A host-guest complex of cucurbit[7]uril and uric acid was formed in an aqueous solution before adding a small amount into Au NP solution for quantitative surface-enhanced Raman spectroscopy (SERS) sensing using a modular spectrometer.
This work describes a rapid and highly sensitive method for the quantitative detection of an important biomarker, uric acid (UA), via surface-enhanced Raman spectroscopy (SERS) with a low detection limit of ~0.2 μM for multiple characteristic peaks in the fingerprint region, using a modular spectrometer. This biosensing scheme is mediated by the host-guest complexation between a macrocycle, cucurbit[7]uril (CB7), and UA, and the subsequent formation of precise plasmonic nanojunctions within the self-assembled Au NP: CB7 nanoaggregates. A facile Au NP synthesis of desirable sizes for SERS substrates has also been performed based on the classical citrate-reduction approach with an option to be facilitated using a lab-built automated synthesizer. This protocol can be readily extended to multiplexed detection of biomarkers in body fluids for clinical applications.
Uric acid, which is the end product of metabolism of purine nucleotides, is an important biomarker in blood serum and urine for the diagnosis of diseases such as gout, preeclampsia, renal diseases, hypertension, cardiovascular diseases and diabetes1,2,3,4,5. Current methods for uric acid detection include colorimetric enzymatic assays, high performance liquid chromatography and capillary electrophoresis, which are time-consuming, expensive and require sophisticated sample preparation6,7,8,9.
Surface-enhanced Raman spectroscopy is a promising technique for routine point-of-care diagnosis as it allows selective detection of biomolecules via their vibration fingerprints and offers numerous advantages such as high-sensitivity, rapid response, ease of use and no or minimal sample preparation. SERS substrates based on noble metal nanoparticles (e.g., Au NPs) can enhance the Raman signals of the analyte molecules by 4 to 10 orders of magnitude10 via strong electromagnetic enhancement caused by surface plasmon resonance11. Au NPs of tailored sizes can be easily synthesized as opposed to the time-consuming fabrication of complex metal nanocomposites12, and thus are widely used in biomedical applications owing to their superior properties13,14,15,16. Attachment of macrocyclic molecules, cucurbit[n]urils (CBn, where n = 5-8, 10), onto the surface of Au NPs can further enhance the SERS signals of the analyte molecules as the highly symmetric and rigid CB molecules can control the precise spacing between the Au NPs and localize the analyte molecules at the center or in close proximity to the plasmonic hotspots via formation of host-guest complexes (Figure 1)17,18,19,20. Previous examples of SERS studies using Au NP: CBn nanoaggregates include nitroexplosives, polycyclic aromatics, diaminostilbene, neurotransmitters and creatinine21,22,23,24,25, with the SERS measurements either being performed in a cuvette or by loading a small droplet onto a custom-made sample holder. This detection scheme is particularly useful to rapidly quantify biomarkers in a complex matrix with a high reproducibility.
Herein, a facile method to form host-guest complexes of CB7 and an important biomarker UA, and to quantify UA with a detection limit of 0.2 μM via CB7-mediated aggregations of Au NPs in aqueous media was demonstrated using a modular spectrometer, which is promising for diagnostic and clinical applications.
1. Synthesis of Au NPs
2. Characterization of Au NPs
3. Formation of CB7-UA complexes
4. SERS sensing of UA
5. Data analysis
In the presented Au NP synthesis, the UV-Vis spectra show a shift of the LSPR peaks from 521 nm to 529 nm after 10 growing steps (Figure 4A,B) while the DLS data shows a narrow size distribution as the size of Au NPs increase from 25.9 nm to 42.8 nm (Figure 4C,D). The average sizes of G0, G5 and G10 measured from TEM images (Figure 4E) are 20.1 ± 2.1 nm, 32.5 ± 2.3 nm and 40.0 ± 2.2 nm respectively, with 200 particles counted in each case. These results indicate this protocol is effective in synthesizing uniform and narrowly dispersed Au NPs.
In the presented SERS studies, host-guest complexes of CB7 and UA were formed with empty CB7 mediating the formation of precise plasmonic nanojunctions within the Au NP: CB7 nanoaggregates, as supported by the characteristic UA signals in the SERS spectrum (Figure 5A).
The assignments for the Raman peaks of CB (marked by +) and UA (marked by *) are shown in Table 2. Conversely, no SERS signals of UA can be observed in the absence of CB7, illustrating the key role of CB7 in triggering the aggregation of Au NPs.
