Development of xMAP Assay for Detection of Six Protein Toxins

Jul 9, 2012 - xMAP technology was used for simultaneous identification of six protein toxins (staphylococcal enterotoxins A and B, cholera toxin, rici...
2 downloads 12 Views 903KB Size
Download PDF
Letter pubs.acs.org/ac

Development of xMAP Assay for Detection of Six Protein Toxins Maria A. Simonova,* Tatiana I. Valyakina, Elena E. Petrova, Ravilya L. Komaleva, Natalia S. Shoshina, Larisa V. Samokhvalova, Olga E. Lakhtina, Igor V. Osipov, Galina N. Philipenko, Evgeniy K. Singov, and Evgeniy V. Grishin Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Moscow, Russia 117997 S Supporting Information *

ABSTRACT: xMAP technology was used for simultaneous identification of six protein toxins (staphylococcal enterotoxins A and B, cholera toxin, ricin, botulinum toxin A, and heat labile toxin of E. coli). Monoclonal antibody-conjugated xMAP microspheres and biotinilated monoclonal antibodies were used to detect the toxins in a sandwich immunoassay format. The detection limits were found to be 0.01 ng/mL for staphylococcal enterotoxin A, cholera toxin, botulinum toxin A, and ricin in model buffer (PBS-BSA) and 0.1 ng/mL for staphylococcal enterotoxin B and LT. In a complex matrix, such as cow milk, the limits of detection for staphylococcal enterotoxins A and B, cholera toxin, botulinum toxin A, and ricin increased 2- to 5-fold, while for LT the detection limit increased 30-fold in comparison with the same analysis in PBS-BSA. In the both PBS-BSA and milk samples, the xMAP test system was 3−200 times (depending on the toxin) more sensitive than ELISA systems with the same pairs of monoclonal antibodies used. The time required for a simultaneous analysis of six toxins using the xMAP system did not exceed the time required for ELISA to analyze one toxin. In the future, the assay may be used in clinical diagnostics and for food and environmental monitoring.

P

All the toxins listed above differ in structure, physicochemical properties, and toxic mechanisms. They have similar ways of ingress into the organism, namely, through environmental media (foodstuffs, water, soil, etc.) and certain similar symptoms at various poisoning stages (gastroenterological disorders and manifestations of general intoxication). Combined poisoning of the human organism by more than one toxin may also occur. The toxins listed above are highly hazardous for human health. Confirming the clinical diagnosis in case of a disease and prevention of disease propagation require specific sensitive methods for the toxin determination. Two fundamentally different approaches are used for the detection of biological toxins: detection of the originator or detection of the toxins it produces. In many cases, e.g., with staphylococcal enterotoxins, ricin, and botulinus toxins, poisoning symptoms can also occur in the absence of an originator. Therefore, the detection of biotoxins is an important stage in the diagnostics of potential infectious diseases that occur without characteristic symptoms. Taking into consideration that the dose of a toxin that causes a disease is generally extremely low, laboratory diagnostics of toxins is aimed determining residual amounts, where the determination sensitivity is a few picograms per milliliter of a sample being analyzed.2 Toxins are traditionally determined using a biological method, by observing their pathologic effect on unicellular

