Clinica Chimica Acta 477 (2018) 166–172
Contents lists available at ScienceDirect
Clinica Chimica Acta
journal homepage: www.elsevier.com/locate/cca
A facile gold nanoparticle–based ELISA system for detection of osteopontin
in saliva: Towards oral cancer diagnostics
T
Debolina Chakrabortya, Thangaraj Soundara Vivekab, Krishnamurthy Arvindb,
Vidyarani Shyamsundarc, Murhekar Kanchanb, Sruthi Ann Alexa, N. Chandrasekarana,
Ramshankar Vijayalakshmib,⁎, Amitava Mukherjeea,⁎
a
b
c
Centre for Nanobiotechnology, VIT University, Vellore, India
Department of Preventive Oncology (Research and Molecular Diagnostics), Cancer Institute (WIA), Chennai, India
Centre for Oral Cancer Prevention Awareness and Research, Sree Balaji Dental College and hospital, Bharath University, Chennai, India
A R T I C L E I N F O
A B S T R A C T
Keywords:
Oral cancer
Osteopontin
Gold nanorod
Gold nanosphere
Nano ELISA
Sensitivity
In the current study, we emphasize that osteopontin is overexpressed in oral squamous cell carcinoma.
Overexpression of osteopontin levels was confirmed by mRNA quantification studies and immunohistochemistry
analysis. Based on this, a gold nanoparticle–based ELISA system was developed for non-invasive osteopontin
detection. The incorporation of AuNRs (Gold nanorods) or AuNSs (Gold nanospheres) in the conventional ELISA
improved the sensitivity of analyses. A considerably lowered detection limit in case of AuNR (detection limit:
0.02 ng mL− 1) and AuNS (detection limit: 0.03 ng mL− 1) modified assay was obtained as compared to commercially available OPN ELISA kit (detection limit: 0.14 ng mL− 1). The modified ELISA had a wide linear detection range (0.31–20 ng mL− 1), good reproducibility, and specificity against the tested interferents in the
saliva. Finally, the nanoELISA was validated with osteopontin spiked in artificial and normal saliva samples and
observed to show good recovery (95.4–97.85%), which indicates the application potential of the developed kit
for real sample analysis.
1. Introduction
Oral cavity cancer is the eight most common cancers worldwide
with a high prevalence among men. There are approximately 300,000
new cases annually worldwide [1,2]. In south central Asia, oral cancer
ranks among the three most common types of cancer. The age standardized incidence rate of oral cancer is 12.6/100,000 in India. Oral
cancer survival rate is <60% and there has been no improvement in
survival for the past 5 decades [3], due to late diagnosis frequently in
up to 50% of the patients with lymph node metastasis during presentation [4]. Initial staging of oral cancer based on TNM (T stands for
“Tumour Size” – size of the primary tumour, measured in cm or mm; N
stands for “Nodes” – indicating the extent of spread of the cancer to the
regional lymph nodes and M stands for “metastasis” – indicating if the
cancer has spread to the other organs of the body from its primary location) classification, which is currently in practice, appears insufficient
to accurately predict the prognosis and is not adequately helpful to
tailor treatment. Biomarkers are therefore an important need to predict
treatment response and to customise therapy options [5]. Since most of
the Oral Squamous cell carcinoma (OSCC) develops from visible lesions
⁎
in the oral cavity, the most preferred biomarker detection medium includes biological fluids like blood and saliva. Saliva holds a promising
future in search of newer clinical markers as it is easily accessible, noninvasive, accurate and cost effective to investigate the malignant molecular pathology by secreted form of biomarkers from tumour.
Osteopontin (OPN) is an extracellular matrix (ECM) associated cytokine like sialic acid rich phosphoglycoprotein [6–8]. It is member of
the SIBLING (small integrin binding ligand and N-linked glycoprotein)
family. Several recent reports reveal that osteopontin is expressed in
tumour educated stromal cells and leads to cancer progression [9]. It
plays an important role in tumour invasion, tumour growth, angiogenesis and metastasis by up regulating several signalling pathways
that lead to overexpression of target proteins like Matrix Mettaloproteins 2/9, urokinase plasminogen activator (uPA) and vascular endothelial growth factor (VEGF) [10]. However, OPN is important for
normal physiological processes like wound healing, bone resorption,
tissue remodeling, immune responses, and vascularisation as well.
Recent clinical studies show that OPN is overexpressed in tumour
tissues and serum samples from patients of various cancers [11]. Elevated levels of OPN in plasma has been associated with unfavourable
Corresponding authors.
E-mail addresses: r.vijayalakshmi@cancerinstitutewia.org (R. Vijayalakshmi), amitav@vit.ac.in (A. Mukherjee).
http://dx.doi.org/10.1016/j.cca.2017.09.009
Received 10 July 2017; Received in revised form 28 August 2017; Accepted 11 September 2017
Available online 14 September 2017
0009-8981/ © 2017 Elsevier B.V. All rights reserved.
Clinica Chimica Acta 477 (2018) 166–172
D. Chakraborty et al.
Green Master (Rox) (Roche) was used according to the manufacturer's
instructions on a 7500 Real Time PCR System (Applied Biosystems) for
qPCR based quantitations.
outcome in several cancers [12]. OPN has been shown to have a definite
prognostic significance in head and neck cancers patients treated with
radiotherapy previously [13]. OPN is associated with tumour hypoxia
and a malignant phenotype. Despite a large body of evidence to show
OPN detection, there is still no validated and certified OPN ELISA based
test that can be used for cross study comparisons. OPN values have been
different significantly using different ELISA systems applied [14].
