Introduction

Lipopolysaccharide (LPS) is a critical component of the outer membrane of Gram-negative bacteria, with functional and structural roles. To date, LPS and the type 4B secretion system (T4BSS) are the only Coxiella burnetii virulence factors currently defined by infection of immunocompetent animals1,2. Indeed, the virulence potential of C. burnetii, the causative agent of human Q fever, in animal models of infection is directly related to LPS length2. In vitro propagation of C. burnetii leads to LPS phase variation, a process that involves the truncation of phase I, smooth LPS into phase II, rough LPS with lipid A, select core sugars but no O-antigen3,4,5. C. burnetii LPS phase variation is associated with passage history, as natural isolates first display phase I LPS that shifts to an intermediate, then truncated (phase II) form as early as 10 passages in embryonated hen’s eggs3,4 or, as tested more recently, the synthetic acidified citrate cysteine medium 2 (ACCM-2)6. The Nine Mile lineage of C. burnetii is commonly utilized for basic experimentation, with Nine Mile I RSA493 (NMI) expressing phase I LPS (fully virulent), Nine Mile Crazy RSA514 (NMC) expressing intermediate, semi-rough LPS (attenuated), and NMII RSA439 (NMII) expressing phase II LPS (avirulent). Moos and Hackstadt demonstrated NMII avirulence by observing a lack of recovery of NMII from the spleens of intraperitoneally infected (≤108 organisms) guinea pigs via passage in embryonated eggs2. Further studies identified a large chromosomal deletion in NMII that irreversibly eliminates genes cbu0678-cbu0698 involved in LPS O-antigen biosynthesis7. The avirulence of NMII paired with the large chromosomal deletion led to the exemption of NMII, clone 4 as a non-select agent by the US Centers for Disease Control and Prevention’s (CDC) Division of Select Agents and Toxins (DSAT) (Fig. S1). A 2018 study by Beare, et al. demonstrated the role of various C. burnetii genes in determining LPS length, including cbu0533, a predicted undecaprenyl-phosphate alpha-N-glucosamine phosphotransferase enzyme that was associated with LPS truncation in the NMII strain of C. burnetii. Here, we describe the natural reversion of cbu0533 mutation and its impact on LPS length.

Results

LPS elongation occurs in C. burnetii NMII, clone 4 in vitro and in vivo

To investigate how LPS length influences virulence, guinea pigs were intraperitoneally infected with one of the following Nine Mile C. burnetii strains: NMI, NMC, or NMII (Fig. 1A). The LPS phase of the initial inoculum used in all experiments, as well as from the outgrowth of splenic strains collected at fourteen days post-inoculation, was determined by silver stain (Fig. S2A) and immunoblot (Fig. S2B). NMII starting inocula did not display intermediate LPS banding or specific intermediate LPS reactivity via silver stain or immunoblot, as previously reported8. Unexpectedly, strains were recovered from 3 of 4 NMII-infected animals and further analyses revealed intermediate LPS reactivity via silver stain and immunoblot analysis (Fig. 1B, C). We coined this phenomenon “LPS elongation”, generating NMII-E (NMII-E1 and NMII-E2 are from two separate splenic cultures). To determine if LPS elongation was limited to in vivo models, long-term axenically passaged NMII was evaluated. NMII was passaged weekly in ACCM-1, with sample collection every 5 passages for 50 weeks (Fig. 1D). LPS analysis revealed the appearance of intermediate LPS reactivity, by passage 15 (silver stain; Fig. 1E) and passage 2 (immunoblot analysis; Fig. 1F). Intermediate LPS reactivity was stable and detectable through the final passage (P51). All described experiments, apart from the initial ACCM-1 long-term study, were performed in BSL-3/ABSL-3 under DSAT select agent guidelines.

Fig. 1: C. burnetii LPS elongation occurs in vivo and in vitro.
figure 1

The in vivo experimental process is described in A. Steps 1 and 2 were conducted with ACCM-1 media. Control and recovered splenic isolate LPS from NMII strain infection are depicted following silver stain (B) and immunoblot (C). Animal group n = 4. The in vitro experimental overview is described in D. LPS profiles of NMC control, NMII inocula, and ACCM-1 passaged strains following ACCM outgrowth are depicted following silver stain (E) and immunoblot with anti-intermediate LPS LPS antibodies (F). Images represent single propagated culture samples. Panels A and D were created with BioRender.com. Source data are provided as a source data file.

