Li, HaoQi

Development and Optimization of a Sensitive and Specific Quantitative PCR Assay for Borrelia lonestari

HaoQi (Esther) Li

Naval MedicalResearchCenter

Infectious Diseases Department

Rickettsial Diseases Department

Mentors: Dr. Allen Richards, Mrs. Ju Jiang

Teacher Advisors: Dr. Barbara Wood, Mr. Fred Lampazzi
Abstract

Borrelia lonestari is a spiral-shaped bacterium recently discovered in the lone star tick, Amblyomma americanum, located throughout the southeastern United States. This spirochete is suspected of inducing symptoms in humans commonly associated with Lyme disease, such as rash, fever, and fatigue. Due to these common symptoms, the diagnosis of the B. lonestari infection becomes very challenging. Previous methods to detect B. lonestari included polymerase chain reaction (PCR) and restriction fragment length polymorphism (RFLP) analysis. However advances in biotechnology have introduced quantitative real-time PCR (qPCR) as a more accurate and efficient detection procedure. This study,therefore,reports the development of a qPCR assay that is highly sensitive and specific for detecting B. lonestari.

Using the programs ClustalW and GeneDoc, a unique region between 594–719bp in the B. lonestari flagellin gene was identified,allowing primers and a molecular Beacon probe to be designed. A plasmid containing the target B. lonestari flagellin gene sequence was constructed with a TOPO XL PCR Cloning Kit.

After calculating the copies of the cloned plasmid, a serial dilution (1010-100 copies/uL) was made for a standard curve to quantitatively demonstrate the sensitivity of the assay. By using various concentrations of the primers, the probe, MgCl2, and by changing the annealing temperature, an optimal condition was established. The limit of detection of the assay was determined to be one copy/uL. Seven related Borreliaspecies and twenty-three non-related bacterial genomic DNA were used to verify the specificity. The assay only responded positively to B. lonestari,thus demonstrating that the assay of the study was indeed specific. Therefore, the newly developed qPCR assay was demonstratedto be a sensitive and specific tool for detecting B. lonestari.

Introduction

Bacteria Borrelia lonestari belong to the same genus as Borrelia burgdorferi, the causative agent of the potentially fatal Lyme disease. B. lonestariinduces Southern Tick-Associated Rash Illness (STARI), resembling symptoms of Lyme disease that are similar to those of many common illnesses (Moore, Varela, Yabsley, Davidson, & Little, 2003). Found throughout the southern and southeastern United States, STARI includes rash, fever, and migraines (Burkot, Mullen, Anderson, Schneider, Happ, & Zeidner, 2001). The commonplace symptoms of STARI in nature, along with the absence of a current method of diagnosis, often make diagnosing the STARI disease extremely difficult. With the advancement of new technology, this study successfully developed a sensitive and specific molecular assay for B. lonestari.

Previous research has established that B. lonestari is vectored by the Lone Star tick Amblyomma americanum. Burkot, Mullen, Anderson, Schneider, Happ, & Zeidner (2001) conducted research using traditional polymerase chain reaction (PCR) to detect B. lonestari bacteria in A. americanum and A. maculatum ticks from Alabama. With PCR for the flagellin and 16s rRNA, two positive B. lonetsari samples were both found in A. americanum ticks.

Not only does B. lonestari infect humans and ticks, they are also found in other mammals such as the white-tailed deer (Moore, Varela, Yabsley, Davidson, & Little, 2003). While studying the presence of B. lonetsari in white-tailed deer as “the first evidence of” the bacteria in a vertebrate other than humans, Moore, Varela, Yabsley, Davidson, & Little (2003) developed a B. lonestari PCR assay to test the samples. However in this study, a more efficient and accurate assay was developed, utilizing the newest technology – quantitative realtime Polymerase Chain Reaction (qPCR).

In order to study the flagellin gene sequences of various Borrelia species, Fukunaga, Okada, Nakao, Konishi, & Sato (1996) performed phylogenetic analyses to map out the Borrelia taxonomy and to distinguish the Lyme disease agents from related speciesthrough PCR-Restriction Fragment Length Polymorphism (PCR-RFLP) with 580 bp fragments. The researchers concluded that B. lonestari bacteria are also related to the New World tick-borne relapsing fever Borreliae. The difference in gene sequences allows an assay to be produced specifically to a certain Borrelia species;and in this experiment, an assay specific to B. lonestari was located.

