- Email: [email protected]
Whispering gallery mode imaging for the multiplexed detection of biomarkers
Sensors and Actuators B 160 (2011) 1262–1267
Contents lists available at SciVerse ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
Whispering gallery mode imaging for the multiplexed detection of biomarkers Heath A. Huckabay, Robert C. Dunn ∗ Ralph N. Adams Institute for Bioanalytical Chemistry, University of Kansas, 2030 Becker Drive, Lawrence, KS 66047, United States
a r t i c l e
i n f o
Article history: Received 11 April 2011 Received in revised form 15 September 2011 Accepted 20 September 2011 Available online 28 September 2011 Keywords: Whispering gallery mode Total internal reflection Label-free Immunoassay
a b s t r a c t An approach is presented for combining sensitive fluorescence imaging with whispering gallery mode (WGM) resonators to develop a label-free, multiplexed approach for biosensing. The evanescent field generated by a total internal reflection microscope is used to couple light into a field of microsphere resonators dispersed on a glass coverslip. The microspheres are labeled with the fluorescent dye Alexa 633 which acts as a reporter of the WGM resonance for each microsphere. As the excitation wavelength is scanned, a distinct ring of increased fluorescence intensity is observed when the resonance condition is satisfied. For sensing applications, shifts in the WGM resonance in response to changes in the effective refractive index can be used to track analyte binding. Calibration plots of WGM shifts with refractive index indicate the dye functionalized microspheres have a sensitivity of ∼50 nm/RIU. To illustrate the biosensing capabilities, two markers of ovarian cancer were detected. Antibodies specific for CA-125 and TNF-␣ were linked to 38 m and 53 m spheres, respectively, and used to detect their biomarkers in solution using fluorescence imaging of the WGM resonance shifts. Both assays resulted in linear response curves and the detection limits for the important ovarian cancer marker CA-125 is more than threefold better than commercial ELISA kits. These results illustrate the feasibility of combining WGM resonators with fluorescence imaging for large scale multiplexing of biomarker detection. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Whispering gallery mode resonators are small dielectric structures that can confine light through continuous total internal reflection at the resonator interface [1–4]. When light is coupled into small spherical structures such as glass microspheres, whispering gallery mode (WGM) resonances are observed at frequencies where the circumnavigation distance around the sphere is an integer multiple of the coupled wavelength [1,3]. At a WGM resonance, the circulating light coherently drives itself leading to greatly enhanced storage of light within the resonator. Whispering gallery mode resonators are emerging as useful platforms for the development of biosensors [1,3,5–7]. These approaches generally take advantage of the dependence of the WGM resonant wavelength with effective refractive index. For a resonator much larger than the optical wavelength, the resonant condition is given by [8,9]:
r =
2rneff m
(1)
Abbreviations: WGM, whispering gallery mode; Q factor, quality factor; TIR, total internal reflection; CA-125, cancer antigen-125; TNF-␣, tumor necrosis factor-␣. ∗ Corresponding author. Tel.: +1 785 864 4313. E-mail address: [email protected] (R.C. Dunn). 0925-4005/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2011.09.060
where r is the resonant wavelength, r is the resonator radius, neff is the effective refractive index surrounding the sphere, and m is an integer. This relationship shows that the WGM resonance wavelength is directly proportional to the effective refractive index, neff . For biosensing applications, recognition elements such as antibodies can be immobilized to the resonator surface which, upon binding of their associated antigen, alters the effective refractive index surrounding the resonator [10,11]. This leads to shifts in the WGM resonance which provides a label-free approach for biosensing. Using this general approach, WGM resonators have been utilized in the detection of a wide array of species from proteins [12–15] and DNA [16,17] to bacteria [18]. The biosensing performance of WGM resonators is strongly linked to the quality (Q) factor of the resonator [7,19–21]. A high Q factor is associated with a narrow resonance and an increased effective path length, both of which are highly desirable for optimal sensor performance. While microfabricated WGM resonators have Q-factors on the order of 104 [22], simple microsphere resonators such as high index glass spheres can have Q-factors approaching 109 [23]. The formation of spheres from melts results in an exquisitely smooth surface which reduces losses and results in ultrahigh Q factors [3,21,23]. In addition to the superior sensing capabilities associated with their high Q factors, microspheres are generally inexpensive, easy to fabricate, and available in a range of materials. However, difficulties in multiplexing the detection of analytes using microsphere resonators has limited their applications in sensing platforms and shifted focus to microfabricated
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
resonators. Microfabricated ring or disk resonators are, in general, more easily integrated into photonics and fluidics systems, have favorable scalability, and are amenable to multiplexed detection schemes [13,14,20,24]. However, as mentioned, the Q factors of these devices are generally orders of magnitude smaller than those of microsphere resonators which restricts their detection limits. Ideally, one wants to combine the unmatched Q factors of microspheres with the multiplexed detection capabilities of microfabricated devices. Here we present a whispering gallery mode imaging (WGMI) scheme that uses sensitive fluorescence imaging of microsphere resonances to provide multiplexed detection capabilities. The WGMI approach utilizes an inverted total internal reflection (TIR) microscope to evanescently couple light from a tunable diode laser into a field of microsphere resonators. The surface of each microsphere is labeled with a fluorescent dye which acts as a reporter for the WGM resonance. On resonance, a distinctive ring of fluorescence is observed as the WGM excites the surface attached dye. As the wavelength of the tunable diode laser is scanned, therefore, the WGM resonance of each microsphere in the field of view is measured through the enhanced fluorescent ring, which is imaged from above with an upright microscope. The feasibility of biosensing using this approach is demonstrated by detecting two biomarkers of ovarian cancer – cancer antigen-125 (CA-125) and tumor necrosis factor-alpha (TNF-␣) [25–30]. Monoclonal antibodies specific for CA-125 and TNF-␣ were attached to 38 m and 53 m glass microspheres, respectively. Using the relative intensity of an attached fluorescent marker to track changes in the resonant wavelength, linear calibration curves were measured for both antigens. For CA-125, we demonstrate detection limits more than threefold better than commercial ELISA kits. Using this approach, the identity of the detected analyte can be encoded in the sphere size and/or location, providing vastly improved multiplexing capabilities. Moreover, the chemistry associated with attaching antibodies to glass beads is well developed and all immobilization procedures are carried out before the spheres are deposited onto the substrate surface. Using this approach simplifies the chemistry associated with multiplexing the detection over other WGM techniques. This method, therefore, provides a sensitive and straightforward approach for multiplexed WGM detection that takes advantage of the superior optical properties of microsphere resonators.
