Sentinel Wraps: Real-Time Monitoring of Food Contamination by Printing DNAzyme Probes on Food Packaging Hanie Youseﬁ,†,‡ M. Monsur Ali,§ Hsuan-Ming Su,† Carlos D.M. Filipe,*,‡ and Tohid F. Didar*,†,⊥,∥ †
Department of Mechanical Engineering, ‡Department of Chemical Engineering, §Biointerfaces Institute, ⊥School of Biomedical Engineering, and ∥Institute for Infectious Disease Research (IIDR), McMaster University, Hamilton, Ontario L8S 4L8, Canada S Supporting Information *
ABSTRACT: Here, we report the development of a transparent, durable, and ﬂexible sensing surface that generates a ﬂuorescence signal in the presence of a speciﬁc target bacterium. This material can be used in packaging, and it is capable of monitoring microbial contamination in various types of food products in real time without having to remove the sample or the sensor from the package. The sensor was fabricated by covalently attaching picoliter-sized microarrays of an E. coli-speciﬁc RNA-cleaving ﬂuorogenic DNAzyme probe (RFD-EC1) to a thin, ﬂexible, and transparent cyclo-oleﬁn polymer (COP) ﬁlm. Our experimental results demonstrate that the developed (RFD-EC1)COP surface is speciﬁc, stable for at least 14 days under various pH conditions (pH 3−9), and can detect E. coli in meat and apple juice at concentrations as low as 103 CFU/mL. Furthermore, we demonstrate that our sensor is capable of detecting bacteria while still attached to the food package, which eliminates the need to manipulate the sample. The developed biosensors are stable for at least the shelf life of perishable packaged food products and provide a packaging solution for real-time monitoring of pathogens. These sensors hold the potential to make a signiﬁcant contribution to the ongoing eﬀorts to mitigate the negative public-health-related impacts of food-borne illnesses. KEYWORDS: ﬂexible surface for biosensing, hands-free bacterial detection, food safety, real-time monitoring, packaged food biosensing, inkjet printing of ﬂuorogenic DNAzyme ccording to the World Health Organization (WHO),1 food-borne pathogens result in approximately 600 million illnesses and 420,000 death per year. Perhaps the most troubling aspect of these statistics is that approximately 30% of these deaths involve children younger than 6 years old. These illnesses account for 33 million disability-adjusted life years (DALY), which is a term that represents the number of healthy-living-years lost as a result of illness.1 This ﬁgure represents not only a signiﬁcant loss in quality of life but also a considerable increase in ﬁnancial burden for the aﬀected population. Bacteria are among the most signiﬁcant pathogens in terms of causing food-borne illnesses, and they are also the leading cause of foodcontamination-related deaths.2 Prior research has demonstrated that the majority of bacterial contamination occurs during the food distribution process as a result of poor storage conditions and the absence of adequate food quality monitoring methods.3,4 Thus, there is an urgent need to develop simple and convenient methods of detecting bacterial pathogens in packaged food, as such methods could be eﬀective in reducing food-borne illnesses and minimizing their associated public health costs.
© XXXX American Chemical Society
Conventional bacterial detection methods fall into three main categories: (1) culture and colony counting, (2) immunological analysis, and (3) nucleic acid sequence-based ampliﬁcation methods.5 These methods are advantageous as they provide high sensitivity, but they also have a number of drawbacks, such as requiring highly trained individuals and specialized analytical equipment, as well as being very timeconsuming. Over the past few years, researchers have attempted to address these problems by creating prototypes that use silica beads, antibodies, bacteriophages, and electrodes.6−9 Although these approaches have been successful in drastically improving the ease of bacterial detection, their application remains limited due to the need for active user interventions, such as sample extraction and cell lysing, or the need for complicated analytical techniques, such as Raman spectroscopy.6−9 Furthermore, a few “smart” detection systems, which are embedded within food packaging, have also been developed to continuously Received: November 12, 2017 Accepted: March 27, 2018
DOI: 10.1021/acsnano.7b08010 ACS Nano XXXX, XXX, XXX−XXX
Cite This: ACS Nano XXXX, XXX, XXX−XXX
Figure 1. Illustration of highly sensitive DNAzyme sensors cleaving in the presence of live E. coli cells. Amine-terminated DNAzyme probes were covalently attached to thin, ﬂexible, and transparent epoxy ﬁlms (left). The cleavage of the ribonucleotide connecting the ﬂuorogenic and quencher substrates of the DNAzyme in the presence of the target produced by live E. coli cells (right).
