We demonstrated a simple method for self-reference and label free biosensing based on a capillary sensing element and common optoelectronic devices. The capillary sensing element is illuminated by a light-emitting diode (LED) light source and detected by a webcam. Part of gold film that deposited on the tubing wall is functionalized to carry on the biological information in the excited SPR modes. The end face of the capillary was monitored and separate regions of interest (ROIs) were selected as the measurement channel and the reference channel. In the ROIs, the biological information can be accurately extracted from the image by simple image processing. Moreover, temperature fluctuation, bulk RI fluctuation, light source fluctuation and other factors can be effectively compensated during detection. Our biosensing device has a sensitivity of 1145%/RIU and a resolution better than 5.287 × 10−4 RIU, considering a 0.79% noise level. We apply it for concanavalin A (Con A) biological measurement, which has an approximately linear response to the specific analyte concentration. This simple method provides a new approach for multichannel SPR sensing and reference-compensated calibration of SPR signal for label-free detection.
© 2017 Optical Society of America
Because surface plasmon resonance (SPR) sensors can perceive small refractive index (RI) changes in the surface of the dielectric layer and have the advantages of label-free detection, high-throughput capacity, real-time monitoring and non-destructive measurement, they are widely used in biology, medicine, chemistry and many other fields [1–5]. In recent decades, many new SPR sensors have been introduced, such as prism-based SPR sensors , waveguide SPR sensors , fiber-optic SPR sensors  and grating-based SPR sensors . However, for all of these sensors, their sensitivities are susceptible to change by external factors, such as the fluctuation of the light-emitting diode (LED) light source, the change in bulk RI and temperature differences [10–14]. It is difficult to make an accurate evaluation through detection of a single SPR signal. Therefore, it is crucial to form reference channels or to create a favorable and stable environment for the SPR sensor, especially for detection of slight RI variation caused by biomolecular interaction. To compensate for signal fluctuation caused by external factors, many SPR sensors have been developed. Nenninger et al. described a dual-channel SPR biosensor with two sensor surfaces for monitoring of antibody–antigen binding . A second channel is used to compensate for changes in the RI of the bulk solution caused by analyte concentration or temperature difference. Berger et al. presented a prism-based SPR device containing multisensing channels with recognition elements for specific analytes in liquid phase . Shao et al. proposed an SPR sensor based tilt fiber grating . This sensor can compensate for the thermo-optical effect of the sensor and the temperature effect of samples. Zhang et al. described a temperature-compensation SPR sensor that provides two SPR resonance wavelengths by using a different coating . Ahnet al. proposed a fiber optic waveguide-coupled SPR sensor with a novel scheme of self-referencing . Using this sensor, the light source fluctuation can be compensated for without using an additional reference channel. However, all these reported SPR devices still have some defects and can be improved. Some of them are based on bulk and specialized devices that require sophisticated skills and intricate production methods. The dual-channel SPR sensors often use two sensing surfaces that do not sense in the same location. In addition, most of the dual-channel SPR sensors eliminate only one disturbing factor, such as temperature change or bulk RI change.
In this paper, we describe a compact SPR sensing method based on a capillary sensing element for self-reference biosensing. A short piece of capillary coated with a gold film is applied for SPR sensing, which has been used in our previous work . A web camera is used to capture the image of the measurement channel and reference channel. Unlike the previous implementation of the measurement channel and reference channel using two or nine optical fibers, several ROIs of single capillary end face are selected as the measurement channels and reference channels [20,21]. Because the gold film is partly functionalized, only a part of the SPR modes carried biological information without suffering from temperature fluctuation, bulk RI fluctuation, light source fluctuation and other factors. Through comparing the light intensities of the measurement channel and reference channel, we can obtain the biological information that is carried by the SPR mode. We calibrate the device using sodium chloride solutions with different RIs and tested its compensation capability when the temperature and the power of the LED are unstable. The capillary-based SPR device is used to detect specific binding of antibody and antigen to verify its capabilities of self-reference and biosensing. As a result, a compact biosensing device is easily achieved without specialized equipment and materials or complex structure, which may be useful for promoting and popularizing the application of SPR biosensing technology. Moreover, this method provides a new approach for self-reference SPR biosensing.
