Abstract

Fifty-three southern USA Borrelia isolates were characterized using randomly amplified polymorphic DNA fingerprinting analysis (RAPD). Twenty-nine types were recognized among 37 B. andersonii strains, seven types among eight B. bissettii strains, and seven types among seven B. burgdorferi sensu stricto strains. Strain TXW-1 formed a separate RAPD type. Nearly complete sequences of the rrs genes from 17 representative southern Borrelia were determined. The similarity values were found to be 96–100% within the B. burgdorferi sensu lato (s.l.) complex, 94–99% among the relapsing fever borreliae, and 93–99% between the two complexes. Phylogenetic analysis indicated that all the Borrelia strains we analyzed could be divided into two parts: the B. burgdorferi s.l. complex and the relapsing fever borreliae complex. TXW-1 segregated with the North American relapsing fever borreliae and formed a separate subbranch.

Introduction

Lyme borreliosis is the most prevalent tick-borne disease in temperate regions throughout the world. It is caused by certain genospecies of the Borrelia burgdorferi sensu lato (s.l.) complex and transmitted by several species of the Ixodes ricinus species complex. More than 80% of Lyme borreliosis cases in the USA occur along the northeastern and mid-Atlantic seaboards [1]. Different clinical symptoms of Lyme borreliosis have been associated with distinct genospecies of B. burgdorferi s.l. [2].

Relapsing fever is another medically important tick-borne borreliosis and is characterized by occurrence of one or more spells of fever after the initial febrile attack. There are two forms of relapsing fever: louse-borne or epidemic relapsing fever caused by B. recurrentis and transmitted by the human body louse, and tick-borne or endemic relapsing fever caused by at least 16 distinctive Borrelia species and transmitted by several species of argasid ticks. Some of the latter Borrelia exhibit complete specificity for their tick vectors. Approximately 20 species of relapsing fever borreliae have been identified to date [3,4].

The genus Borrelia is comprised of two species complexes, B. burgdorferi complex and relapsing fever borreliae, based on genetic differences, specificity of the spirochete–vector relationship, and the type of disease they cause. This genus can be distinguished from other spirochetal phylogenetic groups by base signature analysis of rrs[5].

At least 11 genospecies have been described in the B. burgdorferi s.l. complex. Three of them, B. burgdorferi sensu stricto (s.s.), B. garinii, and B. afzelii, have been isolated from patients. Genetic analysis of ospC and clinical studies indicate that the major ospC groups found in human erythema migrans (EM) lesions are only a subset of the ospC groups found in ticks [6,7] and there is only a minority (10/58) of ospC groups in the three genospecies (B. burgdorferi s.l., B. garinii, and B. afzelii) that are believed to be invasive and able to colonize and persist in organs other than skin. Strains in other groups may cause EM, but do not appear to be invasive and disseminate to other organs.

Three genospecies within the B. burgdorferi s.l. complex have been identified in the southern USA [8–10]. However, strains from the same genospecies sometimes are difficult to differentiate based on restriction fragment length polymorphism (RFLP) analysis of rrf-rrl intergenic spacer. Moreover, rrf-rrl intergenic spacer is only a DNA fragment, and does not reflect the whole genomic situation, nor does the intergenic spacer relate to pathogenicity of Borrelia. Randomly amplified polymorphic DNA (RAPD) fingerprinting analysis is a good genomic indicator and also may provide information related to pathogenicity. Finally, the taxonomic status of some southern strains remains uncertain and needs analysis using other molecular techniques.

RAPD has been used for molecular typing of various microorganisms [11] including B. burgdorferi[12–14]. This technique uses low-stringency polymerase chain reaction (PCR) amplification with a single primer with an arbitrary sequence to generate strain-specific arrays of anonymous DNA fragments. Since RAPD is a highly discriminatory method and easy to perform, it is a powerful tool to distinguish the different Borrelia species from each other as well as to recognize Borrelia strains within each of the species.

In the present study, we use the RAPD method to identify and classify southern Borrelia between and within genospecies, explore the relationship between Borrelia and their tick vectors and mammal reservoirs, and assess the geographic diversity within different Borrelia species. Also, representative strains from each RAPD group were chosen and their rrs genes were sequenced to differentiate spirochetes from B. burgdorferi s.l. and relapsing fever borreliae.

Materials and methods

Spirochete isolates and culture conditions, extraction of spirochete DNA, RAPD, and RAPD fingerprinting analysis

The isolates employed in this study are described in Table 1. Most of them have been characterized previously by rrf-rrl intergenic spacer and ospC[9,10]. Several strains including B-31, SH-2-82 (B. burgdorferi s.l.), DN127, 25015 (B. bissettii), 21038 (B. andersonii), 20047 (B. garinii) were used as positive controls. All southern isolates analyzed were cloned by a range dilution and growth on 1.3% BSK-H agar plates. These strains were third passage. Cultures were incubated at 34°C for 1–2 weeks until the cell density reached about 2×106 cells ml−1.