Full size image

Cbu0533 mutation reversion causes LPS elongation in C. burnetii NMII, clone 4

Previously it has been shown that mutation of cbu0533, a gene encoding a protein with homology to undecaprenyl-phosphate alpha-N-glucosamine phosphotransferase enzymes, could be responsible for LPS truncation9. To investigate differences in cbu0533 in NMII-derived strains with elongated LPS compared to NMII, whole genome sequencing (WGS) of splenic-recovered (NMI-E1/2) and ACCM-1 strains was performed. Sequencing showed that cbu0533 of both virulent NMI and NMC (100% and 99.7% of reads, respectively) encoded a five leucine stretch beginning at AA168, while NMII unambiguously encoded four leucines, rendering CBU0533 inactive (Table 1). In contrast, two splenic strains exhibiting elongated LPS, NMII-E1 and 2, encoded mixed cbu0533 leucine profiles, with 62.7% and 10.5% encoding five leucine residues, respectively. In ACCM-1 passaged strains expressing elongated LPS, cbu0533 exhibited five leucine reads at a 21.1% (passage 26) and 11.9% (passage 51) frequency in contrast to the NMII inocula which encoded four leucine reads. A NMII RSA439 clone 4 ∆cbu0335 mutant (NMII ∆0335) was analyzed due to its extensive passage history in ACCM-2 (>20 passages) and was used to determine if cbu0533 reversion occurs in ACCM-2-grown bacteria. Like NMII, cbu0533 sequence reads in the NMII ∆0335 strain encoded four leucines. In contrast, 99.7% of NMC reads aligned with the NMI sequence encoding five leucine codons. These cbu0533 sequences corresponded with intermediate and phase II reactivity as determined via silver stain and immunoblot. The presence of large NMII and NMC deletions was confirmed by PCR for all NMII and NMC strains evaluated (Figs. S3 and S4).

Table 1 Next-generation sequencing analysis of various C. burnetii strains following LPS elongation
Full size table

Cbu0533-induced LPS elongation enhances the virulence of C. burnetii NMII, clone 4

To both characterize the phenomenon of LPS elongation and identify functional consequences, we infected guinea pigs with the following C. burnetii strains: NMI, NMC, NMII, recovered splenic strains (NMII-E1 and E2), NMII inocula from the long-term ACCM-1 axenic passaging study (NMII P0), ACCM-1 strain passage 50 (NMII P50), NMI ∆dot/icm (a mutated strain unable to replicate intracellularly and cause disease in vivo1), and NMII ∆0533 (a NMII strain lacking the cbu0533 gene, thus theoretically unable to undergo LPS elongation). Analysis of LPS extracts derived from these inocula revealed expected LPS banding patterns (Fig. S2C, D). NMI exhibited phase I LPS while NMC, NMII-E1, NMII-E2, and ACCM P50 had intermediate-length LPS. Importantly, NMII and NMII ∆0533 strains displayed phase II but not intermediate LPS.

Guinea pig body temperatures were monitored for two weeks following inoculation, and NMI, NMC, and NMII-E1 and E2-infected animals experienced fever (39.5 ◦C) at the low (105 GE) infectious dose (Fig. 2A). As expected, neither NMII, NMI ∆dot/icm, NMII ∆0533-infected, nor saline mock-infected animals experienced fever following inoculation with 105 GE This was also reflected by body temperature change readings normalized to starting body temperature (Fig. 2B). At a higher infective dose (106 GE), NMII P0, NMII P50, and NMII-infected animals also experienced fever (Fig. 2C). NMII passaged in vitro (P50) caused transient, low-grade fever in a few guinea pigs, while NMII and NMII P0-infected animals experienced fever of intermediate magnitude and duration. NMII ∆0533-infected animals did not experience fever (Fig. 2C, D). The inability of NMII ∆0533 to cause disease is notable, as it is unable to elongate its LPS via reversion of cbu0533, directly implicating this gene and its subsequent biological function in restoring virulence to a previously avirulent strain in vivo. In addition to induction of fever, significant weight loss was observed in animals following inoculation with both a high and low dose of NMI (Fig. 2E and Fig. S5A) but was only observed in animals inoculated with NMC, NMII, NMII P0, NMII-E1, or E2 at the high dose (Fig. 2F and Fig. S5B). Similarly, splenomegaly was evidenced in all inoculated animals compared to saline mock-infected animals at the low infection dose, albeit to different magnitudes (Fig. 2G). At the high infective dose, only NMI, NMC, and NMII-E1-inoculated animals exhibited statistically significant splenomegaly compared to saline mock-infected animals (Fig. 2H). Compared to NMI-infected animals, the spleen weights of NMII, NMI ∆dot/icm, NMII∆0533, NMII-E1 and E2, NMII P0, and NMII P50 were significantly decreased. In light of attenuation despite elongated LPS in NMII P50, WGS revealed cbu0113 mutations that increased in frequency as C. burnetii NMII was passaged in ACCM-1 (Table S2). Cbu0113 encodes a T4BSS effector protein10,11 that may be responsible for this finding.

