- Email: [email protected]
Sensitive detection of C-reactive protein using optical fiber Bragg gratings
Biosensors and Bioelectronics 65 (2015) 251–256
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Sensitive detection of C-reactive protein using optical fiber Bragg gratings Sridevi S.a, K.S. Vasu b, S. Asokan a,c, A.K. Sood b,n a
Department of Instrumentation and Applied Physics, Indian Institute of Science, Bangalore 560012, India Department of Physics, Indian Institute of Science, Bangalore 560012, India c Robert Bosch Centre for Cyber Physical Systems, Indian Institute of Science, Bangalore 560012, India b
art ic l e i nf o
a b s t r a c t
Article history: Received 17 July 2014 Received in revised form 28 September 2014 Accepted 15 October 2014 Available online 23 October 2014
An accurate and highly sensitive sensor platform has been demonstrated for the detection of C-reactive protein (CRP) using optical fiber Bragg gratings (FBGs). The CRP detection has been carried out by monitoring the shift in Bragg wavelength (ΔλB ) of an etched FBG (eFBG) coated with an anti-CRP antibody (aCRP)-graphene oxide (GO) complex. The complex is characterized by Fourier transform infrared spectroscopy, X-ray photoelectron spectroscopy and atomic force microscopy. A limit of detection of 0.01 mg/L has been achieved with a linear range of detection from 0.01 mg/L to 100 mg/L which includes clinical range of CRP. The eFBG sensor coated with only aCRP (without GO) show much less sensitivity than that of aCRP–GO complex coated eFBG. The eFBG sensors show high specificity to CRP even in the presence of other interfering factors such as urea, creatinine and glucose. The affinity constant of ∼ 1.1 × 1010 M−1 has been extracted from the data of normalized shift (ΔλB /λB ) as a function of CRP concentration. & Elsevier B.V. All rights reserved.
Keywords: C-reactive protein Fiber Bragg gratings Graphene oxide Bragg wavelength
1. Introduction C-reactive protein (CRP) is an acute phase protein, found in blood plasma and its concentration increases up to three orders during the inflammatory process occurring in the human body (Gabay and Kushner, 1999). CRP concentration below 1 mg/L represents low risk, between 1 and 3 mg/L represents average risk and above 3 mg/L represents the high risk of cardiovascular diseases (Lee et al., 2011). Half-life of plasma CRP is approximately 19 h (Mold et al., 1999) and the CRP level decreases rapidly after recovery or treatment. Thus, CRP has emerged as an effective biomarker for cardiovascular diseases, infections in human body and can be used as prototypical inflammatory predictor (Li and Fang, 2004; May and Wang, 2007). The structure of human CRP is described as ‘pentraxin’ in which five identical homologous subunits are non-covalently associated around a central pore. CRP binds to the phosphocholine present on microbes, dead and damaged cells to initiate the removal process. The commonly used CRP detection techniques are radial immunodiffusion (RID), radio immunoassay (RIA), immunonephelometry (IN), immunoturbidimetry (IT), immunofluorescence, immuno chemiluminescence and standard enzyme immunoassay (as n
Corresponding author. Tel.: +91 80 2293 2964; fax: +91 80 2360 2602. E-mail address: [email protected] (A.K. Sood).
http://dx.doi.org/10.1016/j.bios.2014.10.033 0956-5663/& Elsevier B.V. All rights reserved.
ELISA) and the detection limit in these techniques ranges between 0.1 and 0.2 mg/L (Centi et al., 2009). The preparation of monoclonal and polyclonal antibodies used in the above-mentioned methods varies depending on the immune response in different animals and causes the variable specificity and sensitivity of CRP detection (Dajani et al., 2001). The photometric methods like ELISA can be affected by the colour and composition of the medium used in assays (Meyer et al., 2006). Many other methods such as surface plasmon resonance (SPR) (Meyer et al., 2006; Hu et al., 2006), quartz crystal microbalance (QCM) (Kim et al., 2009), piezoelectric micro-cantilever method (Wee et al., 2005; Gan et al., 2012), electrochemistry (Centi et al., 2009; Buch and Rishpon, 2008), electrochemical impedance spectroscopy (EIS) (Bryan et al., 2013) and micro particle-tracking velocimetry (MPTV) (Fan et al., 2010) have been developed for CRP detection. Laser induced fluorescence (Lin et al., 2013) and chemiluminescence (Islam and Kang, 2011) methods have also used for the detection of CRP. Christodoulides et al. (2005) have reported the CRP detection using human saliva rather than the blood serum in ELISA tests. Further, for the detection of CRP, carbon nanomaterials such as carbon nanofibers, graphene oxide (GO) and reduced graphene oxide (RGO) have also been used in EIS and chemiluminescence methods (Gupta et al., 2014; Lee et al., 2012). Fiber Bragg gratings are efficient optical devices in the field of biosensors (Tripathi et al., 2012; Guo et al., 2014), with the Bragg
252
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
wavelength (λB ) and the transmission intensity as probing parameters. Sensitivity of FBG sensors can be enhanced by etching the cladding layer and exposing the core directly to the surrounding medium. Recently, it has been shown that the GO coated etched fiber Bragg gratings (eFBG) can be used for the detection of carbohydrate-protein interactions (Sridevi et al., 2014). In the present study, an optical platform has been demonstrated for accurate detection of CRP using GO coated eFBG sensors. GO flakes are attached with anti-CRP antibodies (aCRP) through the formation of covalent functionalization before coating onto the eFBG sensors. The present eFBG sensors have shown a high specificity to CRP even in the presence of other interfering factors such as urea, creatinine and glucose, with the limit of detection of 0.01 mg/L. Further, the sensors have shown excellent performance in the linear range of detection from 0.05 mg/L to 100 mg/L.
2. Materials and experimental methods 2.1. Materials CRP detection experiments have been carried out with the Quantia CRP-UV kit, which is generally used in turbidimetric immunoassay for CRP detection, purchased from M/S Tulip Diagnostics (P), Ltd. Basically, the kit contains ready to use solution of aCRP of IgG class which are highly specific to CRP, a calibrator which contains the human CRP in lyophilized serum (traceable to W.H.O. international reference standard (85/506) for CRP). The Quantia CRP-UV kit has been designed to measure CRP concentrations in the range of 6–100 mg/L with a detection limit of 3 mg/L by using the UV–vis absorption spectroscopy as probing method. No prozone effect has been observed with CRP concentration upto 1 g/L. N-(3-dimethylaminopropyl)-N-ethylcarbodiimde hydrochloride (EDC), N- hydroxysuccinimide (NHS) and the interfering factors such as urea, creatinine and glucose, of analytical grade, purchased from Sigma-Aldrich Chemical Co. are used without further purification. Graphite powder, purchased from Superior Graphite.Co. (Riverside, Chicago) for graphite oxide preparation. 2.2. Preparation of graphene oxide The graphite oxide (GO) is prepared (Marcano et al., 2010) by stirring 2 g of graphite flakes in acid (H2SO4 : H3PO4 ¼110 mL : 12 mL) followed by very slow addition of KMnO4 (11.1 g) and this one-pot mixture is stirred for 3 days at room temperature. After 3 days, 20 mL of H2O2 is added to the mixture to terminate the oxidation process. The resultant material is then washed repeatedly with 1M HCl and distilled water and then dried under vacuum to obtain graphite oxide. 100 μg/mL GO aqueous dispersion has been prepared by sonicating 1 mg of graphite oxide in 10 mL of deionized (DI) water.