1

Origins, genospecies, and RAPD types of Borrelia strains analyzed in this study

Genospecies RAPD types Isolate Host Source Location 
 Combined AP13 1283 1254     
B. burgdorferi s.s. A1 B1 C1 B31 I. scapularis  Shelter Island, New York 
 A2 B2 C2 SH-2-82 I. scapularis  New York 
 A3 B3 C3 SI-1 P. gossypinus Bladder Sapelo Island, McIntosh County, Georgia 
 A5 B5 C5 SI-3 S. hispidus Bladder Sapelo Island, McIntosh County, Georgia 
 A6 B6 C6 SI-14 I. affinis female Drag Sapelo Island, McIntosh County, Georgia 
 A6 B7 C6 SCI-2 P. gossypinus Ear clip St. Catherines Island, Liberty County, Georgia 
 A7 B6 C6 SCI-4 I. scapularis male Drag St. Catherines Island, Liberty County, Georgia 
 A4 B4 C4 MI-2 P. gossypinus Bladder, ear clip Merritt Island, Brevard County, Florida 
 A8 B8 C7 SM-1 P. gossypinus Ear clip St. Marys, Camden County, Georgia 
B. bissettii 10 A9 B9 C8 DN127 I. pacificus  California 
 11 A10 B10 C8 25015 I. scapularis larva White-footed mouse Dutchess County, New York 
 12 A11 B11 C9 MI-6 S. hispidus Ear clip Merritt Island, Brevard County, Florida 
 13 A12 B12 C10 MI-8 S. hispidus Ear clip Merritt Island, Brevard County, Florida 
 14 A13 B13 C11 MI-9 P. gossypinus Ear clip Merritt Island, Brevard County, Florida 
 15 A10 B14 C12 SCGT-8a I. minor male Woodrat Georgetown County, South Carolina 
 16 A9 B15 C13 SCGT-10 N. floridana Ear clip Georgetown County, South Carolina 
 17 A14 B16 C14 AI-1 S. hispidus Bladder, ear clip Amelia Island, Nassau County, Florida 
 14 A13 B13 C11 FD-1 S. hispidus Bladder, ear clip Favor Dykes, Flagler County, Florida 
 48 A35 B33 C36 SCW-30h I. minor nymph Carolina wren Wedge Plantation, Charleston County, South Carolina 
B. andersonii 18 A15 B17 C15 21038 I. dentatus  New York 
 19 A16 B18 C16 MOR-1 I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 20 A17 B19 C16 MOR-2 I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 21 A18 B20 C17 MOS-1b I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 21 A18 B20 C17 MOS-1c I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 21 A18 B20 C17 MOS-1d I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 22 A19 B18 C18 MOS-1e I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 23 A20 B20 C17 MOS-1f I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 22 A19 B18 C18 MOS-1g I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 24 A21 B18 C19 MOS-2a I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 25 A18 B21 C20 MOS-2b I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 24 A21 B18 C19 MOS-3a I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 25 A18 B21 C20 MOS-3b I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 26 A18 B20 C21 MOS-3c I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 26 A18 B20 C21 MOS-4 I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 27 A22 B22 C22 MOD-1 2 I. dentatus nymphs Rabbit Dowd Farm, Bollinger County, Missouri 
 28 A22 B20 C23 MOD-3 4 I. dentatus larvae Rabbit Dowd Farm, Bollinger County, Missouri 
 29 A23 B20 C24 MOD-5 I. dentatus nymph Rabbit Dowd Farm, Bollinger County, Missouri 
 30 A24 B20 C25 MOK-1b I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 31 A25 B20 C26 MOK-1c I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 32 A26 B23 C26 MOK-1d I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 33 A27 B24 C27 MOK-1e I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 34 A27 B24 C26 MOK-1f I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 35 A28 B25 C28 MOK-1g I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 36 A25 B26 C27 MOK-2-IDa I. dentatus larva Rabbit Koch Farm, Bollinger County, Missouri 
 37 A29 B27 C27 MOK-2-IDb I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 26 A18 B20 C21 MOK-2-IDc I. dentatus larva Rabbit Koch Farm, Bollinger County, Missouri 
 38 A30 B28 C29 MOK-3a I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 39 A31 B18 C30 MOK-3b I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 40 A17 B25 C16 MOK-3c I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 40 A17 B25 C16 MOK-3d I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 41 A32 B29 C31 MOJ-1 I. dentatus larva Rabbit Jenkins Farm, Bollinger County, Missouri 
 42 A32 B20 C32 MOG-1b I. dentatus nymph Rabbit Gipsy, Wayne County, Missouri 
 43 A33 B30 C33 MON-1 I. dentatus nymph Rabbit NUT Junction, Cape Giradeau County, Missouri 
 44 A28 B18 C25 MOH-1 I. dentatus female Rabbit Happy Farm, Bollinger County, Missouri 
 45 A32 B31 C34 MOH-2 I. dentatus male Rabbit Happy Farm, Bollinger County, Missouri 
 46 A18 B32 C17 BC-1 I. dentatus nymph Drag Macon, Bibb County, Georgia 
 47 A34 B20 C35 SI-10 I. scapularis female Drag Sapelo Island, McIntosh County, Georgia 
B. garinii 49 A36 B34 C37 20047 I. ricinus  France 
Borrelia sp. 50 A37 B35 C38 TXW-1 D. variabilis male Coyote Webb County, Texas 
Genospecies RAPD types Isolate Host Source Location 
 Combined AP13 1283 1254     
B. burgdorferi s.s. A1 B1 C1 B31 I. scapularis  Shelter Island, New York 
 A2 B2 C2 SH-2-82 I. scapularis  New York 
 A3 B3 C3 SI-1 P. gossypinus Bladder Sapelo Island, McIntosh County, Georgia 
 A5 B5 C5 SI-3 S. hispidus Bladder Sapelo Island, McIntosh County, Georgia 
 A6 B6 C6 SI-14 I. affinis female Drag Sapelo Island, McIntosh County, Georgia 
 A6 B7 C6 SCI-2 P. gossypinus Ear clip St. Catherines Island, Liberty County, Georgia 
 A7 B6 C6 SCI-4 I. scapularis male Drag St. Catherines Island, Liberty County, Georgia 
 A4 B4 C4 MI-2 P. gossypinus Bladder, ear clip Merritt Island, Brevard County, Florida 
 A8 B8 C7 SM-1 P. gossypinus Ear clip St. Marys, Camden County, Georgia 
B. bissettii 10 A9 B9 C8 DN127 I. pacificus  California 
 11 A10 B10 C8 25015 I. scapularis larva White-footed mouse Dutchess County, New York 
 12 A11 B11 C9 MI-6 S. hispidus Ear clip Merritt Island, Brevard County, Florida 
 13 A12 B12 C10 MI-8 S. hispidus Ear clip Merritt Island, Brevard County, Florida 
 14 A13 B13 C11 MI-9 P. gossypinus Ear clip Merritt Island, Brevard County, Florida 
 15 A10 B14 C12 SCGT-8a I. minor male Woodrat Georgetown County, South Carolina 
 16 A9 B15 C13 SCGT-10 N. floridana Ear clip Georgetown County, South Carolina 
 17 A14 B16 C14 AI-1 S. hispidus Bladder, ear clip Amelia Island, Nassau County, Florida 
 14 A13 B13 C11 FD-1 S. hispidus Bladder, ear clip Favor Dykes, Flagler County, Florida 
 48 A35 B33 C36 SCW-30h I. minor nymph Carolina wren Wedge Plantation, Charleston County, South Carolina 
B. andersonii 18 A15 B17 C15 21038 I. dentatus  New York 
 19 A16 B18 C16 MOR-1 I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 20 A17 B19 C16 MOR-2 I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 21 A18 B20 C17 MOS-1b I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 21 A18 B20 C17 MOS-1c I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 21 A18 B20 C17 MOS-1d I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 22 A19 B18 C18 MOS-1e I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 23 A20 B20 C17 MOS-1f I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 22 A19 B18 C18 MOS-1g I. dentatus larva Rabbit Swinton, Stoddard County, Missouri 
 24 A21 B18 C19 MOS-2a I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 25 A18 B21 C20 MOS-2b I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 24 A21 B18 C19 MOS-3a I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 25 A18 B21 C20 MOS-3b I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 26 A18 B20 C21 MOS-3c I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 26 A18 B20 C21 MOS-4 I. dentatus nymph Rabbit Swinton, Stoddard County, Missouri 
 27 A22 B22 C22 MOD-1 2 I. dentatus nymphs Rabbit Dowd Farm, Bollinger County, Missouri 
 28 A22 B20 C23 MOD-3 4 I. dentatus larvae Rabbit Dowd Farm, Bollinger County, Missouri 
 29 A23 B20 C24 MOD-5 I. dentatus nymph Rabbit Dowd Farm, Bollinger County, Missouri 
 30 A24 B20 C25 MOK-1b I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 31 A25 B20 C26 MOK-1c I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 32 A26 B23 C26 MOK-1d I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 33 A27 B24 C27 MOK-1e I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 34 A27 B24 C26 MOK-1f I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 35 A28 B25 C28 MOK-1g I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 36 A25 B26 C27 MOK-2-IDa I. dentatus larva Rabbit Koch Farm, Bollinger County, Missouri 
 37 A29 B27 C27 MOK-2-IDb I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 26 A18 B20 C21 MOK-2-IDc I. dentatus larva Rabbit Koch Farm, Bollinger County, Missouri 
 38 A30 B28 C29 MOK-3a I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 39 A31 B18 C30 MOK-3b I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 40 A17 B25 C16 MOK-3c I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 40 A17 B25 C16 MOK-3d I. dentatus nymph Rabbit Koch Farm, Bollinger County, Missouri 
 41 A32 B29 C31 MOJ-1 I. dentatus larva Rabbit Jenkins Farm, Bollinger County, Missouri 
 42 A32 B20 C32 MOG-1b I. dentatus nymph Rabbit Gipsy, Wayne County, Missouri 
 43 A33 B30 C33 MON-1 I. dentatus nymph Rabbit NUT Junction, Cape Giradeau County, Missouri 
 44 A28 B18 C25 MOH-1 I. dentatus female Rabbit Happy Farm, Bollinger County, Missouri 
 45 A32 B31 C34 MOH-2 I. dentatus male Rabbit Happy Farm, Bollinger County, Missouri 
 46 A18 B32 C17 BC-1 I. dentatus nymph Drag Macon, Bibb County, Georgia 
 47 A34 B20 C35 SI-10 I. scapularis female Drag Sapelo Island, McIntosh County, Georgia 
B. garinii 49 A36 B34 C37 20047 I. ricinus  France 
Borrelia sp. 50 A37 B35 C38 TXW-1 D. variabilis male Coyote Webb County, Texas 

