Francisella tularensis is a highly virulent, zoonotic bacterium that causes significant natural disease and is of concern as an organism for bioterrorism. Serologic testing of wildlife is frequently used to monitor spatial patterns of infection and to quantify exposure. Cottontail rabbits (Sylvilagus spp.) are a natural reservoir for F. tularensis in the US, although very little work has been done experimentally to determine how these animals respond to infection; thus, information gathered from field samples can be difficult to interpret. We characterized clinical disease, bacteremia, pathology, and antibody kinetics of North American cottontail rabbits experimentally infected with five strains of F. tularensis. Rabbits were infected with four field strains, including MA00-2987 (type A1b), WY96-3418 (type A2), KY99-3387, and OR96-0246 (type B), and with SchuS4 (type A1a), a widely used, virulent laboratory strain. Infection with the different strains of the bacterium resulted in varied patterns of clinical disease, gross pathology, and histopathology. Each of the type A strains were highly virulent, with rabbits succumbing to infection 3–13 d after infection. At necropsy, numerous microabscesses were observed in the livers and spleens of most rabbits, associated with high bacterial organ burdens. In contrast, most rabbits infected with type B strains developed mild fever and became lethargic, but the disease was infrequently lethal. Those rabbits infected with type B strains that survived past 14 d developed a robust humoral immune response, and F. tularensis was not isolated from liver, spleen, or lung of those animals. Understanding F. tularensis infection in a natural reservoir species can guide serosurveillance and generate new insights into environmental maintenance of this pathogen.

Francisella tularensis is an intracellular, Gram-negative, zoonotic bacterium that causes significant disease in humans and domestic and wild animals and is of concern as an organism for bioterrorism (Tarnvik and Berglund 2003). Tularemia is characterized by multisystemic disease and has been classified into six diverse forms that reflect the point of entry of the organism into the body (Foley and Nieto 2010). Tularemia is transmitted by ticks and flies, water exposure, contaminated food, and aerosol dispersion and is found in Europe, North America, and Asia (Oyston et al. 2004; Petersen et al. 2009; Foley and Nieto 2010). There are two main subspecies of F. tularensis: tularensis and holarctica, also referred to as type A and type B, respectively (Kugeler et al. 2009; Nakazawa et al. 2010). Type A is endemic only in North America and is transmitted to humans primarily by ticks and biting flies, purportedly from a wildlife reservoir, such as rabbits, or through direct contact with infected animals (Kugeler et al. 2009; Petersen et al. 2009). Type A is highly virulent and, depending on the route of administration, the 50% lethal dose in mice may be as low as 1 colony-forming unit (cfu; Oyston et al. 2004). Type A is further classified into two subpopulations: A1 is predominant in the central US, and A2 is more common in the western US (Farlow et al. 2005; Petersen et al. 2008). Type B F. tularensis is endemic in the Northern Hemisphere and is commonly isolated in Europe. Type B strains typically have an aquatic life cycle and have been linked to mosquito transmission (Oyston et al. 2004; Triebenbach et al. 2010; Thelaus et al. 2014). Two additional species of Francisella have been described: novicida and mediasiatica (Nigrovic and Wingerter 2008). These species are far less virulent than type A and B and are much less common (Keim et al. 2007). Francisella novicida is endemic in North America and was recently found in Australia, although it is rarely isolated; F. mediasiatica is primarily found in central Asia and the former republics of the Soviet Union (Oyston et al. 2004; Keim et al. 2007).

Previous experimental infections have described clinical presentation and histopathology associated with aerosol infection with F. tularensis and determined that rabbits are a potential model for human tularemia; however, neither study evaluated this organism in the natural reservoir species, cottontail rabbits (Sylvilagus spp.) (Baskerville and Hambleton 1976; Reed et al. 2011). Despite extensive literature asserting rabbits as the primary reservoir species for F. tularensis, there is a significant gap in understanding of transmission patterns, immune response, and the potential for environmental maintenance of F. tularensis. We provide an initial characterization of clinical disease, bacteremia, pathology, organ burden, and antibody kinetics of North American cottontail rabbits experimentally infected with three type A and two type B strains of F. tularensis. Additionally, we characterized the long-term humoral immune response and the ability to clear infection in cottontail rabbits infected with two type B stains.

Experimental design and animals

Two studies were conducted with a total of 46 cottontail rabbits. In the acute-phase study, we infected 20 cottontail rabbits with one of five strains of F. tularensis (four rabbits per strain) to evaluate morbidity, mortality, gross pathology and histopathology, and organ burden. In the long-term study, we used an additional 20 cottontail rabbits, 10 of which were challenged with each of two type B strains and euthanized 2–12 wk postinfection to assess humoral immune responses over time and the ability to clear infection. In each phase, two or three additional rabbits were sham inoculated, housed in the same room as inoculated rabbits, and served as handling controls to evaluate the potential for airborne transmission of F. tularensis, which has been suspected in field studies with European hares (Lepus europaeus; Gyuranecz et al. 2010).