2.4. Fabrication of FBG and preparation of eFBG Phase-mask technique is used to inscribe the gratings in germania doped core of a single mode photosensitive silicate fiber (M/S Nufern) of diameter 125 μm . KrF excimer UV laser of wavelength 248 nm (pulse energy 6 mJ and repetition rate of 200 Hz) and a phase mask of 1069 nm pitch (M/s Stocker Yale, Inc.) are used for the grating inscription. 40% HF solution is used for etching the cladding layer around the grating region (∼3 mm ) of the FBG followed by washing with DI water. In all our experiments the Bragg wavelength (λB) is monitored using an optical interrogator (Micron Optics, SM130) with a wavelength repeatability of 1 pm. 2.5. Functionalization of the eFBG The etched portion of eFBG is subsequently treated with H2O:H2O2:NH4OH (5:1:1) solution, which is followed by washing with DI water to make the eFBG surface hydrophilic in nature. The etched portion of eFBG is dipped into 200 μL of aqueous dispersion of aCRP–GO complex and allowed to dry at room temperature. After drying the aCRP–GO complex coated eFBG is thoroughly washed for 15 min to remove unbound particles. In all the CRP detection experiments, the eFBG sensors coated with aCRP–GO complex were incubated in bovine serum albumin (BSA, 10 μg/ml) for 15 min to block the unoccupied area by the aCRP molecules in aCRP–GO complex. This was followed by washing away the unattached BSA by DI water. The Bragg wavelength value corresponding to the aCRP–GO complex coated eFBG after dipping in BSA is taken as reference value (λB0) for the CRP detection experiments. CRP aqueous solutions of various concentrations ranging from 10−3 mg/L to 100 mg/L are prepared for the experiments. For the experiments with interfering factors, different concentrations of aqueous solutions of urea, glucose and creatinine ranging from 1 mg/L to 10 g/L are prepared. 2.6. Experimental setup Teflon block patterned with micro-fluidic channels of volume 200 μL is used to dip the eFBG region coated with aCRP–GO complex in different concentrations of CRP solution of 200 μL . After 10 min dip, the eFBG sensor is taken out to measure the Bragg wavelength and the same procedure was followed at each concentration of CRP solution varying from 1 μg/L to 100 mg/L. All our measurements throughout the experiment were carried out
2.3. Preparation of aCRP–GO complex The preparation of aCRP–GO complex is carried out by stirring 1 mL of GO aqueous dispersion for 30 min at 25 °C after addition of 2 mg of EDC and 4 mg of NHS. Subsequently, 200 μL of aCRP solution is added and this mixture is kept in the orbital shaker for 10 h to prepare aCRP–GO complex. For the Fourier transform infrared (FTIR) spectroscopy and X-ray photoelectron spectroscopy (XPS) experiments, the aqueous dispersions of aCRP–GO complex is filtered using anodisc filter to form a thin film. The atomic force microscopy (AFM) experiments have been carried out by imaging the same GO flake deposited on the 300 nm SiO2/Si before and after the functionalization with aCRP.
Fig. 1. FTIR spectra of GO and aCRP–GO complex. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
without removing the eFBG sensors from the clamp to avoid bending effect in measuring the Bragg wavelength. For one complete set of experiments with different concentrations of CRP, the
253
same FBG sensor is used and disposed. The concentrations plotted in Figs. 4a and b and 5 are the cumulative concentrations.
3. Results and discussion 3.1. Characterization of aCRP–GO complex
Fig. 2. XPS spectra of GO and aCRP–GO complex. The spectral component shown by dotted line corresponds to C–N bond.
a
c
b
d
EDC is the most commonly used coupling reagent for the amide bond formation between –COOH and –NH2 groups (Cao et al., 2006). The addition of EDC to the GO aqueous dispersion activates the – COOH groups present on GO flakes. The presence of NHS increases the coupling efficiency of EDC (Cao et al., 2006). The EDC-NHS activated –COOH groups at the edges of GO flakes bind with –NH2 groups in the aCRP to form an amide bond by releasing the water molecules. Fig. 1 shows the FTIR spectra of GO and aCRP–GO complex. The pristine GO sample (red colour solid line) shows characteristic bands at 1060, 1229 and 1405 cm 1 (C–O vibrations in epoxy bonds), 1630 cm 1 (CQC vibrations), 1737 cm 1 (CQO vibrations in carboxylic group) and 3415 cm 1 (O–H vibrations) (Marcano et al., 2010; Hu et al., 2014). After the functionalization
Fig. 3. AFM image of the GO flake (a) before and (b) after covalent functionalization with aCRP. (c) Height profile of before (black color solid line) and after (red color solid line) covalent functionalization. (d) SEM image of aCRP–GO complex coated eFBG sensor after CRP detection experiments. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
254
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
with aCRP, the band at 1737 cm 1 (corresponds to CQO vibrations in carboxylic groups) disappears and new bands appear at 1163 and 1238 cm 1 (corresponds to C–N vibration), 1395 and 1446 cm 1 (corresponds to N–H vibrations), 1535 cm 1 (corresponds to amideII mode, an out-of-phase combination of the N–H inplane bend and the C–N stretching vibration) and 3269 cm 1 (corresponds to N–H vibrations) confirms the functionalization of GO flakes with aCRP through amide bond formation (Hu et al., 2014; Lu and Li, 2008). Further, the XPS experiments were carried out to confirm the complex formation. Fig. 2 shows the C1s core level spectrum of the XPS spectra of GO before (black color solid spheres) and after (black color open circles) the functionalization with aCRP. Before the functionalization GO shows typical C1s spectrum and is deconvoluted into 5 bands centered at 284.1 (CQC), 284.9 (C–C and C–H), 286.5 (C– OH and C–O–C), 287.1 (CQO) and 288.4 eV (O ¼ C–O). In the deconvoluted C1s spectrum of aCRP–GO complex a peak appears at 285.6 (shown by dotted line in Fig. 2) corresponding to C–N apart from the conventional GO peaks at 284.1 (CQC), 284.6 (C–C and C–H), 286.8 (C–OH and C–O–C), corroborating amide bond formation between the –COOH groups in GO flakes and the –NH2 groups in aCRP. Fig. 3a and b shows the tapping mode AFM images of GO flake deposited on SiO2/Si substrate before and after functionalization of aCRP. From the height profile (shown in Fig. 3c), the average height of the GO flake before functionalization is 1.5 nm and after the functionalization it changes to 4.5 nm with increased roughness on the GO flakes. The change in average height and appearance of brighter spots on the GO flake after treated with aCRP confirm the functionalization. Fig. 3d shows the scanning electron microscope image of aCRP–GO complex coated eFBG. 3.2. Detection of CRP Fig. 4a shows the shift in Bragg wavelength (ΔλB ) as a function of concentration of CRP for (i) only aCRP and (ii) aCRP–GO complex coated eFBG sensors. The observed shift in Bragg wavelength for aCRP–GO complex coated eFBG sensors (∼6.3 pm per one order increase in the CRP concentration) is ∼5 times more than that of eFBG sensors coated with only aCRP (1.2 pm per one order increase in the CRP concentration). Also, the limit of detection in aCRP–GO complex coated eFBG sensors (0.01 mg/L) shows that the presence of GO enhances the