rotein toxins are produced by various types of living organisms, such as bacteria, fungi, plants, and animals. These proteins usually possess enzyme activity and can affect the human organism at exceptionally low concentrations. Bacterial toxins include staphylococcal enterotoxins A and B (SEA and SEB), cholera toxin (CT), botulinus toxin A (BoNTA), and heat labile E. coli enterotoxin (LT). Plant toxins include ricin found in castor beans. Staphylococcal enterotoxins A and B produced by Staphylococcus aureus bacteria are among the most common reasons of food poisoning. Cholera toxin produced by Vibrio cholerae is responsible for the development of symptoms of cholera, a severe disease whose main symptom is abundant diarrhea. Its related heat labile enterotoxin, which is produced by the toxicogenic E. coli strains, is responsible for the symptoms of the so-called “traveler’s diarrhea”, a less hazardous but more common disease than cholera. The evolution of symptoms of botulism, a severe acute toxico-infectious disease, arises from the ingress of botulinum neurotoxin of clostridia bacteria, Clostridium botulinum, into the organism. The most common human food botulism is most often caused by type A toxin.1 BoNTA has extremely strong neurotoxic and hemagglutinating effects. Ricin is a highly toxic glycoprotein that is accumulated in ripe Ricinus communis beans. It is the most toxic representative of plant lectins. Ricin inhibits protein synthesis in eukaryotic cells and is toxic regardless of the ways of ingress into the organism. Symptoms of ricin poisoning include the retina hemorrhage, nausea and vomiting, strong abdominal pain, bloodstained diarrhea, convulsions, prostration, and collapse. © XXXX American Chemical Society

Received: June 6, 2012 Accepted: July 9, 2012

A

dx.doi.org/10.1021/ac301525q | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Letter

sulfo-NHS/EDC mediated coupling. Briefly, a carboxylated microsphere stock suspension (5.0 × 106 microspheres) was diluted with 0.08 mL of 100 mM monobasic sodium phosphate (PB), pH 6.2. 0.01 mL of EDC (50 g/L in water), and 0.01 mL of sulfo-NHS (50 g/L in water) were added to the microspheres suspension. Following 20 min of incubation, the activated microspheres were washed twice with 0.25 mL of 50 mM MES, pH 5.0 by centrifugation. The activated and washed microspheres were suspended in 0.1 mL of MES. 25, 5, or 1 μg of mAbs were added to the resuspended microspheres. After 2 h of coupling at room temperature in the dark, the microspheres were washed with 0.5 mL of PBS with 0.1% BSA, 0.02% Tween-20, and 0.05% azide (PBS-TBN). Coupled microspheres were stored at 2−8 °C in the dark. Microsphere sets 29, 36, 55, 68, 87, and 93 were conjugated to the anti-LT, anti-CT, anti-SEB, anti-BoNTA, anti-SEA, and antiricin antibodies, respectively. xMAP Sandwich Immunoassays. A volume of 0.05 mL of antibody-coupled microspheres cocktail (5 × 103/mL for each set) was added to wells of the MABVN1250 plate (Millipore). Volumes of 0.05 mL of toxin containing or toxin free samples were added to in the each well. The plate was incubated 1 h at room temperature in the dark. After incubation the plate was washed 3 times with wash buffer (1% BSA in PBS, 0.2 mL per well). A volume of 0.1 mL per well of the biotinilated antibodies cocktail was added. After incubation for 1 h at room temperature in the dark, the plate was washed 3 times with wash buffer. A volume of 0.1 mL per well of streptavidin-Rphycoerythrin solution was added. After incubation for 0.5 h at room temperature in the dark, the plate was washed 3 times with wash buffer and 1 time with PBS. After this, the microspheres in the wells were resuspended in PBS (0.1 mL per well), and sandwich immunoassays were measured on the Luminex 200 instrument with the PMT on the standard setting. At least 100 microspheres of each set were measured per data point of the dose−response experiment in each sample. Data were collected using the Luminex IS 2.3 software. The background fluorescent signals were automatically subtracted from the results. Following the analysis of samples spiked with the toxins, limits of detection (LODs) were determined as the lowest concentration tested with a signal greater than 3 times the standard deviation of the mean background fluorescence. Assays run in PBS-BSA were considered true measures of performance. ELISA sandwich immunoassays were performed as described in the Supporting Information.