2. Design and construction of biosensing device
2.1 Experimental setup
The structure of the capillary-based SPR device is schematically illustrated in Fig. 1. It consisted of three components including the LED light source, the sensing element packaged with the flow cell and a webcam. An LED light source is used to illuminate the capillary with a single light source. We design simple control circuitry to adjust the light intensity of the LED light source. The light from the LED light source is converged by a lens to ensure uniform illumination of the capillary end face. The 625 nm emission wavelength of the LED light source is selected to match the strongest SPR absorption in the spectrum to provide maximum sensitivity. We fabricate the capillary-based SPR sensing elements with the light guide capillary (LTSP250350 from Polymicro Technologies, LLC, Phoenix, Arizona; core and cladding diameters 250 and 350 µm; numerical aperture 0.22, 3 cm length). The polyimide coating of the capillary is stripped to 1 cm in length in the center and immersed in hydrofluoric acid solution (20%) for 2 hours to remove the cladding. After washing and drying, the capillary is coated with chromium/gold layers (Cr 5 nm, Au 50 nm) using a magnetron sputtering system (K575XD from E.M. Technologies Ltd. Ashford, Kent). All end faces of the capillary are polished with emery paper. The capillary-based SPR sensing element is then packaged in a flow cell. The flow cell was fabricated using centrifuge tubes of 0.2mL via typical drilling and cutting processes and sample solutions were allowed to flow through the flow cell at a flow rate of 0.2 mL/min via a peristaltic pump (BT200-2J).A webcam is used to collect the light transmitted through the SPR sensing elements and the end face of the capillary sensing element can be imaged within the active area of the Complementary Metal Oxide Semiconductor (CMOS) detector imager chip. Through transmitting the data to a laptop by a USB cable, the images can be displayed on the screen and processed by laptop.
2.2 Characters of the capillary sensing element
Because the tubing wall of the capillary in sensing region is coated with gold, power of some light reached the sensing region will be absorbed by SPR effect when the sensing region contacts with sample. We test the capillary sensing element with four samples with different RI (1.3334, 1.3395, 1.3460 and 1.3518) to investigate the relationship between the SPR absorption and samples. The corresponding absorption spectra of the capillary sensing element are recorded by a wavelength modulation system which consists of an ocean optics halogen lamp, a capillary SPR sensor and mini-spectrometer as shown in Fig. 2(a). From the responding spectra in Fig. 2(b), we can find the absorption spectrum shifts to longer wavelengths for higher RI. Because a LED with wavelength of 625nm is used as a light source, we focus on light absorbance at a single wavelength (625 nm). As shown in Fig. 2(b), the transmission of the light at 625nm increase with the RI, i.e., the sensing element contacting with higher RI sample is less able to absorb the power of light at 625nm [Fig. 2(c)]. Because the power output of the sensing element will change with the change in RI, it means that it is possible to quantification of the analytes with the calibration of the SPR device.
As shown in Fig. 3(a), different from general silica capillary, the light guiding capillary has a doped silica cladding which makes the light guided by total internal reflection and determines the propagation modes in tubing wall. In order to know the light propagation in detail, we simulate the optical field distribution of the light in the capillary using MATLAB and Rsoft software. We select a beam of light to simulate its 3D optical field distribution for simplifying the calculation. The length of the capillary and wavelength of the light are selected as 1 mm and 625nm, respectively. First, the optical field distribution of the cross sections is simulated using Rsoft. The images of the optical field distribution of the cross sections are saved in a series of bitmapfiles, which are used to generate the 3D figure of the optical field distribution using MATLAB and 3D reconstruction as shown in Fig. 3(b). From some cross sections in different positions, we can find part of rays (shown here marked in purple) spread along the tubing wall and coiled around the tubing wall after the light enters the capillary, while other rays (shown in yellow ring marked in dark blue) bounce in the tubing wall and still propagate along their directions of incidence. The above two type rays are actually skew ray and meridional ray.
