Open Access
Add to BibSonomy
Abstract
Due to the global challenge of donor kidney shortage, expanding the pool of deceased donors has been proposed to include expanded criteria donors. However, the lack of methods to precisely measure donor kidney injury and predict the outcome still leads to high discard rates and recipient complications. As such, evaluation of deceased donor kidney quality is critical prior to transplantation. Biomarkers from donor urine or serum provide potential advantages for the precise measure of kidney quality. Herein, simultaneous detection of secretory leukocyte peptidase inhibitor (SLPI) and interleukin 18 (IL-18), two important kidney injury biomarkers, has been achieved, for the first time, with an ultra-high sensitivity using surface enhanced Raman scattering (SERS). Specifically, black phosphorus/gold (BP/Au) nanohybrids synthesized by depositing Au nanoparticles (NPs) onto the BP nanosheets serve as SERS-active substrates, which offer a high-density of inherent and accessible hot-spots. Meanwhile, the nanohybrids possess biocompatible surfaces for the enrichment of target biomarkers through the affinity with BP nanosheets. Quantitative detection of SLPI and IL-18 were then achieved by characterizing SERS signals of these two biomarkers. The results indicate high sensitivity and excellent reproducibility of this method. The limits of detection reach down to 1.53×10−8 mg/mL for SLPI and 0.23×10−8 mg/mL for IL-18. The limits of quantification are 5.10×10−8 mg/mL and 7.67×10−9 mg/mL for SLPI and IL-18. In addition, simultaneous detection of these biomarkers in serum was investigated, which proves the feasibility in biologic environment. More importantly, this method is powerful for detecting multiple analytes inheriting from excellent multiplexing ability of SERS. Giving that the combined assessment of SLPI and IL-18 expression level serves as an indicator of donor kidney quality and can be rapidly and reproducibly conducted, this SERS-based method holds great prospective in clinical practice.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Kidney transplantation, offering improved survival and quality of life in contrast to dialysis, has been the optimal treatment for patients with end stage renal disease (ESRD). Unfortunately, there is an enormous supply and demand imbalance for donor organs, which greatly restricts the access to kidney transplantation. According to the report, the ratio between the transplant recipients and the patients on the waiting list is only 1: 30 in China [1]. Meanwhile, kidney discard rates are found to be relatively high because of the procurement of low-quality kidneys. These facts drive the attempts to expand the criteria for donated kidneys that are potentially viable, and utilize kidneys from donation after deceased donors with acute kidney injury (AKI). However, donors with AKI may have inferior transplant outcomes in comparison with donors without AKI, such as an increased risk of delayed graft function (DGF). Many clinicians and patients are thus conservative to accept kidneys from donors with AKI. As such, there is an urgent need to objectively identify the risk factors of a donor kidney and evaluate its quality, in terms of the acceptance and rejection of an organ.
Currently, quality of donor kidney is still challenging to be identified due to the lack of precise measures. The most commonly used methods include clinical scores, kidney biopsy histopathology and machine perfusion systems [2]. However, clinical scoring systems suffer from poor accuracy overall, rendering unnecessary organ discard. Machine perfusion parameters are not suitable to serve as standalone criteria due to the relatively weak association with kidney quality, while kidney biopsy is invasive. Consequently, the development of new methods to assess donor kidney quality is of significant importance. Lately, biomarkers in blood serum and urine have been increasingly recognized to be related with kidney injury [3]. Assessing kidney injury in deceased donors prior to transplant by virtue of these biomarkers, would provide information in terms of acceptance and allocation of an organ. Biomarkers, such as, neutrophil gelatinase–associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), liver-type fatty acid binding protein (L-FABP) and IL-18, have been identified to be the most important biomarkers of AKI [4,5]. Especially, NGAL and IL-18 are deemed to be useful for detecting AKI in the early period. Notably, there is no perfect biomarker due to the heterogeneous etiology of AKI and each of these identified biomarkers has its strengths and weaknesses. Recently, SLPI was reported to be upregulated in human kidney biopsies with AKI and, thus, it can be a newly promising biomarker [6–8]. More importantly, due to the heterogeneous etiology, combined measurement of biomarker panels may be a better idea for characterizing AKI in contrast to single marker. In consequence, the combined evaluation of an identified marker IL-18, with new biomarkers, such as SLPI, may further improve the measure accuracy of donor kidney quality.
Currently, there is not a gold standard method for the detection of biomarkers involved in donor kidney injury. Reported detection techniques of AKI related biomarkers include fluorescence spectroscopy [9], ELISA [10,11], electrochemistry [12] and liquid chromatography [13]. The comparison study of these methods in terms of sensitivity, reliability and simplicity was shown in Supplement 1, Table S1. Whereas, these approaches are still confronted with some shortcomings, such as limited sensitivity, multiplexing ability, complicated sample preparation or expensive cost. Surface enhanced Raman scattering (SERS), a vibrational spectroscopy with ultra-sensitivity that can reach down to single molecular detection level, is emerging as another option [14–16]. Meanwhile, SERS exhibits unique and rich spectral bands with narrow widths of ∼1 nm, which is inherently superior for multiplex analysis. SERS has been widely utilized for the sensitive detection of nucleic acids [17,18], proteins [19–22], and cells [23–25], typically by employing gold or silver nanoparticles (NPs) as SERS generators because of their strong localized surface plasmon resonance (LSPR) absorption with high extinction coefficient in the visible range. However, individual NPs suffer from relatively faint Raman vibration enhancement, thus suppressing detection sensitivity. In contrast, metallic NPs assembled in a well-defined manner are capable of stronger signal enhancement due to the enhanced electromagnetic fields, namely hot-spots, at the nanoscale interparticle gaps between adjacent NPs through LSPR coupling. Although many efforts have been devoted to the study of self-assembly of metallic NPs, it remains a major challenge to fabricate controllable and reproducible hot-spots. Up to now, a substrate-supported self-assembly strategy has drawn much attention owing to the easy operation and highly accessible surface hot-spots. Silica, nanowires, carbon nanotubes, graphene are commonly used as support materials to accommodate metallic NPs.