organisms, cell cultures, chicken embryos, as well as on laboratory animals, such as mice or guinea pigs. These methods are very laborious, take a lot of time (about 2−6 days) and have high self-cost: the single sample is tested on several animals; the tests are carried out only in laboratories authorized for handing animals and by employees who have undergone specialized training.3 Detection of toxins is also performed by instrumental methods using equipment with high detection sensitivity: mass spectrometry, high-performance liquid chromatography, capillary electrophoresis, etc. The sensitivity of these methods is comparable to that of biological methods, but their use is limited to specially equipped laboratories that possess expensive instrumentation and highly qualified personnel, which makes them inaccessible to the majority of scientists.4 Currently, immunoassay methods are most popular in the practical diagnostics of bacterial toxins. Enzyme-linked immunosorbent assay (ELISA) methods are used most widely.4,5 This is because the method has high sensitivity, is specific, enables determinations in samples of nearly any nature, is easy to perform and interpret the results, and the analysis cost is comparatively small. The flow cytometry method is also popular in immunochemical assay. Antibodies covalently bound with microsphere surfaces can specifically bind analytes, including toxins, which are then identified by detecting molecules conjugated with the fluorochrome. The xMAP technology6 developed by Luminex employs specific microspheres that differ in the content of red and infrared fluorescent dyes. Microspheres with a particular color ratio belong to the same type which is identified by an ID number, and each such type of microspheres has its own location of the map (dot blot) (so-called spectral address). The variety of the commercial microsphere types makes it possible to perform up to 500 various tests in one reaction volume. In addition, microspheres of different types can be combined, and even then they maintain their characteristic features. In this work, we performed a study aimed at the development of a test system based on xMAP technology for simultaneous determination of SEA and SEB, CT, LT, ricin, and BoNTA using monoclonal antibodies (MAbs) received earlier by us and our collaborators.

■

MATERIALS AND METHODS Materials. Carboxylated xMAP microspheres were obtained from Luminex Corporation. Phosphate buffered saline (PBS), Tween-20, and bovine serum albumin (BSA) were purchased from Sigma-Aldrich. 1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC), N-hydroxysulfosuccinimide (sulfo-NHS), and NHS-LC-biotin were purchased from Pierce. Streptavidin-conjugated phycoerythrin (SA-PE) was purchased from One Lambda. Antimouse phycoerythrin-conjugated antibodies were purchased from Dako (Sweden). SEA, SEB, LT, and BoNTA was kindly gifted by Dr. Yu. V. Vertiev (Gamaleya Research Institute of Epidemiology and Microbiology, Russian Academy of Medical Sciences, Moscow, Russia). Ricin was a kind gift from Dr. F. A. Brovko (Branch of Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences, Pushchino, Moscow Region, Russia). CT was purchased from Sigma-Aldrich. Monoclonal antibodies against SEA, SEB, CT, LT, ricin, and BoNTA were prepared and biotinilated as described in the Supporting Information. Preparation of Antibody-Conjugated Microspheres. The capture MAbs to each toxin were coupled to the xMAP microspheres using the recommended Luminex procedure of

■

RESULTS AND DISCUSSION In this work, we were developing the method for multiplex xMAP analysis of six toxins. It is well-known that antibodies used in multiplex analysis must not have the cross-reactivity toward other antigens being analyzed. Yet another requirement for the antibodies in multiplex analysis is that the background signal must be as low as possible. To develop a test system based on multiplex analysis, we used the one pair of antibodies to BoNTA, the 1 pair of antibodies to ricin, the 4 pairs of antibodies to SEA, the 5 pairs of antibodies to SEB, the 16 pairs of antibodies to CT, and the 5 pairs of antibodies to LT. At the first stage, we selected the antibody pairs most optimal for the multiplex xMAP “sandwich” analysis. Selection of Antibodies Pairs for Development of xMAP Assay. At first, selection of the binding antibodies was based on the maximum specificity for each of the toxins, along with selection of the detecting antibodies for the corresponding B

dx.doi.org/10.1021/ac301525q | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Letter