2.3 Principle of the biosensing device
Because the SPR absorption depends on the dielectric properties of the thin layer near the surface of gold film, the biomolecular interactions on the gold film can be analyzed with the SPR sensing. As shown in Fig. 4, we functionalize the bottom half of the gold film of capillary sensor with antibody to be a measurement channel for specific binding with antigen and monitor the brightness of the bottom half of the end face. Meanwhile, we use the top half of the gold film as a reference channel for compensating the environment disturbance (from temperature, bulk RI, light source and others) and monitor the brightness of the top half of the end face.
Because SPR modes were generated by skew ray and meridional ray in capillary tubing wall, the propagating traces of these SPR modes can be broadly classified into two types: SPR modes generated by skew ray follow a helical path coiled around the capillary and SPR modes generated by meridional ray follow a straight path. Under uniform illumination, these SPR modes are evenly distributed in the capillary and the end face of the capillary is illuminated as a bright ring. Because all of these SPR modes suffer from the same environment disturbances, the brightness of the lower half of the end face and the top half of the end face are equally affected by environment disturbances. As shown in Fig. 4, most of the SPR modes generated by skew ray pass through the functionalized sensing region and are affected by biomolecular interactions after traveling relatively long distances. However, for the SPR modes generated by skew ray, only part of them which present in the bottom half of the capillary pass through the functionalized sensing region and are affected by biomolecular interactions while others never pass through the functionalized sensing region. It means that the brightness of the bottom half of the end face is more affected by biomolecular interactions than the brightness of the top half of the end face.
As a result, through subtracting the brightness of the top half of the end face and the bottom half of the end face, brightness fluctuation caused by environment disturbances can be eliminated and the remains brightness can be used as measurement signal which is determined by biomolecular interactions.
The immobilization process for RNase B is performed as depicted in Figs. 5(a) and 5(b). The capillary based SPR sensing element is ultrasonic cleaned by deionized water, ethanol, dichloromethane and ethanol. After a subsequent wash, for immobilizing a layer of RNase B onto the surface of the bottom half of the gold film, alkanethiol self-assembled monolayer on gold film are prepared by the sensing element semi-submerged in an ethanolic solution of 11-mercaptoundecanic acid at room temperature for 24 hours. Then the sensing element is treated with an aqueous solution containing N-hydroxysuccinimide (NHS, 0.5M)/1-Ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride (EDC, 0.55 M) at 4°C for 30 minutes. After being dried by nitrogen, the sensing element is dipped in SPA (0.1mg/ml in PBS buffer, pH 7.4) and bovine serum albumin (BSA, 1 mg/mL in PBS buffer, pH 7.4) at room temperature for 50 and 30 minutes, respectively. Then, the sensing element has been functionalized and is mounted onto the flow cell for the measurements.
Furthermore, to validate the immobilization process for RNase B and specific binding process have been successfully conducted on surface of sensing region, we treat a capillary sensor in flow cell and monitor its response in real-time. The experiment result is shown in Fig. 5(c), with the introduction of the antigen, the response of the sensor has increased, which means that the antigen is fixed on the surface of the gold film. Then, the BSA is pumped into flow cell as a blocking agent and the sensor signal increase further which means that the unreacted binding site is filled by BSA. Finally, with the introduction of antibody, the signal increased significantly, which means that a specific combination has been achieved between the antigen and antibody.