Lately, black phosphorus (BP), as a new type of two-dimensional (2D) nanomaterial, has attracted a growing interest in the electronic and optoelectronic fields due to its layer-dependent band gap, high carrier mobility, and peculiar in-plane anisotropy. BP nanosheets also exhibit excellent optical and biocompatible properties, offering new fascinating opportunities for biomedical applications [26,27]. Previous reports highlighted the utilization of BP nanosheets for efficient photothermal therapy as well as photoacoustic bioimaging owing to its broad absorption across the ultraviolet, visible, and near-infrared (NIR) regions. Additionally, giving that 2D materials have been burgeoningly employed as the supports for NPs, the assembly of metal NPs on BP nanosheets, yielding BP/metal nanohybrids, can serve as highly efficient SERS substrates. The SERS activity of BP/metal nanohybrids can be attributed to the hot-spots at the interstices among metal NPs as well as chemical enhancement from the charge transfer between BP nanosheets and molecules. Meanwhile, SERS effect is highly distance dependent, thus requiring analyte molecules to be adsorbed or extremely close (within a few nanometers) to a SERS-active surface. Attractively, BP nanosheets exhibits a high affinity toward the biomolecules through the electrostatic interactions, dispersion interactions, hydrophobic interactions and π-π stacking interactions, which is beneficial for the surface enrichment of proteins and thus localizing them in hot-spots [28,29]. As such, these fascinating features make the nanohybrids of BP nanosheets and metal NPs promising SERS active substrates for sensitive and multiple biomolecule detection. So far, only several works have performed the combination of BP nanosheets with metal NPs to enhance photothermal, photodynamic, and antibacterial activities, and to serve as nanoprobes for SERS imaging of live cells. While BP-based nanohybrids for SERS biosensing has not yet been a focus.
Herein, AKI related biomarkers SLPI and IL-18 was simultaneously evaluated for the first time using a label-free SERS technique as illustrated in Fig. 1. Au NPs were attached on the surfaces of BP nanosheets through an in-situ reduction process, yielding the BP/Au nanohybrids. Such nanohybrids were further utilized as SERS substrates for the detection SLPI and IL-18 both in buffered solution as well as blood serum. This method presents three distinct advantages over conventional detection methods: (1) BP/Au nanohybrids display high and stable SERS activity, which are crucial for sensitive, quantitative, and reliable detection; (2) Affinity of BP facilitates surface enrichment of biomarkers on nanohybrids, which is easy and time-saving; (3) SERS fingerprints is readily to detect multiple biomarkers through one single measurement.
Fig. 1. Schematic illustration of SERS detection of biomarkers SLPI and IL-18 using BP/Au nanohybrids. (The dashed line indicates the characteristic band at 1000 cm−1 corresponding to SLPI. While IL-18 shows no obvious band at 1000 cm−1)
2. Experimental section
2.1 Materials
All Chemicals and reagents used were analytical grade and deionized water (18.25 M Ω·cm) was used in all experiments. Chloroauric acid tetrahydrate (HAuCl4·4H2O), sodium citrate dihydrate (C6H5Na3O7·2H2O ≥ 99.0%) were purchased from MACKLIN (Shanghai, China). Black phosphorus nanosheets dispersion (4 mg/mL, 1–10 layers) was obtained from XFNANO Materials Tech Co. Ltd. (Nanjing, China) and stored in a dark N2-filled box. R6G was purchased from Tianjin Guangfu Fine Chemical Research Institute. Fetal bovine serum (FBS, 10%) was obtained from Hangzhou Sijiqing Biological Engineering Materials Co., Ltd. (Hangzhou, Zhejiang, China). Tris-HCl buffer (0.1 M, pH=8.0) in Qingdao Jieshikang Biotechnology Co., Ltd. SLPI and IL-18 were provided by Changhai hospital and analyzed by Wuhan Seville Biotechnology Co., Ltd.
2.2 Synthesis of SERS substrates
2.2.1 Synthesis of Au NPs
20 mL of deionized water was added to a conical flask and heated to be boiling under continuously stirring. Then, 600 µL of 1% HAuCl4 and 1.2 mL of 1% trisodium citrate solution were added in turn, and the mixture was kept boiling for 18 min with stirring. Afterwards, the heat was turned off and the mixture was continuously stirred until it cooled to the room temperature. The obtained sample was kept at 4 °C in a refrigerator.