In this study, qPCR serves the crucial purpose of data collection by detecting the amount of B. lonetsari DNA in samples through the fluorescence emitted by the newly designed assay. This method is based on the direct proportional relationship between the amount of DNA amplification and the intensity of the detected fluorescence. Thus, in the presence of target DNA, the fluorescence increases logarithmically until it exceeds the traditional threshold level of 30 fluorescence units, where background fluorescence is eliminated (Jiang, Temenak, & Richards, 2003). The number of cycles in which the fluorescence of the sample crosses the threshold is called the Cycle of Threshold (Ct). Samples without the target DNA would not have high logarithmic curves and should not cross the threshold level.

Fluorescence is activated by the fluorescent dye FAM Reporter on the molecular Beacon probe during the DNA polymerase elongation process. The uniqueness of the Beacon probe is that before the DNA polymerase elongation, the Beacon probe is curled up in a hairpin structure made by the complimentary sequences at both ends of the targeted probe; this structure makes the Beacon probe more effective than the linear TaqMan probe. On the 5’-end of the Beacon probe is the FAM Reporter that naturally fluoresces. However, before elongation, the FAM Reporter is in the vicinity of the dye quencher (Black Hole Quencher was used in this study) on the 3’-end of the probe, thus its light is extinguished. Just before elongation, the complimentary target gene attracts the Beacon probe and breaks the complimentary handle of the molecular hairpin as the probe anneals to the template DNA. Since the hairpin has been stretched out, the quencher can no longer cover the fluorescence from the reporter, and the FAM reporter is activated. This activation is only possible in the presence of the targeted template DNA (Jiang, Temenak, & Richards, 2003).

The quantitative real-time Polymerase Chain Reaction is a tool that has shown to be both an efficient and accurate method of detecting the presence of targeted bacterial DNA by measuring the intensity of fluorescence from the activated probe. Utilizing this tool, Jiang, Chan, Temenak, Dasch, Ching, & Richards (2004), in a study to develop a sensitive and specific qPCR assay for Orientia tsutsugamushi, stated the four clear advantages of qPCR: “1) faster results, 2) potential for automation for high throughput, 3) multiplexing, and 4) quantitative information useful for clinical monitoring of appropriate response to treatment and prognosis, and for experimental studies.”

In anotherstudy by Jiang, Temenak, & Richards (2003) to develop qPCR assays to detect R. prowazekii and B. recurrentis, two assays were tested in one reaction tube, in order to make a more efficient diagnosis. Even though both R. prowazekii and B. recurrentis could be recognized in one tube, the B. recurrentis assay detected the extraneous geneB. duttoni at normal annealing temperature. However, this extraneous strand was absent when the annealing temperature was raised. In this study, since there was only one bacterium of interest and the B. lonestari assay reactions were not mixed with other assays, it was not necessary to add another probe, and the accuracy of the results were higher due to less cross-contamination between the two assays.

This study used a simplified version of the methods described in Jiang’s research in 2004. Since qPCR testing was sufficient and accurate enough with the four advantages for data collecting, the study did not utilize PCR-RFLPtesting as Fukunaga, et. al., did in 1996. In Jiang’s research (2004), the sensitivity of the specific O. tsutsugamushi assay was consistent in that three to ten copies of DNA target sequence per microliter were detected.

The goal of this study was to develop a qPCR assay that is sensitive to 10copies/L and specific for B. lonestari. To do so, genetic sequences unique to B. lonestari were first identified and then synthesized. The designed assay incorporated both the Beacon probe and the primers. B. lonestari-specific primers on either end of the Beacon probe aided the attachment of DNA polymerases during the start of elongation. In order to test the sensitivity, the cloned flagellin plasmid DNA was tested using the designed assay of different concentrations. Plotting the Cycle of Threshold (Ct) values of each against their respective concentrations, a standard curve was drawn, showing the sensitivity of the assay. If the assay was specific as proposed, it would only detect B. lonestari flagellin samples and no other kinds of DNA. In this study, because thirty non-B. lonestari DNA were tested to demonstrate the assay specificity, none of the thirty DNA samples were expected to cross the threshold with positive Ct values.

