Integration of a 3D-printed read-out platform with a quantum dot-based immunoassay for detection of the avian influenza A (H7N9) virus

Meng Xiao a, Liping Huang a, Xiaohui Dong bc, Kaixin Xie a, Haicong Shen a, Caihong Huang a, Wei Xiao a, Meilin Jin *b and Yong Tang *ad
aDepartment of Bioengineering, Guangdong Province Engineering Research Center for antibody drug and immunoassay, Jinan University, Guangzhou 510632, PR China. E-mail: tyjaq7926@163.com; Tel: (+86)-20-85227003
bState Key Laboratory of Agricultural Microbiology, College of Veterinary Medicine, Huazhong Agricultural University, Wuhan, 430070, PR China. E-mail: jinmeilin@mail.hzau.edu.cn
cWuhan Keqian Biology Co., Ltd, Wuhan, 430000, PR China
dInstitute of Food Safety and Nutrition, Jinan University, Guangzhou, 510632, PR China

Received 2nd December 2018 , Accepted 13th February 2019

First published on 19th February 2019


Outbreaks and potential epidemics of the highly pathogenic avian influenza virus pose serious threats to human health and the global economy. As such, its timely and accurate detection is critically important. In the present study, positive hybridoma cells (6B3) were obtained, which were used to secrete high-titer avian influenza virus (AIV) H7N9 monoclonal antibodies (H7N9 mAb). Based on these mAbs, quantum dot-based lateral flow immunochromatographic strips (QD-LFICS) were developed for AIV H7N9 detection. Under optimized conditions, results from a commercial fluorescent strip reader indicated that the limit of detection of QD-LFICS was 0.0268 HAU. To achieve rapid on-site testing, a mini 3D-printed read-out platform was fabricated to allow observation of QD-LFICS by the naked eye. More importantly, QD-LFICS were found to be practical and specific for the detection of actual samples compared with a real-time polymerase chain reaction.


1. Introduction

Influenza A is an acute respiratory-tract infection caused by the influenza-A virus, and is characterized by its contagiousness and ease of transmission. The type-A influenza virus is classified into subtypes according to its surface glycoproteins: hemagglutinin (HA) and neuraminidase (NA).1,2 Up until now, 17 HA subtypes and 10 NA subtypes have been found.3

HA is a critically important factor for viral infectivity; structurally, it comprises “head” and “stalk” regions.4,5 The head region is attached to sialic acid-containing receptors that are expressed on the surface of host cells. Continuous infection results in release of nucleocapsids into cytoplasm.4 Humans are infected mainly by the highly pathogenic H5, H7, and H9 subtypes of avian influenza viruses (AIVs), which carry a high mortality rate.6–8 For example, the H5N1 subtype of the AIV has a mortality rate >60%.6 The avian influenza A (H7N9) virus was first reported in China on 1 April 2013 by the World Health Organization (WHO). It is also highly pathogenic in humans, causing serious respiratory symptoms that resulted in the deaths of 39 patients by the end of 2013.9,10 Given these sobering statistics, timely detection and effective management of potentially epidemic infectious diseases are essential.

Much work has gone into developing capabilities for the detection and monitoring of AIVs: isolation and identification of viruses; hemagglutination assays/hemagglutination inhibition assays; enzyme-linked immunosorbent assays (ELISAs); real-time polymerase chain reaction (PCR); digoxin-labeled probes.11–13 Isolation and identification remain the “gold standard” for virus detection but require a complex and time-consuming process. With the rapid development of molecular-diagnostic methods, PCR has broad applications in AIV detection. In PCR, enzymatic reactions are used to amplify target sequences exponentially. However, high levels of amplification mean that trace amounts of contaminants could be also amplified exponentially as templates, thus resulting in possible false-positive results. Moreover, enzymatic amplification is susceptible to inhibitors and inherent biases, which may prevent accurate quantification.14–16 In addition, isolation of nucleic acids requires trained technicians, specialized instrumentation, and a clean environment.

Lateral flow immunochromatographic strips (LFICS) have been used for point-of-care testing because they are rapid, simple, affordable, and accurate.17,18 Gold nanoparticles (GNPs) and colored latex nanoparticles are often used as probes to develop colorimetric sensors.19,20 However, usually the color changes of LFICS do not provide sufficient signal intensity, and LFICS have low sensitivity. Considering that there is a low viral load in the early stages of infection, timely detection and effective management of potentially epidemic infectious diseases are essential.19,21 Compared with fluorescent dyes, quantum dots (QDs) have many advantageous physiochemical characteristics: high quantum yields (QYs), large molar extinction coefficients, tunable emission wavelengths, broad absorption cross-sections, and strong photostability.22–25 Based on those excellent optical properties, QDs have been used widely in medical diagnostics, molecular imaging, and chemical analyses.26–28 In this work, we screened a high-titer AIV H7N9 monoclonal antibody hybridoma cells, and developed a QD-based immunoassay to improve the performance of LFICS for H7N9 detection.