2.2.2 Synthesis of BP/Au nanohybrids
BP/Au nanohybrids were synthesized through an in-situ reduction method. Specifically, 600 µL of HAuCl4 (1 wt%) was added to 20 mL of boiling water. Afterwards, 80 µL of 4 mg/mL BP nanosheets solution was first ultrasonically shaken for 5 min and then added drop by drop every 3 seconds. Next, 1.2 mL of trisodium citrate solution (1 wt%) was added and the mixture was stirred for 2 h. Finally, the product was obtained through the centrifugation at 9000 rpm for 20 min. The obtained product was dispersed in deionized water and kept at 4 °C in the dark.
2.3 SERS-based detection
For the SERS detection of biomarkers, 5 µL of SLPI and IL-18 solutions with different concentrations were added to 5 µL of BP/Au nanohybrids. The mixtures were then subject to SERS requirement. The SERS spectra were obtained at the excitation of 638 nm. For each sample, SERS spectra from ten random locations were measured to get an averaged spectrum.
2.4 Instruments
The UV-vis absorbance spectra were recorded with an Evolution 350 UV-vis (Thermo Scientific, China). The surface morphology and the size of the prepared BP/Au nanohybrids were characterized by a JEM-2100F (JEOL, Japan) high-resolution transmission electron microscope (TEM) at an acceleration voltage of 200 kV. The size distribution was measured with the dynamic light scattering experiments (Zetasizer, Nano-ZS90, Malvern, UK). SERS spectra were recorded by a microscopic Raman spectrometer (XploRA PLUS, HORIBA) with an incident laser wavelength of 638 nm and 10x objective lens.
3. Results and discussion
3.1 Characterization of Au NPs and BP/Au nanohybrids
Au NPs were prepared according to a previous reported method, which were then subject to absorption and transmission electron microscopy (TEM) measurements. As shown in Fig. 2(a), an obvious absorption band peaked at 520 nm in the absorption spectrum of Au NPs, which was originated from the surface plasmon resonance of Au NPs. The TEM image in Fig. 2(c) showed that Au NPs were spherical and monodispersed with an average size of 27 nm. BP/Au nanohybrids were synthesis by an in-situ reduction method, in which BP nanosheets served as the support material, while HAuCl4 as the precursor of gold and sodium citrate as the reductant. BP nanosheets and the obtained nanohybrids were also characterized by absorption and TEM. As shown in Fig. 2(a), pure BP nanosheets exhibited no obvious adsorption band, while BP/Au nanohybrids presented an obvious SPR band at 545 nm [30], which can be attributed to the loaded Au NPs onto BP nanosheets. In contrast to Au NPs, the SPR band of nanohybrids exhibited an obvious red-shift and broadening. This can be attributed to the fact that Au NPs on BP nanosheets were slightly packed to each other, which is beneficial for generating multiple hot-spots, and consequently, improving SERS sensitivity for the biosensing. TEM images further confirmed the construction of BP/Au nanohybrids. BP nanosheets showed a typical sheet structure with a lateral size of 600 nm −1 µm as revealed by Fig. 2(d). After the in-situ reduction process, BP nanosheets were observed to be attached with spherical Au NPs as indicated by Fig. 2(e) and (f). Au NPs were densely located on surface of the BP nanosheets and relatively uniform in size with an average diameter of about 30 nm. In the meantime, the coverage of Au NPs on surface of BP nanosheets was restricted within the range of the BP nanosheets, while no Au NPs were observed outside. According to TEM images, the average sizes of Au NPs and Au NPs on BP nanosheets were calculated to be 19.93 nm and 19.63 nm (as shown in the Supplement 1, Fig. S1 and Table S2). Au NPs on BP nanosheets have the common size with Au NPs. These facts demonstrated that Au NPs were efficiently assembled on the BP nanosheets. Additionally, BP/Au nanohybrids were required to keep their structure and property for subsequent biosensing. To verify the colloidal stability of nanohybrids, hydrodynamic diameters were characterized by dynamic light scattering (DLS). As shown in Fig. 2(b), no obvious change was observed in the hydrodynamic diameter of nanohybrids, suggesting that the nanohybrids scarcely aggregated and preserved their morphology for at least 24 h. As such, BP/Au nanohybrids displayed excellent colloidal stability.
Fig. 2. (a) UV–vis absorption spectra of Au NPs, BP nanosheets, BP/Au nanohybrids. Insert: photographs of Au NPs (left) and BP/Au nanohybrids (right). (b) Hydrodynamic Size distribution of BP/Au nanohybrids measured by DLS (0, 24 h). (c) TEM image of Au NPs. (d) TEM image of BP nanosheets. (e-f) TEM images of BP/Au nanohybrids with different magnifications.
