Isolation and Genetic Characterization of the Human Relapsing Fever Spirochete Borrelia persica from a Dog with Improved Cultivation Techniques

,


Introduction
Spirochetes of the relapsing fever (RF) group Borrelia are the causative agents of RF in humans and companion animals [1,2].Tick-borne relapsing fever (TBRF) is caused by different Borrelia species which are mostly transmitted by argasid ticks from the genus Ornithodoros with some exceptions such as Borrelia miyamotoi transmitted by ixodid ticks [3,4].Borrelia persica was reported as the causative agent of TBRF in several countries including Iran, Pakistan, China, Uzbekistan, Tajikistan, and Egypt, and it is the only reported endemic RF species in Israel [5][6][7][8][9][10].B. persica is transmitted by the tick Ornithodoros tholozani, whose geographic distribution extends from India and Central Asia to Egypt and largely overlaps the distribution of B. persica [11,12].Infection with B. persica has been reported in various wild and domestic animal species including the domestic dog (Canis lupus familiaris), cat (Felis catus), European badger (Meles meles), striped hyena (Hyaena hyaena), golden jackal (Canis aureus), red fox (Vulpes vulpes), rock hyrax (Procavia capensis), and several rodent species including the social vole (Microtus socialis), fat sand rat (Psammomys obesus), and Cairo spiny mouse (Acomys cahirinus) [13][14][15][16].Movement of infected hosts between TBRF endemic and TBRF-free areas poses the risk of introduction of the infection into new countries [10,17].
The reported incidence of TBRF in civilians in Israel has declined from 0.35 cases per 100,000 inhabitants in the years 1975-1985 to 0.11/100,000 from 1986 to 2003.Nevertheless, the incidence among Israeli military personnel remained stably high with an annual average of 6.4 cases per 100,000 [20].The mortality rates of Old World TBRF in untreated human patients are in the ranges of 4%-10% and are largely influenced by coexisting conditions, such as malnutrition, dehydration, and coinfection.Although mortality rates decrease with prompt antibiotic treatment to less than 2%, severe lifethreatening complications, such as aseptic meningitis, acute respiratory distress syndrome, and acute neonatal death were reported in human patients infected with B. persica [21,22].In dogs and cats, RF caused by B. persica is associated with clinically evident illness and clinicopathological findings, such as fever, lethargy, anorexia, anemia, and thrombocytopenia, and can be fatal [23].Clinical disease in pets was reported in infection with other RF species including B. turicatae, B. hermsii, and B. hispanica [24][25][26].
One of the most challenging aspects of borreliosis research is the ability to successfully cultivate long-term stable cultures of RF species.Some of the RF species are notoriously difficult to cultivate in vitro [27,28].Early attempts to cultivate Borrelia spirochetes were made by inoculating blood from naturally infected subjects into laboratory animals and serial passages of infected blood to other animals [29].Over the past decades, several cultivation media were developed for cultivating Borrelia spp.including the Barbour-Stoenner-Kelly (BSK) medium, modified Kelly-Pettenkofer (MKP), and variations of these media, which made in vitro cultivation of many Borrelia species possible [30,31].Although some RF Borrelia spp.have been adapted to BSKII medium, B. persica could not be cultured in vitro until 2015, when the pathogen was isolated from the blood of a cat diagnosed with this infection in Israel and from ticks collected in a cave where humans were infected in Israel [32].These strains were successfully isolated using a new modification of the MKP medium, named the Pettenkofer-LMU Bp medium (Pett./LMUBp.).The first isolation and the first two passages of these strains were done into human serum and then transferred into tubes containing Pett./ LMU Bp. medium.
Glycerol is used as a cryoprotective additive at different concentrations for conserving and freezing a variety of microorganism including viruses, bacteria (including rickettsia and mycoplasma), yeasts, algae, and protozoa [33].Nevertheless, although glycerol was found to have a positive cryoprotective effect on diverse microorganisms, it was observed to have a toxic effect on others, with deleterious effect on infectivity, such as in the case of several serovars of the spirochete Leptospira interrogans, Trypanosoma spp., Anaplasma phagocytophilum, and Plasmodium spp.[34][35][36][37][38].
In this study, we describe the first isolation and in vitro cultivation of B. persica from a dog and new modifications to simplify its isolation and cryopreservation.

