|Home | About | Journals | Submit | Contact Us | Français|
Francisella tularensis is a category A select agent based on its infectivity and virulence but disease mechanisms in Francisella tularensis infection remain poorly understood. Murine pulmonary models of infection were therefore employed to assess and compare dissemination and pathology and to elucidate the host immune response to infection with the highly virulent Type A F. tularensis strain Schu4 versus the less virulent Type B live vaccine strain (LVS). We found that dissemination and pathology in the spleen was significantly greater in mice infected with F. tularensis Schu4 compared to mice infected with F. tularensis LVS. Using gene expression profiling to compare the response to infection with the two F. tularensis strains, we found that there were significant differences in the expression of genes involved in the apoptosis pathway, antigen processing and presentation pathways, and inflammatory response pathways in mice infected with Schu4 when compared to LVS. These transcriptional differences coincided with marked differences in dissemination and severity of organ lesions in mice infected with the Schu4 and LVS strains. Therefore, these findings indicate that altered apoptosis, antigen presentation and production of inflammatory mediators explain the differences in pathogenicity of F. tularensis Schu4 and LVS.
Francisella tularensis causes a fatal disseminated infection upon inhalation. While the majority of clinically diagnosed disease is due to arthropod bites or handling of infectious material, interest in Francisella pneumonic infection has been renewed due to the fear of bioterrorism and its history of weaponization [1, 2]. Pneumonic tularemia is the most severe form of the disease and can result in mortality if treatment is not initiated early in infection. In addition, relapse after the completion of treatment is a primary concern with bacteria of such high virulence [3–5]. Despite a substantial revival in F. tularensis research, the mechanisms of pathogenesis and dissemination remain to be elucidated. Greater knowledge of the host-pathogen interaction will aid in development of protective vaccines and effective chemotherapeutics.
F. tularensis Schu4 and LVS were selected for these studies because of the known difference in virulence associated with these strains, which provides a model with which to assess differences in host interaction and response genes [6–10]. Schu4 and LVS belong to the F. tularensis subspecies tularensis and holarctica respectively [11, 12]. Subspecies tularensis, referred to as type A, represents the most virulent of the Francisella subgroups whereas subspecies holarctica, referred to as type B, tends to be less deadly in humans . LVS is the vaccine strain of F. tularensis, which was derived from a less virulent type B isolate . While F. tularensis LVS retains its virulence in mice, lethal infection requires challenge with greater than103 CFU by the pulmonary route, whereas challenge with Schu4 causes a consistent lethal infection with fewer than 102 CFU [9, 7]. In addition time-to-death in the murine pulmonary infection model differs, with Schu4 infection typically resulting in death by 120 hours post infection, while LVS infected mice survive up to 14 days following infection [13, 7, 9, 14]
To understand how infection with virulent F. tularensis leads to a rapidly disseminating and lethal infection, studies have been performed in a variety of different infection models. In vitro studies aimed at characterizing the transcriptional response to F. tularensis using multiple cell types have revealed some insights into the host pathogen interaction [15–18]. Andersson et al. examined the whole lung transcriptional response to infection with type A Francisella isolate FSC033, and found limited host gene expression in the first 4 days of infection, suggesting a subversion of host recognition and delayed immune responses until immediately before death . Schu4 mutant strains have also been used to assess host-pathogen interactions [19–22]. The mutant bacterial strains are generally less virulent in mouse models of infection. However, studies with mutant strains of F. tularensis have not yet been employed to study the overall host response to infection. Rather, they have been more instrumental in assessing the role of specific bacterial components in establishing infection leading to pathology in the lung.
