Abstract
Tularemia is a potentially fatal disease that is caused by the highly infectious and zoonotic pathogen Francisella tularensis. Despite the monomorphic nature of sequenced F. tularensis genomes, there is a significant degree of plasticity in the organization of genetic elements. The observed variability in these genomes is due primarily to the transposition of direct repeats and insertion sequence (IS) elements. Since current methods used to genotype F. tularensis are time-consuming and require extensive laboratory resources, IS elements were investigated as a means to subtype this organism. The unique spatial location of specific IS elements provided the basis for the development of a differential IS amplification (DISA) assay to detect and distinguish the more virulent F. tularensis subsp. tularensis (subtypes A.I and A.II) and subsp. holarctica (type B) strains from F. tularensis subsp. novicida and other near neighbors, including Francisella philomiragia and Francisella-like endosymbionts found in ticks. Amplicon sizes and sequences derived from DISA showed heterogeneity within members of the subtype A.I and A.II isolates but not the type B strains. These differences were due to a 312-bp fragment derived from the IS element ISFtu1. Analysis of wild-type F. tularensis isolates by DISA correlated with pulsed-field gel electrophoresis genotyping utilizing two different restriction endonucleases and provided rapid results with minimal sample processing. The applicability of this molecular typing assay for environmental studies was demonstrated by the accurate identification and differentiation of tick-borne F. tularensis. The described approach to IS targeting and amplification provides new capability for epidemiological investigations and characterizations of tularemia source outbreaks.
INTRODUCTION
Francisella tularensis is a facultative intracellular bacterium and the causative agent of the zoonotic disease tularemia (10). This Gram-negative microbe is highly infectious, with as few as 10 organisms being capable of causing disease in numerous mammalian species, including humans. Accordingly, F. tularensis is considered a potential biological weapon and is classified by the Centers for Disease Control and Prevention (CDC) as a select-agent pathogen (29). The three widely accepted subspecies of F. tularensis include F. tularensis subsp. tularensis (also known as type A), F. tularensis subsp. holarctica (also known as type B), and F. tularensis subsp. mediasiatica, whereas classification of the fourth subspecies, F. tularensis subsp. novicida, remains in contention (13, 14, 20). Differentiation among these subspecies is important due to their variation in virulence and for the tracking of tularemia outbreaks. Although F. tularensis subsp. mediasiatica was reported to exhibit a moderate degree of virulence in mammals, the geographic distribution in Central Asia and limited accessibility to isolates have restricted the knowledge of this subspecies' pathogenicity (4). F. tularensis subsp. novicida is generally considered an opportunistic microbe, whereas F. tularensis subsp. tularensis and F. tularensis subsp. holarctica are capable of causing a severe form of tularemia. Moreover, type A strains (F. tularensis subsp. tularensis) can potentially cause a life-threatening illness. The virulent type A and type B clades show remarkable geographic associations: F. tularensis subsp. holarctica is prevalent throughout the Northern Hemisphere, while F. tularensis subsp. tularensis is restricted primarily to North America (25). Recently, subtypes A.I and A.II of F. tularensis subsp. tularensis have been correlated with certain regions in North America and differences in virulence, with subtype A.I infection being reported to be more often fatal (18). Therefore, the rapid detection and typing of F. tularensis are critical for epidemiological tracking, clinical disease management, and biodefense monitoring.
Numerous molecular studies have been evaluated for the detection and identification of F. tularensis from biological reservoirs such as lagomorphs and ticks. However, amplification of the 16S rRNA gene and other targets has produced misleading results due to their cross-reactivity with near neighbors and Francisella-like endosymbionts known to exist in ticks (3, 11, 17). In general, examination of the population structure of F. tularensis at the genetic level has required more robust and time-consuming approaches, such as pulsed-field gel electrophoresis (PFGE), multilocus variable-number tandem-repeat analysis, and whole-genome detection of single-nucleotide polymorphisms (SNP). The value of these studies has been significant. For example, phylogenetic analyses of repeats and SNP content suggested that F. tularensis subsp. holarctica originated from North America and was introduced multiple times into Eurasia (15, 33). Furthermore, comparative genome hybridization and sequencing were used to demonstrate a unique deletion present within an F. tularensis isolate found in an Iberian Peninsula tularemia outbreak (2, 7, 8).
Other genetic markers with the potential for use in molecular typing have been identified. Seventeen contiguous regions were previously identified in F. tularensis subsp. holarctica that were dispersed in F. tularensis subsp. tularensis using paired-end sequencing (8). A recent genomic comparison between four F. tularensis subsp. tularensis isolates (SCHU S4, FSC 198, NE-061598, and FSC 033) confirmed that these subtype A.I strains are clonal, with a high degree of chromosomal synteny (23). Nevertheless, distinctions among these strains could be made based on inversions and rearrangements that were predictably mediated by direct repeats and insertion sequence (IS) elements (23). Additional computational analysis suggested that several of the contiguous regions could be used to differentiate between type A and type B strains. However, since few wild-type isolates have been sequenced, the application of this approach to characterize new foci of tularemia outbreaks was unknown. To address this challenge, a molecular assay was developed using differential insertion sequence amplification (DISA) to characterize wild-type F. tularensis isolates, and results were compared to the typing patterns obtained with PFGE using two different restriction endonucleases.
MATERIALS AND METHODS
Bacterial strains.
