Rickettsia massiliae and Rickettsia conorii Israeli Spotted Fever Strain Differentially Regulate Endothelial Cell Responses

Abstract

Rickettsiae primarily target microvascular endothelial cells. However, it remains elusive how endothelial cell responses to rickettsiae play a role in the pathogenesis of rickettsial diseases. In the present study, we employed two rickettsial species with high sequence homology but differing virulence to investigate the pathological endothelial cell responses. Rickettsia massiliae is a newly documented human pathogen that causes a mild spotted fever rickettsiosis. The “Israeli spotted fever” strain of R. conorii (ISF) causes severe disease with a mortality rate up to 30% in hospitalized patients. At 48 hours post infection (HPI), R. conorii (ISF) induced a significant elevation of IL-8 and IL-6 while R. massiliae induced a statistically significant elevated amount of MCP-1 at both transcriptional and protein synthesis levels. Strikingly, R. conorii (ISF), but not R. massiliae, caused a significant level of cell death or injury in HMEC-1 cells at 72 HPI, demonstrated by live-dead cell staining, annexin V staining and lactate dehydrogenase release. Monolayers of endothelial cells infected with R. conorii (ISF) showed a statistically significant decrease in electrical resistance across the monolayer compared to both R. massiliae-infected and uninfected cells at 72 HPI, suggesting increased endothelial permeability. Interestingly, pharmacological inhibitors of caspase-1 significantly reduced the release of lactate dehydrogenase by R. conorii (ISF)-infected HMEC-1 cells, which suggests the role of caspase-1 in mediating the death of endothelial cells. Taken together, our data illustrated that a distinct proinflammatory cytokine profile and endothelial dysfunction, as evidenced by endothelial cell death/injury and increased permeability, are associated with the severity of rickettsial diseases.

About the Authors:

Jeremy Bechelli

Affiliation: Department of Pathology, University of Texas Medical Branch, Galveston, Texas, United States of America

Claire Smalley

Affiliation: Department of Pathology, University of Texas Medical Branch, Galveston, Texas, United States of America

Natacha Milhano

Affiliation: Department of Pathology, University of Texas Medical Branch, Galveston, Texas, United States of America

David H. Walker

Affiliation: Department of Pathology, University of Texas Medical Branch, Galveston, Texas, United States of America

Rong Fang

* E-mail: [email protected]

Affiliation: Department of Pathology, University of Texas Medical Branch, Galveston, Texas, United States of America

Introduction

Rickettsiae are Gram-negative obligately intracellular bacteria with a predilection for infecting vascular endothelial cells [1]. Rickettsiae primarily target the vascular endothelium of small and medium sized vessels leading to vasculitis and ultimately edema in vital organs. The typical clinical manifestations of infections caused by spotted fever group rickettsiae include fever, rash, and frequently tache noire, and progress to encephalitis and pneumonitis in severe infections [2]. Pathophysiological effects of rickettsial infection on endothelial cells include an increase in vascular permeability, vascular inflammation, and pro-inflammatory cytokine production [3,4]. Rickettsia-infected endothelial cells have been reported to produce several pro-inflammatory cytokines and chemokines such as IL-6 [5], IL-8 and MCP-1 [6]. These responses are associated with the activation of endothelial cells [7] and enhanced endothelial cell permeability [8–10]. Furthermore, human and mouse endothelial cells exhibit anti-rickettsial activity under cytokine stimulation including TNF-α, IFN-γ and RANTES activation. Until now, it has never been completely understood how to distinguish the endothelial cell responses accounting for host immunity from those contributing to disease pathogenesis during rickettsial infection. Answering this question is crucial for better understanding the endothelial cell pathobiology in severe rickettsioses.

The bacterial agents Rickettsia massiliae and R. conorii are two genetically related rickettsial species with significantly different virulence. R. massiliae and R. conorii chromosomes exhibit >98% identity in coding sequence [11]. Interestingly, the clinical consequences of infections caused by R. massiliae dramatically differ from those by R. conorii. Previously believed to be nonpathogenic, R. massiliae has been recently documented to cause human infections that have presented as mild spotted fever rickettsioses in Argentina, France, and Italy [12]. R. conorii is the etiological agent of Mediterranean spotted fever (MSF), which is considered as one of the most severe and life threatening rickettsial infections. Among four strains of R. conorii, the Israeli spotted fever strain of R. conorii (ISF) is believed to be the most virulent with a case fatality rate up to 32.3% in hospitalized patients [13]. Therefore, R. massiliae and R. conorii (ISF) were employed in the present study to investigate the contributions of endothelial cell responses to the pathogenesis of rickettsial diseases.

Moreover, R. massiliae and R. conorii (ISF) occur in the same geographic regions [14]. Because serological cross-reactivity occurs across spotted fever group rickettsiae [15] and the primary means of diagnosis is through serum antibody assays, the accurate distinction between infections caused by these two species requires the identification of the actual infecting bacterium. This cross reactivity between the species and the overlap in geographic distributions highlight the need to better understand the pathological differences between these rickettsial species. A correct diagnosis is critical to predicting the pathological complications that would arise due to infection, and would allow physicians to anticipate complications and the correct response in the clinic.