2.5 Image process
The webcam captures the image of the end face of the capillary. Then, the data are transmitted to the laptop through a USB cable. We select two ROIs at the end faces of the capillary-based SPR sensing element as the measurement channel and reference channel. Thus, we design an image processing program by LabVIEW software, which is developed for ROI selection, data acquisition, data storage and data processing in real time. The processing flow of the images is shown in Fig. 6. The brightness information is extracted from the image data through a conversion of the colored image to grayscale images. Because every image point has a grayscale value between 0% (white) and 100% (black) to show its brightness, the light intensity of the ROIs is calculated by integrating the grayscale value. The images are captured and processed every second and are presented as intensity-time coordinates in real time.
3. Results and discussion
3.1 Self-reference function test
The responses of the measurement channel and reference channel to the environmental factors (variations of bulk RI, temperature, and LED) are investigated. Sodium chloride solutions with RIs of 1.328, 1.336, 1.341, 1.346 and 1.352 were introduced into the flow cell. The RIs of the sodium chloride solutions (calibration samples) were calibrated with an Abbe refractometer (WAY-2S). With an increase in the RI value of the sample in the flow cell, intensities of the measurement channel and reference channel increase at the same rate, as shown in Fig. 7(a). This means that the measurement channel and reference channel have the same responses to bulk RI. Moreover, the sensitivity and resolution of the SPR biosensing device can be calculated from the calibration curve. By calculating the slope of the calibration curve, we determine that the device provides a sensitivity of 1145%/RIU and a resolution better than 5.287 × 10−4 RIU, considering a 0.79% noise level. The responses of the measurement channel and reference channel to temperature fluctuation is shown in Fig. 7(b). We introduce deionized water at 22°C (room temperature) and 30°C (warm temperature) into the flow cell. From the response curve shown in Fig. 7(b), we can see that the intensity fluctuations of the measurement channel and reference channel are synchronized. Because water at 30°C is introduced into the flow cell at 100s, the intensity decreases gradually between 100s and 120s. Then, the intensity increase quickly and resume the initial intensity after rapidly pumping water at 22°C into the flow cell. Because the measurement channel and reference channel are in same capillary and flow cell, they encounter the same temperature fluctuation. Figure 7(c) shows the real-time responses of the measurement channel and reference channel when the light intensity of the LED is fluctuating. The intensity fluctuation of the LED is controlled by slide rheostat. In Fig. 7(c), the intensities of the measurement channel and reference channel have nearly identical fluctuation during the test. The reason is that the measurement channel and reference channel were illuminated by the same LED.
As a result, the experiment results indicate that the measurement channel and reference channel have similar responses to environment factors (bulk RI, temperature and light source). Thus, the response of the capillary-based SPR device to environment factors can be eliminated by subtracting the responses of the measurement channel and reference channel in real time.
3.2 Real time biosensing
To test the biosensing of the capillary based SPR biosensing device, the functionalized device is tested for real-time detection of Con A. First, PBS buffer is pumped into the flow cell for 5 min to obtain a stable baseline signal. The response of the device is investigated for the specific binding of Con A when diluted with PBS. The intensity increased due to the specific binding to RNase B. After that, PBS is pumped into the flow cell to remove the unbound Con A molecules. Urea solution (8.0 M) is then used to strip the surface-bound Con A and to effectively regenerate the sensing region after the sensing element is rinsed with PBS. Figure 8 shows the real-time relative intensity response of the device to the introduction of Con A (0.5 mg/mL, 1 mg/mL, 1.5 mg/mL and 2.0 mg/mL solutions, respectively). As shown in Fig. 8, with a Con A sample pumped into the flow cell, both the measurement channel and reference channel of the capillary-based SPR device exhibited specific responses to analytes. It can be explained as: Some SPR modes which are generated by skew rays and pass through the functionalized region can reach the whole end face of the capillary. The power of these SPR modes are affected by the Con A molecules binds to Rnase B, which leads to a brightness variation of the whole end face of the capillary. However, the measurement channel has a more significant intensity variation than that of the reference channel. What’s more, the higher the concentrations of the Con A samples are, the greater the response difference are between the measurement channel and reference channel. The reason is probably that some SPR modes generated by meridional rays pass through the functionalized region and only reach the bottom half of the end face of the capillary. As the Con A molecules binds to Rnase B on the surface of the bottom half of the gold film, it affects the power of these SPR modes and causes a brightness variation of the bottom half of the end face (measurement channel).