Materials and Methods

Assay primers and probe. The core of this project was to search for the most effective assay of B. lonestari. The assay, which included the primers and probe, was first picked out by analyzing the most unique sequence in the B. lonestari flagellin gene. The gene sequence (accession code AY850063) was obtained from NCBI GenBank (NCBI; Bethesda, MD). NCBI’s Basic Local Alignment Search Tool (BLAST) was used to identify 34 highly related sequences. The sequences were aligned using ClustalW (EBI; Cambridge, UK) and the variations of the sequences were colored-coded using GeneDoc (PSC; Pittsburgh, PA). Regions of uniqueness were identified, and it was found that an 18bp region between 667-668bp was deleted only in the B. lonestari flagellin gene sequence(see Table 1. The Deleted Region of the Assay Probe). In addition to this deletion, three other unique base pairs and two semi-unique base pairs were present in the primer. Around this target area, the software program Beacon Designer 4.0 (Premier Biosoft; Palo Alto, CA) was used to determine the best combination of primers and probe for qPCR analysis. The output primers also covered multiple unique base pairs: three unique and two semi-unique base pairs in the forward primer, and two unique and three semi-unique base pairs in the reverse primer. The Beacon probe created for this study had a FAM Reporter on the 5’ end and a Black Hole Quencher 1 (BHQ-1) on the 3’ end. Oligonucleotides of primers and probe (see Table 2. Borrelia lonestari Primers and Probe Sequences Designed;see Figure 1. Developed Primers and Probes on the 990bp B. lonestari Flagellin Gene) were synthesized by Sigma Genosys (The Woodlands, TX).

Cloning of the B. lonestari flagellin gene. The gene was cloned to make enough copies of the gene for future optimization tests. The entire B. lonestari flagellin gene was cloned with bacteria rather than alternative methods, such as PCR, because the clones were stored for two weeks, and bacteria provided better environments for the clones. Primers for the full gene sequence of B. lonestari flagellin were designed with Beacon Designer 4.0 by choosing common regions after alignment of similar sequences and were synthesized by MWG (High Point, NC) (see Table 2. Borrelia lonestari Primers and Probe Sequences Designed).

Using this pair of primers, a 960 bp amplicon was amplified by nested PCR and verified by agarose gel electrophoresis. QIAquick PCR Purification Kit (Qiagen; Valencia, CA) was used to purify the PCR product according to the manufacturer’s instructions. The purified PCR product was then cloned using TOPO XL PCR Cloning Kit (Invitrogen; Carlsbad, CA) as per the manufacturer’s instructions. After lysing the resulting bacterial cells via boiling for 10 minutes of 10μL culture, both qPCR and conventional PCR were employed to identify the positive clones.

Following the manufacturer’s instructions, the QIAprep Spin Miniprep Kit (Qiagen; Valencia, CA) was used to isolate the plasmid from two of the five positive bacterial cultures. The concentrations of those two plasmid were measured using the Eppendorf BioPhotometer (Eppendorf; Westbury, NY).

Optimization tests. The purpose of optimization tests was to yield the best results by balancing the right concentrations of qPCR reactants. Optimization tests were performed with qPCR for conditions of the assay primers (range 0.1μM – 0.7μM), assay probe (range 0.2μM – 0.7μM), MgCl2 (ranges 3mM – 7mM and 3.5mM – 7.5mM), and annealing temperature (range 56˚C - 66˚C). For each test, 101 copies/μl and 102 copies/μl of DNA templates were used. Optimal conditions were determined based on the lowest Cycle of Threshold (CT) values of logarithmic fluorescence using the Smart Cycler (Cepheid; Sunnyvale, CA) (see Table 3. Optimal Final Conditions for qPCR). To ensure minimal contamination, a DNA and RNA-free clean room was used to prepare the qPCR master mixes. Three negative controls were prepared with Milli-Q water; two negative controls were prepared in the biosafety cabinet, not in the clean room; and the third negative control was prepared in the clean room. Thermal cycling parameters included a pre-hold of 50˚C for twominutes, a hold at 95˚C for two minutes, followed by 50 two-step cycles of 94˚C/5seconds and 60˚C/30seconds.

Assay sensitivity and specificity. In order to verify the efficacy of the created assay, two tests were run: one forsensitivity (i.e. the minimum level that can still be detected) and the other for specificity (i.e. the accuracy of the assay). To determine the assay sensitivity, the plasmid with target B. lonestari flagellin gene was chosen for use in standard dilutions based on its concentration (~6.93x1010 copies/μL). Serial ten-fold and half-log dilutions of the plasmid were made with Tris EDTA (TE) buffer. The specificity of the assay was tested using seven related Borrelia species, and twenty-three non-related bacterial genomic DNA (see Table 4. List of Bacteria Sequences Tested for B. lonestari Assay Specificity).