Whole-cell DNA was extracted using slight modifications of published procedures [15].

RAPD-PCR reactions were performed using the method described previously with minor modifications [14]. RAPD-PCR reactions were performed in volumes of 50 µl containing 10 mM Tris–HCl (pH 8.8 at 25°C), 50 mM KCl, 4 mM MgCl2, 0.1% Triton® X-100, 200 µM dATP, 200 µM dGTP, 200 µM dCTP, 200 µM dTTP, 2 U Taq DNA polymerase in storage buffer A (Promega), 0.4 µM of each primer and 20 ng of extracted DNA using previously published primers 1254, 1283, and AP13 [14]. Reactions were performed in a GeneAmp® PCR System 9700 (PE Biosystems). Briefly, first at 94°C for 2 min and then three cycles of denaturation at 94°C for 5 min, annealing at 36°C for 5 min, and extension at 72°C for 5 min and then 30 cycles of denaturation at 94°C for 1 min, annealing at 36°C for 1 min, and extension at 72°C for 2 min and a final incubation at 72°C for 10 min.

The amplified DNA fragments were separated in 1% agarose gels containing 0.5 µg ml−1 of ethidium bromide in 0.5×Tris–borate–EDTA. For each experimental run, bacteriophage DNA digested with BstEII was included and used as a size marker for the amplified fragments and as a reference for the normalization of different gels. The gels were photographed using an Eagle Eye II System (Stratagene). The PCR-RAPD patterns were measured and analyzed by using EagleSight® software (version 3.2) as recommended by the manufacturer. Each band on the gel was scored as 1 (present) or 0 (absent). The bands with a faint intensity which were not reproducible in duplicate experiments were excluded in the final analysis. Any clear and neat band which was repeatable in duplicate experiments was scored. The scores for three experiments (one experiment performed with each of the three primers) yielded the data matrix. The banding patterns obtained from three primers were combined. The combined patterns were used for the similarity estimation and phylogenetic analysis. The similarity among strains was estimated by neighbor-joining [16]. Isolates with similarity of more than 0.95 were assigned to the same RAPD type. RAPD patterns were converted into strings of 129 characters, representing all possible migration positions of unambiguous RAPD bands. Strings were analyzed by neighbor-joining [16], maximum likelihood [17], parsimony methods [18], and UPGMA (the unweighted pair group method with arithmetic means analysis) [19] using PAUP (Phylogenetic analysis using parsimony) [18].

