Letters in Applied Microbiology ISSN 0266-8254

Editor’s Choice

ORIGINAL ARTICLE

Comparison of eleven commercially available rapid tests for detection of Bacillus anthracis, Francisella tularensis and Yersinia pestis  ska1, K. Zacharczuk1, D. Jacob2 and R. Grunow2 A.A. Zasada1, K. Formin 1 Department of Bacteriology, National Institute of Public Health – National Institute of Hygiene, Warsaw, Poland 2 Robert Koch-Institut, Centre for Biological Threats and Special Pathogens, Berlin, Germany

Significance and Impact of the Study: Rapid detection of highly pathogenic bacteria causing anthrax, plague and tularemia is crucial for the limitation of negative effects of a potential release (natural, accidental or deliberate). In the study, commercially available rapid tests for detection of Bacillus anthracis, Yersinia pestis and Francisella tularensis were investigated in terms of sensitivity, specificity and ease-toperform. The study showed problems which could be faced during testing and results interpretation. Conclusions from this study should be helpful not only in selection of the most appropriate test for particular group of First Responders but also in undertaking decisions in situation of a contamination suspicion which have high social and economical impacts.

Keywords Bacillus anthracis, field assays, first responders, Francisella tularensis, immunofiltration, lateral flow, rapid tests, Yersinia pestis. Correspondence Aleksandra Anna Zasada, Department of Bacteriology, National Institute of Public Health – National Institute of Hygiene, Chocimska 24, 00-791 Warsaw, Poland E-mail: [email protected] 2014/2300: received 10 November 2014, revised 8 January 2015 and accepted 8 January 2015 doi:10.1111/lam.12392

Abstract Yersinia pestis, Bacillus anthracis and Francisella tularensis cause serious zoonotic diseases and have the potential to cause high morbidity and mortality in humans. In case of natural outbreaks and deliberate or accidental release of these pathogens rapid detection of the bacteria is crucial for limitation of negative effects of the release. In the present study, we evaluated 11 commercially available rapid test kits for the detection of Y. pestis, B. anthracis and F. tularensis in terms of sensitivity, specificity and simplicity of the procedure. The results revealed that rapid and easy-to-perform lateral flow assays for detection of highly pathogenic bacteria have very limited sensitivity. In contrast, the immunofiltration assays showed high sensitivity but limited specificity and required a too complicated procedure to be applied in the field by nonlaboratory workers (e.g. First Responders like fire, police and emergency medical personnel). Each sample - whether tested negative or positive by the rapid tests - should be retested in a reference laboratory using validated methods.

Introduction Yersinia pestis, Bacillus anthracis and Francisella tularensis cause serious zoonotic diseases in humans called plague, anthrax and tularemia, respectively. Under natural conditions, infection may occur through direct contact with infected animals, animals’ tissues, bodily fluids, through ingestion of meat from infected animals, drinking contaminated water, via arthropods bites or through inhalation of infectious droplets (Bossi et al. 2004; Gierczy nski 2010; Harik 2013). These three pathogens have

the potential to cause high morbidity and mortality in humans – especially if carried by aerosols – and are, for this reason, listed as category A biothreat agents according to the Centers for Disease Control and Prevention (CDC). A category A agent is an organism that poses a risk to national security because it can be easily disseminated or transmitted from person to person, results in high mortality rates, has the potential for public health impact, might cause public panic and social disruption and requires special action for public health preparedness (CDC 2014). Responding effectively to situations arising

Letters in Applied Microbiology 60, 409--413 © 2015 The Society for Applied Microbiology

409

Rapid tests for highly pathogenic bacteria

A.A. Zasada et al.