Eighty-eight wild-type and four reference strains of F. tularensis were evaluated, as shown in Fig. 2 and 3. The wild-type isolates were obtained from multiple sources, including public health and diagnostic laboratories in the United States. The identities of all F. tularensis isolates were confirmed according to the testing protocols utilized by Laboratory Response Network reference laboratories. The reference F. tularensis strains included SCHU S4 (subtype A.I), WY96-3418 (subtype A.II), the live vaccine strain (LVS) (type B), and F. tularensis subsp. novicida (Table 1). Twenty-three non-Francisella species representing 21 different species as well as one Francisella philomiragia strain were included in an exclusivity panel to assess specificity (Table 1). Select-agent strains were transferred to the University of Nebraska Medical Center, Omaha, NE, according to the requirements of the Select Agent Program as outlined in the Animal and Plant Health Inspection Service/CDC form 2, guidance document for request to transfer select agents and toxins (5). The manipulation of viable culture material was performed by authorized individuals within a biosafety level 3 laboratory certified for select-agent work by the U.S. Department of Health and Human Services, using laboratory biosafety criteria as described by the National Institutes of Health and the CDC (9). All Francisella isolates were grown on chocolate agar plates (Remel, Lenexa, KS) and incubated at 37°C with 5% CO2 for 3 days before processing.
Fig. 2.
Dendrogram of PFGE patterns obtained with PmeI-digested F. tularensis subsp. tularensis subtype A.I and the apparent clade A.I.a and A.I.b subpopulations (A) as well as F. tularensis subsp. tularensis subtype A.II and F. tularensis subsp. holarctica type B (B). A representative PmeI-digested F. tularensis A.I strain from each of the five clusters is shown in the upper part of B for comparison. Distance is displayed in relative units.
Fig. 3.
Dendrogram of PFGE patterns obtained with BamHI-digested F. tularensis subsp. tularensis subtype A.I (A) as well as F. tularensis subsp. tularensis subtype A.II and F. tularensis subsp. holarctica type B (B). A representative BamHI-digested F. tularensis A.I strain from each of the six clusters is shown in the upper part of B for comparison. Distance is displayed in relative units.
Table 1.
Reference strains of Francisella and non-Francisella strains used in this study
| Species | Strain designation(s)e |
|---|---|
| Francisella tularensis subsp. tularensisa | SCHU S4, NR-643b |
| Francisella tularensis subsp. tularensisa | WY96-3418, NR-644b |
| Francisella tularensis subsp. holarctica | ATCC 29684, LVS, NR-14b |
| Francisella tularensis subsp. novicidaa | ATCC 15482, Utah 112, NR-13b |
| Francisella philomiragia | ATCC 25015 |
| Acinetobacter baumannii | Wild typec |
| Bacillus atrophaeusd | ATCC 9372 |
| Brucella canis | Wild typec |
| Burkholderia cepacia | ATCC 25608 |
| Corynebacterium pseudodiphtheriticum | ATCC 10701 |
| Enterobacter cloacae | ATCC 13047 |
| Enterococcus faecalis | ATCC 29212 |
| Escherichia coli | ATCC 35218 |
| Escherichia coli | Wild typec |
| Haemophilus influenzae | ATCC 10211 |
| Klebsiella pneumoniae | ATCC BAA-1705 |
| Micrococcus luteus | ATCC 49732 |
| Neisseria lactamica | ATCC 23970 |
| Proteus vulgaris | ATCC 49132 |
| Pseudomonas aeruginosa | ATCC 10145 |
| Serratia marcescens | Wild typec |
| Staphylococcus aureus | ATCC 25923 |
| Staphylococcus aureus | Wild typec |
| Staphylococcus epidermidis | ATCC 14990 |
| Streptococcus pneumoniae | ATCC 6305 |
| Streptococcus pyogenes | ATCC 12384 |
| Yersinia pestisa | ATCC 19428, NR-18b |
| Wolbachia persica | ATCC VR-331 |
Received through the submission of APHIS/CDC form 2, request to transfer select agents and toxins.
Strain designation from the NIH Biodefense and Emerging Infections Research Resources Repository (BEI Resources).
Wild-type strain from the University of Nebraska Medical Center stock collection.
Previously known at Bacillus subtilis var. niger.
Abbreviations: LVS, live vaccine strain; ATCC, American Type Culture Collection.
DNA isolation.
The extraction of chromosomal DNA from all bacterial species was performed by using CTAB (cetyltrimethyl ammonium bromide) according to standard procedures (34). For the environmental samples, ticks and insects were homogenized in cold phosphate-buffered saline (PBS) within containment by using a Mini Beadbeater-8 instrument (Bio Spec Products, Inc., Bartlesville, OK) according to the manufacturer's recommendations. DNA was then isolated from the arthropod and insect extracts by using the QIAamp DNA blood minikit (Qiagen, Valencia, CA). Pond water and soil samples were subjected to DNA extraction, as was performed for the tick extracts but without homogenization. The DNA from all environmental samples was concentrated following extraction, and approximately 1 ng of DNA was utilized for the PCR assays. If an inhibitor was detected, the DNA was diluted, and 0.1 ng of DNA was used instead. Dermacentor and Amblyomma ticks were collected from canine or bovine sources as well as by flagging or dragging and were identified to species level according to previously reported keys (16, 35).
PFGE.