Fig. 2: Guinea pig infection reveals increased virulence following LPS elongation.
figure 2

Guinea pig body temperature (A) and change in body temperature (B) following a 105 GE C. burnetii infectious dose. Guinea pig body temperature (C) and change in body temperature (D) following a 106 GE C. burnetii infectious dose. Fever is indicated by a dotted line at 39.5 °C. Body weight change following a 105 (E) or 106 GE (F) infectious dose at the time point associated with most severe weight loss in the NMI control group (days 5 and 4, respectively). Black dotted lines indicate the mean percentage in saline control groups. For A–F, groups are indicated as follows: closed circle:saline, X:NMI, open square:NMC, open circle:NMII, yellow circle: NMI ∆dot/icm, yellow circle: NMII ∆0533, purple circle:NMII-E1, purple square: NMII-E2. Spleen weight normalized to body weight following a 105 (G) or 106 GE (H) infectious dose following euthanasia (day 14). Gray blocks represent mean saline group values. I Splenic outgrowth following three ACCM-2 passages as assessed via qPCR following a 105 GE infectious dose. J Splenic outgrowth following 3 ACCM-2 passages as assessed via qPCR following a 106 GE infectious dose. K LPS profiles of splenic outgrowth following NMII, NMII P0, and NMI infection (105 GE) as assessed via immunoblot. L LPS profiles of splenic outgrowth following NMII ∆0533, NMI ∆dot/icm, and NMII-E2 infection (106 GE) as assessed by immunoblot. For E–H, differences in group means were assessed using two-sample Welch t-tests, allowing for unequal variances between groups. For each comparison, we computed Wald-type 95% confidence intervals and described statistical significance with two-sided p-values. For all graphs, horizontal bars indicate the group means and error bars represent the group standard deviation. Animal group n = 6. * indicates p < 0.05 and ** indicates p < 0.1. Source data are provided as a source data file.

Full size image

To confirm the presence of LPS elongation, bacteria were axenically propagated from spleen homogenates from infected guinea pigs in ACCM-2 and subsequently harvested for LPS extraction and analysis. C. burnetii was isolated from individual animals from all groups except for saline mock-infected animals and growth was observed in fourteen of fifty-four cultures in the low-dose experiment and twenty-four of forty-eight cultures in the high-dose experiment (Table S1). GE was quantified for splenic cultures upon ACCM-2 passage and corresponded with the presence of splenic outgrowth for the low (Fig. 2I) and high (Fig. 2J) dose groups. The LPS profiles of recovered bacteria were then examined and NMII, NMII P0, NMII P50, NMII-E1, and NMII-E2 strains exhibited intermediate LPS reactivity while NMI strains exhibited phase I LPS banding and reactivity (Fig. 2K and S6A–C). Although animals infected with NMII ∆0533 did not display signs of disease, bacteria could be recovered from the spleen; however, NMII ∆0533 did not display intermediate banding (Fig. S6D) or reactivity (Fig. 2L), consistent with this strain’s inability to produce elongated LPS. Similarly, splenic strains of NMI ∆dot/icm (high dose) exhibited a mixture of LPS including phase I, intermediate, and phase II banding pattern detected by silver stain (Fig. S6D) and specific phase I and intermediate LPS reactivity detected by immunoblot (Fig. 2L).