Nanoparticle-based sensing system has been used very frequently
for cancer diagnosis [15] for detecting circulating tumour cells (CTC),
circulating nucleic acids (CNA), circulating proteins [16], etc. Immunosensors made of L-cysteine-capped lanthanum hydroxide [17],
nanostructutred zirconium oxide [18], and nanostructured hafnium
oxide [19] have shown promising results in the detection of cyfra-21-1,
an oral cancer biomarker. On the other hand, unique optical properties,
improved biocompatibility, and modifiable surface chemistry of gold
nanoparticles (AuNPs) have contributed remarkably in the development of nano-biosensors [20–23]. These gold NP-based biosensors utilize diverse principles such as SERS [24], interaction-based alteration in
dynamic light scattering [25], colorimetric response [26], and many
more for detecting cancer. Though these methods tend to enhance the
sensitivity of the assay, a sensor that is more facile in nature as well as
minimizes the instrumentation cost would be most preferable. Thus,
conventional ELISA still remains one of the best analytical assays with
high-end accuracy and least complexity involvement. The incorporation
of NPs in conventional ELISA protocol can be useful in enhancing the
sensitivity and shortening the incubation time of the assay [27].
In the current study, we developed a non-invasive AuNP-based
ELISA for OPN detection in saliva samples with amplified response
when compared with conventional ELISA without increasing the complexity of the pre-defined protocol. To the best of our knowledge, this is
the first work that establishes OPN as a biomarker in saliva and summarises the comparative sensing capabilities of two different types of
AuNP bioconjugates i.e. with gold nanospheres (AuNSs) and gold nanorods (AuNRs), for the development of nanoELISA for OPN detection.
The work demonstrates that the use of AuNPs can considerably improve
the limit of detection (LOD) when compared to the commercially
available kit, paving the way for their probable application for oral
cancer prognosis in future.
2.1.1. Patient materials
Formalin-fixed paraffin sections (n = 146) were obtained from oral
tongue cancer patients admitted to the tertiary care centre between
1995 and 2000 with their clinical information. Surgical margin tissues
(n = 6) from prospective patients who underwent wide excision glossectomy and histological normal tissues (n = 2) were obtained to
compare the OPN protein expression. Additionally, RNA of tumour
tissues from prospective tongue cancer patients (n = 68) were collected
after taking an informed consent from the patients as per the guidelines
of Institutional Ethical clearance. Surgical margins were obtained from
these patients (n = 17) and verified that they were histologically nonmalignant. All research involving human participants had been approved by the author's Institutional Review Board (IRB), and all the
clinical investigations had been conducted according to the principles
expressed in the declaration of Helsinki. A written informed consent
was obtained from all the participants, and the contents of the informed
consent was approved by the IRB (Cancer Institute WIA; Protocol 1
HNCOG (Cancer Institute, Women's India Association; Protocol 1 Head
and Neck Co-operative Oncology Network). The finger prints were
obtained for patients, who were illiterate after explaining the protocol,
and a written informed consent was additionally taken from patient's
relative presenting as witness.
Unstimulated saliva was collected from a healthy volunteer after
obtaining an informed consent from Institutional Ethical Committee
Clearance.
2.2. OPN as a biomarker: methodology
2.2.1. mRNA extraction
Briefly, tissues were ground with a mortar and pestle with liquid
nitrogen, and TRIzol was added to powdered tissue and mixed well.
200 μL of chloroform was added, homogenized, and centrifuged. The
aqueous phase was transferred to a new tube, avoiding contact with the
interface, and 500 μL of 100% isopropyl alcohol was added to it, and
this was further processed using RNeasy® Plus Mini kit (Qiagen, Hilden,
Germany) according to the manufacturer's instructions. The RNA samples were then stored at − 80 °C.
2. Materials and methods
2.1. Materials
2.2.2. Real-time PCR analysis
Real-time PCR was performed to measure the mRNA expression of
OPN in primary oral tongue squamous cell carcinoma tissues (n = 68),
adjacent uninvolved non-cancerous tissue called apparently normals
(n = 6), and absolutely normals (n = 11) as described. The primer
sequences used for the study are shown in Table S1. The quantitative
real-time RT-PCR was performed using FastStart Universal SYBR Green
Master (Rox) (Roche) according to the manufacturer's instructions on a
7500 Real-Time PCR system (Applied Biosystems). Universal thermal
cycling conditions used were as follows: 10 min at 95 °C, 40 cycles of
denaturation at 95 °C for 15 s, and annealing and extension at 60 °C for
1 min. Data was collected at every temperature phase during each
cycle. The comparative threshold cycle (Ct) method was used to calculate the fold change. β-Actin gene was used as a reference control to
normalize the expression values.
Triplicates were performed for each gene, and the average expression value was computed for the subsequent analysis. The relative expression level of the genes was calculated using the (2− ddct) method.
Cetyltrimethylammonium bromide (CTAB), sodium borohydride
(NaBH4), N-hydroxysuccinimide (NHS), N-Ethyl-N′-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC), sodium phosphate monobasic (NaH2PO4), tween-20, and 11- mercaptoundecanoic acid (MUA)
were procured from Sigma-Aldrich (India). Trisodium citrate, hydrogen
tetrachloroaurate hydrate (HAuCl4·2H2O), and potassium chloride were
purchased from SRL Pvt. Ltd. (India). Silver nitrate was purchased from
merck. Sulphuric acid and ascorbic acid were procured from SD Fine
Chemicals Ltd. (India). Sodium bicarbonate (NaHCO3), potassium
thiocyanate (KSCN), sodium chloride (NaCl), glucose (C6H12O6), and
blocking buffer were purchased from Himedia Laboratories Pvt. Ltd.