To quantitative analyses the above biosensing results, we define the average value of the intensity between 1500s-2225s as the final response for the measurement channel (MC), reference channel (RC). Figure 9(a) shows the final responses of the measurement channel (MC), reference channel (RC) and their difference value (MC-RC) which was calculated by subtraction. Figure 9(b) shows the final responses of the measurement channel and reference channel as a function of the Con A concentrations. The final response of the measurement channel and reference channel are approximately linear in the range of 0.5-2.0 mg/mL, with sensitivities of 0.8 and 0.4 (mg/mL)−1, respectively. To obtain the pure biological information about the specific binding between Con A and RNase B, the difference value (MC-RC) is used as the correction signal by subtracting the final response of the measurement channel and reference channel. Figure 9(c) shows there is a linear relationship between the intensity difference value (MC-RC) and concentrations of the Con A samples. The sensitivity is evaluated by the fitting curve at 0.4 (mg/mL)−1.
In summary, a compact SPR device based on capillary sensing element is presented. A short piece of capillary with 50nm gold film is used as the SPR sensing elements and illuminated by a LED light source. Because only part of the gold film is functionalized, the biological information is carried by a part of the SPR mode that is excited in the functionalized region. We monitor the end face of the capillary by webcam and separate regions of interest (ROI) are selected as the measurement channel and the reference channel. Through comparing the ROIs, biological information can be extracted accurately from the image by simple image processing. Moreover, temperature fluctuation, bulk refractive index (RI) fluctuation, light source fluctuation and other factors can be effectively compensated for during detection. After calibrating the device with a series of samples, its sensitivity is evaluated as 1145%/RIU and resolution is better than 5.287 × 10−4 RIU. The capillary based SPR biosensing device is used to detect specific binding of antibody and antigen to verify its capabilities of self-reference and biosensing. A dual-channel functionalized with RNaseB responded to the specific detection of Con A with an approximately linear response to the analyte concentration. Moreover, using a micro surface modification technique , the channel of the SPR sensing element can be added by separating the gold surface into blocks and modifying them with different bioprobes. As a result, research for such a capillary SPR sensing element has mainly focused on attaining a simple biosensing method for self-reference and label free analyte detection.
National Natural Science Foundation of China (NSFC) (Grant Nos. 61520106013, 11474043 and 61137005).
References and links
1. D. R. Shankaran, K. V. Gobi, and N. Miura, “Recent advancements in surface plasmon resonance immunosensors for detection of small molecules of biomedical, food and environmental interest,” Sens. Actuators B Chem. 121(1), 158–177 (2007). [CrossRef]
3. K. V. Gobi and N. Miura, “Highly sensitive and interference-free simultaneous detection of two polycyclic aromatic hydrocarbons at parts-per-trillion levels using a surface plasmon resonance immunosensor,” Sens. Actuators B Chem. 103(1), 265–271 (2004). [CrossRef]