Discussion

Since the recognition of C. burnetii LPS phase variation over half a century ago, this phenomenon has fundamentally impacted Q fever diagnostics and laboratory research. Coincident with the establishment of LPS length as a virulence factor, the search began for genetic lesions that might account for phase variation. Early efforts revealed a missing HaeIII fragment of NMII relative to NMI12 and it was reasoned that this could be related to LPS phase variation13. Subsequently, the ~4.4 kb HaeIII fragment of NMI was employed in a hybridization scheme to define large and contiguous DNA fragments of ~18 kb and 29 kb missing in NMII and NMC, respectively14. Vodkin and Williams further proposed that a point mutation might be responsible for phase II transition in strains lacking large deletions14. In 2006, comparative genome hybridization of the NMC and NMII genomes revealed the absence of 20 ORFs (cbu0679 to cbu0698, NMII) and 24 ORFs (cbu0676 to cbu700, NMC), as previously described by Hoover et al.7,15 and depicted in Fig. S1. Deleted regions were enriched with genes involved in O-antigen biosynthesis, including those implicated in the biosynthesis of the sugar virenose7,16. Beare, et al. concluded that the large chromosomal deletion explains the intermediate truncation of NMC LPS but does not explain why NMII produces a truncated LPS compared to NMC. Further, the authors posited that NMII likely has an additional point/frameshift mutation, small deletion, or transposon insertion in a gene early in the LPS biosynthetic pathway that was not detected by CGH. In 2017, the NMII genome was sequenced and polymorphisms in several LPS genes were discovered but not discussed in the manuscript17. In 2018, Beare et al. used site-directed mutation and complementation to reveal complex, multistep pathways involved in C. burnetii LPS phase variation9. Our current work reveals a novel determinant of C. burnetii phase variation that affects virulence in a guinea pig infection model. This involves cbu0533 mutation reversion via probable slipped strand mutation, transforming the inactive, NMII gene, to an active gene both in vitro and in vivo (Fig. 3A, B).

Fig. 3: Genetic basis for cbu0533-AA168 mutation and impact on C. burnetii LPS length.
figure 3

Cbu0533 mutation (likely slipped strand) is depicted in A. The impact of cbu0533 slipped strand mutation and reversion on LPS length is depicted in B.

Full size image

Cbu0533 displays homology to E. coli WecA (rfe) and other bacterial undecaprenyl-phosphate alpha-N-glucosamine phosphotransferases. These enzymes belong to the class of polyisophrenyl-phosphate N-acetyl hexosamine-1-phosphate transferases (PNPT) which are integral membrane proteins found in eukaryotes and prokaryotes18. WecA is a common bacterial PNPT that initiates O-antigen synthesis in a variety of species;18 however, cbu0533 does not complement an E. coli ΔwecA mutant, suggesting cbu0533 has C. burnetii-specific functionality19. Deletion of cbu0533 in C. burnetii results in loss of LPS outer core and O-antigen, suggesting cbu0533 catalyzes the first step in outer core biosynthesis. Further, leucine 168 appears to be a critical residue essential for the proper function of the enzyme CBU0533 as it is directly downstream of the region homologous to the predicted active site of E. coli WecA. The observed LPS elongation in NMII-E strains reported here is characterized by reversion of the cbu0533 mutation deleting AA168. Importantly, all phase II and intermediate LPS-expressing C. burnetii strains analyzed retained large chromosomal deletions of LPS-related genes, indicating that phase II and intermediate LPS-expressing strains with large gene deletions in LPS synthesis machinery cannot produce phase I LPS required for full virulence.

Cbu0533-dependent LPS elongation may be related to specific selective pressures. In a guinea pig intraperitoneal infection model, a robust dose of 106 GE may be required as LPS elongation was not detected in animals infected with lower amounts of bacteria. Additionally, LPS elongation has only been observed in ACCM-1 passaged NMII strains. Elevated oxidative stress may provide pressure for LPS elongation, as ACCM-1 is a more oxidative medium than ACCM-D/220. This idea is supported by data indicating that phase II C. burnetii are more readily labeled with lactoperoxidase than phase I C. burnetii21. Here, Hackstadt proposed a model of LPS length-dependent resistance against host clearance pathways such as chemical modification by small molecules including reactive oxygen species. Under certain conditions, short DNA repeats, such as microsatellites, and tandem repeats are susceptible to slipped strand mispairing22. Further, bacterial adaptation to host environments via slipped strand mispairing has been documented for several bacterial species, including Helicobacter pylori23,24 and Neisseria gonorrhoeae25. Mechanisms of phase variation are frequently attributed to stochastic mutations; however, data from a variety of bacteria suggests regulation of these events are influenced by environmental factors25. Indeed, host-derived oxidative stress causes double-stranded DNA breaks in Pseudomonas aeruginosa genes, leading to mutagenic repair mechanisms and resultant bacterial genetic diversity26. The cbu0533-AA168 mutation may be a random event followed by biological selection or may be regulated more directly; thus, induced and random mutational events may serve as a benefit to the bacterium leading to enhanced phenotypic diversity and survival in distinct host environments27.

