Nucleic Acids Res, 1994,
22, 2990
24 E.J.H. Wee, Ph.D Thesis, The University of Queensland, 2012.
25 C. Lei, Y. Huang, Z. Nie, J.Hu, L. Li, G. Lu, Y. Han, S. Zuo, Angew.
Chem. Int. Ed. 2014, 53, 8358.
26 K.Saha, S.S. Agasti, C. Kim, X. Li, V. M. Rotello Chem. Rev., 2012, 112,
2739
27 L. Sun, C.Yu, J. Irudayaraj, Anal. Chem. 2007, 79, 3981.
28 L.Sun, C.Yu, J.Irudayaraj, Anal. Chem. 2008, 80, 3342.
29 J.T. Krug, G. D. Wang, S. R. Emory, S. M. Nie, J. Am. Chem. Soc. 1999,
121, 9208.
We acknowledge financial support from ARC DECRA (DE
140101056) to Y.W and the National Breast Cancer Foundation of
Australia to M.T. (CG-12-07). These grants have significantly
contributed to the environment to stimulate the research described
here. We thank adveer re al for taking TEM and SEM images.
We also thank Darren Korbie and Erica Lin for providing the serum-
derived DNA.
30 X. M. Qian, X. H. Peng, D. O. Ansari, Q.Yin-Goen, G.Z.Chen, D.M.Shin,
L. Yang, A. N. Yong, M. D. Wang, S.M. Nie, Nat. Biotech. 2008, 26, 83.
31 M. Salehi, L. Schneider, P. Ströbel, A. Marx, J. Packeisen, S. Schlücker,
Nanoscale. 2014, 6, 2361.
Notes and references
32 L. M. Demers, C. A. Mirkin, R. C. Mucic, R.A.Reynolds, R.L.Letsinger,
R. Elghanian, G. Viswanadham, Anal. Chem. 2000, 72, 5535.
33 Y.W.C.Cao, R.C.Jin, C.A.Mirkin, Science, 2002, 297, 1536.
34 Y. Wang, M. Salehi, M.Schütz, S. Schlücker. Chem Commun. 2014, 50,
2711.
35 Y. Wang, S. Rauf, Y.S. Grewal, L.J. Spadafora, M.J. Shiddiky, G.A.
Cangelosi, S. Schlücker, M. Trau. Anal Chem. 2014, 86, 9930.
36 M. Korpal, E.S. Lee G.H.Hu, Y. B. Kang, J. Biol. Chem. 2008, 283,
14910.
1
2
3
4
5
6
7
8
9
A.P. Feinberg, B. Tycko, Nat. Rev. Cancer. 2004, 4, 143-153.
E. D. Pleasance and R. K. Cheetham, et al., Nature, 2010, 463, 191.
C.Whibley and P. D. Pharoah, et al., Nat. Rev. Cancer, 2009, 9, 95.
A. A. Komar, Science, 2007, 315, 466.
C. Greenman and P. Stephens, et al., Nature, 2007, 446, 153.
A. C. Syvanen , Nat. Rev. Genet. 2001 , 2 , 930
P. Y. Kwk , Annu. Rev. Genomics Hum. Genet. 2001 , 2, 235
J. Ragoussis , Annu. Rev. Genomics Hum. Genet. 2009 , 10, 117
Y. Song, Y. Zhang, T. H. Wang, Small, 2013, 9, 1096.
37 M.S. Nicoloso, R.Spizzo, M. Shimizu, S. Rossi, G.A. Calin, Nat. Rev.
Cancer 2009, 9, 293.
38 E.J.H. Wee, K. Peters, S.S. Nair, T. Hulf, S. Stein, S.Wagner, P. Bailey,
S.Y. Lee, W. J. Qu, B. Brewster, J.D. French, A. Dobrovic, G.D. Francis,
S. J. Clark, M.A. Brown, Oncogene 2012, 31, 4182.
10 S. Schlücker, ChemPhysChem. 2009, 10, 1344
11 B. H. Jun, G. Kim, M. S. Noh, H. Kang, Y.K. Kim, M. H. Cho, D. H.
Jeong, Y. S. Lee, Nanomedicine, 2012, 6, 1463.
12 Y.Wang, S. Schlücker, Analyst, 2013, 138, 2224
13 E. Papadopoulou, S. E. Bell, Angew. Chem.Int. Ed, 2011, 50, 9058
14 L. Guerrini, Z. Krpetic, D. Lierop, R. A. Alvarez-Puebla, D. Graham,
Angew.Chem.Int. Ed. 2014, 53, DOI: 10.1002/anie.201408558
4 | Chem. Commun., 2013, 1-3
This journal is © The Royal Society of Chemistry 2012