Development of A Loop-Mediated Isothermal Amplification (LAMP) Assay for Detection of Relapsing Fever Borreliae

Background: This study aimed to develop a loop-mediated isothermal amplification (LAMP) assay for the rapid detection of tick-borne relapsing fever in resource-limited areas. Methods: A set of six primers were designed based on the conserved regions of the Glycerophosphodiester phosphodiesterase (glpQ) gene of Borrelia species. For sensitivity assay, serial dilutions of a recombinant plasmid containing a 219bp sequence of the glpQ were prepared and used as the template DNA. The LAMP reactions containing the six primers and the reagents required for amplification were incubated at 60–65 °C for 60min in a Loopamp real-time turbidimeter. For the specificity test, DNA from 14 other bacteria were included in the assays, and double-distilled water was used as the negative control. Also, DNA from dried blood spots (DBSs) of spirochetemic mice, and blood samples from relapsing fever-suspected patients were examined by the LAMP along a Borrelia-specific nested PCR that targets the rrs-rrl-IGS region. Results: The LAMP detected as low as 90glpQ copies in reactions. The primers reacted with DNA from DBS of spirochetemic mice showing spirochete concentrations of ≤ one per a 1000X microscopic field. In clinical samples, the LAMP assay showed a higher sensitivity compared to nested-PCR. The LAMP specificity was 100%, as the primers did not react with other bacteria DNA. Conclusion: The high sensitivity and specificity of the test, along with the simplicity of the DNA extraction procedure, make the LAMP a reliable and adaptable tool for the diagnosis of tick-borne relapsing fever in rural endemic areas.


Introduction
The genus Borrelia comprises two distinct groups of spirochetes with the difference in diseases they cause. One group includes the causative agents of Lyme disease and the other the relapsing fever borreliae (RFB). Currently, there are 22 confirmed RFB, and six other taxa have been proposed (1). Except for the louse adapted Borrelia recurrentis, the majority of species pathogenic to humans are transmitted by the soft ticks of the genus Ornithodoros. A few species, such as Borrelia miyamotoi and Borrelia lonestari, are vectored by hard ticks and yet share genetic similarities with the RFB (2). Soft tick-borne relapsing fever (STBR) is endemic to Iran (3)(4)(5). Until now, despite the improvement of housing and the removal of the disease from mandatory reporting to the Ministry of Health and Medical Education (MHME), no year has passed without reports of human infections. In Iran, four RFB, including Borrelia persica, Borrelia microti, Borrelia latyschewii and Borrelia baltazardi have been described (5). Borrelia persica is the primary cause of the disease especially in the west and northwest of the country (5)(6)(7), while in the south, epidemiological data and molecular approved human infections indicated B. microti and other B. microti-like borreliae as the other cause of relapsing fever (3,8,9).
In Iran, until recently, confirmation of relapsing fever merely relied on observation of the spirochetes in peripheral blood of febrile patients using darkfield microscopy or Giemsa-stained blood smears. The disease is easily diagnosed by microscopy during fever peaks with a massive spirochetemia. However, between the peaks and in milder diseases, the bacteria are scanty and are hard to identify in blood smears. In Iran, PCR assays using various molecular markers like flaB, glpQ and rrs have successfully detected the Borrelia spp in relapsing fever patients (3,8) and animals (10). The PCR assays exhibit high sensitivities but are not commonly affordable in resource-limited laboratories of rural areas, where most of the relapsing fever infections occur. Hence, we prompted to develop an alternative DNA amplification assay of lower cost for the detection of the disease in these areas. Loop-mediated isothermal amplification (LAMP) has proved as a robust, cheap, highly sensitive/ specific tool for the detection of various pathogenic agents including viruses, bacteria, fungi, and parasites (11)(12)(13)(14). This assay employs a DNA polymerase and a set of 4-6 primers that operate in isothermal conditions forming loop structures that ultimately precipitate in the reaction mixture (13). This assay has also shown to be less prone to inhibition from DNA preparations and allows adaptability to field conditions (15,16).
This study aimed to develop a (LAMP) assay for the rapid detection of relapsing fever borreliae based on the glycerophosphodiester phosphodiesterase (glpQ) gene, a sequence conserved among all relapsing fever borreliae, but absent from Lyme disease spirochetes (17).