1263
in a 5% APTES (v/v) solution in toluene and allowed to tumble overnight. Following amine modification, the spheres were rinsed once in toluene and triply washed in absolute ethanol. Each sphere set was then placed in a 6% glutaraldehyde (v/v) solution in PBS and tumbled for 2 h. The sphere sets were each triply washed in PBS then placed into corresponding antibody solutions in PBS to tumble for 2 h: the 38 m diameter spheres were incubated in 2 g/mL anti-CA125 solution while 53 m diameter microspheres where incubated in 2 g/mL anti-TNF-␣. Following incubation, each sphere solution was washed in PBS and labeled with the fluorescent dye, Alexa 633. Finally, each microsphere set was tumbled in a 5% BSA, 5% sucrose (w/v) solution in PBS for 1 h to block nonspecific adsorption sites. 2.3. WGM measurement For WGM analysis, spheres were dispersed onto a glass coverslip and placed in a CoverWell perfusion chamber containing PBS. Light was coupled into the spheres using the evanescent wave generated with an inverted total internal reflection fluorescence microscope (TIRF-M) (Olympus IX71, Center Valley, PA) equipped with a 60× objective (1.45 NA achromat, Olympus). The output from an external cavity diode laser (TLB-6904, New Focus, Santa Clara, CA) is offset on the back aperture of the objective to create total internal reflection at the substrate interface. The diode laser can be tuned from 632.86 to 633.12 nm with ≤300 kHz linewidth. Fluorescence is collected from above using an Olympus 10× UMPlanFl (0.3 NA) objective lens, filtered (Thorlabs, Newton, NJ), and imaged onto a CCD camera (Cascade 650, Roper Scientific, Tuscon, AR). The laser was scanned using LabVIEW (National Instruments, Austin, TX) software and image collection and processing was controlled using Slidebook software (v4.2.0.3, Intelligent Imaging Innovations, Denver, CO).
2. Materials and methods 2.1. Reagents Soda–lime glass microspheres were obtained from Mo-Sci Corporation (Rolla, MO). Anti-CA-125 mouse monoclonal antibodies (SPM111), 3-aminopropyltriethoxysilane (APTES), glutaraldehyde, bovine serum albumin (BSA), and all solvents were obtained from Fisher Scientific (Hampton, NH). CA-125 antigen was obtained from a commercially available CA-125 screening kit (Panomics BC-1013). Anti-TNF-␣ antibodies were utilized from a commercial TNF-␣ screening kit (ESS0001, Fisher Scientific), and TNF-␣ antigen was purchased from Pierce Biotechnology (Rockford, IL). The amine reactive succinimidyl ester of Alexa Fluor 633 was obtained from Invitrogen Corporation (Carlsbad, CA). CoverWell perfusion chambers were obtained from Grace Bio-Labs (Bend, OR). 2.2. Microsphere preparation The 38 m and 53 m diameter microspheres were separately sonicated in an alconox solution, triply rinsed with 18 M water, and rinsed with absolute ethanol. Each sphere set was placed
Fig. 1. Schematic of the apparatus used to couple WGM excitation with fluorescence imaging. The output of a tunable diode laser is coupled into an inverted total internal reflection microscope through a high numerical aperture (60×, 1.45 NA) objective. By offsetting the excitation beam in the back aperture of the objective, light approaches the sample beyond the critical angle which leads to total internal reflection (TIR) at the substrate surface. The associated evanescent field is used to excite WGMs in spheres which have been labeled with antibodies to detect CA-125 (38 m) and TNF-␣ (53 m). The spheres are also labeled with a fluorescent dye (stars) which acts as a marker of the WGM resonance. On resonance, the evanescent field from light circumnavigating the sphere excites fluorescence from the dye, which is collected from above and imaged onto a CCD.
1264
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
Fig. 2. (Top) Representative fluorescence images of a 38 m diameter microsphere labeled with Alexa 633. As the excitation wavelength is scanned, a distinctive ring of enhanced fluorescence is observed (middle panel) when a WGM resonance is encountered. (Below) Excitation spectrum of fluorescence intensity with excitation wavelength, extracted from a series of fluorescence images.
3. Results and discussion Here we report a general approach for combining WGM microsphere resonators with fluorescence imaging to develop a new approach for the multiplexed detection of analytes. The general approach is outlined schematically in Fig. 1. 3.1. Experimental design As shown in Fig. 1, light from a tunable diode laser with narrow spectral linewidth (≤300 kHz) is launched into an optical fiber and coupled into an inverted total internal reflection (TIR) microscope. The TIR microscope uses a high numerical aperture (1.45 NA) objective to couple light to the substrate surface beyond the critical angle, where total internal reflection takes place [31]. The associated evanescent field is used to launch light into the microsphere resonators on the substrate surface as shown schematically in Fig. 1. As the excitation wavelength of the diode laser is scanned, the WGM resonance from each microsphere in the field of view is detected through fluorescence imaging. This is accomplished by labeling all the microspheres with the same fluorescent dye (stars in Fig. 1) which acts as a reporter of the WGM resonance. When a particular microsphere reaches the resonance condition given
by Eq. (1), a bright ring of enhanced fluorescence is observed as the evanescent field from the WGM resonance excites the surface attached dye. By collecting the fluorescence from above and imaging it onto a CCD camera, the WGM resonance for each sphere in the field of view can be simultaneously measured as the excitation wavelength is scanned. A similar approach has been reported by utilizing quantum-dot embedded polystyrene microspheres excited at single points using low magnification objectives [32,33]. To illustrate this approach, Fig. 2 shows a series of fluorescence images taken of the same 38 m diameter dye labeled soda–lime glass microsphere as the excitation wavelength is scanned. When a WGM resonance is reached (center panel) a distinctive ring of enhanced fluorescence is observed as the evanescent field from the circumnavigating light excites the surface bound dye. With the current coupling arrangement and dye labeling scheme, we routinely measure 2–4-fold enhancements in the fluorescence intensity at resonance which makes them easily identifiable over the off resonance background fluorescence. The excitation spectrum in Fig. 2 plots the fluorescence intensity from the sphere as the excitation wavelength is scanned. This approach, therefore, uses the fluorescence marker to measure the WGM resonance wavelength. Since the WGM resonance is an intrinsic property of the system, this approach is not as susceptible to complications that can arise from
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
1265
Fig. 3. To explore the sensing capabilities of the WGM imaging approach, fluorescence was used to measure the resonance wavelength as the surrounding refractive index was systematically changed by adding aliquots of ethanol to water. A typical calibration plot measured for a 53 m sphere is shown above. As expected from Eq. (1), the plot of r versus neff is linear (R2 = 0.9931), with a measured sensitivity of ∼50 nm/RIU.
photobleaching or variable dye loading that are problematic for intensity based measurements. Typical Q-factors calculated from these spectra range from 2 × 104 to 5 × 104 , which are on the order of Q-factors measured for microfabricated resonators, but orders of magnitude lower than that expected for microspheres [22,23]. The dye attached to the surface of the microsphere acts as a source of loss in this arrangement which lowers the Q-factor. We are currently exploring optimal dye loading conditions to maximize the Q-factor. 3.2. Sensor calibration To demonstrate that the WGMI approach can track changes in WGM resonance wavelength, calibration plots were measured where the refractive index of the bathing solution was systematically changed. Fig. 3 plots a typical calibration curve measured using the fluorescence imaging approach. In these experiments, microsphere resonators were bathed in 18 M water to which injections of absolute ethanol were added to increase the surrounding refractive index [34]. As seen in Fig. 3, a linear response is observed as expected from the relationship between WGM resonance wavelength and refractive index (Eq. (1)). The particular calibration plot shown in Fig. 3 was measured from a microsphere with a Q-factor of ∼2 × 104 and has a measured sensitivity of ∼50 nm/RIU (refractive index units) and a detection limit of ∼2.6 × 10−5 RIU. Linear calibration plots such as that shown in Fig. 3 illustrates the feasibility of using microsphere resonators combined with fluorescence imaging for biosensing. 3.3. Ovarian cancer biomarker detection by WGMI Having demonstrated that the WGMI approach is sensitive to changes in the refractive index, the specific biosensing capabilities are demonstrated using antibodies specific for biomarkers of ovarian cancer. Two separate assays were performed with microspheres modified with antibodies specific for cancer antigen-125 (CA-125) and tumor necrosis factor-␣ (TNF-␣), both implicated in the pathogenesis of ovarian carcinoma [25–30]. As illustrated in Fig. 1, the antibodies are attached to the sphere surface along with the fluorescent dye. Moreover, each antibody is attached to a distinct sphere size, thus encoding the analyte identity into the sphere diameter which is easily measured in the imaging approach. Here monoclonal antibodies specific for CA-125 were immobilized on 38 m spheres while antibodies for TNF-␣ were attached to 53 m spheres.