from natural outbreaks and deliberate or accidental release of these pathogens depends on the ability to rapidly detect and identify the pathogens for a first risk assessment of such an incidence. It is especially important to provide First Responders – firefighters, police officers and emergency medical personnel – with easy-to-perform rapid field assays. In the present study, we evaluated eleven commercially available rapid tests for detection of Y. pestis, B. anthracis and F. tularensis in terms of sensitivity, specificity and simplicity of the procedure. Results and discussion Results of sensitivity and specificity of the tests are presented in Tables 1 and 2, respectively. The most sensitive ones were the ABICAP test kits as they were able to detect Y. pestis and F. tularensis in a concentration of 104 cells ml 1 and B. anthracis in a concentration of 106 cells ml 1. These results are similar to those obtained by Grunow et al. (2008) for ABICAP classic test kit – F. tularensis. We were not able to detect B. anthracis using BADD Anthrax tests even in the highest concentration tested. Yersinia pestis was detected only in the highest investigated concentration using BADD Plague tests. We could detect F. tularensis only in a concentration of 108 cells ml 1 and Y. pestis and B. anthracis in a concentration of 107 cells ml 1 using appropriate BioThreat Alert tests. Using appropriate SMART II test kit Y. pestis, F. tularensis and B. anthracis were detected in concentrations of 106, 107 and 108 cells ml 1, respectively. Although ABICAP test kits were the most sensitive some cross-reactivity with other species of bacteria was also observed. Positive results were obtained with Yersinia enterocolitica, Yersinia pseudotuberculosis, Enterobacter cloacae, Staphylococcus aureus and Escherichia coli strains using ABICAP classic test kit – Y. pestis and with two Bacillus weihenstephanensis strains and one Bacillus cereus strain using ABICAP classic test kit – B.a. sporen. No cross-reactivity was observed in other evaluated test kits. The series of rapid test kits SMART II, BioThreat Alert and BADD allow to obtain results within 15 min, the test procedure is simple and no washing step is necessary, as the tests are using the same principle as the broadly used pregnancy test kits (Gooding 2006). Moreover, the test kits can be stored at temperatures of 2–30, 15–30 and 1– 49°C, respectively, which means that there is no need to refrigerate any of the components of the kits during storage. But the sensitivity of the tests is very limited. In contrast, ABICAP test kits are characterized by high sensitivity, but the procedure is more complex including several washing steps. For this reason the ABICAP tests kits are not suitable for ‘in the field’ testing, especially 410

when conducted by First Responders. Results are given after about 40 min. Some of the components of the test kits need to be stored at 4°C. Limited sensitivity of SMART II, BioThreat Alert and BADD tests kits may be related to the very short reaction time and relative inadequate affinity and avidity of the used antibodies. An additional reason for the limited sensitivity might be the defined low sample volume of 100 ll, which includes the sample buffer in two (BioThreat Alert and BADD) of the three kits. In ABICAP test kits a volume of 500 ll of a sample is tested. The sensitivity of the ABICAP test kits may be high enough since the column format of the assay allows the application of larger volumes (Lucht et al. 2007). Most of the rapid detection methods used in microbiology are immunoassays based on surface antigens of bacteria. However, serological detection is a challenge as antibodies raised against one antigen often cross-react with other antigens (Longchamp and Leighton 1999; Rao et al. 2010). This problem was also shown in ABICAP test kits. Two of the three ABICAP test kits revealed false positive results: the kit for B. anthracis detection crossreacted with some closely related Bacillus species, but the kit for Y. pestis detection cross-reacted not only with other Yersinia species and other members of Enterobacteriaceae but also with unrelated Staph. aureus. The high sensitivity of the ABICAP test kits is related to limited specificity whereas the high specificity of SMART II, BioThreat Alert and BADD test kits is related to a limited sensitivity. An ideal diagnostic test has a high sensitivity combined with a high specificity. But it is extremely difficult to conform to these criteria. It is often necessary to find a reasonable balance between sensitivity and specificity (Lutkenhoner and Basel 2013). In case of ‘in the field’ detection of highly pathogenic bacteria, sensitivity seems to be a priority as positive samples are usually confirmed by more sophisticated methods in a reference laboratory. But poor specificity might cause false positive alarm with all negative public and financial consequences. When rapid tests kits are applied the high probability of false negative results must be kept in mind. Low sensitivity of the tests used in the field makes it necessary that even negative samples need to be retested in a reference laboratory. Therefore, we conclude that there is still a need for improvement of existing and for development of new technologies for rapid field detection of highly pathogenic micro-organisms that will be highly sensitive and easy-to-perform. Materials and methods The evaluated rapid tests included: SMART II Yersinia pestis Anti-F1 detection kit (New Horizons Diagnostics

Letters in Applied Microbiology 60, 409--413 © 2015 The Society for Applied Microbiology

A.A. Zasada et al.

Rapid tests for highly pathogenic bacteria

Table 1 Results of sensitivity evaluation of the rapid tests kits Sample concentration (bact. per ml) 108

Test kit

SMART II Yersinia pestis Anti-F1 detection kit (New Horizons Diagnostics Corporation) Plague BioThreat Alert (Tetracore) BADD Plague (ADVNT) ABICAP classic test kit – Y. pestis (Senova GmbH) SMART II anthrax spores detection kit (New Horizons Diagnostics Corporation) Anthrax BioThreat Alert (Tetracore) BADD Anthrax (ADVNT) ABICAP classic test kit – B.a. sporen (Senova GmbH) SMART II Francisella tularensis detection kit (New Horizons Diagnostics Corporation) Tularemia BioThreat Alert (Tetracore) ABICAP classic test kit – F. tularensis (Senova GmbH) SMART II Francisella tularensis detection kit (New Horizons Diagnostics Corporation) Tularemia BioThreat Alert (Tetracore) ABICAP classic test kit – F. tularensis (Senova GmbH) ++ (bold line), + (clear line),  (thin line),

107

Yersinia pestis ++ + ++  + ++ ++ Bacillus anthracis  

106

105

104

+



103



++

++ ++ + Francisella tularensis ssp. tularensis +   ++ ++ ++ + F. tularensis ssp. holarctica ++ +  ++ ++ ++ +





(very tiny or completly no line).