Male and female cottontail rabbits were wild-trapped along the front range of Colorado, US. Rabbits were housed individually in standard stainless-steel rabbit cages in an ABSL-3 containment facility approved for use of Select Agents. Rabbits were provided ad libitum access to alfalfa pellets, alfalfa hay, and water, and acclimated to the laboratory setting for 3–4 wk before infection, during which time they were treated for fleas and ticks and received a subcutaneous IPTT300 temperature transponder (BioMedic Data Systems, Inc., Seaford, Delaware, USA). Ticks were not observed on any of the rabbits. We used PCR and DNA sequencing to determine rabbit species (Berkman et al. 2009). Sequences of amplified products were compared with representative sequences from Genbank to identify species. All 22 rabbits used in the acute-phase study and 21 of 23 for the long-term study were desert cottontails (Sylvilagus audubonii); the remaining two were mountain cottontails (Sylvilagus nuttallii).

Rabbits were inoculated with F. tularensis intradermally on the right hip with 50 µL containing 25–125 organisms; control animals were inoculated with 50 µL of sterile phosphate-buffered saline (PBS). Before and after inoculation, body weight, temperature, and appetite of each rabbit were evaluated. Weight was measured using a Pesola scale (Ben Meadows Company, Janesville, Wisconsin, USA) with the rabbit wrapped tightly in a towel and placed in a cloth bag. Once daily, rabbits were provided a treat of peaches, pears, or pineapple, and their enthusiasm for those items proved to be an effective means of assessing subtle changes in clinical presentation.

This work was approved by the Animal Care and Use Committee at Colorado State University (approval 13-4209A) and conducted in strict accordance with the National Institutes of Health's Guide for the Care and Use of Laboratory Animals (Committee for the Update of the Guide for the Care and Use of Laboratory Animals 2011).

Bacterial strains and culture methods

Five strains of F. tularensis were used in these experiments; for clarity, each strain name is abbreviated to include its clade (Table 1). All five strains of F. tularensis were provided at unknown passage history and passaged in our laboratory one time. Stocks of strains Schu-A1a, MA-A1b, KY-B, and OR-B were prepared from cultures grown 24–36 h in modified Mueller-Hinton (MMH) broth at 37 C with 5% CO2, and frozen in 15% glycerol (Baker et al. 1985). The WY-A2 strain was grown on cysteine heart agar with 9% chocolatized sheep blood (CHAB), because of the difficulty culturing this organism in MMH broth, for 48 h at 37 C with 5% CO2; after which, the agar plate was flooded with MMH broth and the colonies collected. Glycerol was added to the broth to achieve a final concentration of 15%.

Table 1. 

Strains of Francisella tularensis used in this study for inoculation of cottontail rabbits (Sylvilagus spp.).

Strains of Francisella tularensis used in this study for inoculation of cottontail rabbits (Sylvilagus spp.).
Strains of Francisella tularensis used in this study for inoculation of cottontail rabbits (Sylvilagus spp.).

Bacteremia evaluation

To assess bacteremia, 50 µL of blood collected from the marginal ear vein was immediately diluted with 450 µL of PBS. Serial 10-fold dilutions of this mixture were plated on MMH agar, except for samples from rabbits infected with the WY-A2 strain, which were plated on CHAB agar. Plates were incubated at 37 C with 5% CO2 for 24–72 h; at which time, they were counted and recorded. Colonies derived from whole blood samples were not individually confirmed using PCR; rather, if the number of colonies recorded was consistent with 10-fold serial dilutions and organisms were found disseminated in the liver, spleen, or lungs (confirmed by PCR), the animal was considered bacteremic.

Euthanasia, necropsy, histopathology, and organ burden

Rabbits were euthanized at the end of the study or as necessary because of a moribund condition by an overdose of pentobarbital administered intravenously. All survivors of the acute-phase study were euthanized at day 14, whereas rabbits used in the long-term experiment were euthanized at intervals to assess organ burden at varying time-points after inoculation. All animals were necropsied and gross lesions recorded, including microabscessation, pulmonary consolidation, and splenomegaly. Splenomegaly was assessed qualitatively, rather than by weight, because control-spleen weights for wild-caught cottontail rabbits were not available. The liver, spleen, and lungs were the primary organs evaluated for gross pathology because F. tularensis preferentially traffics to these sites (Dennis et al. 2001; Lamps et al. 2004). We collected liver, spleen, lungs, heart, duodenum, bladder, and kidneys into 10%-buffered formalin. Samples were embedded in paraffin, sectioned at 5 µm, and stained with H&E for histology. Sections were examined on a Nikon Eclipse 51E microscope, and digital micrographs were taken with a Nikon DS-Fi1 camera with a DS-U2 unit and NIS-elements F software (Nikon Instruments Inc., Melville, New York, USA). Images are reproduced without manipulations other than cropping or adjustment of light intensity. Severity of lesions was scored on a scale of zero to six, with zero denoting histologically normal appearance, and six indicating the most severe lesions.

