Recent research milestones in the pathogenesis of human rickettsioses and opportunities ahead

Abstract

Infections caused by pathogenic Rickettsia species continue to scourge human health across the globe. From the point of entry at the site of transmission by arthropod vectors, hematogenous dissemination of rickettsiae occurs to diverse host tissues leading to ‘rickettsial vasculitis’ as the salient feature of pathogenesis. This perspective article accentuates recent breakthrough developments in the context of host–pathogen–vector interactions during rickettsial infections. The subtopics include potential exploitation of circulating macrophages for spread, identification of new entry mechanisms and regulators of actin-based motility, appreciation of metabolites acquired from and effectors delivered into the host, importance of the toxin–antitoxin module in host–cell interactions, effects of the vector microbiome on rickettsial transmission, and niche-specific riboregulation and adaptation. Further research on these aspects will advance our understanding of the biology of rickettsiae as intracellular pathogens and should enable design and development of new approaches to counter rickettsioses in humans and other hosts.

Rickettsia species as pathogenic & endosymbiotic bacteria

The family Rickettsiaceae in the order Rickettsiales now comprises more than 30 named species and subspecies, about half of which are either known pathogens or suspected of causing human disease. By virtue of their natural reservoir niche in arthropod vectors, rickettsiae are also recognized as intracellular endosymbiotic bacteria playing an important role in insect pathology [1]. Pathogenic Rickettsia species capable of causing serious human diseases such as Rocky Mountain spotted fever (RMSF) due to Rickettsia rickettsii, Mediterranean spotted fever caused by R. conorii, epidemic typhus caused by louse-borne R. prowazekii and murine typhus caused by flea-borne R. typhi have generally been investigated in much greater depth as compared with nonpathogenic symbionts [2]. Yet approved vaccines for human use or to prevent these infections in other mammalian hosts such as dogs as companion animals remain elusive [3]. A majority of tick-borne rickettsiae are transmitted to their hosts by hard ticks (Ixodidae), yet available evidence confirms the ability of soft ticks in the family Argasidae to carry and potentially transmit rickettsial species, for example R. bellii (considered to be nonpathogenic to humans) [4,5]. Studies have also established the phenomenon of interspecific competition suggesting a negative impact of rickettsial endosymbionts on the establishment of infection by pathogenic rickettsiae in the tick vector through an interference mechanism preventing the colonization of tick ovaries and the process of transovarial transmission [6]. Rickettsia endosymbionts thus likely play an important role in natural disease transmission via inhibition of vector colonization by pathogens [5,6]. Evidence further suggests an interactive nutritional relationship wherein a symbiont Rickettsia species phylotype G021 capable of de novo folate synthesis potentially serves as a source of essential folate derivatives to its carrier Pacific coast tick Ixodes pacificus [7,8]. A rather intriguing recent finding in this regard is the presence of operational taxonomic units, including multiple strains of Rickettsia and Wolbachia in agricultural spiders, extending the range of insects harboring endosymbiotic rickettsiae [9]. On the other hand, advances in the ease of application of sophisticated molecular tools to epidemiological and/or disease outbreak investigations have also led to the identification of Rickettsia species as either newly emerging or re-emerging pathogens and broadening range of arthropod vectors across the globe [10]. The life cycle of ticks as biological vectors in the natural transmission of rickettsiae to companion animals and human host is summarized in Figure 1.

Figure 1. . Life cycle of the tick vector and natural transmission of rickettsiae.

Red arrows indicate the major steps in the vector life cycle. Infected nymphs feed on either healthy or infected amplifying hosts (e.g., dogs) resulting in the spread and sustainment of infection (green arrows). Blue arrows indicate the transmission of rickettsiae to humans through the nymph or adult tick bite.

