Mouse Model for Sublethal Leptospira interrogans Infection

Abstract

Although Leptospira can infect a wide range of mammalian species, most studies have been conducted in golden Syrian hamsters, a species particularly sensitive to acute disease. Chronic disease has been well characterized in the rat, one of the natural reservoir hosts. Studies in another asymptomatic reservoir host, the mouse, have occasionally been done and have limited infection to mice younger than 6 weeks of age. We analyzed the outcome of sublethal infection of C3H/HeJ mice older than age 10 weeks with Leptospira interrogans serovar Copenhageni. Infection led to bloodstream dissemination of Leptospira, which was followed by urinary shedding, body weight loss, hypothermia, and colonization of the kidney by live spirochetes 2 weeks after infection. In addition, Leptospira dissemination triggered inflammation in the kidney but not in the liver or lung, as determined by increased levels of mRNA transcripts for the keratinocyte-derived chemokine, RANTES, macrophage inflammatory protein 2, tumor necrosis factor alpha, interleukin-1β, inducible nitric oxide synthase, interleukin-6, and gamma interferon in kidney tissue. The acquired humoral response to Leptospira infection led to the production of IgG mainly of the IgG1 subtype. Flow cytometric analysis of splenocytes from infected mice revealed that cellular expansion was primarily due to an increase in the levels of CD4⁺ and double-negative T cells (not CD8⁺ cells) and that CD4⁺ T cells acquired a CD44^high CD62L^low effector phenotype not accompanied by increases in memory T cells. A mouse model for sublethal Leptospira infection allows understanding of the bacterial and host factors that lead to immune evasion, which can result in acute or chronic disease or resistance to infection (protection).

INTRODUCTION

Leptospirosis causes a substantial burden on human and animal health worldwide, with WHO setting the number of severe cases of this zoonotic disease at over 500,000 per year (1, 2). The enzootic cycle of Leptospira involves a number of reservoir hosts that act as maintenance carriers of the spirochete in the ecosystem (3, 4). Chronically infected reservoir hosts shed Leptospira spirochetes in their urine, and transmission to accidental hosts (humans) or reservoir hosts (small rodents, cattle) occurs directly via contact with Leptospira-infected urine or tissues or indirectly through contact with contaminated soil and water (5–7). Leptospira infection is heavily influenced by the host species and by the bacterial serovar involved (8–11). Recognition of pathogen-associated molecular patterns by a variety of host receptors activates the immune system. The outcome of this response may result in bacterial clearance, limited bacterial colonization of a few target organs, or induction of sepsis as the host succumbs to infection and dies (12). Thus, leptospirosis may appear as an acute, potentially fatal infection in accidental hosts or progress into a chronic, largely asymptomatic infection in natural reservoir hosts.

Significant advances in the characterization of the Leptospira interaction with host tissues in vitro have been made, but these have not been systematically translated into the characterization of disease progression in the reservoir host (13). Most studies have been conducted in golden Syrian hamsters, a species particularly susceptible to acute infection (12, 13), with only a few studies focusing on the natural reservoir host. The rat is the accepted experimental natural reservoir host used to study chronic leptospirosis, as rats shed significant numbers of Leptospira organisms in their urine for at least 6 months postinfection (14–17). However, progress toward understanding the bacterial and host factors that lead to persistence has been delayed by the lack of availability of genetic and immunological tools for rat species.

Mice are also asymptomatic reservoir hosts (3, 4). Early findings that some strains are refractory to infection by pathogenic Leptospira spirochetes after about 4 weeks of age led to a relative reluctance to conduct investigations with this model. Although this is true for the development of acute disease in BALB/c mice (18–20), it is also apparent that other strains, such as DBA, C3H, and C57BL/6, can be infected at between 3 and 6 weeks of age and that mice of these strains show signs and symptoms of both acute and chronic infection (8–11, 20–23).

We characterized disease progression in C3H/HeJ mice infected with a sublethal dose of a pathogenic serovar of Leptospira after 10 weeks of age. We quantified the numbers of live leptospires shed in urine and present in kidney and blood, measured the levels of inflammation in liver, lung, and kidney, and completed the study by profiling the antibody and T cell-mediated immune responses to infection in blood and spleen.

