Fabricating Upconversion Fluorescent Probes for Rapidly Sensing

Article pubs.acs.org/JAFC

Fabricating Upconversion Fluorescent Probes for Rapidly Sensing Foodborne Pathogens Wenxiu Pan, Jiewen Zhao, and Quansheng Chen* School of Food and Biological Engineering, Jiangsu University, Zhenjiang 212013, P. R. China

Downloaded by UNIV OF MANITOBA on September 2, 2015 | http://pubs.acs.org Publication Date (Web): September 1, 2015 | doi: 10.1021/acs.jafc.5b02331

S Supporting Information *

ABSTRACT: Rare earth-doped upconversion nanoparticles (UCNPs) have promising potential in the field of food safety because of their unique frequency upconverting capability and high detection sensitivity. Here, we report a rapid and sensitive UCNP-based bacterium-sensing strategy using Escherichia coli. Highly fluorescent and water-soluble UCNPs were fabricated and conjugated with antibodies against E. coli for use as fluorescent probes. The E. coli were successively captured by the fluorescent probes. After the captured cell samples were pelleted, the differences in the fluorescence intensities between sample supernatants and the control were observed to increase linearly with E. coli concentration from 42 to 42 × 106 colony-forming units (cfu)/mL (R2 = 0.9802), resulting in a relatively low limit of detection of 10 cfu/mL. Furthermore, the ability of the bioassay to detect E. coli was also confirmed in adulterated meat and milk samples. KEYWORDS: bacteria, sensing, upconversion nanoparticles, fluorescent probe, Escherichia coli

■

reagents are often used for colorimetric assays.15,16 Alternatively, fluorescence-based assays are rapid and technically simple and can efficiently determine bacterial counts from preenriched samples.17 However, these assays also have drawbacks. The organic dyes used in the assays raise photobleaching, photodamage, and autofluorescence issues.18 Efforts to circumvent these issues have led to the use of quantum dots (QDs); however, the cytoxicity risks associated with heavy metal components in QDs, such as cadmium, remain a major concern for biological applications.19 Similarly, because QD fluorescence is dependent on photon-based emissions, background autofluorescence remains a concern, as shorter wavelength light is used to excite the nanoparticles.20 Therefore, there exists a challenge to develop photostable, sensitive, and nontoxic bacterial labels that can overcome the current limitations. Recently, rare earth (RE)-doped upconversion nanoparticles (UCNPs) have been identified for their unique upconverting capabilities. These nanoparticles emit efficient and sharp visible luminescence via the ladderlike energy levels of RE ions when excited with near-infrared (NIR) light.21−23 The NIR excitation wavelength of UCNPs is proximal to the optical window, from 650 to 950 nm, which is a range in which biological materials have low absorption and scattering coefficients24,25 and, subsequently, low autofluorescence. Hence, UCNP-based fluorescent probes have become appealing.26−28 The low light scattering and autofluorescence backgrounds can greatly improve the signal-to-noise ratio.29 Additionally, UCNPs have been reported to have significant photostability, which potentially enables the long-term monitoring of UCNP-labeled bacteria without significant signal intensity losses.20 Finally,

INTRODUCTION Food safety has emerged as an important global issue with implications for international trade and public health, as foodborne outbreaks from microbial contaminants, chemicals, and toxins have engendered major health crises.1,2 Pathogenic bacteria are able to survive and easily spread in diverse environments, posing serious threats to human health.3,4 Livestock and poultry are prominent reservoirs for a majority of foodborne pathogens. Furthermore, animal byproducts, such as feed supplements, may transmit pathogens to other animals.5 Thus, the presence of these pathogens poses considerable risks to humans, especially pregnant women, infants, children, the infirm, and individuals with compromised immune systems. Escherichia coli, Salmonella spp., and Staphylococcus spp. are considered to be the most common causes of foodborne illnesses.6 While the global incidence of foodborne disease is difficult to ascertain, 1.8 million people were reported to have died from diarrhea-related causes in 2005; a significant proportion of these cases can be attributed to biogeniccontaminated food and drinking water.7 These alarming trends and reports have highlighted the need for the rapid and accurate detection of foodborne pathogenic bacteria. Current approaches for foodborne pathogenic bacteria detection include culture- and colony-counting methods,8 polymerase chain reaction (PCR) assays,9 immunological techniques,10 and fluorescence-based assays.11−13 However, the culture-based methods require enrichment steps to increase the bacterial count to detectable levels.14 PCR assays often require sophisticated instruments, complicated operating processes, and expensive reagents and are somewhat unreliable because of the frequency of false positive results because of the amplification of dead cells. Immunological techniques, such as enzyme-linked immunosorbent assays, are typically lengthy, requiring 2−3 h for the assay process. Furthermore, the techniques require enrichment steps; their use of active enzymes requires special storage conditions, and carcinogenic © XXXX American Chemical Society

