Multiplex Immunochips for High-Accuracy Detection of AFP-L3

ABSTRACT: Alpha-fetoprotein (AFP) is an important tumor biomarker. ... Accordingly, several hospitals have begun to employ the ratio of AFP-L3 to the ...
0 downloads 0 Views 925KB Size
Subscriber access provided by TUFTS UNIV

Article

Multiplex Immunochips for High-Accuracy Detection of AFP-L3% Based on Surface-Enhanced Raman Scattering: Implications for Early Liver Cancer Diagnosis Hao Ma, Xiaoying Sun, Lei Chen, Weina Cheng, Xiao Xia Han, Bing Zhao, and Chengyan He Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b01349 • Publication Date (Web): 03 Aug 2017 Downloaded from http://pubs.acs.org on August 3, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Multiplex Immunochips for High-Accuracy Detection of AFPL3% Based on Surface-Enhanced Raman Scattering: Implications for Early Liver Cancer Diagnosis Hao Ma,† Xiaoying Sun,‡ Lei Chen,§ Weina Cheng,† Xiao Xia Han,*,† Bing Zhao,† and Chengyan He*‡ †

State Key Laboratory of Supramolecular Structure and Materials, Jilin University, Changchun 130012, P. R. China



China-Japan Union Hospital of Jilin University, Changchun 130033, P. R. China

§

Key Laboratory of Preparation and Applications of Environmental Friendly Materials (Jilin Normal University), Ministry of Education, Changchun, 130103, China ABSTRACT: Alpha-fetoprotein (AFP) is an important tumor biomarker. In particular, the overexpression of AFP-L3 is associated with hepatocellular carcinoma (HCC). Accordingly, several hospitals have begun to employ the ratio of AFP-L3 to the total AFP level (AFP-L3%) as a new diagnostic evidence for HCC owing to its high diagnostic accuracy. However, current methods of detection for AFP and AFP-L3 are time-consuming, require multiple samples, and lack in sensitivity and specificity. Herein, we present a novel concept for the early diagnosis of HCC based on the combination of Raman frequency shift and intensity change, and developed surface-enhanced Raman scattering (SERS)-based immunochips via AFP-L3%. In the first step of the study, the frequency shift of 4-mercaptobenzoic acid (MBA) was applied for the quantitative determination of total AFP based on the AFP and anti-AFP interaction on MBA-modified silver chips. 5,5-Dithiobis(succinimidyl-2-nitrobenzoate) (DSNB)-modified immunogold was then incorporated with AFP-L3 antibodies for sandwich immunoreaction on the chips. As a result, we found that a typical Raman band intensity of DSNB presented an exponential linear relationship with the concentration of AFP-L3. Thus, the AFP-L3% can be calculated according to the concentrations of AFP-L3 and total AFP. The most important advantage of the proposed method is the combination of AFP-L3% and frequency shifts of SERS, which exhibits excellent reproducibility and high accuracy, and significantly simplifies the conventional detection procedure of AFP-L3%. Application of the proposed method with the serum of patients with HCC demonstrated its great potential in early liver cancer diagnosis.

Liver cancer represents 6% and 9% of the global cancer incidence and mortality burden, respectively. With an estimated 746,000 deaths in 2012, liver cancer is the second most common cause of death from cancer worldwide.1 Similar to many other diseases, the development of sensitive and specific methods for early diagnosis is an urgent requirement that is essential for the effective treatment of cancer. Glycoproteins play vital roles in many biological processes such as molecular recognition and the immune response. The expression of glycoproteins has been associated with diverse diseases.2 For example, the glycoprotein alpha-fetoprotein (AFP) has been developed as a biomarker for the early diagnosis of liver cancer.3 Many kinds of technologies have been applied for detecting AFP,4-12 including electrochemistry,8,9 fluorescence,10 surfaceenhanced Raman scattering (SERS),11 and surface plasmon resonance.12 To the best of our knowledge, the lowest limit of detection of AFP is 0.46 fg/mL, which is based on the combination of SERS with SiC nanoparticles,11 and chemiluminescence is commonly adopted for a clinical diagnosis, which provides a detection range from 0.5 ng/mL to 3000 ng/mL. However, the detection of AFP with current methods usually leads to false-positive or false-negative results, and the concentration of AFP is also elevated in other liver diseases.13 Therefore, it is quite necessary to develop a novel method with