Agarose-embedded chromosomal DNA for PFGE was prepared and digested with the restriction endonucleases PmeI and BamHI (Fermentas, Inc., Glen Burnie, MD) as previously described (12). Agarose gels (1%, wt/vol) contained Francisella DNA digested with PmeI or BamHI and were run for 17.5 h or 16.5 h, respectively, with an initial switch time of 1.8 s and a final switch time of 10.7 s. Migration profiles of the restriction fragments were normalized to SmaI-digested Staphylococcus aureus NCTC 8325 by using Bionumerics software (Applied Maths, Inc., Austin, TX). Cluster analysis was performed by using the Dice correlation coefficient and the unweighted-pair group mathematical average (UPGMA) clustering algorithm.
PCR amplifications.
The continuous region 10 (CR10) primer set consisted of the oligonucleotides 5′-GCTGATGATTCTAGCCTTAAAGAAG-3′ (CR10 C), 5′-GCCAATCCTACATTTATAGAACCTG-3′ (CR10 S), and 5′-GCCTGGCATAATTACTGTTTTAGC-3′ (CR10 L). The CR16 primer set was comprised of the primers 5′-CTTGTCAGAGTTGGAGTGAAGC-3′ (CR16 C), 5′-CCACTACCTCGAATCTTACACAAAG-3′ (CR16 S), and 5′-GTCAAATCACTTTCCTCTCGTTC-3′ (CR16 L). PCR amplifications were performed by using the indicated CR primers and Platinum DNA polymerase (Invitrogen, Carlsbad, CA) as recommended by the manufacturer. The cycling parameters were 1 cycle for 2.5 min at 95°C and 40 cycles for 30 s at 95°C, 1 min at 60°C, and 3 min at 72°C, followed by 1 cycle for 5 min at 72°C. PCR products were resolved in a 0.7% (wt/vol) agarose gel, stained with ethidium bromide, and visualized with UV light. Representative amplicons were cloned into pCR4-TOPO (Invitrogen, Carlsbad, CA) and sequenced by using M13 forward and reverse primers by the University of Nebraska Medical Center DNA Sequencing Facility.
RESULTS
PCR and chromosomal mapping of amplicons derived from control strains.
Computational analysis of the different continuous regions (CRs) previously described (8) suggested that two loci, namely, CR10 and CR16, could be used to differentiate between F. tularensis type A and type B strains. The CR10 and CR16 primers were based on chromosomal content within these two loci that was common (C) to both SCHU S4 and LVS or specific to either SCHU S4 (S) or LVS (L). The location of the CR primers and the predicted amplicon size were initially determined by in silico analysis for all Francisella species and subspecies in which the genome sequence was available, including F. tularensis subsp. mediasiatica. PCR evaluations for each oligonucleotide pair in the CR10 and CR16 primer sets (C, S, and L) were then performed by using three well-characterized and sequenced strains of F. tularensis, specifically SCHU S4 (GenBank accession no. AJ749949; subtype A.I), WY96-3418 (here referred to as WY96) (GenBank accession no. CP000608; subtype A.II), and LVS (GenBank accession no. AM233362; type B).
Preliminary PCR assessments revealed that primer pair CR10 C and CR10 S and primer pair CR16 C and CR16 S were specific for SCHU S4, generating 1.2-kb and 2.3-kb PCR products, respectively. Conversely, primer pair CR10 C and CR10 L and primer pair CR16 C and CR16 L amplified a single PCR product that differed in size for WY96 (1.3 kb and 1.7 kb, respectively) and LVS (2.1 kb and 0.8 kb, respectively). The amplicon sizes obtained with the CR10 and CR16 primers for these three reference strains correlated with the expected lengths based on the nucleotide sequences deposited in the GenBank database (Fig. 1). These results confirmed the in silico predictions for all F. tularensis subspecies, except F. tularensis subsp. mediasiatica, for which no DNA was available.
Fig. 1.
Chromosomal map of CR10 and CR16 primer pair locations in F. tularensis subsp. tularensis subtypes A.I and A.II and F. tularensis subsp. holarctica type B. The chromosomal regions amplified with the CR10 (A) and CR16 (B) primer sets from the sequenced reference F. tularensis strains are shown. SCHU S4 (GenBank accession no. AJ749949), WY96-3418 (WY96) (GenBank accession no. CP000608), and LVS (GenBank accession no. AM233362) represent subtype A.I, subtype A.II, and type B strains, respectively. The resulting amplicon size is indicated below the appropriate strain in parentheses. The forward and reverse CR (C, S, or L) primers are denoted by black arrows. The corresponding genes in the various strains for each panel are indicated by the same color with the locus identifier number. The beginning and ending nucleotide positions for the mapped loci are denoted below the respective chromosomal regions. Dashed lines designate relative locations and organizations of the corresponding nucleotide sequences in the F. tularensis genomes.
The PCR products obtained for SCHU S4 with primer pair CR10 C and CR10 S and for WY96 and LVS with primer pair CR10 C and CR10 L contained the IS element ISFtu1 (Fig. 1A). However, an approximately 300-bp region of the ISFtu1 gene in SCHU S4 was apparently deleted, unlike the presumably intact ISFtu1 in WY96 and LVS. The absence of deoD in the SCHU S4 CR10-derived amplicon, the truncated ISFtu1 gene, and the nucleotide differences in primers CR10 S and CR10 L indicated that a unique SCHU S4 chromosomal region was amplified compared to the PCR products obtained with WY96 and LVS. Furthermore, an additional IS element, specifically, ISFtu2, was present in the amplicon derived from LVS with CR10 C and CR10 L but absent in the related WY96 PCR product. These findings verify that this chromosomal region differs between subtype A.II and type B strains even when the same CR10 primers are utilized (Fig. 1A).