Amplification of rrs genes, purification of PCR products, sequence analysis, and phylogenetic analysis

The rrs gene was amplified using previously described primers [20]. PCR reactions were performed in a GeneAmp® PCR System 9700 using conditions described previously [20].

The PCR products were purified using the QIAquick® gel extraction kit (Qiagen) as recommended by the manufacturer. The DNA sequence of the rrs gene was determined using an ABI Prism (model 377). Phylogenetic analysis was performed using PAUP.

Nucleotide sequence accession numbers and retrieved sequences

The rrs nucleotide sequences of 17 B. burgdorferi s.l. isolates were deposited in the GenBank database, and their accession numbers are B. burgdorferi S.S. [MI-2 (AF467964), SI-1 (AF467958), SM-1 (AF467960), SCI-2 (AF467961)], B. bissettii [MI-6 (AF467966), MI-8 (AF467967), MI-9 (AF467968), SCGT-8a (AF467969), SCGT-10 (AF467970), AI-1 (AF467962), FD-1 (AF467963)], B. andersonii [SI-10 (AF467959), BC-1 (AF467973), MOK-3a (AF467974), MOS-1b (AF467975), MOD-5 (AF467972)], Borrelia sp. [TXW-1 (AF467976)].

To compare the relationship between our strains and the strains in different genospecies or species, and to conduct phylogenetic analyses, rrs gene sequences of strains from the following species/genospecies (with strain names and database accession numbers) were used. They included B. burgdorferi S.S. [B31T (U03396)], B. garinii [20047T (D67018)], B. afzelii [DK1 (X85190)], B. bissettii [DN127T (L40596), 25015 (AJ224138)], B. andersonii [21038T (L46701)], B. valaisiana [VS116T (X98232)], B. lusitaniae [Poti B2T (X98228)], B. japonica [HO14T (L40597)], B. tanuki [Hk501T (D67023)], B. turdi [Ya501T (D67022)], B. sinica [CMN2 (AB022145)], B. hermsii [HS1T (U42292)], B. recurrentis [A5 (AF107357)], B. turicatae [M2007 (U42299)], B. parkeri [6232 (AF307100)], B. lonestari [Texas20 (U23211)], B. coriaceae [Co53T (U42286)], B. miyamotoi [HT31 (D45192)], B. duttonii [Ma (AF107366)], B. hispanica [UESV/246 (U42294)], B. crocidurae [UESV/1096TEN (U42302)], B. persica [UESV/340 (U42297)], Borrelia sp. [Spain (U28502), Florida dog strain FCB-1 (L37837), LB2001 (AY024345)].

Results

RAPD results

Repeated RAPD analyses were performed with reference strains B-31, SH-2-82, DN127, 25015, and 21038 in order to assess the reproducibility of our RAPD fingerprinting procedure. We obtained similar DNA fingerprints of these strains in repeated RAPD analysis. In addition, all strains mentioned above were used as positive controls in each run of RAPD amplification. No differences were seen between the RAPD patterns for B31 at different passages (3 and 8) (data not shown).

We scored 129 polymorphic characters and used them to construct a data matrix. Considerable genetic heterogeneity was noted among the isolates. Based on the RAPD profiles in three individual amplifications with different primers, a total of 44 RAPD types were identified from the 53 southern strains (Table 1). Twenty-nine RAPD types were found among 37 B. andersonii strains, seven RAPD types in eight B. bissettii strains, and seven B. burgdorferi s.s. strains yielded seven RAPD types. TXW-1 formed a separate RAPD type (Fig. 1).

1

Neighbor-joining phylogenetic tree based on combined gels after three individual amplifications with different primers. The tree was compared with trees produced by maximum likelihood, parsimony, and UPGMA and the four methods produced similar results. Scale bar=calculated distance value. Bootstrap confidence levels above 50% are indicated to the left of each relevant cluster. Numbers in parentheses to the right of each strain indicate different RAPD types and indistinguishable isolates within each genospecies.

1

Neighbor-joining phylogenetic tree based on combined gels after three individual amplifications with different primers. The tree was compared with trees produced by maximum likelihood, parsimony, and UPGMA and the four methods produced similar results. Scale bar=calculated distance value. Bootstrap confidence levels above 50% are indicated to the left of each relevant cluster. Numbers in parentheses to the right of each strain indicate different RAPD types and indistinguishable isolates within each genospecies.

Phylogenetic trees were constructed based on the combined gels obtained after three individual amplifications with different primers. A general phylogenetic tree derived from combined data arranged the 53 Borrelia strains into four genetic groups: B. andersonii, B. bissettii, B. burgdorferi s.s., and TXW-1. Greater genetic diversity occurred among B. burgdorferi s.l. strains (six clusters/seven RAPD types/seven strains) compared to B. andersonii strains (13 clusters/29 RAPD types/37 strains) and B. bissettii strains (four clusters/seven RAPD types/eight strains) (Fig. 1).

The phylogenetic trees derived from individual amplifications with the three different primers agreed with each other at the genospecies level, but not at the strain level (data not shown but are available upon request). For example, compared to B. burgdorferi S.S. strains SI-14 and SCI-4 and B. bissettii strains MI-6, and MI-9, FD-1, more B. andersonii strains shared common banding patterns in the gels after amplification with primer 1283. Similar results were obtained after two other amplifications with primers 1254 and AP13. Although the phylogenetic trees based on individual amplifications using different primers reflect the phylogenetic relationships among strains at the genospecies level, the tree derived from combined data provides a better understanding at the strain level (Fig. 1).