B. persica Strain Collection and
Isolation.Blood was collected in a 5 ml EDTA tube (Sarstedt AG & Co., Numbrecht, Germany) from a sick dog brought for veterinary care in the city of Beer Sheva in Southern Israel.A blood smear, which was made and stained with May-Grünwald-Giemsa by an attending veterinary clinician, showed spirochetemia suspected as infection with a RF Borrelia spp.(Figure 1) [23].One hundred microliters of the blood were transferred to a microfuge tube (SSIbio, Lodi, CA, USA) for DNA extraction.The rest of the blood (∼2 ml) was divided into two microfuge tubes, which were centrifuged for 10 min at 350x g.The precipitate was washed three times with sterile phosphatebuffered saline.The supernatant was discarded after the final wash and the precipitate containing spirochetes was mixed with 3 ml of prewarmed Pettenkofer-LMU Bp medium, as previously described [32] in a 5 ml polypropylene tube (Sarstedt AG & Co., Numbrecht, Germany) and incubated at 37°C with no additional CO 2 supplementation [32].
2.2.Molecular Analysis.DNA was extracted from 100 µl of culture and from the dog's blood using a commercial kit (DNeasy Blood and Tissue Kit, Qiagen, Germany) following the manufacturer's protocol.Real-time and conventional polymerase chain reactions (PCR) were performed on DNA extracted from the culture and blood samples by the amplification of three different genes of RF borreliae.An approximately 346-bp fragment of the flaB gene was amplified as previously reported for other Borrelia species using primers Bfpbu and Bfpcr [39] and an approximately 280 bp fragment of the glycerophosphodiester phosphodiesterase (GlpQ) gene, which is specific for RF borreliae, was amplified using primers 510f and 770r [40].In addition, an approximately 515 bp fragment of the Borrelia 16S ribosomal RNA (16S rRNA) gene was amplified using primers rec4 and rec9 [41].Furthermore, to detect possible coinfection with other tick-borne disease agents, real-time, and conventional PCRs were performed on DNA from the dog's blood sample for the detection of Ehrlichia canis, Babesia spp., and Hepatozoon spp.(Table 1).
The flaB and glpQ real-time PCR protocols were carried out with an initial hold for 3 min at 95°C, followed by 45 cycles of 15 s at 95°C, 30 s at 58°C, and 10 s at 72°C.The melting phase started at 60°C, each step rising by 0.3°C/s, and finishing at 95°C with a hold for 90 s at the first step and 5 s at the subsequent steps.Each reaction was performed in 20 μl reaction volume containing 4 μl of sample DNA, 0.5 μM of each primer, 0.6 μl of Syto9 (Invitrogen, CA, USA), 4.4 μl of ultra-pure water, and 10 μl of DreamTaq PCR Master Mix (Thermo Scientific, Loughborough, UK).All reactions were performed in the Life Technologies StepO-nePlus real-time PCR system (Thermo Fisher Scientific, Waltham, MA, USA).The E. canis real-time PCR protocol was carried out as previously described [43].The flaB conventional PCR protocol was carried out for 45 cycles consisting of denaturation at 94°C for 30 s, annealing at 56°C for 30 s, and extension at 72°C for 30 s, and there was a final extension step consisting of 5 min at 72°C.The 16S rRNA conventional PCR protocol was carried out for 35 cycles consisting of denaturation at 96°C for 1 min, annealing at 94°C for 1 min, and extension at 72°C for 1 min, and there was a final extension step consisting of 5 min at 72°C.The 18S rRNA piroplasmid conventional PCR protocol was carried out, as previously described [44].Each reaction was performed in 25-μl reaction volume containing 4 μl of sample DNA, 1 μM of each primer, and 19 μl of distilled extra pure water.DNA extracted from 3 ml of B. persica culture was used as a positive control.All positive PCR products were sequenced from both directions using the respective forward and reverse primers.Sequencing was performed by the BigDye Terminator cycle sequencing chemistry from Applied Biosystems ABI 3700 DNA Analyzer (ABI, Carlsbad, CA, USA) and the ABI Data Collection and Sequence Analysis software at the Center for Genomic Technologies, Hebrew University of Jerusalem, Israel.DNA sequences were aligned using the Molecular Evolutionary Genetics Analysis software MEGA, version 11.013 [45].Sequences were further compared to the Gen-Bank database using the BLAST algorithm (National Center for Biotechnology Information, Bethesda, USA; http://blast.ncbi.nlm.nih.gov/Blast.cgi).Species-level identification was obtained according to the closest BLAST match when sequences showed identity higher than 99% to the first match GenBank sequences.

