Archives of Clinical Microbiology

Introduction

Carrion’s disease of human is caused by an infection with Bartonella bacilliformis, which is believed to be transmitted by Lutzomyia verrucarum sand flies [1,2]. Due to favorable environments for L. verrucarum, the disease has been confined to regions between 800 and 3000 m above sea level on the Western Andes Mountains in Peru, Columbia, and Ecuador. However, the disease has recently expanded to lower altitudes of the Andes and coastal regions due to increases in human migration and adaptations of sand flies to new geographic areas [3-5]. Also molecular detection of B. bacilliformis in L. maranonensis and ticks in Peru suggested the potential of spreading B. bacilliformis by these arthropods [6,7].

B. bacilliformis is known to penetrate erythrocytes and endothelial cells, where it can replicate and cause the biphasic Carrion’s disease [1]. During the acute phase, or Oroya fever, symptoms such as severe acute-onset febrile illness and hemolytic anemia are developed soon after infection. As much as 80% of erythrocytes can be lysed [8]. Untreated cases of Oroya fever have fatality rates as high as 88% due to the high susceptibility to fatal secondary infections [9]. The chronic phase of infection, known as “verruga peruana”, is associated with chronic skin eruptions which are self-limiting and can develop weeks or months after infection and persist for several years [5]. Moreover, the symptoms of Carrion’s disease can widely vary from person to person. An epidemiology study has revealed that about 20% of patients infected with the bacteria remain asymptomatic and can act as reservoirs for infection [10].

In addition to the wide spectrum of symptoms, the low sensitivity of diagnostic tests has hindered the proper treatment of patients with effective antibiotics, such as ciprofloxacin or doxycycline. Thin blood smears, often used for detection of acute bartonellosis, have very high specificity but are limited by low sensitivities [11]. Culturing of B. bacilliformis from clinical samples is time-consuming, and serological tests such as IFA or immunoblot cannot distinguish between different Bartonella species and require paired acute and convalescent samples for confirmation [12]. A highly sensitive test for early detection of B. bacilliformis in patient samples would improve proper treatment and a better knowledge of the true disease burden.

Polymerase Chain Reaction (PCR)-based approaches have been used for the diagnosis of infection caused by Bartonella spp. and Carrion’s disease [13-18]. PCR amplification of DNA shows promising sensitivity and specificity but requires a specialized equipment and extensive end-user training, it is impractical for diagnosing the disease in remote areas. It was able to diagnose patients in acute phase previously classified as negatives by thin blood smear [19]. Nonetheless, a critical issue is the detection limit of these techniques, raising doubts about its usefulness in the detection of low-bacteremia carriers. However, most of PCR-based assays to detect B. bacilliformis use single-copy genes. Amplification targeting gene segment of multiple copies in the genome have been used to improve the detection limit of DNA amplification method [20,21]. In this study, we identified a segment of 1,162 bp (MTSeq) which is present at three different locations of the B. bacilliformis genome as the amplification target. PCR primers were designed and evaluated for sensitivity and specificity. Our results demonstrated a higher sensitivity for bacterial DNA detection.

Materials and Methods

Design of primers

Oligonucleotide primers used for PCR and qPCR assays were designed based on the KC 583 strain of B. bacilliformis. The primers were synthesized by Eurofins MWG Operon (Huntsville, AL) and are described in Table 1.

Table 1 List of primer sequences for MTSeq cloning and qPCR.

aFor cloning MTSeq to pCR-XL-TOPO vector
bFor qPCR reaction

The Bb-F and Bb-R are forward and reverse primers for cloning the whole MTSeq, respectively. The 1,162 bp sequence is located at 3 different places in the B. bacilliforms genome between base pair 572720 to 573881, 683928 to 685089, and 710084 to 711245 (Figure 1). The qPCR primers for set A, B, and C were Bb-3031 and Bb-3196; Bb-3230 and Bb-3392; and Bb-3177 and Bb-3352, respectively.

Figure 1: The locations of a multiple copy DNA segment in B. bacilliformis genome. Three copies of the 1162 bp DNA segment (black boxes) are present in strain KC 583 (1445021 bp). Their locations are from base pair 572720 to 573881, 683928 to 685089, and 710084 to 711245. The complete DNA sequences of the segment were listed below with location of set B primers underlined.