specific biometrics and sensitive signal expression for the diagnosis of liver cancer. On the other hand, the Lens culinaris agglutinin (LCA)reactive fraction of AFP (AFP-L3) has recently been proposed as a more specific and important biomarker for hepatocellular carcinoma (HCC) diagnosis.14 In general, the concentration of AFP in healthy people is below 10 ng/mL, while AFP-L3 is detected only in the patients with HCC. In oncology, when the AFP-L3% is ≥10%, a diagnosis of HCC should be highly considered. Measurement of AFP-L3 exhibits dramatically improved specificity and sensitivity compared with the total AFP concentration.15 To date, a few studies have been conducted on the detection of AFP-L3 with different methods, including electrochemistry15,16 and the use of a supramolecular glycoprobe.17 Although these methods provide a suitable lower limit of detection of 3.24 pg/mL, they are not applicable for cases with a low total AFP concentration. Clinical trials have indicated that the ratio of AFP-L3 to the total AFP level should be adopted as the new clinical diagnosis standard for the accurate judgment of liver cancer. Obtaining this ratio requires a twostep procedure for detect AFP and AFP-L3 respectively. To our best knowledge, only one study has reported the use of electrochemistry to detect AFP and AFP-L3 simultaneously to obtain the ratio.15 However, this method is too complex, invasive, and expensive for wide clinical application.

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

SERS has long been considered an excellent choice for biological analysis and is a powerful tool owing to its high sensitivity, high selectivity, and fluorescence-quenching properties.18 The majority of work conducted to date on the SERS detection of AFP has been based on employing new SERSactive substrates.19 For instance, many magnetic nanoparticles have been applied to construct a sandwich immunoassay to carry out the separation and enrichment steps simultaneously.20–23 One group reported a silver nanoparticle trimer assembled with AFP, which showed a limit of detection as low as 0.097 aM.24 A paper-based SERS immunoassay showed its ability to detect AFP, and this method could fulfill the detectable requirements of AFP lower than 20 ng/mL.25 Furthermore, one-dimensional NiCo2O4 nanorods were used as the immunoassay substrate, and with this hybrid structure the detection limit was as low as 2.1 fg/mL.26 SERS-active substrates based on a metal−organic framework was introduced for the highly sensitive determination of AFP, which improved the LOD as low as 0.1ng/mL.27 Moreover, multiple techniques based on the combination of SERS and microfluidics, or molecular imprints are also applied for detection of AFP.2,28,29 However, the accurate quantitative determination of biomolecules based on Raman band intensity remains an ongoing challenge. Several research groups, including our own, have observed a peculiar phenomenon of changes in the vibrational frequencies of a SERS-active probe, which can be applied as a novel readout method. For instance, Olive et al.30 first reported a vibrational frequency shift owing to antibody conjugation. In addition, we previously reported a hydrogen binding effect on the vibrational frequency shift.31 Moreover, an Hg2+ sensor was designed by means of a SERS reporter frequency shift.32 In these studies, measurement of the SERS frequency shifts has shown higher reproducibility than that intensity-based methods which suffer from unavoidable intensity variation due to the inhomogeneity of SERS substrates. Therefore, a new trend has emerged in gradually replacing the determination of biomolecules based on intensity emission toward evaluation of the frequency shift instead. To our best knowledge, none of these studies detected AFPL3 based on SERS spectroscopy, and in no case published so far, the combined two concepts of Raman frequency shift and intensity change were used for the early diagnosis of HCC. To this end, in this study we designed a SERS-based sandwich-

Scheme 1. Schematic of the SERS-based immunoassay. Preparation of immunogold (A), antibody capture chip fabrication (B), and SERS spectra of the two immuno-processes (C, D). A clear frequency shift from (B) to (C) is detected, with a new peak in (D).

Page 2 of 7

type immunoassay, which combines two independent SERS probes, 4-mercaptobenzoic acid (MBA) and 5,5dithiobis(succinimidyl-2-nitrobenzoate) (DSNB), for evaluation of their frequency-shift and intensity-monotonic changes (Scheme 1). The frequency shift was then used for the quantitative determination of AFP. We evaluated the ability of determining the AFP-L3% with this method according to its sensitivity and specificity at a low detection limit. We compared the quantitative detection of AFP-L3 with this new immunoassay in serum samples from an HCC patient with respect to reproducibility and accuracy to help to establish a more economical and less invasive method for the detection of AFP-L3% that directly correlates with liver cancers.