Analysis of rrs sequences

There is a high level of rrs sequence conservation at the species level among those we examined. Of 1337 bp of rrs sequences that were compared among strains in the 11 species in the B. burgdorferi s.l. complex, 80 variations were found. Most variations are replacements except for six deletions in B. japonica, one deletion in B. turdi, three deletions in B. sinica, and one insertion in B. andersonii strains 21038, BC-1, MOS-1b, and MOD-5. Some variable signature positions are distinct in different species in this complex (data not shown but are available upon request). Compared to B. burgdorferi s.l. complex, more mismatches in nucleotide sequences exist among strains of relapsing fever borreliae. We found 98 variations among the sequences in strains of 16 species. There was one deletion in strain Spain, one insertion in strain Spain, B. persica, and strain LB2000, and 93 mutations among the 16 species. Some mutations were different among distinct species (data not shown but are available upon request).

Nucleotide identities were in the range of 96–100% within the B. burgdorferi s.l. complex, 94–99% within relapsing fever borreliae, and 93–99% between the two species complexes (data not shown but are available upon request). The relapsing fever spirochete TXW-1 segregated with the North American relapsing fever borreliae and formed a separate subbranch (Fig. 2). The similarities of nucleotide sequences of rrs among TXW-1 and B. turicatae and B. parkeri are 99%. The DNA sequences of rrs from different genospecies/species in B. burgdorferi and relapsing fever borreliae complex are highly homologous, ranging from 95.3 to 99.6%[20]. However, 0.4–4.7% variation in rrs sequences is enough to infer the taxa of Borrelia and a value of 99% identity is sufficient to differentiate species (data not shown but are available upon request) [20–26]. TXW-1 is close to, but separate from B. turicatae, B. parkeri, and the isolate from the Florida dog (Fig. 2).

2

Phylogenetic tree derived from 16S rDNA nucleotide sequences of southern and reference strains of B. burgdorferi s.l. and relapsing fever borreliae. The neighbor-joining tree was constructed with PAUP software and based on a comparison of 1537-bp nucleotide sequences of near complete genes. The scale bar represents the calculated distance value. Bootstrap confidence levels above 50% are indicated to the left of each relevant cluster. Boldface strain designations indicate the novel sequences determined in this study.

2

Phylogenetic tree derived from 16S rDNA nucleotide sequences of southern and reference strains of B. burgdorferi s.l. and relapsing fever borreliae. The neighbor-joining tree was constructed with PAUP software and based on a comparison of 1537-bp nucleotide sequences of near complete genes. The scale bar represents the calculated distance value. Bootstrap confidence levels above 50% are indicated to the left of each relevant cluster. Boldface strain designations indicate the novel sequences determined in this study.

All the strains analyzed here can be divided into two parts: the B. burgdorferi s.l. complex and the relapsing fever borreliae complex (Fig. 2). In the former, there are eight branches. The relapsing fever borreliae consist of six branches. The branching of the relapsing fever borreliae in the phylogenetic tree reflects the geographic distribution and vector diversity of these organisms (Fig. 2). TXW-1 forms a separate subbranch. The genetic distance (indicated by branch length) between TXW-1 and B. turicatae (0.002) is not only greater than the genetic distance between B. parkeri and B. turicatae (0.001), but also greater than the genetic distance between some other members in this complex, such as between B. miyamotoi and B. lonestari (0.001) (data not shown but are available upon request).

Discussion

RAPD analysis has been used widely for molecular typing and genetic characterization of various microorganisms [11]. Its reliability and reproducibility have been confirmed by previous study [12–14] and our present results. Previous studies indicated that one strain among 136 isolates analyzed has one band change when different clones were used [14], but RAPD fingerprints from different colonies still showed more than 95% similarity to each other and belonged to the same RAPD type. Whether this difference is caused by a mixed population of two strains or by colonial variation within one strain was not clear at that time [14]. We think the banding pattern change was most likely caused by two mixed strains rather than colonial variation. Genomic sequence data indicated that B. burgdorferi s.l. strain B31 contained a linear chromosome of 910 725 bp and at least 17 linear and circular plasmids with a combined size of more than 533 000 bp [27,28]. Some plasmids might be spontaneously lost during in vitro cultivation. However, previous study and our present experiment indicated that the number of passages did not affect the results of RAPD analysis [14]. The reason is that RAPD targeted a few DNA sequences that may or may not be located in the lost plasmids. All isolates used in this study have been cloned and were the same passage. Results obtained from RAPD or AP-PCR agree with the phylogenetic analysis derived from other molecular methods, such as multilocus enzyme electrophoresis, DNA–DNA hybridization, and RFLP of certain genes [2,12–14]. Data derived from our RAPD analyses agree with data obtained from pulsed-field gel electrophoresis (PFGE), RFLP and sequence analysis of ospC, rrs, and the rrf-rrl intergenic spacer [9,10,29].

Although RFLP and sequence analysis of the rrf-rrl intergenic spacer are efficient methods for identification of B. burgdorferi s.l. [30], our characterizations using these methods with some of the southern strains have not been totally satisfactory. As noted in our previous report [9], the taxonomic status of several strains, including TXW-1 and MI-8, was uncertain and needed further analysis by other techniques. Strains within genospecies were not differentiated based on RFLP and sequence analysis of the rrf-rrl intergenic spacer. RAPD and sequence analysis of the 16S rRNA gene appear to be the appropriate techniques, since both methods can be used to identify strains not only in the B. burgdorferi s.l. complex, but also in the relapsing fever borreliae complex. Previous data indicated that TXW-1 appeared to be closer to the B. garinii genospecies than to other genospecies [9]. The reason that this similarity incorrectly appeared was that not all genospecies in the B. burgdorferi s.l. complex and all species in the relapsing fever borreliae complex were included in our previous analysis. Thus, TXW-1 appeared incorrectly close to B. garinii. Present studies provide evidence of the novelty of TXW-1 and suggest that it might be a novel strain in the relapsing fever borreliae complex. The taxonomic status of another southern strain, MI-8, was uncertain in our previous report [9], but both RAPD and sequence analysis of rrs indicate that this strain is a member of the B. bissettii genospecies.