(India). Human Osteopontin (OPN) Quantikine ELISA DOST00 kit was
bought from R & D Systems, and osteopontin antibody (with HRP) was
from Biorbyt Ltd. Glycine was procured from Himedia Laboratories Pvt.
Ltd. (India) and α-amylase from human saliva type IX-A was purchased
from Sigma-Aldrich (India). All the glassware was thoroughly cleaned
with aqua regia (HCl:HNO3 = 3:1), followed by rinsing and was finally
dried in a hot-air oven.
Trizol (Invitrogen, Life Technologies, CA, USA) and RNeasy® Plus
Mini kit (Qiagen, Hilden, Germany) was used for RNA extraction from
tongue tissue samples. cDNA conversion using the High Capacity cDNA
Reverse Transcription Kit (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions. FastStart Universal SYBR
2.2.3. Immunohistochemistry (IHC)
The IHC detection of OPN expression was performed on five-micron
sections of FFPE tissues. The sections were deparaffinised in xylene and
rehydrated in absolute ethanol. Endogenous peroxidase activity was
blocked by incubation in 0.3% hydrogen peroxide in distilled water for
167
Clinica Chimica Acta 477 (2018) 166–172
D. Chakraborty et al.
colorimetric TMB assay was done to estimate the concentration of HRP
in the AuNP bioconjugates. According to the manufacturer's information, the anti-OPN tagged with HRP contained anti-OPN and HRP in the
ratio of 3:10. Based on the estimated HRP concentration, the concentrations of anti-OPN bound to AuNSs and AuNRs were estimated to
be 0.004 and 0.0094 ng L− 1, respectively.
20 min and then washed with distilled water. Antigen retrieval was
done with 0.05 M Tris buffer (pH – 9) in a pressure cooker for 15 min.
Sections were pre-incubated with 2% BSA for 30 min and then incubated with primary antibody against OPN (AKm2A1)- sc-21,742
(Santa Cruz, CA, USA) in 1:100 dilution, overnight at 4 °C. OPN expression was observed using the PolyExcel HRP/DAB IHC Detection
System (PathnSitu Biotechnologies, CA, USA). Sections were counterstained with hematoxylin, dehydrated, and mounted in DPX. Positive
controls and negative controls were included appropriately for OPN,
wherein primary antibody was replaced with 2% BSA in negative
control. Immunostaining of the sections were reviewed along with the
corresponding haematoxylin- and eosin-stained sections.
2.5. NanoELISA system using HRP-tagged anti-OPN/AuNP
For evaluation of the sensitivity of the prepared AuNP immunosensor, the following procedure was adopted. 100 μL of assay
diluent was added to the desired number of wells precoated with OPN
capture Ab. Human OPN standards of concentrations (20, 10, 5, 2.5,
1.25, 0.625, and 0.313 ng mL− 1) were serially diluted, and calibrator
diluent was taken as such for the control. 50 μL of each sample (human
OPN standards and control) was added to the wells, and the plate was
kept for 2 h at room temperature. The wells were carefully washed four
times with wash buffer, following which, 200 μL of the HRP-tagged
anti-OPN/AuNP bioconjugates (AuNSs or AuNRs) was added (for conventional ELISA, 200 μL of HRP-labelled polyclonal antibody specific
for human OPN from the kit was added instead). The AuNP bioconjugate was used after blocking, and the volume ratio of AuNPs:blocking
buffer was optimized. The plate was kept for incubation at room temperature (incubation time was optimized based on the response to
OPN), and the washing step was repeated. 200 μL of the substrate solution was added to all the wells and again incubated for 30 min at
room temperature. 50 μL of stop solution was added to stop further
development of color. For thorough mixing of the contents and uniform
color development, the plate was gently tapped. The microplate reader
was set to 450 nm, and the optical density was determined for each well
within 30 min. HRP-tagged anti-OPN/AuNR conjugate showed better
stability than HRP-tagged anti-OPN/AuNS conjugate when kept for a
duration of 14 days (Result not shown), and thus, nanoELISA for saliva
sample analysis was conducted using HRP-tagged anti-OPN/AuNRs.
2.3. Synthesis of spherical- and rod-shaped gold nanoparticles
Spherical citrate-capped AuNSs were prepared using Turkevich
method [28]. Briefly, 25 mL aqueous solution of HAuCl4 of 1 mM
concentration was heated to 100 °C on a magnetic heater with vigorous
stirring. To this, 1.25 mL of 38.8 mM trisodium citrate was added
dropwise, and stirring was continued for another 15 min until the appearance of a wine-red color. The solution thus obtained was allowed to
cool at room temperature (27 °C). Following this, the solution was
centrifuged at 6000 rpm for 15 min, and the supernatant was carefully
collected for further use.
Rod-shaped CTAB-capped AuNRs were prepared using modified Elsayed method [29]. It involves the preparation of two solutions namely,
seed solution and growth solution. For the preparation of seed solution,
0.3 mL of ice-cold NaBH4 (0.01 mM) was added to a 5-mL mixture of
HAuCl4 (0.5 mM) and CTAB (0.2 M) in the volume ratio of 1:1. This
then was kept for a 3-h incubation at room temperature (27 °C). For the
preparation of growth solution, 6 mL of AgNO3 was added to a 200-mL
solution, which contained (0.5 mM) HAuCl4 and (0.1 M) CTAB. Further
to this solution, 1 mL of H2SO 4 (0.5 M) and 1.4 mL of ascorbic acid
(0.0788 M) were added and mixed. Finally, 0.24 mL of the seed solution
was added to the above mixture and left as such for 12 h. Following
this, the solution was centrifuged twice at 9000 rpm for 30 min.
