Rapid and sensitive detection of Yersinia pestis using amplification of plague diagnostic bacteriophages monitored by real-time PCR

Abstract

Background: Yersinia pestis, the agent of plague, has caused many millions of human deaths and still poses a serious threat to global public health. Timely and reliable detection of such a dangerous pathogen is of critical importance. Lysis by specific bacteriophages remains an essential method of Y. pestis detection and plague diagnostics.

Methodology/principal findings: The objective of this work was to develop an alternative to conventional phage lysis tests--a rapid and highly sensitive method of indirect detection of live Y. pestis cells based on quantitative real-time PCR (qPCR) monitoring of amplification of reporter Y. pestis-specific bacteriophages. Plague diagnostic phages phiA1122 and L-413C were shown to be highly effective diagnostic tools for the detection and identification of Y. pestis by using qPCR with primers specific for phage DNA. The template DNA extraction step that usually precedes qPCR was omitted. phiA1122-specific qPCR enabled the detection of an initial bacterial concentration of 10(3) CFU/ml (equivalent to as few as one Y. pestis cell per 1-microl sample) in four hours. L-413C-mediated detection of Y. pestis was less sensitive (up to 100 bacteria per sample) but more specific, and thus we propose parallel qPCR for the two phages as a rapid and reliable method of Y. pestis identification. Importantly, phiA1122 propagated in simulated clinical blood specimens containing EDTA and its titer rise was detected by both a standard plating test and qPCR.

Conclusions/significance: Thus, we developed a novel assay for detection and identification of Y. pestis using amplification of specific phages monitored by qPCR. The method is simple, rapid, highly sensitive, and specific and allows the detection of only live bacteria.

Figure 1. Lytic properties of bacteriophages ϕA1122, L-413C, and P2 vir1 towards Y. pestis CO92 pgm⁻.

The dynamics of lysis was determined in BHI broth at multiplicity of infection of 0.1. Optical density was normalized to the start of infection (1 on the Y axis corresponds to the initial OD₆₀₀ = 0.2).

Figure 2. Determination of lysis speed and burst sizes for bacteriophages ϕA1122, L-413C, and P2 vir1 on Y. pestis CO92 pgm⁻.

Phage burst sizes (an average phage progeny produced by one bacterial cell) correspond to plateaus on the curves.

Figure 3. Parameters of ϕA1122- and L-413C-based qPCR tests for phage DNA and live phage particles determined by linear regression method.

A and B, standard curves plotted for DNA concentrations of ϕA1122 and L-413C, respectively. C and D, standard curves plotted for live phage particles of ϕA1122 and L-413C, respectively.

Figure 4. Dynamics of growth of phages ϕA1122 and L-413C on different concentrations of Y. pestis cells detected by qPCR.

The starting points of phage infection correspond to 100 PFU per 1 µl sample and are normalized to 1. A. The titer rise of ϕA1122. B. L-413C amplification.

Figure 5. qPCR tests on simulated clinical human blood samples.

A. Linear regression of ϕA1122 DNA concentration in blood diluted 1∶20 in comparison with SM buffer data. B. ϕA1122-based detection of Y. pestis in artificially contaminated blood diluted 10-fold with BHI broth. To calculate the actual bacterial loads in the undiluted blood samples, the CFU numbers shown should be multiplied by 10. The starting points of phage infection correspond to 100 PFU per 1 µl sample and are normalized to 10⁰ = 1.