Bacteria species and DNA extraction
We used B. microti strain IR-1, which was maintained via continual passages in NMRI mice for more than 15 years in the Parasitology Department of Pasteur Institute of Iran. Blood samples were obtained from B. microtiinfected mice when spirochetes reached 1.4× 10 6 /ml of blood. DNA extraction from 1ml of blood was performed using a Genomic DNA Purification Kit (Promega, Madison, USA) as described by the manufacturer.

Primer design
Initially . A set of six primers including two loop primers were designed based on the conserved regions of the glpQ sequence, corresponding to the nucleotides 261015-261682 of B. duttonii strain Ly (Fig. 1), by the online software program, Primer Explorer V4 (Eiken Chemical Co., Tokyo, Japan; http://primerexplorer.jp/e/). The theoretical specificity of the designed primers was confirmed by in silico analysis using BLAST and Primer-BLAST software available in NCBI (http://www. ncbi.nlm.nih.gov/). The primers were synthesized by a commercial company (Generay Biotechnology, Shanghai, China). The main primers (glpQ-F3 and glpQ-B3) and (glpQ-FIP and glpQ-BIP) target fragments of 219bp and 161bp size, respectively, and the loop primer (glpQ-LF and glpQ-LB) produce 49 http://jad.tums.ac.ir Published Online: March 31, 2020 amplicons of various size with a ladder-like pattern (Table 1).

Sensitivity assay
A serial 10-fold dilution of the recombinant plasmid ranging from 9×10 8 to 9×10 -1 copy numbers per microliter, equivalent to ≈32ng to ≈32×10 4 fg/µl of DNA was prepared and used in assays.

Specificity assay
For and Leptospira interrogans in all assays. All the bacteria, including L. interrogans, the closely related bacteria to Borrelia species contained glpQ sequence.

Loop Mediated Isothermal Amplification (LAMP) assay
The LAMP reactions contained 40pM of the inner (glpQ-FIP and glpQ-BIP) and 10pM of outer (glpQ-F3 and glpQ-B3) and loop primers (glpQ-LF and glpQ-LB) ( obasic, Toronto, Canada), 0.1% Tween 20 (Acros Organics, Vernon, USA), 8U of Bst DNA polymerase, large fragment (Biolabs, New England, UK), 1μl of serial dilutions of the recombinant plasmid, and DDW to the 25 μl final volume. The reactions were incubated at temperatures ranging from 60-65 °C for 60min in a Loopamp real-time turbidimeter (LA-320C; Teramecs, Kyoto, Japan), followed by heating at 80 °C for 5min for enzyme inactivation. In all assays, DNAs from 14 other bacteria were included and DDW was used as a negative control.
During the optimization of the assay, some amplifications were performed without the loop primers. Also, in some reactions, we added 1μl of Fluorescent detection reagent containing Calcein (Eiken Chemical co., Tokyo, Japan), an indicator of DNA amplification.

Detection of LAMP products
The amplification in LAMP reactions was examined by 1) naked eye observation of white turbidity resulting from the accumulation of magnesium pyrophosphate, a by-product of the reactions, 2) a Loopamp real-time turbidimeter that records the optical density of reactions every 6sec at 650nm (the reactions were considered positive when the turbidity reached ≥ 0.1 within 60min), 3) the color change from orange to green, as an indication of DNA amplification, and 4) gel electrophoresis of amplicons on 2% agarose gels.
A LAMP product resulting from the 9×10 8 dilution was purified using PCR Purification Kit (Bioneer, Daejeon, Korea) and sequenced in both directions by (ABI 3730xl/Bioneer 3730xl, Daejeon, Republic of Korea).