2.6. NanoELISA for spiked artificial and normal saliva
For the preparation of artificial saliva (AS) (pH–6.2), NaCl
(1.5 mg mL− 1), NaHCO3 (1.5 mg mL− 1), NaH2PO4 (0.5 mg mL− 1),
KSCN (0.5 mg mL− 1) and lactic acid (0.9 mg mL− 1) solutions were
prepared in deionised millipore water [32]. The collection of unsimulated whole saliva (WS) was done from a healthy volunteer. 5 mL
of deionised water was used to rinse the mouth and was expectorated
into a sterilized tube. Following this, the saliva sample was centrifuged
at 2500 rpm for 30 min at room temperature (27 °C). The supernatant
was carefully collected and was diluted ten times with Millipore water
for further use. The artificial saliva (AS 1–3) and processed whole saliva
sample (WS 1–3) were spiked with different OPN concentrations (5, 10,
and 20 ng mL− 1) using the reconstituted human OPN standard
(200 ng mL− 1) provided in the kit. As-prepared artificial saliva and
processed whole saliva samples served as the controls (WSC) and (ASC).
For the saliva sample analysis, 50 μL each of different concentrations (0, 5, 10, 20 μg mL− 1) of OPN-spiked simulated or whole saliva
sample was taken in the microplate, and the incubation, washing, and
other necessary steps were kept the same as described in Section 2.5
2.4. Preparation of HRP-tagged anti-OPN/AuNP bioconjugates
Prior to the bioconjugation of AuNPs, the existing capping of AuNPs
was exchanged with MUA. For MUA capping, 25 mL of AuNSs or AuNRs
were taken, and equal volumes of phosphate buffer (10 mM, pH−6.8)
with 0.2 mg mL− 1 of tween- 20 were added and kept for 30-min incubation under mild stirring. To these solutions, MUA prepared in 1:7
ratio of ethanol and water was added and kept for stirring overnight at
room temperature. The MUA-capped nanoparticles were centrifuged
again and redispersed in Millipore water to obtain MUA-capped AuNSs
and AuNRs [30]. Based on the ICP-OES analysis, the concentrations of
MUA-capped AuNSs and AuNRs were found to be 57.92 and 21.27 μM,
respectively.
The coupling agents used for conjugation were EDC and NHS [31].
EDC (5 mM) and NHS (2.5 mM) solutions were prepared separately in
deionised water. 120 μL of EDC and NHS in a volume ratio of 1:1 were
simultaneously introduced to 1200 μL of the respective MUA-modified
nanoparticles (AuNSs or AuNRs). The solution was mixed thoroughly;
following which, 60 μL (25 ng μL− 1) of HRP-tagged anti-OPN was
added. The final mixture was incubated overnight at 37 °C. To remove
the excess unbound antibody, the mixture was centrifuged at 7000 rpm
for 10 min at 4 °C. The respective AuNP pellet was redispered in Millipore water. Non-specific active sites on the surface of the AuNP bioconjugate were blocked using 60 μL of blocking buffer [17]. This was
left for incubation for another 30 min at 37 °C and centrifuged again.
Finally, the pellet obtained was resuspended in phosphate buffer
(10 mM, pH 7.4) and used for the ELISA. After conjugation with HRPtagged anti-OPN, the total gold concentrations in AuNS and AuNR
bioconjugates were observed to be 54.15 and 20.61 μM, respectively. A
2.7. Characterization of AuNP and AuNP bioconjugates
UV–visible spectroscopic analysis was done for MUA-modified
AuNPs and also for the respective HRP-tagged anti-OPN/AuNP conjugates using UV-2600 UV–visible spectrophotometer (Shimadzu,
Tokyo, Japan). Morphology was analysed using transmission electron
microscopy (JEOL, JEM, 2100, Japan). Mean hydrodynamic diameter
and zeta potential measurements of as-synthesized NPs were taken
using a particle size analyzer (90 Plus Particle Analyzer, Brookhaven
Instruments, USA).
168
Clinica Chimica Acta 477 (2018) 166–172
D. Chakraborty et al.
citrate-capped AuNSs showed a characteristic surface plasmon resonance (SPR) peak at 523 nm. Alkanethiol coating around the gold
surface (due to MUA ligand exchange) affects the dispersity of the
AuNPs causing aggregation as a result, and the SPR peak red shifted to
530 nm [30]. Conjugation of HRP-tagged anti-OPN Ab on the MUAmodified AuNSs further increased the degree of aggregation Fig. 2(C).
The Ab coating on the AuNS surface changes the dielectric constant of
the medium, which indirectly affects the light scattering properties of
gold [33]; thus, the SPR peak was further shifted to 549 nm (19 nm red
shift as compared to the MUA-modified AuNSs).
Fig. 2(D) and (E) shows CTAB-capped and MUA-modified AuNRs
with an average aspect ratio of 3.43 ± 2.13 nm and 3.49 ± 2.09 nm,
respectively. Spectroscopic studies (Fig. S2 (B) ESI) for AuNRs indicated
a characteristic weak transverse surface plasmon resonance peak
(TSPR) at 515 nm and a strong longitudinal surface plasmon resonance
(LSPR) peak at 879 nm [29]. MUA modification does not alter the
morphology or the stability of the AuNRs (evident from the TEM image
and spectral study), whereas conjugation of HRP-tagged anti-OPN Ab to
the AuNRs modifies its surface properties, and induces mild aggregation
Fig. 2(F). This in turn generates a plasmon coupling effect, causing a
reduction of the LSPR peak intensity along with peak broadening [34].