Figure 1. Dose−response curves obtained by a simultaneous multiplexed xMAP sandwich assay of six protein toxins, SEA, SEB, CT, ricin, BoNTA, and LT, in PBS-BSA and milk samples. The variation coefficient did not exceed 15%. Calibration curves for low-range concentrations are shown in the insets. The dashed lines correspond to fluorescence level exceeding the signal emitted by negative control microspheres by 2 standard deviations.

toxin featuring the smallest fluorescence background signal. For this purpose, the xMAP analysis was performed using the following scheme: microspheres with the immobilized specific MAbs to all the six toxins were incubated with various concentrations of one of the toxins and with the detecting antibodies specific to the given toxin. The binding antibodies that manifested cross-reactivity as well as the detecting antibodies with high background fluorescence level (above 15 fluorescence units) were excluded from subsequent analysis. Further, the selected binding and detecting antibodies were used in the analysis to determine the specificity of the detecting antibodies. For this purpose, the microspheres conjugated with specific MAbs to SEA, SEB, CT, ricin, BoNTA, and LT selected

at the first stage were incubated with various concentrations of one of the toxins listed above and with the detecting antibodies to all the toxins. The detecting antibodies which possessed cross-reactivity were excluded from the analysis. Further, the sensitivity of multiplex toxin detection in the model buffer (PBS-BSA) was determined. For this purpose, the microspheres conjugated with the selected specific MAbs to SEA, SEB, CT, ricin, BoNTA, and LT were incubated with various concentrations of all the toxins listed above and with the detecting antibodies to all the toxins selected at the previous stages. On the basis of the experiments, the specific pairs of antibodies were selected which ensured the maximum C

dx.doi.org/10.1021/ac301525q | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Letter

Table 1. Limits of Detection (ng/mL) of Six Toxins in Buffer (PBS-BSA) and Milk with xMAP Sandwich Assay and Sandwich ELISA with the Same Pairs of Antibodies Used xMAP assay buffer milk

ELISA

SEA

SEB

CT

ricin

BoNTA

LT

SEA

SEB

CT

ricin

BoNTA

LT

0.01 0.02

0.07 0.1

0.01 0.02

0.01 0.03

0.01 0.04

0.1 3

1 4

8 8

0.2 0.2

0.2 0.3

2 2

2 10

of fluorescent signals is caused by the compounds contained in the milk. The fact that the milk components can interfere with LT determination by immunoassay is known from the literature.7 It was shown in the paper cited above that the inhibition of LT binding with antibodies is caused by the proteins with molecular mass about 400 000. This molecular mass approximately matches that of dimeric class A immunoglobulins, which are known8 to be present in milk and can bind LT, thus neutralizing its toxic effect. The decrease in the sensitivity of LT determination can also be explained by the toxin binding with ganglioside Gm1 present in milk, which neutralizes the toxic activity of LT and CT.9 However, since cow milk gangliosides have comparatively low neutralizing activity and the sensitivity of CT determination in the milk by the xMAP method developed by us remained almost unchanged in comparison with the determination of this toxin in PBS-BSA (Table 1), obviously, in our case the decrease in LT determination sensitivity was most likely due to milk immunoglobulins specific to this toxin. This can be supported by the fact that E. coli bacteria are part of the normal microflora in animals, including cows. Accordingly, the titer of natural immunoglobulins to this toxin in the organism and in its secretions may have increased levels. The cholera toxin is a close structural analogue of LT, hence the majority of antibodies produced in the organism in response to LT should also interact with CT. However, it was shown in our experiments that the sensitivity of CT determination in the milk decreased to a smaller extent than that in the case of LT (Table 1). This phenomenon may be explained by the absence of any effect of interaction of CT with milk immunoglobulins on its binding with the monoclonal antibodies used in the analysis. The particular attention should be paid to the fact that monoclonal antibodies to LT and CT were obtained and selected based on the absence of the crossreactions. Accordingly, the antigenic determinants for which monoclonal antibodies to CT were selected are unique for this particular toxin and are most likely not shielded by natural milk antibodies specific to LT. Staphylococci are also capable of persisting in the animal organisms, and it is known10 that the level of natural antibodies to staphylococcal enterotoxins in milk can have increased values. However, like in the case of CT, we did not observe any considerable changes in the sensitivity of SEA and SEB determination in the milk in comparison with the similar parameters in PBS-BSA. Yet another explanation for this fact is that natural antibodies to the staphylococcal toxins, if present in milk, did not affect the interaction of these toxins with the monoclonal antibodies used in the analysis. In this study we created the test system for the simultaneous determination of the six protein toxins by means of xMAP analysis. The sensitivity of determination of these toxins using the test system we created was much higher than that of the ELISA method with the use of the same antibody pairs. To date, the three studies are known where the protein toxins were determined by xMAP analysis.11−13 The sensitivity of the toxins