Results based on RFLP and sequence analysis of the rrf-rrl intergenic spacer did not differentiate strains within genospecies, especially for B. andersonii[9]. Since RAPD is a strain-specific technique, most southern strains were differentiated very well. These data are useful in analysis of the relationship among genospecies of Borrelia and their tick vector, mammal reservoirs, geographic distribution, and pathogenicity.

Because RAPD analysis randomly samples sequence polymorphisms distributed throughout the genome, the fingerprints are accurate indicators of genetic distances and therefore can be used to establish phylogenetic relationships at interspecific and intraspecific levels. Moreover, because RAPD data take the form of discrete characters, it is easy to add new taxa to phylogenetic distance trees. Compared to results obtained by multilocus enzyme electrophoresis and other molecular methods, the RAPD protocol is simple and the number of polymorphisms almost unlimited.

RAPD types are correlated with geographic locations. For example, B. bissettii strains from Florida and South Carolina were separated into different clusters, and B. burgdorferi S.S. strains from South Carolina and Florida were separated from strains isolated in Georgia. Studies based on RAPD and PFGE analyses indicated that B. burgdorferi S.S. strains from North America separated from strains from Europe [13,14].

Previous studies suggested that RAPD types might be used to infer possible pathogenic potential for B. garinii strains, but not for B. afzelii strains [14]. B. garinii strains from patients with disseminated disease were recognized as different from strains from ticks. However, RAPD patterns of B. afzelii strains from ticks and humans were randomly distributed. In addition, strains from patients with Acrodermatitis Chronica Atrophicans (ACA) were not recognized as different from strains from patients with Erythema Migrans (EM) [14]. These results agree with a study based on PFGE [29]. In our present study using RAPD, southern B. burgdorferi S.S. strains formed several branches (Fig. 1). Southern strain MI-2 formed a separate branch among the B. burgdorferi S.S. strains. This agrees with a recent study based on ospC sequence analysis in which MI-2 most closely aligns with a so-called invasive cluster that possesses many proven human-invasive strains. This suggests that MI-2 might be invasive for humans [10]. Transmission experiments showed that this group of southern spirochetes is able to infect mice and hamsters and that Ixodes scapularis can transmit it (Oliver et al., unpublished data). Further studies are needed to determine if they are pathogenic for humans.

Sequence analysis of rrs has been widely used for species identification, phylogenetic and evolutionary analysis of most bacteria including B. burgdorferi s.l. [18,27–31]. At least seven of 11 genospecies were described based on the sequence analysis of 16S rRNA [18,27–31]. Different species or genospecies of Borrelia isolates obtained from diverse geographic regions and various biological sources fall into individual clusters in the phylogenetic tree based on 16S rRNA sequence analysis [20–26]. Therefore, sequence analysis of the 16S rRNA gene was considered as a common method for identification of most bacteria including B. burgdorferi and represents a reliable method for inferring the taxa of B. burgdorferi s.l. [20–26]. Sequence analysis of the 16S rRNA gene provide a useful tool to identify and classify Borrelia, not only B. burgdorferi s.l., but also relapsing fever borreliae. All genospecies/species that we analyzed in the B. burgdorferi s.l. and relapsing fever borreliae complex were differentiated as distinct entities in the phylogenetic tree based on sequence analysis of rrs.

Most relapsing fever borreliae are transmitted by soft-bodied ticks (Ornithodoros or Argas ticks) and one species by the human body louse [3]. In contrast, four recently identified Borrelia, B. miyamotoi[31], B. lonestari[32], an unusual northeastern USA Borrelia[33], and TXW-1 are transmitted by the hard-bodied ticks, Ixodes persulcatus, Amblyomma americanum, I. scapularis, and probably Dermacentor variabilis, respectively. Interestingly, although B. miyamotoi was described from Japan [31], a genetically closely related species was recently reported in the USA [33]. Thus far, B. lonestari and TXW-1 have been found to occur only in the USA. Although human infection by B. lonestari has been reported [32,34], it is unknown whether the other species cause illnesses in humans.

Studies on the relationship between ospC of Lyme disease spirochetes and vmp of relapsing fever borreliae (B. hermsii) suggest that B. burgdorferi and relapsing fever borreliae might have evolved from a common ancestor, but have adapted to different arthropod and vertebrate hosts [35–37]. In general, Lyme disease spirochetes have adapted to different species of hard ticks and relapsing fever borreliae have adapted to distinct species of soft ticks, and one Borrelia species to a louse. There appears to be a rather loose association between different genospecies of B. burgdorferi and certain species of Ixodes ticks, except for B. andersonii and I. dentatus which appears to be quite specific. Compared to Lyme disease spirochetes, relapsing fever borreliae have a stricter specificity for their tick vectors. Nevertheless, our results and other reports suggest that relapsing fever borreliae can be transmitted by hard ticks. Further studies on the diversity of Borrelia and their adaptation to ticks under experimental conditions are needed to better understand possible co-speciation among them.

Acknowledgements

We thank Dr. Mark Schembri, Editor of FEMS Microbiology Letters, and three anonymous reviewers for constructive suggestions to improve the manuscript. This research was supported in part by Merit Grant R37 AI-24899 from the National Institutes of Health and Cooperative Agreement U50/CCU410282 from the Centers for Disease Control and Prevention. The opinions expressed are the responsibility of the authors and do not necessarily represent the official views of the NIH or CDC.

References

[1]
Centers for Disease Control and Prevention
(
2001
)
Lyme disease-United States, 1999
.
Morbid. Mortal. Wkly. Rep.
 