LFICS photographic devices are notorously bulky and inconvenient, which restricts application of LFICS for on-site detection. Due to their simplicity, low cost, and multifunctionality, three-dimensional (3D) printing has wide applications in biosensors, medical diagnoses, biomedicine, prostheses, and tissue engineering.29,30 With rapid development of advanced manufacturing technologies, fabrication of portable and cost-effective analytical devices has elicited global attention based on 3D printing. By direct-design workflow, prototyping technology that uses powdered metal or bondable plastic materials can be used to build a 3D object, layer by layer.31–33 In contrast with conventional expensive and complex devices, 3D printing-based instruments eliminate the need for dedicated equipment.34 Use of 3D printing technologies has become an increasingly popular tool in point-of-care testing.35 Roda et al. utilized a 3D printing method to print simple accessories to make a smartphone-based immunosensor with thermochemiluminescence detection that was optimized for valproic-acid detection in blood and saliva samples.36

Recently, our research team created surface-enhanced Raman scattering-based- lateral flow immunoassay strips (SERS-LFIAS).37 Although SERS-LFIAS have high sensitivity and good specificity, complex instrumentation and trained operators are needed, which has restricted their application, especially in resource-constrained regions. To meet the requirement of on-site detection of the AIV H7N9, we made a small, portable 3D-printed fluorescence read-out platform to observe LFICS results by the naked eye. Considering that the results will be ambiguous at lower virus concentrations, we designed the entire observation area of LFICS in a dark room (which avoided interference by external light). Also, by adding a filter, the designed 3D-printed read-out platform could filter-out excess light and reduce the background value. Here, a QD-based LFICS was integrated with a read-out platform for the AIV H7N9 on-site detection. Development of 3D-printed read-out platform could be used to observe 0.0312 hemagglutination units (HAU) of H7N9 viruses as well as that of a commercial immunochromatography strip (ICS) reader. This system was two orders of magnitude more sensitive than traditional colloidal gold ICSs.

2. Materials and methods

2.1. Ethics statement

Animal studies were carried out in strict accordance with the animal-welfare guidelines of the World Organization for Animal Health. Animal procedures were carried out in accordance with the Guidelines for the Care and Use of Laboratory Animals of Huazhong Agricultural University (Wuhan, China) and approved by the Hubei Provincial Animal Care and Use Committee (approval number SCXK 2015-0018). Real samples and biological samples were collected and provided from State Key Laboratory of Agricultural Microbiology (Huazhong Agricultural University) using standard procedures.

2.2 Reagents and materials

CdSe/ZnS quantum dot nanobeads (QDNBs) were purchased from Shanghai Kundao Biotech (Shanghai, China). 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) were obtained from Sigma-Aldrich (Saint Louis, MO, USA). H7N9 monoclonal antibody (H7N9 mAb) and real samples of AIV H7N9 were obtained from State Key Laboratory of Agricultural Microbiology (Huazhong Agricultural University). Inactivated AIV H7N9 and inactivated AIV subtypes H1, H5, and H9 were purchased from Harbin Weike Biotechnology Development Company (Harbin, China). Goat anti-mouse immunoglobulin (Ig)G antibody was obtained from Tianjin Sungene Biotech (Tianjing, China). Bovine serum albumin (BSA) was purchased from Sigma-Aldrich. Nitrocellulose (NC) membranes, conjugation pads, sample pads (glass fiber G2), plastic backing, and absorbent pads (H2) were obtained from Shanghai Jiening Biotech (Shanghai, China). Ultrapure water was generated using the Milli-Q™ Ultra-Pure System (Millipore, Burlington, MA, USA). Reagents were of analytical grade and used as received without further purification or modification.

2.3 Instruments

A centrifuge (Beckman Coulter, Krefeld, Germany) was used. A XYZ3200 series dispenser system was obtained from Bio-Dot Scientific Equipment (Burton, MI, USA). A programmable HGS201 strip cutter was purchased from Shanghai Jiening Biotech (Shanghai, China). Fourier-transform infrared (FT-IR) spectrometry was done using a Vertex 70 spectrometer (Bruker, Karlsruhe, Germany). A fluorescent strip reader was purchased from Guangzhou Lanbo Biotechnology (Guangzhou, China). A 3D-printed read-out platform was assembled by our research team.