MATERIALS AND METHODS

Mice.

C3H/HeJ mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Female mice aged 8 to 16 weeks were kept in a pathogen-free environment within the Laboratory Animal Care Unit of the University of Tennessee Health Sciences Center. Animal experimentation guidelines were followed in compliance with the University of Tennessee Health Science Center Institutional Animal Care and Use Committee (IACUC 14-018).

Bacterial strain and culture conditions.

Leptospira interrogans serovar Copenhageni strain Fiocruz L1-130 was used in this study and was originally isolated from the bloodstream of a leptospirosis patient in Brazil (24). The culture was grown in Probumin vaccine-grade solution (Millipore, Billerica, MA) that had been diluted 5-fold in autoclaved distilled water and 100 μg/ml 5-fluorouracil (MP Chemicals, Santa Ana, CA) at 30°C. L. interrogans spirochetes that had been passaged in hamsters for the retention of virulence (kindly provided by David Haake) were grown and enumerated by dark-field microscopy (Zeiss USA, Hawthorne, NY) and by quantitative PCR. L. interrogans (from the 3rd and 4th passages) was inoculated by intraperitoneal injection.

Experimental infection.

About 10⁶ or 10⁷ L. interrogans spirochetes in 200 μl of culture medium were inoculated intraperitoneally into 10-week-old mice (10⁶ L. interrogans spirochetes, passage 3) or 10-, 12-, or 14-week old mice (10⁷ L. interrogans spirochetes, passage 4). Control mice, which were matched with the L. interrogans-inoculated mice for age, were inoculated with the same volume of sterile culture medium. For 15 consecutive days, the animals were weighed, urine was collected, the urine volume (the number of microliters per animal) was recorded, and the rectal body temperature was measured using a MicroTherma 2T handheld thermometer (Braintree Scientific, Braintree, MA). Urine collection was done daily, before the body weight was recorded, after massaging of the bladder within 5 min of first contact. Blood was collected for 5 consecutive days and every other day between days 6 and 15. Blood and urine were stored at −20°C. On day 15 postinfection, the animals were terminated. Urine, blood, kidney, lung, liver, and spleen were collected for further analysis. The kidneys were placed in culture medium, and the supernatants were analyzed 5 to 7 days later for visual detection of Leptospira by dark-field microscopy using an Axio Zeiss Imager A1 light microscope. Kidney tissue and aliquots of kidney cultures were collected for quantification of the leptospiral 16S rRNA gene by quantitative real-time PCR (qPCR). The liver, lung, and kidney were processed for quantification of proinflammatory mediators by quantitative reverse transcription-PCR (qRT-PCR). The spleen was processed for profiling of T cell populations.

qPCR.

DNA was extracted from blood, urine, and kidney with a DNeasy blood and tissue kit (Qiagen, Valencia, CA), a NucleoSpin tissue kit (Clontech, Mountain View, CA), and a NucleoSpin blood kit (Clontech, Mountain View, CA) according to the manufacturers' instructions. Purified DNA was stored at −20°C until use. Kidney tissue and kidney culture supernatants were used for DNA quantification. Leptospira was quantified using a StepOne Plus qPCR system (Life Technologies, Grand Island, NY) and a 6-carboxyfluorescein–6-carboxytetramethylrhodamine (TAMRA) probe and primers from Eurofins (Huntsville, AL). The Leptospira 16S rRNA gene (rrm locus) was amplified using the primers and a probe described previously (25). The qPCR mixture was calibrated using a known number of heat-killed L. interrogans spirochetes. The PCR mixture contained 25 μM each primer, 100 nM the specific probe, and 2 μl of DNA in a total volume of 20 μl. The amplification protocol consisted of 10 min at 95°C, followed by 40 cycles of amplification (95°C for 15 s and 60°C for 1 min). A negative result was assigned where no amplification occurred or if the threshold cycle (C_T) was greater than 36. Reactions were performed in duplicate for each sample. Results were plotted as the number of leptospires per microliter of urine or blood and per microgram of kidney tissue.

qPCR and qRT-PCR.