Table 1. Synthesized Oligonucleotides (5′-3′) Used To Create the Biosensing Agent (DNAzyme) name NH-EC1
LT FS1 RFD-EC1
sequence 5′-NH2TTTTTCACGGATCCTGACAAGGATGTGGTTGTCGAGAC CTGCGACCGGA ACACTACACTGTGTGGGATGGATTTCTTTACAGTTGTGTGCAGCTCCGTCCG-3′ 5′-CTAGGAAGAGTCGGACGGAGCTG-3′ ACTCTTCCTAGCFrAQGGTTCGATCAAGA (F, ﬂuorescein-dT; rA, riboadenosine; Q: dabcyl-dT) 5′-NH2TTTTTCACGGATCCTGACAAGGATGTGGTTGTCGAG ACCTGCGACCGGAACACTACACTGTGTGGGATGGATTTCTTTACAGTTGTGTGCAGCTCCGTCCG-3′ ACTCTTCCTAGCFrAQGGTTCGATCAAGA-3′
monitor food quality.10−13 However, while these devices monitor speciﬁc conditions associated with food spoilage (e.g., humidity, storage time, and presence of speciﬁc chemicals),10−13 they are not designed to monitor the presence of bacteria. Catalytic DNA molecules (DNAzymes) are synthetic, singlestranded DNA molecules that either have catalytic abilities14,15 or can perform speciﬁc reactions. 16−18 Among these DNAzymes, RNA-cleaving molecules are particularly promising for creating methods that can detect a wide variety of targets.19,20 Recently, RNA-cleaving ﬂuorescent DNAzymes (RFD) were generated by selecting speciﬁc bacteria in vitro and then optimized for real-time bacterial detection purposes.21−23 These DNAzymes cleave at a single ribonucleotide embedded within the ﬂuorogenic DNA substrate. The cleavage junction is ﬂanked by a ﬂuorophore and a quencher so that the intact substrate has a very low ﬂuorescence signal prior to cleavage. When the target bacterium is present, the substrate is cleaved by the action of DNAzyme; this cleavage creates separation between the ﬂuorophore and the quencher-labeled sequence, which in turn produces an enhanced ﬂuorescence signal. DNAzymes have been previously immobilized on magnetic beads,18 metal organic frameworks (MOFs),24 and liquid crystals25 in order to detect metal ions. However, there have been no previously documented attempts to attach DNAzymes to ﬂexible polymer-based surfaces. In addition, although previous research has shown that these DNAzyme sensors are responsive to crude extracellular mixtures (CEM)26 and crude intracellular mixtures (CIM)27 of target bacterium, their application for detecting whole bacteria cells has yet to be demonstrated.
Herein, we present a sentinel bacteria-monitoring system that (1) is simple and requires minimal user intervention, (2) is compatible with currently used packaging materials, (3) is stable during long-term detection (days or weeks),28 and (4) provides easily interpreted results. Furthermore, we demonstrate that our system, which uses a ﬂexible plastic surface coated with DNAzyme molecules (Figure 1), is eﬀective for detecting the presence of speciﬁc bacteria in packaged food. The DNAzyme used in this study, which we will refer to as RFD-EC1, has been previously reported to be highly speciﬁc to CIM and CEM of E. coli.21 We used inkjet printing to covalently couple the DNAzyme molecules present in picolitersized droplets to thin, transparent, and ﬂexible COP (cyclooleﬁn polymer) ﬁlms containing epoxy functional groups. COP ﬁlm was identiﬁed as an excellent candidate for biosensor fabrication due to its low ﬂuorescence background, transparency, high ﬂexibility, ease of functionalization, and low cost. The created sensing surfaces, which will be referred to as (RFD-EC1)-COP, were able to report the presence of live E. coli cells (without requiring lysis) in liquid and solid food samples with various pH levels. The developed reporting system is reliable for at least 2 weeks, which is consistent with the standard shelf life of perishable packaged food products.28
RESULTS AND DISCUSSION Fabrication, Characterization, and Stability of Flexible Sensors. Long-term stability is an important requirement for sensing devices used in food product packaging.29 As explained in the Methods section, the (RFD-EC1)-COP was fabricated by modifying RFD-EC1 with amine moieties at the 5′-end to facilitate covalent attachment with the epoxy groups on the ﬂexible COP surfaces. The cleavage site (FS1 in Table 1) was B
DOI: 10.1021/acsnano.7b08010 ACS Nano XXXX, XXX, XXX−XXX
Figure 2. Characterization of the epoxy-coated surfaces before and after DNA/DNAzyme immobilization. (a) Image of thin, ﬂexible, and transparent epoxy-coated COP ﬁlm. (b) Fluorescence image of surfaces coated with amine-modiﬁed DNA probes and control DNA probes before (left) and after (right) wash. (c) Fluorescence quantiﬁcation of immobilized DNA on the surface. (d) Contact angle measurements of the epoxy-coated surfaces before and after DNA immobilization. (e) XPS results for the percentage of nitrogen in the composition of the epoxy-coated surfaces before and after DNA immobilization. (f) Printed DNAzyme probe stability under diﬀerent pH conditions over a 10 day period.
signiﬁcant diﬀerence in the hydrophobicity of surfaces that had been incubated with amine-modiﬁed DNA and unmodiﬁed DNA. Although the hydrophilic nitrogenous bases in a DNA molecule can decrease the contact angle of the surface upon DNA immobilization, such a change could also possibly occur without successful immobilization due to the reaction between the surface epoxide groups and the water in the printing solution, which creates hydrophilic hydroxyl functional groups. Indeed, when the epoxy COP ﬁlms (without any immobilized DNA) were incubated in water, similar contact angles were observed, thus providing support for the above explanation. Overall, the DNA-coated surfaces showed less than 10° reduction in hydrophobicity compared to that of uncoated epoxy surfaces (Figure 2d). The percent compositions of atoms on the COP surfaces were studied using X-ray photoelectron spectroscopy (XPS). The epoxy-coated surfaces that had been incubated with amine-terminated DNA