Detection of Tickborne Relapsing Fever Spirochete, Austin, Texas, USA

Jack D. Bissett; Suzanne Ledet; Aparna Krishnavajhala; Brittany A. Armstrong; Anna Klioueva; Christopher Sexton; Adam Replogle; Martin E. Schriefer; Job E. Lopez


Emerging Infectious Diseases. 2018;24(11):2003-2009. 

In This Article

Materials and Methods

The Patient

The patient was a 34-year-old previously healthy woman whose illness began on March 29, 2017, when she had a headache, myalgias, arthralgias, and malaise. On March 30, 2017, she traveled to California on a previously scheduled trip and was febrile. The patient had a temperature of 104°F that increased to 105°F, at which point she sought medical treatment at an urgent care clinic. Complete blood counts, and levels of electrolytes, blood urine nitrogen, creatinine, and liver enzymes were within references ranges. She was given intravenous fluids, discharged with a diagnosis of a viral illness, and given instructions for symptomatic treatment of this illness.

Over the next 2 days, the patient still had a high fever, which prompted her to return to the urgent care clinic. Given her ongoing signs and symptoms, she was referred to a local hospital emergency department in California where she underwent computed tomographic imaging of her brain and a lumbar puncture for cerebrospinal fluid analysis. Computed tomographic imaging of the brain showed no abnormalities. Analysis of cerebrospinal fluid also failed to demonstrate abnormal findings. It was again concluded that she likely had a viral infection and was discharged from the emergency department with instructions for symptomatic treatment.

On April 2, 2017, she reported a blotchy maculopapular rash that began on her extremities and spread to her trunk. The rash was nonpruritic, persisted for several days, then gradually faded away. The patient returned home to Austin, Texas, with a temperature of 104°F and continued to have a mild headache in conjunction with intermittent fever. She did not have nausea, vomiting, or diarrhea. Given her ongoing symptoms, on April 9, 2017 she sought an evaluation at an acute care hospital emergency department. At the emergency department assessment, a hematoxylin and eosin–stained peripheral thin blood smear was prepared for evaluation of bloodborne pathogens.

Real-Time PCR Analysis

We performed a real-time PCR assay on DNA extracted from the spirochete-positive peripheral thin blood smear. We scraped 10% of the contents of the slide with a scalpel and placed the contents in a tube containing 200 μL of phosphate-buffered saline (GIBCO, Gaithersburg, MD, USA). We then extracted DNA by using a QIACube (QIAGEN, Valencia, CA, USA), a tissue protocol, and an elution of 100 μL. A total of 5 μL of the eluted DNA extract was used per 20-μL final volume reactions with primers and probes specific for the B. turicatae glycerophosphodiester phosphodiesterase (glpQ) gene (forward primer 5′-GCCTGTCAGAATGAAAAA-3′, reverse primer 5′-CACCTCTGTGAGCTATAATT-3′, and probe FAM-5′-TGAGTATGACAAACAAAAAACCACCA-3′-BHQ) and the B. hermsii glpQ gene (forward primer 5′-TCCTGTCAGGGCGAAAAAAT-3′, reverse primer 5′-GCTGGCACCTCTGTGAGCTAT-3′, and probe FAM-5′-AGTCAAAACCAAAAATCACCA-3′-BHQ). The PCR was performed as described.[14] A no template (DNA) sample was used as a negative control, and DNA extracted from B. turicatae and B. hermsii cultures were used as positive controls.


We also performed immunoblotting for relapsing fever group Borrelia spp. and B. turicatae, as described.[3,13,15] We subjected protein lysates from 1 × 107B. turicatae and 1 μg of recombinant Borrelia immunogenic protein A (rBipA) to electrophoresis by using Mini PROTEAN TGX Precast Gels (Bio-Rad, Hercules, CA, USA) and transferred them onto Immobilon polyvinylidene difluoride membranes (Millipore, Billerica, MA, USA). rBipA was produced as a thioredoxin fusion protein to facilitate solubility and is ≈15 kDa larger than the native protein.[13]

We sent a deidentified serum sample collected 50 days after infection to Baylor College of Medicine (Houston, TX, USA) for evaluation. This sample was diluted 1:200 in Tropix I-Block Protein-Based Blocking Reagent (ThermoFisher Scientific, Waltham, MA, USA), and polyvinylidene difluoride membranes were probed for 1 hour. Recombinant protein G conjugated to horseradish peroxidase (ThermoFisher Scientific) diluted 1:4,000 was used as the secondary molecule, and antibody reactivity was detected by chemiluminescence using the Amersham Enhanced Chemiluminescence ECL Western Blotting Detection Reagent (GE Healthcare, Little Chalfont, UK).