Total RNA was extracted from tissues using an RNeasy minikit (Qiagen) and reverse transcribed using a high-capacity cDNA reverse transcription kit (Applied Biosystems). cDNA was subjected to real-time PCR using a StepOne real-time PCR system (Applied Biosystems). qRT-PCR was performed with TAMRA probes (Eurofins Genomics) specific for the keratinocyte-derived chemokine (KC), RANTES, macrophage inflammatory protein 2 (MIP-2), tumor necrosis factor alpha (TNF-α), interleukin-1β (IL-1β), inducible nitric oxide synthase (iNOS), IL-6, gamma interferon (IFN-γ), and β-actin using the comparative C_T method. PCR data are reported as the relative increase in mRNA transcript levels in infected versus control mice corrected for by the respective levels of β-actin, used as an internal standard. The sequences of the probes and primers used are provided in Table S1 in the supplemental material.

Histopathology.

Processing of tissues for histopathology involved formalin fixation, paraffin embedding, sectioning, and staining with periodic acid-Schiff D (PAS-D) in an automated slide processor. We evaluated tissue samples for interstitial inflammation, glomerular size, and tubular damage under an Axio Zeiss Imager A1 light microscope.

Serological assays.

The wells of enzyme-linked immunosorbent assay (ELISA) plates (MaxiSorp; Thermo Fisher, Waltham, MA) were coated with 100 μl of heat-killed L. interrogans bacteria (4 mg/ml) in 100 mM sodium carbonate (pH 9.7). The bacteria were heated for 5 min at 95°C and spun at 12,000 × g. The supernatant was removed, and the pellet was resuspended in phosphate-buffered saline (PBS). The protein concentration was determined using the Lowry assay, and 1 to 5 μg of protein was used to coat each well of the plate. Anti-Leptospira IgM and IgG levels were measured using serum from infected and uninfected mice and horseradish peroxidase-conjugated anti-mouse immunoglobulin secondary antibodies (1:25,000; Santa Cruz Biotechnology, Dallas, TX). Readings were done at 450 nm using a SpectraMax microplate reader (Molecular Devices, Sunnyvale, CA). Furthermore, for the quantitative detection of mouse immunoglobulins, total IgM, IgG, IgG1, and IgG2a concentrations in the sera of mice were determined using Ready-Set-Go ELISA kits according to the manufacturer's instructions (eBioscience).

Immunophenotyping of T cell populations. (i) Cell preparation.

Spleens were harvested from mice, placed in RPMI 1640 (Mediatech, Manassas, VA), and diced with frozen slides in Hanks balanced salt solution medium (Cellgro, Manassas, VA) to produce cell suspensions. Red blood cell lysis was performed using 2 ml/spleen of 1× ACK lysing buffer (Life Technologies). Cells were then washed in RPMI 1640 containing 10% heat-inactivated fetal bovine serum (Atlanta Biotech) and 0.09% sodium azide (Sigma-Aldrich, St. Louis, MO), followed by passage through 70-μm-pore-size and 40-μm-pore-size nylon filters (BD Falcon, Bedford, MA), and the cells were counted using a Luna-FL automated cell counter (Logos Biosystems, Gyunggi-Do, South Korea).

(ii) Flow cytometric cell staining.