Zeta potential measurements were taken to confirm the successful
ligand exchange on the AuNP surface. Citrate ions on AuNS surface
impart a negative zeta value of − 44.6 ± 1.63 mV. MUA modification
changes the surface charge to − 38.8 ± 1.22 mV as it incorporates
carboxylate moieties on the AuNS surface. CTAB-capped AuNRs showed
a zeta potential value of 49.89 ± 1.34 mV (positively charged due to
CTA+ group of the surfactant), which after ligand exchange with MUA,
changed to −45.65 ± 1.55 mV due to the presence of carboxyl group
of MUA.
2.8. Statistical analysis
All the experiments were performed in triplicates, keeping ± 5% as
the error limit, and the results were expressed as their mean values with
the S.D. The experiments were repeated to evaluate its reproducibility.
Statistical significance was determined using one-way ANOVA with
Bonferroni's multiple comparison test in Microsoft Excel.
3. Results and discussion
3.1. Over expression of OPN mRNA and protein in oral tongue cancer
compared to normal
There was a significant overexpression of OPN mRNA in tongue
tumours compared to adjacent margins and normal tissues (Fig. S1).
Tumours (n = 68) showed 23.75-fold overexpression of OPN mRNA
compared to adjacent histologically proven absolute normals (n = 11).
Histologically abnormal adjacent surgical margins (n = 6) showed
7.44-fold overexpression of OPN compared to absolute normal.
OPN protein expression was assessed by immunohistochemistry in
retrospective FFPE sections (n = 146), which comprised of tongue tumours (n = 138), surgical margins: histologically abnormal (n = 6),
and absolute histologically normals (n = 2) The OPN protein was expressed in >50% of the tumour cells in almost all the tumours stained,
compared to margins showing expression of OPN in < 50% of the
margins, and the normals were negative for OPN. (Fig. 1)
3.2. Morphological and spectroscopic analysis of AuNPs
Monodisperse,
spherical-shaped,
citrate-capped
AuNSs
(15.29 ± 1.23 nm) and MUA-modified AuNSs (15.43 ± 1.77 nm) with
mild aggregation can be seen in Fig. 2(A) and (B). The morphology
correlated well with the spectroscopic studies (Fig. S2(A)), where
Fig. 1. (A) Normal Tongue Epithelium showing negative staining for
OPN (10 × magnification) (B) Normal tongue epithelium with inflammatory cells showing negative staining for OPN (40 × magnification) (C) Well-differentiated tongue squamous cell carcinoma
showing intense cytoplasmic staining for OPN (10 × magnification)
(D) cytoplasmic staining for OPN (40 × magnification).
169
Clinica Chimica Acta 477 (2018) 166–172
D. Chakraborty et al.
Fig. 2. TEM image (A) Citrate-capped AuNSs (B) MUA-capped AuNSs, (C) AuNSs after conjugation with anti-OPN-HRP (D) CTAB-capped AuNRs (E) MUA-capped AuNRs (F) AuNRs after
conjugation with anti-OPN-HRP.
bioconjugates were used and 7 folds (LOD: 0.02 ng mL− 1) when AuNR
bioconjugates were incorporated. An overall linearity in the trend in the
immunoassay was observed, and it was observed to be accurate and
much better than conventional ELISA. However, due to the higher
stability of AuNR bioconjugates and their slightly improved detection
limit when compared to AuNS bioconjugates, the former was more
preferred.
The p-value calculated from one-way ANOVA was observed to be
< 0.05, which indicated the statistical significance of the difference in
absorbance obtained between each set of OPN concentrations in the
tested range. An overview of the current immunosensors available for
detecting various other oral cancer biomarkers have been summarised
in Table 1, indicating the comparable sensitivity of our method with the
state-of-the-art techniques. Additionally, the current method is comparatively simpler, given the inherent process complexities and requirement of expensive instrumentation and trained manpower in the
other assays.
To evaluate the reproducibility of the developed method, run-torun, day-to-day, and batch-to-batch precision were analysed for different concentrations of OPN. The relative standard deviation (% RSD)
was found to be 1.80, 1.85, and 2.75, respectively, which was within
the acceptable range (Table S4); thereby, ascertaining the repeatability
of the nanoELISA.
3.3. Optimization of the assay for sensor development
Various parameters were optimized for enhancing the sensitivity of
AuNP-based immunoassay. Primarily, the EDC and NHS concentrations
were chosen as 5 and 2.5 mM as higher concentrations of EDC and NHS
caused AuNP aggregation. After the AuNPs (AuNSs or AuNRs) were
successfully conjugated with the HRP-tagged anti-OPN, the volume
ratios of the fabricated AuNPs:blocking buffer were varied as (1:1, 2:1,
4:1, 20:1, and 40:1). The optimum volume ratio was determined by
measuring the difference in absorption intensity between control i.e.
0 ng mL− 1 and the highest OPN concentration used in the experiment
i.e. 20 ng mL− 1. The highest value was obtained when the fabricated
AuNPs:blocking buffer volume ratio was 20:1 (Table S2). The incubation time for HRP-tagged anti-OPN/AuNPs with OPN antigen was also
optimized in a similar manner. The interaction time periods were varied
as 1 h, 2 h, and 4 h. The highest difference in absorption intensity was
seen for 2-h incubation (Table S3).
Hence, EDC and NHS of concentrations 5 and 2.5 mM, 20:1 as the
optimum AuNP: blocking buffer volume ratio, and 2-h incubation time
for HRP-tagged anti-OPN/AuNPs were set as the optimized conditions
for the immunoassay.