sensitivity of toxin detection. The optimum concentrations of the binding and detecting antibodies were also selected. The data are presented in the Supporting Information. Using the antibodies selected, we carried out subsequent studies on the multiplex analysis capabilities. Sensitivity of Multiplexed xMAP Assay in PBS-BSA. Figure 1 shows the dose−response curves of the multiplex assays for SEA, SEB, CT, ricin, BoNTA, and LT performed in PBS-BSA. The LODs were determined as the concentration measured with a mean fluorescence signal greater than the mean background fluorescence signal plus 3 times the standard deviation. Table 1 shows the xMAP LODs in PBS-BSA for SEA, SEB, CT, ricin, BoNTA, and LT compared with ELISA data obtained using the same pairs of antibodies. It is evident from these data that the toxin detection sensitivity provided by xMAP analysis is considerably higher (by 1−2 orders) than that of the traditional ELISA. Sensitivity of Multiplexed xMAP Assay in Cow Milk. Next we assessed the ability of xMAP microspheres to detect the same analytes in a compound matrix containing a mixture of proteins, carbohydrates, fats, and other organic substances that may affect the analysis. We used the commercial cow milk as the model of such a matrix. The milk samples were spiked with the target toxins, centrifugated, and analyzed using the xMAP sandwich protocol. Figure 1 shows the dose−response curves for the six toxins spiked into the cow milk. PBS-BSA is shown as the reference. The specificity of toxin analysis in milk remained unchanged, as judged by the low background values of the fluorescence signals, i.e., below 15 units. However, it should be noted that both the analysis sensitivity (Table 1) and the fluorescent signals of toxin-containing samples (Figure 1) decreased in milk. The most significant decrease in the intensity of fluorescent signals and in the analysis sensitivity was observed for LT. For the rest of the toxins, the decrease in the analysis sensitivity was not so pronounced (Table 1). It should also be noted that the analysis sensitivity of the corresponding toxin in the milk by the ELISA method did not decrease, except for SEA and LT (see Table 1). The decrease in fluorescent signals in milk may be due to a number of factors. First, the centrifugation of the milk samples might have resulted in the partial entrainment of the toxins into the deposit. Second, it may not be ruled out that the compounds contained in the milk affect the interaction of toxins with the binding antibodies. To check the first assumption, we made an xMAP analysis of the toxins in the milk without preliminary centrifugation of the samples. According to the analysis results, the sensitivity of toxin determination the in noncentrifuged milk samples did not statistically differ (p > 0.1) from that determined in the centrifuged samples (no data are provided), which rejects the first assumption. However, the microspheres were found to undergo a high degree of aggregation in the noncentrifuged samples, which increased considerably the time for data acquisition per sample. Hence, the decrease in the intensity D

dx.doi.org/10.1021/ac301525q | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Letter