3.2 SERS performance of BP/Au nanohybrids
After the characterization, SERS performance of BP/Au nanohybrids was probed by using R6G as a Raman reporter. First, by using lasers at 638 nm and 785 nm, SERS signals of R6G (10−2 mM) enhanced by BP/Au nanohybrids were collected (Supplement 1, Fig. S2), which indicated that both lasers at 638 nm and 785 nm are feasible for SERS excitation. In the follow-up experiments, 638 nm was selected for SERS measurements. Then, a comparative study was conducted to test SERS activities of Au NPs and BP/Au nanohybrids. Both Au NPs and BP/Au nanohybrids showed no obvious SERS background (Supplement 1, Fig. S3). By using the absorbance spectra, the content of gold in Au NPs and nanohybrids was adjusted to be consistent. Au NPs and BP/Au nanohybrids were then used as substrates for R6G detection. Figure 3(a) displayed SERS spectra of R6G at a concentration of 10−2 mM enhanced by Au NPs and BP/Au nanohybrids, respectively. The characteristic SERS band at 609 cm−1 is attributed to the C-C-C ring in-plane vibration mode while the bands at 1362 cm−1 and 1512 cm−1 correspond to in-plane C-C stretching [31]. SERS intensity of these three characteristic bands was plotted in Fig. 3(b). In contrast to Au NPs, BP/Au nanohybrids significantly amplify the Raman signals of R6G by more than 2 times, which can be attributed to the abundant hot-spots at interstices between Au NPs on BP nanosheets. This underlying mechanism of the hot-spots beneficial Raman amplifying were then verified by performing numerical simulation of electromagnetic field distributions of Au NPs and BP/Au nanohybrids using the finite-difference time-domain method. According to the TEM image of BP/Au nanohybrids, the inter-distance between Au NPs was counted to be around 5 nm (Supplement 1, Fig. S4), which was used for the simulation. As shown in Fig. 4, compared to isolated Au NPs, BP/Au nanohybrids offered much stronger electric field at the nanoscale gaps between Au NPs, namely hot-spots, by interparticle plasmon coupling. Since many Au NPs were deposited onto BP nanosheets densely, hot-spots with high-density were created, and correspondingly, improving the sensitivity for future analysis.
Fig. 3. (a) Average SERS spectra of R6G with a concentration of 10−2 mM using Au NPs and BP/Au nanohybrids as enhancing substrates, respectively. (The spectra were placed in parallel for clarity) (b) SERS intensity quantification of characteristic bands of R6G enhanced by Au NPs and BP/Au nanohybrids, respectively. Error bars indicate the standard deviations of ten measurements.
Fig. 4. Numerical simulation of electromagnetic field distributions of (a) BP/Au nanohybrids and (b) isolated Au NPs using the finite-difference time-domain method. Excitation wavelength: 638 nm, mesh setting: 1 nm, boundary condition: PML, thickness of the BP nanosheet: 5 nm, relationship between Au nanoparticles and BP nanosheets: Tangency.
Additionally, the uniformity and reproducibility of SERS signals are critical factors to appraise the performance of SERS substrates, especially for practical application. To analyze the uniformity, 80 spectra detected from randomly selected locations at one single sample were shown in Fig. 5(a), SERS barcode collected from these 80 spectra was produced in Fig. 5(b), which demonstrated that the SERS signals were quite consistent. The intensity of the characteristic SERS band at 1512 cm−1 was plotted in Fig. 5(c) to assess the signal fluctuation. The corresponding relative standard deviation (RSD) value of this band was calculated to be 9.7%, suggesting BP/Au nanohybrids as homogeneous substrates.
Fig. 5. The uniformity and reproducibility of BP/Au nanohybrids. (a) 80 individual SERS spectra of R6G with a concentration of 10−2 mM. (b) SERS barcode of R6G by collecting 80 spectra along the y-axis, and (c) Histogram statistics of SERS intensity at 1512 cm−1 collected from 80 spectra. (d) 50 Raman spectra of 10−2 mM R6G. Scatter plot (e) and histogram statistics (f) of SERS intensity at 1512 cm−1 collected from 50 individual spectra.
In terms of the reproducibility, 5 batches of substrates were prepared under the same experimental condition, and 50 SERS spectra of R6G on these substrates collected (Fig. 5(d)). The intensity of SERS band at 1512 cm−1 was collected and presented in Fig. 5(e) and (f), which suffered from slight fluctuation with a RSD of 13.85%. Giving that the quantification of the reproducibility of a SERS substrate has been widely accepted with a RSD less than 20% [32], BP/Au nanohybrids were demonstrated to offer good reproducibility. Consequently, such presented BP/Au nanohybrids can obtain highly sensitive, homogeneous, and reproducible SERS signals, providing a significant prerequisite for the SERS technology in quantitative detection.
3.3 Separate detection of SLPI and IL-18
The as-prepared BP/Au nanohybrids were utilized to achieve the detection of SLPI and IL-18 biomarkers. Solutions of SLPI and IL-18 with concentrations ranging from 2 to 2×10−8 mg/mL were mixed the nanohybrids, respectively. Mixed solutions were then subject to SERS measurements. Owing to the high affinity of BP nanosheets towards biomolecules, SLPI and IL-18 would efficiently absorbed on the nanohybrids, and thus contributing to intense SERS signals. As shown in Fig. 6, a series of SERS spectra corresponding to different SLPI and IL-18 concentrations presented an escalating trend as the concentration increased and the dominating characteristic SERS bands were assigned in Table 1.
Fig. 6. (a) Concentration-dependent SERS spectra for SLPI detection in buffered solution (the concentration of SLPI ranging from 2 to 2×10−8 mg/mL). (b)Plot of band intensity at 1000 cm−1 as a function of the SLPI concentration. The regression equation is lg(y) = 3.275 + 0.189lg(x) (y represents band intensity at 1000 cm−1, x represents the concentration of SLPI). Error bars indicate the standard deviations of ten measurements. (c) Concentration-dependent SERS spectra for IL-18 detection using BP/Au nanohybrids (the concentration of IL-18 ranging from 2 to 2×10−8 mg/mL). (d) Plot of band intensity at 950 cm−1 as a function of the IL-18 concentration. The regression equation is lg(y) = 4.056 + 0.252 lg(x) (y represents band intensity at 950 cm−1, x represents the concentration of IL-18). Error bars indicate the standard deviations of ten measurements.