Vascular endothelial cells perform a number of functions required to maintain homeostasis. In response to inflammatory stimuli, endothelial cells can be activated to gain new functions such as displaying surface adhesion molecules and chemokines that lead to recruitment and activation of circulating leucocytes [16]. However, inflammation can also cause endothelial cell injury, which disrupts these processes and results in endothelial dysfunction and death. Endothelial injury may lead to impairment of the endothelial cell barrier that retains fluid, plasma proteins and leukocytes within the intravascular space, leading to vascular leakiness [17]. In order to illustrate the contribution of endothelial cell responses to the pathogenesis of, and immunity to rickettsial diseases, we compared the responses of human dermal microvascular endothelial cells, HMEC-1, by a highly virulent rickettsial species, R. conorii (ISF), and a less virulent rickettsial species, R. massiliae. We hypothesized that R. conorii (ISF), the causative agent of a severe spotted fever rickettsiosis, would cause a pathological response (endothelial dysfunction) including increased inflammatory cytokines and cell death, while R. massiliae, a pathogen causing a mild MSF-like illness, would activate endothelial cells to secrete chemokines for interacting with immune leukocytes.

Materials and Methods

Rickettsia culture and preparation

R. conorii (Israeli spotted fever strain) was obtained from the American Type Culture Collection (ATCC). For cell culture propagation, rickettsiae were cultivated and maintained in Vero cell culture. R. massiliae was cultured as described previously [18]. After homogenization, rickettsiae were diluted in a 10% suspension of sucrose-phosphate-glutamate buffer (0.218 mM sucrose, 3.8 mM KH₂PO₄, 7.2 mM K₂HPO₄, 4.9 mM mono- sodium glutamic acid, pH 7.0) and stored at -80°C until used. The concentration of rickettsiae was determined by plaque assay and quantitative real-time PCR, described below. Plaque assay for testing the quantity of viable rickettsiae in stocks was performed as described previously [19].

Cell culture and infection

HMEC-1 cells first described by Ades et al. [20] were cultured in MCDB 131 medium (Gibco, Grand Island, NY) supplemented with L-glutamine (10 mmol/L; Gibco), mouse epidermal growth factor (10 ng/mL; BD Bioscience, San Jose, CA), hydrocortisone (1 μg/mL; Sigma Aldrich, St. Louis, MO), and 10% heat-inactivated fetal bovine serum. Cells were grown in a humidified incubator with 5% CO₂ at 37°C and used at approximately 95% confluence. Cells were inoculated at an MOI of 5 based on real-time PCR data. Intracellular growth occurred in a humidified incubator with 5% CO₂ at 34°C until times indicated for each experiment.

Quantification of rickettsial loads by quantitative real-time PCR

The number of rickettsiae in infected HMEC-1 cells was determined by quantitative real-time PCR as previously descried [21]. We first incubated endothelial cells with of the same number of R. massiliae and R. conorii (ISF) for 3 hours to allow the adhesion and internalization of viable infectious bacteria. At the selected time points as indicated, the infected samples and uninfected controls were washed with PBS and incubated with DNase to remove the dead or unattached bacteria. After all the samples were collected, genomic DNA was extracted following the manufacturer’s guidelines (Qiagen, Valencia, CA). The final concentration of DNA was determined by NanoDrop Spectrophotometer (Thermo Scientific, Waltham, MA). Samples were then analysed by quantitative real-time PCR targeting the Rickettsia-specific citrate synthase gene, using primers C5 (forward–GAG AGA AAA TTA TAT CCA AAT GTT GAT) and C6 (reverse–AGG GTC TTC GTG CAT TTC TT), as described previously [21]. Serial dilutions of plasmids containing a single copy of a fragment of R. conorii (Malish 7 strain) citrate synthase gene were prepared. Real-time PCR assays were performed in triplicate using an iCycler IQ from Bio-Rad (Hercules, CA), with each reaction including a non-template control (NTC). The results were expressed as rickettsial citrate synthase copies per ng of DNA.

Quantitative detection of inflammatory cytokines in infected endothelial cells

HMEC-1 endothelial cells were infected with R. massiliae or R. conorii (ISF) as described or left uninfected as a negative control. Total RNA was prepared using Qiagen RNeasy Mini kit (Valencia, CA) following the manufacturer’s recommendations. Reverse transcription (RT) was performed using isolated and DNase-treated RNA with Bio-Rad iScript cDNA synthesis kit (Hercules, CA). The cDNA samples were then tested for the expression of mRNA of cytokines by real-time quantitative RT-PCR using SYBR Green PCR Master Mix on an iCycler IQ (Bio-Rad, Hercules, CA). Samples were run in triplicate, and GAPDH was used as the housekeeping gene control. The sequences for the sense and antisense primers used to quantify mRNA were: GAPDH 5’-GCACCGTCAAGGCTGAGAAC-3’, 5’TGGTGAAGACGCAGTGGA-3’:

IL-8: 5’-AGGGTCTTCGTGCATTTCTT-3’, 5’-ACACTGCGCCAACACAGAAATTA-3’; MCP-1: 5’-TTTGCTTGAAGTTTCACTGGCATC-3’, 5’-GCTCATAGCAGCCACCTTCATTC-3’

The ∆∆Ct method was used for quantification with results expressed as the mRNA relative ratio (2^-∆∆Ct) [22]. In brief, the change in mRNA levels was determined using the following formula: fold-change 2^{[∆Ct(infected) -∆Ct(control)]}, where ∆Ct (control) = threshold cycle (Ct) for target gene mock infection—Ct for β-actin mock infection; ∆Ct (infected) = Ct for target gene infection—Ct for β-actin infection.