50
,
181
185
.
[2]
Baranton
G.
Postic
D.
Saint Girons
I.
Boerlin
P.
Piffaretti
J.C.
Assous
M.
Grimont
P.A.D.
(
1992
)
Delineation of Borrelia burgdorferi s.l., Borrelia garinii sp. nov., and group VS461 associated with Lyme borreliosis
.
Int. J. Syst. Bacteriol.
 
42
,
378
383
.
[3]
Kelly
R.T.
(
1984
)
Borrelia
. In:
Bergey's Manual of Systematic Bacteriology
  (
Krieg
N.R.
Hott
J.G.
, Eds.), pp.
57
62
.
Williams and Wilkins
,
Baltimore, MD
.
[4]
Barbour
A.G.
Hayes
S.F.
(
1986
)
Biology of Borrelia species
.
Microbiol. Rev.
 
50
,
381
400
.
[5]
Paster
B.J.
Dewhirst
F.E.
Weisburg
W.G.
Tordoff
L.A.
Fraser
G.J.
Hespell
R.B.
Stanton
T.B.
Zablen
L.
Mandelco
L.
Woese
C.R.
(
1991
)
Phylogenetic analysis of the spirochetes
.
J. Bacteriol.
 
173
,
6101
6109
.
[6]
Seinost
G.
Dykhuizen
D.E.
Dattwyler
R.J.
Golde
W.T.
Dunn
J.J.
Wang
I.N.
Wormser
G.P.
Schriefer
M.E.
Luft
B.J.
(
1999
)
Four clones of Borrelia burgdorferi sensu stricto cause invasive infection in humans
.
Infect. Immun.
 
67
,
3518
3524
.
[7]
Baranton
G.
Seinost
G.
Theodore
G.
Postic
D.
Dykhuizen
D.
(
2001
)
Distinct levels of genetic diversity of Borrelia burgdorferi are associated with different aspects of pathogenicity
.
Res. Microbiol.
 
152
,
149
156
.
[8]
Oliver
J.H.
Jr.
Clark
K.L.
Chandler
F.W.J.
Lin
T.
James
A.M.
Banks
C.W.
Huey
L.O.
Banks
A.R.
Williams
D.C.
Durden
L.A.
(
2000
)
Isolation, cultivation, and characterization of Borrelia burgdorferi from rodents and ticks in the Charleston area of South Carolina
.
J. Clin. Microbiol.
 
38
,
120
124
.
[9]
Lin
T.
Oliver
J.H.
Jr.
Gao
L.
Kollars
T.M.
Jr.
Clark
K.L.
(
2001
)
Genetic heterogeneity of Borrelia burgdorferi sensu lato in the southern United States based on restriction fragment length polymorphism and sequence analysis
.
J. Clin. Microbiol.
 
39
,
2500
2507
.
[10]
Lin
T.
Oliver
J.H.
Jr.
Gao
L.
(
2002
)
Genetic diversity of the outer surface protein C gene of southern Borrelia isolates and its possible epidemiological, clinical, and pathogenetic implications
.
J. Clin. Microbiol.
 
40
,
2572
2583
.
[11]
van Belkum
A.
(
1994
)
DNA fingerprinting of medically important microorganisms by use of PCR
.
Clin. Microbiol. Rev.
 
7
,
174
184
.
[12]
Welsh
J.
Pertzman
C.
Postic
D.
Saint Girons
I.
Baranton
G.
McClelland
M.
(
1992
)
Genomic fingerprinting by arbitrarily primed polymerase chain reaction resolves Borrelia burgdorferi into three distinct phyletic groups
.
Int. J. Syst. Bacteriol.
 
42
,
370
377
.
[13]
Foretz
M.
Postic
D.
Baranton
G.
(
1997
)
Phylogenetic analysis of Borrelia burgdorferi sensu stricto by arbitrarily primed PCR and pulsed-field gel electrophoresis
.
Int. J. Syst. Bacteriol.
 
47
,
11
18
.
[14]
Wang
G.Q.
van Dam
A.P.
Spanjaard
L.
Dankert
J.
(
1998
)
Molecular typing of Borrelia burgdorferi sensu lato by randomly amplified polymorphic DNA fingerprinting analysis
.
J. Clin. Microbiol.
 
36
,
768
776
.
[15]
Zingg
B.C.
Brown
R.N.
Lane
R.S.
LeFebvre
R.B.
(
1993
)
Genetic diversity among Borrelia burgdorferi isolates from wood rats and kangaroo rats in California
.
J. Clin. Microbiol.
 
31
,
3109
3114
.
[16]
Saitou
N.
Nei
M.
(
1987
)
The neighbor-joining method: a new method for reconstructing phylogenetic trees
.
Mol. Biol. Evol.
 
4
,
406
425
.
[17]
Felsenstein
J.
(
1989
)
PHYLIP — phylogeny inference package (version 3.2)
.
Cladistics
 
5
,
164
166
.
[18]
Swofford
D.L.
(
2001
)
PAUP: Phylogenetic Analysis using Parsimony (and Other Methods), Version 4
 .
Sinauer Associates
,
Sunderland, MA
.
[19]
Sneath
P.H.A.
Sokal
R.R.
(
1973
)
Numerical Taxonomy
 .
Freeman, San Francisco
,
CA
.
[20]
Le Flèche
A.
Postic
D.
Girardet
K.
Péter
O.
Baranton
G.
(
1997
)
Characterization of Borrelia lusitaniae sp. nov. by 16S ribosomal DNA sequence analysis
.
Int. J. Syst. Bacteriol.
 
47
,
921
925
.
[21]
Marconi
R.T.
Garon
C.F.
(
1992
)
Identification of a third genomic group of Borrelia burgdorferi through signature nucleotide analysis and 16S rRNA sequence determination
.
J. Gen. Microbiol.
 