EXPERIMENTAL SECTION Reagents and samples. MBA, bovine serum albumin (BSA), HAuCl4, and poly(diallyldimethylammonium chloride) (Mw = 200,000−350,000, 20 wt% aqueous solution) were obtained from Sigma-Aldrich. 5,5'-Dithiobis-(2-nitrobenzoic acid) (DTNB), dicyclohexylcarbodiimide (DCCD), 1-(3-dimethylaminopropyl)3-ethylcarbodiimide hydrochloride (EDC), and Nhydroxysuccinimide (NHS) were obtained from J&K Chemical Co. Borate buffer (0.2 M pH=9) and AFP-L3 enzyme-linked immunoassay (ELISA) kits were obtained from Beijing Dingguo Changsheng Biotechnology Co., Ltd. The phosphate buffered saline (PBS; 0.01 M, pH 7.2) used in this study contained 0.8% NaCl, 0.02% KH2PO4, 0.02% KCl, and 0.12% Na2HPO4·12H2O. All chemicals were analytical-grade reagents and used without further purification. Milli-Q water was used in the study. The clinical serum samples were obtained from the patients at the Third Hospital of Jilin University (P. R. China). The sample amounts, patient ages and genders, and disease types are summarized in Table S2. Preparation of antibody capture chips. The Si/Ag/MBA chips were prepared as reported in our previous work,33 and the method is summarized in the Supporting Information. The chips were then immersed in a solution of EDC/NHS. After 2 h, the chips were rinsed three times with Milli-Q water, and dried with nitrogen gas. Subsequently, anti-AFP antibodies were immobilized by pipetting 500 µL of the protein solution onto chips in a plastic centrifuge tube. The reaction was allowed to progress overnight at 4°C or for 4 h at 37°C. Subsequently, the chips were rinsed with PBS solution, and gently dried with nitrogen gas. The blocking buffer (0.1 mg/L BSA in PBS, pH 7.4) was pipetted onto the chips and incubated for 2 h. The chips were then rinsed three times with PBS, and stored in the plastic centrifuge tube under a nitrogen atmosphere. Preparation of immunogold colloids. Initially, we synthesized DSNB by reacting 0.32 g of DTNB (0.807 mmol) in 8 mL absolute DMF under a nitrogen gas atmosphere. Then, 0.38 g of DCCD (1.85 mmol) and 0.22 g of NHS (1.91 mmol) were dissolved in 2 mL DMF. The reaction mixture was stirred at room temperature for 12 h. After filtration and reduced pressure distillation, the crude product was recrystallized from acetone, yielding a yellow powder; 1H nuclear magnetic resonance (NMR; CDCl3, 500 Hz): 8.13 (d, 2H, C6H4), 7.85 (d, 2H, C6H4), 7.97 (s, 2H, C6H4), 2.91 (s, 8H,CH2). The preparation of immunogold colloids consisted of two steps. In step 1, 100 µL of a 1 mM DSNB solution in acetonitrile was added to 1 mL of the colloidal gold synthesized by the Lee and Meisel method,34 and the mixture was left to react for 3–5 h. The labeled colloids were then separated from the solution by centrifugation at 10,000 rpm for 7 min. The supernatant was discarded, and the deep red sediment was resuspended in 1 mL of borate buffer.

ACS Paragon Plus Environment

Page 3 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

In step 2 (Scheme 1A), anti-AFP-L3 antibodies were bound to the gold particles via the succinimidyl group. The anti-AFP-L3 antibodies (20 µL) were added to the 1mL suspension of the labeled colloids. The mixture was incubated at room temperature of 25°C for 2 h. To block the immunogold colloids, we added 1 mL of PBS (0.1% BSA, pH 7.4), and they were stored at 4°C before use. Immunoassay protocol. In general, detection of AFP and AFP-L3 in a clinical laboratory requires two separate analyses. As shown in Scheme 1, we developed a method to combine the two related steps for the simultaneous detection of AFP and AFP-L3. In step 1 (Scheme 1 B to C), the AFP standard curves were constructed following the typical procedure for a sandwich-type assay. Since the background interference from the serum is negligible as shown in Figure S4, here we used PBS instead of serum for AFP dilution. We added 400 µL PBS (KH2PO4/K2HPO4, pH 7.4), 0.1% BSA, and 150 mM NaCl onto the chips. For each chip, 100 µL of AFP solutions of various concentrations were pipetted into a plastic centrifuge tube and allowed to react overnight at 4°C or for 4 h at room temperature. The chips were rinsed three times with PBS buffer and dried gently with nitrogen gas before SERS characterization. After the first SERS characterization (Scheme 1C), the chips were put back in the plastic centrifuge tube, and 500 µL immunogold colloids in PBS was pipetted into the tube and allowed to react for 4 h at room temperature. The same pretreatment described above was performed before the second SERS characterization (Scheme 1D). SERS characterization and analysis. All SERS spectra were measured on a Jobin Yvon/HORIBA LabRam ARAMIS Raman spectrometer equipped with an HeNe laser (632.8 nm). The typical exposure time for each measurement applied in this work was 10 s with one-time accumulation. The laser power on the sample was set to about 15 mW. The ultraviolet-visible (UV-vis) spectra of the samples were obtained with a SHIMADTU ultraviolet spectrophotometer (UV-3600). Transmission electron microscopy (TEM) images were obtained on a JEM-2100F system operated at 200 kV. The NMR spectra were obtained on a Bruker AVANCEIII500 system. The X-ray photoelectron spectra (XPS) were acquired with a Thermo ESCALAB 250 photoelectron spectrometer with Al Ka X-ray radiation.

