Journal of Hazardous Materials 298 (2015) 188–194
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
A novel nanoprobe for the sensitive detection of Francisella tularensis Ji-eun Kim a,1 , Youngmin Seo a,1 , Yoon Jeong a , Mintai P. Hwang b , Jangsun Hwang a , Jaebum Choo a,c , Jong Wook Hong a,c , Jun Ho Jeon d , Gi-eun Rhie d , Jonghoon Choi a,c,∗ a
Department of Bionano Technology, Graduate School, Hanyang University, Seoul 133-791, South Korea Center for Biomaterials, Korea Institute of Science and Technology, Seoul 136-791, South Korea Department of Bionano Engineering, Hanyang University ERICA, Ansan 426-791, South Korea d Division of High-risk Pathogen Research, Center for Infectious Disease, Korea National Institute of Health, Cheongju 363-951, South Korea b c
h i g h l i g h t s • • • •
We prepare apoferritin nanoprobes decorated with antibodies and nanoparticles. We examine nanoprobes for the sensitive detection of Francisella tularensis. 10-fold decrease of minimum concentration of pathogen was achieved. Simultaneous detection of multiple high-risk pathogens was obtained.
a r t i c l e
i n f o
Article history: Received 19 January 2015 Received in revised form 23 April 2015 Accepted 23 May 2015 Keywords: Nanoprobes Francisella tularensis Apoferritin Immunoassay Biosensing
a b s t r a c t Francisella tularensis is a human zoonotic pathogen and the causative agent of tularemia, a severe infectious disease. Given the extreme infectivity of F. tularensis and its potential to be used as a biological warfare agent, a fast and sensitive detection method is highly desirable. Herein, we construct a novel detection platform composed of two units: (1) Magnetic beads conjugated with multiple capturing antibodies against F. tularensis for its simple and rapid separation and (2) Genetically-engineered apoferritin protein constructs conjugated with multiple quantum dots and a detection antibody against F. tularensis for the amplification of signal. We demonstrate a 10-fold increase in the sensitivity relative to traditional lateral flow devices that utilize enzyme-based detection methods. We ultimately envision the use of our novel nanoprobe detection platform in future applications that require the highly-sensitive on-site detection of high-risk pathogens. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Francisella tularensis, a human zoonotic pathogen, is a comparatively small, non-motile, non-spore forming, gram-negative coccobacillus [1,2]. Residing in a wide variety of hosts including rodents, large carnivores, and blood-feeding insects, [3] F. tularensis is generally classified into two sub-types: a highly virulent type A (F. tularensis subsp. tularensis) predominant in North America, and a less virulent type B (F. tularensis subsp. holarctica) present in the northern hemisphere [4,5]. F. tularensis presents a particularly potent threat when transmitted to humans, which occurs through
∗ Corresponding author at: Department of Bionano Technology, Graduate School, Hanyang University, Seoul 133-791, South Korea. Tel.: +82 31 400 5203; fax: +82 31 436 8146. E-mail address: [email protected] (J. Choi). 1 These authors contributed equally. http://dx.doi.org/10.1016/j.jhazmat.2015.05.041 0304-3894/© 2015 Elsevier B.V. All rights reserved.
arthropod bites, the handling of infected animals, the ingestion of contaminated water or food, and the inhalation of infective aerosols [6]. In particular, it is the causative agent of tularemia, a severe infectious disease, and manifests itself with clinical symptoms including skin lesions, severe pain, and secondary infections such as pneumonia [4]. Its extreme infectivity and potential as a biological warfare agent is a cause for concern. Not surprisingly, the Centers for Disease Control and Prevention (CDC) in the USA has identified F. tularensis as a Category-A select agent (CDC Strategic Planning Workshop, 2000) due to its easy transmission, substantial morbidity and mortality in large numbers of people, and its ability to cause widespread panic [4,7]. For several years, a large emphasis has been placed on the sensitive and specific detection of F. tularensis via molecular and immunological techniques. For instance, polymerase chain reaction (PCR) has been used to amplify and thereby detect F. tularensis genes; the introduction of real-time PCR or multiplex PCR has further improved its detection capacity [8]. In another example,
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
enzyme-linked immunosorbent assays (ELISA) based on immune reactions to specific F. tularensis lipopolysaccharide (LPS) have been used to demonstrate limits of detection at approximately 103 –104 CFU/mL [9,10]. While both PCR and ELISA techniques provide high sensitivity, they unfortunately require complicated procedures and are also time consuming [4,11,12]. Immunochromatographic assays, another method that has been explored for the detection of F. tularensis [13], are appealing in that the amount of pathogens can be measured in a simple and rapid manner by applying several drops of sample solution to a test strip. These striptype assays, however, are limited in their relatively low levels of sensitivity; bacterial targets have a sensitivity anywhere from 105 to 106 CFU/mL [14], while that of toxins range from 1 ng/mL to 50 ng/mL. As a means to overcome such limited sensitivity, targets are often cultured for more than 24 h to expand their numbers, presenting an inherent shortcoming when considering the timesensitive nature of F. tularensis detection. Another limitation is in the mode of detection; results, visualized as colored lines upon conjugation with gold colloids, are measured with the naked eye, and may therefore generate biased and erroneous results [15]. Given these limitations, next generation assays for the on-site detection of high-risk pathogens should be sensitive, fast, and accurate. From this perspective, quantum dots (QDs) are attractive. Compared to widely-used fluorophores, which have poor photostability and low intensity, QDs exhibit a high quantum yield and possess size- and composition-dependent tunable emission spectra. Furthermore, given that pathogen detection would be carried out ex-vivo, concerns of in-vivo QD toxicity are minimal. Separately, magnetic particles are also attractive in their ease of use for the separation of targets from a solution, and in their efficiency as a labeling material for biosensing applications [16,17]. In this study, we link the advantages of QDs and magnetic particles with apoferritin, a self-assembled spherical protein nanoparticle with a large surface-to-volume ratio. Fairly homogenous in size (10–15 nm in diameter), apoferritin consists of 24 subunits [18,19], each of which is genetically-modified to express 6x-His tag and Protein G [20]. 6x-His tag is utilized to conjugate multiple Ni-NTA-functionalized QDs, thereby increasing the sensitivity of the system, while Protein G is used to bind the Fc region of a pathogen-specific antibody, thereby conferring specificity. By using such functionalized apoferritin constructs in tandem with magnetic particles that have been conjugated with antibodies specific for inactivated F. tularensis, we develop a novel nanoprobe system characterized by high sensitivity and ease of use. We believe that our nanoprobe system can serve as a quick and simple platform for the sensitive on-site detection of inactivated F. tularensis, and envision its use for other high-risk pathogens.