Beyond cbu0533, additional genes appear to influence C. burnetii virulence during long-term axenic passage. Kersh, et al. observed reduced virulence in a murine infection model following NMI was passaged in ACCM-128. Here, the authors suggested that factors beyond LPS length appear to influence virulence reduction following axenic culture. Our findings bolster this hypothesis, with long-term ACCM-1 passaged strains undergoing LPS elongation with no apparent increase in virulence. WGS revealed cbu0113 mutations that increased in frequency as C. burnetii NMII was passaged in ACCM-1 (Table S2). This gene encodes a T4BSS effector protein10,11 that may be involved in C. burnetii pathogenesis.

NMII has been used in severe combined immunodeficient (SCID) mice for C. burnetii virulence factor evaluation29. Two studies describing the virulence of NMI and NMII in SCID mouse models yielded contrasting findings. Andoh, et al. reported low virulence of NMII30 while Islam, et al. described high virulence of this strain31. Both studies employed the same intraperitoneal infection model and doses of the respective strains; however, the experimental timing was distinct. These disparate results prompt inquiry regarding the potential occurrence of LPS reversion in SCID mice. Further investigation is required to elucidate potential effects on virulence in this context due to LPS reversion. This approach could decipher host factors influencing LPS reversion. Indeed, additional model systems such as mice, Drosophila melanogaster32, and Galleria mellonella33 should be examined for LPS elongation.

The importance of LPS in C. burnetii biology and Q fever pathogenesis is demonstrated not only by reduced virulence of NMII but also by the cbu0533-dependent, virulence-enhancing effect of LPS elongation in NMII-E strains. Due to C. burnetii’s high infectivity, aerosol transmission potential, and a history of causing laboratory-acquired infections34, C. burnetii strains are considered US CDC DSAT select agents and biosafety level 3 (BSL-3) organisms. NMII RSA439, clone 4 is the sole select agent-exempt strain of C. burnetii (https://www.selectagents.gov/exclusions-hhs.html); this is based on the strain’s clonal nature, avirulence, and historic lack of splenic recovery in a guinea pig infection model, and non-revertible nature due to the large chromosomal deletion35. Here, we demonstrate splenic recovery of C. burnetii following infection with NMC or NMII at considerably lower infection doses (105-6 GE). The discrepancy between historic experiments and this study is likely due to variations in the culturing technique, as previously utilized egg yolk cultures may be less sensitive and successful at isolating bacteria from infected tissues compared to optimized axenic growth media. Here, we present the C. burnetii LPS elongation in vitro and in vivo. Our data identify cbu0533 as the gene mediating phase II to intermediate LPS transition and the ensuing increase in virulence potential. The cbu0533 mutation occurring in NMII is characterized by the loss of a leucine-encoding fragment and is likely a slipped strand event. LPS elongation in NMII-E represents mutation reversion to the leucine-encoding region normally lost in wild-type. The resultant impact on LPS length harbors profound implications for laboratory work with C. burnetii, diagnostic sensitivity, and vaccine development. Also, cbu0533 reversion and LPS elongation impart increased virulence potential on NMII strains derived from guinea pigs. Although this prompts regulatory and safety concerns, virulence reestablishment is at an intermediate level. We have observed LPS elongation after two passages in ACCM-1 axenic media and a single infection passage in guinea pigs. Importantly, the guinea pig is considered the most physiologically relevant small animal model of human Q fever due to a dose-fever response similar to that of humans36. Thus, the findings presented here raise concerns regarding the impact of LPS elongation on humans and in additional model systems such as frequently utilized animal model systems and cell culture. As demonstrated here, the reversion of cbu0533 mutation associated with increased virulence may be specific to high-dose guinea pig infections which would justify the exemption status of NMII in the context of cellular culture systems. Although NMII has a decades-long history of safe use at BSL-2 containment we believe it is incumbent that individual laboratories assess whether cbu0533 reversion has occurred in their specific laboratory applications. Our construction of NMII ∆0533 provides an alternative strain that is non-revertible and biologically equivalent to NMII for general laboratory purposes, allowing for the control of LPS elongation and associated virulence enhancement.