RESULTS AND DISCUSSION Preparation and characterization of the gold nanoprobe and chip. As shown in Scheme 1A, DSNB was employed as the linker for the protein and gold nanoparticles (AuNPs). The gold nanoprobe was first decorated with DSNB through covalent binding (Au-S bond). Therefore, with the succinimidyl group from DSNB, the amino group of the anti-AFP-L3 will rapidly connect to the carboxyl group. The UV-Vis spectra of the different complexes in the three steps of immunogold establishment are shown in Fig. 1A. The UV-Vis spectra band of 20-nm AuNPs was located at 520 nm (Fig. 1A, black line). After introduction of DSNB, the UV-Vis band of the DSNB-decorated AuNPs shifted to 522 nm (Fig. 1A, red dotted line) with decreasing intensity. This result confirmed that DSNB had attached to the surface of the AuNPs by strong Au-S bonds. However, this reaction caused some aggregation. When the DSNB/AuNPs were mixed with anti-AFP-L3, the peak was slightly red-shifted again (Fig. 1A, blue dotted line) due to the aggregation and absorption of the antibody onto the modified AuNPs. The TEM image of the anti-AFP-L3-modified AuNPs is shown in Fig. 1B. Based on these results, DSNB was employed as the SERS nanoprobe for the quantitative determination of AFP-L3. The chip was designed and employed for the detection of the total concentration of AFP. Silver island film was prepared with evaporation technology, and its surface was decorated with MBA,

Figure 1. UV-vis spectra (A) and (B) TEM image of the DSNBmodified AuNPs.

Figure 2. SEM images of (A) Ag/MBA/anti-AFP chip and (B) anti-AFP/gold-decorated chip. the probe for the SERS-based determination of AFP. The MBAderived coupling agent was fabricated by EDC-NHS, which is commonly used for activating the carboxyl group. Figure 2A and 2B show the scanning electron microscopy (SEM) images of Ag/MBA/anti-AFP chip and the gold-decorated chip, respectively. As shown in Figure 2B, the immunogold was successfully decorated on the chips through the immunoreaction. The characterization of XPS was in agreement with the observations of SEM images. The results for the three surfaces are shown in Table S1 in the Supporting Information. We evaluated the spectra of carbon, oxygen, nitrogen, gold, and sulfur. As expected, we found that only the chip decorated with AuNPs had two peaks located at 87.5 eV and 83.5 eV. This result indicated that the DSNB-modified AuNPs had decorated the chip owing to the immunoreaction between AFP-L3 and anti-AFP-L3. The antibody capture substrate consisted of anti-AFP bound to the silicon chip via the MBA-derived coupling agent. MBA (10-4 M) chemisorbs to silver through Ag-S bonds. The formation of the resulting silver-bound thiolate and its subsequent coupling to anti-AFP can be confirmed by SERS. The SERS results are shown in Fig. 3. We clearly observed that the signal of the chip had almost the same peak as that of MBA, except for disappearance of the band at 1370 cm-1 and the appearance of the band at around 1573 cm-1. The band at 1370 cm-1 is ascribed to the COO− stretching mode, which disappears when bound with biomacromolecules such as antibodies and proteins. The band at 1573 cm-1 is assigned to an NH-CO mode. The changes in these two peaks verified the formation of the MBA-based antibody coating. We also observed that the intensity of peaks located at 997 cm-1 and 1021 cm-1 was markedly enhanced. These peaks have thus far been largely ignored, and are assigned to non totally symmetric V(CC),35 which is regularly enhanced and has potential for quantifying biological conjunctions. SERS-based determination of AFP and AFP-L3. In our experiment, AFP was detected using the SERS-based chips described in Scheme 1C. First, we added the saturated anti-AFP at a volume of 5 µL. The chips blocked with BSA were then immersed in AFP/PBS. The SERS spectra of the chips immersed in AFP of different concentrations were observed using a 633-nm laser line, and are shown in Fig. 4. The zoomed-in area in Fig. 4B