Cytokine and chemokine concentrations

Culture supernatants from Rickettsia-infected cells and uninfected control samples were filter-sterilized and stored as aliquots at -80°C. The concentrations of IL-6, IL-8, and MCP-1 protein present in the supernatant were measured by ELISA using Ready-Set-Go ELISA kits (eBioscience, San Diego, CA), following the manufacturer’s instructions. The limits of detection of the ELISA for cytokine concentrations were as follows: IL-6, 2 pg/mL; IL-8, 2 pg/mL; MCP-1, 7 pg/mL. Samples were assayed in duplicate and are presented as the average of ≥ 3 independent experiments. Absorbance values were obtained using a VersaMax ELISA microplate reader (Molecular Devices, Sunnyvale, CA), and the chemokine concentration was calculated from values obtained within the linear range of the standard curve.

Annexin V staining and flow cytometry

HMEC-1 cells were washed with PBS before the addition of 0.025% (wt/vol) trypsin (Gibco, Grand Island, NY) to detach and lift cells from the culture surface. Fresh medium was added to the suspension to inactivate trypsin, and the cells were centrifuged at 330 × g for 10 min at 4°C. The pellet was washed with precooled PBS. Cells were first stained using a fixable LIVE/DEAD Near-IR Dead Cell Stain Kit (Life Technologies, Grand Island, NY) in accordance with the manufacturer’s protocol. Annexin staining solution was prepared using the manufactures provided 10× Annexin V binding buffer (10%), annexin V-PE (2%), and water (88%), and incubated with 1×10⁶ cells for 15 min at room temperature in the dark. Cells were washed in binding buffer, fixed with 1% paraformaldehyde containing binding buffer, and stored at 4°C until subjected to fluorescence-activated cell sorting analysis. Flow cytometry was performed on ⩾10,000 cells with a FACSCalibur laser cytometer (Becton-Dickinson, BD Biosciences, San Jose, CA), and data analysis was performed using FlowJo software.

Measurement of lactate dehydrogenase (LDH) release

Culture supernatants from Rickettsia-infected cells and uninfected controls were filter-sterilized, and used in the assay immediately. LDH release was measured using the CytoTox96^® Assay kit according to the manufacturer’s specifications (Promega, Madison WI). Briefly, 50 μl of supernatant was added to 50 μl of reconstituted substrate mixture and incubated for 30 minutes in the dark. After the addition of stop solution, the absorbance was read at 490 nm using a VersaMax ELISA microplate reader (Molecular Devices, Sunnyvale, CA), and % cytotoxicity was calculated as experimental LDH release/maximum LDH release. Maximum LDH release was generated by incubating HMEC-1 cells with 0.05% Triton X-100 for 2 hours at 37°C and filter sterilizing to remove cell debris.

Measurement of transendothelial resistance (TER) by electric cell-substrate impedance sensing (ECIS)

To determine vascular permeability of HMEC-1 cells, electric cell-substrate impedance sensing was employed using the ECIS Model 1600R (Applied Biophysics, Troy, NY, USA) and 8W10E gold-coated electrodes. Electrode arrays were washed with 10 mM cysteine and coated with 0.1% gelatin (Sigma Aldrich, St. Louis, MO) before plating approximately 1 × 10⁵ cells per well. Cells were allowed to attach before connection to the ECIS monitoring system. The resistance was observed until stabilization with average values being approximately 1,700 Ω. HMEC-1 monolayers were then infected with Rickettsia at an MOI of 5 with fresh medium added every 72 hours until completion of the experiment. Resistance was measured at approximately 4-minute intervals and was calculated by averaging the resistance of ten electrodes in each well. Uninfected controls were analyzed in duplicate, while infected monolayers were studied in triplicate.

Caspase-1 and caspase-4 inhibition

In order to determine the mechanisms of LDH release, we employed the irreversible inhibitors specific for caspase-4 (Z-LEVD-FMK, 30 nM) (AG Scientific, San Diego CA) and caspase-1 (Z-YVAD-FMK, 30 nM) (Enzo Life Sciences, Farmingdale NY). The inhibitor was added to a complete HMEC-1 monolayer for 4 hours before infection with R. conorii (ISF) at an MOI of 5. Medium was changed every 24 hours with fresh medium containing caspase inhibitors.

Statistical analysis

For comparison of mean values of different experimental groups, the one-way analysis of variance was determined using GraphPad Prism software version 5.01. Posthoc group pairwise comparisons were conducted using the Bonferroni procedure and overall α level of significance of 0.05.

Results

Plaque formation and growth kinetics of R. massiliae and R. conorii (ISF) in Vero and endothelial cells

To determine the entry and growth of R. massiliae and R. conorii (ISF) in eukaryotic cells, Vero cell monolayers were inoculated at an MOI of 5 and compared at days 5 and 7 post infection (PI) using standard overlay medium for the detection of plaques. Small foci of cell death were observed in monolayers infected with R. conorii (ISF) at day 3 PI (data not shown), increasing in size at day 5 PI, and by day 7 PI, the monolayer was completely destroyed (Fig 1A). Monolayers infected with R. massiliae showed no visible plaques at day 3 PI, and very small plaques became visible at 5 days PI. At 7 days PI, the monolayer infected with R. massiliae showed more plaques, and the size of each plaque was slightly larger than those at 5 days PI (Fig 1A).

[Figure omitted. See PDF.]

Fig 1. R. massiliae and R. conorii (ISF) infection resulted in plaques in Vero cell monolayers and replicated efficiently in the human dermal microvascular cell line, HMEC-1 cells.

