Rickettsia species are obligate intracellular Gram-negative pathogenic bacteria and the etiologic agents of diseases such as Rocky Mountain spotted fever (RMSF), Mediterranean spotted fever, epidemic typhus, and murine typhus. Genome sequencing revealed that R. prowazekii has ~25 % non-coding DNA, the majority of which is thought to be either “junk DNA” or pseudogenes resulting from genomic reduction. These characteristics also define other Rickettsia genomes. Bacterial small RNAs, whose biogenesis is predominantly attributed to either the intergenic regions (trans-acting) or to the antisense strand of an open reading frame (cis-acting), are now appreciated to be among the most important post-transcriptional regulators of bacterial virulence and growth. We hypothesize that intergenic regions in rickettsial species encode for small, non-coding RNAs (sRNAs) involved in the regulation of its transcriptome, leading to altered virulence and adaptation depending on the host niche. We employed a combination of bioinformatics and in vitro approaches to explore the presence of sRNAs in a number of species within Genus Rickettsia. Using the sRNA Identification Protocol using High-throughput Technology (SIPHT) web interface, we predicted over 1,700 small RNAs present in the intergenic regions of 16 different strains representing 13 rickettsial species. We further characterized novel sRNAs from typhus (R. prowazekii and R. typhi) and spotted fever (R. rickettsii and R. conorii) groups for their promoters and Rho-independent terminators using Bacterial Promoter Prediction Program (BPROM) and TransTermHP prediction algorithms, respectively. Strong σ70 promoters were predicted upstream of all novel small RNAs, indicating the potential for transcriptional activity. Next, we infected human microvascular endothelial cells (HMECs) with R. prowazekii for 3 h and 24 h and performed Next Generation Sequencing to experimentally validate the expression of 26 sRNA candidates predicted in R. prowazekii. Reverse transcriptase PCR was also used to further verify the expression of six putative novel sRNA candidates in R. prowazekii. Our results yield clear evidence for the expression of novel R. prowazekii sRNA candidates during infection of HMECs. This is the first description of novel small RNAs for a highly pathogenic species of Rickettsia, which should lead to new insights into rickettsial virulence and adaptation mechanisms.

MicroRNAs (miRNAs) mediate gene silencing by destabilization and/or translational repression of target mRNA. Infection of human microvascular endothelial cells as primary targets of Rickettsia rickettsii, the etiologic agent of Rocky Mountain spotted fever, triggers host responses appertaining to alterations in cellular gene expression. Microarray-based profiling of endothelial cells infected with R. rickettsii for 3 or 24 h revealed differential expression of 33 miRNAs, of which miRNAs129-5p, 200a-3p, 297, 200b-3p, and 595 were identified as the top five up-regulated miRNAs (5 to 20-fold, p ≤ 0.01) and miRNAs 301b-3p, 548a-3p, and 377-3p were down-regulated (2 to 3-fold, p ≤ 0.01). Changes in the expression of selected miRNAs were confirmed by q-RT-PCR in both in vitro and in vivo models of infection. As potential targets, expression of genes encoding NOTCH1, SMAD2, SMAD3, RIN2, SOD1, and SOD2 was either positively or negatively regulated. Using a miRNA-specific mimic or inhibitor, NOTCH1 was determined to be a target of miRNA 200a-3p in R. rickettsii-infected human dermal microvascular endothelial cells (HMECs). Predictive interactome mapping suggested the potential for miRNA-mediated modulation of regulatory gene networks underlying important host cell signaling pathways. This first demonstration of altered endothelial miRNA expression provides new insights into regulatory elements governing mechanisms of host responses and pathogenesis during human rickettsial infections.