2.4 Preparation of H7N9 mAb-QDNBs

Preparation of H7N9 mAbs is described in detail in the ESI. Positive hybridoma cells (6B3) were obtained to produce H7N9 mAbs. H7N9 mAb-QDNBs were conjugated in a specific manner. Briefly, 45 μL of QDNBs were added to 405 μL of double-distilled (dd)H2O. To activate carboxyl groups, 8 μL of EDC (1 mg mL−1) was added to the solution, followed by 4.8 μL of NHS (1 mg mL−1) and mixed for 30 min. Then, 30 μL of H7N9-mAb was added to the solution. The solution was incubated for 1 h, after which 45 μL of 10% BSA was added and incubation allowed for an additional 2 h. Finally, the mixture was centrifuged at 13[thin space (1/6-em)]000 rpm for 30 min to remove unconjugated H7N9 mAb. After removal of the supernatant, the sediment was solubilized in 450 μL of buffer and stored at 4 °C for later use.

2.5 QD-LFICS fabrication for use in H7N9 detection

QD-LFICS were fabricated using a sample pad, conjugation pad, NC membrane, absorption pad, and plastic backing. A dispenser system was used to load a set volume of H7N9 mAb-QDNBs onto the conjugation pad, which was then dried at 37 °C. The dispenser system was used to distribute 1 μL cm−1 of H7N9 mAb (1.0 mg mL−1) and 1 μL cm−1 of goat anti-mouse IgG Ab (1.0 mg mL−1) to a designated area of the NC membrane as the test line (“T-line”) and control line (“C-line”). The NC membrane was dried at 37 °C for further use. Then, all of the parts were assembled on the plastic backing in the following order: sample pad, conjugation pad, NC membrane, and absorption pad. Each part had an overlap of 2 mm to ensure that the applied liquid would migrate continuously through the entire QD-LFICS during testing. Finally, QD-LFICS were cut to a width of 3.8 mm using a programmable HGS201 strip cutter and placed in a plastic case for later use.

2.6 Mechanism of the 3D-printed read-out platform

The 3D-printed read-out platform were designed by SolidWorks software and then fabricated using 3D printing. As shown in Fig. 1, the sample was added to the sample hole to allow the reaction to occur. After the reaction was complete, the strip was placed into the platform. The excitation light was turned on and the results were observable readily through the filter. All QD-LFICS results were recorded using the CCD camera of a mobile smartphone.
image file: c8an02336k-f1.tif
Fig. 1 3D-printed read-out platform.

2.7 QD-LFICS performance for detection of the H7N9 virus

A series of different concentrations diluted using 100 μL of PBS containing 1 μL of the H7N9 virus were added to the sample pad. The liquid migrated towards the absorption pad by capillary action. Use of the 3D-printed read-out platform allowed us to observe preliminary results directly with the naked eye. To evaluate the sensitivity of our QD-LFICS, a professional fluorescence ICS reader was used to record the results after 15 min.

The specificity of QD-LFICS was evaluated using antigens from the H1, H5, and H9 subtypes of AIVs as well as the infectious bronchitis virus (IBV), Newcastle disease virus (IDV), and infectious bursal disease virus (IBDV). Each of these antigens were diluted to 0.25 HAU with PBS and added individually to the QDs-LFICS. After 15 min, results were recorded using a commercially available fluorescence ICS reader.

2.8 QD-LFICS performance for detection of real samples

Fifty H7N9 samples from different chicken organs, including cloacal swabs and throat swabs, were collected. They were added to 1% formaldehyde to inactivate the virus at 4 °C overnight. After dilution, the samples, with antibiotics, were tested using QD-LFICS to evaluate its practicality for detection of real samples. The results were compared with those obtained using reverse transcription-polymerase chain reaction (RT-PCR).

3. Results and discussion

3.1 Mechanism of QDs-LFICS

QD-LFICS comprised five constituent parts (from left to right in Fig. 2): a sample pad for adding liquid samples, a conjugate pad for combining mAb-QDNBs, a NC membrane to allow loading of the H7N9 antibody and goat-anti-mouse IgG, an absorbent pad for liquid absorption, and a plastic backing for assembling the components mentioned above. When the liquid sample was added to the sample pad, it migrated towards the conjugate pad; when the H7N9 mAb-QDNBs were dissolved and added, they migrated towards the absorbent pad. If the liquid sample contained the H7N9 virus, the latter would first bind to H7N9 mAb-QDNBs, and then combine with the H7N9-mAb on the T-line to form a double-antibody sandwich ELISA reaction (positive sample). Residual uncombined H7N9 mAb-QDNBs would bind to the goat anti-mouse IgG on the C-line. In the absence of AIV H7N9 (negative sample), H7N9 mAb-QDNBs could not be captured on the T-line and only a fluorescent signal was observed on the C-line.
image file: c8an02336k-f2.tif
Fig. 2 Detection of the H7N9 virus (schematic). Each strip was composed of one test line and one control line. Positive samples had a fluorescence signal at the T-line and C-line; negative samples had no fluorescence signal at the T-line.