contained a signiﬁcantly higher percentage of nitrogen than the epoxy surfaces that had been incubated with unmodiﬁed DNAs. This ﬁnding conﬁrmed the successful covalent attachment of amineterminated DNA to the COP surfaces (Figure 2e). Next, amine-terminated RFD-EC1 was applied to the COP surfaces, and the covalent attachment was once again conﬁrmed (as explained in the Methods section). In addition, because diﬀerent food products have diﬀerent pH values, we also tested the stability of the attachment between RFD-EC1 and the COP ﬁlm at various pH levels. To this end, we incubated the immobilized probes in solutions with various pH levels (pH 3− 9) for 10 days to examine how harsh storage conditions aﬀect the stability of the covalent attachment and the DNAzyme’s structure (Figure 2f). The sensor was also incubated in NaOH as a positive control. Under such high pH conditions, the alkalilabile ribonucleotide in RFD-EC1 became hydrolyzed, and the
attached at the 3′-end of the DNAzyme, EC1 (Table 1), in order to cause the release of the quencher after the cleavage reaction. This reaction created an increase in the ﬂuorescence signal of the ﬂuorophore in FS1, which remained attached to the EC1 fragment. The enhanced ﬂuorescence signal could then be monitored using a ﬂuorescence imager/scanner (Figure 1). To minimize the amount of the RFD-EC1 substrate required and to ensure the scalability of the fabrication, an inkjet printer with droplet sizes of less than 450 pL was used to precisely and rapidly apply a small volume of RFD-EC1 onto the COP surface. The use of an inkjet printer was advantageous because it allowed for rapid, covalent, large-scale printing, as well as control over the amount of DNAzyme being applied, which might have been more diﬃcult to achieve using other methods, such as microcontact printing.30,31 Prior to immobilizing the RFD-EC1 onto the epoxy-coated COP surface (Figure 2a), covalent DNA conjugation was conﬁrmed by studying immobilization eﬃciency of ﬂuorogenic amine-terminated DNA, which can be visualized using a ﬂuorescence scanner. Using unmodiﬁed DNA as a control (that is, DNA without terminal amine), we then studied various surface characteristics before and after DNA immobilization. Fluorescence images of the surfaces before and after the washing procedures (explained in the Methods section) were obtained using a ChemiDoc imaging system (results shown in Figure 2b). When the ﬂuorescence signals were quantiﬁed (Figure 2c), it was observed that amine-modiﬁed DNA exhibited signiﬁcantly higher immobilization eﬃciency than unmodiﬁed DNA. The hydrophobicity of the COP surfaces was quantiﬁed via contact angle measurements (Figure 2d). Whereas the contact angle decreased signiﬁcantly upon DNA immobilization, analysis of variance (ANOVA) and subsequent Tukey’s tests indicated no C
DOI: 10.1021/acsnano.7b08010 ACS Nano XXXX, XXX, XXX−XXX
Figure 3. Cleavage and real-time activity test of the printed DNAzyme for monitoring E. coli on COP surfaces. (a) Top sections of the sensors were incubated with live E. coli cells, whereas the bottom sections were incubated in reaction buﬀer (n = 5). (b) Real-time monitoring of E. coli using (RFD-EC1)-COP. Samples were incubated with E. coli cells, with NaOH (alkaline treatment) as the positive control and reaction buﬀer as the negative control (n = 5).
following covalent attachment to COP. For this purpose, we performed a test using a variety of bacteria, including both Gram-positive (Pediococcus acidilactici (PA) and Bacillus subtilis (BS)) and Gram-negative bacteria (Yersinia ruckeri (YR) and Achromobacter xylosoxidans (AX)). The bacteria were grown overnight in tryptic soy broth (TSB) growth media, and the cells were then collected and suspended in reaction buﬀer. (RFD-EC1)-COP ﬁlms were cut into small pieces with an area of 1 cm2 and then incubated in the solutions for 2 h at room temperature. The sensors were then removed and washed, and the ﬂuorescence intensities of the surfaces were measured (Figure 4a). The results demonstrated that our sensors produce a signiﬁcantly higher ﬂuorescence signal in the presence of E. coli compared to that with other bacteria, which also conﬁrms
DNA fragment containing the quencher left the molecular assembly, thus causing the ﬂuorescence signal on the spot where RFD-EC1 was printed to be enhanced. The experimental results (Figure 2f) indicated that RFD-EC1 was stable between pH levels of 3 and 9, where the relative ﬂuorescence signal was signiﬁcantly lower than that of the DNAzyme that had been incubated in NaOH (chemical cleavage). This ﬁnding conﬁrms the stability of the covalent attachment and the consistency of the probe concentration and that the DNAzyme cleavage reaction had not taken place. Functionality after Immobilization and Real-Time Fluorescence Assay. To evaluate the response of RFD-EC1 after attachment to the COP surface, RFD-EC1-COP was incubated with E. coli (7.7 × 107 CFU/mL) and suspended in reaction buﬀer (50 mM HEPES, 150 mM NaCl, 30 mM MgCl2, 0.05% Tween 20) for 2 h at room temperature at pH 7.5. A negative control, wherein the DNAzyme-COP surface was incubated in the reaction buﬀer without E. coli, was also conducted. Following incubation, the surfaces were removed, washed, and imaged using the ﬂuorescence imager (ChemiDoc) to determine ﬂuorescence intensity. Figure 3a shows the ﬂuorescence intensities of the RFD-EC1-COP surfaces after incubation in various samples; as can be seen, the areas that had been exposed to E. coli had signiﬁcantly higher ﬂuorescence intensity. These results also indicate that the coupling process (e.g., printing, incubation, and washing) does not aﬀect the functionality of RFD-EC1. The ability of (RFD-EC1)-COP to monitor target bacterium in real time was investigated by exposing our sensor to live E. coli cells (7.7 × 107 CFU/mL in reaction buﬀer) and measuring the ﬂuorescence intensity of the surface at diﬀerent time points. The sensor was incubated in 1 M NaOH solution in order to determine the maximum cleavage signal, and a noninoculated reaction buﬀer was also used as the negative control; the ﬂuorescence intensity showed a 7-fold increase following 2 h of incubation in the presence of the bacterial target, which is similar to the increase observed with alkali cleavage (Figure 3b). Speciﬁcity and Sensitivity. Although RFD-EC1 has previously been reported as being speciﬁc for E. coli, we were interested in investigating whether this speciﬁcity was unaltered