Collection of O. Turicata Ticks

Because access to the alleged exposure site was not available, we selected a field site in a public park near the suspected exposure site. We determined that the park was in Austin by using the Jurisdictions Web Map maintained by the Enterprise Geospatial Service Program of the City of Austin ( Collection efforts were performed in July and November 2017. We placed CO2 tick traps with dry ice as bait in locations with promising O. turicata tick habitats, as described.[7] As ticks emerged from leaf litter, we stored them in vials labeled according to collection site and date. We collected 20 nymphal ticks from the first location in July and November 2017. We identified the second location in November 2017 and collected 30 nymphs from this site.

Tick Feedings and Isolation of Spirochetes

All animal studies were approved by the Institutional Animal Care and Use Committee at Baylor College of Medicine. The laboratory animal program follows standards and guidelines established by the Association for Assessment and Accreditation of Laboratory Animal Care and the National Institutes of Health Office of Laboratory Animal Welfare. Animal husbandry was provided by trained veterinary staff and animal care technicians.

We randomly selected 10 O. turicata nymphs from each collection vial and allowed them to feed on ICR mice obtained from the Institute of Cancer Research (Philadelphia, PA, USA).[16] Animals were sedated by inhalation of isoflurane (Henry Schein, Melville, NY, USA), and ticks were placed on the shaved abdomen of mice and allowed to feed to repletion. Upon completion of the blood meal, ticks were stored in TTP TubeSpin Bioreactor Tubes (MidSci, St. Louis, MO, USA) and housed at 25°C and a relative humidity of 85%.

We examined the 3 mice for spirochete infection for 10 days by nicking the tip of the tail and expressing a drop of blood onto a microscope slide. A coverslip was placed over the blood and examined at 20× magnification by using a CX33 Trinocular Dark Field Microscope (Olympus, Center Valley, PA, USA). When spirochetes were observed, we obtained a terminal blood sample from the sedated mouse by using cardiac puncture. We centrifuged ≈500 μL of blood at 5,000 × g for 10 min and then inoculated 50 μL of serum into 4 mL of modified Barbour–Stoenner–Kelly (mBSK) medium.[17] Cultures were grown at 35°C in an atmosphere of 5% CO2 and examined for spirochetes every 4 days. When exponential growth was reached, we passaged cultures into two 50-mL culture tubes containing fresh mBSK medium for isolation of genomic DNA and generation of stocks (stored in glycerol). For DNA isolation, we produced spirochete pellets by centrifuging the 50-mL culture tubes at 8,000 × g for 20 min and extracted genomic DNA as described.[18] We designated the 3 isolates from the mice as BRP1, BRP1a, and BRP2.

Plasmid Analysis and Genetic Typing of Relapsing Fever Spirochete Isolates

We performed reverse-field electrophoresis to resolve plasmid content, as described.[9] In addition to BRP1, BRP1a, and BRP2, we compared genomic DNA plasmid profiles with the 91E135 isolate, which originated in Crockett County, Texas.[9] Total DNA was used from the 91E135 isolate passaged 5 times and from the Austin isolates passaged 2 times after initial isolation from mice.

We subjected 500 ng of genomic DNA from each isolate to electrophoresis at 100 V for 15 min and at 80 V for ≈40 hours by using a PPI-200 Programmable Power Inverter (MJ Research, Inc., Waltham, MA, USA). Gels were stained with GelRed Nucleic Acid Stain (Phenix Research Products, Candler, NC, USA) according to the manufacturer's instructions.

We performed multilocus sequencing for relapsing fever Borrelia flaB, rrs, and gyrB genes. Amplicons were generated by using specific primer sets (Table). For the rrs gene, we used primers UniB and FD3 to generate the amplicon, and the remaining internal rrs gene primers were used for sequencing. PCR conditions were an initial incubation at 94°C for 2 min, followed by 35 cycles at 94°C for 30 s, annealing at 55°C for 30 s, and an extension at 72°C for 3 min. After the last cycle, an extension was performed at 72°C for 5 min. We analyzed PCR products by agarose gel electrophoresis to confirm the expected molecular size. Each amplicon was sequenced by Lone Star Laboratories (Houston, TX, USA) by using specific primers (Table). Chromatograms were analyzed, and poor sequences were trimmed by using Vector NTI software (Life Technologies, Grand Island, NY, USA). We performed a BLAST ( search with assembled contiguous DNA segments (contigs) to speciate isolates.