Samples (100 mg) of lung, liver, spleen, and kidney were collected in a vial with 0.9 mL of Mueller-Hinton broth containing 15% glycerol and two stainless-steel BBs, immediately homogenized in a mixer mill and frozen at −80 C. Samples were thawed and 10-fold serial dilutions were made from 10−1 to 10−3. Samples (100 µL) from each of the three dilutions were plated on MMH agar plates (except rabbits infected with the WY-A2 strain were plated on CHAB agar). Plates were incubated at 37 C with 5.0% CO2 for 24–48 h, and colony counts were recorded. For each rabbit with a positive culture, DNA was extracted from one bacterial colony derived from the spleen and its identity as F. tularensis was confirmed by PCR using a protocol similar to that of Long et al. (1993).

Serology

An enzyme-linked immunosorbent assay was developed to detect serum antibodies to F. tularensis. This assay followed World Health Organization procedures (Tarnvik 2007), with exceptions described shortly. Briefly, NUNC polysorp 96-well plates (Thermo Scientific, Rochester, New York, USA) were coated overnight at room temperature with 100 µL of coating buffer containing 3 µg/mL F. tularensis LPS obtained from BEI Resources (Manassas, Virginia, USA), blocked with 5% nonfat dry milk for 30 min and rinsed five times with 300 µL of washing buffer. Serum samples were heat treated for 30 min at 56 C to ensure inactivation of any residual organisms, diluted 1:1,000 in incubation buffer, and duplicate wells were loaded with 100 µL. Positive and negative rabbit sera were used as controls in each assay. After a 1-hr incubation, the plate was emptied and rinsed five times with 300 µL of washing buffer. Goat anti-rabbit horseradish-peroxidase conjugate (Jackson ImmunoResearch, West Grove, Pennsylvania, USA) was added to each well and incubated for 1 h. The plate was emptied and rinsed five times with washing buffer, and substrate was applied (TMB Peroxidase Substrate, KPL, Gaithersburg, Maryland, USA). The reaction was allowed to proceed for 15–20 min, stopped by addition of 50 µL of 1N hydrochloric acid, and optical density determined using a plate reader with a 450 nm filter (Bio-Rad model 680 plate reader, Hercules, California, USA). The cutoff for a positive sample was 3 SDs above the mean of the values from the negative-control sera.

Statistical and survival analyses

STATA software (Stata, Statistical Software: Release 11.2, College Station, Texas, USA) was used for descriptive statistics and survival analysis. Median survival times (in days) and 95% confidence intervals were calculated using a Kaplan-Meier survival function. Univariate nonparametric analysis was conducted using the log-rank test to compare the survival function (risk of death) among cottontail rabbits infected with different strains. Fisher's exact test was used to compare histopathologic findings, frequency of dissemination, and organ burdens in the acute-phase study.

Acute-phase experiment

All rabbits inoculated with type A strains became ill, lost about 10% of their body weight, and either died or required euthanasia within 14 d postinfection (dpi) (Table 2). Fever was observed in all rabbits starting at 2 dpi and persisted through the time of death in all but two rabbits. Baseline body temperature before infection ranged from 38.3 C to 40 C, and peak body temperature after challenge ranged from 41.7 C to 42.8 C. Upon necropsy at 5 dpi, rabbit 1 (Schu-A1a) had a large intrathoracic abscess, likely as a preexisting infection; F. tularensis was not confirmed by PCR in samples from that rabbit, and it was removed from the study. Bacteremia was detected on one day in two of the four rabbits inoculated with either the MA-A1b or Schu-A1a strain, but not in rabbits inoculated with WY-A2 strain. Despite the low rate of bacteremia detected by once daily sampling, the type A organisms were widely disseminated in all but one rabbit, indicating hematogenous spread throughout the body. Gross lesions were detected in all 11 rabbits inoculated with type A strains of F. tularensis (Fig. 1A). The most common gross lesions were microabscesses randomly scattered throughout the liver and/or spleen and splenomegaly (Table 2). Histopathologic lesions were observed in rabbit 4 in the liver, spleen, and lung at 13 dpi with Schu-A1a (Fig. 2). These lesions are representative of severely infected animals (Table 3).

Table 2. 