Figure 4. (a) Speciﬁcity test of DNAzyme biosensors using Gramnegative and Gram-positive bacteria (n = 5). (b) Limit of detection of (RFD-EC1)-COP sensors. (RFD-EC1)-COP incubated in reaction buﬀer containing diﬀerent concentration of E. coli cells for 4 h and overnight (n = 3). The dotted line indicates no ﬂuorescence diﬀerence between buﬀer incubated and E. coli inoculated substrates (RF = 1). D
DOI: 10.1021/acsnano.7b08010 ACS Nano XXXX, XXX, XXX−XXX
ACS Nano that RFD-EC1 maintains its speciﬁcity toward E. coli cells after immobilization onto a substrate (Figure 4a). Previous applications of DNAzyme have produced various limits of detection (LOD) ranging from 103 to 105 CFU/ mL.26,32 However, these methods cannot be used in food packaging because they either require a high volume of sample, need further manipulation, or have yet to show that they are able to successfully covalently attach RFD-EC1 onto suitable substrates.33,34 Therefore, we were interested in using our (RFD-EC1)-COP sensors to determine the LOD of E. coli. To this end, we cultured E. coli cells in TSB media and then collected the live cells from the growth media before serial dilution to prevent bacterial growth during the incubation period. The on-package (RFD-EC1)-COP sensors were incubated with live E. coli in reaction buﬀer for either 4 h or overnight (Figure 4b). Overnight incubation of the sensor yielded an LOD of 103 CFU/mL, and a 4 h incubation period yielded an LOD of 104 CFU/mL. These results demonstrate that the sensing surfaces are capable of detecting small concentrations of bacteria (103 CFU/mL), which is promising in terms of applying them to monitor bacterial contamination in packaged food products. Monitoring Contamination of Food Samples. Because the main goal of this work was to determine whether (RFDEC1)-COP can be used for the continuous monitoring of contamination in packaged food products, we tested the performance of our package sensors in detecting and reporting E. coli in food and drinks. First, we checked the food samples for the presence of bacteria by culturing homogenized samples on TSA (tryptic soy agar) plates. Single colonies were then isolated and cultured on selective MacConkey agar. The results indicated that none of the isolated colonies was capable of lactose fermentation, which proves the absence of E. coli (Supporting Information Table S1). Additionally, we selected 20 diﬀerent colonies for 16S rRNA sequencing. Table S2 in the Supporting Information shows the list of bacteria that were initially found in our samples. This list, which was compiled based on the best hits for sequenced colonies, once again conﬁrmed the absence of E. coli in our samples. We then studied the stability of the (RFD-EC1)-COP ﬁlms by conducting tests to ensure that no false positive results would occur during long-term incubation with food samples containing no target bacterium. For this test, (RFD-EC1)-COP ﬁlms with an area of 1 cm2 were placed in contact with various solid food samples (raw beef and sliced apples) and on the inner wall of an apple juice container for 14 days at 4 °C and room temperature (Figure 5a). No signiﬁcant increase in ﬂuorescence intensity was observed for these surfaces, which demonstrates the high stability of the covalent attachment and the structure of the (RFD-EC1)-COP sensors over time and with diﬀerent food matrices that do not contain the target bacterium. In addition, we incubated the sensors in NaOH as a positive control and in reaction buﬀer (RB) as a negative control. Figure 5b shows that the signal associated with the food samples (noninoculated) was signiﬁcantly smaller than the signal associated with the full cleavage of RFD-EC1. Before spiking the food samples with bacteria, we serially diluted the E. coli and incubated it in reaction buﬀer for over 12 h. During this time, we monitored the reaction buﬀer solutions for turbidity in order to detect any bacterial growth during this incubation period (Supporting Information Figure S3). The results showed that, when E. coli was diluted in concentrations equal to or lower than 105 CFU/mL, there was no signiﬁcant growth
Figure 5. (RFD-EC1)-COP biosensors applied for monitoring contamination in real time. (a) Meat and apple samples wrapped in ﬁlm containing sensors. (b) Relative ﬂuorescence signal of the (RFD-EC1)-COP biosensors incubated in raw meat, sliced apple, apple juice, 1 M NaOH, and reaction buﬀer for 14 days at both 4 and 25 °C (n = 6). (c) Biosensors that had been stored with food for 14 days were incubated with E. coli-inoculated apple juice, meat, and sliced apple for 4 h. Fresh noninoculated food supply, 1 M NaOH, or reaction buﬀer was used as the control (n = 6). (d) Representative bright-ﬁeld (top) and ﬂuorescence (bottom) images of sensors on the solid food sample wrapped in food packaging. Food samples were inoculated with bacteria mixture (PA, BS, YR, AX) with E. coli (right) and the same bacteria mixture without E. coli (left) at 4 °C.