Corporation Baltimore, Maryland), Plague BioThreat Alert (Tetracore Rockville, Maryland), BADD Plague (ADVNT Scottsdale, Arizona), ABICAP classic test kit – Y. pestis (Senova GmbH, Weimar Germany), SMART II anthrax spores detection kit (New Horizons Diagnostics Corporation), Anthrax BioThreat Alert (Tetracore), BADD Anthrax (ADVNT), ABICAP classic test kit – B.a. sporen (Senova GmbH), SMART II Francisella tularensis detection kit (New Horizons Diagnostics Corporation), Tularemia BioThreat Alert (Tetracore), ABICAP classic test kit – F. tularensis (Senova GmbH). The ABICAP test kits represent immunofiltration assays whereas the other test kits represent lateral flow assays. The test kits were selected for evaluation as they are the ones most used in European laboratories offering rapid detection of highly pathogenic bacteria. For investigation the following bacterial strains were used: Y. pestis EV76, B. anthracis UD III/7, F. tularensis ssp. tularensis ATCC 6223, F. tularensis ssp. holarctica ATCC 29684. To prepare reference materials, each bacterial species was inoculated into PBS to a density of approx. OD600 = 10 directly from colonies. Therefore, B. anthracis was cultivated at 37°C over 7 days on sheep blood agar plates (SBA). Approximately 80% of the inoculated material was spores. The sporulation was controlled by spore staining using 5% malachite green solution over an incubation time of 45 min at 90°C according to a modified protocol by Rackette (Beyer 2009). After staining, spores appeared green and vegetative cells appeared red. The percentage of spores in a

solution was determined by microscopic counting of spores and vegetative cells in about 10 fields of view. For Y. pestis and F. tularensis we used colonies grown over 2– 3 days at 37°C on SBA or cysteine heart agar plates for F. tularensis, respectively. To estimate the number of colony forming units (cfu) of each suspension, dilutions were plated onto SBA (or cysteine heart agar plates for F. tularensis). Each sample was tested twice. Bacterial suspensions were inactivated by heat at 60°C for at least 22 h. The exception was B. anthracis which was treated by 1% peracetic acid for 30 min. with subsequent washing. All inactivated suspensions were checked for sterility by one-tenth of each volume over 14 days. After inactivation, cell numbers in each suspension were determined using a counting chamber and a real-time polymerase chain reaction (real-time PCR). Suspensions of all inactivated samples were analysed for cross-contamination with other target bacteria by real-time PCR and by immunological methods (direct immune fluorescence and enzyme-linked immunosorbent assay - ELISA). For evaluation of sensitivity of the tests 10-fold dilutions of the samples were prepared in PBS buffer for all test kits except ABICAP where samples were diluted in the sample dilution buffer provided in the kit. The adequate reactivity of the inactivated reference material was checked in standard laboratory enzyme linked with immunosorbent assays (ELISA, data not shown). For evaluation of specificity of the tests suspensions of 108 cells ml 1 of the following bacterial strains were prepared: Y. enterocolitica O:8 biotype 1B, Y. enterocolitica

Letters in Applied Microbiology 60, 409--413 © 2015 The Society for Applied Microbiology

411

412 +

coli



cloacae

Enterobacter

+

aureus

coccus faecalis

Enterococcus pneumoniae

Streptococcus

nt

nt

HD3

thuringiensis

nt

nt

nt

nt

++ (bold line), + (clear line),  (thin line),

GmbH)

F. tularensis (Senova

ABICAP classic test kit –

(Tetracore)

Tularemia BioThreat Alert

Diagnostics Corporation)

detection kit (New Horizons

SMART II Francisella tularensis

sporen (Senova GmbH)

nt

nt

ABICAP classic test kit – B.a.

nt

nt

BADD Anthrax (ADVNT)

Alert (Tetracore)

Anthrax BioThreat

Corporation)

Horizons Diagnostics

detection kit (New

SMART II anthrax spores

Y. pestis (Senova GmbH)

Nt

Nt

Nt

nt

(very tiny or completly no line), nt – not tested.

nt

nt

nt

nt

nt

nt

nt

nt 

berculosis

Escherichia

Bacillus

nt



O3

pseudotu-

Staphylo-

ABICAP classic test kit –

+

O8

enteroclitica

enteroclitica

Yersinia

BADD Plague (ADVNT)

(Tetracore)

Plague BioThreat Alert

Diagnostics Corporation)

detection kit (New Horizons

SMART II Yersinia pestis Anti-F1

Test kit

Y.