Preparation of DBSs from Borrelia-infected mice
Amounts of 200µl blood from B. microtiinfected mice with various degrees of the spirochetemia (225±61.34, 71±36.71, 3.3±1.63, 1.1±1.44, and 0.8±0.78 spirochetes per microscopic field) were dotted on 30 DNA banking cards (DBC) (Kawsar Biotech Co, Tehran, Iran). Blood from non-infected mice was used as controls on DBCs. The blood spots were allowed to dry at room temperature, and the DBCs were kept in the same condition until used.

Loop Mediated Isothermal Amplification (LAMP) and PCR with DBS
Circles of 2mm from DBSs were cut and washed three times with DNA extraction buffer (provided by the manufacturer), followed by DDW. The circles were allowed to dry at room temperature and then used in LAMP and PCR assays as described above except that instead of template DNA, DBS circles were included in the reactions.
The sensitivity and specificity of the LAMP assay were calculated in comparison with microscopy as the gold standard assay using Med-Calc (2018 MedCalc Software bvba) software available online (https://www.medcalc.org/calc/diagnostic_tes t.php).

Loop Mediated Isothermal Amplification (LAMP) and nested-PCR amplification of clinical specimens
DNA was extracted from 39 sera of febrile patients residing in the relapsing fever endemic areas in the south and west of Iran using a commercial DNA extraction kit as described above. The DNA samples were examined by the LAMP, and a Borrelia-specific nested PCR that amplifies the rrs-rrl-IGS region using the outer primers F, 5´-GTATG TTTAGTGAGGGGGGTG-3´ and R, 5´-GG ATCATAGCTCAGGTGGTTAG-3´ and inner nested primers F, 5´-AGGGGGGTGAAGTC GTAACAAG-3´ and R, 5´-GTCTGATAAACC TGAGGTCGGA-3´ (18). This PCR has exhibited high sensitivity in detecting relapsing fever borreliae (3,8,19,20). In all assays, for specificity test, DNAs from other bacteria species were included and DDW was used as a negative control. The LAMP reactions were checked for DNA amplification as described above, and the 51 http://jad.tums.ac.ir Published Online: March 31, 2020 PCR products were resolved on 2% agarose gels, visualized under UV and photographed.

Glycerophosphodiester Phosphodiesterase Gene (glpQ ) cloning
Amplification of the glpQ sequence from the recombinant plasmid (pTZ57R/T-glpQ) with the primers glpQ-F3 and glpQ-B3yielded the expected 219bp indicating the insertion of this sequence in the plasmid.

Loop Mediated Isothermal Amplification (LAMP) assay with recombinant pTZ57R/ T-glpQ
We observed amplification of glpQ sequence in the LAMP reactions by naked eye observation of turbidity and the Loopamp real-time turbidimeter down to 9×10 3 copies of recombinant plasmid equivalent to 0.32fg DNA ( Fig.  2A). However, gel electrophoresis revealed amplifications in two lower dilutions of 9×10 2 and 9×10 1 , equivalent to 3.2fg and 0.32fg DNA, respectively (Fig. 2B). Our LAMP assay was 100% specific as the primers reacted with none of the other 14 bacterial DNA, including L. introgans. In the reactions containing calcein and the Borrelia DNA, the color turned from orange to green.
The 137bp sequence resulted following the sequencing of the LAMP amplicon matched with the glpQ sequence of B. microti strain IR-1 (acc. No. JF825473) corresponding to nucleotides 261158-261295 of the whole genome sequence of B. duttonii Ly (acc. No. CP000796). The results were the same within the temperature range of 60-65 °C. In the reactions, in abscence of loop primers, the optimum time for isothermal amplification was 60min, whereas, in those with the loop primers, the incubation time reduced to 45min.