sensitivity when compared to the eFBG sensors coated with only aCRP molecules (the limit of detection is 1 mg/L). This is due to two reasons: (i) immobilization of antibodies (aCRP molecules) on the surface of eFBG sensors through binding aCRP molecules to GO via amide bond formation (facilitated by EDC-NHS) and the efficacy of GO in binding to the hydrophilic surface of eFBG are main reasons for higher sensitivity in the eFBG sensors coated with aCRP–GO complex. As mentioned before, GO binds strongly to the hydrophilic surface of eFBG through hydrogen bonding between the –OH groups present on the hydrophilic etched FBG surface and the –COOH, –OH, –O– groups on GO. In direct coating of aCRP molecules on eFBG sensor, the bonding between aCRP molecules and the surface of eFBG sensor is physical and weak when compared to the binding of aCRP–GO complex onto the surface of eFBG sensor. (ii) The change in Bragg wavelength is directly proportional to the change in effective refractive index (Δneff ), i.e. ΔλB = 2Δneff Λ, where Λ is grating period and neff is a function of ncore and nclad given by (Kashyap, 1999; Sridevi et al., 2014)
⎤ ⎡⎛ n 2 − n 2 ⎞⎛ ⎞ eff clad ⎟⎜ ncore − nclad ⎟ ⎥ neff = nclad ⎢⎜⎜ 2 + 1 ⎟ ⎜ ⎟ 2 ⎥⎦ ⎢⎣⎝ ncore − neff nclad ⎠⎝ ⎠
(1)
Fig. 4. (a) The shift in Bragg wavelength as a function of CRP concentration in (i) aCRP (purple color open circles) and (ii) aCRP–GO complex (olive color green squares) coated eFBG sensors. Inset: Shift in Bragg wavelength as a function of concentration of only interfering factors such as urea (blue color open squares), glucose (magenta color open circles) and creatinine (red color open stars). (b) The shift in Bragg wavelength as a function of CRP concentration in the presence of urea (1800 mg/L) – blue color open squares; glucose (3800 mg/L) – magenta color open circles and creatinine (5800 mg/L) – red color open stars. The olive color solid spheres represents the data for shift in Bragg wavelength as a function of CRP concentration in the absence of interfering factors. Inset: difference in ΔλB for different concentrations of CRP as a function of CRP concentration in the presence of urea (blue color open squares), glucose (magenta color open circles) and creatinine (red color open stars). (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
Since most of the cladding layer has been etched and the graphene oxide was coated on the eFBG, Δneff depends on the values of nclad and Δnclad . From Fig. 4a, for aCRP–GO coated eFBG sensor the shift in Bragg wavelength is 20.15 pm for CRP concentration of 1 mg/L. The corresponding change in the clad refractive index (Δnclad ) is 2 10 5, calculated using Eq. (1). Note that for GO coated eFBG, nclad = nGO = 1.7. For the same value of Δnclad ( ¼2 10 5), the calculated Δneff (using Eq. (1)), for the eFBG sensor without GO coating is 5 10 6 and the corresponding ΔλB is ∼5.3 pm. This shows, for the same value of Δnclad achieved after CRP attachment to aCRP, Δneff is higher for eFBG sensor with GO coating and hence GO coated eFBG is a better platform for sensing. Further, CRP detection experiments have been carried out in the presence of different interfering compounds such as urea, glucose and creatinine. For these experiments, first the interfering limits of urea (2000 mg/L), glucose (4000 mg/L) and creatinine (6000 mg/L) have been confirmed by performing the detection experiments with aCRP–GO complex coated eFBG sensors, using different concentrations of each interfering compound separately. The inset in Fig. 4a shows the ΔλB values for different concentrations of aqueous solutions of urea, glucose and creatinine. It is observed that ΔλB value for aCRP–GO complex coated eFBG sensors
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
Fig. 5. Normalized shift of Bragg wavelength as a function of CRP concentration (i) aCRP (purple color open circles) and (ii) aCRP–GO complex (olive color green squares) coated eFBG sensors in the absence of interfering factors. The solid lines represent the Langmuir adsorption isotherm. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
very high concentration of creatinine (5800 mg/L) as compared to other interfering factors glucose (3800 mg/L) and urea (1800 mg/ L), in order to check the interference effect, a slightly higher error bar is observed in the measurements of ΔλB values for creatinine. The inset of Fig. 4b shows the difference in ΔλB plotted as a function of concentration of CRP in the presence of urea, glucose and creatinine. It is observed that the ΔλB values for CRP in the presence of interfering compounds varies from 0.2 pm to 3 pm with respect to the ΔλB values for CRP in the absence of interfering compounds. This difference in ΔλB values is close to the variations in measuring the base Bragg wavelength value with our instrument. The observed negligible difference in ΔλB for CRP in the presence of urea, glucose and creatinine shows the specificity of the aCRP–GO complex coated eFBG sensors to CRP. Fig. 5 shows the normalized shift in Bragg wavelength as a function of CRP concentration in the absence of interfering factors and the data follows the Langmuir type adsorption isotherm given by (Sridevi et al., 2014)
ΔλB λB0
is less than 2 pm even after addition of 2000 mg/L of urea, 4000 mg/L of glucose and 6000 mg/L of creatinine. From the knowledge of above data, 1800 mg/L of urea has been added to the CRP samples of different concentrations varying from 10 3 mg/L to 100 mg/L for CRP detection experiments. It is observed that the presence of urea does not affect the sensitivity (as shown in Fig. 4b) of the present aCRP–GO complex coated eFBG sensors in detecting the CRP. Likewise, similar experiments have been performed by adding 3800 mg/L of glucose and 5800 mg/L of creatinine to various concentrations of CRP solutions. It is found that the aCRP–GO complex coated eFBG sensors show high specificity to CRP even in the presence of these interfering compounds. Fig. 4b shows ΔλB as a function of different concentrations of CRP in the presence of urea, glucose and creatinine. From Fig. 4b, it can be seen that the sensitivity of the present eFBG devices for CRP detection in the presence of any interfering compound is same as (with in error limits) that for CRP detection in the absence of interfering compound (we have performed these experiments three times for each interfering compound and calculated the average ΔλB values). Further, the difference in ΔλB , defined as ΔλB(CRP) − ΔλB(CRP + (urea/ glucose/creatinine)) , has been measured for different concentrations of CRP in the presence of interfering compounds (urea, glucose and creatinine). Since we have used
255
= S[log C + log 7.389KA]
(2)
where S is structural factor is given by 0.58kCmax ; Cmax is the maximum surface density of aCRP and k is a constant specifying the sensor characteristics, C is the concentration of CRP expressed in M (for CRP, the concentration of 1 mg/L ¼9.524 nM) and KA is the affinity constant. The normalized shift in Bragg wavelength as a function of CRP concentration in Fig. 5 is fitted to Eq. (2) and the extracted affinity constant KA is 1.1 1010 M 1 for eFBG sensors coated with aCRP–GO complex and 5.2 109 M 1 for eFBG sensors coated with only aCRP molecules.