Monitoring bacterial growth and dissemination along with pathology and in vivo transcriptional profiling of the host response to infection has provided important advances in understanding the host-pathogen interaction for organisms such as Listeria , Mycobacteria , and Yersinia . Therefore, we believe this is also an appropriate technique for assessing the host response to infection in the F. tularensis mouse model of infection. The present work is to our knowledge the first comprehensive comparative study to define the host transcriptional response to F. tularensis infection following dissemination from the lungs to secondary sites of infection. In the present study, bacterial burden was monitored, pathology was assessed, and global gene expression was examined throughout the course of infection with F. tularensis, comparing infection with the Schu4 and LVS strains in a murine model. Here we report significant differences in pathology and regulation of expression of host immune response genes following infection with the Schu4 and LVS strains of F. tularensis.
F. tularensis Schu4 and LVS were provided by Dr. J. Peterson (Centers for Disease Control, Fort Collins, CO). Schu4 and LVS were cultured in modified Mueller-Hinton broth at 37 °C with constant shaking overnight, supplemented with 10% glycerol and aliquoted into 1 ml samples, frozen at −80 °C, and thawed just before use. Frozen stocks were titered by enumerating viable bacteria from serial dilutions plated on modified Mueller-Hinton agar as previously described . The number of viable bacteria in frozen stock vials varied <5% over a 10-month period.
Six week-old female C57BL/6 mice were purchased from Jackson Laboratories, Bar Harbor, Maine. All mice were housed in sterile micro-isolator cages in the laboratory animal resources facility or in the Rocky Mountain Regional Biocontainment Laboratory BSL-3 facility at Colorado State University (Fort Collins, CO) and provided water and food ad libitum. All research involving animals was conducted in accordance with the Animal Care and Use Committee approved animal guidelines and protocols.
Mice were infected with either F. tularensis Schu4 or F. tularensis LVS via intranasal (i.n.) or aerosol routes as described previously [27, 28] depending on the objective of the study. For pathology, qRT-PCR, and bacterial burden studies mice were infected via intranasal route. Mice were anesthetized with ketamine-xylazine and 10 μL inocula was administered to each of the nares in sequential droplets allowing mice to inhale the fluid (20 μL total). Infected mice were monitored for morbidity twice daily and were euthanized at pre-determined endpoints. For global transcriptional profiling, mice were exposed to F. tularensis Schu4 or F. tularensis LVS by exposure in a Glas-Col Inhalation Exposure System (Glas-Col, Inc, Terre Haute, IN). Exposure was conducted by aerosolizing approximately 3.5 × 107 CFU in a volume of 5 cubic feet over a period of 30 min, followed by a 20 minute period of cloud decay.
C57BL/6 mice (n = 4 per group per time point) were infected i.n. with the F. tularensis Schu4 strain (102 CFU) or the F. tularensis LVS strain (104 CFU) and then sacrificed at 48 and 120 hours after exposure. Lung and spleen tissues were removed, divided and placed in 10% neutral buffered formalin for histopathology or in sterile PBS for bacterial quantification. Organs for histopathological examination were fixed, imbedded in paraffin, sectioned, and stained with hematoxylin and eosin.
Samples of lung and spleen tissues were homogenized in 5 mL sterile PBS using a stomacher (Teledyne Tekmar, Mason, OH). Bacterial CFU per mL of organ homogenate were determined by plating serial 10-fold dilutions of organ homogenates on modified Mueller-Hinton agar and incubating at 37 °C for 72 hours. RT-PCR was carried out on RNA samples from the lungs using a 16s primer set and approach adapted from Cole et al. . Relative detection of 16s molecules was determined using the ΔCT method.
RNA was stabilized and recovered from mouse organs by the addition of TRIzol reagent and organic partitioning. Total RNA was extracted from the TRIzol by the addition of chloroform (1:1) to achieve a bi-phase separation, then precipitated by the addition of isopropanol and subjected to Dnase treatment, and purified using a Qiagen RNeasy kit (Valencia, CA). Messenger RNA was converted to cDNA using poly(T) primers and amplified in the presence of modified dUTPs using the AminoAllyl Message Amplification Kit (Ambion, Foster City, CA). Indirect labeling of cDNA for hybridization was conducted by conjugating Cy3 dye with modified dUTPs in a subsequent reaction.