Plasmid and genomic DNA template

The MTSeq of B. bacilliformis KC 583 was cloned into pCR-XLTOPO vector, and the closed circular plasmid (pCR-XL-TOPOMTSeq) was purified using standard Qiagen plasmid mini kit (Qiagen, stockach, Germany) following the manufacturer’s instruction. The pure pCR-XL-TOPO-MTSeq was quantified using a Nano-drop 2000 microsample spectrophotometer (Thermo Scientific, Wilmington, DE) and used as a standard for the selection of the best primer set to be used in the qPCR assay. The genomic DNA of B. bacilliformis KC 583 was used as the template in qPCR as described below. The genomic DNA from other bacteria (O. tsutsugamushi, R. typhi, R. conorii, R. rickettsi, C. burnetii, T. cruzi) transmitted by a variety of bloodsucking arthropod vectors capable of infecting humans were used to test the specificity of the qPCR assay. The copy number of each genomic DNA for different bacteria were determined similar to what was described above using a cloned plasmid containing a specific gene of each bacteria (Figure 2).

Figure 2: MTSeq PCR amplification plot.

Quantitative PCR

Conditions of quantitative PCR reactions were carried out as following. Each 20 μL reaction mixture included 0.5 μM of forward primer, 0.5 μM of reverse primer, 1×RT2 SYBR Green ROX qPCR Mastermix (Qiagen, Stockach, Germany), and DNA template. The qPCR was performed using the 7500 Fast Realtime PCR (Applied Biosystems, Foster City, CA). Enzyme activation at 95°C for 3 min was followed by 40 cycles of 15 sec at 95°C and 1 min at 60°C. Each sample was measured in duplicate and nuclease-free water was used as negative controls. Specificity of the product was determined by the melting curve (Figure 3).

Figure 3: Sandfly PCR amplification plot.

Detection of B. bacilliformis DNA targets in plasma samples

Normal human plasma was spiked with pCR-XL-TOPO-Bb- MTSeq for a final concentration of 25,000, 5,000, and 0 cp/mL. Total DNA in the spiked plasma samples was extracted using the QIAamp DNA Mini Kit (Qiagen, Stockach, Germany) according to the manufacturer’s manual. A total of 200 μL plasma was used for extraction and eluted in 25 μL. The DNA extractions were performed independently three times.

Detection of B. bacilliformis in experimentally infected sand flies

Previously collected L. longipalpis female sand flies fed with blood containing B. bacilliformis were used to extract DNA [22]. DNA extraction of the whole genome of B. bacilliformis was performed using the DNeasy Blood & Tissue Kit (Qiagen, Stockach, Germany) following the manufacturer’s protocol. Extracted DNA from six individual flies on Day 0, 1, 3, 5, 7, and 9 were used in duplicate for qPCR quantification. Ct values above 36 are considered negative.

Results

Selection of the best primer set

Ten-fold serial dilutions of plasmid DNA containing the MTSeq from 106 to 10 copies were used to evaluate the performance of each primer set. For all three primer sets, the limit of detection was 10 copies in the reaction. Table 2 shows slope, y-intercept, R2 value, and efficiency of the standard curve based on the average of two independent trials of ten-fold serial dilutions. Although these 3 primer sets performed similarly to detect 10 copies/reaction, their efficiency differed. Since high amplification efficiency will yield an increased PCR product and result in lower Ct values, primer set B with 99.96% efficiency was selected for the rest of this study.

Table 2 Comparison of qPCR results for 3 primer Setsa.

aThe qPCR was performed using the 7500 Fast Real-time PCR. Enzyme activation at 95°C for 3 min was followed by 40 cycles of 15 sec at 95°C and 1 min at 60°C.

Specificity of the qPCR

The specificity of the qPCR was evaluated by using the genomic DNA from other bacteria including Orientia tsutsugamushi (Karp, Kato, Gilliam, and TA763), Rickettsia species (R. typhi, R. conorii, and R. rickettsii), Coxiella burnetii, and Trypanosoma cruzi. About 106 copies of each genomic DNA was used as template and all tested negative with primer set B. The specificity of the primer set B was also analyzed in silico using the Primer-Blast program on the National Center for Biotechnology Information website. Another B. bacilliformis strain ATCC 35685D-5 had three identical segments. Two Bartonella species contained sequences very similar to the primer set B were also identified. B. ancashensis strain 20.00 had two segments with one mismatch to the forward primer sequences and two mismatches to the reverse primer sequences. B. sp. WD16.2 also had two DNA segments with two mismatches to the forward primer and two mismatches to the reverse primer.

