Development of a highly sensitive nucleic acid amplification-based detection for human leptospirosis infection

Leptospirosis, a zoonotic disease, is caused by infection with pathogenic strains of Leptospira [1]. The disease is distributed mainly in tropical and subtropical areas, particularly in agricultural regions [2]. Patients with leptospiral infection can experience various clinical manifestations, from nonspecific febrile illness to multi-organ failure and death. Leptospirosis symptoms are commonly confused with other causes of acute febrile illnesses, including dengue fever, melioidosis, influenza, scrub typhus and malaria, depending on the geographic overlapping of endemic pathogens [3,4]. This has caused the diagnosis of leptospirosis to rely on clinical manifestations of the patients in areas of endemicity. In addition, the primary diagnosis of the disease, such as clinical history, is rarely confirmed owing to the lack of availability and affordability of the diagnostic tools. Consequently, laboratory support is essential.

Several molecular-based detection techniques for Leptospira spp. have been developed recently, such as conventional PCR, high-resolution melting post-PCR analysis and quantitative PCR (qPCR). An emerging method for clinical diagnosis is polymerase chain displacement reaction (PCDR). PCDR is an approach combining conventional PCR with strand displacement amplification. The technique requires DNA polymerase that performs 5′ to 3′ strand displacement activity and lacks exonuclease activity [5,6]. In PCDR, at least two pairs of primers are required for the reaction. The amplification of all primers in the reaction is initiated simultaneously. As a result, the inner downstream nucleic acid strands are displaced during polymerization by the outer primers [6]. The displaced nucleic acid strands are employed as additional template strands, thus considerably increasing the sensitivity of the assay. Recently, a fast and highly sensitive quantitative PCDR (qPCDR)-based approach has been successfully applied for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral RNA detection [7].

The present study aimed to demonstrate the feasibility of applying a highly sensitive qPCDR-based technique for human leptospirosis detection. The sensitivity and specificity of the qPCDR-based technique using four oligonucleotide primers targeting the same conserved region of the lipL32 gene were determined. Finally, the efficiency of the qPCDR-based approach was compared with the conventional qPCR assay.

Materials & methods

Ethics statement

Ethical approval for this study was obtained from the Research Ethics Committee of the Faculty of Medicine, Prince of Songkla University (REC63-157-19-2).

Primer design

Primers used in this study were designed based on three criteria: consisting of two pairs of primers, one pair of outer primers and one set of inner primers; annealing to DNA belonging to pathogenic Leptospira spp. based on the consensus sequence of the lipL32 gene but not to intermediate and saprophyte groups; and targeting the same DNA region. Seventy-seven sequences from pathogenic species carrying the lipL32 gene were downloaded from GenBank for the partial or complete gene (Supplementary Table 1). The number (n) for each species was as follows: L. borgpetersenii (n = 16), L. kirschneri (n = 6), L. interrogans (n = 39), L. santarosai (n = 9), L. weilii (n = 2) and L. noguchii (n = 5). Multiple sequence alignments were performed using the ClustalW in BioEdit version 7.0.4.1. The primers were designed by Amplifx software and validated by Primer-Blast. All primer pairs are shown in Supplementary Figure 1.

Leptospira spp., bacterial strains & DNA extraction

Pathogenic Leptospira species; L. interrogans serovars Autumnalis (NR-20161) and non-pathogenic Leptospira species; L. biflexa were used in this study. In addition, one clinical isolate of other pathogenic organisms that are common causes of febrile illness or septicemia (i.e., Burkhoderia pseudomallei, Escherichia coli, Klebsiella pneumoniae, Staphylococcus aureus, Pseudomonas aeruginosa, Salmonella typhi, Plasmodium falciparum and P. vivax) were kindly provided by the Department of Pathology, Faculty of Medicine, Songklanagarind Hospital, Songkhla, Thailand. The DNA extractions were carried out from laboratory cultures by the Genomic DNA Extraction Kit (Promega, WI, USA). The resulting DNA samples were kept at -20°C.