A constant CB7 concentration of 20 μM was used in the SERS titration of UA throughout so as to ensure the in situ formation of reproducible plasmonic nanostructures (i.e., SERS substrates). The high sensitivity of the detection scheme presented in this protocol was demonstrated by the observation of clear SERS signals from the UA peaks at 640 cm-1 and 1130 cm-1 (attributed to skeletal ring deformation and C-N vibration respectively) down to ~0.2 μM (Figure 5B−D), which is known as the detection limit. In addition, very strong correlations (R2 > 0.98) between the SERS intensity and log concentration of UA were obtained by power law for both peaks, with linear regions found in the range of 0.2 to 2 μM (Figure 5E,F). It should be noted that linear correlations between the SERS intensity and log concentration can be approximated for a narrow range of analyte concentrations, whereas the SERS signal approaches 0 when the log concentration approaches negative infinity (i.e., the analyte concentration approaches 0), as observed in our data. The SERS signals are also highly reproducible as evidenced by the small error bars shown in Figure 5E,F.
Figure 1: Schematic illustration of the precise plasmonic nanojunctions within self-assembled Au NP: CB7 nanoaggregates. Inset shows a zoom-in of the plasmonic nanojunctions where the aggregation is mediated by empty CB7 while UA is enriched on the surface of Au NPs via host-guest complexation. It is noted that the scheme is not drawn to scale. Please click here to view a larger version of this figure.
Figure 2: (a) Schematic illustration and (b) photograph of the automated Au NP synthesizer. Please click here to view a larger version of this figure.
Figure 3: Schematic illustration of the Raman system. Please click here to view a larger version of this figure.
Figure 4: Representative characterization of Au NPs. (A) UV-Vis spectra of Au NPs and (B) zoom-in spectra showing the shifting of the LSPR peaks as the number of growing steps increases to 10. (C) Hydrodynamic size of Au NPs and (D) corresponding plot of particle size as a function of number of growing steps. (E) TEM images of Au NPs, showing sizes of Au seeds and Au NPs after 5 and 10 growing steps. Please click here to view a larger version of this figure.
Figure 5: Representative SERS results of UA detection within Au NP: CB7 nanoaggregates. (A) SERS spectra of UA in the presence or absence of CB7. Raman peaks of CB7 and UA are marked by + and * respectively. (B) Full-range, (C) 600 – 700 cm-1 zoom-in and (D) 1100 – 1180 cm-1 zoom-in SERS spectra of UA with concentrations from 0 to 20 μM. Main Raman peaks of UA are marked by *. Spectra were baseline corrected and offset for clarity. (E,F) Corresponding plots of the SERS peak intensity against concentration of UA. Please click here to view a larger version of this figure.
Conc. of UA stock solution (µM) | Vol. of UA stock solution added (mL) | Vol. of water added (mL) | Conc. of new UA stock solution (µM) |
400 | 5 | 5 | 200 |
200 | 5 | 5 | 100 |
100 | 4 | 6 | 40 |
40 | 5 | 5 | 20 |
20 | 5 | 5 | 10 |
10 | 4 | 6 | 4 |
4 | 5 | 5 | 2 |
Table 1: Sequential dilutions of UA solution.
CB7 | UA | ||
SERS peak (cm-1) | Peak assignment | SERS peak (cm-1) | Peak assignment |
446 | Ring scissor mode | 491 | C-N-C ring vibration |
831 | Ring deformation | 640 | Skeletal ring deformation |
1375 | Symmetric C-N stretching | 896 | N-H bending |
1420 | Asymmetric C-N stretching | 1020 | Ring vibration |
– | – | 1130 | C-N vibration |
– | – | 1202 | N-C-C stretching and bending |
Table 2: Assignments for the Raman peaks of CB7 and UA2,4,29.
The automated synthesis method described in the protocol allows Au NPs of increasing sizes to be reproducibly synthesized. Although there are some elements that still need to be carried out manually, such as the fast addition of sodium citrate during the seed synthesis and checking periodically to ensure that the PEEK tubing is secure, this method allows Au NPs of large sizes (up to 40 nm), which would usually require multiple manual injections of HAuCl4 and sodium citrate, to be synthesized via continuous addition over a long period of time.
Further characterization can be performed to elucidate the fundamental property of the CB complexes. For instance, the formation of host-guest complexes can typically be confirmed using 1H nuclear magnetic resonance (NMR), which should show upfield shift and broadening of signals in case of complexation21,22,25. Yet 1H NMR is not applicable to UA due to its lack of non-exchangeable protons. Alternative techniques such as 13C NMR and mass spectrometry could also be employed to characterize the complexation. Binding constants between CB7 and UA can be measured using titration techniques, such as UV-Vis spectroscopy titration and isothermal titration calorimetry (ITC)21,22,25. Meanwhile molecular modelling based on force-field and density functional theory (DFT) models can be computed to obtain theoretical insights into the binding geometry of the host-guest complexes21,22,25,29. Moreover IR and Raman spectra can be computed by frequencies calculations21,25,29.