Alveolar type II pneumocytes (ATII) and alveolar macrophages (AM) play a crucial role in the lung’s innate immune response. Burkholderia pseudomallei (BP) and Burkholderia mallei (BM) are facultative Gram-negative bacilli that cause melioidosis and glanders, respectively. The inhalation of these pathogens can cause lethal disease and death in humans. We sought to compare the pathogenesis of and host responses to BP and BM through contact with human primary ATII cells and monocytes-derived macrophages (MDM). We hypothesized that because BP and BM induce different disease outcomes, each pathogen would induce distinct, unique host immune responses from resident pulmonary cells. Our findings showed that BP adhered readily to ATII cells compared to BM. BP, but not BM, was rapidly internalized by macrophages where it replicated to high numbers. Further, BP induced significantly higher levels of pro-inflammatory cytokine secretion from ATII cells (IL-6, IL-8) and macrophages (IL-6, TNFα) at 6h post-infection compared to BM (p<0.05). Interestingly, BM induced the anti-inflammatory cytokine, IL-10, in ATII cells and macrophages at 6h post-infection, with delayed induction of inflammatory cytokines at 24h post-infection. Because BP is flagellated and produces LPS, we confirmed that it stimulated both Toll-like receptor (TLR) 4 and TLR5 via NF-κb activation while the non-flagellated BM stimulated only TLR4. These data show the differences in BP and BM pathogenicity in the lung when infecting human ATII cells and macrophages and demonstrate the ability of these pathogens to elicit distinct immune responses from resident lung cells which may open new targets for therapeutic intervention to fight against these pathogens.

Ehrlichia canis infections have been reported in humans in Venezuela and Costa Rica. In this study, 506 healthy residents and 114 dogs from four municipalities (Cauca, Colombia) were surveyed and blood samples collected. Antibodies to E. canis in human and canine sera were evaluated using the Tandem repeat protein 19 (TRP19) peptide ELISA and indirect immunofluorescence assay (IFA). Ehrlichia canis TRP19 antibodies were detected in only 1/506 human sera, but the single positive sample was negative by IFA. The majority (75/114; 66%) of dogs surveyed had antibodies to the E. canis TRP19 peptide by ELISA, and eight randomly selected sera were further confirmed by E. canis IFA. Genomic DNA samples obtained from 73 E. canis TRP19 ELISA–positive dog blood samples were examined by PCR targeting the 16S ribosomal ribonucleic acid (rRNA) gene. Ehrlichia canis 16S rRNA was amplified in 30 (41%) of the dogs, and 16 amplicons were selected for DNA sequencing, which confirmed that all were E. canis. A second PCR was performed on the 16 confirmed E. canis 16S rRNA PCR–positive samples to determine the TRP36 genotype by amplifying the trp36 gene. TRP36 PCR amplicon sequencing identified nine dogs infected with the U.S. E. canis TRP36 genotype (56%), one dog with the Brazilian genotype (6%), and six dogs with the Costa Rican genotype (38%). Moreover, these molecular genotype signatures were consistent with serologic analysis using TRP36 genotype–specific peptides. Notably, there was no serologic evidence of E. canis infection in humans, suggesting that E. canis infection in dogs in Cauca is not associated with zoonotic human infection.

Microvascular endothelial cells (ECs) represent the primary target cells during human rickettsioses and respond to infection via the activation of immediate–early signaling cascades and the resultant induction of gene expression. As small noncoding RNAs dispersed throughout the genome, microRNAs (miRNAs) regulate gene expression post-transcriptionally to govern a wide range of biological processes. Based on our recent findings demonstrating the involvement of fibroblast growth factor receptor 1 (FGFR1) in facilitating rickettsial invasion into host cells and published reports suggesting miR-424 and miR-503 as regulators of FGF2/FGFR1, we measured the expression of miR-424 and miR-503 during R. conorii infection of human dermal microvascular endothelial cells (HMECs). Our results revealed a significant decrease in miR-424 and miR-503 expression in apparent correlation with increased expression of FGF2 and FGFR1. Considering the established phenomenon of endothelial heterogeneity and pulmonary and cerebral edema as the prominent pathogenic features of rickettsial infections, and significant pathogen burden in the lungs and brain in established mouse models of disease, we next quantified miR-424 and miR-503 expression in pulmonary and cerebral microvascular ECs. Again, R. conorii infection dramatically downregulated both miRNAs in these tissue-specific ECs as early as 30 min post-infection in correlation with higher FGF2/FGFR1 expression. Changes in the expression of both miRNAs and FGF2/FGFR1 were next confirmed in a mouse model of R. conorii infection. Furthermore, miR-424 overexpression via transfection of a mimic into host ECs reduced the expression of FGF2/FGFR1 and gave a corresponding decrease in R. conorii invasion, while an inhibitor of miR-424 had the expected opposite effect. Together, these findings implicate the rickettsial manipulation of host gene expression via regulatory miRNAs to ensure efficient cellular entry as the critical requirement to establish intracellular infection.