3.2 Design of the 3D-printed read-out platform

The 3D-printed read-out platform consisted of a battery case, a core component, and shell (Fig. 3). All parts were designed by SolidWorks software and then fabricated using 3D printing. The top of the 3D-printed read-out platform was a 600 nm optical filter, which was used as a visual observation window; the interior was composed of two light-emitting diode (LED) lights (3 W, 365 nm), and the latter were added to a condenser lens to realize the gathering light as an exciting light source. The bottom of the 3D-printed read-out platform was a battery box. The whole device was supported by a 9 V lithium ion battery (Fig. S1). The optical filter could filter-out excess light (which could reduce the background value) and the LED lights with a specific wavelength were used to excite the QDs. The LED lights and battery were connected through a wire, and a switch was used to control the circuit. The wire connection was attached using a melt adhesive without the current shorting-out.
image file: c8an02336k-f3.tif
Fig. 3 3D-printed read-out platform (schematic). (a) Shell of the 3D-printed read-out platform; (b) the interior was composed of core component and optical filter; (c) the bottom was a battery box.

3.3 Characterization of QDNBs

Transmission electron microscopy (TEM) revealed that the average size of QDNBs was ∼80 nm (Fig. 4a). Energy-dispersive spectroscopy (EDS) was carried out to analyze the elemental composition of QDNBs. EDS indicated the peaks of Cd, Se, Zn, and S due to a CdSe core and ZnS shell (Fig. S2). Due to the polystyrene encapsulation of QDs, there were obvious peaks for C and O elements. Next, we studied the UV/Vis absorption spectra and fluorescence emission spectra of QDNBs. The UV-vis spectrum of mAb-QDNBs showed an obvious absorption band at ∼280 nm, which was attributed to the absorption peak of mAbs. The emission peak of QDNBs was centered at 605 nm, and there was no significant change in the emission peak of the mAb-QDNBs (Fig. 4b). The particle size of mAb-QDNBs was increased slightly compared with that of QDNBs, indicating that the mAb was labeled on the QDNBs surface (Fig. 4c). The surface functionalities of QDNBs was revealed by FT-IR spectroscopy. As shown in Fig. 4d, the peaks around 3303 cm−1 and 1645 cm−1 coincided with the stretching vibrations of N–H and C[double bond, length as m-dash]O groups, indicating the existence of the CO–NH– group in mAb-QDNBs. Compared with the FT-IR spectra of mAbs and QDNBs, the new stretching vibrations of N–H and C[double bond, length as m-dash]O groups appeared in mAb-QDNBs, further demonstrating that mAbs had been conjugated with QDNBs. Collectively, these data showed that the fluorescence emission peaks of QDNBs were narrow and stable, and that mAb-QDNBs fluorescent probes had been prepared.
image file: c8an02336k-f4.tif
Fig. 4 Characterization of QDNBs. (a) TEM images of QDNBs; (b) corresponding UV-vis and photoluminescence emission spectra of QDNBs and mAb-QDNBs; (c) changes in particle size of QDNBs before and after mAb labeling. (d) FT-IR spectra of QDNBs, mAbs and mAb-QDNBs.

To further confirm the stability of mAbs after conjugation with QDNBs, scanning electron microscopy (SEM) images for the corresponding of H7N9 virus (1 HAU) were recorded. In the absence of AIV H7N9, no immunocomplexes were formed in the test zone (Fig. 5a). For a positive sample (0.5 HAU H7N9), immunocomplexes were observed clearly in the NC membrane (Fig. 5b). Those results indicated that QDs-LFICS had excellent performance for detection of AIV H7N9.


image file: c8an02336k-f5.tif
Fig. 5 Fluorescence spectra and SEM images of QD-LFICS (a) in the absence of AIV H7N9 and (b) in the presence of AIV H7N9 (1 HAU).

3.4 Conditions and feasibility for use of QD-LFICS

The fluorescence intensity of the reaction was measured at different time points. The maximum fluorescence intensity was reached and plateaued after 15 min (Fig. 6a). Therefore, the optimum reaction time was determined to be 15 min. In addition, different concentrations of H7N9 mAb-QDNBs were dispensed onto the conjugation pad by a dispenser system. A total of 1 HAU of the H7N9 virus was added to QD-LFICS and the fluorescence intensity on the T-line was analyzed. As shown in Fig. 6b, T-line fluorescence increased with increasing concentration of the fluorescent probe. The correlation coefficient (R2) was 0.992, indicating a strong, positive correlation between fluorescence intensity and the concentration of H7N9 mAb-QDNBs.
image file: c8an02336k-f6.tif
Fig. 6 (a) The reaction time and corresponding fluorescence intensity. (b) Correlation between Ab-QDNBs and fluorescence intensity. (c) Optimization of anti-H7N9-Ab concentration on the T-line and labeled H7N9-mAb on QDNBs (d).