Fig. 4. (A) Calibration plot of WGM resonance versus CA-125 (in Units/mL) measured using 38 m microsphere resonators modified with anti-CA-125 antibodies and the fluorescent dye Alexa 633. Binding of CA-125 to the immobilized antibody alters the effective refractive index, leading to shifts in the WGM resonance. The linear calibration plot (R2 = 0.9677) has a sensitivity of ∼6 pm/(U/mL). (B) Calibration plot of WGM resonance versus TNF-␣ (in ng/mL) measured using 53 m microsphere resonators modified with anti-TNF-␣ antibodies and the fluorescent dye Alexa 633. The linear calibration plot (R2 = 0.9798) has a sensitivity of ∼108 pm/ng/mL.
Fig. 4 plots the shifts in WGM resonant wavelength as CA-125 was injected into the perfusion chamber. A linear shift in the resonant wavelength is observed with a measured CA-125 sensitivity of ∼6 pm/U/mL and detection limit of ∼1.5 U/mL. This detection limit is threefold better than the 5 U/mL detection limit of commercial ELISA kits (Panomics BC-1013). Also shown in Fig. 4 is a similar calibration plot for the detection of TNF-␣ using 53 m spheres functionalized with monoclonal antibodies for TNF-␣. The linear response of TNF-␣ shown in Fig. 4 has sensitivity of ∼108 pm/ng/mL and a detection limit of ∼240 pg/mL. Interestingly, the sensitivity is approximately an order of magnitude higher than previous microresonator reports using other antibody-antigen sensing modalities. This may arise from differences in the specific KD values for the antibody–antigen pairs used in these studies [35–37]. However, the shorter excitation wavelength used here may also influence microresonator sensor performance through changes in mode volume and mode density. Experiments are underway to quantify these effects. The calibration plots shown in Fig. 4 demonstrate the specific detection of biomarkers using the WGMI approach. Coupling WGM resonators with sensitive fluorescence imaging offers many advantages for the multiplexed detection of biomarkers. The approach is label-free, sensitive and rapid. Since the analyte identity is encoded in the sphere size or location, every sphere can be labeled with the same dye and excited with the same laser system. Since the WGM resonance wavelength is an intrinsic property of the system, the assay does not suffer from complications arising from dye photobleaching or variable labeling. Microspheres are readily
1266
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
available, inexpensive, come in a range of materials and sizes, and have extremely large Q-factors which are desirable for sensing applications. Finally, their small size makes them compatible with many of the developments taking place in the lab-on-a-chip field, which can further reduce assay costs. 4. Conclusions Fluorescence imaging is combined with total internal reflection excitation of whispering gallery modes in microspheres to develop a new approach for multiplexed biosensing. In this approach, microspheres are functionalized with a fluorescent dye which acts as a reporter of the WGM resonance. Light is coupled into a field of microspheres using the evanescent field created in a total internal reflection microscope. As the excitation wavelength is scanned, WGM resonances are detected through fluorescence imaging. As a particular sphere reaches a WGM resonance, a distinct ring of enhanced fluorescence is observed as the evanescent wave from light circumnavigating the sphere excites the surface attached dye. This approach enables the simultaneous measurement of WGM resonances from every microsphere in the field of view, thus enabling the development of highly multiplexed biosensors. To demonstrate the feasibility of biosensing using the WGMI approach, antibodies specific for CA-125 and TNF-␣ were linked to 38 m and 53 m spheres, respectively. CA-125 and TNF-␣ are commonly used markers for the screening of ovarian cancer and by linking their antibodies to differentially sized microspheres, detection specificity is encoded into sphere diameter. Binding of their associated antigens shifts the WGM resonance of the sphere, which is detected using fluorescence imaging. Calibration curves measured in separate assays reveals a linear response for both antigens with the detection limits for CA-125 over threefold better than that obtained with commercial ELISA approaches. These results confirm that fluorescence imaging can be used to simultaneously detect WGMs from a field of microsphere resonators where analyte identity can be encoded into sphere size and/or location, thus enabling significant multiplexing capabilities. Acknowledgments We gratefully acknowledge support from NSF (CBET 1133814) and the Madison and Lila Self Foundation. HAH is grateful to Kevin Armendariz for assistance with LabVIEW programming. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.snb.2011.09.060. References [1] S. Arnold, R. Ramjit, D. Keng, V. Kolchenko, I. Teraoka, Microparticle photophysics illuminates viral bio-sensing, Faraday Discuss. 137 (2008) 99–113, 65-83. [2] K.J. Vahala, Optical microcavities, Nature 424 (2003) 839–846. [3] S. Soria, S. Berneschi, M. Brenci, F. Cosi, G.N. Conti, S. Pelli, G.C. Righini, Optical microspherical resonators for biomedical sensing, Sensors 11 (2011) 785–805. [4] I.M. White, H. Zhu, J.D. Suter, N.M. Hanumegowda, H. Oveys, M. Zourob, X. Fan, Refractometric sensors for lab-on-a-chip based on optical ring resonators, IEEE Sens. J. 7 (2007) 28–35. [5] J.D. Suter, X. Fan, Overview of the optofluidic ring resonator: a versatile platform for label-free biological and chemical sensing, in: EMBC 2009, Ann. Int. Conf. IEEE (2009) 1042–1044. [6] S. Wang, A. Ramachandran, J. Clarke, S.J. Ja, D. Goad, L. Wald, E.M. Flood, E. Knobbe, J.V. Hryniewicz, S.T. Chu, D. Gill, W. Chen, O. King, B.E. Little, A universal biosensing platform based on optical micro-ring resonators, Biosens. Bioelectron. 23 (2008) 939–944. [7] F. Vollmer, S. Arnold, Whispering-gallery-mode biosensing: label-free detection down to single molecules, Nat. Methods 5 (2008) 591–596.