4. Conclusions The limit of detection of 0.01 mg/L has been achieved in our GO coated eFBG sensors. Even in the presence of various interfering factors, aCRP–GO complex coated eFBG sensors have shown high specificity to CRP. The limit of detection and linearity range in our present methodology are compared with other methods reported in the literature in Table 1. It can be seen that the eFBG sensors coated with GO flakes functionalized with aCRP are highly sensitive optical devices for CRP detection and compare extremely well with other methods. Further, the present experiments show that eFBG coated with graphene oxide can be a very sensitive platform technology for many biosensors.
Table 1 Limit of detection and the linear range of detection of CRP in various experimental methods. Experimental method
Experimental read out
Limit of detection
FBG
Bragg wavelength
SPR
SPR angle shift
10 μg/L 2 mg/L 1 mg/L 0.130 μg/L
Linear range of detection
Reference
0.01–100 mg/L
Present work
2-5 mg/L 1–26 mg/L 0.130 μg/L–5.016 mg/L –
Meyer et al. (2006) Hu et al. (2006) Kim et al. (2009)
0.01–0.2 mg/L 0.3–10 mg/L 0.1–50 mg/L
Gan et al. (2012) Gan et al. (2012) Centi et al. (2009)
QCM
Resonant frequency
Piezoelectric Microcantelivers
Resistance
ELISA Electrochemistry
Resonant frequency Absorption intensity Current
10 μg/L
0.01–1 mg/L
Buch and Rishpon (2008)
EIS
Charge transfer resistance
18 μg/L
0.052–2 mg/L
Bryan et al. (2013)
MPTV
Brownian velocity
–
Fan et al. (2010)
Laser induced fluorescence Chemiluminescence
Fluorescence Intensity Relative chemiluminescence emission intensity
100 μg/L 9.2 mg/L 200 pg/L
10–150 mg/L
Lin et al. (2013) Islam and Kang (2011)
10 μg/L 3 ng/L 0.1 mg/L 54 μg/L
0.93 μg/L
2 × 10−4–0.2 μg/L –
Wee et al. (2005)
Lee et al. (2012)
256
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
Acknowledgments Professor A.K. Sood thank the Department of Science and Technology for support. Professor S. Asokan thank the Robert Bosch Centre for Cyber Physical Systems, for partial support.
References Bryan, T., Luo, X., Bueno, P.R., Davis, J.J., 2013. Biosens. Bioelectron. 39, 94–98. Buch, M., Rishpon, J., 2008. Electroanalysis 20, 2592–2594. Cao, Y.-C., Huang, Z.-L., Liu, T.-C., Wang, H.-Q., Zhu, X.-X., Wang, Z., Zhao, Y.-D., Liu, M.-X., Luo, Q.-M., 2006. Anal. Biochem. 351, 193–200. Centi, S., Sanmartin, L.B., Tombelli, S., Palchetti, I., Mascini, M., 2009. Electroanalysis 21, 1309–1315. Christodoulides, N., Mohanty, S., Miller, C.S., Langub, M.C., Floriano, P.N., Dharshan, P., Ali, M.F., Bernard, B., Romanovicz, D., Anslyn, E., Fox, P.C., McDevitt, J.T., 2005. Lab. Chip 51, 2391–2395. Dajani, L.K., Paus, E., Warren, D.J., 2001. Clin. Chem. 47, 867–873. Fan, Y.-J., Sheen, H.-J., Liu, Y.-H., Tsai, J.-F., Wu, T.-H., Wu, K.-C., Lin, S., 2010. Langmuir 26, 13751–13754. Gabay, C., Kushner, I., 1999. New Engl. J. Med. 340, 448–454. Gan, N., Wang, L., Zhou, H., Li, T., Sang, W., Hu, F., Cao, Y., 2012. Int. J. Electrochem. Sci. 7, 11564–11577. Guo, T., Liu, F., Liu, Y., Chen, N.-K., Guan, B.-O., Albert, J., 2014. Biosens. Bioelectron. 55, 452–458.
Gupta, R.K., Periyakaruppan, A., Meyyappan, M., Koehne, J.E., 2014. Biosens. Bioelectron. 59, 112–119. Hu, W.P., Hsu, H.Y., Chiou, A., Tseng, K.Y., Lin, H.Y., Chang, G.L., Chen, S.J., 2006. Biosens. Bioelectron. 21, 1631–1637. Hu, W., He, G., Chen, T., Guo, C.X., Lu, Z., Selvaraj, J.N., Liu, Y., Li, C.M., 2014. Chem. Commun. 50, 2133–2135. Islam, Md.S., Kang, S.H., 2011. Talanta 84, 752–758. Kashyap, R., 1999. Fiber Bragg Gratings. Academic Press, San Diego. Kim, N., Kim, D.K., Cho, Y.J., 2009. Sens. Actuators B 143, 444–448. Lee, J.S., Joung, H.-A., Kim, M.-G., Park, C.B., 2012. ACS Nano 6, 2978–2983. Lee, W.B., Chen, Y.H., Lin, H.I., Shiesh, S.C., Lee, G.B., 2011. Sens. Actuators B 157, 710–721. Li, J.-J., Fang, C.-H., 2004. Med. Hypotheses 62, 499–506. Lin, Y.-J., Yang, J.-Y., Shu, T.-Y., Lin, T.-Y., Chen, Y.-Y., Su, M.-Y., Li, W.-J., Liu, M.-Y., 2013. J. Chromatogr. A 1315, 188–194. Lu, Q., Li, C.M., 2008. Biosens. Bioelectron. 24, 767. Marcano, D.C., Kosynkin, D.V., Berlin, J.M., Sinitskii, A., Sun, Z., Slesarev, A., Alemany, L.B., Lu, W., Tour, J.M., 2010. ACS Nano 4, 4806. May, A., Wang, T.J., 2007. Expert Rev. Mol. Diagn. 7, 793–804. Meyer, M.H.F., Hartmann, M., Keusgen, M., 2006. Biosens. Bioelectron. 21, 1987–1990. Mold, C., Gewurz, H., Du Clos, T.W., 1999. Immunopharmacology 42, 23–30. Sridevi, S., Vasu, K.S., Jayaraman, N., Asokan, S., Sood, A.K., 2014. Sens. Actuators B 195, 150–155. Tripathi, S.M., Bock, W.J., Mikulic, P., Chinnappan, R., Ng, A., Tolba, M., Zourob, M., 2012. Biosens. Bioelectron. 35, 308–312. Wee, K.W., Kang, G.Y., Park, J., Kang, J.Y., Yoon, D.S., Park, J.H., Kim, T.S., 2005. Biosens. Bioelectron. 20, 1932–1938.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Sensitive detection of C-reactive protein using optical fiber Bragg gratings Sridevi S.a, K.S. Vasu b, S. Asokan a,c, A.K. Sood b,n a
Department of Instrumentation and Applied Physics, Indian Institute of Science, Bangalore 560012, India Department of Physics, Indian Institute of Science, Bangalore 560012, India c Robert Bosch Centre for Cyber Physical Systems, Indian Institute of Science, Bangalore 560012, India b
art ic l e i nf o
a b s t r a c t
Article history: Received 17 July 2014 Received in revised form 28 September 2014 Accepted 15 October 2014 Available online 23 October 2014
An accurate and highly sensitive sensor platform has been demonstrated for the detection of C-reactive protein (CRP) using optical fiber Bragg gratings (FBGs). The CRP detection has been carried out by monitoring the shift in Bragg wavelength (ΔλB ) of an etched FBG (eFBG) coated with an anti-CRP antibody (aCRP)-graphene oxide (GO) complex. The complex is characterized by Fourier transform infrared spectroscopy, X-ray photoelectron spectroscopy and atomic force microscopy. A limit of detection of 0.01 mg/L has been achieved with a linear range of detection from 0.01 mg/L to 100 mg/L which includes clinical range of CRP. The eFBG sensor coated with only aCRP (without GO) show much less sensitivity than that of aCRP–GO complex coated eFBG. The eFBG sensors show high specificity to CRP even in the presence of other interfering factors such as urea, creatinine and glucose. The affinity constant of ∼ 1.1 × 1010 M−1 has been extracted from the data of normalized shift (ΔλB /λB ) as a function of CRP concentration. & Elsevier B.V. All rights reserved.