Methods

Ethics

All experiments were performed in accordance with the RML Institutional Biosafety Committee and DSAT regulations. Animals were housed in approved animal biosafety level 3 (ABSL-3) facilities and manipulated under ABSL-3 standard operating procedures approved by the Rocky Mountain Laboratories Institutional Biosafety Committee and an Institutional Animal Care and Use Committee-approved protocol (2021-067-E). Animal experiments and procedures were performed in an Association for Assessment and Accreditation of Laboratory Animal Care-accredited National Institutes of Health-National Institute of Allergy and Infectious Diseases animal facility.

Coxiella burnetii strains and propagation

All C. burnetii strains were grown in acidified citrate cysteine medium-1, -2 or -D (ACCM-1, ACCM-2, or ACCM-D)6 at 37 °C, 2.5% O2, and 5% CO2 and were subsequently stored at −80 °C in cell freezing medium (DMEM with 10% fetal bovine serum and 10% dimethyl sulfoxide). The cbu0335 and cbu0533 genes were deleted in NMII RSA439, clone 4 using ACCM-2 and ACCM-D axenic media, respectively, as previously described (2). For experiments based on long-term axenic passaging, ACCM-1 was inoculated with C. burnetii NMII RSSA439, clone 4, and passaged weekly. Samples were then expanded with a single passage in ACCM-2 media and LPS was extracted as described below. For all experiments, wild-type NMII was derived from a Vero cell infected C. burnetii NMII RSA439 strain harvested at 28 days post-inoculation. Oligonucleotide primers and plasmids used in this study are listed in Supplementary Table 3.

Biosafety and dual-use research of concern oversight

All manipulations of phase I and intermediate LPS-expressing C. burnetii and infected animal tissue were performed in a BSL-3 laboratory following standard operating procedures approved by the Rocky Mountain Laboratories Institutional Biosafety Committee. C. burnetii expressing phase I and elongated LPS were handled as select agents as defined by the US Division of Select Agents and Toxins (DSAT). These samples were classified as select agents and were manipulated and stored accordingly. Specifically, axenic NMII RSA439, clone 4 manipulations were performed in BSL-2 containment, and all animal infection studies and downstream analysis were performed in ABSL-3/BSL-3 containment. The work described in this report was evaluated per the HHS policies for dual-use research of concern (DURC) and potential pandemic pathogens (P3CO). The NIH Dual Research of Concern review entity reviewed and approved the project and this manuscript before publication. The committee determined that this research did not meet the definition of DURC because the results, information, and technology are not reasonably anticipated to be directly misapplied to have a significant impact on public health.

Characterization of C. burnetii LPS and relative splenic bacterial load

C. burnetii LPS was extracted by modified hot phenol extraction and visualized by silver stain and immunoblot9. All samples utilized in silver stains and immunoblots were derived from the indicated experiments and were run in parallel. The BioRad Precision Plus Protein Dual Color Standard was utilized as a molecular weight marker in these experiments. C. burnetii stocks used for infection were quantified using qPCR to enumerate genome equivalents (GE)6. For enumeration of relative bacterial load, spleens were homogenized and homogenates were used to inoculate ACCM-2 media as previously described1, with the following modifications. Briefly, 200 uL of spleen homogenate was added to 1.8 mL of ACCM-2 in 6-well plates. After 7 days of growth, C. burnetii DNA was extracted from bacteria contained in 1.4 mL of each ACCM-2 culture using the Qiagen DNeasy Blood and Tissue Kit. Quantification of genome equivalents (GE) in extracted DNA was conducted by qPCR using primers specific for C. burnetii groEL. Ct values over 35 were considered background amplification in this assay.

Guinea pigs

Female Hartley guinea pigs were obtained from Charles River (strain code 051) at 4 to 6 weeks of age. Female animals were utilized to control for known impacts of sex on C. burnetii pathogenesis/virulence and in accordance with historical studies. Animals were acclimated for at least a week before experimental manipulation. To minimize potentially confounding sex-associated factors, female guinea pigs were utilized in these studies. Animals were housed in individually ventilated plastic cages (Allentown; two animals per cage) with hardwood Sani-chip bedding (PJ Murphy). A high-fiber guinea pig diet (Envigo global high-fiber guinea pig diet; Teklad, cat n. 2041) and chlorinated, reverse osmosis filtered tap water were administered ad libitum. A 12-h light–dark cycle was maintained in animal housing facilities which were kept at 68–72 °F and 40–60% relative humidity with a 50% set point. Experimental group numbers ranged from 4 to 6 animals and are described in corresponding figure legends. All animals were housed in approved animal biosafety level 3 (ABSL-3) facilities and manipulated under ABSL-3 standard operating procedures approved by the Rocky Mountain Laboratories Institutional Biosafety Committee and an Institutional Animal Care and Use Committee-approved protocol. Animal experiments and procedures were performed in an Association for Assessment and Accreditation of Laboratory Animal Care-accredited National Institutes of Health-National Institute of Allergy and Infectious Diseases animal facility.