Acute phase experiment: Summary of clinical response, pathology, and microbiology for wild-caught desert cottontails (Sylvilagus audubonii) experimentally infected with Francisella tularensis, Colorado, USA.

Acute phase experiment: Summary of clinical response, pathology, and microbiology for wild-caught desert cottontails (Sylvilagus audubonii) experimentally infected with Francisella tularensis, Colorado, USA.
Acute phase experiment: Summary of clinical response, pathology, and microbiology for wild-caught desert cottontails (Sylvilagus audubonii) experimentally infected with Francisella tularensis, Colorado, USA.
Figure 1. 

Microabscessation resulting from Francisella tularensis infection in Colorado, USA, cottontail rabbits (Sylvilagus spp.). (A) Liver and spleen from rabbit 3 at 7 d after challenge with Schu-A1a strain of F. tularensis. (B) Liver and spleen from rabbit 20 at 14 d after challenge with strain OR-B.

Figure 1. 

Microabscessation resulting from Francisella tularensis infection in Colorado, USA, cottontail rabbits (Sylvilagus spp.). (A) Liver and spleen from rabbit 3 at 7 d after challenge with Schu-A1a strain of F. tularensis. (B) Liver and spleen from rabbit 20 at 14 d after challenge with strain OR-B.

Close modal
Figure 2. 

Histopathology associated with laboratory infection of Schu-A1a strain of Francisella tularensis in a cottontail rabbit (Sylvilagus spp.). Micrographs of (A) spleen, (B) liver, and (C) lung, at 40× from rabbit 4 at 13 d following challenge.

Figure 2. 

Histopathology associated with laboratory infection of Schu-A1a strain of Francisella tularensis in a cottontail rabbit (Sylvilagus spp.). Micrographs of (A) spleen, (B) liver, and (C) lung, at 40× from rabbit 4 at 13 d following challenge.

Close modal
Table 3. 

Acute phase experiment: Histopathology for wild-caught desert cottontails (Sylvilagus audubonii) experimentally infected with Francisella tularensis, Colorado, USA.

Acute phase experiment: Histopathology for wild-caught desert cottontails (Sylvilagus audubonii) experimentally infected with Francisella tularensis, Colorado, USA.
Acute phase experiment: Histopathology for wild-caught desert cottontails (Sylvilagus audubonii) experimentally infected with Francisella tularensis, Colorado, USA.

All three animals infected with Schu-A1a had severe focal (rabbit 2) or multifocal to coalescing, necrotizing splenitis. Two animals also had moderate, necrotizing hepatitis, whereas only one animal (rabbit 4) had evidence of pneumonia, characterized by multifocal to coalescing infiltration of macrophages, heterophils, fewer lymphocytes, and marked pulmonary vasculitis. All four animals infected with strain MA-A1b had moderate to severe, necrotizing hepatitis and moderate splenitis. Only one (rabbit 5) had apparent pneumonia, presenting as focal, necrotizing lesions, whereas rabbit 7 had very severe lung edema, presumably an agonal change. The WY-A2-infected animals all had multifocal, mild to moderate, necrotizing hepatitis and mild splenitis. In three animals, there was mild leukocytosis in the pulmonary vasculature but no frank parenchymal inflammation. Of the four animals infected with OR-B, and terminated at 14 dpi, two had either focal (rabbit 17) or multifocal (rabbit 20) severe, necrotizing pneumonia. All four animals had multifocal, mild to moderate splenitis and hepatitis, either necrotizing or granulomatous in character. In contrast, two of the animals infected with KY-B appeared histologically unremarkable (rabbits 13 and 14), whereas rabbits 15 and 16 had minimal to mild hepatitis, splenitis, and alveolitis.

In contrast to the response to each of the type A strains, none of eight rabbits inoculated with the two type B strains manifested severe, overt clinical disease nor did they succumb to their infection within 14 d; however, gross lesions were detected on three of the eight rabbits (Fig. 1B). Elevated body temperature was observed starting at 2 dpi and persisted until 8 dpi in all of the rabbits inoculated with either strain of type B F. tularensis (data not shown). Bacteremia was not detected in any of these animals, and the risk of gross lesions was limited in comparison to rabbits inoculated with type A strains (Table 2). The frequency of dissemination in rabbits inoculated with type B strains was not different from those inoculated with type A strains (Fisher's exact test, P = 0.65), but the magnitude of organ burdens in lung, liver, and spleen was significantly less in rabbits infected with type B strains when comparing tissues with >105 cfu/g to all others (Fisher's exact test, P<0.001). When comparing histopathologic lesions between rabbits inoculated with type A versus type B strains, we compared rabbits with “severe, widespread, diffuse” changes in liver, spleen, or lung to those with “normal, minimal, mild, or moderate” changes in the same tissues and found 18% and 0%, respectively (Fisher's exact test, P = 0.07).