Keywords: C-reactive protein Fiber Bragg gratings Graphene oxide Bragg wavelength
1. Introduction C-reactive protein (CRP) is an acute phase protein, found in blood plasma and its concentration increases up to three orders during the inflammatory process occurring in the human body (Gabay and Kushner, 1999). CRP concentration below 1 mg/L represents low risk, between 1 and 3 mg/L represents average risk and above 3 mg/L represents the high risk of cardiovascular diseases (Lee et al., 2011). Half-life of plasma CRP is approximately 19 h (Mold et al., 1999) and the CRP level decreases rapidly after recovery or treatment. Thus, CRP has emerged as an effective biomarker for cardiovascular diseases, infections in human body and can be used as prototypical inflammatory predictor (Li and Fang, 2004; May and Wang, 2007). The structure of human CRP is described as ‘pentraxin’ in which five identical homologous subunits are non-covalently associated around a central pore. CRP binds to the phosphocholine present on microbes, dead and damaged cells to initiate the removal process. The commonly used CRP detection techniques are radial immunodiffusion (RID), radio immunoassay (RIA), immunonephelometry (IN), immunoturbidimetry (IT), immunofluorescence, immuno chemiluminescence and standard enzyme immunoassay (as n
Corresponding author. Tel.: +91 80 2293 2964; fax: +91 80 2360 2602. E-mail address: [email protected] (A.K. Sood).
http://dx.doi.org/10.1016/j.bios.2014.10.033 0956-5663/& Elsevier B.V. All rights reserved.
ELISA) and the detection limit in these techniques ranges between 0.1 and 0.2 mg/L (Centi et al., 2009). The preparation of monoclonal and polyclonal antibodies used in the above-mentioned methods varies depending on the immune response in different animals and causes the variable specificity and sensitivity of CRP detection (Dajani et al., 2001). The photometric methods like ELISA can be affected by the colour and composition of the medium used in assays (Meyer et al., 2006). Many other methods such as surface plasmon resonance (SPR) (Meyer et al., 2006; Hu et al., 2006), quartz crystal microbalance (QCM) (Kim et al., 2009), piezoelectric micro-cantilever method (Wee et al., 2005; Gan et al., 2012), electrochemistry (Centi et al., 2009; Buch and Rishpon, 2008), electrochemical impedance spectroscopy (EIS) (Bryan et al., 2013) and micro particle-tracking velocimetry (MPTV) (Fan et al., 2010) have been developed for CRP detection. Laser induced fluorescence (Lin et al., 2013) and chemiluminescence (Islam and Kang, 2011) methods have also used for the detection of CRP. Christodoulides et al. (2005) have reported the CRP detection using human saliva rather than the blood serum in ELISA tests. Further, for the detection of CRP, carbon nanomaterials such as carbon nanofibers, graphene oxide (GO) and reduced graphene oxide (RGO) have also been used in EIS and chemiluminescence methods (Gupta et al., 2014; Lee et al., 2012). Fiber Bragg gratings are efficient optical devices in the field of biosensors (Tripathi et al., 2012; Guo et al., 2014), with the Bragg
252
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
wavelength (λB ) and the transmission intensity as probing parameters. Sensitivity of FBG sensors can be enhanced by etching the cladding layer and exposing the core directly to the surrounding medium. Recently, it has been shown that the GO coated etched fiber Bragg gratings (eFBG) can be used for the detection of carbohydrate-protein interactions (Sridevi et al., 2014). In the present study, an optical platform has been demonstrated for accurate detection of CRP using GO coated eFBG sensors. GO flakes are attached with anti-CRP antibodies (aCRP) through the formation of covalent functionalization before coating onto the eFBG sensors. The present eFBG sensors have shown a high specificity to CRP even in the presence of other interfering factors such as urea, creatinine and glucose, with the limit of detection of 0.01 mg/L. Further, the sensors have shown excellent performance in the linear range of detection from 0.05 mg/L to 100 mg/L.
2. Materials and experimental methods 2.1. Materials CRP detection experiments have been carried out with the Quantia CRP-UV kit, which is generally used in turbidimetric immunoassay for CRP detection, purchased from M/S Tulip Diagnostics (P), Ltd. Basically, the kit contains ready to use solution of aCRP of IgG class which are highly specific to CRP, a calibrator which contains the human CRP in lyophilized serum (traceable to W.H.O. international reference standard (85/506) for CRP). The Quantia CRP-UV kit has been designed to measure CRP concentrations in the range of 6–100 mg/L with a detection limit of 3 mg/L by using the UV–vis absorption spectroscopy as probing method. No prozone effect has been observed with CRP concentration upto 1 g/L. N-(3-dimethylaminopropyl)-N-ethylcarbodiimde hydrochloride (EDC), N- hydroxysuccinimide (NHS) and the interfering factors such as urea, creatinine and glucose, of analytical grade, purchased from Sigma-Aldrich Chemical Co. are used without further purification. Graphite powder, purchased from Superior Graphite.Co. (Riverside, Chicago) for graphite oxide preparation. 2.2. Preparation of graphene oxide The graphite oxide (GO) is prepared (Marcano et al., 2010) by stirring 2 g of graphite flakes in acid (H2SO4 : H3PO4 ¼110 mL : 12 mL) followed by very slow addition of KMnO4 (11.1 g) and this one-pot mixture is stirred for 3 days at room temperature. After 3 days, 20 mL of H2O2 is added to the mixture to terminate the oxidation process. The resultant material is then washed repeatedly with 1M HCl and distilled water and then dried under vacuum to obtain graphite oxide. 100 μg/mL GO aqueous dispersion has been prepared by sonicating 1 mg of graphite oxide in 10 mL of deionized (DI) water.