(A) R. massiliae and R. conorii (ISF) were grown in Vero cells at 34°C for 5 and 7 days to quantify the number of plaques present as revealed by crystal violet staining (B). R. massiliae and R. conorii (ISF) were grown in HMEC-1 cells for times indicated. Data are representative of three independent experiments (A and B).

https://doi.org/10.1371/journal.pone.0138830.g001

To establish the growth kinetics of cultured Rickettsia species in human endothelial cells, confluent HMEC-1 cell monolayers were inoculated with an MOI of 5, and the total DNA was extracted at 24, 48, 72, and 96 hours PI. Rickettsia massiliae and R. conorii (ISF) were both able to invade and replicate in the HMEC-1 cells and grow with very similar kinetic curves up to 96 hours PI (Fig 1B). This finding allows for valid comparisons during infection by assuring that any differences observed in pathogenicity between these Rickettsia species would not be due to the quantity of bacteria.

Differential regulation of cytokine and chemokine production in infected endothelial cells by R. conorii (ISF) and R. massiliae

To determine the potential differences in regulation of chemokine and cytokine production by endothelial cells infected with different species of Rickettsia, the levels of IL-8 and MCP-1 mRNA were quantified using RT-PCR at 24 and 48 hours PI. After 24 hours of infection, we did not observe any significant upregulation of MCP-1 or IL-8 mRNA in HMEC-1 cells infected by either species of rickettsiae compared to uninfected controls (Fig 2A and 2B). Interestingly, R. massiliae, but not R. conorii (ISF), induced a greater upregulation of MCP-1 mRNA at 48 hours PI compared to uninfected controls (Fig 2A). In contrast, IL-8 mRNA was dramatically upregulated in response to R. conorii (ISF), but not R. massiliae, at 48 hours PI (Fig 2B).

[Figure omitted. See PDF.]

Fig 2. R. massiliae induced upregulation of MCP-1 mRNA levels and R. conorii (ISF) induced IL-8 mRNA levels in HMEC-1 cells.

Real-time quantitative PCR analysis of the expression of MCP-1 mRNA (A) and IL-8mRNA (B) in HMEC-1 cells infected at an MOI of 5. Bar graphs indicate the average and standard error of three independent experiments. White bars represent uninfected cells, grey bars represent R. massiliae- infected cells, and R. conorii (ISF)-infected cells are represented by black bars. *, p <0.05.

https://doi.org/10.1371/journal.pone.0138830.g002

To further evaluate the cytokine and chemokine secretion by endothelial cells infected with rickettsiae at the post transcriptional level, we performed ELISAs for MCP-1, IL-8, and IL-6. We did not detect any increased secretion of MCP-1, IL-8 or IL-6 by endothelial cells infected with either rickettsial species until 48 hours PI compared to uninfected controls (Fig 3A, 3B and 3C). After 48 hours PI, R. massiliae induced a higher level of MCP-1 in infected HMEC-1 cells compared to uninfected controls, while R. conorii (ISF) did not cause any significant secretion of MCP-1 in these cells (Fig 3A). Infection with R. massiliae did not induce increased secretion of IL-8 and IL-6 at 48 HPI compared to uninfected controls (Fig 3B and 3C). In contrast, R. conorii (ISF) induced significantly elevated levels of both IL-8 and IL-6 at 48 hours PI compared to uninfected controls. Furthermore, there is no significant difference in the amount of MCP-1 or IL-8 between these two rickettsial species (Fig 3A and 3B). The levels of IL-6 induced by R. conorii (ISF) were significantly greater than those by R. massiliae (Fig 3C).

[Figure omitted. See PDF.]

Fig 3. R. conorii (ISF) induced significant secretion of IL-8 and IL-6, while R. massiliae induced significant production of MCP-1 in infected HMEC-1 cells.

The production levels of MCP-1 (A), IL-8 (B) and IL-6 (C) in the supernatant of Rickettsia-infected HMEC-1 cells was assessed by ELISA. Bar graphs indicate the average and standard error of three independent experiments. White bars represent uninfected cells, grey bars represent R. massiliae-infected cells, and R. conorii (ISF)—infected cells are represented by black bars. *, p < 0.05, ** n.s. = non-statistically significant.

https://doi.org/10.1371/journal.pone.0138830.g003

Endothelial cell death induced by R. conorii (ISF) at the late stage of infection

Rickettsiae inhibit apoptosis of infected endothelial cells through NF-κB activation at 24 and 48 HPI [23, 24]. Here, we determined the viability of endothelial cells infected by each rickettsial species using Live/Dead® viability dye. In line with the previous studies, we did not find any significant increase in death of endothelial cells infected with either R. massiliae or R. conorii (ISF) compared to uninfected controls until 72 HPI (Fig 4A). However, at 72 hours of infection, cell death was identified in endothelial cells infected with R. conorii (ISF) (91%). Cells infected with R. massiliae showed levels of cell death similar to uninfected controls (Fig 4A). As a positive control, cells treated with staurosporine also showed significant cell death (85%). To confirm these results, we further determined the cell viability using annexin V staining. Interestingly, endothelial cells infected with R. massiliae for 72 hours showed no increase in the amount of surface exposed phosphatidylserine, which was detected by its binding to the protein annexin V. In contrast, R. conorii (ISF) induced a dramatic shift in the intensity of exposed phosphatidylserine compared to uninfected controls (Fig 4B).

[Figure omitted. See PDF.]

Fig 4. R. conorii (ISF), but not R. massiliae, induced cell death at 72 HPI in HMEC-1 cells: (A) Cells were infected at an MOI of 5 and stained with Live/Dead fixable dye for 30 minutes before fixation in paraformaldehyde.