For the amplicons derived from primer pairs CR16 C and CR16 S as well as CR16 C and CR16 L, ISFtu1 was present in SCHU S4 and WY96, respectively, but absent in LVS (Fig. 1B). The smaller PCR product obtained for LVS with primers CR16 C and CR16 L reflects the absence of this IS element. When the CR16-derived chromosomal regions in SCHU S4 and WY96 were compared, ISFtu1 and the adjacent fnuDI gene were inverted, with an apparent duplication of fnuDI in the latter strain. Since LVS is an attenuated strain, the chromosomal contents of the amplified CR10 and CR16 regions were examined in all available F. tularensis subsp. holarctica genomes. These assessments indicated that the targeted regions shared considerable nucleotide sequence identity in type B isolates. Collectively, the chromosomal mapping of the CR10 and CR16 primers in the three control strains concurs with the amplicon lengths obtained and confirms that the amplified regions differ in contents among the F. tularensis subtype A.I, subtype A.II, and type B strains.
PFGE genotyping.
PFGE analysis was shown previously to type and subtype F. tularensis (12, 31) and was used as the reference methodology to evaluate a collection of 88 wild-type and 4 well-characterized F. tularensis isolates. PFGE was performed by using two different restriction endonucleases (PmeI and BamHI). In these PFGE analyses, F. tularensis strains SCHU S4, WY96, and LVS were used as subtype A.I, subtype A.II, and type B controls, respectively. Overall, PFGE analyses of PmeI-digested (Fig. 2) and BamHI-digested (Fig. 3) F. tularensis chromosomal DNA were in agreement, identifying 56 type A.I, 15 type A.II, and 21 type B isolates. PFGE using PmeI segregated the subtype A.I isolates largely into two groups, clades A.I.a and A.I.b, as was previously defined by Kugeler and associates (18). However, PFGE analysis utilizing BamHI did not group the subtype A.I isolates into the same clade A.I.a and A.I.b subgroups. A detailed analysis of the clustering patterns obtained with PmeI and BamHI identified five and six subpopulations in the subtype A.I strains, respectively. The PFGE patterns of the 15 subtype A.II isolates were more divergent than those of the subtype A.I strains, as previously reported (18). PFGE analysis with both PmeI and BamHI indicated the presence of six discernible subtype A.II groups, whereas all 21 type B isolates were indistinguishable. For F. tularensis subsp. novicida and F. philomiragia, PFGE patterns obtained with PmeI and BamHI indicated that they were both outliers and distantly related to F. tularensis type A and type B strains (data not shown). Together, these data validate the PFGE patterns obtained with the use of either PmeI or BamHI to identify the F. tularensis subspecies or subtype.
Subspecies identification and subtyping by PCR.
The CR10 and CR16 primers were utilized in a PCR to type a collection of wild-type F. tularensis isolates. Typing results obtained with the CR-derived amplicons were compared to results of PFGE genotyping with PmeI and BamHI. These results are summarized in Table 2. The CR-derived amplification products differentiated the F. tularensis strains into types and subtypes that corroborated the results of genotyping by PFGE.
Table 2.
Comparison of typing results obtained by PCR with the CR primers and by PFGE with PmeI or BamHI for F. tularensis subsp. tularensis (subtypes A.I and A.II), F. tularensis subsp. holarctica (type B), and F. tularensis subsp. novicida isolates
| No. of F. tularensis isolates | CR primer pair | Amplicon size (kb) | Type determined by: |
||
|---|---|---|---|---|---|
| PCR-based typing with CR primers | PFGE-based typing with PmeI | PFGE-based typing with BamHI | |||
| 43a | CR10 C + CR10 S | 1.2 | A.I | A.I | A.I |
| 14 | CR10 C + CR10 S | 1.5 | A.I | A.I | A.I |
| 54a | CR16 C + CR16 S | 2.3 | A.I | A.I | A.I |
| 3 | CR16 C + CR16 S | 2.0 | A.I | A.I | A.I |
| 15b | CR10 C + CR10 L | 1.3 | A.II | A.II | A.II |
| 14b | CR16 C + CR16 L | 1.7 | A.II | A.II | A.II |
| 1 | CR16 C + CR16 L | 1.4 | A.II | A.II | A.II |
| 21c | CR10 C + CR10 L | 2.1 | B | B | B |
| 21c | CR16 C + CR16 L | 0.8 | B | B | B |
| 1d | CR10 C + CR10 L | 0.4 | F. tularensis subsp. novicida | F. tularensis subsp. novicida | F. tularensis subsp. novicida |
| 1d | CR10 C + CR10 S | ||||
| 1d | CR16 C + CR16 L | ||||
| 1d | CR16 C + CR16 S | ||||
Includes F. tularensis subsp. tularensis strain SCHU S4 (subtype A.I).
Includes F. tularensis subsp. tularensis strain WY96-3418 (subtype A.II).
Includes F. tularensis subsp. holarctica LVS (type B).