Results for plasma samples spiked with MTSeq containing plasmid

Plasmid DNA (pCR-XL-TOPO-Bb-MTSeq) was spiked into normal human plasma to mimic patient samples, and qPCR was performed after DNA extraction using primer set B. The detection limit by qPCR was 79.5 copies of MTSeq (equals to 26.5 copies of organism) per reaction based on standard curves obtained from diluted plasmid (Table 3).

Table 3 Detection limit of Bb-MTSeq in normal human plasma spiked with plasmid pCR-XL-TOPO-Bb-MTSeq.

aAssuming 100% recovery for the extraction steps
bCopy numbers are based on standard curves obtained from diluted MTSeq -XL-TOPO-Bb-MTSeq plasmid and averages of 3 independent trials.

Results for sand flies fed with B. bacilliformis inoculated blood

L. longipalpis female sand flies fed on blood containing B. bacilliformis were subject to DNA extraction [19]. DNA from sandflies on Day 0, 1, 3, 5, 7, 9 was used in duplicate for qPCR. All 6 samples from Day 1 and 3 were positive for B. bacilliformis, while the 6 samples from day 0, 5, 7, and 9 were negative (Table 4).

Table 4 Comparison of qPCR results for sand flies fed on B. bacilliformis inoculated blood.

aDNA extraction of the whole genome of B. bacilliformis was performed using the DNeasy Blood & Tissue Kit following the manufacturer’s protocol. Extracted DNA from six individual flies on Day 0, 1, 3, 5, 7, and 9 were used in duplicate for qPCR quantification. Ct values above 36 are considered negative.
bqPCR results from Angkasekwinai et al. [19].

Sensitivity of qPCR targeting MTSeq as compared to that of pap31 for genomic DNA detection

Three independent qPCR experiments were performed. Each experiment had 48 reactions. 24 reactions had genomic DNA of 25 copies of MTSeq (equivalent to 8.3 copies of the bacteria genome) and the other 24 reactions had 10 copies of MTSeq (equivalent to 3.3 copies of the bacteria genome). Our results clearly demonstrated that targeting MTSeq was more sensitive than targeting Pap31 (Table 5). At 8.3 copies of bacteria genome per reaction, 18 and 8 reactions out of 36 showed positive results with primers targeting MTSeq and pap31, respectively. But at 3.3 copies per reaction, 12 and 2 reactions out of 36 were tested positive with MTseq and pap31 primers, respectively.

Table 5 Comparison of qPCR results for genomic DNA using primers targeting MTSeq and pap31.

agenomic DNA copy numbers
bThree independent qPCR runs were performed. Each run had total of 48 reactions. 24 qPCR reactions had 8.3 copies of genomic DNA, 12 were tested with MTseq primer set B and 12 were tested with pap31 primers. The other 24 qPCR reactions were performed with the present of 3.3 copies of genomic DNA

Discussions

Diagnosis of Carrion’s disease is currently limited by low sensitivity of thin blood smears, time-consuming cultures of B. bacilliformis, and unspecific serological tests [23]. PCR assays are the most sensitive techniques to diagnose B. bacilliformis. The purpose of this study was to develop a qPCR using a multiplecopy DNA segment to detect B. bacilliformis in clinical samples. The qPCR assay was found to detect 10 copies of plasmid containing the MTSeq equivalent of 3.3 organisms of B. bacilliformis per reaction. Angkasekwinai et al. [22]. reported that the limit of detection using a single-copy gene (pap31) for qPCR was 18 and 50 copies of bacteria per reaction in samples containing genomic DNA only and genomic DNA in presence of human DNA, respectively. Our assay was able to detect 3.3 copies of bacteria using genomic DNA only and 26.5 copies of bacteria (79.5 copies of MTSeq) using genomic DNA in presence of human DNA. Gomes et al. [24]. evaluated three different PCR approaches (16S rRNA, fla, and its genes) and found that targeting 16S rRNA performed the best with a similar detection limit. Therefore, the results in this study demonstrated a higher sensitivity by targeting this 1,162 bp multi-copy DNA sequence. Primer-Blast results showed potential non-specific DNA amplification with two other B. species. To further confirm the assay’s specificity against highly homologous Bartonella species, additional assays will be performed using genomic DNA of B. ancashensis strain 20.00 and B. sp. WD16.2.