Primer testing by PCR

The optimization of each primer set was verified through PCR amplification of the L. interrogans, which was used as a positive control. The PCR conditions were optimized by varying MgCl₂ concentration and primer annealing temperature. Each primer pair consisting of LipL32F1/LipL32R1 and LipL32F2/LipL32R2 was performed in an individual tube. A total of 25 μl of reaction mixtures was prepared as follows: 1× PCR buffer, 1–3 mM of MgCl₂, 0.25 mM of dNTP, 0.25 μm of each primer 0.5 unit of Taq DNA polymerase (Invitrogen, CA, USA) and 3 μl of L. interrogans gDNA. The PCR cycling condition was set as follows: initial denaturation at 95°C for 5 min, 35 cycles of denaturation at 94°C for 1 min, annealing at gradient temperatures varying from 54°C to 60°C for 1 min (each primer pair) and elongation at 72°C for 1 min. The last cycle was followed by heating at 72°C for 5 min.

qPCDR & qPCR assay

The qPCDR and qPCR assay were prepared in each 20 μl reaction volume, which contained 2 μl of gDNA of L. interrogans used as the template in the reaction. The qPCDR reaction mixtures were prepared as follows: 1× strand displacement polymerase (SD polymerase) reaction buffer, 3 mM MgCl₂, 0.25 mM deoxyribonucleotide triphosphate (dNTP), 0.4× SYBR Green I intercalating dye and 3U of SD polymerase (Bioron GmbH, Römerberg, Germany). The optimal concentration of each primer pair was 0.075 μm (LipL32F1/LipL32R1 and LipL32F2/LipL32R2). In this study, a low concentration of primers for the qPCDR test was used to avoid primer-dimer artifacts from primer excess. The qPCDR step was carried out as follows: initial denaturation at 92°C (2 min), 45 cycles of denaturation at 92°C (30 s), annealing at 60°C (30 s) and extension at 68°C (30 s). The master mix of qPCR contained two primers (250 nM each): LipL32F2 and LipL32R2, 10 μl of EvaGreen Supermix (Bio-Rad, CA, USA) and 2 μl of template DNA in a total volume of 20 μl. The qPCR step was carried out as follows: initial denaturation at 98°C (2 min), 45 cycles of denaturation at 98°C (5 s), annealing at 60°C (20 s) and extension at 72°C (20 s). All amplifications were performed using the LightCycler^® 480 system (Roche Diagnostics GmbH, Mannheim, Germany).

Sensitivity & specificity analysis of qPCDR & qPCR

The gDNA of L. interrogans was prepared by tenfold dilution with PCR-grade water, from 1 to 10^-5 ng/μl or 2 × 10⁵-2.0 gDNA copies/μl (as the size of the genome of L. interrogans strain is about 4.6 Mb, 1 genome is ∼5 fg gDNA). All dilutions were tested in duplicate, while a negative control was included in the run. The sensitivity of the two methods was compared by the crossing point (Cp) values. Lower Cp values imply higher amounts of the target nucleic acid. Testing for sensitivity could be reported in the form of the lowest concentration or limit of detection. The specificity was evaluated by non-target DNA templates from non-pathogenic Leptospira spp. and other pathogenic organisms that are common causes of febrile illness or septicemia (i.e., S. aureus, E. coli, P. aeruginosa, K. pneumoniae, B. pseudomallei, S. typhi, P. falciparum and P. vivax).

Detection of pathogenic Leptospira DNA diluted with human DNA

Human DNA was extracted from a 5 ml blood sample obtained from an individual using the QIAamp DNA mini blood kit (Qiagen, Hilden, Germany). Genomic DNA extracted from L. interrogans serovars Autumnalis was spiked into human DNA as diluent by tenfold serial dilution. The sensitivity of qPCDR and qPCR for pathogenic Leptospira spp. detection was determined as described above.

Results

Optimization of amplification conditions

The optimal condition for PCR analysis was as follows: dNTP concentrations of 0.25 mM, 3 mM of MgCl₂ and 0.25 µM of each primer. The annealing temperature for each set of primers specific for L. interrogans serovars Autumnalis was 60°C. The amplified product of outer primer (LipL32F1/LipL32R1) was 479 bp and 321 bp for inner primer (LipL32F2/LipL32R2), respectively.