2. Experimental methods 2.1. Materials and reagents The gene for human heavy chain apoferritin was purchased from OriGene (FTH1:NM-002032, OriGene Technologies, Inc., Rockville, MD). The T7-based pET-28b(+) vector was purchased from Novagen (Merck KGaA, Darmstadt, Germany). The gene for protein G was provided by GeneScript (GenScript USA Inc., Piscataway, NJ) in a pUC57 plasmid-cloning vector. Myristoyl-2-hydroxy-sn-glycero-3-phosphocholine (MHPC; #855575), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N(methoxy polyethylene glycol)-2000 (DPPEPEG2000; #880120), and 1,2-dioleoyl-snglycero-3-N-(5-amino-1-carboxypentyl) iminodiacetic acid succinyl nickel salt (Ni-NTA; #790404) were all purchased from Avanti (Avanti Polar Lipids, Inc., Alabaster, AL). Quantum dots (QDs; QD Solution Nanodot HE-series 100–620 nm)
189
were purchased from QD solution (Ecoflux Inc., Cheongju-si, Korea). Monoclonal mouse anti-Francisella tularensis LPS (T14, FB11) antibody was purchased from HyTest (HyTest Ltd., Turku, Finland) while anti-Bacillus anthracis spore antigen antibody (SA26,SA27) and anti-Yersinia pestis antibody (YPF19, 6031) were purchased from Abcam (Abcam Inc. Cambridge, MA, USA). The information of antibodies used are described in Table 1. Dynabeads Protein G (1003D) and Dynabeads My One Carboxylic Acid (65011) were purchased from Life Technologies (Thermo Fisher Scientific Inc., Waltham, MA). NTA-Atto 488 and NTA-Atto 647N were purchased from Sigma-Aldrich (Sigma–Aldrich Co., Switzerland). 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and Nhydroxysuccinimide (NHS) were purchased from Sigma–Aldrich (Sigma–Aldrich Co., St. Louis, MO, USA). Inactivated Francisella tularensis, Bacillus anthracis, and Yersinia pestis were all prepared and provided by the Division of High-Risk Pathogen Research, Center for Disease Control & Prevention, National Institute of Health (Cheongju, Korea). 2.2. Genetic modification of apoferritin The gene for human heavy chain apoferritin was purchased from OriGene (FTH1:NM-002032) and inserted into the T7-based pET28b(+) vector flanked by restriction enzymes NdeI and BamHI. The gene for protein G was excised from the pUC57 plasmid cloning vector and inserted into the pET-28b(+) vector. The gene sequences for protein G and 6x-His tag were inserted near the C-terminus for extraluminal orientation. The vectors for the expression of apoferritin were induced by heat-shock treatment in DH5␣ (NEB #C2987H), and self-assembled in BL21 bacterial cells (Koram Deo Lab Seoul, Korea). Finally, the bacterial cells were lysed via tip sonication and apoferritin was purified using Ni-NTA His-Bind Resin (Qiagen #70666) and a Ni-NTA buffer kit (Novagen #70899-3). Apoferritin was sent to Cosmo Genetech (Seoul, Korea) for gene sequence analysis [20]. 2.3. Water-solubilization and Ni-NTA functionalization of quantum dots 1-Myristoyl-2-hydroxy-sn-glycero-3-phosphocholine (MHPC; Avanti Polar Lipids, Inc., #855575), 1,2-distearoyl-sn-glycero3-phosphoethanolamine-N-(methoxy polyethylene glycol)-2000 (DPPEPEG2000; Avanti Polar Lipids, Inc., #880120), and 1,2dioleoyl-snglycero-3-N-(5-amino-1-carboxypentyl) iminodiacetic acid succinyl nickel salt (Ni-NTA; Avanti Polar Lipids, Inc., #790404) were added in a 80%, 15%, 5% molar ratio to a chloroform solution of QDs. The resulting solution was added drop-wise to 2 mL of distilled water, briefly sonicated for 1 min, and continually heated for 1 h at 90 ◦ C to evaporate excess chloroform and facilitate the exchange of solvent into distilled water. The subsequent watersolubilized QD solution (surface modified with MHPC, PEG, and Ni-NTA) was sonicated for 30 min to ensure single particle suspension and was then spun down at 14,000 × g for 10 min. The supernatant was collected and filtered through a 0.2-m syringe filter for further removal of any aggregates. To measure the quantum yield of the water-solubilized and functionalized QDs, a baseline was set using a reference sample (DPBS), after which the ratio of photons emitted as photoluminescence to those absorbed by the QDs upon excitation at 365 nm was measured (ProtoFlex Quantum Efficiency Measurement System QE-1100) [20]. 2.4. Preparing apoferritin nanoprobes decorated with functionalized QDs or fluorescent dyes Genetically modified apoferritin constructs and functionalized QDs were mixed at the ratio of 1:6 (v/v) in DPBS and incubated
190
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
Table 1 The information of antibodies used in the study. Pathogen
Antibody
Type
Host
Company
Bacillus antracis
Anti-Bacillus anthracis spore antigen antibody [SA27] ab2281 Anti-Bacillus anthracis spore antigen antibody [SA26] ab1994 Francesella Tularensis LPS [T14] Francesella Tularensis LPS [FB11] Anti-Yersinia pestis antibody [YPF19] ab8275 Anti-Yersinia pestis antibody [6031] ab13639
Mono Mono Mono Mono Mono Mono
Mouse Mouse Mouse Mouse Mouse Mouse
Abcam Abcam HyTest HyTest Abcam Abcam
Francisella tularensis Yersina pestis
at room temperature for two hours to promote the conjugation of QDs to apoferritin. Excess QDs were removed from the solution via filtration. Specifically, centrifugal filters were first blocked with 5% BSA to minimize the adsorption of apoferritin-QD constructs, after which the filters were centrifuged at 14,000 g for 10 min to remove the extra BSA. Apoferritin nanoprobes were then added to the BSAblocked centrifugal filters and centrifuged once more at 1000 g for 5 min to ensure the removal of residual QDs. For dual color imaging, NTA Atto 488 or NTA Atto 647N were diluted to 10000:1 in DPBS and conjugated to the genetically modified apoferritin particles following the same protocol described above.
Table 2 The series of pathogen mixtures prepared for the blind test. Unidentified sample number
Blinded samples
Mixing ratio (v/v)
1 2 3 4 5 6 7 8
F. tularensis: B. anthracis F. tularensis: B. anthracis F. tularensis: Y. pestis F. tularensis: Y. pestis F. tularensis: Y. pestis: B. anthracis F. tularensis: Y. pestis: B. anthracis F. tularensis: Y. pestis: B. anthracis F. tularensis: Y. pestis: B. anthracis
7:3 3:7 7:3 3:7 1:1:1 1:0:0 0:0:1 0:1:0
2.5. Assembly and characterization of nanoprobes Purified intermediate apoferritin was subsequently analyzed with dynamic light scattering (Malvern Zetasizer, Malvern Instruments Ltd., Worcestershire, UK) to generate a size distribution profile and TEM image. Water-solubilized and Ni-NTA-functionalized QDs were then added to apoferritin at a 10-fold molar excess and incubated at room temperature for 2 h. Size distribution profiles were obtained for resulting nanoprobes, and TEM images were taken after sample preparation (JEM-3010, JEOL, Ltd., Tokyo, Japan). 2.6. Conjugation of antibodies to apoferritin nanoprobes or magnetic beads Anti-Bacillus anthrasis spore antigen antibody (SA27), Francesella Tularensis LPS (FB11) antibody, and anti-Yersinia pestis antibody (6031) were used as various detection antibodies and added to the apoferritin nanoprobes at the ratio of 10:1 (v/v) – the proper concentrations of antibodies were determined separately via ELISA (4 g/mL of antibody and 10 g/mL of apoferritin
nanoparticles). After reacting the detection antibodies and apoferritin for an hour at room temperature, excess antibodies were removed from the solution via filtration as described in the preparation of apoferritin nanoprobes decorated with functionalized QDs. After the addition of apoferritin nanoprobes to the BSA-blocked centrifugal filters, residual antibodies were removed by an additional centrifugation step at 1000 × g for 5 min. On the other hand, functionalized magnetic beads were prepared by conjugating capture antibodies that recognize different epitopes against the same pathogens to magnetic beads;anti-Bacillus anthrasis spore antigen antibody (SA26), Francesella Tularensis LPS (T14) antibody, and anti-Yersinia pestis antibody (YPF19) were used as capture antibodies. Dynabeads Protein G (1003D), with an average diameter of 2.8 m, were reacted with the capture antibodies at a ratio of 1:10 (v/v), 4 g/mL of antibody and 150 g/mL of particles for an hour at room temperature. Unreacted protein G sites were blocked with 5% BSA for an hour at room temperature, followed by a washing step with 1x TBST (0.1% tween 20) buffer.
Fig. 1. Scheme for the sensitive detection of inactivated F. tularensis. Sandwich immunoassay for the detection of F. tularensis with apoferritin nanoprobes and magnetic beads.
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
191
Fig. 2. Characterization of apoferritin nanoprobes. (a) TEM image of genetically-modified self-assembled apoferritin. (b) TEM images of QD-conjugated apoferritin. Inset is a high resolution EM image of the apoferritin-QD assembly. Scale bar is 10 nm. (c) Size distribution of apoferritin by DLS measurements, n = 3.