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

shows obvious frequency shifts among the samples of different concentrations (from 1 to 1000 ng/mL) compared with the blank sample. Repeated measurements showed that each peak position of 4-MBA on one chip varied by less than 0.1 cm-1, whereas the frequency shift of each concentration ranged from 0.6 ± 0.1 cm-1 to 3 ± 0.1 cm-1 for 1 ng/mL to 1000 ng/mL, respectively. The change of the peak frequency shift as a function of AFP concentration is shown in Fig. 4C, revealing a linear change in frequency with the logarithmic concentration of AFP by fitting the equation y = 0.82719x + 0.51749, where x represents the logarithmic concentration of AFP and y is the absolute value of the frequency shift. In addition, the proposed chips could detect AFP at a concentration as low as 0.5 ng/mL. This result indicated that the proposed SERS-based immunochip is feasible for the quantitative determination of AFP in the concentration range tested (0.5 to 1000 ng/mL). However, we also found that detection became saturated with an increasing equilibrium concentration up to 1000 ng/mL. In addition, higher concentrations also caused a frequency shift, and the wavenumber of the designated peak infinitely approached 1071 cm-1, which is in contrast to the linear correlation mentioned above. This result indicates that the quantitative analysis would not be possible outside of the tested concentration range (0.5 to 1000 ng/mL), and that 1000 ng/mL would be the threshold value.

Figure 3. Raman spectrum of the antibody capture substrate before (A) and 4 h after (B) adding the antibodies.

Page 4 of 7

According to previous work, the observed frequency shift may be attributed to a change in the chemical environment,30,36 orientation of the reporter,37 and/or charge transfer.31,38 With respect to the influence of the chemical environment, Olivo30 suggested that this frequency shift could be attributed to a pressure-induced shift. However, by high-pressure Raman spectra, we only observed a frequency shift of 3 cm-1 when the pressure level reached the gigaPascal range; hence, it does not seem possible that this is a pressure-induced shift. In the present system, upon exposure of the chip to the AFP solution in PBS buffer at different concentrations, the designed peak at 1075 cm-1 exhibited a downshift, corresponding to an in-plane ring-breathing mode and vibration of the C-S bond. Our proposed mechanism for this frequency shift is outlined in Scheme 2. The peak located at 1075 cm-1 (blue line in Scheme 2) from the SERS spectra could be ascribed to antiAFP/MBA. Binding with AFP caused deformation of the phenyl ring-S-silver (ph-S-Ag) surface complex because of the high molecular weight of AFP. Thus, the fingerprint of MBA in this condition will only cause a change to the band resulting from a change in the polarizability of the ph-S-Ag complex. According to the SERS spectra of the saturated concentration of AFP, we infer that the peak of AFP/anti-AFP/MBA is located at 1071 cm-1 (red line in Scheme 2), which is ascribed to deformation of the ph-SAg complex. Therefore, we consider that the frequency change is a comprehensive reflection of two modes in ph-S-Ag. In Scheme 2, parts A, B, and C show the multi-peak fitting of the characteristic peaks at 10 ng/mL, 100 ng/mL, and 1000 ng/mL, respectively. It is obvious that the SERS intensity of AFP/anti-AFP/MBA becomes stronger and the intensity of anti-AFP/MBA becomes weaker with an increase of AFP concentration. Therefore, the intensity changes of the two peaks result in a frequency shift of the band. As mentioned above, the frequency shift-dependent analytical approach has relatively high reproducibility because the shifted bands directly correlate with the target concentration. As shown in Figure 4 C, the reproducibility of the proposed method for AFP detection is remarkable even within 3 wavenumbers.

Scheme 2. Multi-peak fitting of the characteristic peaks at three typical AFP concentrations: (A) 10 ng/mL, (B) 100 ng/mL, and (C) 1000 ng/mL. The red line indicates the peak located at 1075 cm-1, and the blue line is a peak located at 1071 cm-1. Figure 4. (A) SERS spectrum of the modified substrate after adding AFP of various concentrations. (B) Zoomed-in image of the dotted line in (A). (C) Semi-log plot of the absolute peak shift for the peak at around 1075 cm-1 to quantify the AFP concentration. Each data point represents an average of 5 measurements, and each error bar indicates the standard deviation.