Growth Characteristics.
The growth dynamics of B. persica strain HU-D03 in Pett./LMUBp were evaluated for longterm stability and propagation characteristics.The long-term stability of the isolate was evaluated by inoculation of 100 µl of the culture's first passage (P1) in 3 ml Pett./LMUBp medium in triplicates and subsequent incubation at 37°C.The cultures were checked every 24-48 hr, and the viable cell density was measured at day 3 or 4, using the improved Neubauer counting chamber (Merck KGaA, Darmstadt, Germany).Only motile spirochetes were included in the counting.When bacterial growth reached cell density of at least 1× 10 6 /ml with highly motile spirochetes, 100 µl of the culture were transferred into 3 ml of fresh medium, and passages were made every 3 or 4 days up to passage 30.Spirochete growth dynamic was studied in repeated reactions (six times) as follows: 100 µl of B. persica stock passage (P7) were introduced into 3 ml Pett./LMUBp medium and incubated at 37°C.The cultures underwent spirochete counts daily for 11 days and analysis of bacterial growth was carried out including cell counts and evaluation of spirochete viability [46].Generation time was calculated using the equation G = t/3.3(log b/B), where G is generation time, b is the number of cells at the end of the exponential phase, B is the number of cells at the beginning of exponential phase, and t is the time interval between the beginning and the end of exponential phase [47].Data on growth dynamics were analyzed using Microsoft Excel Software (Microsoft Corporation 2013).
2.4.Culture Maintenance.Four days after the spirochete's inoculation into the culture medium as passage 1 (P1), 100 µl of culture were transferred to 3 ml of prewarmed medium and designated as passage 2 (P2) of B. persica strain HU-DO3.Additional passages from live cultures were further made by transferring 100 µl of culture to 3 ml of prewarmed medium every 3-4 days and testing for viability by dark-field microscopy.

Transboundary and Emerging Diseases
Cultures of the passages were incubated at 37°C, and a portion of 100 µl of each passage starting from P2 was frozen directly at −80°C without any cryopreserving additive at day 3 of the culture, in sterile 1.5 ml microfuge tubes (Labcon, Langenhagen, Germany).Freezing was carried out following testing for viability when reaching a cell density of 1 × 10 6 spirochetes/ml or above, possessing high viability as measured by motility speed and intact morphology.The thawing process was made by quick defrosting of the aliquots by warming the tube to body temperature (37°C) in a laboratory water bath and when completely defrosted, immediate transfer of 100 µl into prewarmed medium at 37°C.