Limit of detection

The sensitivity of the qPCDR and qPCR approaches for L. interrogans serovars Autumnalis detection was evaluated. SD polymerase employed for the amplification of the target DNA was a DNA polymerase with 5′ to 3′ strand displacement activity and lacking exonuclease activity. This polymerase has been successfully applied to DNA amplification techniques, such as isothermal DNA amplification, conventional PCR and PCDR [6,8–12]. Under the optimal conditions, qPCDR yielded lower Cp values compared with qPCR. The ΔCp between the two approaches was about 3–5 cycles (Table 1). The limit of detection of the qPCDR-based approach was 2 copies/μl of L. interrogans serovars Autumnalis within 35 cycles, while that of the qPCR-based approach was only 20 copies/μl (Figure 1A & B). Thus, qPCDR provided at least tenfold enhancement in sensitivity compared with qPCR. In addition, all reactions were negative using DNA from non-pathogenic Leptospira spp. and one representative each of S. aureus, E. coli, P. aeruginosa, K. pneumoniae, B. pseudomallei, S. typhi, P. falciparum and P. vivax (Figure 2).

Evaluation of qPCDR & qPCR assays

gDNA of L. interrogans serovars Autumnalis was diluted with human DNA by tenfold serial dilution. The concentration of gDNA was prepared in the range of 2 × 10⁵ to 2 gDNA copies/μl. In this study, qPCDR yielded lower Cp values compared with qPCR. The ΔCp between the two approaches was about 6–7 cycles. The limit of detection of the qPCDR-based approach was 2 × 10³ copies/μl of L. interrogans serovars Autumnalis within 36 cycles, while that of the qPCR-based approach was only 2 × 10⁴ copies/μl within 36 cycles (Table 1 & Figure 3A & B). Thus, qPCDR provided at least a tenfold enhancement in sensitivity compared with qPCR.

Discussion

During a pathogenic Leptospira endemic, the availability of highly sensitive diagnostic tools is essential for the mass screening of individuals. A new hybrid technique for nucleic acid amplification, qPCDR, is offered for the sensitive detection of a target pathogenic DNA. qPCDR is a new approach to targeting nucleic acid amplification that combines qPCR and strand displacement isothermal amplification [5,6]. Basically, the approach requires a conventional dsDNA denaturation and DNA polymerase with the following features: 5′ to 3′ strand displacement activity and the absence of 5′ to 3′ exonuclease activity. In the reaction, at least two to three pairs of primers are necessary. The amplification of all primers in the reaction, from the most outer to the most inner ones, is initiated at the same time. As using the DNA polymerase capable of strand displacement, the inner downstream nucleic acid strands are not degraded but displaced during polymerization by the outer primers. Thus, the displaced nucleic acid strands can be employed as template strands for additional amplification.

In the present study, four-primer qPCDR for the detection of pathogenic Leptospira spp. was used. The application of four-primer qPCDR enabled the detection of the target pathogenic DNA concentration tenfold lower than that required by qPCR under similar conditions. The Cp values were decreased by 4–7 cycles in four-primer qPCDR compared with qPCR, implying a shorter reaction time needed by qPCDR. In concordance with previous reports [5,6,9,12], the qPCDR-based approach exhibited higher sensitivity (at least tenfold) and greater efficiency than the qPCR-based assay. To be noted, the technique requires routine standard qPCR equipment, and no additional steps are necessary.

Conclusion

qPCDR enhanced the detection efficiency of pathogenic Leptospira spp. compared with a conventional qPCR approach. The qPCDR increased the sensitivity and decreased the detection time against the conventional amplification approach. Therefore, the qPCDR developed in this study is a promising tool for detecting pathogenic Leptospira spp.

Future perspective

Leptospirosis causes a broad range of symptoms in humans, frequently leading to misdiagnosis with other diseases. Without treatment, it can lead to organ dysfunction, including the kidneys, brain, lungs and liver, and even death. The accurate, reliable and timely diagnosis of leptospirosis is, therefore, essential. Molecular-based diagnostics provide a sensitive method for Leptospira spp. detection and an actionable diagnosis in the acute care setting. The sensitivity and specificity of the molecular-based approaches can be further improved by selecting a combination of target genes that are conserved and have multiple copies. Finally, the molecular-based technology should be translated to point-of-care settings such as lab-on-chip platforms, which offer integration and portability of the assay to facilitate clinical diagnostics in remote locations where the accessibility of laboratory equipment is limited.