Full mouse genome version 4.0.3 (Operon Biotechnology, Huntsville, AL) cDNA spotted microarrays were obtained from the Genomics Proteomics Core of the Rocky Mountain Regional Center of Excellence (http://www.rmrce.colostate.edu/). The 70mer oligonucleotide cDNAs were printed on poly-amine coated slides (ArrayIt Corporation, Sunnyvale, CA) and post-processed by UV cross linking and blocking with 10% BSA and 3X SSC at 42 °C. Dye coupled cDNA, was combined with yeast tRNA (10 mg/mL), and hybridization buffer (formamide, 20XSSC and 10%SDS) and heated. Single channel (Cy3) hybridization was carried out in triplicate for each sample. Slides were scanned using the Genepix 4000B (Molecular Devices, Sunnyvale, CA) fluorescent scanner and analyzed using Genepix Pro 6.0 software. Background fluorescence was corrected for by subtracting background from foreground intensity values. Technical replicates were averaged before normalizing to the global mean intensity values from the entire data set. Log transformation, t-test, ANOVA, principal component analysis and Benjamini and Hochberg false discovery correction were applied to the data using the Genesifter software (Geospiza, Seattle, WA). Genes considered to be differentially expressed were induced or repressed by 1.5 fold or higher and had a p-value of 0.01 or lower. Clustering was conducted using Cluster software  (http://rana.lbl.gov/eisen/?page_id=42). Functional enrichment analysis was conducted using the DAVID Bioinformatics Database [31, 32] (http://david.abcc.ncifcrf.gov/). Response to each strain was then compared to controls to examine changes in expression of genes during the progression of the infections. The complete dataset is available through the Gene Expression Omnibus (GEO) database using accession # GSE22203.
Quantitative real time PCR was used to assess bacterial burden in infected tissues, validate microarray data, and monitor molecular markers of disease. Briefly, cDNA synthesis from total RNA was carried out using First Strand cDNA Synthesis Kit (Invitrogen, Carlsbad, CA). Briefly, 1 mg of total RNA was combined with random hexamer and oligo (dT) primers and heated in 10ml total volume for 5 minutes. 10 μL of buffered enzyme mix (2 μL 10X buffer, 4 μL MgCl2 (6 mM), 2 μL DTT (0.1 M), 1 μL RNAse out, and 1 μL Superscripttm) was added and incubated at 25 °C for 10 minutes, 50 °C for 50 minutes, and 85 °C for 5 minutes. Platinum SYBR Green qPCR Supermix-UDG (Invitrogen, Carlsbad, CA) was combined with gene specific primers (5nmol) and 50ng of template (cDNA) and run in triplicate on an IQ5 thermocycler (Bio-Rad, Hercules, CA). The transcripts encoding 18S rRNA, GapDH, and β-actin were used to monitor consistency in biological replicates. Other genes described in the text where employed to confirm the expression trends identified by microarray analysis. Resulting data from each condition was compared to controls in an independent fashion using the ΔCT method.
To assess possible differences in dissemination to the spleen, and pathology between F. tularensis strains Schu4 and LVS, mice were infected by the intranasal route. The intranasal route of infection was chosen for the dissemination and pathology studies in order to facilitate equalizing the bacterial burden in the lungs at the 48h time point. To accomplish this, it was necessary to accurately administer higher challenge doses of F. tularensis LVS than for F. tularensis Schu4. In addition, we also employed a higher intranasal challenge dose of F. tularensis LVS because there was minimal lung pathology noted when mice were subjected to low-dose aerosol challenge with F. tularensis LVS (data not shown). Accordingly, by assuring that mice had equivalent bacterial burdens at the appropriate time points after infection, we were able to directly compare the efficiency of bacterial dissemination from the lungs and the associated organ pathology.