F. tularensis subsp. novicida Utah 112 (ATCC 15482).
The majority of subtype A.I isolates analyzed with primers CR10 C and CR10 S generated PCR products that were 1.2 kb in size (43 strains). An amplicon of this length was also observed for the subtype A.I prototype strain, SCHU S4. The remainder of the subtype A.I isolates (14 strains) produced a 1.5-kb amplicon, indicating that this locus differed by approximately 300 bp from the other subtype A.I strains (Table 2). Primer pair CR10 C and CR10 L produced a 1.3-kb amplicon for all of the subtype A.II isolates and a 2.1-kb amplicon for all of the type B strains (Table 2). For F. tularensis subsp. novicida, a 0.4-kb PCR product was produced with primers CR10 C and CR10 L, and no amplicon was observed with primer pair CR10 C and CR10 S.
The utilization of primers CR16 C and CR16 S produced either a 2.3-kb (54 strains) or a 2.0-kb (3 strains) amplicon for the F. tularensis subtype A.I isolates, with the larger PCR product being amplified from the SCHU S4 genome (Table 2). For the subtype A.II strains, primers CR16 C and CR16 L produced either a 1.7-kb or a 1.4-kb PCR product, with the generation of the larger 1.7-kb amplicon from WY96 (Table 2). All type B isolates produced a 0.8-kb amplicon with primers CR16 C and CR16 L (Table 2). No amplicons were obtained with either primer pair CR16 C and CR16 S or CR16 C and CR16 L for F. tularensis subsp. novicida.
Overall, these results demonstrated that the same PCR product profile was generated with all four CR primer pairs for both the reference and wild-type strains of the same subtype and/or type. To reduce the number of PCR samples and further test the capability of the DISA assay, the CR10 and CR16 primer pairs were reduced to only two sets of primers, with each set containing the three associated primers (CR10 C, CR10 S, and CR10 L as well as CR16 C, CR16 S, and CR16 L). These tests produced the same PCR product profiles as those of the four primer pairs but with half as many reactions.
Amplification specificity and sensitivity of CR primers.
The CR primer specificity for F. tularensis was evaluated by using an exclusivity panel (n = 24) (Table 1) that included chromosomal DNA from the near neighbors F. philomiragia and Wolbachia persica. Neither the CR10 nor CR16 primer pairs resulted in any prominent PCR products for these bacterial samples, with the exception of W. persica, in which a 250-bp amplicon was generated by use of primers CR10 C and CR10 L.
The applicability of PCR amplification with the CR primers in environmental studies to detect the presence of F. tularensis was examined by using arthropods, pond water, and soil. Fifty-four ticks, including Dermacentor andersoni, Dermacentor variabilis, and Amblyomma americanum, species known to transmit F. tularensis in North America (26), were assessed for the presence of F. tularensis.
PCR with the CR primers revealed that one D. andersoni tick and one D. variabilis tick produced products consistent with the presence of F. tularensis, whereas the remainder of the ticks did not produce any amplicons. Importantly, the PCR product sizes obtained with the CR primers provided subspecies and subtype details. Primer pair CR10 C and CR10 L and primer pair CR16 C and CR16 L produced amplicons indicating the presence of F. tularensis subsp. tularensis subtype A.II in the D. andersoni tick. Amplification products derived with primer pair CR10 C and CR10 L and primer pair CR16 C and CR16 L and DNA isolated from the D. variabilis tick were characteristic of F. tularensis subsp. holarctica (type B). The amplicons derived from these two ticks were sequenced, and the data corresponded to the appropriate F. tularensis subtype and/or type as determined by the DISA assay. Results from two commercially available PCR-based test kits (Idaho Technology, Inc., Salt Lake City, UT) that detect F. tularensis were also positive using DNA extracts from these two ticks. However, the commercial assays could not type the F. tularensis isolates. No viable F. tularensis was recovered from these two tick specimens by culturing on chocolate agar plates, and no other environmental samples were positive with either the commercial or CR primer pairs. Together, these data demonstrated that the CR primers were specific for F. tularensis and capable of detection and subspecies or subtype identification for both the clinical and environmental isolates examined.
As a control to verify the presence of Francisella-like endosymbionts in the ticks, PCR amplifications were performed by using primers known to amplify target genes in Francisella-like bacteria (3, 11, 24, 30). The Francisella-like endosymbiont W. persica served as a positive control. In addition, commercially available test kits (Idaho Technology, Inc.) that detect only F. tularensis subspecies were utilized in these assessments. Ticks that produced PCR products with the primers designed to amplify Francisella-like bacteria were further evaluated with the commercial F. tularensis-specific assays as well as with the DISA assay. If no amplicons were produced with the F. tularensis-specific assays but PCR products were produced with the primers for Francisella-like bacteria, the respective amplicons were sequenced. The subsequent nucleotide sequences were compared to those deposited in the databases, confirming the presence of Francisella-like endosymbionts and the absence of F. tularensis or F. philomiragia in the associated ticks. These findings demonstrated that the CR primers did not amplify DNA from Francisella-like endosymbionts known to exist in ticks.
The limit of detection for the PCR amplification of F. tularensis with the CR primers under standard PCR conditions described in Materials and Methods was determined to be approximately 20 copies of the 1.9-Mbp F. tularensis chromosome. Therefore, PCR with the CR primers provided a detection of F. tularensis at a level naturally present in environmental sources and achieved an identification of the subspecies or subtype.
Sequence analysis of amplicons.
Representative PCR products obtained with the CR primers for each size group were cloned and sequenced to determine content. The DNA sequences of the PCR products generated with the CR10 and CR16 primers for these strains were in agreement with the expected lengths (Table 2) and mapped loci (Fig. 1). The size difference observed for the 1.2-kb and 1.5-kb amplicons in the F. tularensis subtype A.I clade obtained with primers CR10 C and CR10 S was due to a 312-bp deletion in the smaller PCR product (Table 3).