The type A strains of F. tularensis were considerably more virulent in cottontail rabbits than the type B strains were because none of the rabbits infected with the type A strains (n = 11) survived to day 14, whereas all of those inoculated with type B strains (n = 8) survived to day 14. There were significant differences in risk of mortality for rabbits inoculated with these two types of F. tularensis (log-rank test, P<0.001) (Table 2). Rabbits infected with type A strains were 14 times more likely to develop microabscesses in the liver or spleen compared with rabbits infected with type B strains (Fisher's exact test odds ratio [OR] 95% confidence interval [CI] = 1.6–111.8, P = 0.024). There was no difference in the odds of splenomegaly between rabbits challenged with type A and type B strains of F. tularensis (Fisher's exact test OR 95% CI = 0.2–8.2, P = 1.0).

Uninfected control rabbits remained healthy and active throughout the experiment and did not exhibit elevation in body temperature or weight loss.

Long-term experiment

Clinical and microbiologic responses of rabbits inoculated with KY-B or OR-B are summarized in Table 4. Three of the 10 rabbits inoculated with KY-B were euthanized because of clinical disease between 8 dpi and 10 dpi; however, F. tularensis was detected postmortem in only two of these animals. The third (rabbit 20) was euthanized because of a periocular abscess determined (by Gram stain) to be unrelated to infection with F. tularensis. The remaining seven rabbits inoculated with KY-B developed moderate fever for several days but did not manifest overt disease. Mild, transient fever was also observed in all rabbits inoculated with OR-B, and two of those 10 animals were euthanized during the experiment because of conditions not related to infection. Only one rabbit of the 20 inoculated with KY-B or OR-B was found to be bacteremic. That was detected at 5 dpi, and the animal (rabbit 16, KY-B) was found moribund at 10 dpi and euthanized.

Table 4. 

Long-term experiment: Summary of clinical response, pathology, and microbiology for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.

Long-term experiment: Summary of clinical response, pathology, and microbiology for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.
Long-term experiment: Summary of clinical response, pathology, and microbiology for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.

Of the 10 animals infected with OR-B, three were terminated at 14–17 dpi. Two of these had multifocal, subacute, necrotizing pneumonia, whereas the third had granulomatous splenitis. Of the remaining seven animals, terminated at 28–84 dpi, all but one had mild to severe hepatitis, in two cases, accompanied by amyloid depositions and marked Kupffer cell hypertrophy. The same six animals also had multifocal, moderate to severe, mostly necrotizing pneumonia. Of the 10 animals infected with KY-B, four were terminated at 8–14 dpi. Three had necrotizing splenitis, varying from mild (rabbit 20) to very severe (rabbit 14). Two (rabbits 16 and 18) had multifocal, mild-to-moderate, subacute, necrotizing pneumonia, whereas one (rabbit 14) had multifocal to coalescing, necrotizing hepatitis. Of the remaining six animals in this group, terminated at 28–84 dpi, only two had histopathologic evidence of severe pneumonia. Other mild lesions observed in two animals included interstitial nephritis, most likely a lesion unrelated to the experimental infection, and mild hepatitis.

No histopathologic comparisons were made among or within groups at the varying time points as the sample sizes were not sufficient to allow comparisons. Histopathologic findings are summarized in Tables 5 and 6.

Table 5. 

Long-term experiment: Histopathology scores (strain KY-B) for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.

Long-term experiment: Histopathology scores (strain KY-B) for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.
Long-term experiment: Histopathology scores (strain KY-B) for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.
Table 6. 

Long-term experiment: Histopathology scores (strain OR-B) for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.

Long-term experiment: Histopathology scores (strain OR-B) for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.
Long-term experiment: Histopathology scores (strain OR-B) for wild-caught cottontails (Sylvilagus spp.) experimentally infected with Francisella tularensis, Colorado, USA.

Humoral immune response was evaluated only in the long-term experiment, and all rabbits surviving past 14 dpi developed antibodies. Irrespective of the infecting strain, peak antibody production occurred at 14–21 dpi, then typically declined slightly before leveling out and remaining stable for the duration of the experiment (Fig. 3).

Figure 3. 

Enzyme-linked immunosorbent assay (ELISA) antibody responses of wild-caught cottontail rabbits (Sylvilagus spp.) infected with (A) Francisella tularensis strain OR-B, and (B) strain KY-B between 8 d and 84 d after infection.

Figure 3. 

Enzyme-linked immunosorbent assay (ELISA) antibody responses of wild-caught cottontail rabbits (Sylvilagus spp.) infected with (A) Francisella tularensis strain OR-B, and (B) strain KY-B between 8 d and 84 d after infection.