Bacterial load in lung and spleen tissue at different times of infection was determined by molecular detection of F tularensis 16S RNA and confirmed by direct plating of organ homogenates for F. tularensis colony detection. At 24 hours after inoculation, the bacterial load of Schu4 and LVS in the lungs was similar based on 16S RNA (Fig. 1A) and colony counting. For example, the lungs contained 4.4 ± 0.30 Log10 CFU Schu4 and 4.9 ± 0.30 Log10 CFU LVS at 24 hours of infection. In contrast, by 120 hours of infection, there was a significantly higher bacterial load in the lungs of mice infected with Schu4 (8.6 ± 0.09 Log10 CFU Schu4 versus 7.3 ± 0.49 Log10 CFU LVS). Although the bacterial load in the spleen was relatively low, Schu4 was detectable by 16S RNA as early as 24 hours after infection, while LVS was not detectable by 16S RNA until 48 hours after infection (Fig. 1B). At 120 hours after infection, Schu4 and LVS were detected in the spleen, though the splenic bacterial burden was significantly (p<0.001) higher in Schu4 infected mice (8.3 ± 0.26 Log10 CFU Schu4 versus 5.8 ± 0.28 Log10 CFU LVS). Overall, Schu4 demonstrated greater growth in the lungs, quicker dissemination to the spleen, and more rapid growth in the spleen compared to LVS. This observation is consistent with the known, more rapid disease progression and virulence of Schu4 compared to LVS [9, 7]. Further these data suggest that rapidity of dissemination to secondary sites is related to the extent of infection in the lungs.
In lung tissues collected from mice 48 hours after infection with either Schu4 or LVS, there was mild to moderate perivascular edema, extravasation of erythrocytes and mild leukocyte margination in pulmonary vessels (Fig. 2A). In addition, there were perivascular interstitial accumulations of granulocytes, monocytes and macrophages, which were accompanied by minimal cell degeneration and necrosis. At 120 hours of infection both F. tularensis Schu4 and LVS infected lungs had pathologic changes consisting of multifocal interstitial edema and increased infiltration of granulocytes and monocytes and macrophages along with intra-alveolar accumulations of macrophages (Fig. 2B). While infection with both Schu4 and LVS induced lung pathology, the lungs from F. tularensis LVS-infected mice lacked abscesses and had less well-defined lesions with minimal necrosis compared to lungs from F. tularensis Schu4-infected mice, which were characterized by prominent abscesses and well-defined lesions with an increased amount of necrosis (Figs. 2B, C).
Spleen tissues from F. tularensis Schu4 and LVS infected mice collected 48 hours after infection were histologically unremarkable and indistinguishable from spleens of uninfected animals (Figs. 2A, B). By 120 hours after infection (Fig. 2C), spleens from F. tularensis LVS infected mice had mild lymphocyte depletion of the white pulp and multifocal accumulations of granulocytes and macrophages in the red pulp, with little evidence of necrosis. In contrast, spleens from F. tularensis Schu4-infected mice had almost complete destruction of parenchymal structures due to diffuse severe necrosis, fibrin-deposition and massive lymphocyte depletion. The marked increase in spleen pathology in F. tularensis Schu4-infected mice was the most notable histological difference between infections caused by the two strains of bacteria.
Whole genome transcriptional profiling of lungs and spleen tissues collected at 12, 24, 48, and 120 hours of infection from mice infected via aerosol with Schu4 or LVS was conducted to investigate the global host response to infection with each bacterium. Low dose aerosol inoculation was used for the transcriptional studies in mice because this route is believed to more closely approximate human infection by inhalation of F. tularensis than other routes of infection. Genes that were considered to be differentially regulated had a variance <0.01 (ANOVA) and were up or down-regulated > 1.5 fold compared to uninfected mice. The complete dataset is available through the Gene Expression Omnibus (GEO) Accession # GSE22203.