Results
A 10 years old, female spayed boxer dog was admitted to a veterinary clinic due to lethargy.It had a rectal temperature of 39.5°C indicating mild fever.A complete blood count showed mild normocytic normochromic anemia (Hematocrit 34.5%, reference range (RI): 37.3%-61.7%;mean cell volume (MCV) 69.1 fL, RI: 61.6-73.5 fL; mean corpuscular hemoglobin concentration (MCHC) 36.6 g/dl, RI: 32-37.9g/dl) and severe thrombocytopenia (5,000 per µl; RI: 148,000-484,000 per µl).Blood smear microscopic evaluation showed no red blood cells regeneration, confirmed the thrombocytopenia, and demonstrated infection with spirochetes suspected as RF Borrelia spp.(Figure 1).Serum biochemistry panel showed only a mild elevation in alanine transaminase (ALT) activity (139 U/L; RI: 10-125 U/l).Real-time PCR of blood was positive for the Borrelia flaB, glpQ, and 16S rRNA genes, and negative for E. canis, Babesia and Hepatozoon species.Sequencing analysis of the flaB and 16S rRNA genes showed 100% identity to B. persica GenBank accession numbers KX258795.1 and KU565880.1 previously amplified from an O. tholozani tick and from a cat in Israel, respectively.Furthermore, sequence analysis of the glpQ-gene showed 99.6% identity to B. persica GenBank accession MH923346.1 previously amplified from an O. tholozani in Israel.
A cell density of 2 × 10 6 viable spirochetes per ml was observed by dark-field microscopy 3 days after the initial cultivation of B. persica HU-D03 (P1) in Pett./LMUBp medium which originated from the blood of a spirochetemic febrile dog.Growth behavior dynamics are shown in Figure 2.During the first 4 days, the spirochete motility was intense and the microorganisms presented in ring forms and clear helical shapes.The highest cell densities were measured on day 8, with an average of 3.5 × 10 7 /ml.From day 6, the motility decreased, and the ring formation declined.From day 8, the number of motionless and partially dissolved spirochetes increased gradually.
Generation time was calculated based on the mean cell numbers during the exponential phase, as shown by the logarithmic representation of the spirochete growth (Figure 3).The generation time of B. persica strain HU-D03 in Pett./LMUBp medium was calculated based on two-time intervals in the exponential growth phase: The mean cell number/milliliter at the end of exponential phase (day 4) was 1.4 × 10 7 and the mean cell number/milliliter at the beginning of exponential phase (day 2) was 1.4 × 10 6 .Based on the equation G = t/3.3log(b/B), the generation time was calculated as 14.6 hr.Transboundary and Emerging Diseases Successful propagation of thawed cryoprotectant-free B. persica strain HU-D03 was achieved four times using different passages (P5, P6, P7, and P24).These aliquots were frozen for 2-3 months before thawing.To evaluate the spirochetes' ability to propagate well after prolonged freezing, two additional freeze and thaw cycles of one of the cultures (P5) were made showing that the isolates revived and grew successfully.
Sequencing data were obtained for both blood and culture (P3).For the Borrelia flaB gene, 706-bp sequences amplified from the blood and from culture were found 100% identical to each other and 100% identical to B. persica (KX258795.1)and deposited in GenBank under accession number OR900368.Phylogenetic analysis based on a 706bp segment of the flaB gene sequence (Figure 4) revealed that both blood and culture isolations of B. persica sequences   Sequencing data of a 235-bp amplified fragment of the Borrelia glpQ-gene which was deposited in GenBank as accession number OR900369 showed that sequences obtained from the blood and culture (P3) were 100% identical to each other and 99.6% identical to B. persica (accession number MH923346.1).
Phylogenetic analysis of the 235-bp glpQ-segment (Figure 5 Sequencing data of a 484-bp sequence of the Borrelia 16S rRNA-gene obtained from both dog blood and culture (P3) and deposited in GenBank as accession number OR890442 showed that sequences from the blood and culture were 100% identical to each other and 100% identical to B. persica (accession number KU565880.1).Phylogenetic analysis of a 407-bp segment of the Borrelia 16S rRNAgene (Figure 6) showed that sequences from this study clustered with other