of bacteria in this buﬀer. The food samples were then spiked with a complex mixture of E. coli, PA, BS, YR, and AX cells (detection mixture), and another sample was spiked with the same mixture of bacteria, only without E. coli (control). Both sets of samples were then incubated at both 4 °C and room temperature for 4 h (Figure 5c). The total amount of bacteria in each food sample was 1 CFU/mg, and the sensors were only in contact with approximately 10% of the total area of the samples. This provided a substantial challenge for detection as not all of the bacteria (or CEM) would have been able to reach the slides, which is required to activate the sensor. Following incubation, the ﬁlms were tested for ﬂuorescence intensity and compared to the control samples (Figure 5c). We also imaged the sensors while they were still in contact with the food samples using a simple scanning system. This imaging system consisted of an SLR camera which provided a larger ﬁeld of view and enabled us to image the whole sensor on the food product (ﬂuorescence and bright-ﬁeld) (Figure 5d). These scans revealed that the (RFD-EC1)-COP surfaces in the samples that had been spiked with the bacterial mixture containing E. coli showed a signiﬁcant increase in ﬂuorescence intensity compared to the control samples (bacterial mixture without E. coli). Our experimental results conﬁrm that the sensors described in this paper possess great potential for use in food packaging and create surfaces that can speciﬁcally detect and report the presence of target bacteria in complex environments. E
DOI: 10.1021/acsnano.7b08010 ACS Nano XXXX, XXX, XXX−XXX
continuous agitation at 250 rpm. The ﬁnal concentrations of all the bacteria were adjusted to OD600 of ∼1. Live cells were collected by centrifugation at 5000 rpm for 5 min and then added to the reaction buﬀer (1× RB; 50 mM HEPES, 150 mM NaCl, 30 mM MgCl2, 0.05% Tween 20, pH 7.5) for use in the subsequent experiments. Preparation of RFD-EC1. NH-EC1 was enzymatically ligated to FS1 as follows: 500 pmol of FS1 was phosphorylated in 100 μL of 1× PNK buﬀer A, which contained 10 units of PNK enzyme, for 35 min at 37 °C. The enzyme was inactivated by ﬁrst heating it at 90 °C for 5 min and then cooling it to room temperature (RT) over a 20 min period. Next, an equal number of NH-EC1 and LT sequences was added to the FS1 mixture, vortexed, and spun down. The mixture was heated at 90 °C for 1 min and then cooled for 20 min until it reached RT. Twenty microliters of PEG4000 and 20 μL of T4 DNA ligase buﬀer were then added to this mixture. After adding 4 μL of T4 DNA ligase (20 units), the volume was adjusted to 200 μL by adding ddH2O. This solution was mixed by pipetting. The tube was then incubated at RT for 2 h, after which the DNA molecules were isolated via ethanol precipitation. The ligated DNA molecules (RFD-EC1) were puriﬁed via electrophoresis using 10% denaturing gel (dPAGE), dissolved in ddH2O, quantiﬁed using a nanoquant plate (TECAN), and stored at −20 °C until use. The ﬁnal concentration for the DNAzyme solution at time of storage was 3 μM. The functionality of the DNAzyme probes was tested prior to further processes by adding E. coli cell lysate to the mixture and measuring the increase in ﬂuorescence intensity throughout the incubation period. The produced DNAzyme probes were tested with bacteria samples to monitor the real-time functionality of our DNAzyme probes (Supporting Information Figure S2). Covalent Immobilization of RFD-EC1 and DNA onto the Surfaces. Five microliters of DNAzyme probes (15 pmol) was mixed with 5 μL of 2× printing buﬀer (autoclaved sodium phosphate buﬀer at pH 7.5). A Scienion printer was used to apply the reaction solution onto epoxy-coated ﬁlms (400 pL of 3 μM DNAzyme solution on each spot), which was followed by overnight incubation (14 h) at room temperature with 75% relative humidity. After the overnight incubation, the slides were washed thoroughly to make sure unattached DNAzymes were removed. The washing process consisted of a 2 min rinse cycle using autoclaved Milli-Q water, which was followed by a 1 min rinse cycle using PBS buﬀer with a pH of 7.5. The covalent attachment between the RFD-EC1 and the epoxy-COP substrates was conﬁrmed by comparing their ﬂuorescence intensity with that of two diﬀerent control substrates: in the ﬁrst group, DNAzyme without amine modiﬁcation was printed onto epoxy-coated ﬁlms; in the second group, RFD-EC1 was printed onto non-epoxycoated COP (Supporting Information Figure S1a). The same process was repeated for amine-modiﬁed DNA and unmodiﬁed DNA probes. Surface Characterization. To prepare the coated surfaces for characterization experiments (XPS and contact angle measurements), we submerged the surfaces in the desired solutions containing our probes (RFD-EC1, water, and DNA samples) in order to ensure that all of the surfaces were evenly coated and that the contact angle and XPS results would be reliable. Contact Angle Measurement. The contact angles of the water droplets on the substrates were measured using a contact angle measurement system obtained from Future Digital Scientiﬁc Corp. (Biointerface Institute, McMaster University). A microneedle was used to dispense 2 μL droplets of deionized (dI) H2O onto all substrates before and after DNA immobilization. X-ray Photoelectron Spectroscopy. XPS measurements were performed using a Physical Electronics (PHI) Quantera II spectrometer equipped with an Al anode source for X-ray generation, and a quartz crystal monochromator was used to focus the generated X-rays (Biointerfaces Institute, McMaster University). For XPS measurements, DNA was hand printed to cover a large surface area in order to allow for proper analysis. A minimum of three areas containing DNA was analyzed on each substrate. Food Sample Preparation and Initial Bacteria Identiﬁcation. Apple juice, apple chips, and beef slices were purchased from three diﬀerent local supermarkets and prepared for the food sample studies.