Close modal

Control rabbits

Control rabbits had no detectable antibodies to F. tularensis nor was any organism detected when tissues were homogenized and plated after necropsy. Because of the territorial nature of cottontail rabbits, it is unfeasible to cohouse them, which may be necessary to evaluate transmission among rabbits. Two of the three mock-infected control animals, terminated on day 84, had a large hepatic granuloma, in one case, with caseation and mineralization of the central core. The most likely cause of this type of lesion is aberrant parasite migration, as often observed in wild-caught cottontail rabbits (H.B.O. pers. obs.).

We provide an initial assessment of the pathogenesis of infection in cottontail rabbits experimentally infected with several field isolates of F. tularensis. Understanding the dynamics of organism dissemination, gross pathology and histopathology, organ burden, and mortality, as well as humoral immune response and ability to clear infection, is a crucial start when attempting to determine the role of cottontail rabbits in the maintenance and spread F. tularensis.

The acute-phase experiment identified the MA-A1b strain to be the most virulent of those tested in cottontail rabbits (all four infected rabbits succumbed to disease by day 5), followed successively by WY-A2 and Schu-A1a (Table 2). As anticipated, the type B strains (OR-B and KY-B) were distinctly less virulent than type A strains, although capable of causing mortality in some instances (KY-B). These findings support what has been found in laboratory mice, but contrasts with findings in humans because the type B strains typically result in higher mortality than A2 strains in people (Molins et al. 2010; Reese et al. 2011). Among human cases of tularemia reported in the US, the A1b strain resulted in 24% mortality, compared with 4% with A1a strains, 0% with A2 strains, and 7% with type B strains (Staples et al. 2006; Reese et al. 2010). Increases in body temperature in our study were observed more rapidly following inoculation and were of much higher magnitude than those reported by Reed et al. (2011).

The long-term experiment demonstrated that cottontail rabbits can develop a robust humoral immune response following intradermal challenge with F. tularensis. The level of protection afforded by this antibody response against a virulent challenge with F. tularensis is unknown, although the intracellular nature of this pathogen may render antibodies insufficient (Celli and Zahrt 2013).

Bacteremia was difficult to detect in both experiments, despite 90% of rabbits having bacteria in liver, spleen, or lungs, which is suggestive of hematogenous spread. To evaluate the presence of organisms in the bloodstream, rabbits were bled once daily (on the aforementioned days) in the morning. This suggests that bacteremia was transient and thus, not readily detected with only one bleed per day. We suspect that the organism was sequestered in microabscesses, specifically on the liver and spleen, and at various points, these abscesses would rupture resulting in a “bacterial seeding event,” which facilitated organism dissemination. Alternatively, the numbers of organisms in the bloodstream may have been below our limit of detection (100 cfu/mL).

This study was subject to several potential limitations; perhaps most important, we used wild-caught animals that were undoubtedly stressed by captivity. We attempted to mitigate that problem by acclimatizing them for several weeks before challenge and handling them as gently and a infrequently as possible. Although the laboratory is an unfamiliar environment, it is arguably less stressful than their natural environment. Additionally, our goal was to study the infection in a natural host, and the animals we used had been exposed to and perhaps harbored a variety of pathogens. Finally, the inoculating dose and route were not necessarily representative of those experienced in nature. There is no precise estimate from the literature indicating the amount of F. tularensis that would be transferred from an arthropod vector to a mammalian host, so we used a low dose administered intradermally in an attempt to mimic what a tick might deliver. Despite these challenges, we believe our findings are novel and reliable as we pursue further understanding of F. tularensis infection in cottontail rabbits.

The primary F. tularensis strains in the US are type A, which have been associated with cottontail rabbits as the primary reservoir; transmission to humans thought to result from interaction with, or ingestion of, contaminated rabbit carcasses or an arthropod vector, primarily ticks and biting flies (Eisen 2007; Petersen et al. 2009; Reese et al. 2011). Despite this longstanding notion, our understanding of cottontail rabbits and their role in F. tularensis transmission and maintenance is inadequate (Farlow et al. 2005; Foley and Nieto 2010; Telford and Goethert 2011). All rabbits infected with field isolates of type A strains, MA-A1b and WY-A2, rapidly succumbed to infection (maximum survival of 8 d). This level of virulence is not typical in reservoir hosts, which must survive infection to maintain the agent in nature. Furthermore, bacteremia was detected in only two of 16 rabbits challenged with field isolates, further challenging the notion that cottontail rabbits are a competent reservoir host. These characteristics question the assertion that cottontail rabbits are reservoir hosts and suggest that cottontails are an incidental host that is extremely susceptible to disease. Foley and Nieto (2010) articulate a similar message, “although the rabbit is commonly cited as the reservoir of tularemia, it is more likely that the actual reservoir is either an environmental fomite or the arthropod vector itself. Nevertheless, the rabbit is often associated with human exposure risk and may be more appropriately termed an amplification host rather than a reservoir.” In many instances, large rabbit die-offs are indicative of the presence of F. tularensis, which often spills over into the human population, “rabbits and hares may only be the epidemiological bridge and are not necessarily an element of natural focality” (Telford and Goethert 2011). It may be that rabbits could be used as a sentinel species for tularemia as prairie dogs are used for plague, Yersinia pestis, in the US (Cully et al. 2000; Lowell et al. 2009). Further research is needed to elucidate the role cottontail rabbits have in the maintenance and transmission of F. tularensis among wildlife populations and to humans.