Discussion
This study describes the first cultivation of the RF spirochete B. persica isolated from a dog.B. persica is considered difficult to culture in medium and before 2015, there were no cultivated isolates grown in vitro of this spirochete, and it was only possible to pass it in laboratory animals for research purposes [7,48].In 2015, B. persica was isolated successfully from a cat and an O. tholozani tick and these are the only existing isolates of this Borrelia spp.until the current canine isolation described herein [32].Cultivation described previously required the initial use of freshly prepared human serum and revival of the isolates following freezing in glycerol was difficult and often unsuccessful in the author's experience [32].The current study has found a simplified successful and reliable method for directly freezing and thawing of B. persica isolates without using glycerol or other preservatives.
Clinical disease due to RF Borrelia spp.has been reported in domestic dogs from Israel and Iran infected with B. persica, dogs from North America infected with B. turicatae and B. hermsii, and dogs from Europe infected with B. hispanica [23,24,26,[49][50][51].Previous reports described the successful isolation and cultivation of the North American RF species, B. turicatae, and B. hermsii, from spirochetemic dogs.The  New sequences derived from this study are marked with black diamond squares.The GenBank accession numbers, species of infected host, and country of origin are included for each sequence.The Tamura-3-parameter model was used in the construction of this phylogram with bootstrap performed on 1,000 replicates, and values higher than 70% are indicated.6 Transboundary and Emerging Diseases cultivation method described for B. turicatae included direct isolation into a BSKII medium in one report [25]; however, isolation of B. hermsii and B. turicatae in other reports required two in-vivo steps of inoculating laboratory mice (Mus musculus) with the dogs' infected blood samples or sera [23,24].The growth dynamics of the isolated canine strain of B. persica showed similar characteristics to the growth dynamics found for the previous cat and O. tholozani tick isolates, with approximately 2 days of lag phase and 3 days of log phase.Furthermore, the maximal cell density was observed on day 8 for all mentioned isolates.However, the cell density reached during growth of the canine isolate was approximately 10 times higher than those reported for the cat and tick isolates [30].Future studies will be directed to compare the growth dynamics of B. persica and other RF Borrelia spp.using the same medium and under the same conditions in order to improve the characterization of the growth of the new canine B. persica isolate.
The generation time of Borrelia spp.varies and range 7-20 hr [52].The generation time of the isolated canine strain of B. persica from this study (14.6 hr) was shorter than previously reported for the cat and tick isolates (19.1 hr), but longer than reported for other RF Borrelia spp.such as B. hermsii (7 hr) and B. recurrentis (8-9 hr) [32,[53][54][55].
In contrast to the well-established knowledge on the genetic structure, epidemiology, clinical features, and virulence factors of Borrelia burgdorferi that belongs to the Lyme disease group, knowledge on some of the RF Borrelia group species, and B. persica in particular, is lacking and only a few RF species have been comprehensively studied [56][57][58].The ability to grow an in vitro culture of B. persica is of great importance to the study of the pathogenesis of this spirochete.Therefore, the successful methods of isolation, cultivation, and preservation reported in this study may provide an opportunity to study the infection mechanisms of B. persica in detail.
Phylogenetic analysis based on flaB, glpQ-, and 16S rRNA gene sequences of the isolated canine strain of B. persica in this study revealed identity or very close genetic proximity to other B. persica sequences reported previously from human patients, ticks, and animals.Furthermore, a close genetic proximity of B. persica genotype III to B. caucasica, which causes TBRF in Northeastern Europe and the Caucasus area, was observed with particular attention to the flaB and 16S rRNA sequences, as reported previously [59].Comprehensive genetic studies using techniques such as whole genome sequencing analysis often require isolation and cultivation of the studied species to reach optimal outcomes [59,60].The development of reliable isolation and cultivation methods will enable future studies on the genetic and molecular basis of pathogenicity of RF species.Furthermore, it will allow the comparison of RF species, such as B. persica to other closely related species such as B. caucasica.
Cryopreservation of in vitro cultures is of considerable importance in the study of RF spirochetes.Glycerol preservation has been shown to be toxic in some cases.In a previous study, the cryoprotective effect of glycerol on the growth of several serovars of L. interrogans was compared to dimethyl sulphoxide (DMSO), which is also used frequently for bacterial preservation.Glycerol was found to have a deleterious effect on bacterial growth 10 times more than DMSO, and all serovars tested including L. interrogans serovar Canicola, L. interrogans serovar Hardjo, L. interrogans serovar Icterohaemorrhagiae, and L. interrogans serovar Pomona were inhibited by the addition of 2% glycerol, and partially inhibited with lower concentrations as low as 0.2% [34].Furthermore, glycerol was found to  Transboundary and Emerging Diseases be toxic to Streptococcus thermophilus at 10% and to have no beneficial effect on crystallinity when compared to synthetic cryoprotective polymers [61].The relative similarity between Leptospira and Borrelia spirochetes warrants the need for further research in order to study the effect of different cryoprotectant additives on the survival of B. persica in frozen cultures.
The results of this study showed that the revival of frozen cultures was surprisingly unaffected by the freezing method used in this study, without any cryoprotectant additive and after direct freezing at −80°C.Isolate aliquots were frozen for 2-3 months before successful defrosting and several passages of the same isolate with freeze-thaw-cycles did not affect the isolates viability.Cryopreservation of B. persica and perhaps other RF Borrelia spp.isolates with no cryproprotectants may be beneficial for further studies on these pathogens.