Cell viability was analyzed by mixing 18 μl of splenocytes with 2 μl of acridine orange-propidium iodide staining solution in a Luna-FL automated cell counter, and 3 × 10⁶ cells per tube were stained as follows: cells were incubated in 0.5 μg Fc block (BD Biosciences) for 15 min at 4°C in staining buffer (Ca²⁺- and Mg²⁺-free PBS containing 3% heat-inactivated fetal bovine serum, 0.09% sodium azide [Sigma-Aldrich, St. Louis, MO], 5 mM EDTA), washed twice, and incubated with the appropriate marker for surface staining in the dark for 30 min at 4°C. The following are the surface markers used: CD3 clone 17A2 conjugated with fluorescein isothiocyanate (FITC) (1:200; BD Biosciences), CD4 clone RM4-5 conjugated with phycoerythrin (PE) (1:150; BD Biosciences), CD8 clone 53-6.7 conjugated with allophycocyanin (APC)-Cy7 (1:150; BD Biosciences), CD62L clone MEL-14 conjugated with PE-Cy7 (1:150; BD Biosciences), CD44 clone IM7 conjugated with APC (1:150; BD Biosciences), and Pacific blue as the dump channel. For flow cytometry, cells were acquired on an LSR II flow cytometer (BD Immunocytometry Systems, San Jose, CA) equipped with 405-nm, 488-nm, 561-nm, and 640-nm excitation lasers. Data were collected using BD FACSDiva software (BD Biosciences) and analyzed using FlowJo software (TreeStar, Ashland, OR). Fluorescence-minus-one (FMO) controls were used for gating analyses to distinguish positively stained from negatively stained cell populations. Compensation was performed using single-color controls prepared from BD Comp Beads (eBioscience) for cell surface staining. Compensation matrices were calculated and applied using FlowJo software (TreeStar). Biexponential transformation was adjusted manually when necessary. Cells were gated on the basis of their forward and side scatter profiles. Data analyses were performed with acquisition of a minimum of 100,000 events.

Statistical analysis.

Data are averages or means for 4 or 5 mice per group ± standard deviations. Statistical analyses were performed using a two-tailed paired t test (weight loss) and an unpaired t test (T cell responses, mRNA transcript levels in tissues).

RESULTS

Ten- and 12-week-old (but not 14-week-old) mice infected with 10⁷ leptospires shed bacteria in urine as early as 5 days postinfection.

We used 10⁷ L. interrogans bacteria (passage 4) to infect 10-, 12-, and 14-week-old C3H/HeJ mice (Fig. 1) and determined the numbers of copies of the Leptospira 16S rRNA gene in urine by quantitative PCR. Ten-week-old mice started shedding ∼2,500 leptospires in urine on day 5 postinfection, and shedding peaked at day 10 (∼10⁴ L. interrogans 16S rRNA gene copies/μl urine) and continued until termination on day 15 postinfection; 12-week-old mice started shedding ∼2,000 leptospires in urine on day 8 postinfection, and shedding peaked on days 13 to 15 (∼4,000 L. interrogans 16S rRNA gene copies/μl of urine); 14-week-old mice shed negligible numbers of leptospires in urine on day 8 postinfection (<500 L. interrogans 16S rRNA gene copies/μl), and shedding of these numbers of leptospires continued until day 13 (Fig. 1).

FIG 1.

Shedding at 10, 12, and 14 weeks of Leptospira in urine of C3H/HeJ mice infected intraperitoneally with 10⁷ L. interrogans spirochetes (passage 4). Urine was collected daily within 5 min of first contact from day 0 until day 15. Data represent quantitative PCR analysis of the average number of 16S rRNA genes (rrm locus) of L. interrogans per microliter of urine for 4 or 5 mice per group per day. Data represent one of two independent experiments.

Leptospires disseminated from blood to kidney in a biphasic manner, and bacterial shedding persisted after 6 days of infection.