Results and Discussion
Assay primers and probe. The synthesized primers and probe were tested on two unknown concentrations of B. lonestari samples, and the results for both were positive, verifying the ability of the assay to detect B. lonestari (see Figure 2. Initial Testing of Assay Primers and Probe).

Cloning of the B. lonestari flagellin gene. All five bacteria cultures obtained from DNA cloning of B. lonestari flagellin gene demonstrated positive results when assayed by B. lonestari qPCR (see Figure 3. Positive Results for All Five Transformant Bacteria Colonies) and PCR, indicating that the plasmid transformation had been successful.

Assay sensitivity. When serial dilutions of the plasmid range from 107 to 100 copies/μL were tested eight times, assay sensitivity demonstrated the ability to detect 100 copies/μL, and the assay consistently detected 101copies/μL (Ct value = 37.88). The R2 value of the standard curve obtained was 0.986 (see Figure 4. Standard Curve, FAM Log, and FAM Ct for Assay Sensitivity).

Assay specificity. All seven Borrelia related and twenty-three unrelated genomic DNA samples were shown to be negative (see Table 4. List of Bacteria Sequences Tested for B. lonestari Assay Specificity), which meant that the assay was indeed specific to B. lonestari.

Conclusion

In this experiment, primers and probe were developed to create a specific and sensitive B. lonestari qPCR assay. Through qPCR, the sensitivity threshold of the assay was shown to be onecopy/μL and the assay consistently detected 10copies/L of B. lonestari target DNA. Testing against 30 related and unrelated bacterial sequences producing negative results,verifying the specificity of assay.

Future research should include testing of clinical samples by B. lonestari qPCR in order to further support conclusions made in this study. Utilizing this qPCR assay should reduce diagnosis time and increase diagnosis accuracy of B. lonestari infection in STARI patients, expediting proper treatment process.

References

Burkot, T. R., Mullen, G. R., Anderson, R., Schneider, B. S., Happ, C. M., & Zeidner, N. S. (2001, May/June). Borrelia lonestari DNA in adult Amlyomma americanum ticks, Alabama. Emerging Infectious Diseases, 7(3), 471-473.

Fukunaga, M., Okada, K., Nakao, M., Konishi, T., & Sato, Y. (1996, October). Phylogenetic analysis of Borrelia species based on flagellin gene sequences and its application for molecular typing of lyme disease Borreliae. International Journal of Systematic Bacteriology, 46(4), 898-905.

Jiang, J., Chan, T.-C., Temenak, J. J., Dasch, G. A., Ching, W.M., & Richards, A. L. (2004). Development of a Quantitative Real-Time Polymerase Chain Reaction assay specific for Orientia tsutsugamushi. American Society of Tropical Medicine and Hygiene, 70(4), 351-356.

Jiang, J., Temenak, J. J., & Richards, A. L. (2003). Real-time PCR duplex assay for Rickettsia prowazekii and Borrelia recurrentis. In K. E. Hechemy, T. Avsic-Zupanc, J. E. Childs, & D. A. Raoult (Eds.), Rickettsiology present and future directions (pp. 302-310). New York: New YorkAcademy of Sciences. From Rickettsiology Present and Future Directions.

Moore IV, V. A., Varela, A. S., Yabsley, M. J., Davidson, W. R., & Little, S. E. (2003, January). Detection of Borrelia lonestari, putative agent of southern tick-associated rash illness, in white-tailed deer (Odocoileus virginianus) from the southeastern United States. Journal of Clinical Microbiology, 41(1), 424-427.

Acknowledgements

I would like to thank my mentors Dr. Allen L. Richards and Dr. Ju Jiang, my research sponsor Dr. Wood, the TJHSST DNA Techlab Director Mr. Lampazzi, and the TJHSST Mentorship Director Mr. Pearce for their continuous guidance and support through my project. In addition, I would like to thank the NavyMedicalResearchCenter, the Science & Engineering Apprentice Program, and the TJHSST Mentorship Program for giving me the opportunity to do this research. Finally, I would like to thank my family for encouraging for me to pursue science.