The total number of differentially expressed genes in the lung and spleen paralleled the bacterial burden. Infection with Schu4 resulted in differential regulation of 3,958 and 5,442 genes in lungs and spleen respectively, compared to uninfected mice. A similar range of differences in global responses was also observed in LVS infected mice, which resulted in 2,230 differentially regulated genes in the lungs and 9,388 differentially modulated open reading frames in the spleen. Global gene expression response data from all time points of infection with LVS and Schu4 were interrogated to identify ontologies and pathways that were over-represented in the host response to infection (Fig. 3A and 3B). Genes associated with inflammation, host-pathogen interactions, cellular activation/differentiation, host antimicrobial activity, and leukocyte receptor signaling constituted the majority of the host response to infection with both strains of F. tularensis (Table 1).
Infection with either strain resulted in the down-regulation of Il-1β expression immediately following infection in the lungs. Il-1β is a potent inflammatory cytokine and its suppression may be a key mechanism in Francisella infection. Upregrulation of Tgfβ1 and Ptger1 expression was noted 48 hours post infection in Schu4 infected mice, and expression of these immunosuppressive cytokines may be key to the rapid dissemination of Schu4. For example, Tgfβ1 and Ptger1 have both been shown to play a role in the suppression of host defenses in the lungs of LVS infected mice and in human dendritic cells infected with Schu4 [33, 34],. There was also altered expression of several MHC genes and the killer cell lectin-like receptor family genes, including (Ly49/Klra) and H2-Ab1, H2-Bl, H2-M10.2, H2-M3, H2-Q8. The Ly49/Kllra killer cell lectin-like receptors have been shown to be vital for recognition and activation/inhibition of natural killer cells [35, 36]. The fact that Schu4 and LVS infection both decreased the expression of these receptors adds further evidence to the notion that F. tularensis evades the host innate immune response by suppressing key mediators of this response.
Although the overall host response to infection with F. tularensis Schu4 and LVS is similar, unique host transcriptional responses to infection with Schu4 or infection with LVS infection were identified. Further inspection of the transcriptional response to Schu4 revealed notable differences in the transcription of immunologically important genes relative to their expression in LVS-infected mice. These differentially expressed genes included genes encoding components involved in apoptosis, antimicrobial activity, inflammatory response, cellular activation and differentiation, leukocyte receptors, and cell signaling (Table 2).
Genes associated with apoptosis and antimicrobial activity had different expression patterns in Schu4 as compared to LVS and uninfected mice. For example, expression of the pro-apoptotic genes Bad, Bnip2, Bnip3l, Pdcd2, Pdcd4 and Pdcd6, and the anti-apoptotic genes Bcl2 was repressed in Schu4 infected lungs compared to LVS-infected lungs. Similarly, in the spleen there was also repression of apitd1, Bclaf1 and Casp6 expression. Inhibition of apoptosis has been shown to be an important mechanism for replication and survival during infection of other bacteria such as C. burnetii [37–39].
The antimicrobial activity response in the lungs of F. tularensis Schu4 infected mice was dominated by altered expression of Adam2 and Adam9, cathepsin D, L, S and Z, thrombomodulin, thrombospondin 1 and 2, and Timp3. In the spleen, Adam15, Defb1, and Defb21 showed increased expression, while cathepsin B, D and E had significantly reduced expression. The transcriptional response of these genes indicated a reduction in tissue remodeling and breakdown, intracellular protein metabolism, and breakdown of antigenic proteins for MHC-II presentation. Defensins are intrinsically antimicrobial but the isoforms induced during infection have been shown to have little effect on Francisella using human alveolar cells in vitro .
The transcription of CD4, CD52, CD74 (Ii, Invariant chain), and B lymphocyte markers CD37 and CD79B (Igβ) involved in cellular activation and differentiation were uniquely upregulated in Schu4 infection. The increased expression of these particular components involved in MHC-II antigen presentation is consistent with augmented cell-mediated immunity. As antigen presentation is a tightly regulated process , these data in addition to the cathepsin data above may implicate F. tularensis induced alterations in processing and presentation of antigens during infection with F. tularensis Schu4.