[8] M.L. Gorodetsky, V.S. Ilchenko, Optical microsphere resonators: optimal coupling to high-Q whispering-gallery modes, J. Opt. Soc. Am. B: Opt. Phys. 16 (1999) 147–154. [9] J.C. Knight, G. Cheung, F. Jacques, T.A. Birks, Phase-matched excitation of whispering-gallery-mode resonances by a fiber taper, Opt. Lett. 22 (1997) 1129–1131. [10] I. Teraoka, S. Arnold, F. Vollmer, Perturbation approach to resonance shifts of whispering-gallery modes in a dielectric microsphere as a probe of a surrounding medium, J. Opt. Soc. Am. B 20 (2003) 1937–1946. [11] F. Vollmer, D. Braun, A. Libchaber, M. Khoshsima, I. Teraoka, S. Arnold, Protein detection by optical shift of a resonant microcavity, Appl. Phys. Lett. 80 (2002) 4057–4059. [12] H. Zhu, J. Suter, I. White, X. Fan, Aptamer based microsphere biosensor for thrombin detection, Sensors 6 (2006) 785–795. [13] A.L. Washburn, L.C. Gunn, R.C. Bailey, Label-free quantitation of a cancer biomarker in complex media using silicon photonic microring resonators, Anal. Chem. 81 (2009) 9499–9506. [14] A.L. Washburn, M.S. Luchansky, A.L. Bowman, R.C. Bailey, Quantitative, labelfree detection of five protein biomarkers using multiplexed arrays of silicon photonic microring resonators, Anal. Chem. 82 (2009) 69–72. [15] H.Y. Zhu, P.S. Dale, C.W. Caldwell, X.D. Fan, Rapid and label-free detection of breast cancer biomarker CA15-3 in clinical human serum samples with optofluidic ring resonator sensors, Anal. Chem. 81 (2009) 9858–9865. [16] X.D. Fan, J.D. Suter, I.M. White, H.Y. Zhu, H.D. Shi, C.W. Caldwell, Label-free quantitative DNA detection using the liquid core optical ring resonator, Biosens. Bioelectron. 23 (2008) 1003–1009. [17] F. Vollmer, S. Arnold, D. Braun, I. Teraoka, A. Libchaber, Multiplexed DNA quantification by spectroscopic shift of two microsphere cavities, Biophys. J. 85 (2003) 1974–1979. [18] H.-C. Ren, F. Vollmer, S. Arnold, A. Libchaber, High-Q microsphere biosensor—analysis for adsorption of rodlike bacteria, Opt. Express 15 (2007) 17410–17423. [19] A.M. Armani, R.P. Kulkarni, S.E. Fraser, R.C. Flagan, K.J. Vahala, Label-free, singlemolecule detection with optical microcavities, Science 317 (2007) 783–787. [20] I.M. White, X. Fan, On the performance quantification of resonant refractive index sensors, Opt. Express 16 (2008) 1020–1028. [21] N.M. Hanumegowda, C.J. Stica, B.C. Patel, I. White, X. Fan, Refractometric sensors based on microsphere resonators, Appl. Phys. Lett. 87 (2005) 201107. [22] X. Fan, I.M. White, H. Zhu, J.D. Suter, H. Oveys, Overview of novel integrated optical ring resonator bio/chemical sensors, in: A.V. Kudryashov, A.H. Paxton, V.S. Ilchenko (Eds.), SPIE, San Jose, CA, USA, 2007, pp. 64520M–64520. [23] M.L. Gorodetsky, A.A. Savchenkov, V.S. Ilchenko, Ultimate Q of optical microsphere resonators, Opt. Lett. 21 (1996) 453–455. [24] D.X. Xu, A. Densmore, A. Delâge, P. Waldron, R. McKinnon, S. Janz, J. Lapointe, G. Lopinski, T. Mischki, E. Post, P. Cheben, J.H. Schmid, Folded cavity SOI microring sensors for high sensitivity and real time measurement of biomolecular binding, Opt. Express 16 (2008) 15137–15148. [25] C. Tian, M. Markman, R. Zaino, R.F. Ozols, W.P. McGuire, F.M. Muggia, P.G. Rose, D. Spriggs, D.K. Armstrong, CA-125 change after chemotherapy in prediction of treatment outcome among advanced mucinous and clear cell epithelial ovarian cancers: a gynecologic oncology group study, Cancer 115 (2009) 1395–1403. [26] P. Kenemans, C.A. Yedema, G.G. Bon, S. von Mensdorff-Pouilly, CA 125 in gynecological pathology—a review, Eur. J. Obstet. Gynecol. Reprod. Biol. 49 (1993) 115–124. [27] B. Dobrzycka, S.J. Terlikowski, O. Kowalczuk, M. Kinalski, Circulating levels of TNF-alpha and its soluble receptors in the plasma of patients with epithelial ovarian cancer, Eur. Cytokine Netw. 20 (2009) 131–134. [28] D. Gupta, C.G. Lis, Role of CA125 in predicting ovarian cancer survival—a review of the epidemiological literature, J. Ovarian Res. 2 (2009) 13. [29] B.B. Hermsen, S. von Mensdorff-Pouilly, J. Berkhof, P.J. van Diest, J.J. Gille, F.H. Menko, M.A. Blankenstein, P. Kenemans, R.H. Verheijen, Serum CA-125 in relation to adnexal dysplasia and cancer in women at hereditary high risk of ovarian cancer, J. Clin. Oncol. 25 (2007) 1383–1389. [30] B. Dobrzycka, S.J. Terlikowski, M. Garbowicz, W. Niklinska, P.S. Bernaczyk, J. Niklinski, M. Kinalski, L. Chyczewski, Tumor necrosis factor-alpha and its receptors in epithelial ovarian cancer, Folia Histochem. Cytobiol. 47 (2009) 609–613. [31] D. Axelrod, Total internal reflection fluorescence microscopy, in: J.C. John, H. William Detrich III (Eds.), Methods in Cell Biology, Academic Press, 2008, pp. 169–221. [32] H. Beier, G. Coté, K. Meissner, Whispering gallery mode biosensors consisting of quantum dot-embedded microspheres, Ann. Biomed. Eng. 37 (2009) 1974–1983. [33] S. Pang, R.E. Beckham, K.E. Meissner, Quantum dot-embedded microspheres for remote refractive index sensing, Appl. Phys. Lett. 92 (2008) 221108. [34] D.R. Lide, CRC Handbook of Chemistry and Physics, CRC Press, Boca Raton, FL, 2011. [35] J. Maynard, G. Georgiou, Antibody engineering, Annu. Rev. Biomed. Eng. 2 (2000) 339–376. [36] P. Rathanaswami, J. Babcook, M. Gallo, High-affinity binding measurements of antibodies to cell-surface-expressed antigens, Anal. Biochem. 373 (2008) 52–60. [37] G.Q. Liu, Y. Liu, W. Zhang, X. Yu, H.W. Zhang, R. Zhao, D. Shangguan, Y. Li, B.F. Shen, Quartz crystal biosensor for real-time kinetic analysis of interaction between human TNF-alpha and monoclonal antibodies, Sens. Actuators B 99 (2004) 416–424.
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
Biographies Heath A. Huckabay is a graduate student in Department of Chemistry at University of Kansas. His research interests include optical sensor development and spectroscopic imaging. He received his B.Sc. in Forensic Chemistry in 2006 from Sam Houston State University in Huntsville, Texas.
1267
Robert C. Dunn is a professor of chemistry at University of Kansas and member of the Ralph N. Adams Institute for Bioanalytical Chemistry. His research specializes in using novel optical approaches for analyzing samples with high spatial resolution and low detection limits.