2.7. Pathogen capture via apoferritin nanoprobes and magnetic beads Inactivated pathogens were diluted in reaction buffer (1% BSA, 0.02% tween 20) at different concentrations and reacted with functionalized magnetic beads for an hour at room temperature. Pathogens bound to the functionalized magnetic beads were then
isolated with the use of a magnet and washed with 1x TBST (0.1% tween 20). Magnetic beads binding the pathogens were subsequently mixed with functionalized apoferritin nanoprobes for an hour at room temperature, ultimately resulting in the sandwich-targeting of pathogen. To remove unreacted functionalized apoferritin nanoprobes, the particle mixture was washed once with 1 mL of 1x TBST (0.1% Tween) and then in triplicate
Fig. 3. Visualization of inactivated F. tularensis capture by apoferritin nanoprobes and magnetic beads. (a) TEM images of immunocomplexes composed of antibody-conjugated magnetic beads and apoferritin nanoprobes. (b) Red fluorescence corresponds to the QDs from apoferritin-pathogen-magnetic bead immuno constructs, thereby indicating the presence of F. tularensis.
192
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
with 200 L of 1x TBST. After washing, the samples were excited at 360 nm and measured for their fluorescence at 620 nm. 2.8. Specificity validation of apoferritin nanoprobes A blind test to validate the specificity of the apoferritin nanoprobes was performed by preparing samples mixed with inactivated F. tularensis and two other inactivated pathogens (B. anthracis and Y. pestis). The series of blinded samples are described in Table 2. Mixed pathogen samples were tested via a sandwich immunoassay consisting of antibody-functionalized magnetic beads and apoferritin nanoprobes specific for F. tularensis. 2.9. Capturing of dual pathogens in a sample mixture Antibodies against Francisella tularensis LPS and Bacillus anthracis spore antigen were conjugated to carboxylic acid magnetic beads via EDC/NHS chemistry. After incubating antibodies and magnetic beads in a 1:10 ratio (v/v) at room temperature for an hour, 5% BSA/PBS solution was added to block unconjugated carboxyl groups on the magnetic beads. Unreacted constituents in solution were removed by washing with 1xTBST (0.1% Tween) buffer. Recombinant inactivated Francisella tularensis and Bacillus anthracis were diluted in PBS at different concentrations, mixed together, and incubated with the functionalized magnetic beads at room temperature for an hour. The magnetic beads – bound with antigens – were separated from solution using a magnet, after which fluorescent probes were added to sandwich-target the antigens. Unreacted fluorescent probes were removed from the samples by serially washing the immunoassayed samples with 1xTBST (0.1% Tween), once with 1 mL and 3 times with 200 L. The average fluorescent intensity of samples was measured in 96 well plates and images of the immunocomplexes were obtained using a fluorescent microscope. 3. Results and discussion Apoferritin, a globular spherical protein naturally found in physiological conditions, is formed via the self-assembly of 24 heavy and/or light subunits. In this study, we use genetically-modified heavy chain apoferritin (H subunit) in tandem with functionalized magnetic beads for the sensitive detection of inactivated F. tularensis (Fig. 1). Specifically, the H subunit of apoferritin is modified to express both protein G and 6x-His tag. While protein G, a cell-surface protein on streptococcal bacteria, confers directionality to a F. tularensis antibody by specifically binding to the Fc region, the 6x-His tag allows for the conjugation of Ni-NTA-functionalized QDs. Modified apoferritin has a spherical morphology (Fig. 2a) as identified via transmission electron microscopy (TEM), and has an average outer diameter of approximately 15 nm as calculated from the analysis of TEM images and dynamic light scattering (DLS) measurements (Fig. 2c). Several QDs are observed assembled around the modified apoferritin (Fig. 2b); approximately 6 QDs attach per apoferritin unit, thereby enabling a 4.5-fold increase in fluorescence intensity compared to a single QD [20]. Concurrently, the magnetic beads are functionalized with another set of antibodies against inactivated F. tularensis. Together, the magnetic beads and apoferritin nanoprobes form an immunosandwich in the presence of inactivated F. tularensis, and produce a quantifiable fluorescent readout. A standard curve of known concentrations of inactivated pathogen was constructed by measuring the fluorescence intensities from the apoferritin-pathogen-magnetic bead sandwich constructs. The limit of detection of the apoferritin nanoprobe magnetic bead system is being compared to the experiment using a dye-conjugated antibody. TEM images show the apoferritin-pathogen-magnetic
Fig. 4. Detection of inactivated F. tularensis via apoferritin nanoprobes and magnetic beads. (a) The limit of detection for measuring inactivated F. tularensis with apoferritin nanoprobes and magnetic beads. (b) The specificity of the system is assessed via blinded samples containing different ratios of F. tularensis, B. anthracis, and Y. pestis (F: Francisella tularensis, B: Bacillus anthracis, Y: Yersinia pestis)
bead immunocomplexes (Fig. 3a). Pathogen detection was further corroborated through fluorescence images of the immunocomplexes, in which a strong red fluorescence is observed in the presence of pathogen compared to the negative control (Fig. 3b). It is worth noting that traditional immunochromatographic assays for F. tularensis have a detection limit of approximately 105 ∼ 107 CFU/mL. In our apoferritin nanoprobe system, however, the use of specific antibodies and multiple QDs to one apoferritin unit increased the detection sensitivity to approximately 104 CFU/mL of F. tularensis (Fig. 4a). Furthermore, the specificity of our system was assessed via blind tests of samples including inactivated F. tularensis and two other pathogens – Y. pestis and B. anthracis (Fig. 4b) (concentration of 104 CFU/mL each). Given that sample number 6 – F. tularensis alone – showed a comparative level of fluorescence intensity to that of the original assay, the near absence of fluorescence for sample numbers 7 and 8 – Bacillus anthracis and Yersinia pestis, respectively – confirm the specificity of our system. Other combinations, in which F. tularensis is mixed with other bacteria, showed a weaker, but still detectable, level of fluorescence intensity. Finally, while F. tularensis alone is a serious threat to public health, its combination with other high-risk pathogens presents an even graver risk. The ability to detect multiple pathogens simultaneously is therefore highly desirable. From this perspective, we assess the capacity of our system for the simultaneous detection of two pathogens: F. tularensis and B. anthracis (Fig. 5). Apoferritin nanoprobes are prepared in the same manner as before, but with different antibodies and with the use of fluorescent molecules
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
193
Fig. 5. Scheme of dual capturing of pathogens by apoferritin nanoprobes.
rather than QDs. Upon mixing F. tularensis and B. anthracis in equal amounts, ranging from 102 CFU/mL to 106 CFU/mL individually, apoferritin nanoprobes with antibodies against F. tularensis (green fluorescence) or B. anthracis (red fluorescence) are used to measure their respective limits of detection. The limit of detection for F. tularensis was approximately 104 CFU/mL, similar to that of single antigen spiked samples, while that of B. anthracis was 104 CFU/mL as well (Fig. 6a and Fig. 4a). Increasing amounts of
pathogen resulted in an increase in apoferritin-pathogen-magnetic bead immuno complex formation, as observed via bright field and fluorescence images (Fig. 6b). Additionally, we observed similar fluorescent intensity at each pathogen concentration, indicative of the detection of similar numbers of pathogens. To further confirm that our system can be used to accurately detect the amount of each pathogen from a non-equal mixture of two pathogens, 10 times more F. tularensis than B. anthracis were mixed together.