CONCLUSION The DNAzyme-based sensing surfaces described in this paper possess several characteristics that make them ideal for the realtime monitoring of contamination particularly in packaged food products and for creating target-speciﬁc sensing surfaces in general. These characteristics include (1) the ability to detect target bacterium without lysis; (2) eliminating the need for complicated user interventions, such as liquid handling, pipetting, conveying ﬂow, or the use of laboratory equipment; (3) real-time response to the presence of target bacterium; (4) maintaining an “ON” signal, which acts as a register of contamination, even if the target moves away from the sensor; and (5) high stability in a wide variety of food storage conditions. As DNAzyme probes can be generated through in vitro selection to react to diﬀerent bacteria, the RNA-cleavagebased detection mechanism described in this paper would be highly suitable for detecting a wide range of bacterial targets. Furthermore, the literature contains several reports that document low-cost attachments for cell phones that allow a user to detect ﬂuorescence signals.35,36 These new cell phone attachments, in combination with the sensing surfaces reported in this paper, provide a packaging solution that could potentially mitigate a signiﬁcant portion of the negative public health impacts associated with food-borne illness. Finally, in addition to detecting pathogens on packaged food or bottled drinks, the developed sensing surfaces could also be useful for a variety of other applications, for example, in food preparation surfaces or in materials for dressing wounds. METHODS Materials. The amine-modiﬁed and unmodiﬁed DNA and the ligation template oligonucleotides (NH-EC1 and LT in Table 1) were purchased from Integrated DNA Technologies (IDT), and the ﬂuorogenic substrate (FS1) was purchased from Yale University (sequence is provided in Table 1). Epoxy-coated COP foils were purchased from PolyAn molecular surface engineering. NaCl, MgCl2, Tween 20, Na2PO4, NaHPO4, tryptic soy broth, KCl, Na2CO3 (99.99%), NaHCO3, NaOH, and HEPES (4-(2-hydroxyethyl)-1piperazineethanesulfonic acid) were purchased from Sigma-Aldrich. ATP, PEG4000, T4 DNA ligase, polynucleotide kinase (PNK), and their respective buﬀers were purchased from Thermo Scientiﬁc, Canada. E. coli K12 strains are regularly maintained in our laboratory, whereas other bacteria types (PA, BS, YR, AX) were provided by Dr. Yingfu Li’s laboratory at McMaster University. In addition, PCR 16S rRNA primers were generously provided by Dr. Alexander Hynes’ laboratory at McMaster University. A ChemiDoc imaging system, Zeiss and Olympus microscopes, and an imaging setup consisting of an SLR camera and an appropriate ﬁlter were used to image the DNAzyme slides, and a Scienion FLEXARRAYER was used to print DNAzyme onto the epoxy-coated COP slides. Monarch DNA puriﬁcation kits were acquired from NewEngland BioLabs. Meat, apple, and apple juice samples, as well as Glad plastic wraps, were purchased from local supermarkets. Bacteria Preparation. E. coli K12, BS, YR, PA, and AX were cultured overnight (14 h at 37 °C and continuous agitation at 250 rpm) in TSB culture media. In order to measure the colony formation unit (CFU/mL), a fresh culture was conducted until the OD600 reached ∼1. Next, a serial dilution (10-fold) was conducted with 1 mL volume of the bacterial culture. Then, 100 μL from the dilution tubes was spread onto a TSA plate and incubated at 37 °C overnight. This process was done in triplicate. After incubation, the CFUs were then counted and averaged to obtain the number of CFU/mL. E. coli cell concentration within the culture was calculated as 7.7 × 108 CFU/ mL. Likewise, the other bacterial colonies were also transferred to TSA plates and incubated for 14 h at 37 °C. A single colony was then taken and inoculated into 2 mL of TSB and grown for 14 h at 37 °C with F
DOI: 10.1021/acsnano.7b08010 ACS Nano XXXX, XXX, XXX−XXX
ACS Nano Their pH values were measured and reported as follows: pH 3 for apple and apple juice, and pH 6 for meat. The samples were incubated with the prepared DNAzyme biosensors at two diﬀerent temperatures (4 and 25 °C) for 14 days in order to study their stability. The bacterial populations in the samples were characterized by culturing the homogenized samples on TSA plates, and the isolated colonies were subsequently cultured on MacConkey agar plates to assess lactose fermentation. At the same time, 16S rRNA from 20 selected colonies was ampliﬁed using PCR with universal rRNA primers following DNA extraction with the Monarch kit. Ampliﬁcation was veriﬁed via agarose gel electrophoresis, and the ampliﬁed regions were sequenced at the McMaster Genomic Facility. Following sequencing, the results were then processed using the bioinformatics software Geneious (R11, Biomatters) for the purpose of identifying bacterial species. After the 14 days storage period, the samples were taken out of the storage boxes, and the slides were ﬁrst washed with water before a second washing cycle with PBS at pH 7.5. After being washed, the ﬂuorescence intensity of the sensors was measured to make sure there was no cleavage or false positive signals. In addition, all homogenized samples were once again cultured on MacConkey agar after their storage time to conﬁrm the absence of E. coli. Food Contamination Test and Scanning Setup. To prepare the bacteria samples for the food contamination experiment, 105 CFU/mL of each bacteria sample was taken directly from the culture media and diluted in reaction buﬀer without any further processing. The serial dilutions were cultured on TSA plates to identify the cell concentrations (CFU/mL). Figure S3 in the Supporting Information shows that the bacteria did not continue to grow in the reaction buﬀer at a concentration of 105 CFU/mL. After 14 days of storage, the samples (food wrapped in our sensors) were incubated in one of two mixtures for 240 min: mixture one contained 105 CFU/mL of live E. coli, AX, BS, YR, and PA cells, and mixture two consisted of the same solution, only without E. coli (nontarget bacteria). The samples were spiked with 1 CFU/mg of bacteria. Following incubation, the sensors were imaged without being separated from the food using a simple broad view ﬂuorescence detection system for large view analysis, which was designed at McMaster University. The broad view ﬂuorescence detection system consisted of an SLR camera and a 3D-printed ﬁlter cube ﬁtted with a 492 nm excitation ﬁlter and a 525 nm emission ﬁlter. This system provided a large ﬁeld of view that allowed us to image both the whole sensor on the E. coli inoculated food and the uncleaved sensor on the noninoculated food (control) in the same frame. Statistical Analysis. Data are presented as means ± SD, and each experimental condition was repeated at least three times. Statistical signiﬁcance was assessed using one-way ANOVA tests, which were followed by posthoc analysis via Tukey’s tests. For all comparisons, P values less than 0.05 were considered statistically signiﬁcant.
Tohid F. Didar: 0000-0002-8757-8002 Author Contributions
T.F.D. and C.F conceived the research and supervised the experiments. H.Y. performed the experiments and analyzed the data with help from the other authors. H.Y. also wrote the paper with input from the other authors. M.A. helped with the design of the DNAzyme probes. H.S. helped with the preparation and immobilization of the DNAzyme probes and the data analysis. All authors have read, commented upon, and approved the ﬁnal version of this paper. Notes
The authors declare no competing ﬁnancial interest.