Table 3.
Sequence analysis of representative amplicons obtained by PCR with the CR10 primer set relative to F. tularensis SCHU S4, WY96-3418, or LVSa
| F. tularensis isolate | Type | CR10 primer pair | Amplicon size (kb) | Insertion | SNPb |
|---|---|---|---|---|---|
| SCHU S4 | A.I | C + S | 1.2 | ||
| NE-061598 | A.I | C + S | 1.2 | ||
| NC-52797-99 | A.I | C + S | 1.2 | nt 896/A→T; nt 955/T→C | |
| NE-071905 | A.I | C + S | 1.5 | 312 bp at nt 539-850c | nt 512/T→C |
| WY-0705409 | A.I | C + S | 1.5 | 312 bp at nt 539-850c | nt 1135/A→G; nt 1244/T→C |
| WY96-3418 | A.II | C + L | 1.3 | ||
| WY-00W4114 | A.II | C + L | 1.3 | nt 755/T→C | |
| WY-HumanNC | A.II | C + L | 1.3 | ||
| LVS | B | C + L | 2.1 |
With primer pair CR16 C and CR16 S, either a 2.3-kb or a 2.0-kb amplicon was generated for the subtype A.I strains, while primers CR16 C and CR16 L resulted in either a 1.7-kb or a 1.4-kb PCR product for the subtype A.II isolates. In both cases, the smaller amplification product was again due to a 312-bp deletion (Table 4). The 312-bp deleted region in the smaller CR10- and CR16-derived amplicons correlated with the middle region of the ISFtu1 transposase gene. Single-nucleotide polymorphisms (SNPs) were also noted for some of the CR-derived PCR products, as indicated in Tables 3 and 4.
Table 4.
Sequence analysis of representative amplicons obtained by PCR with the CR16 primer set relative to F. tularensis SCHU S4, WY96-3418, or LVSa
| F. tularensis isolate | Type | CR16 primer pair | Amplicon size (kb) | Insertion | Deletion | SNPc | |
|---|---|---|---|---|---|---|---|
| SCHU S4 | A.I | C + S | 2.3 | ||||
| NE-061598 | A.I | C + S | 2.3 | ||||
| WY-07F10908 | A.I | C + S | 2.0 | T at nt 1193 | 312 bp at nt 1282-1593b | nt 1202/T→C | |
| NE-052608 | A.I | C + S | 2.0 | 312 bp at nt 1282-1593b | |||
| WY-07F8058 | A.I | C + S | 2.0 | T at nt 1193 | 312 bp at nt 1282-1593b | nt 1202/T→C | |
| WY96-3418 | A.II | C + L | 1.7 | ||||
| WY-00W4114 | A.II | C + L | 1.7 | ||||
| WY-HumanNC | A.II | C + L | 1.7 | ||||
| WY-07F13554 | A.II | C + L | 1.4 | 312 bp at nt 691-1002b | nt 567/C→T; nt 643/A→G; nt 1126/G→C; nt 1228/T→C; nt 1389/T→C; nt 1650/T→C | ||
| LVS | B | C + L | 0.8 | ||||
| WY-05W8084 | B | C + L | 0.8 | ||||
| UT-70101201 | B | C + L | 0.8 | nt 290/A→T |
DISCUSSION
As with other virulent pathogens, such as Bacillus anthracis and Yersinia pestis, F. tularensis is highly monomorphic, with greater than 99% sequence similarity between type A and type B strains (27, 28). Genetically monomorphic pathogens may have recently passed through a population bottleneck, reducing or abolishing genetic diversity (1). Despite the considerable conservation of overall genomic content among F. tularensis subspecies, the organizations of discrete genetic elements differ (8). The observed translocations and rearrangements appear to be mediated by the movement of IS elements, most notably by the highly abundant ISFtu1 gene (27). Seven different types of IS elements have been found within SCHU S4 and the closely related subtype A.I strain NE-061598, including ISFtu1 (50 copies); ISFtu2 (16 copies); ISFtu3 and ISFtu6 (3 copies each); and ISFtu4, ISFtu5, and ISSod13 (1 copy each) (23). IS elements appear to be the primary driving force for the evolution of the Francisella chromosome, contributing to genome decay and subsequent intracellular dependence.
Although there are similar numbers of ISFtu1 genes in F. tularensis subsp. tularensis (type A) and F. tularensis subsp. holarctica (type B) and many more copies of ISFtu2 in type B than in type A strains, genomic rearrangements in type B strains are much less frequent (32). The low level of genetic diversity and high frequency of transposable elements in the type B clade support the hypothesis that F. tularensis subsp. holarctica emerged after F. tularensis subsp. tularensis and, therefore, is more clonal. In addition, the wider geographic dispersion of F. tularensis subsp. holarctica supports the correlation between this pathogen either having a higher capacity to survive in different environmental niches than F. tularensis subsp. tularensis or causing less morbidity in the associated host. The availability of a rapid PCR-based method for examining additional wild-type isolates may provide new insights into the mechanism for the observed geographic associations between these two subspecies.
