Sensitivity of Real-Time PCR Performed on Dried Sera Spotted on Filter Paper for Diagnosis of Leptospirosis

Abstract

In remote countries, leptospirosis confirmation is difficult because it requires the shipment of frozen samples to reference laboratories. The sensitivity of leptospirosis real-time PCR performed on filter paper-dried serum samples stored at ambient temperature was evaluated at 2 × 10² equivalent leptospires/ml (eqLeptospires/ml). This easy alternative procedure can be used to enhance the surveillance of leptospirosis.

TEXT

Leptospirosis is a worldwide zoonosis due to pathogenic Leptospira species (1). This severe but neglected disease is an emerging public health concern with a higher incidence in tropical areas (2). The epidemiology and burden of the disease in most countries where leptospirosis is endemic are poorly understood due to difficulties in case confirmation. Leptospirosis is caused by pathogenic bacteria called leptospires that are transmitted directly or indirectly from animals to humans. Leptospirosis diagnosis is challenging (3) because the clinical spectrum of the disease is large and unspecific, ranging from subclinical to severe forms sometimes leading to death. Moreover, the culture of leptospires is long and tedious, serology is neither sensitive nor specific during the first week of illness when most patients present to health facilities (4), and paired serum samples are required for serological diagnosis confirmation. PCR and real-time PCR assays have been developed for acute-phase leptospirosis confirmation (5) but are rarely available in remote and resource-poor countries. Case confirmation requires fresh frozen shipment of samples to reference laboratories, but this is expensive, subject to restrictive regulations, and usually impossible from remote areas. Filter paper (FP)-dried serum samples are not subjected to dangerous goods regulation (6) and represent a cheap and convenient tool for sample transportation (7).

FP-dried serum samples have been used for the detection of a wide range of pathogens, including viruses (8), bacteria (9), and protozoa (10). In this study we evaluated the sensitivity of leptospirosis PCR performed on FP-dried sera.

Genomic DNA (gDNA) of five Leptospira reference strains (Leptospira interrogans, serovar Icterohaemorrhagiae, strain Copenhageni Fiocruz L1-130; L. interrogans, serovar Autumnalis, strain Ballico; L. interrogans, serovar Canicola, strain Hond Utrecht; Leptospira kirschneri, serovar Cynopteri, strain 3522 C; and L. interrogans, serovar Muenchen, strain München C 90) were provided by the URMITE laboratory (Marseille, France) and quantified using a Qubit double-stranded DNA (dsDNA) HS assay kit with a Qubit 1.0 fluorometer. The number of gDNA copies/ml and the number of equivalent leptospires/ml (eqLeptospires/ml) were deducted from the DNA concentration based on a genomic size of 4.6 Mb and considering the fact that leptospires contain 5 gDNA copies of the genome (11). Each gDNA was spiked into healthy human serum samples that had previously tested negative for leptospirosis by PCR, and 10-fold dilutions were performed to obtain final concentrations ranging from 10¹ to 10⁵ gDNA copies/ml or 2 to 2 × 10⁴ eqLeptospires/ml. For each dilution, 2 FPs (LDA²², Ploufragan, France) were spotted with 180 μl (12 precut circles with 15 μl of serum) of spiked serum and stored at room temperature. The first FP was processed 1 day after storage and the second FP 30 days after storage. Unspiked serum and nonspotted FPs were processed as negative controls.

Twenty-eight fresh serum samples collected from patients previously tested positive and 5 from patients tested negative for leptospirosis by the PCR described below were also spotted on 2 FP and stored for 30 days at room temperature. All positive samples were obtained from patients with mild leptospirosis. The study was approved by the Ethics Committee of French Polynesia under reference 61/CEPF.

DNA was extracted from FP as previously reported using the Easymag extraction system (bioMérieux) (12). Amplification was performed in duplicate in 2 independent experiments using a TaqMan PCR targeting the lipL32 gene on a CFX96 Bio-Rad thermocycler with primers and probes previously described (13). PCR was considered positive for a cycle threshold (C_T) value of <40 for the two experiments.

Results for PCR detection of leptospires on FP-dried reference strains and FP-dried sera are reported in Tables 1 and 2.

TABLE 1.

PCR results of 10-fold dilutions of different strains of leptospires

Strain and no. of gDNA (copies/ml)a	No. of eqLeptospires/mlb	Mean CTc
		Fresh sera	FP day 1	FP day 30
L. interrogans, serovar Icterohaemorrhagiae, strain Copenhageni Fiocruz L1–130
105	2 × 104	31.4	32.1	31.3
104	2 × 103	34.5	35.1	34
103	2 × 102	37.1	38.5	38.6
102	2 × 101	NA	NA	NA
101	2 × 100	NA	NA	NA
L. interrogans, serovar Autumnalis, strain Ballico
105	2 × 104	29.4	32.8	31.7
104	2 × 103	35	35.7	34.8
103	2 × 102	38.1	38.5	37.7
102	2 × 101	NA	37.6	NA
101	2 × 100	NA	NA	NA
L. interrogans, serovar Canicola, strain Hond Utrecht
105	2 × 104	30.9	32	33.5
104	2 × 103	34.5	36	36.5
103	2 × 102	38.2	36.7	38.1
102	2 × 101	NA	NA	NA
101	2 × 100	NA	NA	NA
L. kirschneri, serovar Cynopteri, strain 3522 C
105	2 × 104	31.3	33.2	35
104	2 × 103	36.1	36	38.1
103	2 × 102	39.1	35.3	39.2
102	2 × 101	NA	37.1	NA
101	2 × 100	NA	NA	NA
L. interrogans, serovar Muenchen, strain München C 90
105	2 × 104	26.3	27.3	32.7
104	2 × 103	29.8	30.8	36.7
103	2 × 102	33.5	33.5	36.8
102	2 × 101	36.5	38.2	NA
101	2 × 100	NA	NA	NA