Contents lists available at SciVerse ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
Whispering gallery mode imaging for the multiplexed detection of biomarkers Heath A. Huckabay, Robert C. Dunn ∗ Ralph N. Adams Institute for Bioanalytical Chemistry, University of Kansas, 2030 Becker Drive, Lawrence, KS 66047, United States
a r t i c l e
i n f o
Article history: Received 11 April 2011 Received in revised form 15 September 2011 Accepted 20 September 2011 Available online 28 September 2011 Keywords: Whispering gallery mode Total internal reflection Label-free Immunoassay
a b s t r a c t An approach is presented for combining sensitive fluorescence imaging with whispering gallery mode (WGM) resonators to develop a label-free, multiplexed approach for biosensing. The evanescent field generated by a total internal reflection microscope is used to couple light into a field of microsphere resonators dispersed on a glass coverslip. The microspheres are labeled with the fluorescent dye Alexa 633 which acts as a reporter of the WGM resonance for each microsphere. As the excitation wavelength is scanned, a distinct ring of increased fluorescence intensity is observed when the resonance condition is satisfied. For sensing applications, shifts in the WGM resonance in response to changes in the effective refractive index can be used to track analyte binding. Calibration plots of WGM shifts with refractive index indicate the dye functionalized microspheres have a sensitivity of ∼50 nm/RIU. To illustrate the biosensing capabilities, two markers of ovarian cancer were detected. Antibodies specific for CA-125 and TNF-␣ were linked to 38 m and 53 m spheres, respectively, and used to detect their biomarkers in solution using fluorescence imaging of the WGM resonance shifts. Both assays resulted in linear response curves and the detection limits for the important ovarian cancer marker CA-125 is more than threefold better than commercial ELISA kits. These results illustrate the feasibility of combining WGM resonators with fluorescence imaging for large scale multiplexing of biomarker detection. © 2011 Elsevier B.V. All rights reserved.
1. Introduction Whispering gallery mode resonators are small dielectric structures that can confine light through continuous total internal reflection at the resonator interface [1–4]. When light is coupled into small spherical structures such as glass microspheres, whispering gallery mode (WGM) resonances are observed at frequencies where the circumnavigation distance around the sphere is an integer multiple of the coupled wavelength [1,3]. At a WGM resonance, the circulating light coherently drives itself leading to greatly enhanced storage of light within the resonator. Whispering gallery mode resonators are emerging as useful platforms for the development of biosensors [1,3,5–7]. These approaches generally take advantage of the dependence of the WGM resonant wavelength with effective refractive index. For a resonator much larger than the optical wavelength, the resonant condition is given by [8,9]:
r =
2rneff m
(1)
Abbreviations: WGM, whispering gallery mode; Q factor, quality factor; TIR, total internal reflection; CA-125, cancer antigen-125; TNF-␣, tumor necrosis factor-␣. ∗ Corresponding author. Tel.: +1 785 864 4313. E-mail address: [email protected] (R.C. Dunn). 0925-4005/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2011.09.060
where r is the resonant wavelength, r is the resonator radius, neff is the effective refractive index surrounding the sphere, and m is an integer. This relationship shows that the WGM resonance wavelength is directly proportional to the effective refractive index, neff . For biosensing applications, recognition elements such as antibodies can be immobilized to the resonator surface which, upon binding of their associated antigen, alters the effective refractive index surrounding the resonator [10,11]. This leads to shifts in the WGM resonance which provides a label-free approach for biosensing. Using this general approach, WGM resonators have been utilized in the detection of a wide array of species from proteins [12–15] and DNA [16,17] to bacteria [18]. The biosensing performance of WGM resonators is strongly linked to the quality (Q) factor of the resonator [7,19–21]. A high Q factor is associated with a narrow resonance and an increased effective path length, both of which are highly desirable for optimal sensor performance. While microfabricated WGM resonators have Q-factors on the order of 104 [22], simple microsphere resonators such as high index glass spheres can have Q-factors approaching 109 [23]. The formation of spheres from melts results in an exquisitely smooth surface which reduces losses and results in ultrahigh Q factors [3,21,23]. In addition to the superior sensing capabilities associated with their high Q factors, microspheres are generally inexpensive, easy to fabricate, and available in a range of materials. However, difficulties in multiplexing the detection of analytes using microsphere resonators has limited their applications in sensing platforms and shifted focus to microfabricated
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
resonators. Microfabricated ring or disk resonators are, in general, more easily integrated into photonics and fluidics systems, have favorable scalability, and are amenable to multiplexed detection schemes [13,14,20,24]. However, as mentioned, the Q factors of these devices are generally orders of magnitude smaller than those of microsphere resonators which restricts their detection limits. Ideally, one wants to combine the unmatched Q factors of microspheres with the multiplexed detection capabilities of microfabricated devices. Here we present a whispering gallery mode imaging (WGMI) scheme that uses sensitive fluorescence imaging of microsphere resonances to provide multiplexed detection capabilities. The WGMI approach utilizes an inverted total internal reflection (TIR) microscope to evanescently couple light from a tunable diode laser into a field of microsphere resonators. The surface of each microsphere is labeled with a fluorescent dye which acts as a reporter for the WGM resonance. On resonance, a distinctive ring of fluorescence is observed as the WGM excites the surface attached dye. As the wavelength of the tunable diode laser is scanned, therefore, the WGM resonance of each microsphere in the field of view is measured through the enhanced fluorescent ring, which is imaged from above with an upright microscope. The feasibility of biosensing using this approach is demonstrated by detecting two biomarkers of ovarian cancer – cancer antigen-125 (CA-125) and tumor necrosis factor-alpha (TNF-␣) [25–30]. Monoclonal antibodies specific for CA-125 and TNF-␣ were attached to 38 m and 53 m glass microspheres, respectively. Using the relative intensity of an attached fluorescent marker to track changes in the resonant wavelength, linear calibration curves were measured for both antigens. For CA-125, we demonstrate detection limits more than threefold better than commercial ELISA kits. Using this approach, the identity of the detected analyte can be encoded in the sphere size and/or location, providing vastly improved multiplexing capabilities. Moreover, the chemistry associated with attaching antibodies to glass beads is well developed and all immobilization procedures are carried out before the spheres are deposited onto the substrate surface. Using this approach simplifies the chemistry associated with multiplexing the detection over other WGM techniques. This method, therefore, provides a sensitive and straightforward approach for multiplexed WGM detection that takes advantage of the superior optical properties of microsphere resonators.