138
,
533
536
.
[22]
Marconi
R.T.
Liveris
D.
Schwartz
I.
(
1995
)
Identification of novel insertion elements, restriction fragment length polymorphism patterns, and discontinuous 23S rRNA in Lyme disease spirochetes: phylogenetic analyses of rRNA genes and their intergenic spacers in Borrelia japonica sp. nov. and genomic group 21038 (Borrelia andersonii sp. nov.) isolates
.
J. Clin. Microbiol.
 
33
,
2427
2434
.
[23]
Fukunaga
M.
Hamase
A.
Okada
K.
Nakao
M.
(
1996
)
Borrelia tanukii sp. nov. and Borrelia turdae sp. nov. found from ixodid ticks in Japan rapid species identification by 16S rRNA gene-targeted PCR analysis
.
Microbiol. Immunol.
 
40
,
877
881
.
[24]
Fukunaga
M.
Koreki
Y.
(
1996
)
A phylogenetic analysis of Borrelia burgdorferi sensu lato isolates associated with Lyme disease in Japan by flagellin gene sequence determination
.
Int. J. Syst. Bacteriol.
 
46
,
416
421
.
[25]
Wang
G.
van Dam
A.P.
Le Fleche
A.
Postic
D.
Peter
O.
Baranton
G.
de Boer
R.
Spanjaard
L.
Dankert
J.
(
1997
)
Genetic and phenotypic analysis of Borrelia valaisiana sp. nov. (Borrelia genomic groups VS116 and M19)
.
Int. J. Syst. Bacteriol.
 
47
,
926
932
.
[26]
Postic
D.
Ras
N.M.
Lane
R.S.
Hendson
M.
Baranton
G.
(
1998
)
Expanded diversity among California Borrelia isolates and description of Borrelia bissettii sp. nov. (formerly Borrelia group DN127)
.
J. Clin. Microbiol.
 
36
,
3497
3504
.
[27]
Fraser
C.M.
Casjens
S.
Huang
W.
Sutton
G.G.
Clayton
R.
Lathigra
R.
White
O.
Ketchum
K.A.
Dodson
R.
Hickey
E.K.
Gwinn
M.
Dougherty
B.
Tomb
J.-F.
Fleischmann
R.D.
Richardson
D.
Peterson
J.
Kerlavage
A.R.
Quackenbush
J.
Salzberg
S.
Hanson
M.
van Vugt
R.
Palmer
N.
Adams
M.D.
Gocayne
J.
Weidman
J.
Utterback
T.
Watthey
L.
McDonald
L.
Artiach
P.
Bowman
C.
Garland
S.
Fujii
C.
Cotton
M.D.
Horst
K.
Roberts
K.
Hatch
B.
Smith
H.O.
Venter
J.C.
(
1997
)
Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi
.
Nature
 
390
,
580
586
.
[28]
Casjens
S.
Palmer
N.
van Vugt
R.
Huang
W.M.
Stevenson
B.
Rosa
P.
Lathigra
R.
Sutton
G.
Peterson
J.
Dodson
R.J.
Haft
D.
Hickey
E.
Gwinn
M.
White
O.
Fraser
C.M.
(
2000
)
A bacterial genome in flux: the twelve linear and nine circular extrachromosomal DNAs in an infectious isolate of the Lyme disease spirochete, Borrelia burgdorferi
.
Mol. Microbiol.
 
35
,
490
516
.
[29]
Busch
U.
Hizo-Teufel
C.
Böhmer
R.
Fingerle
V.
Roeler
D.
Wilske
B.
Preac-Mursic
V.
(
1996
)
Borrelia burgdorferi sensu lato strains isolated from cutaneous Lyme borreliosis biopsies differentiated by pulsed-field gel electrophoresis
.
Scand. J. Infect. Dis.
 
28
,
583
589
.
[30]
Postic
D.
Assous
M.V.
Grimont
P.A.D.
Baranton
G.
(
1994
)
Diversity of Borrelia burgdorferi sensu lato evidenced by restriction fragment length polymorphism of rrf (5S) rrl (23S) intergenic spacer amplicons
.
Int. J. Syst. Bacteriol.
 
44
,
743
752
.
[31]
Fukunaga
M.
Koreki
Y.
(
1995
)
The flagellin gene of Borrelia miyamotoi sp. nov. and its phylogenetic relationship among Borrelia species
.
FEMS Microbiol. Lett.
 
134
,
255
258
.
[32]
Barbour
A.G.
Maupin
G.O.
Teltow
G.J.
Carter
C.J.
Piesman
J.
(
1996
)
Identification of an uncultivable Borrelia species in the hard tick Amblyomma americanum: possible agent of a Lyme disease-like illness
.
J. Infect. Dis.
 
173
,
403
409
.
[33]
Scoles
G.A.
Papero
M.
Beati
L.
Fish
D.
(
2000
)
A relapsing fever group spirochete transmitted by Ixodes scapularis ticks
.
Vector Borne Zoonotic Dis.
 
1
,
21
34
.
[34]
James
A.M.
Liveris
D.
Wormser
G.P.
Schwartz
I.
Montecalvo
M.A.
Johnson
B.J.B.
(
2001
)
Borrelia lonestari infection after a bite by an Amblyomma americanum tick
.
J. Infect. Dis.
 
183
,
1810
1814
.
[35]
Marconi
R.T.
Samuels
D.S.
Schwan
T.G.
Garon
C.F.
(
1993
)
Identification of a protein in several Borrelia species which is related to OspC of the Lyme disease spirochetes
.
J. Clin. Microbiol.
 
31
,
2577
2583
.
[36]
Carter
C.J.
Bergstrom
S.
Norris
S.J.
Barbour
A.G.
(
1994
)
A family of surface-exposed proteins of 20 kilodaltons in the genus Borrelia
.
Infect. Immun.
 
62
,
2792
2799
.
[37]
Margolis
N.
Hogan
D.
Cieplak
W.
Jr.
Schwan
T.G.
Rosa
P.A.
(
1994
)
Homology between Borrelia burgdorferi OspC and members of the family of Borrelia hermsii variable major proteins
.
Gene
 
143
,
105
110
.