Despite the subtype A.I and A.II strains having similar numbers of IS elements (6, 20), there is considerable overall variation in chromosome architectures (18, 23). The current study revealed that a 312-bp insertion or deletion derived from ISFtu1 was responsible for the size heterogeneity within the CR-derived amplicons in the subtype A.I and A.II strains but that discrimination between these two clades was still apparent. In Mycobacterium, Salmonella, and other pathogens the locations of IS elements have been sufficiently stable for use as markers in restriction fragment length polymorphism studies and typing for epidemiological purposes (21). In Francisella, the spatial location of IS elements and their involvement in the plasticity of the genome can also be exploited. Our future studies will determine whether the DISA method will provide a means to detect and monitor the emergence of new substrains of this highly infectious pathogen.
Type A.I, A.II, and B strains of F. tularensis are present in North America, with hard ticks being considered the major biological vector of tularemia in this region (26). Since F. tularensis survival is sustained throughout the tick's life cycle from larva to nymph to adult, and since tick morphogenesis requires a blood meal, pathogen transmission to a new host may occur at each stage. Therefore, a single infected tick may be responsible for establishing a new or recurring focus of disease in the environment. The multidirectional nature of zoonosis, the presence of Francisella-like endosymbionts in ticks, and the broad host range of F. tularensis have complicated epidemiological investigations into diseases caused by this pathogen. A field-deployable assay would be useful for the characterization of infection outbreaks and the identification of their point source or location. The PCR-based DISA method targeting IS elements provides a reliable approach for the rapid evaluation of ticks, animals, and humans potentially infected with F. tularensis. Furthermore, an accurate surveillance of F. tularensis vectors and reservoirs will provide a better understanding of this pathogen's biology.
Molecular-based assays for the detection of infectious diseases have continued to evolve in order to address the problem of false-positive results due to the cross-hybridization of primers with unrelated sequences with unexpected homology. Some of the most effective changes have been the incorporation of multiple targets into the assay, with the criterion for positivity being the amplification of more than one target. Although several real-time PCR assays have been developed to differentiate isolates of F. tularensis that are of types A and B (19) or types A.I and A.II (22), a more comprehensive method was needed to simultaneously and accurately genotype this species. The DISA assay offers this capability and requires only two sets of primers, unlike typing by PCR-based multilocus variable-number tandem repeat analysis. Furthermore, the DISA method takes advantage of the presence of multiple inherent targets within Francisella, and this approach will be used for the future development of a real-time molecular assay.
ACKNOWLEDGMENTS
We thank the following individuals for the new contribution of strains used in this study: David Byrd, Missouri State Public Health Laboratory, Jefferson City, MO; Wanda Lyon, AFIOH/SDE Laboratory, Brooks City Base, TX; and Lori Smith, Utah Department of Health, Salt Lake City, UT. Additionally, we thank Tricia Aden at the Nebraska Public Health Laboratory for her technical help and Elizabeth Wallace for her assistance in collecting environmental ticks.
The U.S. Army Space and Missile Defense Command, Frederick, MD, is the awarding and administering acquisition office that supported this work with a grant (W9113M-07-1-0004) to the University of Nebraska Medical Center (S.H.H.).
The views expressed in this article are those of the authors and do not reflect the official policy or position of the U.S. Air Force, U.S. Army, U.S. Navy, Department of Defense, or U.S. government.
Footnotes
Present address: Armed Forces Radiobiology Research Institute, Uniformed Services University of the Health Sciences, Bethesda, MD 20889.
Published ahead of print on 25 May 2011.
REFERENCES
- 1. Achtman M. 2008. Evolution, population structure, and phylogeography of genetically monomorphic bacterial pathogens. Annu. Rev. Microbiol. 62:53–70 [DOI] [PubMed] [Google Scholar]
- 2. Barabote R. D., et al. 2009. Complete genome sequence of Francisella tularensis subspecies holarctica FTNF002-00. PLoS One 4:e7041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Barns S. M., Grow C. C., Okinaka R. T., Keim P., Kuske C. R. 2005. Detection of diverse new Francisella-like bacteria in environmental samples. Appl. Environ. Microbiol. 71:5494–5500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Broekhuijsen M., et al. 2003. Genome-wide DNA microarray analysis of Francisella tularensis strains demonstrates extensive genetic conservation within the species but identifies regions that are unique to the highly virulent F. tularensis subsp. tularensis. J. Clin. Microbiol. 41:2924–2931 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Centers for Disease Control and Prevention Office of Inspector General Department of Health and Human Services (HHS) 2005. Possession, use, and transfer of select agents and toxins, final rule. Fed. Regist. 70:13293–13325 [PubMed] [Google Scholar]
- 6. Champion M. D., et al. 2009. Comparative genomic characterization of Francisella tularensis strains belonging to low and high virulence subspecies. PLoS Pathog. 5:e1000459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Dempsey M. P., et al. 2007. Genomic deletion marking an emerging subclone of Francisella tularensis subsp. holarctica in France and the Iberian Peninsula. Appl. Environ. Microbiol. 73:7465–7470 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Dempsey M. P., et al. 2006. Paired-end sequence mapping detects extensive genomic rearrangement and translocation during divergence of Francisella tularensis subsp. tularensis and Francisella tularensis subsp. holarctica populations. J. Bacteriol. 188:5904–5914 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Department of Health and Human Services 2007. Biosafety in microbiological and biomedical laboratories, 5th ed U.S. Government Printing Office, Washington, DC [Google Scholar]