We thank Jeannine Petersen (US Centers for Disease Control and Prevention) for donating the F. tularensis isolates. We also thank Sarah Bevins, Tom Gidlewski, and Dennis Kohler for insightful discussions. This work was funded through a cooperative agreement 11-7100-0331-CA from the US Department of Agriculture.

Baker
CN
,
Hollis
DG
,
Thornsberry
C.
1985
.
Antimicrobial susceptibility testing of Francisella tularensis with a modified Mueller-Hinton broth
.
J Clin Microbiol
22
:
212
215
.
Baskerville
A
,
Hambleton
P.
1976
.
Pathogenesis and pathology of respiratory tularaemia in the rabbit
.
Brit J Exp Pathol
57
:
339
347
.
Berkman
LK
,
Saltzgiver
MJ
,
Heist
EJ
,
Nielsen
CK
,
Roy
CL
,
Scharine
PD.
2009
.
Hybridization and polymorphic microsatellite markers for two lagomorph species (genus Sylvilagus): Implications for conservation
.
Conserv Genet Resour
1
:
419
424
.
Celli
J
,
Zahrt
TC.
2013
.
Mechanisms of Francisella tularensis intracellular pathogenesis
.
Cold Spring Harbor Perspect Med
3
:
a010314
.
Committee for the Update of the Guide for the Care and Use of Laboratory Animals, National Research Council of the National Academies
.
2011
.
Guide for the care and use of laboratory animals
. 8th Ed.
National Academy Press
,
Washington, DC
,
220
pp.
Cully
JF
,
Carter
LG
,
Gage
KL.
2000
.
New records of sylvatic plague in Kansas
.
J Wildl Dis
36
:
389
392
.
Dennis
DT
,
Inglesby
TV
,
Henderson
DA
,
Bartlett
JG
,
Ascher
MS
,
Eitzen
E
,
Fine
AD
,
Friedlander
AM
,
Hauer
J
,
Layton
M
,
et al
.
2001
.
Tularemia as a biological weapon—Medical and public health management
.
JAMA
285
:
2763
2773
.
Eisen
L.
2007
.
A call for renewed research on tick-borne Francisella tularensis in the Arkansas-Missouri primary national focus of tularemia in humans
.
J Med Entomol
44
:
389
397
.
Farlow
J
,
Wagner
DM
,
Dukerich
M
,
Stanley
M
,
Chu
M
,
Kubota
K
,
Petersen
JM
,
Keim
P.
2005
.
Francisella tularensis in the United States
.
Emerg Infect Dis
11
:
1835
1841
.
Foley
JE
,
Nieto
NC.
2010
.
Tularemia
.
Vet Microbiol
140
:
332
338
.
Gyuranecz
M
,
Szeredi
L
,
Makrai
L
,
Fodor
L
,
Meszaros
AR
,
Szepe
B
,
Fuleki
M
,
Erdelyi
K.
2010
.
Tularemia of European brown hare (Lepus europaeus): A pathological, histopathological, and immunohistochemical study
.
Vet Pathol
47
:
958
963
.
Keim
P
,
Johansson
A
,
Wagner
DM.
2007
.
Molecular epidemiology, evolution, and ecology of Francisella
.
Ann N Y Acad Sci
1105
:
30
66
.
Kugeler
KJ
,
Mead
PS
,
Janusz
AM
,
Staples
JE
,
Kubota
KA
,
Chalcraft
LG
,
Petersen
JM.
2009
.
Molecular epidemiology of Francisella tularensis in the United States
.
Clin Infect Dis
48
:
863
870
.
Lamps
LW
,
Havens
JM
,
Sjostedt
A
,
Page
DL
,
Scott
MA.
2004
.
Histologic and molecular diagnosis of tularemia: A potential bioterrorism agent endemic to North America
.
Mod Pathol
17
:
489
495
.
Long
GW
,
Oprandy
JJ
,
Narayanan
RB
,
Frontier
AH
,
Porter
KR
,
Nacy
CA.
1993
.
Detection of Francisella tularensis in blood by polymerase chain reaction
.
J Clin Microbiol
31
:
152
154
.
Lowell
JL
,
Eisen
RJ
,
Schotthoefer
AM
,
Xiaocheng
L
,
Monternieri
JA
,
Tanda
D
,