3.4. Sensitivity assessment and precision analysis of the nano-ELISA system
3.5. Interference studies of the nanoELISA system
To assess the sensitivity of the nano-ELISA system, the prepared
OPN standards (0.31, 0.63, 1.25, 2.5, 5, 10, and 20 ng mL− 1) were
tested with AuNP bioconjugates as well as with HRP-labelled polyclonal
antibody specific for human OPN under the optimized conditions.
Three different standard curves for OPN concentrations vs. optical
density were plotted, and a good linear correlation (R2 = 0.99) was
obtained for all the cases (Fig. 3). The signal enhancements (as compared to conventional ELISA) of the two different bioconjugates were
nearly same. The experimentally calculated LOD (range:
0.313–20 ng mL− 1) for conventional ELISA was 0.14 ng mL− 1, which
was seen to be improved by ~5 folds (LOD: 0.03 ng mL− 1) when AuNS
The performance of the immunosensor was evaluated in the presence of other interferents that are commonly found in natural saliva.
Fig. 4 shows the effect on the color development in the ELISA system
when different concentrations of these interferents were used., i.e. potassium chloride (0.6 mg mL− 1), sodium chloride (1.16 mg mL− 1),
glucose (7 mg mL− 1), glycine (0.17 mM), α-amylase (120 kU L− 1),
vitamin-C (0.687 μmol L− 1) and (OPN (20 ng mL− 1) were taken, and
the immunoassay was performed. The response obtained in the case of
the target analyte i.e. OPN was ~16–17 times higher as compared to
170
Clinica Chimica Acta 477 (2018) 166–172
D. Chakraborty et al.
Fig. 3. Standard curves for (A) conventional ELISA, (B) Nano-ELISA with GNPs, (C)NanoELISA with GNRs.
the tested interferents (much higher concentration).
3.6. Spiked artificial and whole saliva sample analysis
Since HRP-tagged anti-OPN/AuNRs were found to be more stable
when tested for a period of 14 days and showed improved LOD as
compared to the HRP-tagged anti-OPN/AuNSs conjugates, the immunoassay study in the spiked saliva samples was performed with HRPtagged anti-OPN/AuNRs. From the values depicted in Table 2, it can be
inferred that when different concentrations of OPN spiked in simulated
saliva sample were analysed, a corresponding response that correlated
well with the standard curve were obtained. The percentage recovery
was calculated to be 97 ± 0.7, 95.7 ± 0.52, and 97.85 ± 0.22 for OPN
concentrations, 5, 10, and 20 ng mL− 1, respectively.
To assess the performance of the HRP-tagged anti-OPN/AuNRs
under the influence of real matrix, the same study was repeated with
spiked whole saliva sample, and the results are summarised in Table 2.
From the signal obtained, it was evident that the AuNR-based sensor
system responded well under natural conditions. The recovery percentage obtained was almost similar to that of spiked artificial saliva
sample, i.e. 95.4 ± 0.7, 96.5 ± 0.43, and 97.6 ± 0.17 for OPN concentrations, 5, 10, and 20 ng mL− 1, respectively. The accuracy of the
developed procedure is thus proved, and it also indicated that the developed nanoELISA would be a promising non-invasive alternative
when used for real sample analysis in the future.
Fig. 4. Effect on response generation in the presence of interferents.
wide linear detection range (0.31–20 ng mL− 1), and remarkably low
LOD. The developed nanoELISA was also found to be highly selective
against the tested interferents and exhibited excellent reproducibility.
The method also showed good recovery rates for artificial and normal
saliva samples spiked with OPN. The nanoELISA technique can be validated further in larger series of samples from oral cancer patients.
Additionally, other secretory biomarkers can be explored in combination with OPN to form a panel of markers that can be tested in future for
early detection of oral cancer.
4. Conclusions and future outlook
Acknowledgement
A biomarker that is excess in production and plays an important role
in pathogenesis and progression of oral cancer tissue can be ideally
explored for testing in saliva since it has been useful like a liquid biopsy
in oral cancer diagnostics. There was a significant overexpression of
OPN mRNA in tongue tumours as compared to normal, which correlated well with the overexpression of OPN protein that was observed in
IHC studies. Therefore, a non-invasive AuNP-based ELISA system for
OPN detection was developed, which exhibited high sensitivity, had
The authors would like to acknowledge SAIF IIT-Madras for the ICPOES analysis. This research did not receive any specific grant from
funding agencies in public, commercial, or not-for-profit sectors.
Conflict of interest
The authors declare that they do not have any conflict of interest.
Table 1
Characteristics of nanoELISA as compared to existing immunosensors for oral cancer.
S No.
1
2
3
4
5
Nanoparticles
Cys-La(OH)3/ITO
APTES/ZrO2-RGO/ITO
APTES/nHfO2/ITO
Gold NP layers
AuNS- ELISA
AuNR-ELISA
Biomarker
Detection range
−1
CYFRA-21-1
CYFRA-21-1
CYFRA-21-1
CEA
OPN
0–18 ng mL
2–22 ng mL− 1
2–18 ng mL− 1
2–64 ng mL− 1
0.31–20 ng mL− 1
171
Limit of detection
−1
0.001 ng mL
0.12 ng mL− 1
0.21 ng mL− 1
2 ng mL− 1
0.03 ng mL− 1
0.02 ng mL− 1
Reference
[17]
[18]
[19]
[35]
Our work
Clinica Chimica Acta 477 (2018) 166–172
D. Chakraborty et al.
Table 2
OPN estimation in spiked artificial and whole saliva sample.
