Efficient Detection of Pathogenic Leptospires Using 16S Ribosomal RNA

Abstract

Pathogenic Leptospira species cause a prevalent yet neglected zoonotic disease with mild to life-threatening complications in a variety of susceptible animals and humans. Diagnosis of leptospirosis, which primarily relies on antiquated serotyping methods, is particularly challenging due to presentation of non-specific symptoms shared by other febrile illnesses, often leading to misdiagnosis. Initiation of antimicrobial therapy during early infection to prevent more serious complications of disseminated infection is often not performed because of a lack of efficient diagnostic tests. Here we report that specific regions of leptospiral 16S ribosomal RNA molecules constitute a novel and efficient diagnostic target for PCR-based detection of pathogenic Leptospira serovars. Our diagnostic test using spiked human blood was at least 100-fold more sensitive than corresponding leptospiral DNA-based quantitative PCR assays, targeting the same 16S nucleotide sequence in the RNA and DNA molecules. The sensitivity and specificity of our RNA assay against laboratory-confirmed human leptospirosis clinical samples were 64% and 100%, respectively, which was superior then an established parallel DNA detection assay. Remarkably, we discovered that 16S transcripts remain appreciably stable ex vivo, including untreated and stored human blood samples, further highlighting their use for clinical detection of L. interrogans. Together, these studies underscore a novel utility of RNA targets, specifically 16S rRNA, for development of PCR-based modalities for diagnosis of human leptospirosis, and also may serve as paradigm for detection of additional bacterial pathogens for which early diagnosis is warranted.

Fig 1. L. interrogans 16S rRNA primers offered highest analytical sensitivity.

A) Total RNA samples isolated from cultured spirochetes were converted into cDNA and amplification cycles (cycle threshold or Ct value) of various L. interrogans target genes are assessed in qRT-PCR assays in the presence (gray bars) or absence (black bars) of hamster cDNA. Data represent results from three independent experiments. B) 16S rRNA primers display a high PCR efficiency. L. interrogans cDNA in RNase-free water (320 ng/μL) was serially diluted to tenfold (10⁻¹ to 10⁻⁹) and subjected to qRT-PCR assays using 16S-1 primers. Amplification cycles (left panel) were used to calculate standard curve (middle panel), which indicated detection to 10⁻⁹ dilutions with an amplification efficiency of 91.2%. A melt curve analysis (right panel) showed a melting temperature of 82°C without any non-specific amplification.

Fig 2. RNA-based detection is more sensitive than DNA.

Aliquots of L. interrogans bacteria were serially diluted from 10⁶ to 1 bacterium per milliliter of human blood and used for either RNA-based qRT-PCR (upper panel) or DNA-based qPCR (lower panel). Data represent results from three independent experiments.

Fig 3. High sensitivity and specificity of 16S rRNA qPCR assay for pathogenic and nonpathogenic leptospiral species and serovars.

Total RNA was extracted from 17 high or intermediate pathogenic, and five non-pathogenic leptospiral species, Borrelia burgdorferi, group A Streptococcus, and Escherichia coli, as well as from uninfected human blood or hamster liver, as described in the materials and methods and converted into cDNA. Equal amount (10 ng) of cDNA templates from each bacterial species were subjected to qRT-PCR assays using 16S-1 primers and amplification cycles (Ct values) were measured. Note that the sensitivity of detection is the best for all tested highly pathogenic species or serovars followed by intermediate species while non-pathogenic strains display the lowest sensitivity (an average of 15 Ct values or 10,000 fold less detectability). All tested non-target bacterial species or mammalian samples remained undetectable. Data represent results from three independent experiments.

Fig 4. Stability of Leptospira 16S transcripts in human blood.

A) Transcript stability in the blood treated with an RNA stabilization agent. Aliquots of human blood were spiked with leptospires (100 cells/ml), mixed with an RNase stabilization agent (TRIzol), and each aliquot was stored at room temperature for various times (0–120 hours). Following storage, levels of 16S rRNA transcripts were measured using qRT-PCR assays. Data represent results from three independent experiments. B) Transcript stability in the blood stored at various temperatures in the absence of any RNA stabilization agent. Spiked samples were prepared as described above and stored either at room temperature or at various cold temperatures (4°C, -20°C, and -80°C) up to 14 days, and transcript levels were monitored by qRT-PCR analyses. Transcript levels of “0 hour” were considered as 100%, which served as baseline controls, which displayed significant differences in transcript levels in groups marked by an asterisk (ANOVA, p<0.05). Data represent results from three independent experiments.