ACKNOWLEDGMENTS The authors would like to thank Dr. Yingfu Li for his signiﬁcant contributions to the development of the DNAzyme probes and for providing us with bacteria samples. We also want to thank Dr. Alexander Hynes for his generous help with the 16S ribosomal RNA sequencing and the analysis of the sequencing results. In addition, we are grateful to Dr. Ravi Selvaganapathy for providing us access to his lab’s imaging system. T.F.D. acknowledges the support from Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant, as well as start-up funds from McMaster University. Finally, we would like to thank the Canadian Foundation for Innovation, and the Ontario Ministry of Research and Innovation for Infrastructure funding to the Biointerfaces Institute at McMaster University. REFERENCES (1) World Health Organization. WHO Estimates of the Global Burden of Foodborne Diseases: Foodborne Disease Burden Epidemiology Reference Group 2007−2015; WHO, 2015; 255 pp. (2) World Health Organization. Estimating the Global Burden of Foodborne Diseases: A Collaborative Eﬀort= Evaluation De La Charge Mondiale Des Maladies D’origine Alimentaire: Une Action Concertée. Wkly. Epidemiol. Rec. 2009, 84, 203−211. (3) Mølbak, K.; Højlyng, N.; Jepsen, S.; Gaarslev, K. Bacterial Contamination of Stored Water and Stored Food: A Potential Source of Diarrhoeal Disease in West Africa. Epidemiol. Infect. 1989, 102, 309−316. (4) Kaneko, K.-I.; Hayashidani, H.; Ohtomo, Y.; Kosuge, J.; Kato, M.; Takahashi, K.; Shiraki, Y.; Ogawa, M. Bacterial Contamination of Ready-to-Eat Foods and Fresh Products in Retail Shops and Food Factories. J. Food Prot. 1999, 62, 644−649. (5) Singh, R.; Mukherjee, M. D.; Sumana, G.; Gupta, R. K.; Sood, S.; Malhotra, B. Biosensors for Pathogen Detection: A Smart Approach Towards Clinical Diagnosis. Sens. Actuators, B 2014, 197, 385−404. (6) Srivastava, S. K.; Hamo, H. B.; Kushmaro, A.; Marks, R. S.; Grüner, C.; Rauschenbach, B.; Abdulhalim, I. Highly Sensitive and Specific Detection of E. Coli by a Sers Nanobiosensor Chip Utilizing Metallic Nanosculptured Thin Films. Analyst 2015, 140, 3201−3209. (7) Tokel, O.; Yildiz, U. H.; Inci, F.; Durmus, N. G.; Ekiz, O. O.; Turker, B.; Cetin, C.; Rao, S.; Sridhar, K.; Natarajan, N.; et al. Portable Microfluidic Integrated Plasmonic Platform for Pathogen Detection. Sci. Rep. 2015, 5, 9152. (8) DuVall, J. A.; Borba, J. C.; Shafagati, N.; Luzader, D.; Shukla, N.; Li, J.; Kehn-Hall, K.; Kendall, M. M.; Feldman, S. H.; Landers, J. P. Optical Imaging of Paramagnetic Bead-DNA Aggregation Inhibition Allows for Low Copy Number Detection of Infectious Pathogens. PLoS One 2015, 10, e0129830. (9) Basu, P. K.; Indukuri, D.; Keshavan, S.; Navratna, V.; Vanjari, S. R. K.; Raghavan, S.; Bhat, N. Graphene Based E. Coli Sensor on Flexible Acetate Sheet. Sens. Actuators, B 2014, 190, 342−347.
ASSOCIATED CONTENT S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.7b08010. Additional data on DNAzyme covalent attachment and the stability tests, DNAzyme reactivity with bacterial CIM in solution, bacteria growth while diluted in reaction buﬀer, culture tests from food samples for detecting the presence of E. coli, and DNA sequencing results that list the bacteria that are initially available in food matrices (PDF)
AUTHOR INFORMATION Corresponding Authors
*E-mail: ﬁ[email protected]
. *E-mail: [email protected]
Carlos D.M. Filipe: 0000-0002-7410-3323 G
DOI: 10.1021/acsnano.7b08010 ACS Nano XXXX, XXX, XXX−XXX
ACS Nano (10) Kuswandi, B.; Restyana, A.; Abdullah, A.; Heng, L. Y.; Ahmad, M. A Novel Colorimetric Food Package Label for Fish Spoilage Based on Polyaniline Film. Food Control 2012, 25, 184−189. (11) Tan, E. L.; Ng, W. N.; Shao, R.; Pereles, B. D.; Ong, K. G. A Wireless, Passive Sensor for Quantifying Packaged Food Quality. Sensors 2007, 7, 1747−1756. (12) Nopwinyuwong, A.; Trevanich, S.; Suppakul, P. Development of a Novel Colorimetric Indicator Label for Monitoring Freshness of Intermediate-Moisture Dessert Spoilage. Talanta 2010, 81, 1126− 1132. (13) Wang, S.; Liu, X.; Yang, M.; Zhang, Y.; Xiang, K.; Tang, R. Review of Time Temperature Indicators as Quality Monitors in Food Packaging. Packag. Technol. Sci. 2015, 28, 839−867. (14) Mori, M.; Eki, T.; Takahashi-Kudo, M.; Hanaoka, F.; Ui, M.; Enomoto, T. Characterization of DNA Synthesis at a Restrictive Temperature in the Temperature-Sensitive Mutants, Tsft5 Cells, That Belong to the Complementation Group of Ts85 Cells Containing a Thermolabile Ubiquitin-Activating Enzyme E1. Involvement of the Ubiquitin-Conjugating System in DNA Replication. J. Biol. Chem. 1993, 268, 16803−16809. (15) Gong, L.; Zhao, Z.; Lv, Y.