Yersinia

Bacterial strain

Table 2 Results of specificity evaluation of the rapid tests kits

nt

nt

nt

nt

nt

nt

nt

HD7

B. thuringiensis

nt

nt

nt



nt

nt

nt

nt

WSBC 10405

anensis

weihensteph-

Bacillus

nt

nt

nt

++

nt

nt

nt

nt

WSBC 10392

phanensis

B. weihenste-

nt

nt

nt

+++

nt

nt

nt

nt

cereus D17

Bacillus

nt

nt

nt

nt

nt

nt

nt

F2-1

cereus

B.

Rapid tests for highly pathogenic bacteria A.A. Zasada et al.

Letters in Applied Microbiology 60, 409--413 © 2015 The Society for Applied Microbiology

A.A. Zasada et al.

O:3 biotype 4, Y. pseudotuberculosis I, E. cloacae 148a, E. coli ATCC 25922, S. aureus ATCC 25923, Enterococcus faecalis A, Streptococcus pneumoniae T3, Bacillus thuringiensis HD3, B. thuringiensis HD7, B. cereus D17, B. cereus F2-1, B. weihenstephanensis WSBC 10392, B. weihenstephanensis WSBC 10405. All the tests were conducted according to manufacturers’ instructions and read visually. The tests were carried out in the framework of external quality assurance exercises conducted by the QUANDHIP project (CHAFEA agreement no. 2010 21 02) financed by the European Commission as part of the Health Programme (reference QUANDHIP). Acknowledgements The studies have arisen from the Project QUANDHIP (CHAFEA agreement no. 2010 21 02) which has been funded by the European Commission in the framework of the Health Programme. Conflict of Interests The authors declared no conflict of interests with respect to the research, authorship and/or publication of this article. References Beyer, W. (2009) Bacillus spp. In Mikrobiologische Diagnostik eds Neumeister, B., Geiss, H.K., Braun, R.W. and Kimmig, P. pp. 386–397. Stuttgart-New York: Georg Thieme Verlag. Bossi, P., Tegnell, A., Baka, A., Van Lock, F., Hendriks, J., Werner, W., Maidhof, H. and Gouvras, G. (2004) BICHAT guidelines for the clinical management of plague and bioterrorism-related plague. Euro Surveill 9. Available

Rapid tests for highly pathogenic bacteria

at: http://www.eurosurveillance.org/em/v09n12/0912232.asp (accessed on: 10 November 2014) Centers for Disease Control and Prevention. (2014) Emergency preparedness and response. Bioterrorism Agents/Diseases. Available at: http://emergency.cdc.gov/agent/agentlistcategory.asp (accessed on: 10 November 2014) Gierczy nski, R. (2010) Diagnostics and molecular epidemiology of Bacillus anthracis. Post Microbiol 49, 165–172. Gooding, J.J. (2006) Biosensor technology for detecting biological wafare agents: recent progress and future trends. Anal Chim Acta 559, 137–151. Grunow, R., Miethe, P., Conlan, W., Finke, E.-J., Friedewald, € ur€ S. and Porsch-Ozc€ umez, M. (2008) Rapid detection of Francisella tularensis by the immunoaffinity assay ABICAP in environmental and human samples. J Rapid Methods Autom Microbiol 16, 30–54. Harik, N.S. (2013) Tularemia: epidemiology, diagnosis, and treatment. Pediatr Ann 42, 288–292. Longchamp, P. and Leighton, T. (1999) Molecular recognition specificity of Bacillus anthracis spore antibodies. J Appl Microbiol 87, 246–249. Lucht, A., Formenty, P., Feldmann, H., Gotz, M., Leroy, E., Bataboukila, P., Grolla, A., Feldmann, F. et al. (2007) Development of an immunofiltration-based antigendetection assay for rapid diagnosis of Ebola virus infection. J Infect Dis 196(Suppl 2), S184–S192. Lutkenhoner, B. and Basel, T. (2013) Predictive modeling for diagnostic tests with high specificity, but low sensitivity: a study of the glycerol test in patients with suspected Meniere’s disease. PLoS One 8, e79315. QUANDHIP http://www.quandhip.info/Quandhip/EN/Home/ Homepage_node.html (accessed on: 10 November 2014) Rao, S.S., Mohan, K.V.K. and Atreya, C.D. (2010) Detection technologies for Bacillus anthracis: prospects and challenges. J Microbiol Methods 82, 1–10.

Letters in Applied Microbiology 60, 409--413 © 2015 The Society for Applied Microbiology

413

Comparison of eleven commercially available rapid tests for detection of Bacillus anthracis, Francisella tularensis and Yersinia pestis.

Yersinia pestis, Bacillus anthracis and Francisella tularensis cause serious zoonotic diseases and have the potential to cause high morbidity and mort...
101KB Sizes 0 Downloads 10 Views