To improve the performance of QD-LFCS, different concentrations of H7N9-mAb on the T-line were optimized. After reacting with 1 HAU of the H7N9 virus, the fluorescence intensity was measured. As shown in Fig. 6c, the fluorescence signal of the QD-LFICS reached a plateau at a H7N9 mAb concentration of 1.0 μg cm−1, and there was no non-specific adsorption in the control group. Therefore, the concentration of the T-line was set to 1.0 μg cm−1. Furthermore, we optimized the amount of H7N9 mAb labeled on QDNBs. Different amounts of H7N9 mAb were added to label the QDNBs, and the fluorescence intensity was measured subsequently using the strip test. Afterwards, the H7N9 virus was added to the QD-LFICS, and the strip was run and tested with a fluorescence reader. As shown in Fig. 6d, the fluorescence intensity was highest when the 30 μL of H7N9 mAb was used to bind QDNBs. Hence, we used 30 μL of H7N9 mAb as the optimal labeling amount to prepare mAb-QDNBs.

3.5 QD-LFICS performance for detection of the H7N9 virus

Next, we sought to analyze the performance range of QD-LFICS for detection of the H7N9 virus. To achieve this aim, a series of concentrations of the H7N9 virus were analyzed under optimal conditions. Finally, fluorescent signals on the T-line and C-line of the LFICS could be observed by the naked eye using our fabricated 3D-printed read-out platform. As shown in Fig. 7a, the fluorescent images from the 3D-printed read-out platform were captured using a mobile smartphone. The fluorescence intensity on the T-line became ambiguous if the H7N9 virus concentration was 0.015 HAU. To evaluate the sensitivity of QD-LFICS for detection of the H7N9 virus, the fluorescence intensity of the T-line and C-line was read using a commercial fluorescence ICS reader. As shown in Fig. 7b, the fluorescence peaks indicated that the fluorescence intensity on the T-line was associated clearly with the concentration of the H7N9 virus. Furthermore, we took the peak integral area on the T-line as an ordinate variable and the concentration of the H7N9 virus as an independent variable to create a standard cure. As shown in Fig. 6c, the linear range of detection (LRD) for the H7N9 virus was 0.0312–0.5 HAU, whereas the limit of detection (LoD) was 0.0268 HAU. These results suggested that QD-LFICS were suitable and sensitive for detection of the H7N9 virus.
image file: c8an02336k-f7.tif
Fig. 7 QD-LFICS for H7N9 detection. (a) Fluorescence images from the 3D-printed read-out platform for different H7N9 concentrations. (b) Fluorescence peaks on the T- or C-lines of each test readout using a fluorescence ICS reader. (c) The standard curve used for H7N9 detection. Each value represents the mean of three replicates (n = 3).

Furthermore, we assessed the precision and stability of QD-LFICS for detection of the H7N9 virus. A series of concentrations of the H7N9 virus was assessed by analyzing the coefficient of variation (CV) between and within assays. The CV generated from triplicate experiments from duplicate QD-LFICS batches and the same batch indicated high stability (Table S1). More specifically, CVs were <12.8%, indicating the stability of QD-LFICS and their ability to be used for detection of the H7N9 virus.

3.6 Comparison of analytical methods

The comparison of different methods for AIV detection is shown in Table 1. Although ELISA was more sensitive, the process of ELISA was complex: protein coating, sample addition, washing, blocking, and enzyme/fluorescent labeling. Compared with other methods shown in Table 1, our method achieved a low LoD and did not need sophisticated instrumentation, such as a commercial biosensor for rapid influenza diagnostic tests, microplate readers, or impedance analyses. When compared to immunochromatographic strips for AIV detection, QD-LFICS were sensitive, portable and cost-effective without the need for complicated instruments. More importantly, QD-LFICS could be integrated with the 3D-printed read-out platform, thereby aiding application for rapid monitoring of potential outbreaks of influenza in poultry. Yeo et al.38 prepared eurpium nanoparticle-based rapid fluorescent immunochromatographic strips (FICTs) to detect the H7 subtype virus of an AIV. The LoD was 40 HAU, which was not as good as the sensitivity of our assay (0.0268 HAU). Therefore, these advantages highlight the convenience and sensitivity of QD-LFICS for detection of the H7N9 virus.
Table 1 Method comparison for AIV detection
Method Detection target LOD Instrument Time Ref.
Lateral flow strip
Europium nanoparticle H7 subtype virus 40 HAU Portable fluorescent strip reader 15 min 38
Red dye 53 H7N1 20 HAU Diagnostic device 10–15 min 39
Au/Fe3O4 core–shell nanoparticles AIV H7 103.5 EID50 By the naked eye 15 min 40
Colloidal gold H7N9 2.5[thin space (1/6-em)]log10[thin space (1/6-em)]EID50 By the naked eye 15 min 41
Surface-enhanced Raman scattering H7N9 HA 0.0018 HAU Portable Raman spectrometer RM-3000 15 min 37
QDs H7N9 HA 0.0268 HAU 3D printed read-out platform 15 min This work
 