Fig. 6. Dual capturing of pathogens by apoferritin nanoprobes. (a) Individual limits of detection from a solution of equal amounts of inactivated F. tularensis and B. anthracis. (b) Fluorescent images of captured F. tularensis (green color) and B. anthracis (red color), each labeled with two different colored sets of apoferritinn nanoprobes. (c) Quantified fluorescent intensities of captured inactivated F. tularensis and B. anthracis when mixed together in a 10:1 ratio (F. tularensis: B. anthracis). (d) Fluorescent images of captured inactivated F. tularensis and B. anthracis from a solution of 10:1 F. tularensis to B. anthracis. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
194
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
The fluorescence intensities of 106 CFU/mL of F. tularensis and 105 CFU/mL of B. anthracis (Fig. 6c–d) yielded comparable values to those obtained from each concentration in equally-mixed samples (Fig. 6a–b). Overall, we demonstrate the capacity of our apoferritin probe system for the simultaneous detection of multiple pathogens without interference between the different targets. It is worth noting that our apoferritin nanoprobes utilize antibodies specific to live pathogens that are not fully applicable to their inactivated counterparts. Furthermore, we make use of commercial antibodies, which present limitations with regards to their binding efficiency. In other words, the lowest concentration detectable with our system is dependent on the performance of the commercial antibody. Recent reports employing target-specific peptides as an alternative to antibodies demonstrate an improved level of antigen binding by ferritin probes [18,21,22]. Similar approaches for the development of F. tularensis-specific peptides should increase the breath and sensitivity of our apoferritin nanoprobes. 4. Conclusions The reported apoferritin nanoprobes in this study have been shown to increase fluorescence intensity and were used to improve the limit of detection for high-risk pathogens. We demonstrate a specificity and an increased sensitivity towards F. tularensis via apoferritin nanoprobes carrying an antibody for inactivated F. tularensis. Specifically, approximately 10 times lower levels of F. tularensis were detected using our system than were via traditional antibody-fluorescent dye combinations. Furthermore, we utilize our system to simultaneously detect multiple pathogens, and in effect demonstrate the ease with which different apoferritin nanoprobes specific for various pathogens can be constructed. It should be noted that the stability of nanoprobes packaged in an assay should be further evaluated before any implementation towards on-site detection systems. Promisingly, in an ongoing study, functionalized apoferritin nanoprobes stored for three weeks show a comparable level of conjugation efficiency and limit of detection to freshly-prepared samples. Increasing the stability of detection probes without decreasing the sensitivity is another important aspect that we are actively investigating for the practical application of our system for on-site detection capabilities. Ultimately, we envision that the high sensitivity of our system, coupled with a simple yet rapid measurement capability, present promising opportunities in applications that require the on-site detection of high-risk pathogens. Acknowledgments This work was supported by the Basic Science Research Program through the National Research Foundation (NRF) funded by the Ministry of Education (grant number: 2013R1A1A1012653), by the Korea Centers for Disease Control and Prevention (grant number: 2014-E45002-00) and the grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute(KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI14C3266).
References [1] T.F.J. Pikula, M. Beklova, Z. Holesovska, J. Pikulova, Geographic information systems in epidemilogy-Ecology of common vole and distribution of natural foci of Tularaemia, Acta Vet. Brno 71 (2002) 379–387. [2] J. Pikula, F. Treml, M. Beklova, Z. Holesovska, J. Pikulova, Ecological conditions of natural foci of tularaemia in the Czech Republic, Eur. J. Epidemiol. 18 (2003) 1091–1095. [3] J.M. Petersen, M.E. Schriefer, Tularemia: emergence/re-emergence, Vet. Res. 36 (2005) 455–467. [4] D.T. Dennis, T.V. Inglesby, D.A. Henderson, J.G. Bartlett, M.S. Ascher, E. Eitzen, A.D. Fine, A.M. Friedlander, J. Hauer, M. Layton, S.R. Lillibridge, J.E. McDade, M.T. Osterholm, T. O’Toole, G. Parker, T.M. Perl, P.K. Russell, K. Tonat, B. Working Group on Civilian, Tularemia as a biological weapon: medical and public health management, JAMA: J. Am. Med. Assoc. 285 (2001) 2763–2773. [5] P.C. Oyston, A. Sjostedt, R.W. Titball, Tularaemia: bioterrorism defence renews interest in Francisella tularensis, Nat. Rev. Microbiol. 2 (2004) 967–978. [6] J. Ellis, P.C. Oyston, M. Green, R.W. Titball, Tularemia, Clin. Microbiol. Rev. 15 (2002) 631–646. [7] W.D. Sawyer, H.G. Dangerfield, A.L. Hogge, D. Crozier, Antibiotic prophylaxis and therapy of airborne tularemia, Bacteriol. Rev. 30 (1966) 542–550. [8] E.M. Elnifro, A.M. Ashshi, R.J. Cooper, P.E. Klapper, Multiplex PCR: optimization and application in diagnostic virology, Clin. Microbiol. Rev. 13 (2000) 559–570. [9] E. Vinogradov, M.B. Perry, J.W. Conlan, Structural analysis of Francisella tularensis lipopolysaccharide, Eur. J. Biochem./FEBS 269 (2002) 6112–6118. [10] P. Schmitt, W. Splettstosser, M. Porsch-Ozcurumez, E.J. Finke, R. Grunow, A novel screening ELISA and a confirmatory Western blot useful for diagnosis and epidemiological studies of tularemia, Epidemiol. Infect. 133 (2005) 759–766. [11] D.R. Christensen, L.J. Hartman, B.M. Loveless, M.S. Frye, M.A. Shipley, D.L. Bridge, M.J. Richards, R.S. Kaplan, J. Garrison, C.D. Baldwin, D.A. Kulesh, D.A. Norwood, Detection of biological threat agents by real-time PCR: comparison of assay performance on the R.A.P.I.D., the LightCycler, and the Smart Cycler platforms, Clin. Chem. 52 (2006) 141–145. [12] K. Tomioka, M. Peredelchuk, X. Zhu, R. Arena, D. Volokhov, A. Selvapandiyan, K. Stabler, J. Mellquist-Riemenschneider, V. Chizhikov, G. Kaplan, H. Nakhasi, R. Duncan, A multiplex polymerase chain reaction microarray assay to detect bioterror pathogens in blood, J. Mol. Diagn.: JMD 7 (2005) 486–494. [13] M.J. Hepburn, A.J. Simpson, Tularemia: current diagnosis and treatment options, Expert Rev. Anti-Infect. Ther. 6 (2008) 231–240. [14] R. Grunow, W. Splettstoesser, S. McDonald, C. Otterbein, T. O’Brien, C. Morgan, J. Aldrich, E. Hofer, E.J. Finke, H. Meyer, Detection of Francisella tularensis in biological specimens using a capture enzyme-linked immunosorbent assay, an immunochromatographic handheld assay, and a PCR, Clin. Diagn. Lab. Immunol. 7 (2000) 86–90. [15] A.H. Peruski, L.F. Peruski Jr., Immunological methods for detection and identification of infectious disease and biological warfare agents, Clin. Diagn. Lab. Immunol. 10 (2003) 506–513. [16] W. Shen, B.D. Schrag, M.J. Carter, G. Xiao, Quantitative detection of DNA labeled with magnetic nanoparticles using arrays of MgO-based magnetic tunnel junction sensors, Appl. Phys. Lett. 93 (2008) 033903. [17] T.J. Lowery, R. Palazzolo, S.M. Wong, P.J. Prado, S. Taktak, Single-coil multisample proton relaxation method for magnetic relaxation switch assays, Anal. Chem. 80 (2008) 1118–1123. [18] S.H. Lee, H. Lee, J.S. Park, H. Choi, K.Y. Han, H.S. Seo, K.Y. Ahn, S.S. Han, Y. Cho, K.H. Lee, J. Lee, A novel approach to ultrasensitive diagnosis using supramolecular protein nanoparticles, FASEB J.: Off. Publ. Fed. Am. Soc. Exp. Biol. 21 (2007) 1324–1334. [19] J.Y. Ahn, H. Choi, Y.H. Kim, K.Y. Han, J.S. Park, S.S. Han, J. Lee, Heterologous gene expression using self-assembled supra-molecules with high affinity for HSP70 chaperone, Nucleic Acids Res. 33 (2005) 3751–3762. [20] M.P. Hwang, J.W. Lee, K.E. Lee, K.H. Lee, Think modular: a simple apoferritin-based platform for the multifaceted detection of pancreatic cancer, ACS Nano 7 (2013) 8167–8174. [21] E.J. Lee, K.Y. Ahn, J.H. Lee, J.S. Park, J.A. Song, S.J. Sim, E.B. Lee, Y.J. Cha, J. Lee, A novel bioassay platform using ferritin-based nanoprobe hydrogel, Adv. Mater. (2012) 4739–4744, 4730. [22] K.C. Kwon, J.H. Ryu, J.H. Lee, E.J. Lee, I.C. Kwon, K. Kim, J. Lee, Proteinticle/gold core/shell nanoparticles for targeted cancer therapy without nanotoxicity, Adv. Mater. 26 (2014) 6436–6441.