To do a comprehensive analysis of disease progression, we used 10⁶ L. interrogans serovar Copenhageni Fiocruz L1-130 bacteria (passage 3) to infect 10-week-old C3H/HeJ mice. We collected urine and determined the body weight, body temperature, and urine volume on a daily basis for 15 days after infection with 10⁶ spirochetes. Blood was collected daily until day 5 and every other day until day 13. Bacterial DNA was purified from urine and blood, and the number of Leptospira 16S rRNA gene (rrm locus) copies was determined by quantitative PCR. At termination (day 15), the kidneys were collected for histopathology by PAS-D staining, bacterial DNA was purified from the kidney, and the remaining tissue was placed in culture for determination of Leptospira viability. Dissemination of Leptospira in blood was observed by quantification of ∼10⁴ L. interrogans 16S rRNA gene copies on days 4 to 7 postinfection (Fig. 2A). Analysis of our records showed that infected mice produced an average ± standard deviation (SD) of 57 ± 18 μl of urine over the course of infection (days 0 to 14), whereas uninfected controls produced an average ± SD of 70 ± 30 μl. Infected mice started shedding bacteria in urine on day 6 postinfection, and shedding peaked at day 12 (∼10⁶ L. interrogans 16S rRNA gene copies/μl urine) and continued until the mice were terminated (Fig. 2B). Body weight records showed that infected mice lost a significant amount of weight on days 11, 12, and 13 (average, 18.64 g) compared to the amount of weight lost by the controls (average, 20.08 g) (P = 0.0044) (Fig. 3A), with the infected mice losing about 10% of their body weight overall. Body temperature records showed that infected mice became hypothermic on day 12, with temperatures declining to an average of 35.54°C for infected mice from an average of 38.078°C for the controls (Fig. 3B). Histopathological analysis of the kidneys from infected mice by staining with PAS-D showed scattered interstitial mononuclear cells and focal nodular mononuclear cell infiltration, a reduction in glomerular size to about 1/3 of that for the uninfected controls, and tubular damage (Fig. 4A). About 10⁶ leptospires per microgram of kidney tissue were detected by qPCR (Fig. 4B). Tissues from the same kidneys were placed in culture to assess spirochete viability. By dark-field microscopy of kidney cultures, we observed spirochetes with V-shaped flapping and rotationally swimming and forward-swimming live spirochetes 1 week after termination (Fig. 4C). We quantified about 10³ to 10⁶ leptospires per μl of kidney culture (Fig. 4D), which demonstrated colonization of the kidney by live Leptospira at termination. We conclude that doses of 10⁶ and 10⁷ L. interrogans leptospires lead to sublethal infection and kidney colonization of 10- and 12-week-old mice but not 14-week-old mice at 2 weeks after infection.

FIG 2.

Biphasic pattern of Leptospira dissemination: spread into the bloodstream precedes shedding in urine. Groups of 10-week-old C3H/HeJ mice were inoculated intraperitoneally with 10⁶ L. interrogans spirochetes (passage 3). Blood was collected daily from day 0 to day 5 and on alternate days until day 13; urine was collected daily within the 5 min of first contact until day 14. Data represent quantitative PCR analysis of the average number of 16S rRNA genes (rrm locus) of L. interrogans per microliter of blood (A) or urine (B) for 4 or 5 mice per group per day. Data represent one of two independent experiments.

FIG 3.

Weight loss, hypothermia, and urine volume following intraperitoneal infection of 10-week-old C3H/HeJ mice with 10⁶ L. interrogans spirochetes (passage 3). Data are average body weight (A), rectal body temperature (B), and volume of urine collected daily within 5 min of first contact (C) for 4 or 5 mice per group per day. Data represent one of two independent experiments. *, P = 0.0044, paired two-tailed t test.

FIG 4.

Histopathology of the kidney and colonization by live Leptospira at 15 days postinfection. (A) Light microscopy (Zeiss Axio microscope) of PAS-D-stained kidney sections from control and infected mice showing a loss of structure and a reduction of the glomerular size, a marked infiltration of mononuclear lymphocytes, and tubular damage in infected kidney. Magnification, ×40. Arrows, tubules; arrowheads, glomeruli; curved arrow, infiltration of mononuclear lymphocytes. (C) Dark-field microscopy (DFM) of culture of kidney from control and infected mice at 7 days posttermination. (B and D) Data represent quantitative PCR analysis of the number of 16S rRNA genes of L. interrogans per microgram of kidney (B) and per microliter of kidney culture (D) at 7 days posttermination for 4 or 5 mice per group.

Leptospira infection triggers inflammation in the kidney but not in the liver or lung and induces production of IgG1.