Bacterial DNA extracted from harvested L. longipalpis female sand flies fed on blood containing B. bacilliformis was also tested by this newly developed qPCR. Although there was some variation between the six sand flies from each day, all 6 samples were determined to be positive on Days 1 (Ct values between 29.34 to 32.76) and 3 (Ct values between 33.78 to 37.82) and negative on Days 0, 5, 7, and 9. Previously, Angkasekwinai et al. [22]. used qPCR targeting the single-copy pap31 gene and found that only 86% (n = 15) or 63% (n = 11) of samples were determined positive for day 1 and 3, respectively. This study found that 100% (n = 6) of samples were positive for both day 1 and 3. The results demonstrated that this method can be used to detect the presence of B. bacilliformis in sand flies, making this qPCR a useful tool to monitor the migration of B. bacilliformis infected sand fly.

In conclusion, this qPCR method using the multiple-copy DNA sequence has higher sensitivity as compared to previous assays. This will enhance our ability to diagnose an individual for B. bacilliformis infection and allow rapid and appropriate treatment of acute Carrion’s disease. In the future, we would like to conduct qPCR assay on patient samples and vector samples to verify its utility. We will also develop loop-mediated isothermal amplification assay targeting this multi-copy DNA

© Under License of Creative Commons Attribution 3.0 License sequence for a robust and easy-to-perform assay suitable to be used in the endemic regions for asymptomatic surveillance studies.

Acknowledgements

This research was supported by Naval Medical Research Center, research work unit 6000.RAD1.J.A0310. The views expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense, or the U.S. Government. Chien-Chung Chao and Wei-Mei Ching are employees of the U.S. Government. This work was prepared as part of their official duties. Title 17 U.S.C. §105 provides that ‘Copyright protection under this title is not available for any work of the United States Government.’ Title 17 U.S.C. §101 defines a U.S. Government work as a work prepared by military service member or employee of the U.S. Government as part of that person’s official duties.