1263
in a 5% APTES (v/v) solution in toluene and allowed to tumble overnight. Following amine modification, the spheres were rinsed once in toluene and triply washed in absolute ethanol. Each sphere set was then placed in a 6% glutaraldehyde (v/v) solution in PBS and tumbled for 2 h. The sphere sets were each triply washed in PBS then placed into corresponding antibody solutions in PBS to tumble for 2 h: the 38 m diameter spheres were incubated in 2 g/mL anti-CA125 solution while 53 m diameter microspheres where incubated in 2 g/mL anti-TNF-␣. Following incubation, each sphere solution was washed in PBS and labeled with the fluorescent dye, Alexa 633. Finally, each microsphere set was tumbled in a 5% BSA, 5% sucrose (w/v) solution in PBS for 1 h to block nonspecific adsorption sites. 2.3. WGM measurement For WGM analysis, spheres were dispersed onto a glass coverslip and placed in a CoverWell perfusion chamber containing PBS. Light was coupled into the spheres using the evanescent wave generated with an inverted total internal reflection fluorescence microscope (TIRF-M) (Olympus IX71, Center Valley, PA) equipped with a 60× objective (1.45 NA achromat, Olympus). The output from an external cavity diode laser (TLB-6904, New Focus, Santa Clara, CA) is offset on the back aperture of the objective to create total internal reflection at the substrate interface. The diode laser can be tuned from 632.86 to 633.12 nm with ≤300 kHz linewidth. Fluorescence is collected from above using an Olympus 10× UMPlanFl (0.3 NA) objective lens, filtered (Thorlabs, Newton, NJ), and imaged onto a CCD camera (Cascade 650, Roper Scientific, Tuscon, AR). The laser was scanned using LabVIEW (National Instruments, Austin, TX) software and image collection and processing was controlled using Slidebook software (v4.2.0.3, Intelligent Imaging Innovations, Denver, CO).
2. Materials and methods 2.1. Reagents Soda–lime glass microspheres were obtained from Mo-Sci Corporation (Rolla, MO). Anti-CA-125 mouse monoclonal antibodies (SPM111), 3-aminopropyltriethoxysilane (APTES), glutaraldehyde, bovine serum albumin (BSA), and all solvents were obtained from Fisher Scientific (Hampton, NH). CA-125 antigen was obtained from a commercially available CA-125 screening kit (Panomics BC-1013). Anti-TNF-␣ antibodies were utilized from a commercial TNF-␣ screening kit (ESS0001, Fisher Scientific), and TNF-␣ antigen was purchased from Pierce Biotechnology (Rockford, IL). The amine reactive succinimidyl ester of Alexa Fluor 633 was obtained from Invitrogen Corporation (Carlsbad, CA). CoverWell perfusion chambers were obtained from Grace Bio-Labs (Bend, OR). 2.2. Microsphere preparation The 38 m and 53 m diameter microspheres were separately sonicated in an alconox solution, triply rinsed with 18 M water, and rinsed with absolute ethanol. Each sphere set was placed
Fig. 1. Schematic of the apparatus used to couple WGM excitation with fluorescence imaging. The output of a tunable diode laser is coupled into an inverted total internal reflection microscope through a high numerical aperture (60×, 1.45 NA) objective. By offsetting the excitation beam in the back aperture of the objective, light approaches the sample beyond the critical angle which leads to total internal reflection (TIR) at the substrate surface. The associated evanescent field is used to excite WGMs in spheres which have been labeled with antibodies to detect CA-125 (38 m) and TNF-␣ (53 m). The spheres are also labeled with a fluorescent dye (stars) which acts as a marker of the WGM resonance. On resonance, the evanescent field from light circumnavigating the sphere excites fluorescence from the dye, which is collected from above and imaged onto a CCD.
1264
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
Fig. 2. (Top) Representative fluorescence images of a 38 m diameter microsphere labeled with Alexa 633. As the excitation wavelength is scanned, a distinctive ring of enhanced fluorescence is observed (middle panel) when a WGM resonance is encountered. (Below) Excitation spectrum of fluorescence intensity with excitation wavelength, extracted from a series of fluorescence images.
3. Results and discussion Here we report a general approach for combining WGM microsphere resonators with fluorescence imaging to develop a new approach for the multiplexed detection of analytes. The general approach is outlined schematically in Fig. 1. 3.1. Experimental design As shown in Fig. 1, light from a tunable diode laser with narrow spectral linewidth (≤300 kHz) is launched into an optical fiber and coupled into an inverted total internal reflection (TIR) microscope. The TIR microscope uses a high numerical aperture (1.45 NA) objective to couple light to the substrate surface beyond the critical angle, where total internal reflection takes place [31]. The associated evanescent field is used to launch light into the microsphere resonators on the substrate surface as shown schematically in Fig. 1. As the excitation wavelength of the diode laser is scanned, the WGM resonance from each microsphere in the field of view is detected through fluorescence imaging. This is accomplished by labeling all the microspheres with the same fluorescent dye (stars in Fig. 1) which acts as a reporter of the WGM resonance. When a particular microsphere reaches the resonance condition given
by Eq. (1), a bright ring of enhanced fluorescence is observed as the evanescent field from the WGM resonance excites the surface attached dye. By collecting the fluorescence from above and imaging it onto a CCD camera, the WGM resonance for each sphere in the field of view can be simultaneously measured as the excitation wavelength is scanned. A similar approach has been reported by utilizing quantum-dot embedded polystyrene microspheres excited at single points using low magnification objectives [32,33]. To illustrate this approach, Fig. 2 shows a series of fluorescence images taken of the same 38 m diameter dye labeled soda–lime glass microsphere as the excitation wavelength is scanned. When a WGM resonance is reached (center panel) a distinctive ring of enhanced fluorescence is observed as the evanescent field from the circumnavigating light excites the surface bound dye. With the current coupling arrangement and dye labeling scheme, we routinely measure 2–4-fold enhancements in the fluorescence intensity at resonance which makes them easily identifiable over the off resonance background fluorescence. The excitation spectrum in Fig. 2 plots the fluorescence intensity from the sphere as the excitation wavelength is scanned. This approach, therefore, uses the fluorescence marker to measure the WGM resonance wavelength. Since the WGM resonance is an intrinsic property of the system, this approach is not as susceptible to complications that can arise from
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
1265
Fig. 3. To explore the sensing capabilities of the WGM imaging approach, fluorescence was used to measure the resonance wavelength as the surrounding refractive index was systematically changed by adding aliquots of ethanol to water. A typical calibration plot measured for a 53 m sphere is shown above. As expected from Eq. (1), the plot of r versus neff is linear (R2 = 0.9931), with a measured sensitivity of ∼50 nm/RIU.
photobleaching or variable dye loading that are problematic for intensity based measurements. Typical Q-factors calculated from these spectra range from 2 × 104 to 5 × 104 , which are on the order of Q-factors measured for microfabricated resonators, but orders of magnitude lower than that expected for microspheres [22,23]. The dye attached to the surface of the microsphere acts as a source of loss in this arrangement which lowers the Q-factor. We are currently exploring optimal dye loading conditions to maximize the Q-factor. 3.2. Sensor calibration To demonstrate that the WGMI approach can track changes in WGM resonance wavelength, calibration plots were measured where the refractive index of the bathing solution was systematically changed. Fig. 3 plots a typical calibration curve measured using the fluorescence imaging approach. In these experiments, microsphere resonators were bathed in 18 M water to which injections of absolute ethanol were added to increase the surrounding refractive index [34]. As seen in Fig. 3, a linear response is observed as expected from the relationship between WGM resonance wavelength and refractive index (Eq. (1)). The particular calibration plot shown in Fig. 3 was measured from a microsphere with a Q-factor of ∼2 × 104 and has a measured sensitivity of ∼50 nm/RIU (refractive index units) and a detection limit of ∼2.6 × 10−5 RIU. Linear calibration plots such as that shown in Fig. 3 illustrates the feasibility of using microsphere resonators combined with fluorescence imaging for biosensing. 3.3. Ovarian cancer biomarker detection by WGMI Having demonstrated that the WGMI approach is sensitive to changes in the refractive index, the specific biosensing capabilities are demonstrated using antibodies specific for biomarkers of ovarian cancer. Two separate assays were performed with microspheres modified with antibodies specific for cancer antigen-125 (CA-125) and tumor necrosis factor-␣ (TNF-␣), both implicated in the pathogenesis of ovarian carcinoma [25–30]. As illustrated in Fig. 1, the antibodies are attached to the sphere surface along with the fluorescent dye. Moreover, each antibody is attached to a distinct sphere size, thus encoding the analyte identity into the sphere diameter which is easily measured in the imaging approach. Here monoclonal antibodies specific for CA-125 were immobilized on 38 m spheres while antibodies for TNF-␣ were attached to 53 m spheres.
