Surface plasmon resonance imaging of pathogens: the Yersinia pestis paradigm
- 1.2k Downloads
Yersinia pestis, causing deadly plague, is classified as a group A bioterrorism bacterium. Some recent DNA-based methods were used for detection of bioterrorism agents.
Y. pestis was used as a model organism to develop an immunosensor based on surface plasmon resonance imaging (SPRi) using monoclonal antibody against Y. pestis F1 antigen. The experimental approach included step-by-step detection of Y. pestis membrane proteins, lysed bacteria, intact bacteria, mock-infected powder and mock-infected clinical specimens. SPRi detected on average 106 intact Y. pestis organisms in buffer, in mock-infected powder and in a 1:4 mixture with HEL cells.
This study offers the proof-of-concept of the SPRi-based detection of a human pathogen in both environmental and clinical specimens.
KeywordsSurface plasmon resonance imaging Yersinia pestis Plague Detection
surface plasmon resonance imaging
phosphate buffered saline
human embryonic lung cells
biosafety level 3 laboratory
polymerase chain reaction
Plague is a deadly zoonosis caused by the bacterium Yersinia pestis . It remains a public health problem in many tropical countries including subtropical African countries [1, 2] and it is re-emerging in North Africa [3, 4]. There are no longer any plague foci in Europe, though plague had caused devastating epidemics for two millennia . Also, Y. pestis has been classified as a group A bioterrorism agent . Currently, the detection of Y. pestis in environmental and clinical specimens, is based on the culture of Y. pestis, a process requiring at least 24 h and a biosafety level 3 laboratory (BSL3) ; and PCR-based detection of DNA sequences , whose specificity has recently been challenged by the observation of pla sequences, thought to be Y. pestis-specific, in other organisms and non-infected rodents . Also, immunochromatography detection of the Y. pestis-specific F1 antigen [9, 10, 11] is used for research, as this diagnostic assay is not widely available. Recently, some new approaches such as high-throughput RT-PCR-coupled ESI–MS assay and Luminex were applied to the detection of bioterrorism agents [12, 13]. These techniques required principally amplified DNA and apparently remained time-consuming.
In this study, Y. pestis was used as a model organism to test whether surface plasmon resonance imaging (SPRi) could be used as a novel technique for the rapid detection of pathogens in environmental and clinical specimens. SPRi has advantages (high-throughput, real time, label-free, multi-detection and sensitive) which could be applied to the detection of organisms, such as Y. pestis. Several studies of immune reactions (cells—antibodies, peptides—antibodies) have been conducted with this technique [14, 15]. Recently, this technique was used to detect the plant pathogenic bacterium Acidovorax avenae subsp. citrulli . However, until now, SPRi has not been used to detect human pathogenic bacteria.
In this study, we challenged the proof-of-concept that SPRi could be used for the rapid detection of highly pathogenic organisms in environmental and clinical specimens, using Y. pestis as a model organism. We developed a step-by-step experimental approach to test membrane proteins, lysed bacteria, intact bacteria (Y. pestis Orientalis YPA, Medievalis 6B4), mock-infected powder and mock-infected clinical specimens.
Materials and instruments
CS-SPRi Biochips and CS-SPRi Slides covered by a thin layer of gold and functionalized NHS groups were purchased from HORIBA (Palaiseau, France). The ligand used in this study was a mouse monoclonal antibody (mAb) against the F1 antigen of Y. pestis [YPF19] (4.3 mg/mL) purchased from GenWay Biotech, Inc. (Gentaur, Belgium). A mouse non-immune control serum was produced and purified in our laboratory (URMITE, Marseille, France). The protocol to collect serum from non-immune mice has been approved by the French National Ethic Committee for Animals under the reference number 60-12112012. Sodium acetate, ethanolamine and glycine were purchased from Sigma-Aldrich (Saint-Quentin Fallavier, France), while phosphate buffered saline (PBS) was obtained from bioMérieux (La Balme-les-Grottes, France).
Ligands diluted in 10 mM sodium acetate, pH 5 at different concentrations (mAb: 1, 0.5, 0.25 mg/mL; control serum: 1 mg/mL) were automatically deposited onto the chip (6 spots for each ligand with a distance of 0.7 mm between each spot) using a 300 nm diameter ceramic needle controlled by the mechanical SPRi-Arrayer (HORIBA, Palaiseau, France). Needle rinsing with distilled water for 3 s, followed by drying with compressed air for 3 s, were automatically repeated 3 times both before and after each ligand was deposited. The antibody was immobilised at room temperature in a humid chamber set to 60% relative humidity. The chip was air-dried and placed in the chip box at 4°C until use.