(8) Cruz, J. R.; Gil, L.; Cano, F.; Caceres, P.; Pareja, G. Acta Paediatr. Scand. 1988, 77, 658−662. (9) Laegreid, A.; Otnaess, A. B.; Fuglesang, J. Pediatr. Res. 1986, 20, 416−421. (10) Valle, J.; Vadillo, S.; Piriz, S.; Gomez-Lucia, E. FEMS Microbiol. Immunol. 1991, 3, 53−58. (11) Anderson, G. P.; Taitt, C. R. Biosens. Bioelectron. 2008, 24, 324− 328. (12) Kim, J. S.; Anderson, G. P.; Erickson, J. S.; Golden, J. P.; Nasir, M.; Ligler, F. S. Anal. Chem. 2009, 81, 5426−5432. (13) Kim, J. S.; Taitt, C. R.; Ligler, F. S.; Anderson, G. P. Sens. Instrum. Food Qual. Saf. 2010, 4, 73−81.

determination in those studies varied considerably. In fact, the sensitivity of the toxins determination in the model buffer varied as follows: for ricin, from 1.612 to 3.311 ng/mL; for CT, from 0.00411 to 3.212 ng/mL; for SEB, from 0.06413 to 1.612 ng/mL. It was also demonstrated11 that the sensitivity of botulinic anatoxin A determination in the model buffer was 1.1 ng/mL. We failed to find data on the determination of SEA and LT by xMAP analysis in the literature. In general, the test system we developed demonstrates higher sensitivity of the toxins detection (except for CT) than the analogues known in the literature. Furthermore, the sensitivity of CT, botulinic anatoxin A, and ricin detection only in the model buffer was assessed.11,12 The sensitivity of multiplex detection in milk is only known for SEB;13 it amounts to 1.6 ng/mL. Thus, we have shown that it is possible to create a specific and sensitive test system for the detection of six protein toxins based on sandwich xMAP analysis. The time required for analyzing toxins by this method does not exceed the time required to analyze one toxin by the ELISA method. In prospect, the test system developed can be used in clinical diagnostics and in monitoring of foodstuffs and environmental objects.

■

ASSOCIATED CONTENT

* Supporting Information S

Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Fax: +74953350812. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors are grateful to Yu. V. Vertiev for staphylococcal toxins and the BoNTA supply, I. D. Vinogradova for help with BoNTA manipulations, F. A. Brovko for the ricin supply, Anna Shepelyakovskaya for antibodies to the staphylococcal toxins supply, S. G. Abbasova and N. V. Rudenko for antibodies to BoNTA and the ricin supply, and S.Veselyi for assistance during the preparation of the manuscript. This work has been supported in part by Advokate State Program/086 64 2007 1/6 01 2008/FSUE SC Signal (Russia)

■

REFERENCES

(1) Dembek, Z. F.; Smith, L. A.; Rusnak, J. M. In Medical Aspects of Biological Warfare; Dembek, Z. F., Ed.; Department of Defense, Office of The Surgeon General, U.S. Army, Borden Institute: Washington, DC, 2007; pp 337−353. (2) Scarlatos, A.; Welt, B.; Cooper, B. J. Food 2005, 70, 121−124. (3) Murray, P. R.; Baron, E. J.; Pfaller, M. A.; Jorgensen, J. H.; Yolken, R. H. Manual of Clinical Microbiology; ASM Press: Washington, DC, 2003. (4) Technologies and Techniques for Early Warning Systems to Monitor and Evaluate Drinking Water Quality: State-of-the-Art Review; Office of Research and Development National Homeland Security Research Center Research Report, United States Environmental Protection Agency: Washington, DC, 2005. (5) Peruski, A. H.; Peruski, L. F. Clin. Diag. Lab. Immunol. 2003, 10, 506−513. (6) http://www.luminexcorp.com (7) Otnaess, A. B.; Halvorsen, S. Acta Pathol. Microbiol. Scand. 1980, 88, 247−253. E

dx.doi.org/10.1021/ac301525q | Anal. Chem. XXXX, XXX, XXX−XXX