In a traditional method for HCC diagnosis, such as chemiluminescence (detailed information shown in SI), the diagnosis is always established immediately following the determination of enhanced AFP levels. However, many other hepatopathies such as cirrhosis also result in an increase in the concentration of AFP in the patient serum, which can mislead the diagnosis. For this reason, we propose the use of AFP-L3% instead, which shows higher

ACS Paragon Plus Environment

Page 5 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 5. (A) SERS spectra of chips with various AFP-L3 concentrations from 8 ng/mL to 1000 ng/mL. (B) Intensity ratio (I1331/I1071) as a function of AFP-L3 concentration. (C) Details of the dotted area in (A). Each data point represents an average of 5 measurements, and each error bar indicates the standard deviation. sensitivity and accuracy as a marker for HCC than AFP at early diagnosis. The ratio of AFP-L3 to total AFP is a new independent predictor factor, which could effectively avoid the disadvantages of AFP detection in obtaining false positives or false negatives. Thus, to obtain the AFP-L3% value, not only the concentration of AFP is necessary but the determination of AFP-L3 is also obligatory. Here, we introduced anti-AFP-L3-modified AuNPs to capture the AFP-L3 on the chips, which we refer to as immunogold chips. DSNB was used as a reporter and a linker between the antibody and AuNPs. There should be hot spots between Ag-Ag and AuAu nanoparticles, which enabled the sandwich immunogold chips to display high Raman scattering enhancement,but the hot spots between Ag-Au nanoparticles would be inhibited due to the large distance (three layers of proteins) between them. The concentration-dependent SERS spectra of the immunogold-decorated chips are shown in Fig. 5A. The strong SERS band at 1331 cm-1 was chosen as a signal of immunogold, because of the strong scattering cross-section of its symmetric NO2 stretch of DSNB. The detection of a strong SERS signal further implied that it would be possible to determine AFP-L3 at a low concentration. To correct for any detected signals that do not correspond to changes in the concentration of AFP-L3, an internal standard is required. In general, a material that responds to changes in the experimental conditions in the same manner as the analyte is considered to serve as the best internal standard. Hence, we chose the peak of MBA at 1071 cm-1 as the internal standard, which is in the same condition as the reporter. The plot of the SERS intensity ratio (I1331/I1071) versus the AFP-L3 concentration and the detailed area of the DSNB peaks are shown in Fig. 5B. The intensity of the band at 1331 cm-1 gradually increased with an increase in the AFP-L3 concentration. After fitting and uniformization, the intensity ratio (I1331/I1071) change showed a linear relationship with the change in AFP-L3 concentration in the range of 8 ng/mL to 1000 ng/mL by

fitting the equation y = 0.07481 + 0.0012x, where x represents the concentration of AFP-L3 and y is the intensity ratio (I1331/I1071). The lowest detectable concentration was found to be 0.5 ng/mL. According to the equation above, we infer that the proposed immunogold-decorated chip can be applied for the quantitative analysis of AFP-L3 concentration in the range of 0.5 ng/mL to 1000 ng/mL. This result suggests that it would be possible to accurately measure AFP-L3% even if the sample has an extremely low or no concentration of AFP. Clinical sample analysis. To verify the clinical application potential of the proposed SERS-based immunoassay, the chip was employed to test AFP-L3% in a clinical human serum sample obtained from a 50-year-old male patient with HCC. When we obtained the sample, the concentration of AFP (540.6 ng/mL) had previously been determined by the chemiluminescence method. Therefore, we could directly compare the performance of the new method to the conventional detection method. We carried out AFP sensing directly using the SERS-based chip. For AFP-L3 determination, we diluted the sample 100× with BSA to achieve an AFP level within the response range of ELISA (0.1–8 ng/mL). The results of the two immunosteps are shown in Fig. 6A, indicating a downshift of the MBA peak of 2.78 ± 0.5 cm-1. According to the standard curve, this shift value corresponds to an AFP concentration of 543.5 ng/mL. The intensity ratio (I1331/I1071) was calculated to be 0.7265 (Fig. 6A-c), and using the internal standard curve, this value corresponds to an AFP-L3 concentration of 542.3 ng/mL. The value measured by ELISA was 5.27 ng/mL. According to the standard curve shown in Fig. S1, this result indicated that the concentration of the actual sample was approximately 527 ng/mL. According to the AFP-L3% value (99.5%) , the patient would be diagnosed with HCC, confirming the diagnostic result of the hospital based on AFP measurement with the chemiluminescent method. These results demonstrate the acceptable accuracy of our SERS-based assay compared with the conventional method.

Figure 6. (A) Raman spectra of the (a) anti-AFP/MBA/Ag substrate alone, and after (b) exposure to the serum of an HCC patient and (c) anti-AFP-L3. (B) Clinical sample test results by the proposed SERS-based immunoassay compared with the diagnosis results.

ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Moreover, 32 clinical samples with the different AFP concentrations from the serum of patients with cancer or benign diseases were tested with the results shown in Fig. 6 B, which also support the feasibility and sensitivity of the proposed SERS-based assay for the early diagnosis of HCC. Typically, the SERS results of two hepatocirrhosis patients with AFP concentrations of 351.7 and 1.4 ng/mL respectively were shown in Figure S2 and S3. All basic information of the patients was summarized in Table S2.

CONCLUSION In summary, a SERS-based immunochip was designed using two Raman reporter modified substrates. Based on assessment of the frequency shift of MBA, we accomplished the quantitative determination of AFP with a good linear relationship to concentration. Our method was also applied to the determination of AFPL3, a more sensitive and specific biomarker for HCC, with an internal standard. The proposed approach significantly simplifies the conventional detection procedure of AFP-L3% with excellent reproducibility and high accuracy and it shows acceptable accuracy with potential for reduced economic cost. We have provided a new tool for multiple-homolog detection based on SERS with high sensitivity. The SERS-based immunochip for AFP-L3% is expected to be of great convenience for not only the early diagnosis of HCC but also for other hepatopathies, with great reference value.

ASSOCIATED CONTENT Supporting Information Detailed procedure of preparation of Si/Ag/MBA chips; information about the chemiluminescence method; Standard curve of ELISA; binding energies (eV) and assignments for XPS Spectra of the chips; SERS spectra of the chips for two clinical samples with hepatocirrhosis; summary of characteristics of HCC patients, BLD patients, and healthy people; control SERS spectra of MBAmodified substrates for AFP detection in serum. The Supporting Information is available free of charge on the ACS Publications website.

AUTHOR INFORMATION Corresponding Author *[email protected] (X.X.H); [email protected] (C.Y.H)

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENT This work was supported by the National Natural Science Foundation (Grant Nos. 21273091, 81572082, 21403082 and 21611130173) of P. R. China,

REFERENCES (1) International Agency for Research on Cancer, 2014,World Cancer Report 2014: Epidemiology of liver cancer; Lyon, 2014. (2) Ye, J.; Chen, Y.; Liu, Z. Angew. Chem., Int. Ed. 2014, 53, 10386-10389. (3) Waidely, E.; Al-Yuobi , A. O.; Bashammakh, A. S.; ElShahawi, M. S.; Leblanc, R. M. Analyst. 2016, 141, 36-44. (4) Katoh, H.; Nakamura, K.; Tanaka, T.; Satomura, S.; Matsuura, S. Anal. Chem. 1998, 70, 2110-2114. (5) Wu, B.Y.; Wang, H. F.; Chen, J. T.; Yan, X. P. J. Am. Chem. Soc. 2011, 133, 686-688.