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
A novel nanoprobe for the sensitive detection of Francisella tularensis Ji-eun Kim a,1 , Youngmin Seo a,1 , Yoon Jeong a , Mintai P. Hwang b , Jangsun Hwang a , Jaebum Choo a,c , Jong Wook Hong a,c , Jun Ho Jeon d , Gi-eun Rhie d , Jonghoon Choi a,c,∗ a
Department of Bionano Technology, Graduate School, Hanyang University, Seoul 133-791, South Korea Center for Biomaterials, Korea Institute of Science and Technology, Seoul 136-791, South Korea Department of Bionano Engineering, Hanyang University ERICA, Ansan 426-791, South Korea d Division of High-risk Pathogen Research, Center for Infectious Disease, Korea National Institute of Health, Cheongju 363-951, South Korea b c
h i g h l i g h t s • • • •
We prepare apoferritin nanoprobes decorated with antibodies and nanoparticles. We examine nanoprobes for the sensitive detection of Francisella tularensis. 10-fold decrease of minimum concentration of pathogen was achieved. Simultaneous detection of multiple high-risk pathogens was obtained.
a r t i c l e
i n f o
Article history: Received 19 January 2015 Received in revised form 23 April 2015 Accepted 23 May 2015 Keywords: Nanoprobes Francisella tularensis Apoferritin Immunoassay Biosensing
a b s t r a c t Francisella tularensis is a human zoonotic pathogen and the causative agent of tularemia, a severe infectious disease. Given the extreme infectivity of F. tularensis and its potential to be used as a biological warfare agent, a fast and sensitive detection method is highly desirable. Herein, we construct a novel detection platform composed of two units: (1) Magnetic beads conjugated with multiple capturing antibodies against F. tularensis for its simple and rapid separation and (2) Genetically-engineered apoferritin protein constructs conjugated with multiple quantum dots and a detection antibody against F. tularensis for the amplification of signal. We demonstrate a 10-fold increase in the sensitivity relative to traditional lateral flow devices that utilize enzyme-based detection methods. We ultimately envision the use of our novel nanoprobe detection platform in future applications that require the highly-sensitive on-site detection of high-risk pathogens. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Francisella tularensis, a human zoonotic pathogen, is a comparatively small, non-motile, non-spore forming, gram-negative coccobacillus [1,2]. Residing in a wide variety of hosts including rodents, large carnivores, and blood-feeding insects, [3] F. tularensis is generally classified into two sub-types: a highly virulent type A (F. tularensis subsp. tularensis) predominant in North America, and a less virulent type B (F. tularensis subsp. holarctica) present in the northern hemisphere [4,5]. F. tularensis presents a particularly potent threat when transmitted to humans, which occurs through
∗ Corresponding author at: Department of Bionano Technology, Graduate School, Hanyang University, Seoul 133-791, South Korea. Tel.: +82 31 400 5203; fax: +82 31 436 8146. E-mail address: [email protected] (J. Choi). 1 These authors contributed equally. http://dx.doi.org/10.1016/j.jhazmat.2015.05.041 0304-3894/© 2015 Elsevier B.V. All rights reserved.
arthropod bites, the handling of infected animals, the ingestion of contaminated water or food, and the inhalation of infective aerosols [6]. In particular, it is the causative agent of tularemia, a severe infectious disease, and manifests itself with clinical symptoms including skin lesions, severe pain, and secondary infections such as pneumonia [4]. Its extreme infectivity and potential as a biological warfare agent is a cause for concern. Not surprisingly, the Centers for Disease Control and Prevention (CDC) in the USA has identified F. tularensis as a Category-A select agent (CDC Strategic Planning Workshop, 2000) due to its easy transmission, substantial morbidity and mortality in large numbers of people, and its ability to cause widespread panic [4,7]. For several years, a large emphasis has been placed on the sensitive and specific detection of F. tularensis via molecular and immunological techniques. For instance, polymerase chain reaction (PCR) has been used to amplify and thereby detect F. tularensis genes; the introduction of real-time PCR or multiplex PCR has further improved its detection capacity [8]. In another example,
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
enzyme-linked immunosorbent assays (ELISA) based on immune reactions to specific F. tularensis lipopolysaccharide (LPS) have been used to demonstrate limits of detection at approximately 103 –104 CFU/mL [9,10]. While both PCR and ELISA techniques provide high sensitivity, they unfortunately require complicated procedures and are also time consuming [4,11,12]. Immunochromatographic assays, another method that has been explored for the detection of F. tularensis [13], are appealing in that the amount of pathogens can be measured in a simple and rapid manner by applying several drops of sample solution to a test strip. These striptype assays, however, are limited in their relatively low levels of sensitivity; bacterial targets have a sensitivity anywhere from 105 to 106 CFU/mL [14], while that of toxins range from 1 ng/mL to 50 ng/mL. As a means to overcome such limited sensitivity, targets are often cultured for more than 24 h to expand their numbers, presenting an inherent shortcoming when considering the timesensitive nature of F. tularensis detection. Another limitation is in the mode of detection; results, visualized as colored lines upon conjugation with gold colloids, are measured with the naked eye, and may therefore generate biased and erroneous results [15]. Given these limitations, next generation assays for the on-site detection of high-risk pathogens should be sensitive, fast, and accurate. From this perspective, quantum dots (QDs) are attractive. Compared to widely-used fluorophores, which have poor photostability and low intensity, QDs exhibit a high quantum yield and possess size- and composition-dependent tunable emission spectra. Furthermore, given that pathogen detection would be carried out ex-vivo, concerns of in-vivo QD toxicity are minimal. Separately, magnetic particles are also attractive in their ease of use for the separation of targets from a solution, and in their efficiency as a labeling material for biosensing applications [16,17]. In this study, we link the advantages of QDs and magnetic particles with apoferritin, a self-assembled spherical protein nanoparticle with a large surface-to-volume ratio. Fairly homogenous in size (10–15 nm in diameter), apoferritin consists of 24 subunits [18,19], each of which is genetically-modified to express 6x-His tag and Protein G [20]. 6x-His tag is utilized to conjugate multiple Ni-NTA-functionalized QDs, thereby increasing the sensitivity of the system, while Protein G is used to bind the Fc region of a pathogen-specific antibody, thereby conferring specificity. By using such functionalized apoferritin constructs in tandem with magnetic particles that have been conjugated with antibodies specific for inactivated F. tularensis, we develop a novel nanoprobe system characterized by high sensitivity and ease of use. We believe that our nanoprobe system can serve as a quick and simple platform for the sensitive on-site detection of inactivated F. tularensis, and envision its use for other high-risk pathogens.