C. burnetii infection, euthanasia, tissue collection, and processing

On the day of infection, animals were sedated by isoflurane inhalation using an anesthetic vaporizer with activated charcoal adsorption filters (VetEquip Inc, cat. n. 901801 and 931401) and an IPTT-300 transponder (BioMedic Data Systems) was implanted subcutaneously above the shoulder of each animal in a longitudinal orientation using a large bore needle. Guinea pigs were infected with 105-6 GE of C. burnetii in USP-grade saline via intraperitoneal injection. Negative control animals were mock infected with USP-grade saline for each experiment. Body weights, body temperatures, and behavioral/clinical changes were recorded daily following infections. Body temperatures were collected using a DAS-8007-P reader (BioMedic Data Systems) at a consistent daily time. A temperature of ≥39.5 °C was defined as fever1. For all experiments, guinea pigs were humanely euthanized on day 14 post-inoculation. Blood was collected by cardiac puncture using Vacutainer® blood collection tubes and needles (BD). Following euthanasia, spleens were excised and placed into conical tubes containing 5 mL sterile phosphate-buffered saline (PBS; Gibco, pH: 7.4, cat. no. 10010023). Spleens were dissociated using disposable 15- or 50-mL tissue grinders (VWR International, cat. n. 47732-446 and -450). Spleen cellularity was determined using a Scepter automated cell counter (Millipore, cat. n. PHCC20060) with size exclusion parameters (6 to 36 μm). Spleen suspensions were stored at −20 °C for subsequent analysis.

Splenic bacterial culture and LPS analysis

Bacteria isolated from spleens were expanded in ACCM-2 as indicated above. For initial cultures, 200 uL of spleen homogenate was added to 1.8 mL of ACCM-2 in 6-well plates. After 7 days of growth, cultures were passaged 1/10 into new 6-well plates. After an additional 7 days of growth, cultures were passed into T25 flasks (1 mL of passage 2 culture into 6 mL ACCM-2) and propagated for an additional 7 days. DNA was extracted from bacteria contained in 1.4 mL of each ACCM-2 culture using the Qiagen DNeasy Blood and Tissue Kit and qPCR was conducted as described above. LPS was extracted and analyzed by silver stain and immunoblot as described above.

PCR and next-generation sequence analysis of C. burnetii strains

Genomic DNA (gDNA) was isolated from C. burnetii strains using the PowerMicrobial Maxi DNA isolation kit (Mo Bio) or Qiagen DNeasy Blood and Tissue kit with an additional step of boiling for 30 minutes before the physical disruption of the cells. Genomic DNA was sequenced using an Illumina MiSeq instrument to generate read pairs as previously described17,37. PCR was performed on isolated gDNA with Accuprime Pfx or Taq (Invitrogen) using the oligonucleotide pairs listed in Supplementary Table 3. Genetic analysis was performed using Geneious software.

Statistical analysis

Statistical analyses were conducted using GraphPad Prism version 7.0 (GraphPad Software, La Jolla, CA, USA). Statistical evidence for differences in group means was assessed using two-sample Welch t-tests, allowing for unequal variances between groups. For each comparison, we computed Wald-type 95% confidence intervals and described statistical significance with two-sided p-values, where we represent p-values as equal to or below 0.05 with a single asterisk (*) and p-values equal to or below 0.01 with a double asterisk (**). For splenic bacterial outgrowth data, we utilized the same analysis after transforming all samples above our qPCR amplification threshold (Ct > 35) to the minimum threshold value. For each comparison, we describe statistical significance with two-sided p-values, where we represent p-values as equal to or below 0.05 with a single asterisk (*), p-values equal to or below 0.01 with a double asterisk (**), and p-values equal to or below 0.001 with a triple asterisk (***).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.