Received: May 11, 2015 Revised: August 23, 2015 Accepted: August 25, 2015

A

DOI: 10.1021/acs.jafc.5b02331 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article


Journal of Agricultural and Food Chemistry

Figure 1. Scheme of this proposed fluorescence bioassay platform. Eclipse fluorescence spectrophotometer (Varian Inc.) modified with an external 0−1300 mW adjustable continuous wave 980 nm laser (Beijing Hi-Tech Optoelectronic Co., Beijing, China). Fourier transform infrared (FT-IR) spectra were obtained using a Nicolet Nexus 470 Fourier transform infrared spectrophotometer (Thermo Electron Co.) with the KBr method. Synthesis of Water-Soluble Upconversion Nanoparticles (UCNPs). Oleic acid-capped NaYF4:Yb,Er UCNPs were synthesized according to the method reported in refs 32−34 with a few modifications. In a typical experiment, 2 mL of RECl3 [0.2 M; RE = Y (78%), Yb (20%), or Er (2%)] in methanol was added to a 50 mL flask containing 3 mL of oleic acid and 7 mL of 1-octadecene. The prepared solution was heated to 160 °C for 30 min and cooled to room temperature. Thereafter, 5 mL of a methanol solution of NH4F (1.6 mmol) and NaOH (1 mmol) was added, and the solution was stirred for 30 min. After the methanol evaporated, the solution was heated to 300 °C under argon for 1.5 h and cooled to room temperature. The resulting nanoparticles were precipitated by the addition of ethanol, collected by centrifugation, washed with methanol and ethanol several times, and finally dried in an oven at 60 °C. The obtained oleic acid-capped UCNPs dispersed well in nonpolar solvents. However, for biological applications, hydrophilic UCNPs are required for compatibility with biomolecules,35 such as antibodies. Hydrophobic UCNPs were surface-modified via a ligand exchange process as described in ref 20. Briefly, a mixture of 2 mmol of sodium citrate in 15 mL of diethylene glycol was first heated to 110 °C under argon for 30 min. Oleic acid-capped UCNPs (10 mg) were dispersed in cyclohexane. Toluene was added to the mixture, and the reaction mixture was heated to 160 °C to evaporate the cyclohexane and toluene. After the evaporation, the reaction mixture was further maintained at 160 °C for 3 h. Subsequently, water-soluble UCNPs were collected by centrifugation, washed with ethanol and ultrapure water several times, and finally dispersed in ultrapure water. Bioconjugation of UCNPs with Anti-E. coli Antibodies. The bioconjugation of UCNPs with anti-E. coli antibodies was performed by EDC/NHS linking chemistry. Here, UCNPs (5 mg/mL in water) were activated with EDC (25 μL, 2 mg/mL) and NHS (12.5 μL, 2

UCNPs have low toxicities in vitro and in vivo, which are therefore suitable for biological applications.30,31 The main objective of this study was to develop a photostable and selective fluorescent probe for rapidly sensing pathogens in food. Figure 1 presents the scheme of this proposed fluorescence bioassay platform. Specific procedures are outlined as follows. (1) Upconversion nanoparticles (UCNPs) were synthesized and functionalized. (2) The resultant water-soluble UCNPs were conjugated with anti-E. coli antibodies to produce biological fluorescent probes. (3) A fluorescence standard curve was prepared with different concentrations of E. coli. (4) Independent food samples were tested. As an efficient, specific, and technically simple biological probe, these selective sensors can be used for rapidly detecting pathogens in food.