Spiked OPN (ng mL− 1)
0
5
10
20
Measured OPN concentration (ng mL− 1)
Normalised OPN concentration (ng mL− 1)
% Recovery
Artificial saliva
Artificial saliva
Whole saliva
Artificial saliva
Whole saliva
–
4.85 ± 0.04
9.57 ± 0.07
19.57 ± 0.05
–
4.77 ± 0.05
9.65 ± 0.10
19.52 ± 0.01
–
97 ± 0.7
95.7 ± 0.52
97.85 ± 0.22
–
95.4 ± 0.7
96.5 ± 0.43
97.6 ± 0.17
0.05 ± 0.03
4.9 ± 0.03
9.62 ± 0.05
19.62 ± 0.04
Whole
saliva
0.03 ± 0.04
4.8 ± 0.06
9.68 ± 0.07
19.55 ± 0.02
[17] S. Tiwari, P.K. Gupta, Y. Bagbi, T. Sarkar, P.R. Solanki, L-cysteine capped lanthanum hydroxide nanostructures for non-invasive detection of oral cancer biomarker, Biosens. Bioelectron. 89 (2017) 1042–1052.
[18] S. Kumar, J.G. Sharma, S. Maji, B.D. Malhotra, Nanostructured zirconia decorated
reduced graphene oxide based efficient biosensing platform for non-invasive oral
cancer detection, Biosens. Bioelectron. 78 (2016) 497–504.
[19] S. Kumar, S. Kumar, S. Tiwari, S. Augustine, S. Srivastava, B.K. Yadav,
B.D. Malhotra, Highly sensitive protein functionalized nanostructured hafnium
oxide based biosensing platform for non-invasive oral cancer detection, Sensors
Actuators B Chem. 235 (2016) 1–10.
[20] Y. Si, Z. Sun, N. Zhang, W. Qi, S. Li, L. Chen, H. Wang, Ultrasensitive electroanalysis
of low-level free microRNAs in blood by maximum signal amplification of catalytic
silver deposition using alkaline phosphatase-incorporated gold nanoclusters, Anal.
Chem. 86 (2014) 10406–10414.
[21] N. Zhang, Y. Si, Z. Sun, L. Chen, R. Li, Y. Qiao, H. Wang, Rapid, selective, and
ultrasensitive fluorimetric analysis of mercury and copper levels in blood using
bimetallic gold–silver nanoclusters with “silver effect”-enhanced red fluorescence,
Anal. Chem. 86 (2014) 11714–11721.
[22] N. Zhang, Y. Si, Z. Sun, S. Li, S. Li, Y. Lin, H. Wang, Lab-on-a-drop: biocompatible
fluorescent nanoprobes of gold nanoclusters for label-free evaluation of phosphorylation-induced inhibition of acetylcholinesterase activity towards the ultrasensitive detection of pesticide residues, Analyst 139 (2014) 4620–4628.
[23] L. Zhang, S. Li, M. Dong, Y. Jiang, R. Li, S. Zhang, X. Lv, L. Chen, H. Wang,
Reconstituting redox active centers of heme-containing proteins with biomineralized gold toward peroxidase mimics with strong intrinsic catalysis and electrocatalysis for H2O2 detection, Biosens. Bioelectron. 87 (2017) 1036–1043.
[24] I.H. El-Sayed, X. Huang, M.A. El-Sayed, Surface plasmon resonance scattering and
absorption of anti-EGFR antibody conjugated gold nanoparticles in cancer diagnostics: applications in oral cancer, Nano Lett. 5 (2005) 829–834.
[25] X. Liu, Q. Dai, L. Austin, J. Coutts, G. Knowles, J. Zou, H. Chen, Q. Huo, A one-step
homogeneous immunoassay for cancer biomarker detection using gold nanoparticle
probes coupled with dynamic light scattering, J. Am. Chem. Soc. 130 (2008)
2780–2782.
[26] C.D. Medley, J.E. Smith, Z. Tang, Y. Wu, S. Bamrungsap, W. Tan, Gold nanoparticlebased colorimetric assay for the direct detection of cancerous cells, Anal. Chem. 80
(2008) 1067–1072.
[27] A. Ambrosi, F. Airo, A. Merkoçi, Enhanced gold nanoparticle based ELISA for a
breast cancer biomarker, Anal. Chem. 82 (2009) 1151–1156.
[28] I.E. Paul, A. Rajeshwari, T. Prathna, A.M. Raichur, N. Chandrasekaran,
A. Mukherjee, Colorimetric detection of melamine based on the size effect of
AuNPs, Anal. Methods 7 (2015) 1453–1462.
[29] L. Feng, Z. Xuan, J. Ma, J. Chen, D. Cui, C. Su, J. Guo, Y. Zhang, Preparation of gold
nanorods with different aspect ratio and the optical response to solution refractive
index, J. Exp. Nanosci. 10 (2015) 258–267.
[30] D. Li, Q. He, Y. Cui, L. Duan, J. Li, Immobilization of glucose oxidase onto gold
nanoparticles with enhanced thermostability, Biochem. Biophys. Res. Commun. 355
(2007) 488–493.
[31] D. Bartczak, A.G. Kanaras, Preparation of peptide-functionalized gold nanoparticles
using one pot EDC/sulfo-NHS coupling, Langmuir 27 (2011) 10119–10123.
[32] L. Liu, C. Qiu, Q. Chen, S. Zhang, Corrosion behavior of Zr-based bulk metallic
glasses in different artificial body fluids, J. Alloys Compd. 425 (2006) 268–273.
[33] J.C.Y. Kah, K.W. Kho, C.G.L. Lee, C.J.R. Sheppard, Z.X. Shen, K.C. Soo, M.C. Olivo,
Early diagnosis of oral cancer based on the surface plasmon resonance of gold nanoparticles, Int. J. Nanomedicine 2 (2007) 785.