For flow cytometry experiments, ⩾10,000 cells were analyzed. (B) Annexin V staining of HMEC-1 cells following 72 hours of infection with R. massiliae or R. conorii (ISF). Cell counts were normalized to mode. (C) Lactate dehydrogenase (LDH) activity assay of monolayer supernatants demonstrating significantly increased levels of endothelial cell cytotoxicity after infection with R. conorii (ISF) for 72 hours. Bar graphs (A and C) indicate the average and standard error of three independent experiments. White bars represent uninfected cells, grey bars represent R. massiliae-infected cells, R. conorii (ISF)-infected cells are represented by black bars, and cells treated with staurosporine are represented by the checkered bar. *, p < 0.05.

https://doi.org/10.1371/journal.pone.0138830.g004

To further investigate the endothelial cell death caused by R. conorii (ISF), we analyzed cell culture supernatants for the presence of LDH as an indicator of cytotoxicity. Following 48 hours of infection, no significant increase in cytotoxicity occurred in cells infected with either R. conorii (ISF) or R. massiliae. Interestingly, R. conorii (ISF) induced a significantly greater level of cytotoxicity at 72 HPI (92%) compared to uninfected controls (17%) (Fig 4C). In contrast, R. massiliae infection did not cause any significant LDH release from infected endothelial cells (24%).

Increased endothelial cell permeability caused by infection with R. conorii (ISF) but not R. massiliae

To examine the effect of rickettsial infection on endothelial cell permeability, we used ECIS technology to measure TER in real time. Compared to uninfected controls, R. conorii (ISF) infection caused a marked decrease in electrical resistance across an HMEC-1 monolayer starting at 60 HPI up to 100 HPI (Fig 5A). In contrast, the cellular resistance in R. massiliae-infected endothelial cell monolayers did not show any significant difference from uninfected controls. Medium was changed at 72 HPI causing a spike in resistance before returning back to baseline levels near 1500 Ohms (Fig 5A).

[Figure omitted. See PDF.]

Fig 5. R. conorii (ISF), but not R. massiliae, increased endothelial cell monolayer permeability.

(A) Representative ECIS graph demonstrating the loss of electrical resistance across an HMEC-1 monolayer in real-time. (B) Average resistance of endothelial cell monolayers after infection with rickettsiae for designated time points. Bar graphs (B) indicate the average and standard error of three independent experiments. White bars represent uninfected cells, grey bars represent R. massiliae-infected cells, and R. conorii (ISF)—infected cells are represented by black bars. *, p < 0.05.

https://doi.org/10.1371/journal.pone.0138830.g005

To further determine the difference in endothelial cell permeability caused by different species of rickettsiae, we statistically compared the average resistance of these infected endothelial cell monolayers measured in three independent experiments. At time points consistently used throughout this study, there were no significant difference between R. massiliae-infected cells and uninfected controls (Fig 5B). At 72 and 96 hours of infection with R. conorii (ISF), there was a significant decrease in electrical resistance indicating increased endothelial monolayer permeability. Cells infected with R. massiliae demonstrated a trend of having increased resistance although this was not statistically significant. Thus, increased endothelial cell permeability caused by R. conorii (ISF) at 60 HPI is potentially associated with endothelial cell injury and/or death measured by dead-live cell staining, annexin V staining and LDH release at the late stage of infection.

Endothelial cell death induced by R. conorii (ISF) is partially rescued by the inhibition of caspase-1

Since our results showed that R. conorii-induced endothelial cell death was characterized by a significant release of LDH activity, we further investigated the mechanisms involved in cell injury manifested as release of LDH by these infected HMEC-1 cells. It is known that caspase-1/4 mediates LDH release [25]. To examine the role of caspase-1 or caspase-4 involvement in cell injury in rickettsiae-infected endothelial cells, we examined LDH release caused by R. conorii (ISF) in infected HMEC-1 cells treated with inhibitors of caspase-1 or -4. We did not observe any significant change in LDH release in uninfected cells treated with inhibitors of caspase-1 or caspase-4 compared to untreated cells, which excludes the occurrence of non-specific effects of these inhibitors on cell injury and LDH release. Infection with R. conorii (ISF) caused a significant increase in LDH release compared to uninfected controls (42% and 17%, respectively), consistent with the results described in Fig 4B. Interestingly, as shown in Fig 6, the caspase-1 inhibitor significantly reduced the amount of endothelial cell cytotoxicity during infection with R. conorii (ISF) compared to cells without the inhibitor (33% and 42%, respectively). In contrast, caspase-4 inhibitor did not show a reduction in cytotoxicity (44%). These data suggest that caspase-1, but not caspase-4, partially mediates endothelial cell injury associated with LDH release.

[Figure omitted. See PDF.]

Fig 6. Endothelial cell death induced by R. conorii (ISF) is partially dependent on caspase-1.

Cells were treated with caspase-1 and caspase-4 inhibitors and then infected with R. conorii (ISF) for 72 hours. Fresh medium and caspase inhibitors were added daily, and the removed supernatant was used immediately for the LDH activity assay. The bar graph indicates the average and standard error of three independent experiments. *, p < 0.05, inh = inhibitor, Rc = R. conorii (ISF), casp = caspase.

https://doi.org/10.1371/journal.pone.0138830.g006

Discussion

Rickettsia are known to subvert host cell signaling to avoid cell death early in the infection by prolonging endothelial cell viability through inhibiting apoptosis in a NF-ĸB-dependent manner [26]. This impairment of programed death of the infected cells allows further intracellular rickettsial growth [27]. The later stages of infection involving increased endothelial permeability and cell death are poorly understood. To our knowledge, this study for the first time demonstrated that infection with highly virulent R. conorii (ISF), but not low virulence R. massiliae, led to endothelial cell injury and death, which was associated with increased endothelial permeability. Interestingly, we determined that endothelial cell injury/death caused by R. conorii (ISF) was mediated in part through caspase-1. Moreover, highly virulent R. conorii (ISF) induced a robust pro-inflammatory cytokine production profile by endothelial cells, including IL-8 and IL-6, while low virulence R. massiliae stimulated significant MCP-1 production by infected endothelial cells. These findings suggest that cell injury- and death-associated endothelial dysfunction and pro-inflammatory endothelial responses contribute to the pathogenesis of severe rickettsioses. Therefore, this new information sheds light on how pathological response of endothelial cells mediates the pathogenesis of severe rickettsioses.