Suspensions of Y. pestis strain YPA (an Orientalis biotype, CSUR P100) in PBS were sonicated 5 times for 1 min on ice at an amplitude of 30 W with Q700 Sonicator (Qsonica, LLC, DENTA LABO, Avignon, France). The tubes were centrifuged for 5 min at 4.000×g. Supernatant was ultra-centrifuged for 1 h at 100.000×g. The pellet containing membrane proteins was suspended in 500 µL PBS or 500 µL 0.2% Triton X-100, 30 mM Tris HCl pH 8 and 2 mM MgCl2 and was then left overnight at 4°C to solubilize the membrane proteins. Following the same procedure, Escherichia coli was used as a negative control.
Five hundred µL of various concentrations of Y. pestis YPA were broken with acid-washed glass beads in a screw-cap tube using a FastPrep®-24 Instrument (MP Biomedicals, Illkirch, France) at a speed 4.0 m/V for 40 s. The tube was then centrifuged for 30 s at 6.700×g and the supernatant was analysed with SPRi. Bartonella quintana, prepared according to the same protocol, was used as a negative control.
Virulent Y. pestis YPA and Y. pestis Medievalis 6B4 were cultured on Columbia agar and 5% sheep blood (bioMérieux) at 32°C, 5% CO2 for 3–5 days. E. coli and Staphylococcus aureus used as negative controls were cultured in the same medium at 37°C. Virulent Y. pestis was handled in a BSL3. Bacteria were inactivated with 70% ethanol. The SPRi specificity test was carried out with Y. pestis YPA, Y. pestis Medievalis, E. coli and S. aureus. SPRi sensitivity was tested with different concentrations of Y. pestis YPA.
A “sandwich” test (mAb/Y. pestis/mAb) was developed on SPRi in order to enhance the sensitivity of the SPRi assay. On a chip with 1 mg/mL immobilized mAb, Y. pestis YPA (1.2 × 101 to 1.2 × 107 CFU/mL) was tested within 10 min, followed by an injection of mAb of 1/500. The area under the curve for each injection was analysed using GraphPad PRISM V6 software (GraphPad Software, Inc., USA). The first phase (bacterial injection) from 0 to 11.5 min, the second phase (antibody injection) from 11.5 to 22 min and the entire process from 0 to 22 min were analysed.
Y. pestis YPA mixed at different concentrations (108, 106, 104 CFU/mL) with flour powder was tested on SPRi to estimate whether this technique could detect the pathogen in environmental samples in mimicking a bioterrorist alert. Powder mixed with either PBS or E. coli were used as negative controls. The experiment was repeated three times.
Mock-infected clinical specimens
Y. pestis YPA mixed with HEL cells at ratio 1:1, 1:10, 1:100 was used as a model to evaluate the capability of SPRi to detect the pathogen in infected clinical specimens. A suspension of non-infected HEL cells was used as negative control. This experiment was performed in triplicate.
The experiments were conducted using the SPRi-Plex II system and monitored using SPRi P5.0.2-View software (HORIBA, Palaiseau, France). The running buffer for the SPRi-Plex II system was 10 mM PBS. Initial buffer flow rate was 500–1,000 µL/min to fill the fluid system for 15 min. Once the chip was inserted into the machine, the analysis cell was filled at a flow rate of 750–1,000 µL/min, followed by a flow rate of 50 µL/min and a temperature of 37°C was used for all experiments. After system stabilization, plasmon images were acquired by software and system mirror and camera. The study area deposits were detected on the previously acquired high-contrast image and spot and spot family definitions were performed. The plasmon curve and resonance angle were determined for each spot. The mirror system was shifted by the resonance angle and the experiments were conducted with this value.
The surface of the chip was saturated for 10 min with 500 µL of 1 M ethanolamine pH 9 and regenerated for 10 min with 500 µL of 10 mM glycine pH 1.85 to remove any non-covalent bindings (unfixed antibodies and ethanolamine). The system was calibrated by comparing the reflectivity of 12.5 mM PBS and of 10 mM PBS buffer. Control serum not targeting Y. pestis was defined as a reference surface. This was used to make subtracted curves of each spot family in real time to eliminate non-specific signals. A 400 µL-volume of each sample was loaded into the system (200 µL for analysis, 200 µL for the carrier fluid). The interaction of each sample with the surface of the chip was measured for 10 min. Changes in reflectivity were monitored in real time on the graph and on the chip image. After each experiment, the chip was regenerated with 10 mM glycine pH 1.85.
The data were analysed using SPRi-Analysis software V1.2. The reflectivity change subtracted from the negative control plus two standard deviations was considered as positive. The data were analysed by means of a t-test in GraphPad PRISM V6 for p value (GraphPad Software, Inc., USA).