Ten-week-old C3H/HeJ mice infected with 10⁶ L. interrogans spirochetes were sacrificed at 15 days postinfection; kidney, liver, and lung tissues were collected and processed for quantification of mRNA of proinflammatory mediators (Fig. 5); and blood was collected to assess the production of Leptospira-specific IgM and IgG. IgG was isotyped for IgG1 and IgG2a (Fig. 6). A stark image of tissue inflammation in the kidney was represented by the significant upregulation of chemokines and cytokines in the infected mice compared to the levels of these immune modulators in the uninfected controls, while the level of inflammation in the liver and lung was not statistically significantly different from that in the uninfected controls. Expression of mRNA for the chemoattractant chemokines produced by neutrophils, monocytes, and memory T helper cells was upregulated in the kidneys of infected mice (for KC/CXCL1, P = 0.0376; for RANTES/CCL5, P = 0.0170; for MIP-2/CXCL2 P = 0.0052). Expression of mRNA for the effector cytokines produced by macrophages and many other cell types was also upregulated in the kidneys of infected mice (for TNF-α, P = 0.0026; for IL-1β, P = 0.0334; for IL-6, P = 0.0257; for IFN-γ, P = 0.0012). In addition, the iNOS enzyme, which is produced by macrophages and is toxic for bacteria, was upregulated only in kidneys from infected mice (P < 0.0001). At 2 weeks postinfection, we detected IgM and IgG specific to Leptospira in serum from infected mice (Fig. 6A). Determination of the overall concentration of immunoglobulin in serum showed that infected mice produced IgG antibodies predominantly of the IgG1 isotype (Fig. 6B).

FIG 5.

Leptospira infection triggers inflammation in the kidney but not in the liver or lung in C3H/HeJ mice. Data represent quantitative real-time and reverse transcription-PCR analyses of mRNA of proinflammatory mediators in the liver, lung, and kidney. Values are mean ± SD mRNA transcript levels (β-actin mRNA was used as an internal standard) for 4 or 5 mice sacrificed at 15 days after intraperitoneal inoculation. Ctrl, uninfected control mice; Inf, infected mice. P values were determined by an unpaired t test. *, P < 0.05; **, P < 0.005; ****, P < 0.0001.

FIG 6.

Leptospira infection induces a Th2-biased antibody response, represented by larger amounts of IgG1 than IgG2a in serum from C3H/HeJ mice at 15 days after intraperitoneal inoculation. (A) Inactivated whole-cell extracts from L. interrogans cultures were coated on ELISA plates, and sera from uninfected (control [Ctrl]) and infected (Inf) mice were analyzed for IgM and IgG antibodies specific to Leptospira. Data are presented as the optical density at 450 nm (OD450) and represent the means ± SDs for 4 mice per group. (B) Total IgM (sIgM), IgG (sIgG), IgG1 (sIgG1), and IgG2a (sIgG2a) were quantified in serum from infected mice. Data are presented as concentrations (in micrograms per milliliter) and represent the means ± SDs for 4 mice per group. P values were determined by a two-tailed unpaired t test. **, P = 0.0028 for IgG1 versus IgG2a.

Leptospira infection induces proliferation of CD4⁺ (but not CD8⁺) and double-negative T cells in spleen.

Infection with 10⁶ Leptospira spirochetes induced splenomegaly at 15 days after infection, as determined by a 2-fold increase in the numbers of splenocytes in the infected group compared to the numbers in the controls (data not shown). Flow cytometric analysis revealed that splenomegaly was caused by expansion of the T cell (CD3⁺) population (Fig. 7A) and showed a significant increase in the percentages of CD4⁺ and double-negative T cells and a decrease in the percentages of CD8⁺ T cells (Fig. 7B to D). We analyzed the frequency of double-negative T cell populations, since these cells have been considered potent effector cells that expand after infection with other bacterial pathogens (26). As shown in Fig. 7D, an increase in the frequency of CD3⁺ double-negative T cells was observed in the Leptospira-infected group in comparison with the frequency in control uninfected mice. The CD4⁺ and CD8⁺ T cell populations were profiled for the expression of activation molecules CD62L and CD44. CD62L is expressed in naive T cells, but its expression declines upon activation. In contrast, CD44 is expressed at low levels on naive T cells, and expression is upregulated during activation of effector T cells. We found that in 10-week-old naive mice, 18 to 20% of splenic CD4⁺ T cells expressed high levels of CD44 and low levels of CD62L. At 2 weeks postinfection with L. interrogans, the frequency of the naive T cell population decreased in the Leptospira-infected group in comparison to the frequency in the uninfected group (Fig. 7E); more than 25% of CD4⁺ T cells acquired a CD44^high CD62L^low effector phenotype (Fig. 7F), and no difference in memory CD4⁺ T cell populations was observed (Fig. 7G). No differences in the naive, effector, and memory CD8⁺ T cell populations were observed (Fig. 7H to J).