2. Experimental methods 2.1. Materials and reagents The gene for human heavy chain apoferritin was purchased from OriGene (FTH1:NM-002032, OriGene Technologies, Inc., Rockville, MD). The T7-based pET-28b(+) vector was purchased from Novagen (Merck KGaA, Darmstadt, Germany). The gene for protein G was provided by GeneScript (GenScript USA Inc., Piscataway, NJ) in a pUC57 plasmid-cloning vector. Myristoyl-2-hydroxy-sn-glycero-3-phosphocholine (MHPC; #855575), 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N(methoxy polyethylene glycol)-2000 (DPPEPEG2000; #880120), and 1,2-dioleoyl-snglycero-3-N-(5-amino-1-carboxypentyl) iminodiacetic acid succinyl nickel salt (Ni-NTA; #790404) were all purchased from Avanti (Avanti Polar Lipids, Inc., Alabaster, AL). Quantum dots (QDs; QD Solution Nanodot HE-series 100–620 nm)
189
were purchased from QD solution (Ecoflux Inc., Cheongju-si, Korea). Monoclonal mouse anti-Francisella tularensis LPS (T14, FB11) antibody was purchased from HyTest (HyTest Ltd., Turku, Finland) while anti-Bacillus anthracis spore antigen antibody (SA26,SA27) and anti-Yersinia pestis antibody (YPF19, 6031) were purchased from Abcam (Abcam Inc. Cambridge, MA, USA). The information of antibodies used are described in Table 1. Dynabeads Protein G (1003D) and Dynabeads My One Carboxylic Acid (65011) were purchased from Life Technologies (Thermo Fisher Scientific Inc., Waltham, MA). NTA-Atto 488 and NTA-Atto 647N were purchased from Sigma-Aldrich (Sigma–Aldrich Co., Switzerland). 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and Nhydroxysuccinimide (NHS) were purchased from Sigma–Aldrich (Sigma–Aldrich Co., St. Louis, MO, USA). Inactivated Francisella tularensis, Bacillus anthracis, and Yersinia pestis were all prepared and provided by the Division of High-Risk Pathogen Research, Center for Disease Control & Prevention, National Institute of Health (Cheongju, Korea). 2.2. Genetic modification of apoferritin The gene for human heavy chain apoferritin was purchased from OriGene (FTH1:NM-002032) and inserted into the T7-based pET28b(+) vector flanked by restriction enzymes NdeI and BamHI. The gene for protein G was excised from the pUC57 plasmid cloning vector and inserted into the pET-28b(+) vector. The gene sequences for protein G and 6x-His tag were inserted near the C-terminus for extraluminal orientation. The vectors for the expression of apoferritin were induced by heat-shock treatment in DH5␣ (NEB #C2987H), and self-assembled in BL21 bacterial cells (Koram Deo Lab Seoul, Korea). Finally, the bacterial cells were lysed via tip sonication and apoferritin was purified using Ni-NTA His-Bind Resin (Qiagen #70666) and a Ni-NTA buffer kit (Novagen #70899-3). Apoferritin was sent to Cosmo Genetech (Seoul, Korea) for gene sequence analysis [20]. 2.3. Water-solubilization and Ni-NTA functionalization of quantum dots 1-Myristoyl-2-hydroxy-sn-glycero-3-phosphocholine (MHPC; Avanti Polar Lipids, Inc., #855575), 1,2-distearoyl-sn-glycero3-phosphoethanolamine-N-(methoxy polyethylene glycol)-2000 (DPPEPEG2000; Avanti Polar Lipids, Inc., #880120), and 1,2dioleoyl-snglycero-3-N-(5-amino-1-carboxypentyl) iminodiacetic acid succinyl nickel salt (Ni-NTA; Avanti Polar Lipids, Inc., #790404) were added in a 80%, 15%, 5% molar ratio to a chloroform solution of QDs. The resulting solution was added drop-wise to 2 mL of distilled water, briefly sonicated for 1 min, and continually heated for 1 h at 90 ◦ C to evaporate excess chloroform and facilitate the exchange of solvent into distilled water. The subsequent watersolubilized QD solution (surface modified with MHPC, PEG, and Ni-NTA) was sonicated for 30 min to ensure single particle suspension and was then spun down at 14,000 × g for 10 min. The supernatant was collected and filtered through a 0.2-m syringe filter for further removal of any aggregates. To measure the quantum yield of the water-solubilized and functionalized QDs, a baseline was set using a reference sample (DPBS), after which the ratio of photons emitted as photoluminescence to those absorbed by the QDs upon excitation at 365 nm was measured (ProtoFlex Quantum Efficiency Measurement System QE-1100) [20]. 2.4. Preparing apoferritin nanoprobes decorated with functionalized QDs or fluorescent dyes Genetically modified apoferritin constructs and functionalized QDs were mixed at the ratio of 1:6 (v/v) in DPBS and incubated
190
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
Table 1 The information of antibodies used in the study. Pathogen
Antibody
Type
Host
Company
Bacillus antracis
Anti-Bacillus anthracis spore antigen antibody [SA27] ab2281 Anti-Bacillus anthracis spore antigen antibody [SA26] ab1994 Francesella Tularensis LPS [T14] Francesella Tularensis LPS [FB11] Anti-Yersinia pestis antibody [YPF19] ab8275 Anti-Yersinia pestis antibody [6031] ab13639
Mono Mono Mono Mono Mono Mono
Mouse Mouse Mouse Mouse Mouse Mouse
Abcam Abcam HyTest HyTest Abcam Abcam
Francisella tularensis Yersina pestis
at room temperature for two hours to promote the conjugation of QDs to apoferritin. Excess QDs were removed from the solution via filtration. Specifically, centrifugal filters were first blocked with 5% BSA to minimize the adsorption of apoferritin-QD constructs, after which the filters were centrifuged at 14,000 g for 10 min to remove the extra BSA. Apoferritin nanoprobes were then added to the BSAblocked centrifugal filters and centrifuged once more at 1000 g for 5 min to ensure the removal of residual QDs. For dual color imaging, NTA Atto 488 or NTA Atto 647N were diluted to 10000:1 in DPBS and conjugated to the genetically modified apoferritin particles following the same protocol described above.
Table 2 The series of pathogen mixtures prepared for the blind test. Unidentified sample number
Blinded samples
Mixing ratio (v/v)
1 2 3 4 5 6 7 8
F. tularensis: B. anthracis F. tularensis: B. anthracis F. tularensis: Y. pestis F. tularensis: Y. pestis F. tularensis: Y. pestis: B. anthracis F. tularensis: Y. pestis: B. anthracis F. tularensis: Y. pestis: B. anthracis F. tularensis: Y. pestis: B. anthracis
7:3 3:7 7:3 3:7 1:1:1 1:0:0 0:0:1 0:1:0
2.5. Assembly and characterization of nanoprobes Purified intermediate apoferritin was subsequently analyzed with dynamic light scattering (Malvern Zetasizer, Malvern Instruments Ltd., Worcestershire, UK) to generate a size distribution profile and TEM image. Water-solubilized and Ni-NTA-functionalized QDs were then added to apoferritin at a 10-fold molar excess and incubated at room temperature for 2 h. Size distribution profiles were obtained for resulting nanoprobes, and TEM images were taken after sample preparation (JEM-3010, JEOL, Ltd., Tokyo, Japan). 2.6. Conjugation of antibodies to apoferritin nanoprobes or magnetic beads Anti-Bacillus anthrasis spore antigen antibody (SA27), Francesella Tularensis LPS (FB11) antibody, and anti-Yersinia pestis antibody (6031) were used as various detection antibodies and added to the apoferritin nanoprobes at the ratio of 10:1 (v/v) – the proper concentrations of antibodies were determined separately via ELISA (4 g/mL of antibody and 10 g/mL of apoferritin
nanoparticles). After reacting the detection antibodies and apoferritin for an hour at room temperature, excess antibodies were removed from the solution via filtration as described in the preparation of apoferritin nanoprobes decorated with functionalized QDs. After the addition of apoferritin nanoprobes to the BSA-blocked centrifugal filters, residual antibodies were removed by an additional centrifugation step at 1000 × g for 5 min. On the other hand, functionalized magnetic beads were prepared by conjugating capture antibodies that recognize different epitopes against the same pathogens to magnetic beads;anti-Bacillus anthrasis spore antigen antibody (SA26), Francesella Tularensis LPS (T14) antibody, and anti-Yersinia pestis antibody (YPF19) were used as capture antibodies. Dynabeads Protein G (1003D), with an average diameter of 2.8 m, were reacted with the capture antibodies at a ratio of 1:10 (v/v), 4 g/mL of antibody and 150 g/mL of particles for an hour at room temperature. Unreacted protein G sites were blocked with 5% BSA for an hour at room temperature, followed by a washing step with 1x TBST (0.1% tween 20) buffer.
Fig. 1. Scheme for the sensitive detection of inactivated F. tularensis. Sandwich immunoassay for the detection of F. tularensis with apoferritin nanoprobes and magnetic beads.
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
191
Fig. 2. Characterization of apoferritin nanoprobes. (a) TEM image of genetically-modified self-assembled apoferritin. (b) TEM images of QD-conjugated apoferritin. Inset is a high resolution EM image of the apoferritin-QD assembly. Scale bar is 10 nm. (c) Size distribution of apoferritin by DLS measurements, n = 3.