Page 6 of 7

(6) Peng, M. P.; Ma, W.; Long, Y. T. Anal. Chem. 2015, 87, 58915896. (7) Zhang, F.; Zhu, J.; Li, J. J.; Zhao, J. W. J. Mater. Chem. C. 2015, 3, 6035-6045. (8) Li, Y. J.; Ma, M. J.; Zhu, J. J. Anal. Chem. 2012, 84, 1049210499. (9) Qin, X.; Liu, L.; Xu, A.; Wang, L.; Tan, Y.; Chen, C.; Xie, Q. J. Phys. Chem. C. 2016, 120, 2855-2865. (10) Wu, Y.; Wei, P.; Pengpumkiat, S.; Schumacher, E. A.; Remcho, V. T. Anal. Chem. 2015, 87, 8510-8516. (11) Zhou, L.; Zhou, J.; Feng, Z.; Wang, F.; Xie, S.; Bu, S. Analyst. 2016, 141, 2534-2541. (12) Tawa, K.; Kondo, F.; Sasakawa, C.; Nagae, K.; Nakamura, Y.; Nozaki, A.; Kaya, T. Anal. Chem. 2015, 87, 3871-3876. (13) Choi, J. Y.; Jung, S. W.; Kim, H. Y.; Kim, M.; Kim, Y.; Kim, D. G.; Oh, E. J. World. J. Gastroenterol. 2013, 19, 339-346. (14) Nakagawa, T.; Miyoshi, E. J.; Yakushijin, T.; Hiramatsu, N.; Igura, T.; Hayashi, N.; Taniguchi, N.; Kondo, A. J. Proteome. Res. 2008, 7, 2222-2233. (15) Liu, X.; Jiang, H.; Fang, Y.; Zhao, W.; Wang, N.; Zang, G. Anal. Chem. 2015, 87, 9163-9169. (16) Li, Y.; Zhong, Z.; Chai, Y.; Song, Z.; Zhuo, Y.; Su, H.; Liu, S.; Wang, D.; Yuan, R. Chem. Commun. 2012, 48, 537-539. (17) He, X. P.; Hu, X. L.; Jin, H. Y.; Gan, J.; Zhu, H.; Li, J.; Long, Y. T.; Tian, H. Anal. Chem. 2015, 87, 9078-9083. (18) Han, X. X.; Jia, H. Y.; Wang, Y. F.; Lu, Z. C.; Wang, C. X.; Xu, W. Q.; Zhao, B.; Ozaki, Y. Anal. Chem. 2008, 80, 2799-2804. (19) Han, X. X.; Zhao, B.; Ozaki, Y. Anal. Bioanal. Chem. 2009, 394, 1719-1729 (20) Ge, M.; Wei, C.; Xu, M.; Fang, C.; Yuan, Y.; Gu, R.; Yao, J. Anal. Methods. 2015, 7, 6489-6495. (21) Chon, H.; Lee, S.; Yoon, S. Y.; Chang, S. I.; Lim, D. W.; Choo, J. Chem. Commun. 2011, 47, 12515-12517. (22) He, Y.; Wang, Y.; Yang, X.; Xie, S.; Yuan, R.; Chai, Y. ACS Appl. Mater. Interfaces. 2016, 8, 7683-7690. (23) Song, C.; Yang, Y.; Yang, B.; Min, L.; Wang, L. J. Mater. Chem B. 2016, 4, 1811-1817. (24) Wu, X.; Fu, P.; Ma, W.; Xu, L.; Kuang, H.; Xu, C. RSC Adv. 2015, 5, 73395-73398. (25) Chen, Y.; Cheng, H.; Tram, K.; Zhang, S.; Zhao, Y.; Han, L.; Chen, Z.; Huan, S. Analyst, 2013, 138, 2624-2631. (26) Wang, X.; Zhou, L.; Wei, G.; Jiang, T.; Zhou, J. RSC Adv. 2016, 6, 708-715. (27) Hu, Y.; Liao, J.; Wang, D.; Li, G. Anal. Chem. 2014, 86, 3955-3963. (28) Lee, M.; Lee, K.; Kim, K. H.; Oh, K. W.; Choo, J. Lab Chip, 2012, 12, 3720-3727. (29) Muhammad, P.; Tu, X.; Liu, J.; Wang, Y.; Liu, Z. ACS Appl. Mater. Interfaces, 2017, 9, 12082–12091. (30) Kho, K. W.; Dinish, U. S.; Kumar, A.; Olivo, M. ACS Nano. 2012, 6, 4892-4902. (31) Wang, Y.; Ji, W.; Sui, H.; Kitahama, Y.; Ruan, W.; Ozaki, Y.; Zhao, B. J. Phys. Chem. C. 2014, 118, 10191-10197. (32) Chen, L.; Zhao, Y.; Wang, Y.; Zhang, Y.; Liu, Y.; Han, X. X.; Zhao, B.; Yang, J. Analyst, 2016, 141, 4782-4788. (33) Wang, X.; Wang, Y.; Sui, H.; Zhang, X.; Su, H.; Cheng, W.; Han, X. X.; Zhao, B. J. Phys. Chem. C. 2016, 120, 13078-13086. (34) Lee, P. C.; Meisel, D. J. Phys. Chem. 1982, 86, 3391-3395. (35) Smith, G.; Girardon, J. S.; Paul, J. F.; Berrier, E. Phys. Chem. Chem. Phys. 2016, 18, 19567-19573. (36) Tang, B.; Wang, J.; Hutchison, J. A.; Ma, L.; Zhang, N.; Guo, H.; Hu, Z.; Li, M.; Zhao, Y. ACS nano. 2016, 10, 871-879. (37) Guerrini, L.; Pazos, E.; Penas, C.; Vazquez, M. E.; Mascarenas, J. L.; Alvarez-Puebla, R. A. J. Am. Chem. Soc. 2013, 135, 1031410317. (38) Wang, Y.; Yu, Z.; Ji, W.; Tanaka, Y.; Sui, H.; Zhao, B.; Ozaki, Y.Angew. Chem., Int. Ed. 2014, 53, 13866-13870.

ACS Paragon Plus Environment

Page 7 of 7

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Table of Content

ACS Paragon Plus Environment