Other methods
IM-SPR H7N9 HA 144 (copies per mL) IM-SPR biosensor <10 min 42
Digital ELIAS H7N9 AIV 7.8 fg mL−1 Microplate reader 2.5 h 43
Enzyme-induced metallization H7N9 AIV 25 pg mL−1 Microplate reader 30 min 10
Sandwich ELISA H7N9 NA 6.25 ng mL−1 Microplate reader 44
Colorimetric immunoassay H5N1 AIV 1.0 pg mL−1 Microplate reader 45
Fluorescence H5N1 AIV 0.4 HAU Portable fluorescence system <30 min 46
Impedance H5N1 AIV 0.5 HAU Impedance analyze 1 h 47


To better understand the sensitivity of QD-LFICS, we prepared traditional colloidal gold ICSs for detection of AIV H7N9 using identical materials. The visual signals of colloidal gold ICSs became obscured with decreasing AIV H7N9 concentration according to the naked eye (Fig. S3). The LRD of colloidal gold ICSs was 0.32–2.56 HAU (Fig. S4), and LoD was 0.32 HAU. Compared with colloidal gold ICSs, the sensitivity of QD-LFICS was 12-fold higher.

3.7 Analyses of QD-LFICS specificity

To evaluate the specificity of QD-LFICS, we detected 0.25 HAU of the following AIV antigens: H1, H5 and H9 subtypes of the H7N9 virus, as well as the IBV, IDV and IBDV. As shown in Fig. 8, only the AIV H7N9 virus produced a strong fluorescence signal; all other AIVs gave nearly negligible fluorescence signals. These results reflected the high specificity of QD-LFICS for detection of the AIV H7N9.
image file: c8an02336k-f8.tif
Fig. 8 Analyses of QD-LFICS specificity for detection of AIV H7N9.

3.8 QD-LFICS for detection of real samples

To investigate the applicability of QD-LFICS in real samples, samples of the H7N9 virus from different chicken organs were tested using QD-LFICS and real-time PCR. As shown in Table 2, the total coincidence rate between QD-LFICS and real-time PCR was 98%. The negative coincidence rate and positive coincidence rate between PCR and QD-LFICS was 100% and 94.7% respectively. The positive coincidence rate (94.7%) indicated that the sensitivity of QD-LFICS was slightly lower than that of RT-PCR. In general, as a novel immunoassay, QD-LFICS has great potential for use in the rapid diagnosis and prevention of AIVs.
Table 2 Sample tests using QD-LFICS and real-time PCR
Samples Real-time PCR
  Positive Negative Total Total coincidence rate (%)
QD-LFICS Positive 18 0 18 98
Negative 1 31 32
Total 19 31 50


4. Conclusion

A QDs-based immunoassay integrated with a 3D-printed read-out platform was developed to detect AIV H7N9. QD-LFICS was sensitive and specific for detection of the H7N9 virus. When compared with the traditional hemagglutination assay, QD-LFICS sensitivity was enhanced nearly 16-fold. In addition, results for detection of real samples were consistent with those from more conventional real-time PCR. Importantly, these results were also highly accurate. Most importantly, detection results could be observed using the 3D-printed read-out platform, which improved the capacity for rapid screening of AIVs. QD-LFICS was found to be rapid, simple, lightweight and portable. Hence, these qualities mean that QD-LFICS could be adapted for detecting H7N9 and preventing its spread. In the future, simple replacement of mAbs means that QD-LFICS could be used as versatile tools to detect other types of pathogens.

Ethics statement

All animal studies were carried out in strict accordance with the animal welfare guidelines of the World Organization for Animal Health. All animal procedures were undertaken in accordance with the Guidelines for Care and Use of Laboratory Animals of Huazhong Agricultural University (Wuhan, China) and approved by the Hubei Provincial Animal Care and Use Committee (approval number SCXK 2015-0018). Real samples and biological samples were collected and provided from State Key Laboratory of Agricultural Microbiology (Huazhong Agricultural University) by standard procedures.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

This work was supported by the National Key Research and Development Program of China (2016YFD0500600).