Table 1. Assignment of Raman characteristic peaks in biomarker solution.
For quantitative analysis of the SLPI, band intensity at 1000 cm−1 attributed to Phe v12 and $\rho{\rm CH}_{2}$ [33,35,36] was extracted to characterize the trend. As shown in Fig. 6(b), SERS intensity showed a good linear relationship with the SLPI concentration ranging from 2 to 2×10−8 mg/mL. The linear regression equation was shown in Table 2, with a correlation coefficient (r) of 0.99. The limit of detection (LOD) using BP/Au nanohybrids was calculated to be 1.53×10−8 mg/mL. As a control, SERS detection using Au NPs only reached a concentration of 2×10−6 mg/mL, which was lower than that based on BP/Au nanohybrids, as shown in Supplement 1, Fig. S5. As such, the sensitivity of SLPI detection was improved by 2 orders through the use of BP/Au nanohybrids in contrast to Au NPs. Moreover, our SERS-based detection method is more sensitive than ELISA. Similarly, BP/Au nanohybrids were also used to detect IL-18 with various concentrations by characterizing the SERS band intensity at 950 cm−1. The results were displayed in Fig. 6(d) and Table 2, which indicated that SERS intensity showed a good linear dependence to IL-18 concentration and the LOD reached to 0.23×10−8 mg/mL. Additionally, the sensitivity of BP/Au nanohybrids based detection was improved by two orders of magnitude in contrast to that of Au NPs. The comparison study of sensitivity offered by such presented BP/Au nanohybrids and recently reported SERS substrates was shown in the following Supplement 1, Table S3. As such, a highly increased sensitivity was achieved by using BP/Au nanohybrids as SERS generators, which can be utilized for further multiplex detection.
Table 2. Linear response for SERS detection of biomarkers SLPI and IL-18.
3.4 Separate detection of SLPI and IL-18 in real sample
To validate this detection method in the biologic environment, here, SLPI and IL-18 with various concentrations were mixed in fetal bovine serum (FBS, 10%) and followed by exposure to BP/Au nanohybrids, respectively. SERS spectrum of FBS (10%) using BP/Au nanohybrids was shown as Supplement 1, Fig.S6. Several characteristic bands at 1021 cm−1 and 1453 cm−1 can be observed in the SERS spectrum of FBS, which can be attributed to Phe ν18a/C–N stretch and δ(CH). The intensity of these bands were very weak and meanwhile not overlapped with characteristic bands of SLPI or IL-18. Figure 7(a) displayed the concentration-dependent calibration curve of SLPI assay. It can be seen that SERS band intensity at 1000 cm−1 increased with the increase of SLPI concentration, indicating a similar trend as for the result in buffered solution. The linear relationship was regressed as the equation of lg(y) = 3.904 + 0.201lg(x) with a correlation coefficient of 0.979. Meanwhile, similar results were observed for the detection of IL-18 in serum, the SERS intensity at 950 cm−1 showed a linear relationship with the concentration of IL-18, which was depicted by the function of lg(y) = 3.967 + 0.231lg(x) with a correlation coefficient of 0.992. The correlation coefficients for experiments in serum were observed to be slightly lower than that in buffered solution, which may be attributed to the presence of various proteins in serum. Nevertheless, we can still conclude that such SERS-based detection method is applicable in real biologic environments. The recovery experiments were then performed by using serum samples mixed with SLPI and IL-18, respectively. SERS signals were used to calculate the concentrations of SLPI and IL-18 by utilizing the calibration curves in Fig. 7. The standard concentration, detected concentration and recovery rate of each sample were listed in Table 3. The recoveries of this presented detection method in real samples were in an acceptable range of 87.95% - 97.05%, which proves its feasibility in the biologic environment with high accuracy and sensitivity.
Fig. 7. (a) Concentration-dependent SERS spectra for SLPI detection in buffered solution supplemented with 10% FBS (the concentration of SLPI ranging from 2 to 2×10−8 mg/mL). (b) Plot of band intensity at 1000 cm−1 as a function of the SLPI concentration. The regression equation is lg(y) = 3.904 + 0.201lg(x) (y represents band intensity at 1000 cm−1, x represents the concentration of SLPI). Error bars indicate the standard deviations of ten measurements. (c) Concentration-dependent SERS spectra for IL-18 detection in buffered solution supplemented with 10% FBS (the concentration of IL-18 ranging from 2 to 2×10−8 mg/mL). (d) Plot of band intensity at 950 cm−1 as a function of the IL-18 concentration. The regression equation is lg(y) = 3.967 + 0.231lg(x) (y represents band intensity at 950 cm−1, x represents the concentration of IL-18). Error bars indicate the standard deviations of ten measurements.
Table 3. Recovery results of biomarkers in real sample.