■

MATERIALS AND METHODS

Reagents. The rare earth salts used in this work, including YCl3, YbCl3, and ErCl3, were of 99.99% purity and were purchased from Aladdin Industrial, Inc. (Shanghai, China). Methanol, ethanol, oleic acid, 1-octadecene, cyclohexane, toluene, sodium citrate, diethylene glycol, 1-ethyl 3-[3-(dimethylamino)propyl]carbodiimide hydrochloride (EDC), tryptone, beef extract, NaCl, agar, NH4F, and NaOH were of analytical grade and were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). N-Hydroxysuccinimide (NHS) was obtained from Adamas Reagent Co., Ltd. (Shanghai, China). E. coli strain O157:H7 (NCTC 12900-2) was purchased from American Type Culture Collection (Manassas, VA), and the corresponding antiE. coli antibody (ab75244) was purchased from Abcam Co., Ltd. Instruments. X-ray diffraction (XRD) measurements were performed using a D8-advance instrument (Bruker AXS, Ltd.). The size and morphology of UCNPs, UCNP antibodies, and UCNP antibody−E. coli complexes were determined using a JEM-2100HR transmission electron microscope (TEM, JEOL, Ltd.) operated at 200 kV. Upconversion fluorescence spectra were measured using a Cary B


Article


Journal of Agricultural and Food Chemistry mg/mL) for approximately 3 h at room temperature. The activated UCNPs were purified by centrifugation at 9350g for 15 min and redispersed in ultrapure water. Subsequently, 50 μg of anti-E. coli antibody was incubated with 1 mg of activated UCNPs. After being shaken overnight at 4 °C, unbound UCNPs were removed by centrifugation at 6000g for 5 min at 4 °C. The bioconjugated particles were washed once with ultrapure water and then redispersed in ultrapure water. The centrifugation parameters, i.e., 6000g for 5 min, were optimized in preliminary tests with purely water-soluble UCNPs. The obtained UCNP antibody precipitates were validated by FTIR spectra measurements. Immunoassay Procedure. Bacterial Culture and Preparation. Stock cultures of E. coli (NCTC 12900-2) were inoculated and grown in Luria-Bertani (LB) medium for 24 h at 37 °C. The cultures were pelleted by centrifugation at 5000g for 5 min, washed twice with ultrapure water, and redispersed in ultrapure water. Thereafter, serial 10-fold dilutions were prepared in ultrapure water to produce various concentrations [1−108 colony-forming units (cfu)/mL] of bacteria. Bacterial counts were determined using the colony counting method. Capturing Bacteria with UCNP Antibody Probes. The UCNP− antibody (100 μL, 1 mg/mL) probes were mixed with various concentrations (1−108 cfu/mL) of bacteria and incubated for 2 h at 37 °C with agitation. As a control, 500 μL of ultrapure water was also treated. Thereafter, the reaction suspensions were pelleted by centrifugation at 5000g for 5 min to separate redundant UCNP− antibody probes. The resulting supernatants were used to monitor changes in emission fluorescence intensity. Fluorescence Spectral Measurement. The fluorescence spectra of the obtained supernatants (from 400 to 700 nm) were measured with a fluorescence spectrophotometer equipped with a 980 nm excitation laser at 1.3 W. The intensity differences (ΔI = I0 − I) between the supernatants and the control (I0) were used to generate a standard curve against the bacterial concentration (log cfu/mL). Here, the 657 nm peak intensity emission wavelength was used.