References

  1. V. S. Hinshaw, G. M. Air, A. J. Gibbs, L. Graves, B. Prescott and D. Karunakaran, J. Virol., 1982, 42, 865–872 CAS.
  2. R. C. Murdock, K. M. Gallegos, J. A. Hagen, N. Kelley-Loughnane, A. A. Weiss and I. Papautsky, Lab Chip, 2017, 17, 332–340 RSC.
  3. S. C. B. Gopinath, K. Awazu, M. Fujimaki and K. Shimizu, PLoS One, 2013, 8(12) CrossRef , e69121.
  4. M. Imai and Y. Kawaoka, Curr. Opin. Virol., 2012, 2, 160–167 CrossRef CAS PubMed.
  5. Y. A. Watanabe, T. Ito, M. S. Ibrahim, Y. Arai, K. Hotta and P. Hoang Vu Mai, et al. , Biosens. Bioelectron., 2015, 65, 211–219 CrossRef CAS PubMed.
  6. P. K. Chan, Clin. Infect. Dis., 2002, 34(Suppl. 2), S58 CrossRef PubMed.
  7. Y. P. Lin, M. Shaw, V. Gregory, K. Cameron, W. Lim and A. Klimov, et al. , Proc. Natl. Acad. Sci. U. S. A., 2000, 97, 9654–9658 CrossRef CAS PubMed.
  8. J. H. Han, D. Lee, C. H. C. Chew, T. Kim and J. J. Pak, Sens. Actuators, B, 2016, 228, 36–42 CrossRef CAS.
  9. S. C. B. Gopinath, T. H. Tang, Y. Chen, M. Citartan, J. Tominaga and T. Lakshmipriya, Biosens. Bioelectron., 2014, 61, 357–369 CrossRef CAS PubMed.
  10. Z. Wu, C. H. Zhou, J. J. Chen, C. Xiong, Z. Chen, D. W. Pang and Z. L. Zhang, Biosens. Bioelectron., 2015, 68, 586–592 CrossRef CAS PubMed.
  11. G. Boivin, I. Hardy and A. Kress, J. Clin. Microbiol., 2001, 39, 730–732 CrossRef CAS PubMed.
  12. L. Kaiser, M. S. Briones and F. G. Hayden, J. Clin. Virol., 1999, 14, 191–197 CrossRef CAS PubMed.
  13. S. Cui and G. Tong, J. Vet. Diagn. Invest., 2008, 20, 567–571 CrossRef PubMed.
  14. P. Yager, G. J. Domingo and J. Gerdes, Annu. Rev. Biomed. Eng., 2008, 10, 107 CrossRef CAS PubMed.
  15. H. T. Ngo, N. Gandra, A. M. Fales, S. M. Taylor and V.-D. Tuan, Biosens. Bioelectron., 2016, 81, 8–14 CrossRef CAS PubMed.
  16. X. Fu, Z. Cheng, J. Yu, P. Choo, L. Chen and J. Choo, Biosens. Bioelectron., 2016, 78, 530–537 CrossRef CAS PubMed.
  17. M. J. Raeisossadati, N. M. Danesh, F. Borna, M. Gholamzad, M. Ramezani, K. Abnous and S. M. Taghdisi, Biosens. Bioelectron., 2016, 86, 235–246 CrossRef CAS PubMed.
  18. L. M. Hu, K. Luo, J. Xia, G. M. Xu, C. H. Wu, J. J. Han, G. G. Zhang, M. Liu and W. H. Lai, Biosens. Bioelectron., 2017, 91, 95–103 CrossRef CAS PubMed.
  19. X. Gong, J. Cai, B. Zhang, Q. Zhao, J. Piao, W. Peng, W. Gao, D. Zhou, M. Zhao and J. Chang, J. Mater. Chem. B, 2017, 5, 5079–5091 RSC.
  20. L. G. Lee, E. S. Nordman, M. D. Johnson and M. F. Oldham, Biosensors, 2013, 3, 360–373 CrossRef CAS PubMed.
  21. I. Y. Goryacheva, P. Lenain and S. De Saeger, TrAC, Trends Anal. Chem., 2013, 46, 30–43 CrossRef CAS.
  22. Y. Volkov, Biochem. Biophys. Res. Commun., 2015, 468, 419–427 CrossRef CAS PubMed.
  23. T. R. Pisanic Ii, Y. Zhang and T. H. Wang, Analyst, 2014, 139, 2968–2981 RSC.
  24. M. Xiao, Q. Fu, H. Shen, Y. Chen, W. Xiao, D. Yan, X. Tang, Z. Zhong and Y. Tang, Talanta, 2018, 178, 644–649 CrossRef CAS PubMed.
  25. E. Morales-Narvaez, T. Naghdi, E. Zor and A. Merkoci, Anal. Chem., 2015, 87, 8573–8577 CrossRef CAS PubMed.