- 10. Ellis J., Oyston P. C., Green M., Titball R. W. 2002. Tularemia. Clin. Microbiol. Rev. 15:631–646 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Escudero R., et al. 2008. Molecular method for discrimination between Francisella tularensis and Francisella-like endosymbionts. J. Clin. Microbiol. 46:3139–3143 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Fey P. D., et al. 2007. Molecular analysis of Francisella tularensis subspecies tularensis and holarctica. Am. J. Clin. Pathol. 128:926–935 [DOI] [PubMed] [Google Scholar]
- 13. Huber B., et al. 2010. Description of Francisella hispaniensis sp. nov., isolated from human blood, reclassification of Francisella novicida (Larson et al. 1955) Olsufiev et al. 1959 as Francisella tularensis subsp. novicida comb. nov. and emended description of the genus Francisella. Int. J. Syst. Evol. Microbiol. 60:1887–1896 [DOI] [PubMed] [Google Scholar]
- 14. Johansson A., et al. 2010. Objections to the transfer of Francisella novicida to the subspecies rank of Francisella tularensis. Int. J. Syst. Evol. Microbiol. 60:1717–1718 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Johansson A., et al. 2004. Worldwide genetic relationships among Francisella tularensis isolates determined by multiple-locus variable-number tandem repeat analysis. J. Bacteriol. 186:5808–5818 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Jones E. K., Clifford C. M., Keirans J. E., Khols G. M. 1972. The ticks of Venezuela (Acarina: Ixodoidea) with a key to the species of Amblyomma in the Western Hemisphere. Brigham Young University, Provo, UT [Google Scholar]
- 17. Kugeler K. J., et al. 2005. Discrimination between Francisella tularensis and Francisella-like endosymbionts when screening ticks by PCR. Appl. Environ. Microbiol. 71:7594–7597 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Kugeler K. J., et al. 2009. Molecular epidemiology of Francisella tularensis in the United States. Clin. Infect. Dis. 48:863–870 [DOI] [PubMed] [Google Scholar]
- 19. Kugeler K. J., Pappert R., Zhou Y., Petersen J. M. 2006. Real-time PCR for Francisella tularensis types A and B. Emerg. Infect. Dis. 12:1799–1801 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Larsson P., et al. 2009. Molecular evolutionary consequences of niche restriction in Francisella tularensis, a facultative intracellular pathogen. PLoS Pathog. 5:e1000472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Mahillon J., Chandler M. 1998. Insertion sequences. Microbiol. Mol. Biol. Rev. 62:725–774 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Molins C. R., Carlson J. K., Coombs J., Petersen J. M. 2009. Identification of Francisella tularensis subsp. tularensis A1 and A2 infections by real-time polymerase chain reaction. Diagn. Microbiol. Infect. Dis. 64:6–12 [DOI] [PubMed] [Google Scholar]
- 23. Nalbantoglu U., et al. 2010. Large direct repeats flank genomic rearrangements between a new clinical isolate of Francisella tularensis subsp. tularensis A1 and Schu S4. PLoS One 5:e9007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Noda H., Munderloh U. G., Kurtti T. J. 1997. Endosymbionts of ticks and their relationship to Wolbachia spp. and tick-borne pathogens of humans and animals. Appl. Environ. Microbiol. 63:3926–3932 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Nubel U., et al. 2006. Population structure of Francisella tularensis. J. Bacteriol. 188:5319–5324 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Petersen J. M., Mead P. S., Schriefer M. E. 2009. Francisella tularensis: an arthropod-borne pathogen. Vet. Res. 40:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Petrosino J. F., et al. 2006. Chromosome rearrangement and diversification of Francisella tularensis revealed by the type B (OSU18) genome sequence. J. Bacteriol. 188:6977–6985 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Rohmer L., et al. 2007. Comparison of Francisella tularensis genomes reveals evolutionary events associated with the emergence of human pathogenic strains. Genome Biol. 8:R102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Rotz L. D., Khan A. S., Lillibridge S. R., Ostroff S. M., Hughes J. M. 2002. Public health assessment of potential biological terrorism agents. Emerg. Infect. Dis. 8:225–230 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Scoles G. A. 2004. Phylogenetic analysis of the Francisella-like endosymbionts of Dermacentor ticks. J. Med. Entomol. 41:277–286 [DOI] [PubMed] [Google Scholar]
- 31. Staples J. E., Kubota K. A., Chalcraft L. G., Mead P. S., Petersen J. M. 2006. Epidemiologic and molecular analysis of human tularemia, United States, 1964-2004. Emerg. Infect. Dis. 12:1113–1118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Titball R. W., Petrosino J. F. 2007. Francisella tularensis genomics and proteomics. Ann. N. Y. Acad. Sci. 1105:98–121 [DOI] [PubMed] [Google Scholar]
- 33. Vogler A. J., et al. 2009. Phylogeography of Francisella tularensis: global expansion of a highly fit clone. J. Bacteriol. 191:2474–2484 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Wilson K. 1997. Preparation of genomic DNA from bacteria. In Ausubel F. M., et al. (ed.), Current protocols in molecular biology. John Wiley & Sons, Inc., New York, NY: [DOI] [PubMed] [Google Scholar]
- 35. Yunker C. E., Clifford K. J. E. C. M., Easton E. R. 1986. Dermacentor ticks (Acari: Ixodoidea: Ixodidae) of the New World: a scanning electron microscope atlas. Proc. Entomol. Soc. Wash. 88:609–627 [Google Scholar]