2.4. Fabrication of FBG and preparation of eFBG Phase-mask technique is used to inscribe the gratings in germania doped core of a single mode photosensitive silicate fiber (M/S Nufern) of diameter 125 μm . KrF excimer UV laser of wavelength 248 nm (pulse energy 6 mJ and repetition rate of 200 Hz) and a phase mask of 1069 nm pitch (M/s Stocker Yale, Inc.) are used for the grating inscription. 40% HF solution is used for etching the cladding layer around the grating region (∼3 mm ) of the FBG followed by washing with DI water. In all our experiments the Bragg wavelength (λB) is monitored using an optical interrogator (Micron Optics, SM130) with a wavelength repeatability of 1 pm. 2.5. Functionalization of the eFBG The etched portion of eFBG is subsequently treated with H2O:H2O2:NH4OH (5:1:1) solution, which is followed by washing with DI water to make the eFBG surface hydrophilic in nature. The etched portion of eFBG is dipped into 200 μL of aqueous dispersion of aCRP–GO complex and allowed to dry at room temperature. After drying the aCRP–GO complex coated eFBG is thoroughly washed for 15 min to remove unbound particles. In all the CRP detection experiments, the eFBG sensors coated with aCRP–GO complex were incubated in bovine serum albumin (BSA, 10 μg/ml) for 15 min to block the unoccupied area by the aCRP molecules in aCRP–GO complex. This was followed by washing away the unattached BSA by DI water. The Bragg wavelength value corresponding to the aCRP–GO complex coated eFBG after dipping in BSA is taken as reference value (λB0) for the CRP detection experiments. CRP aqueous solutions of various concentrations ranging from 10−3 mg/L to 100 mg/L are prepared for the experiments. For the experiments with interfering factors, different concentrations of aqueous solutions of urea, glucose and creatinine ranging from 1 mg/L to 10 g/L are prepared. 2.6. Experimental setup Teflon block patterned with micro-fluidic channels of volume 200 μL is used to dip the eFBG region coated with aCRP–GO complex in different concentrations of CRP solution of 200 μL . After 10 min dip, the eFBG sensor is taken out to measure the Bragg wavelength and the same procedure was followed at each concentration of CRP solution varying from 1 μg/L to 100 mg/L. All our measurements throughout the experiment were carried out
2.3. Preparation of aCRP–GO complex The preparation of aCRP–GO complex is carried out by stirring 1 mL of GO aqueous dispersion for 30 min at 25 °C after addition of 2 mg of EDC and 4 mg of NHS. Subsequently, 200 μL of aCRP solution is added and this mixture is kept in the orbital shaker for 10 h to prepare aCRP–GO complex. For the Fourier transform infrared (FTIR) spectroscopy and X-ray photoelectron spectroscopy (XPS) experiments, the aqueous dispersions of aCRP–GO complex is filtered using anodisc filter to form a thin film. The atomic force microscopy (AFM) experiments have been carried out by imaging the same GO flake deposited on the 300 nm SiO2/Si before and after the functionalization with aCRP.
Fig. 1. FTIR spectra of GO and aCRP–GO complex. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
without removing the eFBG sensors from the clamp to avoid bending effect in measuring the Bragg wavelength. For one complete set of experiments with different concentrations of CRP, the
253
same FBG sensor is used and disposed. The concentrations plotted in Figs. 4a and b and 5 are the cumulative concentrations.
3. Results and discussion 3.1. Characterization of aCRP–GO complex
Fig. 2. XPS spectra of GO and aCRP–GO complex. The spectral component shown by dotted line corresponds to C–N bond.
a
c
b
d
EDC is the most commonly used coupling reagent for the amide bond formation between –COOH and –NH2 groups (Cao et al., 2006). The addition of EDC to the GO aqueous dispersion activates the – COOH groups present on GO flakes. The presence of NHS increases the coupling efficiency of EDC (Cao et al., 2006). The EDC-NHS activated –COOH groups at the edges of GO flakes bind with –NH2 groups in the aCRP to form an amide bond by releasing the water molecules. Fig. 1 shows the FTIR spectra of GO and aCRP–GO complex. The pristine GO sample (red colour solid line) shows characteristic bands at 1060, 1229 and 1405 cm 1 (C–O vibrations in epoxy bonds), 1630 cm 1 (CQC vibrations), 1737 cm 1 (CQO vibrations in carboxylic group) and 3415 cm 1 (O–H vibrations) (Marcano et al., 2010; Hu et al., 2014). After the functionalization
Fig. 3. AFM image of the GO flake (a) before and (b) after covalent functionalization with aCRP. (c) Height profile of before (black color solid line) and after (red color solid line) covalent functionalization. (d) SEM image of aCRP–GO complex coated eFBG sensor after CRP detection experiments. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
254
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
with aCRP, the band at 1737 cm 1 (corresponds to CQO vibrations in carboxylic groups) disappears and new bands appear at 1163 and 1238 cm 1 (corresponds to C–N vibration), 1395 and 1446 cm 1 (corresponds to N–H vibrations), 1535 cm 1 (corresponds to amideII mode, an out-of-phase combination of the N–H inplane bend and the C–N stretching vibration) and 3269 cm 1 (corresponds to N–H vibrations) confirms the functionalization of GO flakes with aCRP through amide bond formation (Hu et al., 2014; Lu and Li, 2008). Further, the XPS experiments were carried out to confirm the complex formation. Fig. 2 shows the C1s core level spectrum of the XPS spectra of GO before (black color solid spheres) and after (black color open circles) the functionalization with aCRP. Before the functionalization GO shows typical C1s spectrum and is deconvoluted into 5 bands centered at 284.1 (CQC), 284.9 (C–C and C–H), 286.5 (C– OH and C–O–C), 287.1 (CQO) and 288.4 eV (O ¼ C–O). In the deconvoluted C1s spectrum of aCRP–GO complex a peak appears at 285.6 (shown by dotted line in Fig. 2) corresponding to C–N apart from the conventional GO peaks at 284.1 (CQC), 284.6 (C–C and C–H), 286.8 (C–OH and C–O–C), corroborating amide bond formation between the –COOH groups in GO flakes and the –NH2 groups in aCRP. Fig. 3a and b shows the tapping mode AFM images of GO flake deposited on SiO2/Si substrate before and after functionalization of aCRP. From the height profile (shown in Fig. 3c), the average height of the GO flake before functionalization is 1.5 nm and after the functionalization it changes to 4.5 nm with increased roughness on the GO flakes. The change in average height and appearance of brighter spots on the GO flake after treated with aCRP confirm the functionalization. Fig. 3d shows the scanning electron microscope image of aCRP–GO complex coated eFBG. 3.2. Detection of CRP Fig. 4a shows the shift in Bragg wavelength (ΔλB ) as a function of concentration of CRP for (i) only aCRP and (ii) aCRP–GO complex coated eFBG sensors. The observed shift in Bragg wavelength for aCRP–GO complex coated eFBG sensors (∼6.3 pm per one order increase in the CRP concentration) is ∼5 times more than that of eFBG sensors coated with only aCRP (1.2 pm per one order increase in the CRP concentration). Also, the limit of detection in aCRP–GO complex coated eFBG sensors (0.01 mg/L) shows that the presence of GO enhances the sensitivity