Most strikingly, our studies illustrated that highly virulent R. conorii (ISF) induced endothelial injury and cell death during the late stage of infection measured by three independent approaches including annexin V staining, LDH release and fluorescent reactive amines dye (Fig 4A, 4B and 4C). In contrast, the endothelial cell injury and death were not observed in R. massiliae-infected cells throughout the course of infection in vitro. While annexin V staining is a marker of apoptosis, it also occurs in cells undergoing pyroptosis, an inflammatory process of caspase-1-dependent programmed cell death [28]. LDH is a soluble cytosolic enzyme that is released into the culture supernatant following the loss of membrane integrity that results from cell death [29]. LDH release is a useful marker of endothelial injury and was used recently alongside the activation of caspase-1 to distinguish pyroptosis from apoptosis [30, 31]. Most of the studies on endothelial cell survival promoted by rickettsiae focus on the early stage of infection [23, 26], although R. rickettsii was reported to cause cell death on days 5–6 PI [32]. Our research provided convincing evidence that rickettsiae caused endothelial cell death at the late stage of infection. This increased cell death may enhance rickettsial dissemination by forcing viable bacteria to invade neighboring or distant cells. Investigations relating to primary endothelial cell death in vitro and in vivo would provide enhanced understanding of rickettsial pathogenesis and endothelial cell dysfunction during infection.

What are the potential mechanisms by which rickettsiae cause endothelial cell death and injury? It has been considered that the pathogenesis of rickettsial diseases is dependent on bacterial growth, reactive oxygen species produced by endothelial cells, or rickettsial actin-based polymerization causing cell lysis [33]. Our results did not show any significant difference in endothelial cell infection rate and bacterial growth kinetics between highly virulent R. conorii (ISF) and low virulence R. massiliae (unpublished data and Fig 1B). Thus, our findings exclude the possibility that the difference in endothelial cell death and injury caused by these two species of rickettsiae is due merely to the intracellular growth of bacteria. Our study found that inhibition of caspase-1, but not caspase-4, significantly rescued the quantity of endothelial cell death (Fig 6). These results suggest that caspase-1 is, at least partially, involved in mediating endothelial cell death during rickettsial infection. However, whether the endothelial cell death caused by R. conorii (ISF) is pyroptosis or not requires further investigation. It is important to note that inhibition of caspase-1 rescued only a relatively small portion of cells, suggesting that mechanisms other than these associated with caspase-1 are involved in cell injury. It is important to mention that the levels of LDH released by R. conorii (ISF)-infected cells (Fig 4) was much higher than those shown in Fig 6. The culture medium in the experiments using caspase-1/4 inhibitors (Fig 6) was changed every day in order to maintain the concentration and the activity of the caspase inhibitors, while culture medium changes were not performed in experiments in Fig 4 allowing for a greater accumulation of LDH.

Fatal human rickettsioses are characterized by vasogenic cerebral edema and non-cardiogenic pulmonary edema resulting from increased endothelial permeability [34]. It is still not completely understood how rickettsiae cause increased microvascular permeability. Interestingly, only R. conorii (ISF)-infected human endothelial cell monolayers showed significantly increased permeability starting at 72 HPI, when death of these infected endothelial cells occurred (Figs 4 and 5). Endothelial cell barrier disruption or increased permeability has been reported to result from loss of endothelial cell adhesion molecules associated with cytotoxicity [35]. Recent studies also demonstrated that endothelial cell death is associated with pulmonary microvascular albumin leak in sepsis [36]. Thus, endothelial cell death likely plays a role in increased vascular permeability during infection with high virulent rickettsiae.

The secretion of cytokines and chemokines by Rickettsia-infected endothelial cells has been well characterized in in vitro infection models [6], although the effects of these inflammatory responses have never been fully understood. IL-6 and IL-8 have been proposed to be involved in the development of vasculitis induced by rickettsial infection and are secreted in large quantities from HUVECs infected with R. conorii (Malish 7) [37]. Oristrell et al. [5] reported that plasma IL-6 levels are significantly higher in patients diagnosed with MSF than in healthy controls (49.6 pg/ml vs. < 4 pg/ml). Additionally, plasma IL-6 levels are associated with more severe MSF (97.1 pg/ml vs. 32.7 pg/ml). These data are in agreement with Masueto et al, as IL-6 is a potent inducer of the acute phase response [5]. Damas et al. [38] observed that R. conorii induced a greater inflammatory response than R. africae as characterized by increased serum concentrations of IL-8, MCP-1, and soluble forms of adhesion molecules in patients with MSF than African tick bite fever. Clifton et al. [39] also showed that IL-8 is an important neutrophil and monocyte chemokine in R. rickettsii infection in human endothelial cells. Our present study suggests that severe rickettsioses are associated with endothelial pro-inflammatory cytokine production (IL-6 and IL-8), while less pathogenic rickettsial infection of endothelial cells correlates with increased levels of MCP-1. The cytokine profile in infections caused by these two closely related rickettsial species may have further implications in the differential diagnosis of these two rickettsial infections in the future.