2.7. Pathogen capture via apoferritin nanoprobes and magnetic beads Inactivated pathogens were diluted in reaction buffer (1% BSA, 0.02% tween 20) at different concentrations and reacted with functionalized magnetic beads for an hour at room temperature. Pathogens bound to the functionalized magnetic beads were then
isolated with the use of a magnet and washed with 1x TBST (0.1% tween 20). Magnetic beads binding the pathogens were subsequently mixed with functionalized apoferritin nanoprobes for an hour at room temperature, ultimately resulting in the sandwich-targeting of pathogen. To remove unreacted functionalized apoferritin nanoprobes, the particle mixture was washed once with 1 mL of 1x TBST (0.1% Tween) and then in triplicate
Fig. 3. Visualization of inactivated F. tularensis capture by apoferritin nanoprobes and magnetic beads. (a) TEM images of immunocomplexes composed of antibody-conjugated magnetic beads and apoferritin nanoprobes. (b) Red fluorescence corresponds to the QDs from apoferritin-pathogen-magnetic bead immuno constructs, thereby indicating the presence of F. tularensis.
192
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
with 200 L of 1x TBST. After washing, the samples were excited at 360 nm and measured for their fluorescence at 620 nm. 2.8. Specificity validation of apoferritin nanoprobes A blind test to validate the specificity of the apoferritin nanoprobes was performed by preparing samples mixed with inactivated F. tularensis and two other inactivated pathogens (B. anthracis and Y. pestis). The series of blinded samples are described in Table 2. Mixed pathogen samples were tested via a sandwich immunoassay consisting of antibody-functionalized magnetic beads and apoferritin nanoprobes specific for F. tularensis. 2.9. Capturing of dual pathogens in a sample mixture Antibodies against Francisella tularensis LPS and Bacillus anthracis spore antigen were conjugated to carboxylic acid magnetic beads via EDC/NHS chemistry. After incubating antibodies and magnetic beads in a 1:10 ratio (v/v) at room temperature for an hour, 5% BSA/PBS solution was added to block unconjugated carboxyl groups on the magnetic beads. Unreacted constituents in solution were removed by washing with 1xTBST (0.1% Tween) buffer. Recombinant inactivated Francisella tularensis and Bacillus anthracis were diluted in PBS at different concentrations, mixed together, and incubated with the functionalized magnetic beads at room temperature for an hour. The magnetic beads – bound with antigens – were separated from solution using a magnet, after which fluorescent probes were added to sandwich-target the antigens. Unreacted fluorescent probes were removed from the samples by serially washing the immunoassayed samples with 1xTBST (0.1% Tween), once with 1 mL and 3 times with 200 L. The average fluorescent intensity of samples was measured in 96 well plates and images of the immunocomplexes were obtained using a fluorescent microscope. 3. Results and discussion Apoferritin, a globular spherical protein naturally found in physiological conditions, is formed via the self-assembly of 24 heavy and/or light subunits. In this study, we use genetically-modified heavy chain apoferritin (H subunit) in tandem with functionalized magnetic beads for the sensitive detection of inactivated F. tularensis (Fig. 1). Specifically, the H subunit of apoferritin is modified to express both protein G and 6x-His tag. While protein G, a cell-surface protein on streptococcal bacteria, confers directionality to a F. tularensis antibody by specifically binding to the Fc region, the 6x-His tag allows for the conjugation of Ni-NTA-functionalized QDs. Modified apoferritin has a spherical morphology (Fig. 2a) as identified via transmission electron microscopy (TEM), and has an average outer diameter of approximately 15 nm as calculated from the analysis of TEM images and dynamic light scattering (DLS) measurements (Fig. 2c). Several QDs are observed assembled around the modified apoferritin (Fig. 2b); approximately 6 QDs attach per apoferritin unit, thereby enabling a 4.5-fold increase in fluorescence intensity compared to a single QD [20]. Concurrently, the magnetic beads are functionalized with another set of antibodies against inactivated F. tularensis. Together, the magnetic beads and apoferritin nanoprobes form an immunosandwich in the presence of inactivated F. tularensis, and produce a quantifiable fluorescent readout. A standard curve of known concentrations of inactivated pathogen was constructed by measuring the fluorescence intensities from the apoferritin-pathogen-magnetic bead sandwich constructs. The limit of detection of the apoferritin nanoprobe magnetic bead system is being compared to the experiment using a dye-conjugated antibody. TEM images show the apoferritin-pathogen-magnetic
Fig. 4. Detection of inactivated F. tularensis via apoferritin nanoprobes and magnetic beads. (a) The limit of detection for measuring inactivated F. tularensis with apoferritin nanoprobes and magnetic beads. (b) The specificity of the system is assessed via blinded samples containing different ratios of F. tularensis, B. anthracis, and Y. pestis (F: Francisella tularensis, B: Bacillus anthracis, Y: Yersinia pestis)
bead immunocomplexes (Fig. 3a). Pathogen detection was further corroborated through fluorescence images of the immunocomplexes, in which a strong red fluorescence is observed in the presence of pathogen compared to the negative control (Fig. 3b). It is worth noting that traditional immunochromatographic assays for F. tularensis have a detection limit of approximately 105 ∼ 107 CFU/mL. In our apoferritin nanoprobe system, however, the use of specific antibodies and multiple QDs to one apoferritin unit increased the detection sensitivity to approximately 104 CFU/mL of F. tularensis (Fig. 4a). Furthermore, the specificity of our system was assessed via blind tests of samples including inactivated F. tularensis and two other pathogens – Y. pestis and B. anthracis (Fig. 4b) (concentration of 104 CFU/mL each). Given that sample number 6 – F. tularensis alone – showed a comparative level of fluorescence intensity to that of the original assay, the near absence of fluorescence for sample numbers 7 and 8 – Bacillus anthracis and Yersinia pestis, respectively – confirm the specificity of our system. Other combinations, in which F. tularensis is mixed with other bacteria, showed a weaker, but still detectable, level of fluorescence intensity. Finally, while F. tularensis alone is a serious threat to public health, its combination with other high-risk pathogens presents an even graver risk. The ability to detect multiple pathogens simultaneously is therefore highly desirable. From this perspective, we assess the capacity of our system for the simultaneous detection of two pathogens: F. tularensis and B. anthracis (Fig. 5). Apoferritin nanoprobes are prepared in the same manner as before, but with different antibodies and with the use of fluorescent molecules
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
193
Fig. 5. Scheme of dual capturing of pathogens by apoferritin nanoprobes.
rather than QDs. Upon mixing F. tularensis and B. anthracis in equal amounts, ranging from 102 CFU/mL to 106 CFU/mL individually, apoferritin nanoprobes with antibodies against F. tularensis (green fluorescence) or B. anthracis (red fluorescence) are used to measure their respective limits of detection. The limit of detection for F. tularensis was approximately 104 CFU/mL, similar to that of single antigen spiked samples, while that of B. anthracis was 104 CFU/mL as well (Fig. 6a and Fig. 4a). Increasing amounts of
pathogen resulted in an increase in apoferritin-pathogen-magnetic bead immuno complex formation, as observed via bright field and fluorescence images (Fig. 6b). Additionally, we observed similar fluorescent intensity at each pathogen concentration, indicative of the detection of similar numbers of pathogens. To further confirm that our system can be used to accurately detect the amount of each pathogen from a non-equal mixture of two pathogens, 10 times more F. tularensis than B. anthracis were mixed together.