[34] W.M. Park, B.G. Choi, Y.S. Huh, W.H. Hong, S.Y. Lee, T.J. Park, Facile functionalization of colloidal gold nanorods by the specific binding of an engineered protein
that is preferred over CTAB bilayers, ChemPlusChem 78 (2013) 48–51.
[35] F. Zhou, M. Wang, L. Yuan, Z. Cheng, Z. Wu, H. Chen, Sensitive sandwich ELISA
based on a gold nanoparticle layer for cancer detection, Analyst 137 (2012)
1779–1784.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.cca.2017.09.009.
References
[1] M. Nagata, H. Kurita, K. Uematsu, S. Ogawa, K. Takahashi, H. Hoshina, R. Takagi,
Diagnostic value of cyclin-dependent kinase/cyclin-dependent kinase inhibitor expression ratios as biomarkers of locoregional and hematogenous dissemination risks
in oral squamous cell carcinoma, Mol. Clin. Oncol. 3 (2015) 1007–1013.
[2] I. Skrinjar, V. Brailo, D. Vidovic-Juras, V. Vucicevic-Boras, A. Milenovic, Evaluation
of pretreatment serum interleukin-6 and tumour necrosis factor alpha as a potential
biomarker for recurrence in patients with oral squamous cell carcinoma, Med. Oral
Patol. Oral Cir. Bucal. 20 (2015) e402.
[3] J.-Y. Tang, E. Hsi, Y.-C. Huang, N.C.-H. Hsu, P.-Y. Chu, C.-Y. Chai, High LC3 expression correlates with poor survival in patients with oral squamous cell carcinoma, Hum. Pathol. 44 (2013) 2558–2562.
[4] L. Morandi, D. Gissi, A. Tarsitano, S. Asioli, V. Monti, G. Del Corso, C. Marchetti,
L. Montebugnoli, M.P. Foschini, DNA methylation analysis by bisulfite next-generation sequencing for early detection of oral squamous cell carcinoma and highgrade squamous intraepithelial lesion from oral brushing, J. Cranio-Maxillofac.
Surg. 43 (2015) 1494–1500.
[5] M. Krüger, J. Amort, P. Wilgenbus, J.P. Helmstädter, I. Grechowa, J. Ebert,
S. Tenzer, M. Moergel, I. Witte, S. Horke, The anti-apoptotic PON2 protein is Wnt/
β-catenin-regulated and correlates with radiotherapy resistance in OSCC patients,
Oncotarget 7 (2016) 51082.
[6] H. Rangaswami, A. Bulbule, G.C. Kundu, Osteopontin: role in cell signaling and
cancer progression, Trends Cell Biol. 16 (2006) 79–87.
[7] S. Rittling, A. Chambers, Role of osteopontin in tumour progression, Br. J. Cancer
90 (2004) 1877–1881.
[8] G.F. Weber, The metastasis gene osteopontin: a candidate target for cancer therapy,
Biochim. Biophys. Acta (BBA)-Rev. Cancer 1552 (2001) 61–85.
[9] C.H. Österreicher, M. Penz-Österreicher, S.I. Grivennikov, M. Guma, E.K. Koltsova,
C. Datz, R. Sasik, G. Hardiman, M. Karin, D.A. Brenner, Fibroblast-specific protein 1
identifies an inflammatory subpopulation of macrophages in the liver, Proc. Natl.
Acad. Sci. 108 (2011) 308–313.
[10] M. Ahmed, R. Behera, G. Chakraborty, S. Jain, V. Kumar, P. Sharma, A. Bulbule,
S. Kale, S. Kumar, R. Mishra, Osteopontin: a potentially important therapeutic
target in cancer, Expert Opin. Ther. Targets 15 (2011) 1113–1126.
[11] M. Bache, M. Kappler, H. Wichmann, S. Rot, A. Hahnel, T. Greither, H.M. Said,
M. Kotzsch, P. Würl, H. Taubert, Elevated tumor and serum levels of the hypoxiaassociated protein osteopontin are associated with prognosis for soft tissue sarcoma
patients, BMC Cancer 10 (2010) 132.
[12] X. Hou, X. Wu, P. Huang, J. Zhan, T. Zhou, Y. Ma, T. Qin, R. Luo, Y. Feng, Y. Xu,
Osteopontin is a useful predictor of bone metastasis and survival in patients with
locally advanced nasopharyngeal carcinoma, Int. J. Cancer 137 (2015) 1672–1678.
[13] D. Petrik, P.W. Lavori, H. Cao, Y. Zhu, P. Wong, E. Christofferson, M.J. Kaplan,
H.A. Pinto, P. Sutphin, A.C. Koong, Plasma osteopontin is an independent prognostic marker for head and neck cancers, J. Clin. Oncol. 24 (2006) 5291–5297.
[14] D. Vordermark, H.M. Said, A. Katzer, T. Kuhnt, G. Hänsgen, J. Dunst, M. Flentje,
M. Bache, Plasma osteopontin levels in patients with head and neck cancer and
cervix cancer are critically dependent on the choice of ELISA system, BMC Cancer 6
(2006) 207.
[15] M. Perfézou, A. Turner, A. Merkoçi, Cancer detection using nanoparticle-based
sensors, Chem. Soc. Rev. 41 (2012) 2606–2622.
[16] X. Huang, R. O'Connor, E.A. Kwizera, Gold nanoparticle based platforms for circulating cancer marker detection, Nanotheranostics (Sydney, NSW) 1 (2017) 80.
172