crystals (JCPDS Standard Card No. 16-0334); no diffraction peaks corresponding to cubic phase crystals or other impurities were observed. Furthermore, successful bioconjugation, selectivities, sizes, and luminance and spectral properties of UCNPs before and after surface modification were validated by transmission electron microscopy (TEM) and by fluorescence spectral measurements, as presented in Figure 3. The TEM images confirmed the hexagonal UCNP structures and revealed that the particles were uniform with an average diameter of approximately 50 nm before and after surface modification and bioconjugation. The fluorescence spectra of the UCNPs showed the expected characteristic emission peaks at approximately 407, 542, and 657 nm upon NIR (980 nm) excitation, corresponding to blue, green, and red light, respectively (the naked-eye images in the inset show the visible intensity of the UCNPs). The peaks are ascribed to the transitions from the 2H9/2, 4S3/2, and 4F9/2 levels to the 4I15/2 ground state of the Er3+ ion.36−38 Additionally, the fluorescence properties were also retained, as both the UCNP antibodies and the UCNPs showed the same characteristic emission peaks upon NIR excitation. When mixed with bacteria, the fluorescence intensity of the supernatant decreased, as shown in Figure 3D. The inset TEM image shows electron-dense dark spheres with diameters of approximately 50 nm, which correspond to the sizes of the UCNPs; these spheres decorated the outer surface of the bacterial cells. Negligible intensity signals were observed within the cytoplasm of the UCNPlabeled bacteria, indicating that the UCNPs were not internalized. Figure 4 displays the Fourier transform infrared (FT-IR) spectra of UCNPs (A), UCNP−antibody probes (B), and UCNP−antibody−bacterium complexes (C), confirming the presence of antibodies on the UCNPs and the presence of UCNP−antibody probes on the bacteria. The water-soluble UCNPs presented with a single broad peak at 3427 cm−1, corresponding to the stretching vibration of hydroxide radicals (-OH). The characteristic peak at 1629 cm−1 is related to the asymmetric stretching vibration of carboxyl groups (-COOH) of the citrate ligands on the surface of the UCNPs. These two peaks indicated that the carboxyl groups from the ligand exchange were successfully modified on the surface of UCNPs to produce water-soluble UCNPs. When the EDC/NHSprepared antibodies were introduced, three characteristic peaks at 2360, 2335, and 1396 cm−1 appeared. The peaks at 2360 and 2335 cm−1 corresponded to methylene stretching vibrations (-CH2-). The peak at 1396 cm−1 corresponded to carboxyl stretching vibrations (COO-) due to the linking reaction between the water-soluble UCNPs and the antibodies. Furthermore, a new peak was observed at 1540 cm−1 upon comparison of the spectra of the UCNP−antibody−bacterium complex and the UCNP−antibody probe; this peak is attributed to the distinct amide I and amide II vibration modes characteristic of bacterial proteins and peptides. On the basis of these characterizations, the proposed UCNPbased method is suitable for sensing bacteria. Evaluation of Specificity. To evaluate the specificity of the fluorescent probe for E. coli, two commonly encountered foodborne pathogens, Staphylococcus aureus (ATCC 25923) and Salmonella typhimurium (ATCC 14028), were tested. The results are shown in Figure 5 (the concentrations of these bacteria were 102 cfu/mL, and the figure was plotted with a peak intensity at 657 nm). Both S. aureus and Sa. typhimurium induced negligible changes in the fluorescence intensity,

■

RESULTS AND DISCUSSION Characterization. In this work, bacterium-specific antibodies with high selectivity and sensitivity were conjugated onto

Figure 2. X-ray diffraction (XRD) pattern of oleic acid-capped UCNPs.

the surface of UCNPs to yield UCNP−antibody probes, as illustrated in Figure 1. Prior to the conjugation, the precursor UCNPs were first characterized by X-ray diffraction (XRD), as shown in Figure 2. Here, the diffraction peaks of the XRD pattern were identified as pure hexagonal β-phase NaYF4 C


Article



Figure 3. Fluorescence properties of oleic acid-capped UCNPs (A), water-soluble UCNPs (B), bioconjugated UCNPs (C), and UCNP−antibody− bacterium complexes (D). Insets show naked-eye photographs and TEM images of particles.