Fig. 6. Dual capturing of pathogens by apoferritin nanoprobes. (a) Individual limits of detection from a solution of equal amounts of inactivated F. tularensis and B. anthracis. (b) Fluorescent images of captured F. tularensis (green color) and B. anthracis (red color), each labeled with two different colored sets of apoferritinn nanoprobes. (c) Quantified fluorescent intensities of captured inactivated F. tularensis and B. anthracis when mixed together in a 10:1 ratio (F. tularensis: B. anthracis). (d) Fluorescent images of captured inactivated F. tularensis and B. anthracis from a solution of 10:1 F. tularensis to B. anthracis. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
194
J.-e. Kim et al. / Journal of Hazardous Materials 298 (2015) 188–194
The fluorescence intensities of 106 CFU/mL of F. tularensis and 105 CFU/mL of B. anthracis (Fig. 6c–d) yielded comparable values to those obtained from each concentration in equally-mixed samples (Fig. 6a–b). Overall, we demonstrate the capacity of our apoferritin probe system for the simultaneous detection of multiple pathogens without interference between the different targets. It is worth noting that our apoferritin nanoprobes utilize antibodies specific to live pathogens that are not fully applicable to their inactivated counterparts. Furthermore, we make use of commercial antibodies, which present limitations with regards to their binding efficiency. In other words, the lowest concentration detectable with our system is dependent on the performance of the commercial antibody. Recent reports employing target-specific peptides as an alternative to antibodies demonstrate an improved level of antigen binding by ferritin probes [18,21,22]. Similar approaches for the development of F. tularensis-specific peptides should increase the breath and sensitivity of our apoferritin nanoprobes. 4. Conclusions The reported apoferritin nanoprobes in this study have been shown to increase fluorescence intensity and were used to improve the limit of detection for high-risk pathogens. We demonstrate a specificity and an increased sensitivity towards F. tularensis via apoferritin nanoprobes carrying an antibody for inactivated F. tularensis. Specifically, approximately 10 times lower levels of F. tularensis were detected using our system than were via traditional antibody-fluorescent dye combinations. Furthermore, we utilize our system to simultaneously detect multiple pathogens, and in effect demonstrate the ease with which different apoferritin nanoprobes specific for various pathogens can be constructed. It should be noted that the stability of nanoprobes packaged in an assay should be further evaluated before any implementation towards on-site detection systems. Promisingly, in an ongoing study, functionalized apoferritin nanoprobes stored for three weeks show a comparable level of conjugation efficiency and limit of detection to freshly-prepared samples. Increasing the stability of detection probes without decreasing the sensitivity is another important aspect that we are actively investigating for the practical application of our system for on-site detection capabilities. Ultimately, we envision that the high sensitivity of our system, coupled with a simple yet rapid measurement capability, present promising opportunities in applications that require the on-site detection of high-risk pathogens. Acknowledgments This work was supported by the Basic Science Research Program through the National Research Foundation (NRF) funded by the Ministry of Education (grant number: 2013R1A1A1012653), by the Korea Centers for Disease Control and Prevention (grant number: 2014-E45002-00) and the grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute(KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI14C3266).
References [1] T.F.J. Pikula, M. Beklova, Z. Holesovska, J. Pikulova, Geographic information systems in epidemilogy-Ecology of common vole and distribution of natural foci of Tularaemia, Acta Vet. Brno 71 (2002) 379–387. [2] J. Pikula, F. Treml, M. Beklova, Z. Holesovska, J. Pikulova, Ecological conditions of natural foci of tularaemia in the Czech Republic, Eur. J. Epidemiol. 18 (2003) 1091–1095. [3] J.M. Petersen, M.E. Schriefer, Tularemia: emergence/re-emergence, Vet. Res. 36 (2005) 455–467. [4] D.T. Dennis, T.V. Inglesby, D.A. Henderson, J.G. Bartlett, M.S. Ascher, E. Eitzen, A.D. Fine, A.M. Friedlander, J. Hauer, M. Layton, S.R. Lillibridge, J.E. McDade, M.T. Osterholm, T. O’Toole, G. Parker, T.M. Perl, P.K. Russell, K. Tonat, B. Working Group on Civilian, Tularemia as a biological weapon: medical and public health management, JAMA: J. Am. Med. Assoc. 285 (2001) 2763–2773. [5] P.C. Oyston, A. Sjostedt, R.W. Titball, Tularaemia: bioterrorism defence renews interest in Francisella tularensis, Nat. Rev. Microbiol. 2 (2004) 967–978. [6] J. Ellis, P.C. Oyston, M. Green, R.W. Titball, Tularemia, Clin. Microbiol. Rev. 15 (2002) 631–646. [7] W.D. Sawyer, H.G. Dangerfield, A.L. Hogge, D. Crozier, Antibiotic prophylaxis and therapy of airborne tularemia, Bacteriol. Rev. 30 (1966) 542–550. [8] E.M. Elnifro, A.M. Ashshi, R.J. Cooper, P.E. Klapper, Multiplex PCR: optimization and application in diagnostic virology, Clin. Microbiol. Rev. 13 (2000) 559–570. [9] E. Vinogradov, M.B. Perry, J.W. Conlan, Structural analysis of Francisella tularensis lipopolysaccharide, Eur. J. Biochem./FEBS 269 (2002) 6112–6118. [10] P. Schmitt, W. Splettstosser, M. Porsch-Ozcurumez, E.J. Finke, R. Grunow, A novel screening ELISA and a confirmatory Western blot useful for diagnosis and epidemiological studies of tularemia, Epidemiol. Infect. 133 (2005) 759–766. [11] D.R. Christensen, L.J. Hartman, B.M. Loveless, M.S. Frye, M.A. Shipley, D.L. Bridge, M.J. Richards, R.S. Kaplan, J. Garrison, C.D. Baldwin, D.A. Kulesh, D.A. Norwood, Detection of biological threat agents by real-time PCR: comparison of assay performance on the R.A.P.I.D., the LightCycler, and the Smart Cycler platforms, Clin. Chem. 52 (2006) 141–145. [12] K. Tomioka, M. Peredelchuk, X. Zhu, R. Arena, D. Volokhov, A. Selvapandiyan, K. Stabler, J. Mellquist-Riemenschneider, V. Chizhikov, G. Kaplan, H. Nakhasi, R. Duncan, A multiplex polymerase chain reaction microarray assay to detect bioterror pathogens in blood, J. Mol. Diagn.: JMD 7 (2005) 486–494. [13] M.J. Hepburn, A.J. Simpson, Tularemia: current diagnosis and treatment options, Expert Rev. Anti-Infect. Ther. 6 (2008) 231–240. [14] R. Grunow, W. Splettstoesser, S. McDonald, C. Otterbein, T. O’Brien, C. Morgan, J. Aldrich, E. Hofer, E.J. Finke, H. Meyer, Detection of Francisella tularensis in biological specimens using a capture enzyme-linked immunosorbent assay, an immunochromatographic handheld assay, and a PCR, Clin. Diagn. Lab. Immunol. 7 (2000) 86–90. [15] A.H. Peruski, L.F. Peruski Jr., Immunological methods for detection and identification of infectious disease and biological warfare agents, Clin. Diagn. Lab. Immunol. 10 (2003) 506–513. [16] W. Shen, B.D. Schrag, M.J. Carter, G. Xiao, Quantitative detection of DNA labeled with magnetic nanoparticles using arrays of MgO-based magnetic tunnel junction sensors, Appl. Phys. Lett. 93 (2008) 033903. [17] T.J. Lowery, R. Palazzolo, S.M. Wong, P.J. Prado, S. Taktak, Single-coil multisample proton relaxation method for magnetic relaxation switch assays, Anal. Chem. 80 (2008) 1118–1123. [18] S.H. Lee, H. Lee, J.S. Park, H. Choi, K.Y. Han, H.S. Seo, K.Y. Ahn, S.S. Han, Y. Cho, K.H. Lee, J. Lee, A novel approach to ultrasensitive diagnosis using supramolecular protein nanoparticles, FASEB J.: Off. Publ. Fed. Am. Soc. Exp. Biol. 21 (2007) 1324–1334. [19] J.Y. Ahn, H. Choi, Y.H. Kim, K.Y. Han, J.S. Park, S.S. Han, J. Lee, Heterologous gene expression using self-assembled supra-molecules with high affinity for HSP70 chaperone, Nucleic Acids Res. 33 (2005) 3751–3762. [20] M.P. Hwang, J.W. Lee, K.E. Lee, K.H. Lee, Think modular: a simple apoferritin-based platform for the multifaceted detection of pancreatic cancer, ACS Nano 7 (2013) 8167–8174. [21] E.J. Lee, K.Y. Ahn, J.H. Lee, J.S. Park, J.A. Song, S.J. Sim, E.B. Lee, Y.J. Cha, J. Lee, A novel bioassay platform using ferritin-based nanoprobe hydrogel, Adv. Mater. (2012) 4739–4744, 4730. [22] K.C. Kwon, J.H. Ryu, J.H. Lee, E.J. Lee, I.C. Kwon, K. Kim, J. Lee, Proteinticle/gold core/shell nanoparticles for targeted cancer therapy without nanotoxicity, Adv. Mater. 26 (2014) 6436–6441.