when compared to the eFBG sensors coated with only aCRP molecules (the limit of detection is 1 mg/L). This is due to two reasons: (i) immobilization of antibodies (aCRP molecules) on the surface of eFBG sensors through binding aCRP molecules to GO via amide bond formation (facilitated by EDC-NHS) and the efficacy of GO in binding to the hydrophilic surface of eFBG are main reasons for higher sensitivity in the eFBG sensors coated with aCRP–GO complex. As mentioned before, GO binds strongly to the hydrophilic surface of eFBG through hydrogen bonding between the –OH groups present on the hydrophilic etched FBG surface and the –COOH, –OH, –O– groups on GO. In direct coating of aCRP molecules on eFBG sensor, the bonding between aCRP molecules and the surface of eFBG sensor is physical and weak when compared to the binding of aCRP–GO complex onto the surface of eFBG sensor. (ii) The change in Bragg wavelength is directly proportional to the change in effective refractive index (Δneff ), i.e. ΔλB = 2Δneff Λ, where Λ is grating period and neff is a function of ncore and nclad given by (Kashyap, 1999; Sridevi et al., 2014)
⎤ ⎡⎛ n 2 − n 2 ⎞⎛ ⎞ eff clad ⎟⎜ ncore − nclad ⎟ ⎥ neff = nclad ⎢⎜⎜ 2 + 1 ⎟ ⎜ ⎟ 2 ⎥⎦ ⎢⎣⎝ ncore − neff nclad ⎠⎝ ⎠
(1)
Fig. 4. (a) The shift in Bragg wavelength as a function of CRP concentration in (i) aCRP (purple color open circles) and (ii) aCRP–GO complex (olive color green squares) coated eFBG sensors. Inset: Shift in Bragg wavelength as a function of concentration of only interfering factors such as urea (blue color open squares), glucose (magenta color open circles) and creatinine (red color open stars). (b) The shift in Bragg wavelength as a function of CRP concentration in the presence of urea (1800 mg/L) – blue color open squares; glucose (3800 mg/L) – magenta color open circles and creatinine (5800 mg/L) – red color open stars. The olive color solid spheres represents the data for shift in Bragg wavelength as a function of CRP concentration in the absence of interfering factors. Inset: difference in ΔλB for different concentrations of CRP as a function of CRP concentration in the presence of urea (blue color open squares), glucose (magenta color open circles) and creatinine (red color open stars). (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
Since most of the cladding layer has been etched and the graphene oxide was coated on the eFBG, Δneff depends on the values of nclad and Δnclad . From Fig. 4a, for aCRP–GO coated eFBG sensor the shift in Bragg wavelength is 20.15 pm for CRP concentration of 1 mg/L. The corresponding change in the clad refractive index (Δnclad ) is 2 10 5, calculated using Eq. (1). Note that for GO coated eFBG, nclad = nGO = 1.7. For the same value of Δnclad ( ¼2 10 5), the calculated Δneff (using Eq. (1)), for the eFBG sensor without GO coating is 5 10 6 and the corresponding ΔλB is ∼5.3 pm. This shows, for the same value of Δnclad achieved after CRP attachment to aCRP, Δneff is higher for eFBG sensor with GO coating and hence GO coated eFBG is a better platform for sensing. Further, CRP detection experiments have been carried out in the presence of different interfering compounds such as urea, glucose and creatinine. For these experiments, first the interfering limits of urea (2000 mg/L), glucose (4000 mg/L) and creatinine (6000 mg/L) have been confirmed by performing the detection experiments with aCRP–GO complex coated eFBG sensors, using different concentrations of each interfering compound separately. The inset in Fig. 4a shows the ΔλB values for different concentrations of aqueous solutions of urea, glucose and creatinine. It is observed that ΔλB value for aCRP–GO complex coated eFBG sensors
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
Fig. 5. Normalized shift of Bragg wavelength as a function of CRP concentration (i) aCRP (purple color open circles) and (ii) aCRP–GO complex (olive color green squares) coated eFBG sensors in the absence of interfering factors. The solid lines represent the Langmuir adsorption isotherm. (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this article.)
very high concentration of creatinine (5800 mg/L) as compared to other interfering factors glucose (3800 mg/L) and urea (1800 mg/ L), in order to check the interference effect, a slightly higher error bar is observed in the measurements of ΔλB values for creatinine. The inset of Fig. 4b shows the difference in ΔλB plotted as a function of concentration of CRP in the presence of urea, glucose and creatinine. It is observed that the ΔλB values for CRP in the presence of interfering compounds varies from 0.2 pm to 3 pm with respect to the ΔλB values for CRP in the absence of interfering compounds. This difference in ΔλB values is close to the variations in measuring the base Bragg wavelength value with our instrument. The observed negligible difference in ΔλB for CRP in the presence of urea, glucose and creatinine shows the specificity of the aCRP–GO complex coated eFBG sensors to CRP. Fig. 5 shows the normalized shift in Bragg wavelength as a function of CRP concentration in the absence of interfering factors and the data follows the Langmuir type adsorption isotherm given by (Sridevi et al., 2014)
ΔλB λB0
is less than 2 pm even after addition of 2000 mg/L of urea, 4000 mg/L of glucose and 6000 mg/L of creatinine. From the knowledge of above data, 1800 mg/L of urea has been added to the CRP samples of different concentrations varying from 10 3 mg/L to 100 mg/L for CRP detection experiments. It is observed that the presence of urea does not affect the sensitivity (as shown in Fig. 4b) of the present aCRP–GO complex coated eFBG sensors in detecting the CRP. Likewise, similar experiments have been performed by adding 3800 mg/L of glucose and 5800 mg/L of creatinine to various concentrations of CRP solutions. It is found that the aCRP–GO complex coated eFBG sensors show high specificity to CRP even in the presence of these interfering compounds. Fig. 4b shows ΔλB as a function of different concentrations of CRP in the presence of urea, glucose and creatinine. From Fig. 4b, it can be seen that the sensitivity of the present eFBG devices for CRP detection in the presence of any interfering compound is same as (with in error limits) that for CRP detection in the absence of interfering compound (we have performed these experiments three times for each interfering compound and calculated the average ΔλB values). Further, the difference in ΔλB , defined as ΔλB(CRP) − ΔλB(CRP + (urea/ glucose/creatinine)) , has been measured for different concentrations of CRP in the presence of interfering compounds (urea, glucose and creatinine). Since we have used
255
= S[log C + log 7.389KA]
(2)
where S is structural factor is given by 0.58kCmax ; Cmax is the maximum surface density of aCRP and k is a constant specifying the sensor characteristics, C is the concentration of CRP expressed in M (for CRP, the concentration of 1 mg/L ¼9.524 nM) and KA is the affinity constant. The normalized shift in Bragg wavelength as a function of CRP concentration in Fig. 5 is fitted to Eq. (2) and the extracted affinity constant KA is 1.1 1010 M 1 for eFBG sensors coated with aCRP–GO complex and 5.2 109 M 1 for eFBG sensors coated with only aCRP molecules.