^agDNA, genomic DNA based on a genomic size of 4.6 Mb.

^bBased on 5 gDNA copies/leptospire.

^cMean cycle threshold (C_T) of one experiment done in duplicate. FP, filter paper; NA, no amplification.

TABLE 2.

PCR results of 28 positive clinical serum samples spotted on filter paper

Patient no.	CT for fresh seraa	Mean CT forb:
		FP on day 1	FP on day 30
1	30.6	33.5	32.4
2	31.9	34.6	34.3
3	27.7	31.3	29.9
4	30.3	32.9	32.3
5	27.2	29.7	28.9
6	34.4	35.4	35.7
7	31.4	37.8	34.7
8	34.7	37.6	36.9
9	34.7	33.9	35.4
10	35.4	38.0	NA
11	31.7	39.6	37.0
12	29.4	36.0	33.9
13	37.8	37.1	NA
14	35.3	36.7	36.6
15	39.2	37.4	33.5
16	33.6	36.6	34.1
17	28.3	32.7	32.4
18	30.8	37.3	34.9
19	33.5	36.9	36.5
20	35.7	39.6	35.9
21	37.8	NA	NA
22	30.3	34.1	35.5
23	38.9	NA	NA
24	35.6	NA	36.5
25	37.1	NA	38.7
26	37.9	NA	NA
27	33.2	34.8	35.5
28	38.9	NA	NA

^aCycle threshold (C_T) values of <35 are in bold type.

^bMean C_T values of two independent experiments done in duplicate. FP, filter paper; NA, no amplification.

The sensitivity of the PCR on fresh serum was 20 eqLeptospires/ml (10² gDNA copies/ml) for L. interrogans serovar Muenchen and for L. interrogans serovar Autumnalis, L. kirschneri serovar Cynopteri, and L. interrogans serovar Muenchen on FP-dried serum 1 day after storage. Sensitivity of the assay was 2 × 10² eqLeptospires/ml (10³ gDNA copies/ml) for the other strains on both fresh sera and on FP-dried sera after 1 and 30 days of storage.

From known positive patients, the sensitivity of the PCR was 78.6% (22/28) for the results of FP-dried sera stored for 30 days compared to those for fresh sera. A sensitivity of 100% (17/17) on FP stored for 30 days was observed for all samples with a C_T of up to 35 for fresh sera (Table 2).

All positive controls were positive, and all negative controls were negative.

The main challenge in the diagnosis of leptospirosis by PCR is the low bacterial load responsible for false-negative results. PCR for leptospirosis demonstrated a bacterial load ranging from 10² to 10⁶ leptospires/ml (14). According to Bourhy et al., the sensitivity of 4 specific Leptospira PCR performed after spiking of leptospire reference strains ranged from 10² to 10³ leptospires/ml (11). Our spiking experiment demonstrated comparable results, with an average sensitivity of around 2 × 10² eqLeptospires/ml on FP-dried sera stored for 30 days at room temperature. The sensitivity of the assay for the patients' FP-dried sera was 78.6% compared to that for fresh sera. The lower sensitivity observed with FP-dried sera compared to that of fresh sera might be due to DNA damage or poor DNA recovery during extraction. Moreover, all patients tested in this study presented mild disease, often associated with a bacterial load of <10² leptospires/ml. Thus, this method remains valuable for main cases of leptospirosis and particularly for severe forms that exhibit a critical threshold of 10⁴ leptospires/ml (15, 16).

To overcome the problem of sample transportation to reference laboratories, leptospirosis serology by microagglutination testing (MAT) of sera spotted on FP has been evaluated (17). Leptospirosis MAT of sera spotted on FP had some limitations, because a decrease of sensitivity for storage up to 7 days has been reported. In addition, the regionally circulating serovars are often unknown in remote areas, which limits the use of MAT in such situations.

We previously demonstrated the advantage of using FP-dried samples for dengue surveillance in the Pacific, and we diagnosed cases of leptospirosis from samples collected from Micronesian patients with suspected dengue during a dengue outbreak (18).

The sensitivity of this protocol is suboptimal for mild leptospirosis case confirmation, but this easy and cheap procedure is a potentially powerful tool for enhancing the diagnosis and surveillance of leptospirosis in all remote and resource-poor regions.