FIG 7.

CD3⁺ CD4⁺ effector T cell populations are expanded in the spleens of C3H/HeJ mice infected with 10⁶ L. interrogans spirochetes at 15 days after intraperitoneal inoculation. Freshly isolated splenocytes were blocked with anti-mouse CD16/CD32 antibody, incubated with the surface lineage markers FITC-CD3, PE-CD4, APC-Cy7 CD8, PE-Cy7-CD62L, and APC-CD44, and acquired at 100,000 cells per sample in an LSR II flow cytometer. Data are the mean ± SD percentage of splenocytes for 4 or 5 mice per group; data for early effectors (CD44^low and CD62^low) are not presented. Ctr, uninfected control mice; Inf, infected mice. P values were determined by a two-tailed unpaired t test. *, P < 0.05; **, P < 0.01; ***, P < 0.0005.

DISCUSSION

The development of models of Leptospira infection in host reservoir species as versatile as the mouse allows researchers to leverage off a number of genetic and immunological tools to investigate the pathogenesis of Leptospira and the immune responses to this spirochete under acute, sublethal, and chronic infection conditions. A model of infection in mice older than 10 weeks of age is necessary to understand the immunological mechanisms of vaccine protection.

Leptospira can infect a wide range of mammalian species, and several studies have analyzed a number of pathogenic Leptospira interrogans serovars, infection doses, strains of mice, and mouse ages (8–11, 20–23). Young C3H/HeJ mice (3 weeks old) infected with 10⁷ Leptospira interrogans serovar Copenhageni spirochetes or with Leptospira interrogans serovar Icterohaemorrhagiae spirochetes died within 5 days or 12 days of infection of acute disease, respectively (8, 10). In contrast, 6-week-old mice infected with 10⁷ L. interrogans serovar Icterohaemorrhagiae spirochetes survived for 6 months (8), while infection with the same dose of L. interrogans serovar Copenhageni spirochetes resulted in death from acute disease at 7 days postinfection (10). L. interrogans serovar Manila and L. interrogans serovar Icterohaemorrhagiae isolated from human patients were used to infect 9-week-old C3H/HeJ mice at doses that ranged from 10⁶ to 10⁸ spirochetes. Mice infected with 10⁸ L. interrogans serovar Manila and L. interrogans serovar Icterohaemorrhagiae spirochetes died of acute disease on day 4, and mice infected with 10⁶ L. interrogans serovar Manila spirochetes survived past day 28 but showed signs of chronic disease (9). Others have used 10⁸ L. interrogans serovar Icterohaemorrhagiae spirochetes to induce acute infection in young 3-week-old C3H/HeJ mice and to induce sublethal infection in 6-week old C3H/HeJ mice (11). In our model (Fig. 1), 10- and 12-week-old C3H/HeJ mice but not 14-week-old C3H/HeJ mice infected with 10⁷ L. interrogans serovar Copenhageni spirochetes shed 10³ to 10⁴ copies of Leptospira 16S rRNA per microliter of urine at 9 to 10 days after intraperitoneal infection.