Fig. 4. (A) Calibration plot of WGM resonance versus CA-125 (in Units/mL) measured using 38 m microsphere resonators modified with anti-CA-125 antibodies and the fluorescent dye Alexa 633. Binding of CA-125 to the immobilized antibody alters the effective refractive index, leading to shifts in the WGM resonance. The linear calibration plot (R2 = 0.9677) has a sensitivity of ∼6 pm/(U/mL). (B) Calibration plot of WGM resonance versus TNF-␣ (in ng/mL) measured using 53 m microsphere resonators modified with anti-TNF-␣ antibodies and the fluorescent dye Alexa 633. The linear calibration plot (R2 = 0.9798) has a sensitivity of ∼108 pm/ng/mL.
Fig. 4 plots the shifts in WGM resonant wavelength as CA-125 was injected into the perfusion chamber. A linear shift in the resonant wavelength is observed with a measured CA-125 sensitivity of ∼6 pm/U/mL and detection limit of ∼1.5 U/mL. This detection limit is threefold better than the 5 U/mL detection limit of commercial ELISA kits (Panomics BC-1013). Also shown in Fig. 4 is a similar calibration plot for the detection of TNF-␣ using 53 m spheres functionalized with monoclonal antibodies for TNF-␣. The linear response of TNF-␣ shown in Fig. 4 has sensitivity of ∼108 pm/ng/mL and a detection limit of ∼240 pg/mL. Interestingly, the sensitivity is approximately an order of magnitude higher than previous microresonator reports using other antibody-antigen sensing modalities. This may arise from differences in the specific KD values for the antibody–antigen pairs used in these studies [35–37]. However, the shorter excitation wavelength used here may also influence microresonator sensor performance through changes in mode volume and mode density. Experiments are underway to quantify these effects. The calibration plots shown in Fig. 4 demonstrate the specific detection of biomarkers using the WGMI approach. Coupling WGM resonators with sensitive fluorescence imaging offers many advantages for the multiplexed detection of biomarkers. The approach is label-free, sensitive and rapid. Since the analyte identity is encoded in the sphere size or location, every sphere can be labeled with the same dye and excited with the same laser system. Since the WGM resonance wavelength is an intrinsic property of the system, the assay does not suffer from complications arising from dye photobleaching or variable labeling. Microspheres are readily
1266
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
available, inexpensive, come in a range of materials and sizes, and have extremely large Q-factors which are desirable for sensing applications. Finally, their small size makes them compatible with many of the developments taking place in the lab-on-a-chip field, which can further reduce assay costs. 4. Conclusions Fluorescence imaging is combined with total internal reflection excitation of whispering gallery modes in microspheres to develop a new approach for multiplexed biosensing. In this approach, microspheres are functionalized with a fluorescent dye which acts as a reporter of the WGM resonance. Light is coupled into a field of microspheres using the evanescent field created in a total internal reflection microscope. As the excitation wavelength is scanned, WGM resonances are detected through fluorescence imaging. As a particular sphere reaches a WGM resonance, a distinct ring of enhanced fluorescence is observed as the evanescent wave from light circumnavigating the sphere excites the surface attached dye. This approach enables the simultaneous measurement of WGM resonances from every microsphere in the field of view, thus enabling the development of highly multiplexed biosensors. To demonstrate the feasibility of biosensing using the WGMI approach, antibodies specific for CA-125 and TNF-␣ were linked to 38 m and 53 m spheres, respectively. CA-125 and TNF-␣ are commonly used markers for the screening of ovarian cancer and by linking their antibodies to differentially sized microspheres, detection specificity is encoded into sphere diameter. Binding of their associated antigens shifts the WGM resonance of the sphere, which is detected using fluorescence imaging. Calibration curves measured in separate assays reveals a linear response for both antigens with the detection limits for CA-125 over threefold better than that obtained with commercial ELISA approaches. These results confirm that fluorescence imaging can be used to simultaneously detect WGMs from a field of microsphere resonators where analyte identity can be encoded into sphere size and/or location, thus enabling significant multiplexing capabilities. Acknowledgments We gratefully acknowledge support from NSF (CBET 1133814) and the Madison and Lila Self Foundation. HAH is grateful to Kevin Armendariz for assistance with LabVIEW programming. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.snb.2011.09.060. References [1] S. Arnold, R. Ramjit, D. Keng, V. Kolchenko, I. Teraoka, Microparticle photophysics illuminates viral bio-sensing, Faraday Discuss. 137 (2008) 99–113, 65-83. [2] K.J. Vahala, Optical microcavities, Nature 424 (2003) 839–846. [3] S. Soria, S. Berneschi, M. Brenci, F. Cosi, G.N. Conti, S. Pelli, G.C. Righini, Optical microspherical resonators for biomedical sensing, Sensors 11 (2011) 785–805. [4] I.M. White, H. Zhu, J.D. Suter, N.M. Hanumegowda, H. Oveys, M. Zourob, X. Fan, Refractometric sensors for lab-on-a-chip based on optical ring resonators, IEEE Sens. J. 7 (2007) 28–35. [5] J.D. Suter, X. Fan, Overview of the optofluidic ring resonator: a versatile platform for label-free biological and chemical sensing, in: EMBC 2009, Ann. Int. Conf. IEEE (2009) 1042–1044. [6] S. Wang, A. Ramachandran, J. Clarke, S.J. Ja, D. Goad, L. Wald, E.M. Flood, E. Knobbe, J.V. Hryniewicz, S.T. Chu, D. Gill, W. Chen, O. King, B.E. Little, A universal biosensing platform based on optical micro-ring resonators, Biosens. Bioelectron. 23 (2008) 939–944. [7] F. Vollmer, S. Arnold, Whispering-gallery-mode biosensing: label-free detection down to single molecules, Nat. Methods 5 (2008) 591–596.