There were also important differences in the molecular mediators of the inflammatory response in mice infected with Schu4 compared to LVS infected mice. Expression of the genes for IL-13, IL-13Ra2, CCL2, CCL6, CCL22, and CXCL10 were only induced in the lungs of Schu4 infected mice. Upregulated expression of IL-13 is important because of its role as a Th2 related cytokine, which can be associated with downregulation of Th1 immunity. Upregulation of the chemokine genes suggests that Schu4 infection may lead to increased recruitment of monocytes. A similar trend of altered expression of cytokines and chemokines, specifically IL-13, CCL3, CCR3, CCR5, interferon activated genes, prostaglandin D2 synthase 2, prostacyclin I2, prostaglandin reductase 2, and several Ly6-family genes was observed in the spleen of Schu4 infected mice, albeit later time points in infection. Interferon activated gene families as well as prostaglandin signaling has been shown to be involved in the response to virulent Francisella [13, 42]. We found similar involvement of these pathways in response to Schu4 infection in the mouse spleen. Interestingly, the expression of the T-helper 2 type interleukin IL-13, anti-inflammatory cytokines IL-10 and TGFβ, and the down regulation of the proinflammatory cytokines IL-18 and interferon alpha suggest a disruption in the activation of the protective defenses in Schu4 infection compared to LVS infection.
To confirm the transcriptional response of select immunological genes during F. tularensis infection, quantitative real-time PCR (qRT) was performed on lung and spleen tissue from independent infections (Table 3). Analysis revealed that the trends identified by global microarray analysis were 85% and 62% concordant with qRT data in the lung and spleen, respectively. The lower concordance noted in the spleen is attributed to temporal differences in dissemination. The expression of the 12 key pro-inflammatory and anti-inflammatory markers in the lung and spleen was limited in the initial 24 hours of infection. However, 48 hours post infection with LVS shows activation of cytokine and chemokine expression not seen until 120 hours post infection with Schu4. The genes significantly up-regulated as determined by qRT in the lungs during Schu4 infection included the pro-inflammatory chemokines CCL4, CXCL1 and CXCL10; the pro-inflammatory cytokines IL-6 and IL-12a; the gene for inducible nitric oxide, Nos2; and the gene for a type I interferon, IFN-β. A similar trend in the differential expression of these genes was observed in the spleen of Schu4-infected mice at 120 hours. This trend in transcriptional activity indicates a delayed and reduced host response to infection with Schu4 and is consistent with a lack of host recognition or active mechanism of host-response suppression by the Schu4, consistent with previous reports [43, 26, 34].
A critical question in understanding F. tularensis pathobiology is to determine which critical host responses are altered during the first 4–5 days following infection. Whole genome microarrays are an established post-genomic approach that allow the assessment of global host responses in an unbiased fashion. In the present study, we coupled whole genome microarray analysis with analysis of tissue pathology and organ bacterial burden to gain a more complete understanding of disease progression and host response to infection with a fully virulent and a less virulent strain of F. tularensis. By means of this combined approx we were able to identify important host response differences to infection with the two strains of F. tularensis.
Quantification of bacterial burden in the lungs revealed that Schu4 had increased growth compared to LVS, such that by 120 hours the bacterial load of Schu4 in the lungs significantly exceeded that of mice infected with the LVS strain. In addition, F. tularensis Schu4 demonstrated increased dissemination to the spleen, as indicated by detection within 48 hours of infection and significantly increased bacterial burden in the spleen at later time points following infection. Tissue damage was markedly more severe in the spleen following infection with Schu4, particularly at later time points of infection. Notably, both Schu4 and LVS established similar levels of infection in the lung, but eventually the Schu4 infection progressed to more severe pulmonary pathology, presumably due to more rapid replication and avoidance of host immune responses. Efficient dissemination appears to be an important distinction and hallmark of infection with highly virulent strains of F. tularensis [14, 44, 45]. Importantly, the correlation between controlled dissemination and survival has been observed in drug development studies that indicate that drug efficacy is related to control of dissemination to secondary organs such as the spleen .