Signs of sublethal leptospirosis have been described in several strains of mice (A, CBA, C57BL/6), as determined by quantification of the leptospires in blood and urine and by determination of weight loss (23), a high burden of leptospires in the kidney and nephritis (20), and fibrosis in the kidney (22). In C3H/HeJ mice, Leptospira infection appears to occur in a biphasic manner. There is an initial period of bacterial dissemination in blood from days 4 to 7 (Fig. 2A), followed by shedding in urine from days 6 to 15 (Fig. 2B) that peaks at day 11 (10⁶ leptospires). Shedding of Leptospira spirochetes in urine is accompanied by a significant loss of body weight on days 11 to 13 postinfection (Fig. 3A) and a decrease in body temperature on day 12 (Fig. 3B), which appear to signal colonization of the kidney. Differences in the numbers of bacteria shed after infection with 10⁶ and 10⁷ leptospires could be related to the number of passages that the cultures were subjected to before infection. The kinetics of Leptospira dissemination in the C3H/HeJ mouse model are similar to the biphasic sublethal signs of leptospirosis seen in the model of chronic infection developed in C57BL/6 mice, where the radiance of bioluminescent leptospires is detected immediately after infection; peak radiance occurs at about day 4; and the radiance declines until day 7, increases exponentially from day 8 to day 15, and remains high until day 142 (23). The differences in the histopathology of the kidney suggestive of nephritis in infected mice (Fig. 4A) are on par with what was previously described in C57BL/6 mice (20) and the tubular injury and vestigial glomeruli seen in C3H/HeJ mice (10, 21). Colonization of the kidney by live leptospires at 2 weeks postinfection was demonstrated by the direct detection of the 16S rRNA gene in kidney tissue and the subsequent recovery and quantification of viable bacteria from kidney culture (Fig. 4B to D). Quantification of the mRNA for proinflammatory mediators (Fig. 5) showed the involvement of an early innate immune response (KC, RANTES, and MIP-2) as well as acute inflammation (TNF-α, IL-1β, and iNOS) and the cytokines involved in adaptive Th2 (IL-6) and Th1 (IFN-γ) immune responses in the kidney but not in the lung or liver and provides molecular support for our histopathological findings and those of other investigators (22). These findings suggest that the colonization of kidney tissue by pathogenic Leptospira leads to significant inflammatory responses, which further modulate the specific adaptive response. The sublethal infection of C3H/HeJ mice with pathogenic Leptospira, documented from day 1 to day 15 postinoculation, could lead to chronic infection, which was not assessed in this study. Ideally, chronic infection of mice would be determined by quantification of Leptospira shedding in urine for 3 to 6 months, as was done for the rat model (16).

Most bacterial lipopolysaccharides (LPSs) are primarily recognized by Toll-like receptor 4 (TLR4), but Leptospira LPSs are primarily recognized by TLR2 in humans (27, 28). Nevertheless, TLR4 also contributes to the activation of immune cells in mice. C3H/HeJ mice are TLR4 deficient and are susceptible to infection with L. interrogans, as shown in this study (Fig. 1 to 7) and in several others (8–10, 20–23). These data suggest that TLR4 plays a role in preventing immune evasion in mice (11, 27). The acquired humoral immune response to Leptospira infection led to the production of IgG mainly of the IgG1 subtype within 2 weeks of infection (Fig. 6), which suggests a Th2 bias. Flow cytometric analysis of splenocytes from infected mice revealed that cellular expansion was primarily due to an increase in the levels of CD3⁺ CD4⁺ cells and double-negative CD3⁺ (not CD3⁺ CD8⁺) cells and that CD4⁺ T cells acquired a CD44^high CD62L^low effector phenotype (Fig. 7) not accompanied by increases in memory T cells. Although a Th2-biased, B cell-mediated immune response may regulate protective immunity or immune evasion (Fig. 6) (27), our data also show that other CD4⁺ effector T lymphocytes promote the immune response to Leptospira (Fig. 7).

Our goal was to develop a mouse model of sublethal infection that mimics the reservoir host carrier status in mice older than age 10 weeks to produce a tool that can be used to investigate new vaccines or other immunotherapeutic compounds while allowing dissection of the immunological mechanisms of protection. Furthermore, comparison of the immune response to Leptospira following vaccination and the immune response induced by naive infection will help us understand the mechanisms of immune evasion and the pathogenesis of L. interrogans.