3.5 Simultaneous detection of SLPI and IL-18
Herein, SLPI and IL-18 were mixed in serum with the concentration ratios of 1 : 0, 1 : 1, 0 : 1 (denoted as sample A, B and C). Afterwards, the mixed samples were exposed to BP/Au nanohybrids. As shown in Fig. 8, in the SERS spectra of the mixed samples A and B, the characteristic band at 1000 cm−1 suggested the presence of SLPI. While no obvious band at 1000 cm−1 was observed in the spectrum of sample C, indicating that SLPI was not present in sample C. By using the above established calibration curve, detected concentrations in sample A and B were calculated to be 0.184 mg/mL and 0.199 mg/mL, respectively. The results were consistent with the standard concentrations of SLPI in the mixed samples. Giving that both SLPI and IL-18 contributed to the SERS band at 950 cm−1, quantification of IL-18 was performed in two steps. First step is to confirm the presence or absence of IL-18 in mixed samples and the quantification comes second. From the spectrum of sample C, it could be easily identified that only IL-18 was present. The concentration of IL-18 in the sample C was calculated to be 0.2 mg/mL, which was consistent with the standard concentration of IL-18 in sample C. Notably, SERS intensity ratio (I1000 cm−1/ I950 cm−1) of SLPI was found to be constant with a value of 0.09. For the presence of both SLPI and IL-18, the ratio will be definitely smaller than this constant value. As such, by calculating this ratio of the mixed sample B, IL-18 can be identified to be present or not. Herein, the ratio was calculated to be 0.083, suggesting the presence of both SLPI and IL-18. SERS intensity at 950 cm−1 was first subtracted the portion contributed by SLPI and then used to calculate the concentration of IL-18. The concentration of IL-18 in sample B was calculated to be 0.152 mg/mL. The result was relatively smaller than the standard concentration of IL-18 in the sample B. The recovery was 76.0%, which is a moderate and acceptable level. Thus, quantitative simultaneous detection of SLPI and IL-18 has been achieved using BP/Au nanohybrids. In addition, different machine learning algorithms were used for the identification of mixed samples. SERS intensity ratios (I1000 cm−1/ I950 cm−1) of mixed samples were taken as the input feature of support vector machine (SVM), and the output result was shown in the Fig. 9(a). The accuracy of SVM algorithm is about 85.0%. Apart from SVM, one dimensional convolutional neural network (1DCNN) algorithm is used for the identification of mixed samples. Specifically, a classification model for SERS spectra is built based on the classic lenet-5 neural network. By taking 200 SERS spectra of mixed samples as the input. As shown in Supplement 1, Fig. S7, a pool layer is added after every two convolution layers. The maximum pool method is used for training calculation. After continuous training in neural network, feature extraction and qualitative classification of high-dimensional SERS data are realized. Then, the visual confusion matrix is used to calculate the accuracy. Finally, the classification results and actual prediction results are displayed in a matrix. The output result was shown in the Fig. 9(b), and the accuracy is 98.3%. As such, 1DCNN is powerful for the quantitative classification of these mixed samples. Through the combination of machine learning and SERS technology, multiple detection of various biomarkers can be achieved with high sensitivity and accuracy.
Fig. 9. Identification results of (a) SVM and (b) CNN models generated by using the composite SERS spectra of three mixed biomarker samples. Samples indicated by the darker green blocks are identified as the target sample types.
4. Conclusions
In this work, a novel label-free and highly sensitive SERS detection strategy for biomarkers regarding the quality of a kidney about to be implanted has been proposed by using BP/Au nanohybrids as Raman enhancing generators. BP/Au nanohybrids were achieved through the densely deposition of Au NPs onto the BP nanosheets, which present the outstanding and uniform SERS activity benefiting from the abundant hot-spots at interstices between Au NPs on the nanosheets. Additionally, BP/Au nanohybrids display well affinity to biomolecules, which is favorable for surface enrichment of the analytes. BP/Au nanohybrids were then utilized to quantitatively evaluate the expression level of SLPI and IL-18, two important biomarkers of AKI and indicators of quality of a donor kidney. It has been demonstrated that such proposed method can detect the biomarker with an extremely low concentration down to 0.02 ng/mL, much lower than the expression level in donors without AKI. Meanwhile, simultaneous assessment of SLPI and IL-18 was successfully achieved, which proves that this presented method is powerful for the identification and classification of multiple targets in a single assay. The total analysis process is quite simple and time-saving. As such, this highly sensitive SERS-based multiplexing technique is expected to capture subtle changes in the biomarker levels associated with donor kidney injury that are potentially predictive of the graft and patient outcome, which paves the way for objectively assessing the quality of a donor kidney prior to transplantation in clinical practice.
Funding
National Natural Science Foundation of China (61805143); Natural Science Foundation for Exploration Project of Shanghai (19ZR1478200).
Disclosures
The authors declare that they have no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
Supplemental document
See Supplement 1 for supporting content.