4. Conclusions The limit of detection of 0.01 mg/L has been achieved in our GO coated eFBG sensors. Even in the presence of various interfering factors, aCRP–GO complex coated eFBG sensors have shown high specificity to CRP. The limit of detection and linearity range in our present methodology are compared with other methods reported in the literature in Table 1. It can be seen that the eFBG sensors coated with GO flakes functionalized with aCRP are highly sensitive optical devices for CRP detection and compare extremely well with other methods. Further, the present experiments show that eFBG coated with graphene oxide can be a very sensitive platform technology for many biosensors.
Table 1 Limit of detection and the linear range of detection of CRP in various experimental methods. Experimental method
Experimental read out
Limit of detection
FBG
Bragg wavelength
SPR
SPR angle shift
10 μg/L 2 mg/L 1 mg/L 0.130 μg/L
Linear range of detection
Reference
0.01–100 mg/L
Present work
2-5 mg/L 1–26 mg/L 0.130 μg/L–5.016 mg/L –
Meyer et al. (2006) Hu et al. (2006) Kim et al. (2009)
0.01–0.2 mg/L 0.3–10 mg/L 0.1–50 mg/L
Gan et al. (2012) Gan et al. (2012) Centi et al. (2009)
QCM
Resonant frequency
Piezoelectric Microcantelivers
Resistance
ELISA Electrochemistry
Resonant frequency Absorption intensity Current
10 μg/L
0.01–1 mg/L
Buch and Rishpon (2008)
EIS
Charge transfer resistance
18 μg/L
0.052–2 mg/L
Bryan et al. (2013)
MPTV
Brownian velocity
–
Fan et al. (2010)
Laser induced fluorescence Chemiluminescence
Fluorescence Intensity Relative chemiluminescence emission intensity
100 μg/L 9.2 mg/L 200 pg/L
10–150 mg/L
Lin et al. (2013) Islam and Kang (2011)
10 μg/L 3 ng/L 0.1 mg/L 54 μg/L
0.93 μg/L
2 × 10−4–0.2 μg/L –
Wee et al. (2005)
Lee et al. (2012)
256
S. Sridevi et al. / Biosensors and Bioelectronics 65 (2015) 251–256
Acknowledgments Professor A.K. Sood thank the Department of Science and Technology for support. Professor S. Asokan thank the Robert Bosch Centre for Cyber Physical Systems, for partial support.
References Bryan, T., Luo, X., Bueno, P.R., Davis, J.J., 2013. Biosens. Bioelectron. 39, 94–98. Buch, M., Rishpon, J., 2008. Electroanalysis 20, 2592–2594. Cao, Y.-C., Huang, Z.-L., Liu, T.-C., Wang, H.-Q., Zhu, X.-X., Wang, Z., Zhao, Y.-D., Liu, M.-X., Luo, Q.-M., 2006. Anal. Biochem. 351, 193–200. Centi, S., Sanmartin, L.B., Tombelli, S., Palchetti, I., Mascini, M., 2009. Electroanalysis 21, 1309–1315. Christodoulides, N., Mohanty, S., Miller, C.S., Langub, M.C., Floriano, P.N., Dharshan, P., Ali, M.F., Bernard, B., Romanovicz, D., Anslyn, E., Fox, P.C., McDevitt, J.T., 2005. Lab. Chip 51, 2391–2395. Dajani, L.K., Paus, E., Warren, D.J., 2001. Clin. Chem. 47, 867–873. Fan, Y.-J., Sheen, H.-J., Liu, Y.-H., Tsai, J.-F., Wu, T.-H., Wu, K.-C., Lin, S., 2010. Langmuir 26, 13751–13754. Gabay, C., Kushner, I., 1999. New Engl. J. Med. 340, 448–454. Gan, N., Wang, L., Zhou, H., Li, T., Sang, W., Hu, F., Cao, Y., 2012. Int. J. Electrochem. Sci. 7, 11564–11577. Guo, T., Liu, F., Liu, Y., Chen, N.-K., Guan, B.-O., Albert, J., 2014. Biosens. Bioelectron. 55, 452–458.
Gupta, R.K., Periyakaruppan, A., Meyyappan, M., Koehne, J.E., 2014. Biosens. Bioelectron. 59, 112–119. Hu, W.P., Hsu, H.Y., Chiou, A., Tseng, K.Y., Lin, H.Y., Chang, G.L., Chen, S.J., 2006. Biosens. Bioelectron. 21, 1631–1637. Hu, W., He, G., Chen, T., Guo, C.X., Lu, Z., Selvaraj, J.N., Liu, Y., Li, C.M., 2014. Chem. Commun. 50, 2133–2135. Islam, Md.S., Kang, S.H., 2011. Talanta 84, 752–758. Kashyap, R., 1999. Fiber Bragg Gratings. Academic Press, San Diego. Kim, N., Kim, D.K., Cho, Y.J., 2009. Sens. Actuators B 143, 444–448. Lee, J.S., Joung, H.-A., Kim, M.-G., Park, C.B., 2012. ACS Nano 6, 2978–2983. Lee, W.B., Chen, Y.H., Lin, H.I., Shiesh, S.C., Lee, G.B., 2011. Sens. Actuators B 157, 710–721. Li, J.-J., Fang, C.-H., 2004. Med. Hypotheses 62, 499–506. Lin, Y.-J., Yang, J.-Y., Shu, T.-Y., Lin, T.-Y., Chen, Y.-Y., Su, M.-Y., Li, W.-J., Liu, M.-Y., 2013. J. Chromatogr. A 1315, 188–194. Lu, Q., Li, C.M., 2008. Biosens. Bioelectron. 24, 767. Marcano, D.C., Kosynkin, D.V., Berlin, J.M., Sinitskii, A., Sun, Z., Slesarev, A., Alemany, L.B., Lu, W., Tour, J.M., 2010. ACS Nano 4, 4806. May, A., Wang, T.J., 2007. Expert Rev. Mol. Diagn. 7, 793–804. Meyer, M.H.F., Hartmann, M., Keusgen, M., 2006. Biosens. Bioelectron. 21, 1987–1990. Mold, C., Gewurz, H., Du Clos, T.W., 1999. Immunopharmacology 42, 23–30. Sridevi, S., Vasu, K.S., Jayaraman, N., Asokan, S., Sood, A.K., 2014. Sens. Actuators B 195, 150–155. Tripathi, S.M., Bock, W.J., Mikulic, P., Chinnappan, R., Ng, A., Tolba, M., Zourob, M., 2012. Biosens. Bioelectron. 35, 308–312. Wee, K.W., Kang, G.Y., Park, J., Kang, J.Y., Yoon, D.S., Park, J.H., Kim, T.S., 2005. Biosens. Bioelectron. 20, 1932–1938.