Rapid dissemination is an important determinant of disease outcome and likely relies on the initial recognition and control of pathogen replication at the site of infection. A study conducted by Chiavolini et. al showed the importance of the initial inflammatory response in determining survival following F. tularensis infection. For example, survival was predicted by the induction of several inflammatory genes before day 7 of infection with LVS in mice . Since replication of Schu4 was actually higher in the spleens of infected mice than replication of LVS, it is likely that the decrease in cytokine gene expression in the lungs of Schu4 infected mice reflects either failure to activate immune responses, or active immune suppression.
The results of the global analyses of the host response to infection with Francisella Schu4 or LVS strains indicate highly virulent strains are capable of subverting the host innate immune response and cell mediated immunity. In the present study, these altered responses included apoptosis, antigen processing and presentation, the inflammatory response, and leukocyte receptor signaling. The down regulation of multiple host defense mechanisms by F. tularensis is consistent with results reported in previous studies [15, 13, 18, 17, 16][34, 43, 26]. In addition, the transcriptional response to Schu4 and F. tularensis subspecies novicida in human monocytes found that reported that there was less inflammatory gene activation by Schu4 as compared to the less virulent F. novicida strain .
In addition, we found F. tularensis Schu4 versus LVS induced changes in novel gene subsets, particularly IL-13, cathepsins, and most strikingly, the killer cell lectin-like receptor family (Ly49/Klre1). Studies have shown interferon activated macrophages treated with IL-13 have a reduced capacity to inhibit the growth of intracellular bacteria. A previous transcriptional profiling study showed the increased expression of interferon activated genes four days after infection in the lungs of mice infected with Type A FSC033 . Moreover, the killer cell lectin-like receptors have been shown to be vital for recognition and activation/inhibition of natural killer cells [35, 36]. Evidence that infection with F. tularensis Schu4 decreased the expression of these receptors further highlights the immuno-evasive activity of Francisella Schu4 compared to the less virulent F. tularensis LVS. Furthermore, expression of the prostaglandin E1 receptor (Ptger1) confirms recent reports implicating prostaglandin signaling as an important mechanism of Francisella manipulation of the host-response to infection [46, 33, 42].
Our studies also indicate confirm previous studies and indicate that dissemination to secondary sites of infection leading to multi-organ damage and failure are key contributing factors to mortality from F. tularensis infection. We have also identified gene expression patterns that may reflect immune responses to bacterial dissemination from the lung to spleen tissues. These data may also be useful for facilitating the development of diagnostics for monitoring treatment efficacy, the effectiveness of chemotherapeutic or vaccine strategies. For example, gene expression correlates of host evasion during early infection combined with gene expression signatures of dissemination provide a panel of genes that can be used to assess disease progression and severity that can be used as checkpoints of therapeutic efficacy. In addition, as high throughput RNA sequencing becomes more readily available, biomarkers can be correlated to in vivo transcriptional data from the pathogen in an attempt to decipher complex host-pathogen interactions. Importantly the expression of markers that are associated with differences infection with LVS and Schu4 will be useful for assessing immune response to immunotherapeutic drugs. These studies therefore provide a foundation for continued research in this area that will ultimately provide unique opportunities that can be exploited for the development of protective vaccines and effective chemotherapeutics with enhanced efficacy and that prevent relapse of disease.
We are thankful for post-genomics resources and instrumentation, and animal models expertise provided by the Genomics and Proteomics Core and the Animal Models Core in the Rocky Mountain Regional Center of Excellence (AI065357) respectively. We would like to thank Laurel Respicio for technical assistance and reviewing of the manuscript. Funding from the Rocky Mountain Regional Center of Excellence (AI065357) to R.A.S supported this work.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.