-F.; Huan, S.-Y.; Fu, T.; Zhang, X.-B.; Shen, G.-L.; Yu, R.-Q. Dnazyme-Based Biosensors and Nanodevices. Chem. Commun. 2015, 51, 979−995. (16) Lu, Y.; Liu, J. Functional DNA Nanotechnology: Emerging Applications of Dnazymes and Aptamers. Curr. Opin. Biotechnol. 2006, 17, 580−588. (17) Yin, B.-C.; Zuo, P.; Huo, H.; Zhong, X.; Ye, B.-C. Dnazyme SelfAssembled Gold Nanoparticles for Determination of Metal Ions Using Fluorescence Anisotropy Assay. Anal. Biochem. 2010, 401, 47−52. (18) Zhang, H.; Lin, L.; Zeng, X.; Ruan, Y.; Wu, Y.; Lin, M.; He, Y.; Fu, F. Magnetic Beads-Based Dnazyme Recognition and Aunps-Based Enzymatic Catalysis Amplification for Visual Detection of Trace Uranyl Ion in Aqueous Environment. Biosens. Bioelectron. 2016, 78, 73−9. (19) Breaker, R. R.; Joyce, G. F. A DNA Enzyme That Cleaves Rna. Chem. Biol. 1994, 1, 223−229. (20) He, S.; Qu, L.; Shen, Z.; Tan, Y.; Zeng, M.; Liu, F.; Jiang, Y.; Li, Y. Highly Specific Recognition of Breast Tumors by an Rna-Cleaving Fluorogenic Dnazyme Probe. Anal. Chem. 2015, 87, 569−77. (21) Aguirre, S. D.; Ali, M. M.; Kanda, P.; Li, Y. Detection of Bacteria Using Fluorogenic Dnazymes. J. Visualized Exp. 2012, e3961. (22) Li, Y. Advancements in Using Reporter Dnazymes for Identifying Pathogenic Bacteria at Speed and with Convenience. Future Microbiol. 2011, 6, 973−976. (23) Zhang, W.; Feng, Q.; Chang, D.; Tram, K.; Li, Y. In Vitro Selection of Rna-Cleaving Dnazymes for Bacterial Detection. Methods 2016, 106, 66−75. (24) Chen, M.; Gan, N.; Zhou, Y.; Li, T.; Xu, Q.; Cao, Y.; Chen, Y. A Novel Aptamer-Metal Ions-Nanoscale Mof Based Electrochemical Biocodes for Multiple Antibiotics Detection and Signal Amplification. Sens. Actuators, B 2017, 242, 1201−1209. (25) Liao, S.; Ding, H.; Wu, Y.; Wu, Z.; Shen, G.; Yu, R. Label-Free Liquid Crystal Biosensor for L-Histidine: A Dnazyme-Based Platform for Small Molecule Assay. Biosens. Bioelectron. 2016, 79, 650−655. (26) Ali, M. M.; Aguirre, S. D.; Lazim, H.; Li, Y. Fluorogenic Dnazyme Probes as Bacterial Indicators. Angew. Chem., Int. Ed. 2011, 50, 3751−4. (27) Aguirre, S. D.; Ali, M. M.; Salena, B. J.; Li, Y. A Sensitive DNA Enzyme-Based Fluorescent Assay for Bacterial Detection. Biomolecules 2013, 3, 563−77. (28) Wilson, G. S.; Gifford, R. Biosensors for Real-Time in Vivo Measurements. Biosens. Bioelectron. 2005, 20, 2388−2403. (29) Scott, A. O. Biosensors for Food Analysis; Elsevier, 1998. (30) Didar, T. F.; Bowey, K.; Almazan, G.; Tabrizian, M. A Miniaturized Multipurpose Platform for Rapid, Label-Free, and Simultaneous Separation, Patterning, and in Vitro Culture of Primary and Rare Cells. Adv. Healthcare Mater. 2014, 3, 253−260.
(31) Didar, T. F.; Foudeh, A. M.; Tabrizian, M. Patterning Multiplex Protein Microarrays in a Single Microfluidic Channel. Anal. Chem. 2012, 84, 1012−1018. (32) Tram, K.; Kanda, P.; Salena, B. J.; Huan, S.; Li, Y. Translating Bacterial Detection by Dnazymes into a Litmus Test. Angew. Chem., Int. Ed. 2014, 53, 12799−12802. (33) Ali, M. M.; Brown, C. L.; Jahanshahi-Anbuhi, S.; Kannan, B.; Li, Y.; Filipe, C. D.; Brennan, J. D. A Printed Multicomponent Paper Sensor for Bacterial Detection. Sci. Rep. 2017, 7, 12549. (34) Kang, D.-K.; Ali, M. M.; Zhang, K.; Huang, S. S.; Peterson, E.; Digman, M. A.; Gratton, E.; Zhao, W. Rapid Detection of Single Bacteria in Unprocessed Blood Using Integrated Comprehensive Droplet Digital Detection. Nat. Commun. 2014, 5, 5427. (35) Zhu, H.; Yaglidere, O.; Su, T.-W.; Tseng, D.; Ozcan, A. CostEffective and Compact Wide-Field Fluorescent Imaging on a CellPhone. Lab Chip 2011, 11, 315−322. (36) Zhu, H.; Sikora, U.; Ozcan, A. Quantum Dot Enabled Detection of Escherichia Coli Using a Cell-Phone. Analyst 2012, 137, 2541− 2544.
DOI: 10.1021/acsnano.7b08010 ACS Nano XXXX, XXX, XXX−XXX