References
1. Y. Wang, T. Lei, L. Wei, S. Du, L. Girani, and S. Deng, “Xenotransplantation in China: Present status,” Xenotransplantation 26, e12490 (2019). [CrossRef]
2. B. Moeckli, P. Sun, F. Lazeyras, P. Morel, S. Moll, M. Pascual, and L. H. Bühler, “Evaluation of donor kidneys prior to transplantation: An update of current and emerging methods,” Transplant International 32(5), 459–469 (2019). [CrossRef]
3. G. D’Amico, “Influence of clinical and histological features on actuarial renal survival in adult patients with idiopathic IgA nephropathy, membranous nephropathy, and membranoproliferative glomerulonephritis: survey of the recent literature,” American journal of kidney diseases : the official journal of the National Kidney Foundation 20(4), 315–323 (1992). [CrossRef]
4. M. Hamouche, “Revue sur la neutrophil gelatinase associated lipocalin ou NGAL,” Immuno-analyse & Biologie Spécialisée 27(4), 149–152 (2012). [CrossRef]
5. P. P. Reese, I. E. Hall, F. L. Weng, B. Schroppel, M. D. Doshi, R. D. Hasz, H. Thiessen-Philbrook, J. Ficek, V. Rao, P. Murray, H. Lin, and C. R. Parikh, “Associations between Deceased-Donor Urine Injury Biomarkers and Kidney Transplant Outcomes. Journal of the American Society of Nephrology,” J Am Soc Nephrol 27(5), 1534–1543 (2016). [CrossRef]
6. N. G. Ambrosi, F. Y. Caro, F. Osella, L. D. Alvarez, F. Sanchez, F. Toniolo, D. Guerrieri, C. Incardona, D. Casadei, and E. Chuluyan, “SLPI in the perfusion solution helps to identify graft quality in kidney transplants,” Biomark Med 13(11), 895–906 (2019). [CrossRef]
7. L. Averdunk, C. Fitzner, T. Levkovich, D. E. Leaf, M. Sobotta, J. Vieten, A. Ochi, G. Moeckel, G. Marx, and C. Stoppe, “Secretory Leukocyte Protease Inhibitor (SLPI)—A Novel Predictive Biomarker of Acute Kidney Injury after Cardiac Surgery: A Prospective Observational Study,” J Clin Med. 8(11), 1931 (2019). [CrossRef]
8. S. Li, S. Wang, R. Murugan, A. Al-Khafaji, D. J. Lebovitz, M. Souter, S. R. N. Stuart, and J. A. Kellum, “Donor biomarkers as predictors of organ use and recipient survival after neurologically deceased donor organ transplantation,” J. Crit. Care 48, 42–47 (2018). [CrossRef]
9. M. Atif, M. S. AlSalhi, S. Devanesan, V. Masilamani, K. Farhat, and D. Rabah, “A study for the detection of kidney cancer using fluorescence emission spectra and synchronous fluorescence excitation spectra of blood and urine,” Photodiagn. Photodyn. Ther. 23, 40–44 (2018). [CrossRef]
10. N. L. Tateosian, M. J. Costa, D. Guerrieri, A. Barro, J. A. Mazzei, and H. Eduardo Chuluyan, “Inflammatory mediators in exhaled breath condensate of healthy donors and exacerbated COPD patients,” Cytokine 58(3), 361–367 (2012). [CrossRef]
11. J. Bao, Z. Tu, J. Wang, F. Ye, H. Sun, M. Qin, Y. Shi, H. Bu, and Y. P. Li, “A Novel Accurate Rapid ELISA for Detection of Urinary Connective Tissue Growth Factor, a Biomarker of Chronic Allograft Nephropathy,” Transplant Proc 40(7), 2361–2364 (2008). [CrossRef]
12. S. Aitekenov, A. Gaipov, and R. Bukasov, “Review: Detection and quantification of proteins in human urine,” Talanta 223(1), 121718 (2021). [CrossRef]
13. E. Tengstrand, H. Zhang, N. Liu, K. Dunn, and F. Hsieh, “A multiplexed UPLC-MS/MS assay for the simultaneous measurement of urinary safety biomarkers of drug-induced kidney injury and phospholipidosis,” Toxicol Appl Pharmacol 366, 54–63 (2019). [CrossRef]
14. Y. Wang, B. Yan, and L. Chen, “SERS Tags: Novel Optical Nanoprobes for Bioanalysis,” Chem. Rev. 113(3), 1391–1428 (2013). [CrossRef]
15. X. Lu, W. Ren, C. Hu, C. Liu, and Z. Li, “Plasmon-Enhanced Surface-Enhanced Raman Scattering Mapping Concentrated on a Single Bead for Ultrasensitive and Multiplexed Immunoassay,” Anal. Chem. 92(18), 12387–12393 (2020). [CrossRef]
16. X. Chen, L. Qin, S. Kang, and X. Li, “A special zinc metal-organic frameworks-controlled composite nanosensor for highly sensitive and stable SERS detection,” Appl. Surf. Sci. 550, 149302 (2021). [CrossRef]
17. Y. Zhao, X. Fang, M. Bai, J. Zhang, H. Yu, F. Chen, and Y. Zhao, “A microfluidic surface-enhanced Raman scattering (SERS) sensor for microRNA in extracellular vesicles with nucleic acid-tyramine cascade amplification,” Chin. Chem. Lett.621–625 (2021).