4. J. A. Ruemmele, M. S. Golden, Y. Gao, E. M. Cornelius, M. E. Anderson, L. Postelnicu, and R. M. Georgiadis, “Quantitative surface plasmon resonance imaging: a simple approach to automated angle scanning,” Anal. Chem. 80(12), 4752–4756 (2008). [CrossRef] [PubMed]
5. T. M. Chinowsky, M. S. Grow, K. S. Johnston, K. Nelson, T. Edwards, E. Fu, and P. Yager, “Compact, high performance surface plasmon resonance imaging system,” Biosens. Bioelectron. 22(9-10), 2208–2215 (2007). [CrossRef] [PubMed]
6. S. G. Nelson, K. S. Johnston, and S. S. Yee, “High sensitivity surface plasmon resonance sensor based on phase detection,” Sens. Actuators B Chem. 35(1), 187–191 (1996). [CrossRef]
7. T. J. Wang, C. W. Tu, F. Liu, and H. L. Chen, “Surface plasmon resonance waveguide biosensor by bipolarization wavelength interrogation,” IEEE Photonics Technol. Lett. 16(7), 1715–1717 (2004). [CrossRef]
8. B. D. Gupta and R. K. Verma, “Surface plasmon resonance-based fiber optic sensors: principle, probe designs, and some applications,” J. Sens. 2009, 1–12 (2009). [CrossRef]
9. M. Piliarik, M. Vala, I. Tichý, and J. Homola, “Compact and low-cost biosensor based on novel approach to spectroscopy of surface plasmons,” Biosens. Bioelectron. 24(12), 3430–3435 (2009). [CrossRef] [PubMed]
10. D. Monzón-Hernández, J. Villatoro, D. Talavera, and D. Luna-Moreno, “Optical-fiber surface-plasmon resonance sensor with multiple resonance peaks,” Appl. Opt. 43(6), 1216–1220 (2004). [CrossRef] [PubMed]
11. Z. Zhang, P. Zhao, F. Sun, G. Xiao, and Y. Wu, “Self-referencing in optical-fiber surface plasmon resonance sensors,” IEEE Photonics Technol. Lett. 19(24), 1958–1960 (2007). [CrossRef]
12. W. Peng, S. Banerji, Y. C. Kim, and K. S. Booksh, “Investigation of dual-channel fiber-optic surface plasmon resonance sensing for biological applications,” Opt. Lett. 30(22), 2988–2990 (2005). [CrossRef] [PubMed]
13. E. K. Akowuah, T. Gorman, S. Haxha, and J. V. Oliver, “Dual channel planar waveguide surface plasmon resonance biosensor for an aqueous environment,” Opt. Express 18(24), 24412–24422 (2010). [CrossRef] [PubMed]
14. L. L. Obando and K. S. Booksh, “Tuning dynamic range and sensitivity of white-light, multimode, fiber-optic surface plasmon resonance sensors,” Anal. Chem. 71(22), 5116–5122 (1999). [CrossRef]
15. G. G. Nenninger, J. B. Clendenning, C. E. Furlong, and S. S. Yee, “Reference-compensated biosensing using a dual-channel surface plasmon resonance sensor system based on a planar lightpipe configuration,” Sens. Actuators B Chem. 51(1), 38–45 (1998). [CrossRef]
16. C. H. Berger, T. A. M. Beumer, R. P. H. Kooyman, and J. Greve, “Surface plasmon resonance multisensing,” Anal. Chem. 70(4), 703–706 (1998). [CrossRef]
17. L. Y. Shao, Y. Shevchenko, and J. Albert, “Intrinsic temperature sensitivity of tilted fiber Bragg grating based surface plasmon resonance sensors,” Opt. Express 18(11), 11464–11471 (2010). [CrossRef] [PubMed]
20. W. Peng, Y. Liu, P. Fang, X. Liu, Z. Gong, H. Wang, and F. Cheng, “Compact surface plasmon resonance imaging sensing system based on general optoelectronic components,” Opt. Express 22(5), 6174–6185 (2014). [CrossRef] [PubMed]
21. Y. Liu, S. Chen, Q. Liu, J. F. Masson, and W. Peng, “Compact multi-channel surface plasmon resonance sensor for real-time multi-analyte biosensing,” Opt. Express 23(16), 20540–20548 (2015). [CrossRef] [PubMed]
22. I. Turyan, T. Matsue, and D. Mandler, “Patterning and characterization of surfaces with organic and biological molecules by the scanning electrochemical microscope,” Anal. Chem. 72(15), 3431–3435 (2000). [CrossRef] [PubMed]