Tests with membrane proteins
Sensitivity with lysed and intact bacteria
Specificity and effect of mAb concentration
Mock-infected powder and clinical specimens
Rapid and accurate detection of harmful organisms in both clinical and environmental specimens is a constant goal to serve medical diagnosis and protection. SPRi, being reported as a high-throughput method for the detection of molecule interactions, including antigen–antibody interactions, held promise to help in the rapid detection of organisms. The results reported in the present study were validated by numerous controls, indicating that SPRi yielded the specific detection of Y. pestis used as a model organism.
As SPRi-based measures incorporate no-labelled Mab and since all substances could affect the reflectivity index of the solution, it was important to distinguish specific interactions from non-specific ones. To eliminate non-specific binding to antibodies immobilized on the chip, the antibody not targeting the bacteria could be used as a blank . We used the non-immune purified serum as a reference surface and the injection of other bacteria as negative controls. Reference surface interactions were automatically subtracted and results compared with the negative control to identify the specific interactions.
Moreover, specificity was confirmed by the observation that Gram-negative (E. coli, B. quintana) and Gram-positive (S. aureus) bacteria did not give any signal with anti-Y. pestis mAb. Moreover, signal intensity increased with antibody concentration, as has previously been reported . Here, 6.4.106/mL lysed or intact bacteria alone or in mock-infected powder and clinical specimens could be detected by SPRi. SPRi was significantly less sensitive when lysed bacteria were used rather than intact bacteria, by contrast with a previous observation involving A. avenae subsp. citrulli . The sandwich assay, however, was able to 2.4.105 Y. pestis bacteria, in line with a previous study by Puttharugsa et al. .
Altogether, SPRi was able to detect 105 Y. pestis bacteria, an inoculum in the range of that expected to cause deadly plague after natural or criminal exposure to the pathogen. Indeed, the lowest inoculum of Y. pestis that consistently gave a 100% mortality rate in a mouse model was 104 CFU , suggesting that an inoculum of bacteria found in a bioterrorism attack would be >104 CFU. Specimen incubation for a few hours prior to the SPRi assay is a promising way to further improve sensitivity, as previously reported for E. coli O157:H7 . Some new approaches were applied in the detection of bioterrorism agents. These techniques were principally based on amplified DNA-sequence detection including high-throughput RT-PCR-coupled ESI–MS assay and Luminex assay [12, 13]. According to the current state of the art, SPRi is less sensitive than PCR-based techniques  and this sensitivity remains to be improved. Improving SPRi sensitivity is warranted as SPRi has advantages over PCR-based techniques: SPRi is specific whereas it has been shown that pla, a long-standing target for the PCR-based detection of Y. pestis is in fact detectable in non-Yersinia organisms, including the host . In particular, Y. pestis is easily engineered to intentionally modify PCR targets, thus helping the pathogen and bioterrorism agent escape detection. Also, SPRi is not too time-consuming, requiring only 40 min for a direct assay (20 min for infection of analyte, 20 min for negative control) and 1 h for a sandwich assay. A chip with many kinds of mAb could be used to detect different organisms. Moreover, this chip could be reused several times. This makes SPRi appropriate for rapid, cheap, multiplexed detection. Once sensitivity is improved, this technique would be perfectly suitable for sample analysis in the context of a bioterrorism emergency or for routine analysis in an epidemic area.
In conclusion, SPRi is a new technique for the rapid detection of bacteria in environmental and clinical specimens, as illustrated here using Y. pestis as a model organism. Future improvements will be directed towards increasing the sensitivity of the technique.
HTTH, GG, JT, GA, MD and EC: (1) have made substantial contributions to conception and design, or acquisition of data, or analysis and interpretation of data; (2) have been involved in drafting the manuscript or revising it critically for important intellectual content; (3) have given final approval of the version to be published; (4) agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All authors read and approved the final manuscript.
This work was supported by “Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes”, Marseille, France.
Compliance with ethical guidelines
Competing interests The authors declare that they have no competing interests.
- 5.Riedel S (2004) Biological warfare and bioterrorism: a historical review. Proc (Bayl Univ Med Cent) 17:400–406Google Scholar
- 6.Bearden SW, Perry RD (2008) Laboratory maintenance and characterization of Yersinia pestis. In: Current protocols in microbiology. Chapter 5. Wiley, New YorkGoogle Scholar
- 9.Bianucci R, Rahalison L, Massa ER, Peluso A, Ferroglio E, Signoli M (2008) Technical note: a rapid diagnostic test detects plague in ancient human remains: an example of the interaction between archeological and biological approaches (southeastern France, 16th–18th centuries). Am J Phys Anthropol 136:361–367PubMedCrossRefGoogle Scholar
- 11.Rajerison M, Dartevelle S, Ralafiarisoa LA, Bitam I, Dinh TN, Andrianaivoarimanana V et al (2009) Development and evaluation of two simple, rapid immunochromatographic tests for the detection of Yersinia pestis antibodies in humans and reservoirs. PLoS Negl Trop Dis 3:e421PubMedCentralPubMedCrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.