[8] M.L. Gorodetsky, V.S. Ilchenko, Optical microsphere resonators: optimal coupling to high-Q whispering-gallery modes, J. Opt. Soc. Am. B: Opt. Phys. 16 (1999) 147–154. [9] J.C. Knight, G. Cheung, F. Jacques, T.A. Birks, Phase-matched excitation of whispering-gallery-mode resonances by a fiber taper, Opt. Lett. 22 (1997) 1129–1131. [10] I. Teraoka, S. Arnold, F. Vollmer, Perturbation approach to resonance shifts of whispering-gallery modes in a dielectric microsphere as a probe of a surrounding medium, J. Opt. Soc. Am. B 20 (2003) 1937–1946. [11] F. Vollmer, D. Braun, A. Libchaber, M. Khoshsima, I. Teraoka, S. Arnold, Protein detection by optical shift of a resonant microcavity, Appl. Phys. Lett. 80 (2002) 4057–4059. [12] H. Zhu, J. Suter, I. White, X. Fan, Aptamer based microsphere biosensor for thrombin detection, Sensors 6 (2006) 785–795. [13] A.L. Washburn, L.C. Gunn, R.C. Bailey, Label-free quantitation of a cancer biomarker in complex media using silicon photonic microring resonators, Anal. Chem. 81 (2009) 9499–9506. [14] A.L. Washburn, M.S. Luchansky, A.L. Bowman, R.C. Bailey, Quantitative, labelfree detection of five protein biomarkers using multiplexed arrays of silicon photonic microring resonators, Anal. Chem. 82 (2009) 69–72. [15] H.Y. Zhu, P.S. Dale, C.W. Caldwell, X.D. Fan, Rapid and label-free detection of breast cancer biomarker CA15-3 in clinical human serum samples with optofluidic ring resonator sensors, Anal. Chem. 81 (2009) 9858–9865. [16] X.D. Fan, J.D. Suter, I.M. White, H.Y. Zhu, H.D. Shi, C.W. Caldwell, Label-free quantitative DNA detection using the liquid core optical ring resonator, Biosens. Bioelectron. 23 (2008) 1003–1009. [17] F. Vollmer, S. Arnold, D. Braun, I. Teraoka, A. Libchaber, Multiplexed DNA quantification by spectroscopic shift of two microsphere cavities, Biophys. J. 85 (2003) 1974–1979. [18] H.-C. Ren, F. Vollmer, S. Arnold, A. Libchaber, High-Q microsphere biosensor—analysis for adsorption of rodlike bacteria, Opt. Express 15 (2007) 17410–17423. [19] A.M. Armani, R.P. Kulkarni, S.E. Fraser, R.C. Flagan, K.J. Vahala, Label-free, singlemolecule detection with optical microcavities, Science 317 (2007) 783–787. [20] I.M. White, X. Fan, On the performance quantification of resonant refractive index sensors, Opt. Express 16 (2008) 1020–1028. [21] N.M. Hanumegowda, C.J. Stica, B.C. Patel, I. White, X. Fan, Refractometric sensors based on microsphere resonators, Appl. Phys. Lett. 87 (2005) 201107. [22] X. Fan, I.M. White, H. Zhu, J.D. Suter, H. Oveys, Overview of novel integrated optical ring resonator bio/chemical sensors, in: A.V. Kudryashov, A.H. Paxton, V.S. Ilchenko (Eds.), SPIE, San Jose, CA, USA, 2007, pp. 64520M–64520. [23] M.L. Gorodetsky, A.A. Savchenkov, V.S. Ilchenko, Ultimate Q of optical microsphere resonators, Opt. Lett. 21 (1996) 453–455. [24] D.X. Xu, A. Densmore, A. Delâge, P. Waldron, R. McKinnon, S. Janz, J. Lapointe, G. Lopinski, T. Mischki, E. Post, P. Cheben, J.H. Schmid, Folded cavity SOI microring sensors for high sensitivity and real time measurement of biomolecular binding, Opt. Express 16 (2008) 15137–15148. [25] C. Tian, M. Markman, R. Zaino, R.F. Ozols, W.P. McGuire, F.M. Muggia, P.G. Rose, D. Spriggs, D.K. Armstrong, CA-125 change after chemotherapy in prediction of treatment outcome among advanced mucinous and clear cell epithelial ovarian cancers: a gynecologic oncology group study, Cancer 115 (2009) 1395–1403. [26] P. Kenemans, C.A. Yedema, G.G. Bon, S. von Mensdorff-Pouilly, CA 125 in gynecological pathology—a review, Eur. J. Obstet. Gynecol. Reprod. Biol. 49 (1993) 115–124. [27] B. Dobrzycka, S.J. Terlikowski, O. Kowalczuk, M. Kinalski, Circulating levels of TNF-alpha and its soluble receptors in the plasma of patients with epithelial ovarian cancer, Eur. Cytokine Netw. 20 (2009) 131–134. [28] D. Gupta, C.G. Lis, Role of CA125 in predicting ovarian cancer survival—a review of the epidemiological literature, J. Ovarian Res. 2 (2009) 13. [29] B.B. Hermsen, S. von Mensdorff-Pouilly, J. Berkhof, P.J. van Diest, J.J. Gille, F.H. Menko, M.A. Blankenstein, P. Kenemans, R.H. Verheijen, Serum CA-125 in relation to adnexal dysplasia and cancer in women at hereditary high risk of ovarian cancer, J. Clin. Oncol. 25 (2007) 1383–1389. [30] B. Dobrzycka, S.J. Terlikowski, M. Garbowicz, W. Niklinska, P.S. Bernaczyk, J. Niklinski, M. Kinalski, L. Chyczewski, Tumor necrosis factor-alpha and its receptors in epithelial ovarian cancer, Folia Histochem. Cytobiol. 47 (2009) 609–613. [31] D. Axelrod, Total internal reflection fluorescence microscopy, in: J.C. John, H. William Detrich III (Eds.), Methods in Cell Biology, Academic Press, 2008, pp. 169–221. [32] H. Beier, G. Coté, K. Meissner, Whispering gallery mode biosensors consisting of quantum dot-embedded microspheres, Ann. Biomed. Eng. 37 (2009) 1974–1983. [33] S. Pang, R.E. Beckham, K.E. Meissner, Quantum dot-embedded microspheres for remote refractive index sensing, Appl. Phys. Lett. 92 (2008) 221108. [34] D.R. Lide, CRC Handbook of Chemistry and Physics, CRC Press, Boca Raton, FL, 2011. [35] J. Maynard, G. Georgiou, Antibody engineering, Annu. Rev. Biomed. Eng. 2 (2000) 339–376. [36] P. Rathanaswami, J. Babcook, M. Gallo, High-affinity binding measurements of antibodies to cell-surface-expressed antigens, Anal. Biochem. 373 (2008) 52–60. [37] G.Q. Liu, Y. Liu, W. Zhang, X. Yu, H.W. Zhang, R. Zhao, D. Shangguan, Y. Li, B.F. Shen, Quartz crystal biosensor for real-time kinetic analysis of interaction between human TNF-alpha and monoclonal antibodies, Sens. Actuators B 99 (2004) 416–424.
H.A. Huckabay, R.C. Dunn / Sensors and Actuators B 160 (2011) 1262–1267
Biographies Heath A. Huckabay is a graduate student in Department of Chemistry at University of Kansas. His research interests include optical sensor development and spectroscopic imaging. He received his B.Sc. in Forensic Chemistry in 2006 from Sam Houston State University in Huntsville, Texas.
1267
Robert C. Dunn is a professor of chemistry at University of Kansas and member of the Ralph N. Adams Institute for Bioanalytical Chemistry. His research specializes in using novel optical approaches for analyzing samples with high spatial resolution and low detection limits.