  26. S. Su, J. Fan, B. Xue, L. Yuwen, X. Liu, D. Pan, C. Fan and L. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 1152–1157 CrossRef CAS PubMed.
  27. J. Tian, L. Zhou, Y. Zhao, Y. Wang, Y. Peng and S. Zhao, Talanta, 2012, 92, 72–77 CrossRef CAS PubMed.
  28. Y. Suzuki, C. N. Roy, W. Promjunyakul, H. Hatakeyama, K. Gonda and J. Imamura, et al. , Mol. Cell. Biol., 2013, 33, 3036–3049 CrossRef CAS PubMed.
  29. C. G. Zhang, S. J. Chang, K. Settu, C. Jung Chen and J. T. Liu, Sens. Actuators, B, 2019, 279, 267–273 CrossRef CAS.
  30. M. Sharafeldin, A. Jones and J. Rusling, Micromachines, 2018, 9, 394 CrossRef PubMed.
  31. B. Elena, G. Andrea, I. Luca and G. V. Maria, Rapid Prototyping J., 2007, 13, 148–155 CrossRef.
  32. Y. Zhang, S. Ge and J. Yu, TrAC, Trends Anal. Chem., 2016, 85, 166–180 CrossRef CAS.
  33. G. I. Peterson, M. B. Larsen, M. A. Ganter, D. W. Storti and A. J. Boydston, ACS Appl. Mater. Interfaces, 2015, 7, 577–583 CrossRef CAS PubMed.
  34. C. L. Manzanares Palenzuela and M. Pumera, TrAC, Trends Anal. Chem., 2018, 103, 110–118 CrossRef CAS.
  35. A. Roda, M. Guardigli, D. Calabria, M. M. Calabretta, L. Cevenini and E. Michelini, Analyst, 2014, 139, 6494–6501 RSC.
  36. A. Roda, M. Zangheri, D. Calabria, M. Mirasoli, C. Caliceti, A. Quintavalla, M. Lombardo, C. Trombini and P. Simoni, Sens. Actuators, B, 2019, 279, 327–333 CrossRef CAS.
  37. M. Xiao, K. Xie, X. Dong, L. Wang, C. Huang, F. Xu, W. Xiao, M. Jin, B. Huang and Y. Tang, Anal. Chim. Acta, 2019, 1053, 139 CrossRef CAS PubMed.
  38. S. J. Yeo, B. Duong Tuan, G. E. Seo, B. Cuc Thi, D. T. H. Kim and A. Nguyen Thi Viet, et al. , Sci. Rep., 2017, 7, 7933 CrossRef PubMed.
  39. S. J. Yeo, B. T. Cuc, S. A. Kim, D. T. H. Kim, D. T. Bao, T. T. T. Tien, N. T. V. Anh, D. Y. Choi, C. K. Chong, H. S. Kim and H. Park, Biosens. Bioelectron., 2017, 94, 677–685 CrossRef CAS PubMed.
  40. J. Huang, Z. Xie, L. Xie, Z. Xie, S. Luo, X. Deng, L. Huang, T. Zeng, Y. Zhang, S. Wang and M. Zhang, RSC Adv., 2018, 8(25), 14064–14071 RSC.
  41. Z. H. Sun, B. L. Shi, F. F. Meng, R. N. Ma, Q. Y. Hu, T. Qin, S. J. Chen, D. X. Peng and X. F. Liu, Microbiol., 2018, 9, 8 Search PubMed.
  42. Y. F. Chang, W. H. Wang, Y. W. Hong, R. Y. Yuan, K. H. Chen and Y. W. Huang, et al. , Anal. Chem., 2018, 90, 1861–1869 CrossRef CAS PubMed.
  43. Z. Wu, W. J. Guo, Y. Y. Bai, L. Zhang, J. Hu and D. W. Pang, et al. , Anal. Chem., 2018, 90, 1683–1690 CrossRef CAS PubMed.
  44. Y. Yu, X. Zhang, B. Zhao, Y. Sun, X. Zhang and T. Bai, et al. , J. Virol. Methods, 2017, 247, 58–60 CrossRef CAS PubMed.
  45. S. Xu, W. Ouyang, P. Xie, Y. Lin, B. Qin and Z. Lin, et al. , Anal. Chem., 2017, 89, 1617–1623 CrossRef CAS PubMed.
  46. L. Xu, R. Wang, L. C. Kelso, Y. Ying and Y. Li, Sens. Actuators, B, 2016, 234, 98–108 CrossRef CAS.
  47. J. Lin, R. Wang, P. Jiao, Y. Li, M. Liao, Y. Yu and M. Wang, Biosens. Bioelectron., 2015, 67, 546–552 CrossRef CAS PubMed.

Footnotes

These authors contributed equally to the manuscript.
Electronic supplementary information (ESI) available. See DOI: 10.1039/c8an02336k

This journal is © The Royal Society of Chemistry 2019