18. J. Zhang, Y. Yang, X. Jiang, C. Dong, C. Song, C. Han, and L. Wang, “Ultrasensitive SERS detection of nucleic acids via simultaneous amplification of target-triggered enzyme-free recycling and multiple-reporter,” Biosens. Bioelectron. 141, 111402 (2019). [CrossRef]
19. M. Zhang, X. Li, J. Pan, Y. Zhang, L. Zhang, C. Wang, X. Yan, X. Liu, and G. Lu, “Ultrasensitive detection of SARS-CoV-2 spike protein in untreated saliva using SERS-based biosensor,” Biosens. Bioelectron. 190, 113421 (2021). [CrossRef]
20. C. Huo, W. Han, W. Tang, and X. Duan, “Stable SERS substrate based on highly reflective metal liquid-like films wrapped hydrogels for direct determination of small molecules in a high protein matrix,” Talanta 234, 122678 (2021). [CrossRef]
21. Z. Hu, X. Zhou, J. Duan, X. Wu, J. Wu, P. Zhang, W. Liang, J. Guo, H. Cai, P. Sun, H. Zhou, and Z. Jiang, “Aptamer-based novel Ag-coated magnetic recognition and SERS nanotags with interior nanogap biosensor for ultrasensitive detection of protein biomarker,” Sensors and Actuators B Chemical 334, 129640 (2021). [CrossRef]
22. P. Matteini, M. Cottat, F. Tavanti, E. Panfilova, M. Scuderi, G. Nicotra, M. C. Menziani, N. Khlebtsov, M. de Angelis, and R. Pini, “Site-selective surface-enhancd Raman detection of proteins,” ACS Nano 11(1), 918–926 (2017). [CrossRef]
23. Z. Mao, Z. Liu, J. Yang, X. Han, B. Zhao, and C. Zhao, “In situ semi-quantitative assessment of single-cell viability by resonance Raman spectroscopy,” Chem Communications 54, 37–43 (2018). [CrossRef]
24. C. Li, P. Chen, Z. Wang, and X. Ma, “A DNA zyme-gold nanostar probe for SERS-fluorescence dual-mode detection and imaging of calcium ions in living cells,” Sensors and Actuators B: Chemical 347, 130596 (2021). [CrossRef]
25. N. Choi, H. Dang, A. Das, M. S. Sim, I. Y. Chung, and J. Choo, “SERS biosensors for ultrasensitive detection of multiple biomarkers expressed in cancer cells,” Biosens. Bioelectron. 164, 112326 (2020). [CrossRef]
26. X. Wang and S. Lan, “Optical properties of black phosphorus,” Adv. Opt. Photonics 8(4), 618–648 (2016). [CrossRef]
27. M. Luo, Y. Zhou, N. Gao, W. Cheng, X. Wang, J. Cao, X. Zeng, and G. L. Mei, “Mesenchymal stem cells transporting black phosphorus-based biocompatible nanospheres: Active trojan horse for enhanced photothermal cancer therapy,” Chem. Eng. J. 385, 123942 (2020). [CrossRef]
28. H. Zhang, Q. Han, X. Yin, and Y. Wang, “Insights into the binding mechanism of two-dimensional black phosphorus nanosheets-protein associations,” Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 227, 117662 (2020). [CrossRef]
29. W. Zhang, T. Huynh, P. Xiu, B. Zhou, C. Ye, B. Luan, and R. Zhou, “Revealing the importance of surface morphology of nanomaterials to biological responses: Adsorption of the villin headpiece onto graphene and phosphorene,” Carbon 94, 895–902 (2015). [CrossRef]
30. J. Peng, Y. Lai, Y. Chen, J. Xu, L. Sun, and J. Weng, “Sensitive Detection of Carcinoembryonic Antigen Using Stability-Limited Few-Layer Black Phosphorus as an Electron Donor and a Reservoir,” Small 13(15), 1603589 (2017). [CrossRef]
31. S. Bai, D. Serien, A. Hu, and K. Sugioka, “3D Microfluidic Surface-Enhanced Raman Spectroscopy (SERS) Chips Fabricated by All-Femtosecond-Laser-Processing for Real-Time Sensing of Toxic Substances,” Adv. Funct. Mater. 28(23), 1706262 (2018). [CrossRef]
32. M. J. Natan, “Concluding Remarks: Surface enhanced Raman scattering,” Faraday Discuss. 132, 321–328 (2006). [CrossRef]
33. N. R. Agarwal, M. Tommasini, E. Ciusani, A. Lucotti, S. Trusso, and P. M. Ossi, “Protein-Metal Interactions Probed by SERS: Lysozyme on Nanosteuctured Gold Surface,” Plasmonics 13(6), 2117–2124 (2018). [CrossRef]
34. E. Podstawka, Y. Ozaki, and L. M. Proniewicz, “Surface-Enchanced Raman Scattering of Amino Acids and Their Homodipeptide Monolayers Deposited onto Colloilal Gold Surface,” Appl. Spectrosc. 59(12), 1516–1526 (2005). [CrossRef]
35. S. Siddhanta, D. Karthigeyan, P. P. Kundu, T. K. Kundu, and C. Narayana, “Surface enhanced Raman spectroscopy of Aurora kinases: direct, ultrasensitive detection of autophosphorylation,” RSC Adv. 3(13), 4221–4230 (2013). [CrossRef]
36. A. E. Aliaga, C. Garridoa, P. Leyton, G. Diaz F, J. S. Gomez-Jeria, T. Aguayo, E. Clavijo, M. M. Campos-Vallette, and S. Sanchez-Cortes, “SERS and theoretical studies of arginine,” Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy 76(5), 458–463 (2010). [CrossRef]
37. J. A. Huang, M. Z. Mousavi, G. Giovannini, Y. Zhao, A. Hubarevich, M. A. Soler, W. Rocchia, D. Garoli, and F. De Angelis, “Multiplexed Discrimination of Single Amino Acid Residues in Polypeptides in a Single SERS Hot Spot,” Angew. Chem. Int Ed 59(28), 11423–11431 (2020). [CrossRef]
38. X. Wei, D. Zheng, P. Zhang, T. Lin, H. Wang, and Y. Zhu, “Surface-enchanced Raman scattering investigation of bovine serum albumin by Au nanoparticles with different sizes,” JABFM 16(1_suppl), 157–162 (2018). [CrossRef]