SERS is a highly sensitive and selective analytical technique which allows identification of trace analytes via their molecule-specific vibrational fingerprints. SERS is gaining interests across different science disciplines, in particular biomedical studies, due to its greatly enhanced signals, much shorter acquisition time and high tolerance to liquid water (suitable for sensing in biofluids)30,31,32,33,34,35. In contrast to previous reports on UA sensing1,2,3,4,36,37, the rigid structure of CB7 defines precise spacing of 0.9 nm between Au NPs via carbonyl portal binding while the surface-bound CB7 can trap UA molecules within its cavity (Figure 1), resulting in strong and localized plasmonic hotspots, and hence the highly sensitive (down to ~0.2 μM) and reproducible (within 2% error) SERS signals of UA with very strong correlations (R2 > 0.98) between the SERS intensity and log concentration (Figure 5).
In an attempt to optimize the concentration of CB7, we note that 20 μM CB7 was used to ensure the formation of reproducible SERS substrates. In particular, the absolute concentration of CB7 used is dependent on the overall system (i.e., Au NPs, analytes and background molecules, if any)18,22. A higher concentration of CB7 should be used if the aggregation of Au NPs is too slow. Conversely, a lower concentration of CB7 should be used if the sample solution precipitates quickly and leads to shorter measurement windows. The aggregation of Au NPs mediated by CB7 in our experimental setting is expected to follow the diffusion-limited colloidal aggregation (DLCA) kinetics19, in which open and elongated chain-like structures were rapidly formed initially before joining together as quasi-fractal network. DLCA kinetics typically occurs at a high CB: Au NP ratio (by number), which is equal to 106:1 in our case. It should be noted that uric acid is present in bodily fluids (e.g., blood serum, urine) at a higher concentration. For instance, the normal concentration of uric acid is 3.5 – 7.0 mg/dL in blood serum38 and 16 – 100 mg/dL in urine2 respectively (concentration above or below the normal concentration is known as hyperuricemia and hypouricemia)39. Therefore, only a very small amount of sample is needed for biomarker detection in this highly sensitive scheme where a high dilution factor is used to lower the concentration of the sample to a suitable range. This is particularly important for point-of-care monitoring of terminally ill patients whose urine excretion is very low. Highly diluted samples result in larger sample volumes and thus reduce errors in the quantification of biomarkers due to water evaporation and loss of samples due to liquid transfer, while giving other advantages including minimizing the matrix effects25. Due to the selective nature of this probing method, it is limited to analyte molecules that can form host-guest complexes with CB. It should be noted it is possible to observe interferences from other molecules because CB can bind to different guest molecules. Nevertheless, sample purification such as gel electrophoresis and HPLC can be performed prior to SERS measurement.
The detection scheme demonstrated in this protocol has the potential for multiplexed detection of biomarkers in a complex matrix for clinical applications when it is coupled to advanced data analysis techniques.
The authors have nothing to disclose.
TCL is grateful to the support from the Royal Society Research Grant 2016 R1 (RG150551) and the UCL BEAMS Future Leader Award funded through the Institutional Sponsorship award by the EPSRC (EP/P511262/1). WIKC, TCL and IPP are grateful to the Studentship funded by the A*STAR-UCL Research Attachment Programme through the EPSRC M3S CDT (EP/L015862/1). GD and TJ would like to thank the EPSRC M3S CDT (EP/L015862/1) for sponsoring their studentship. TJ and TCL acknowledge Camtech Innovations for contribution to TJ’s studentship. All authors are grateful to the UCL Open Access Fund.
40 nm gold nanoparticles | NanoComposix | AUCN40-100M | NanoXact, 0.05 mg/ mL, bare (citrate) |
Centrifuge tube | Corning Falcon | 14-432-22 | 50 mL volume |
Cucurbit[7]uril | Lab-made | see ref. 19 | |
Gold(III) chloride trihydrate | Sigma aldrich | 520918 | ≥99.9% trace metals basis |
Luer lock disposable syringe | Cole-Parmer | WZ-07945-15 | 3 mL volume |
Luer-to-MicroTight adapter | LuerTight | P-662 | 360 μm outer diameter Tubing to Luer Syringe |
PEEK tubing | IDEX | 1572 | 360 μm outer diameter, 150 μm inner diameter |
PEEK tubing cutter | IDEX | WZ-02013-30 | Capillary Polymer Chromatography Tubing Cutter For 360 µm to 1/32" OD tubing |
Raman spectrometer | Ocean Optics | QE pro | |
Sodium citrate tribasic dihydrate | Sigma aldrich | S4641 | ACS reagent, ≥99.0% |
Sonicator | |||
Standard Probe | Digi-Sense | WZ-08516-55 | Type-K |
Syringe pump | Aladdin | ALADDIN2-220 | 2 syringes, maximum syringe volume 60 mL |
Thermocouple thermometer | Digi-Sense | WZ-20250-91 | Single-Input Thermocouple Thermometer with NIST-Traceable Calibration |
ThermoMixer | Eppendorf | 5382000031 | With an Eppendorf SmartBlock for 50 mL tubes |
Uric acid | Sigma aldrich | U2625 | ≥99%, crystalline |