Loop Mediated Isothermal Amplification (LAMP) and PCR with DBSs
Of the 30 B. microti-positive DBS, all (100%) showed turbidity with the naked eye indicating amplification, while the glpQ-PCR (with the primers glpQ-F3 and glpQ-B3) yielded the expected 219bp band in 20 positive DBSs (66.67 %). The negative PCR reactions belonged to the mice with the lowest level of spirochetemia [1.1±1.44 (n= 4) and 0.8±0.78 (n= 6) spirochete per microscopic field]. Neither LAMP nor PCR amplification was observed with the negative controls (DDW) or reactions containing other bacteria DNA as the template.
By considering microscopy as the gold standard, the sensitivity of glpQ-LAMP and glpQ-PCR were 100% and 66.67%, respectively. The specificity was 100% for both assays.

Loop Mediated Isothermal Amplification (LAMP) and PCR with clinical samples
Of 39 clinical samples, 11 became turbid in the LAMP assay indicating amplification of glpQ gene, whereas PCR amplification of IGS sequence only yielded the expected 540bp band in three specimens (Fig. 3).

Discussion
In Iran, blood analysis of relapsing fever patients by examination of wet smears using dark-field microscopy or Giemsa-stained blood slides has been a common practice for years. This approach was efficient in endemic areas, where the disease commonly appeared in clus-53 http://jad.tums.ac.ir Published Online: March 31, 2020 ters (5), and overlooking the spirochetes in the blood of some febrile patients did not question the identity of the causative agent. However, with the decline of the disease in the endemic regions and reports of sporadic cases from other areas application of more sensitive approaches for the identification RFB became necessary. In Iran, over the past decade, PCR assays by based on various molecular markers, such as glpQ, rrs, and flaB, were developed for the identification of Borrelia infection in Ornithodoros ticks (21,22) or characterization of the tick-originated relapsing fever borreliae (9, 23), but rarely clinical samples were included. Lately, qPCR and conventional PCRs identified a B. microti-like strain, presumably, an ecotype of African B. duttonii, in relapsing fever patients from southern Iran (3,8).
In our previous work, using B. persica-spiked blood samples, we consistently observed bacteria by microscopy in the blood samples with densities ≥ 800-1000 spirochetes/µl (24). Herein, with the DBSs prepared with the spirochetemic murine blood, the LAMP could detect borrelial DNA in blood specimens showing less than one spirochetes per 1000X microscopic field. Our LAMP assay also showed a higher sensitivity in comparison with a nested-PCR amplification of the IGS in the diagnosis of clinical samples. The specificity of the LAMP assay for the diagnosis of B. microti DNA was 100% as the designed primers exhibited no cross-reaction with DNA from the other 14 bacteria used in this study. Our LAMP method requires to be further tested with blood samples from RF patients infected with other Borrelia species to ensure the sensitivity and specificity of the assay.
LAMP method has also shown promise for the detection of borreliae DNA and other pathogens in the tick vectors. In China, the LAMP could detect B. burgdorferi s. l. in ticks, with a higher sensitivity than a conventional PCR (25). Also, in recognizing spotted fever group rickettsia, the LAMP appeared ten times more sensitive than an end-point PCR targeting the same gene fragment (26).

Conclusion
Such a reliable sensitivity and specificity, as well as the simplicity of the DNA extraction procedure, make the LAMP a suitable and adaptable assay for field studies in relapsing fever endemic areas. In resource-limited rural health centers, LAMP can readily check fieldcollected blood samples, preferably on DBS, for relapsing fever borreliae with the least equipment accessible, i.e., a heat block. Having access to a Loopamp real-time turbidimeter is preferable. However, LAMP is a qualitative assay, and the result can also be monitored by color change with calcein in the fluorescent detection reagent or by the turbidity of magnesium pyrophosphate in the reaction tubes. Hence, a heating system that provides a temperature within the range of 60-65 °C would suffice.