Conclusions
This study describes the first isolation of B. persica from a dog.It also indicated that B. persica can survive and propagate after direct in vitro cultivation in Pettenkofer/LMU Bp medium, without initial isolation in human serum and isolates may proliferate in-vitro with stable viability over at least 30 passages.By reaching high cell densities, it is possible to produce a large number of bacteria for further experiments.Furthermore, B. persica organisms preserved in frozen cultures may survive and be revived, even after a relatively long period of time, without the use of cryoprotectant additives.Therefore, the cultivation modifications for B. persica described here provide an alternative to previously reported in vitro methods and to in vivo cultivation in laboratory animals.

FIGURE 2 :
FIGURE 2: Growth curve of the B. persica strain HU-D03 (P9) showing the mean daily cell number/milliliter and standard error.

4
Mean cell number per day Linear polynomial fit to "mean cell number per day"

FIGURE 3 :
FIGURE 3: A logarithmic representation of the values of Figure 2. The continuous line follows the values in the exponential phase of the growth.
) revealed that sequences from this study clustered with other B. persica sequences from a human, cat, and an O. tholozani tick.As in the flaB phylogram, B. persica glpQ-sequences clustered separately from Old-World RF Borrelia spp.including B. hispanica, B. duttonii, B. crocidurae, and B. recurrentis.The Old-World species including B. persica also clustered separately from American RF Borrelia spp., namely B. hermsii, B. parkeri, and B. turicatae.

B
. persica sequences from a human, cat, and an O. tholozani tick, and also with B. caucasica isolated from an O. verrucosus tick.B. persica 16S rRNA sequences showed similar grouping patterns to the flaB and glpQ phylograms, as they clustered separately from Old-World and North American RF Borrelia spp., including B. recurrentis, B. theileri, B. hermsii, B. turicatae, and B. miyamotoi.

FIGURE 5 :
FIGURE 5: A maximum likelihood phylogram comparing 249-bp DNA fragment sequences of the glpQ-gene from the canine blood and cultured B. persica strain included in the study to sequences from other B. persica and other Borrelia spp.(GenBank accessions).New sequences derived from this study are marked with black diamond squares.The GenBank accession numbers, species of infected host, and country of origin are included for each sequence.The Tamura-3-parameter model was used in the construction of this phylogram with bootstrap performed on 1,000 replicates, and values higher than 70% are indicated.

FIGURE 6 :
FIGURE 6: Maximum likelihood phylogram comparing 407-bp DNA fragment sequences of the Borrelia 16S rRNA gene from the canine blood and cultured B. persica strain included in the study to sequences of other B. persica and Borrelia species (GenBank accessions).New sequences derived from this study are marked with black diamond squares.The GenBank accession numbers, species of infected host, and country of origin are included for each sequence.The Kimura-2-parameter model was used in the construction of this phylogram with bootstrap performed on 1,000 replicates.

TABLE 1 :
Target genes and primers used for PCR to detect and characterize B. persica, E. canis, Babesia spp., and Hepatozoon spp.
A maximum likelihood phylogram comparing 706-bp DNA fragment sequences of the flaB gene from the canine blood and cultured B. persica strain included in the study to other B. persica sequences and Borrelia spp.(GenBank accessions).New sequences derived from this study are marked with black diamonds.Note the devision of B. persica genotypes marked in Roman numerals.The GenBank accession numbers, species of infected host, and country of origin are included for each sequence.The Tamura-3-parameter model was used in the construction of this phylogram with bootstrap values deduced from 1,000 replicates, and values higher than 70% are indicated.
Transboundary and Emerging Diseases from this study clustered together with a B. persica genotype I sequence from an O. tholozani tick collected in Israel (Gen-Bank DQ679905.1),and separately from B. persica genotypes II (GenBank DQ679907.1)and III (GenBank DQ673617) previously amplified from human patients in Israel [42].All B. persica sequences clustered separately from other Old-World RF Borrelia spp.including B. recurrentis, B. duttonii, and B. crocidurae, and from North American RF species including B. parkeri, B. turicatae, and B. hermsii which also clustered separately.