Our research suggests that the severe form of Israeli spotted fever compared to the milder Mediterranean spotted fever-like illness caused by R. massiliae is likely due to the enhanced activation of endothelial cells leading to excessive inflammation and greater cell death, leading to increased vascular permeability and ultimately edema in critical organs and death. Because mutant rickettsial strains are unavailable, our study has relied on the comparison of two closely related species with distinct pathological consequences. This results in a limitation of the study in that the underlying mechanism responsible for bacterial pathogenicity are not completely understood. Moreover, several questions still remain unanswered regarding the exact mechanisms of cell death induced by rickettsiae such as the role of reactive oxygen species or lytic enzymes in cell injury. Other potential mechanisms of injury may be associated with the nature of the differences in bacterial pathogenicity. Ongoing work in our laboratory investigating the role of caspase-1 during rickettsial infections is shedding light on the pathogenic mechanisms of severe spotted fever rickettsial infections.

Acknowledgments

This work was supported by the Carmage and Martha Walls Distinguished University Chair in Tropical Diseases and the UTMB Institute for Human Infections and Immunity Endothelial Biology Pilot Grant. Jeremy Bechelli is supported by the UTMB T32-AI060549 Biodefense Training Grant.

Author Contributions

Conceived and designed the experiments: JB CS DHW RF. Performed the experiments: JB CS NM. Analyzed the data: JB CS DHW RF. Contributed reagents/materials/analysis tools: DHW RF. Wrote the paper: JB DHW RF.

Citation: Bechelli J, Smalley C, Milhano N, Walker DH, Fang R (2015) Rickettsia massiliae and Rickettsia conorii Israeli Spotted Fever Strain Differentially Regulate Endothelial Cell Responses. PLoS ONE 10(9): e0138830. https://doi.org/10.1371/journal.pone.0138830

References

1. Walker DH. Rickettsiae. Baron S, ed. Medical Microbiology. 4th ed. University of Texas Medical Branch; 1996.

2. Oteo JA, Portillo A. Tick-borne rickettsioses in Europe. Ticks Tick Borne Dis. 2012;3(5–6):271–8. pmid:23177355

3. Jensenius M, Ueland T, Fournier PE, Brosstad F, Stylianou E, Vene S, et al. Systemic inflammatory responses in African tick-bite fever. J Infect Dis. 2003;187(8):1332–6. pmid:12696016

4. Elghetany MT, Walker DH. Hemostatic changes in Rocky Mountain spotted fever and Mediterranean spotted fever. Am J Clin Pathol. 1999;112(2):159–68. pmid:10439795

5. Oristrell J, Sampere M, Amengual MJ, Font B, Segura F. Plasma interleukin-6 levels in Mediterranean spotted fever. Eur J Clin Microbiol Infect Dis. 2004;23(5):417–8. pmid:15103526

6. Rydkina E, Sahni A, Silverman DJ, Sahni SK. Comparative analysis of host-cell signalling mechanisms activated in response to infection with Rickettsia conorii and Rickettsia typhi. J Med Microbiol. 2007;56(Pt 7):896–906. pmid:17577053

7. Reinhart K, Bayer O, Brunkhorst F, Meisner M. Markers of endothelial damage in organ dysfunction and sepsis. Crit Care Med. 2002;30(5 Suppl):S302–12. pmid:12004252

8. Maruo N, Morita I, Shirao M, Murota S. IL-6 increases endothelial permeability in vitro. Endocrinology. 1992;131(2):710–4. pmid:1639018

9. Yu H, Huang X, Ma Y, Gao M, Wang O, Gao T, et al. Interleukin-8 regulates endothelial permeability by down-regulation of tight junction but not dependent on integrins induced focal adhesions. Int J Biol Sci. 2013;9(9):966–79. pmid:24155670

10. Stamatovic SM, Keep RF, Kunkel SL, Andjelkovic AV. Potential role of MCP-1 in endothelial cell tight junction 'opening': signaling via Rho and Rho kinase. J Cell Sci. 2003;116(Pt 22):4615–28. pmid:14576355

11. Blanc G, Ogata H, Robert C, Audic S, Claverie JM, Raoult D. Lateral gene transfer between obligate intracellular bacteria: evidence from the Rickettsia massiliae genome. Genome Res. 2007;17(11):1657–64. pmid:17916642

12. Cascio A, Torina A, Valenzise M, Blanda V, Camarda N, Bombaci S, et al. Scalp eschar and neck lymphadenopathy caused by Rickettsia massiliae. Emerg Infect Dis. 2013;19(5):836–7. pmid:23697545

13. de Sousa R, Nóbrega SD, Bacellar F, Torgal J. Mediterranean spotted fever in Portugal: risk factors for fatal outcome in 105 hospitalized patients. Ann N Y Acad Sci. 2003;990:285–94. pmid:12860641

14. Harrus S, Perlman-Avrahami A, Mumcuoglu KY, Morick D, Baneth G. Molecular detection of Rickettsia massiliae, Rickettsia sibirica mongolitimonae and Rickettsia conorii israelensis in ticks from Israel. Clin Microbiol Infect. 2011;17(2):176–80. pmid:20331680

15. Maia C, Ferreira A, Nunes M, Vieira ML, Campino L, Cardoso L. Molecular detection of bacterial and parasitic pathogens in hard ticks from Portugal. Ticks Tick Borne Dis. 2014;5(4):409–14. pmid:24745731

16. Pober JS. Warner-Lambert/Parke-Davis award lecture. Cytokine-mediated activation of vascular endothelium. Physiology and pathology. Am J Pathol. 1988;133(3):426–33. pmid:2462353

17. Joris I, Cuénoud HF, Doern GV, Underwood JM, Majno G. Capillary leakage in inflammation. A study by vascular labeling. Am J Pathol. 1990;137(6):1353–63. pmid:2260625

18. Milhano N, Saito TB, Bechelli J, Fang R, Vilhena M, DE Sousa R, et al. The role of Rhipicephalus sanguineus sensu lato saliva in the dissemination of Rickettsia conorii in C3H/HeJ mice. Med Vet Entomol. 2015.