References

1. Maguina C, Gotuzzo E (2000) Bartonellosis-New and old. Infect DisnClin North Am 14: 1-22.
2. Gomes C, Ruiz J (2017) Carrion’s Diseases: the Sound of Silence.nClin Microbiol Rev 31: e00056-17.
3. Kosek M, Lavarello R, Gilman RH, Delgado J, Maguina C, et al.n(2000) Natural history of infection with Bartonella bacilliformis inna nonendemic population. J Infect Dis 182: 865-872.
4. Huarcaya E, Maguina C, Torres R, Rupay J, Fuentes L (2004)nBartonelosis (Carrion’s Disease) in the pediatric population ofnPeru: an overview and update. Braz J Infect Dis 8: 331-339.
5. Lydy SL, Eremeeva ME, Asnis D, Paddock CD, Nicholson WL, et al.n(2008) Isolation and characterization of Bartonella bacilliformisnfrom an expatriate Ecuadorian. J Clin Microbiol 46: 627-637.
6. Ulloa GM, Vasquez-Achaya F, Gomes C, Del Valle LT, Ruiz J, et al.n(2018) Molecular detection of Bartonella bacilliformis innLutzomyia maranonensis in Cajamarca, Peru: A new potentialnvector of Carrion’s disease in Peru? Am J Trop Med Hyg 99:n1229-1233.
7. Del Valle-Mendoza J, Rojas-Jaimes J, Vasquez-Achaya F, Aguilar-nLuis MA, Correa-Nunez G, et al. (2018) Molecular identification ofnBartonella bacilliformis in ticks collected from two species of wildnmammals in Madre de Dios: Peru. BMC Res Notes 11: 405.
8. Coleman SA, Minnick MF (2001) Establishing a direct role for thenBartonella bacilliformis Invasion-associated locus B (IalB) proteinnin human erythrocyte parasitism. Infect Immun 69: 4373-4381.
9. Minnick MF, Anderson BE, Lima A, Battisti JM, Lawyer PG, et al.n(2014) Oroya fever and verruga peruana: Bartonelloses unique tonSouth America. PLoS Negl Trop Dis 8: e2919.
10. Chamberlin J, Laughlin LW, Romero S, Solorzano N, Gordon S, et al.n(2002) Epidemiology of endemic Bartonella bacilliformis: anprospective cohort study in a Peruvian mountain valleyncommunity. J Infect Dis 186: 983-990.
11. Smit PW, Peeling RW, Garcia PJ, Torres LL, Perez-Lu JE, et al. (2013)nDried blood spots for qPCR diagnosis of acute Bartonella bacilliformis infection. Am J Trop Med Hyg 89: 988-990.
12. Chamberlin J, Laughlin L, Gordon S, Romero S, Solorzano N, et al.n(2000) Serodiagnosis of Bartonella bacilliformis infection by indirect fluorescence antibody assay: test development andnapplication to a population in an area of bartonellosis endemicity.nJ Clin Microbiol 38: 4269-4271.
13. Johnson G, Ayers M, McClure SC, Richardson SE, Tellier R (2003)nDetection and identification of Bartonella species pathogenic fornhumans by PCR amplification targeting the riboflavin synthasengene (ribC). J Clin Microbiol 41: 1069-1072.
14. Ciervo A, Ciceroni L (2004) Rapid detection and differentiation ofnBartonella spp. by a single-run real-time PCR. Mol Cell Probes 18:n307-312.
15. Chmielewski T, Brydak-Godowska J, Fiecek B, Rorot U, Sedrowicz E,net al. (2014) Bacterial tick-borne diseases caused by Bartonellanspp., Borrelia burgdorferi sensu lato, Coxiella burnetii, andnRickettsia spp. among patients with cataract surgery. Med ScinMonit 20: 927-931.
16. Korhonen EM, Perez Vera C, Pulliainen AT, Sironen T, Aaltonen K,net al. (2015) Molecular detection of Bartonella spp. in deer kednpupae, adult keds and moose blood in Finland. Epidemiol Infectn143: 578-585.
17. Mazzitelli M, Lamberti AG, Quirino A, Marascio N, Barreca GS, etnal. (2018) Utility of molecular identification and quantitation ofnBartonella species with species-specific real-time PCR fornmonitoring treatment response: A case series. Open Microbiol Jn31: 148-153.
18. Ricapa-Antay F, Diaz-Melon K, Silva-Caso W, Del Valle LJ, Aguilar-nLuis MA, et al. (2018) Molecular detection and clinicalncharacteristics of Bartonella bacilliformis, Leptospira spp., andnRickettsia spp. in the Southeastern Peruvian Amazon basin. BMCnInfect Dis 18: 618.
19. del Valle Mendoza J, Silva Caso W, Tinco Valdez C, Pons MJ, delnValle LJ, et al. (2014) Diagnosis of Carrion’s disease by direct bloodnPCR in thin blood smear negative samples. PLoS One 9: e92283.
20. Reed JL, Walker ZJ, Basu D, Allen V, Nicol MP, et al. (2016) Highlynsensitive sequence specific qPCR detection of Mycobacteriumntuberculosis complex in respiratory specimens. Tuberculosis 101:n114-124.
21. Chen HW, Ching WM (2014) Development of loop-mediatednisothermal amplification assays for rapid and easy detection ofnCoxiella Burnetti. J Microbiol Methods 107: 176-181.
22. Angkasekwinai N, Atkins EH, Johnson RN, Grieco JP, Ching WM, etnal. (2014). Rapid and sensitive detection of Bartonella bacilliformisnin experimentally infected sand flies by loop-mediated isothermalnamplification (LAMP) of the pap31 gene. PLoS Negl Trop Dis 8:ne3342.
23. Sanchez Clemente N, Ugarte-Gil CA, Solorzano N, Maguina C,nPachas P, et al. (2012) Bartonella bacilliformis: a systematic reviewnof the literature to guide the research agenda for elimination.nPLoS Negl Trop Dis 6: e1819.
24. Gomes C, Martinez-Puchol S, Pons MJ, Bazan J, Tinco C, et al.n(2016) Evaluation of PCR approaches for detection of Bartonella bacilliformis in blood samples. PLoS Negl Trop Dis 10: e4529.