Pape
J
,
Schriefer
ME
,
Antolin
MF
,
Gage
KL.
2009
.
Colorado animal-based plague surveillance systems: relationships between targeted animal species and prediction efficacy of areas at risk for humans
.
J Vector Ecol
34
:
22
31
.
Molins
CR
,
Delorey
MJ
,
Yockey
BM
,
Young
JW
,
Sheldon
SW
,
Reese
SM
,
Schriefer
ME
,
Petersen
JM.
2010
.
Virulence differences among Francisella tularensis Subsp. tularensis clades in mice
.
PLoS One
5
:
e10205
.
Nakazawa
Y
,
Williams
RAJ
,
Peterson
AT
,
Mead
PS
,
Kugeler
KJ
,
Petersen
JM.
2010
.
Ecological niche modeling of Francisella tularensis subspecies and clades in the United States
.
Am J Trop Med Hyg
82
:
912
918
.
Nigrovic
LE
,
Wingerter
SL.
2008
.
Tularemia
.
Infect Dis Clin North Am
22
:
489
504
.
Oyston
PCF
,
Sjöstedt
A
,
Titball
RW.
2004
.
Tularaemia: Bioterrorism defence renews interest in Francisella tularensis
.
Nat Rev Microbiol
2
:
967
978
.
Petersen
JM
,
Carlson
JK
,
Dietrich
G
,
Eisen
RJ
,
Coombs
J
,
Janusz
AM
,
Summers
J
,
Beard
CB
,
Mead
PS.
2008
.
Multiple Francisella tularensis subspecies and clades, tularemia outbreak, Utah
.
Emerg Infect Dis
14
:
1928
1930
.
Petersen
JM
,
Mead
PS
,
Schriefer
ME.
2009
.
Francisella tularensis: An arthropod-borne pathogen
.
Vet Res
40
:
7
.
Reed
DS
,
Smith
L
,
Dunsmore
T
,
Trichel
A
,
Ortiz
LA
,
Cole
KS
,
Barry
E.
2011
.
Pneumonic tularemia in rabbits resembles the human disease as illustrated by radiographic and hematological changes after infection
.
PLoS One
6
:
e24654
.
Reese
SM
,
Dietrich
G
,
Dolan
MC
,
Sheldon
SW
,
Piesman
J
,
Petersen
JM
,
Eisen
RJ.
2010
.
Transmission dynamics of Francisella tularensis subspecies and clades by nymphal Dermacentor variabilis (Acari: Ixodidae)
.
Am J Trop Med Hyg
83
:
645
652
.
Reese
SM
,
Petersen
JM
,
Sheldon
SW
,
Dolan
MC
,
Dietrich
G
,
Piesman
J
,
Eisen
RJ.
2011
.
Transmission efficiency of Francisella tularensis by adult American dog ticks (Acari: Ixodidae)
.
J Med Entomol
48
:
884
890
.
Staples
JE
,
Kubota
KA
,
Chalcraft
LG
,
Mead
PS
,
Petersen
JM.
2006
.
Epidemiologic and molecular analysis of human tularemia, United States, 1964–2004
.
Emerg Infect Dis
12
:
1113
1118
.
Tarnvik
A.
2007
.
WHO guidelines on tularaemia: Epidemic and pandemic alert and response
.
World Health Organization
,
Geneva, Switzerland
,
125
pp. .
Tarnvik
A
,
Berglund
L.
2003
.
Tularaemia
.
Eur Respir J
21
:
361
373
.
Telford
SR
,
Goethert
HK.
2011
.
Toward an understanding of the perpetuation of the agent of tularemia
.
Front Microbiol
1
:
150
.
Thelaus
J
,
Andersson
A
,
Broman
T
,
Backman
S
,
Granberg
M
,
Karlsson
L
,
Kuoppa
K
,
Lundmark
E
,
Lundstrom
JO
,
Mathisen
P.
2014
.
Francisella tularensis subspecies holarctica occurs in Swedish mosquitoes, persists through the developmental stages of laboratory-infected mosquitoes and is transmissible during blood feeding
.
Microb Ecol
67
:
96
107
.
Triebenbach
AN
,
Vogl
SJ
,
Lotspeich-Cole
L
,
Sikes
DS
,
Happ
GM
,
Hueffer
K.
2010
.
Detection of Francisella tularensis in Alaskan mosquitoes (Diptera: Culicidae) and assessment of a laboratory model for transmission.”
J Med Entomol
47
:
639
648
.