Figure 5. Specific selectivity evaluation of the proposed method for E. coli (102 cfu/mL) against other bacteria (102 cfu/mL).

antibody probes. The centrifugation parameters were previously optimized to maximize the separation of the complex without disrupting the UCNP−antibody probes. Thereafter, serial dilutions of the supernatants were prepared to examine the fluorescence spectra of the complexes. The 657 nm emission peak was used to monitor the E. coli concentration. In the absence of bacteria (i.e., the control), the fluorescence intensity of the supernatant was at a maximum. In the presence of bacteria, the decrease in the fluorescence intensities (ΔI) of the supernatants varied proportionally with the concentration of bacteria. Selections of fluorescence spectra of bacterium-containing supernatants (log cfu/mL) are shown in Figure 6; the fluorescence intensity rapidly decreased as the E. coli concentration increased from 42 to 42 × 106 cfu/mL. A strong linear correlation (R2 = 0.9802) was obtained between

Figure 4. Fourier transform infrared (FT-IR) spectra of UCNPs (A), UCNP−antibody probes (B), and UCNP−antibody−bacterium complexes (C).

whereas a significant change was observed for E. coli. These results show the specificity of the designed fluorescent probe for E. coli. Bacterial Detection. In a typical experiment, different concentrations of E. coli were incubated under agitation with the UCNP−antibody probes for 2 h at 37 °C. On the basis of the specificity of the antibody for the bacteria, UCNP− antibody−bacterium complexes were formed. The samples were subsequently concentrated and separated by centrifugation at 5000g for 5 min to separate the unbound UCNP− D


Article



Figure 6. (A) Typical recording output for detecting different concentrations of E. coli using the proposed method. (B) Standard curve of the decreased upconversion fluorescence intensity (ΔI = I0 − I) vs the concentration of E. coli.

can be produced on a large scale because of their high photostability and stored under appropriate conditions). Analytical Performance. For comparative purposes, the linear ranges and LODs of several fluorescence-based capturing methods for detecting pathogens are summarized in Table 1. The proposed UCNP-based method has a relatively low LOD compared with those of conventional fluorescence-based capturing methods. Conventional fluorescent dyes are subject to photobleaching, photodamage, and autofluorescence issues; QDs also generate background autofluorescence because a shorter excitation wavelength is used to excite the nanoparticles. The proposed method has overcome these problems and provides a wider detection range and higher sensitivity. Additionally, fluorescent dyes based on antibody−antigen interactions typically achieve better selectivity but cannot be prepared on large scales because of photobleaching and photodamage.44 The prepared UCNPs had reduced photodamage and prolonged photostability, i.e., properties suitable for industrial processing. Analytical Applications. To further evaluate the feasibility and reliability of the proposed UCNP-based method for practical applications, the designed fluorescent probes were used to detect E. coli in solid (pork) and liquid (milk) samples adulterated with specific amounts of E. coli. Prior to probing, a series of pretreatment procedures were performed. For the milk samples, 5 mL of each sample was centrifuged at 7000g for 10 min at 10 °C. The upper layer of cream was removed, and the milk was diluted with distilled water at a 1:20 ratio. Finally, the

Table 1. Comparison of the Linear Ranges and LODs of Several Selected Fluorescence-Capturing Methods for Bacterial Detection material

linear range (cfu/mL)

104−107 10−104 2.4 × 104 to 6.0 × 106 quantum dots (QDs) coated with 103−107 streptavidin CdTe QDs coated with antibody 3 × 102 to 3 × 107 CdSe/ZnS QDs coated with not mentioned antibody UCNPs 42−42 × 106

organic dye organic dye coated with antibody lysozyme AuNCs

LOD (cfu/mL)

ref

104 5 2 × 104

39 40 41

103

42

1 × 102

43

10

44

10

this work

the relative intensity of the upconversion fluorescence and the concentration of E. coli: ΔI = 31.044c + 52.39 (c = log cfu/mL). The sensitivity of the developed immunoassay was also investigated. The limit of detection (LOD) is defined as LOD = 3S0/K, where S0 is the standard deviation of n control measurements (n = 10) and K is the slope of the calibration graph. In this work, the LOD of the proposed E. coli-sensing method was 10 cfu/mL. The total detection time from the addition of the fluorescent probes to the detection solution to obtaining the final result was

Fabricating Upconversion Fluorescent Probes for Rapidly Sensing

Recommend Documents