19. Walker DH, Popov VL, Wen J, Feng HM. Rickettsia conorii infection of C3H/HeN mice. A model of endothelial-target rickettsiosis. Lab Invest. 1994;70(3):358–68. pmid:7511715

20. Ades EW, Candal FJ, Swerlick RA, George VG, Summers S, Bosse DC, et al. HMEC-1: establishment of an immortalized human microvascular endothelial cell line. J Invest Dermatol. 1992;99(6):683–90. pmid:1361507

21. Labruna MB, Whitworth T, Horta MC, Bouyer DH, McBride JW, Pinter A, et al. Rickettsia species infecting Amblyomma cooperi ticks from an area in the state of São Paulo, Brazil, where Brazilian spotted fever is endemic. J Clin Microbiol. 2004;42(1):90–8. pmid:14715737

22. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402–8. pmid:11846609

23. Bechelli JR, Rydkina E, Colonne PM, Sahni SK. Rickettsia rickettsii infection protects human microvascular endothelial cells against staurosporine-induced apoptosis by a cIAP(2)-independent mechanism. J Infect Dis. 2009;199(9):1389–98. pmid:19358671

24. Joshi SG, Francis CW, Silverman DJ, Sahni SK. NF-kappaB activation suppresses host cell apoptosis during Rickettsia rickettsii infection via regulatory effects on intracellular localization or levels of apoptogenic and anti-apoptotic proteins. FEMS Microbiol Lett. 2004;234(2):333–41. pmid:15135541

25. Sollberger G, Strittmatter GE, Kistowska M, French LE, Beer HD. Caspase-4 is required for activation of inflammasomes. J Immunol. 2012;188(4):1992–2000. pmid:22246630

26. Joshi SG, Francis CW, Silverman DJ, Sahni SK. Nuclear factor kappa B protects against host cell apoptosis during Rickettsia rickettsii infection by inhibiting activation of apical and effector caspases and maintaining mitochondrial integrity. Infect Immun. 2003;71(7):4127–36. pmid:12819104

27. Walker DH. Rickettsiae and rickettsial infections: the current state of knowledge. Clin Infect Dis. 2007;45 Suppl 1:S39–44. pmid:17582568

28. Bergsbaken T, Fink SL, Cookson BT. Pyroptosis: host cell death and inflammation. Nat Rev Microbiol. 2009;7(2):99–109. pmid:19148178

29. Chan FK, Moriwaki K, De Rosa MJ. Detection of necrosis by release of lactate dehydrogenase activity. Methods Mol Biol. 2013;979:65–70. pmid:23397389

30. Rayamajhi M, Zhang Y, Miao EA. Detection of pyroptosis by measuring released lactate dehydrogenase activity. Methods Mol Biol. 2013;1040:85–90. pmid:23852598

31. Drent M, Cobben NA, Henderson RF, Wouters EF, van Dieijen-Visser M. Usefulness of lactate dehydrogenase and its isoenzymes as indicators of lung damage or inflammation. Eur Respir J. 1996;9(8):1736–42. pmid:8866602

32. Silverman DJ. Rickettsia rickettsii-induced cellular injury of human vascular endothelium in vitro. Infect Immun. 1984;44(3):545–53. pmid:6724689

33. Valbuena G, Walker DH. Infection of the endothelium by members of the order Rickettsiales. Thromb Haemost. 2009;102(6):1071–9. pmid:19967137

34. Walker DH, Ismail N. Emerging and re-emerging rickettsioses: endothelial cell infection and early disease events. Nat Rev Microbiol. 2008;6(5):375–86. pmid:18414502

35. McAmis WC, Schaeffer RC, Baynes JW, Wolf MB. Menadione causes endothelial barrier failure by a direct effect on intracellular thiols, independent of reactive oxidant production. Biochim Biophys Acta. 2003;1641(1):43–53. pmid:12788228

36. Gill SE, Taneja R, Rohan M, Wang L, Mehta S. Pulmonary microvascular albumin leak is associated with endothelial cell death in murine sepsis-induced lung injury in vivo. PLoS One. 2014;9(2):e88501. pmid:24516666

37. Mansueto P, Vitale G, Cascio A, Seidita A, Pepe I, Carroccio A, et al. New insight into immunity and immunopathology of Rickettsial diseases. Clin Dev Immunol. 2012;2012:967852. pmid:21912565

38. Damas JK, Davì G, Jensenius M, Santilli F, Otterdal K, Ueland T, et al. Relative chemokine and adhesion molecule expression in Mediterranean spotted fever and African tick bite fever. J Infect. 2009;58(1):68–75. pmid:19091423

39. Clifton DR, Rydkina E, Huyck H, Pryhuber G, Freeman RS, Silverman DJ, et al. Expression and secretion of chemotactic cytokines IL-8 and MCP-1 by human endothelial cells after Rickettsia rickettsii infection: regulation by nuclear transcription factor NF-kappaB. Int J Med Microbiol. 2005;295(4):267–78. pmid:16128401