Target cells during infection of mammalian hosts

A longstanding and well-established concept in the context of mammalian host interactions and pathogenesis can best be summarized as ‘rickettsial vascultis’ attributed to the endotheliotropic nature of rickettsiae leading to disseminated infection of microvasculature and culminating in vascular inflammation, damage, dysfunction and compromised permeability leading to fluid imbalance and edema of vital organ systems including the lungs and the brain. Such pathologic sequelae are consistently seen in autopsy specimens from fatal human infections and in both murine and nonhuman primate hosts in experimental laboratory models of spotted fever and typhus group rickettsioses [2,11,12]. In cell culture systems, however, a variety of human and other mammalian cells, including Vero cells, L929 fibroblasts, mouse bone-marrow derived macrophages, human embryonic fibroblasts and bladder carcinoma cells are highly susceptible to infection [13]. It is not surprising that rickettsiae, as intracellular parasites, strategically exploit redundant entry mechanisms to achieve entry into host cells via interactions with identified receptors not entirely specific or exclusive to endothelial cells. The biological basis of rickettsial tropism for microvascular endothelial cells, thus, remains an open question. A plausible, but yet to be tested, explanation to this phenomenon may simply be that circulating rickettsiae, which have now been shown to resist complement-mediated killing in the blood [14,15], primarily target the single layer of endothelial cells lining the vessels as one of the very first host niches that they encounter as obligate intracellular pathogens dependent on a viable host environment. This possibility is amply supported by investigations on samples from both human cases and animal models of infection, wherein rickettsiae have been observed in other host cells, such as hepatocytes and Kupffer cells in the liver and macrophages [16,17]. Another important point of consideration in this regard is that R. akari predominantly targets CD68-positive macrophages in the eschars at the sites of rickettsial inoculation [18], and evidence further documents only low levels of cytotoxicity in mouse-derived macrophages and differential cytokine secretion responses to in vitro infection with R. akari and R. typhi [19]. Similar to R. akari, R. parkeri has also been visualized in mononuclear phagocytes in the eschar at the site of tick bite [20]. Curto et al. have more recently implicated an important role for macrophages in the determination of rickettsial virulence based on the invasion, growth and replication of pathogenic R. conorii in both nonphagocytic epithelial cells and phagocytic human macrophage-like THP1 cells. In contrast, R. montanensis (considered to be only mildly pathogenic or nonpathogenic) survives and grows well in nonphagocytic host cells, but not so in THP1 macrophages due to association with and destruction in lysosomal compartments [21]. The follow-up mechanistic studies further reveal differences in the proteomic profiles of macrophages infected with R. conorii vis-à-vis R. montanensis, marked by a metabolic rewiring of host cells toward an M2-like phenotype characterized by an anti-inflammatory signature during R. conorii infection to allow for its growth and replication [22,23]. In agreement with the established theme of manipulating host cell death mechanisms to enable their parasitic intracellular lifestyle, as has been discussed during R. rickettsii infection of human endothelial cells in a previous issue of Future Microbiology [13], this group of investigators further suggests modulation of host cell transcriptional machinery and downstream gene expression profiles enabling R. conorii to maintain a supportive niche for proliferation and spread. An intriguing possibility emerging from these findings and worthy of critical in-depth analysis is the potential for exploitation of circulating macrophages by invading pathogenic rickettsiae to enable their dissemination and spread through the host. Taken together, it is quite likely that all rickettsiae infect mononuclear phagocytes at the cutaneous portal of entry and subsequently target microvascular endothelial cells following hematogenous dissemination.

Mechanisms underlying host cell entry, phagosomal escape & dissemination/spread

Of the 17 rickettsial surface cell antigen (Sca) proteins identified using a combination of bioinformatics and molecular/biochemical approaches, five in spotted fever rickttesiae have been demonstrated to be functionally involved in the interactions with host cell receptors. These include Sca0 (Outer membrane protein A [OmpA]), Sca1, Sca2, Sca4 and Sca5 (OmpB). While Sca0 (OmpA) and Sca5 (OmpB) are constitutively expressed and conserved throughout the spotted fever group, typhus group rickettsiae only express Sca5 (OmpB) on their surface [2,24]. An important breakthrough in host–pathogen interplay enabling rickettsial adhesion and invasion was the identification of Ku70, a component nuclear protein of Ku heterodimer involved in the repair of DNA double-strand breaks and maintenance of genomic integrity but also expressed on the surface of mammalian cells, as the first host cell receptor engaging rickettsial OmpB to promote rickettsial internalization via clathrin-/caveolin-2-mediated endocytosis [25]. This was followed by a report of α₂β₁ integrin binding with rickettsial OmpA to facilitate rickettsial invasion into host cells [26]. Exchange protein directly activated by cyclic-AMP (Epac) has also been implicated in rickettsial entry into host endothelial cells, but the rickettsial ligand for this interaction remains unknown [27]. In addition, rickettsial interactions with yet another host cell receptor, namely FGF receptor-1 (FGFR1), have been implicated in the internalization of spotted fever group organisms into host endothelial cells, wherein rickettsial OmpA (32 kDa peptide) serves as the ligand leading to caveolin-1-dependent endocytosis [28]. Both OmpB and OmpA are processed post-translationally, revealing a 32-kDa fragment at the C-terminal end of the protein and enabling its function as a membrane anchor domain facilitating the interaction with FGFR1 [14,29–31]. More recently, surface-expressed annexin 2 has been determined to be yet another host cell receptor-mediating rickettsial adhesion to vascular endothelial cells through binding interactions with OmpB as the bacterial ligand [32]. After gaining entry into the host cells, rickettsiae break free from membrane-bound endosomes into the cytoplasm by utilizing the activities of phospholipase A2, phospholipase D and hemolysin C [2,12,33]. As intracellular parasites dependent on a viable host environment to fulfill their metabolic needs for growth and replication, it is not surprising that rickettsiae exploit multiple host–cell surface receptor interactions to ensure their internalization and redundant escape mechanisms to gain access to the host cytosol as the source of energy, nutrients and substrates of the pathways of metabolism.

Following invasion into host cells, not all but a majority of pathogenic Rickettsia species accomplish intracellular movements and intercellular spread by forming polar actin tails to propel themselves in a directional manner. The actin-based motility is generally present in the spotted fever group of rickettsiae, but is either absent (e.g., R. prowazekii) or not as effective (e.g., R. typhi) in typhus group species [34–36]. Spotted fever rickettsiae have been shown to utilize two actin-polymerizing proteins for actin-based motility. Of these, RickA functions by activating host Arp2/3 complex and driving motility during the early phase of infection (15–60 min postinfection) [37,38], whereas Sca2 required for motility later during the infection (>8 h postinfection) functions as a mimic of host formins [39–41]. Sca2 is present in most Rickettsia species, but is interrupted in R. peacockii and R. canadensis and truncated in R. prowazekii [39,42]. A recent study further suggests that R. parkeri exploits intercellular tension between target host cells and a mechanotransduction-based mechanism for spreading from cell to cell. In this scenario, Sca4 acts as a secreted effector of spread by promoting the protrusion engulfment via binding to vinculin and blocking the association with α-catenin, suggesting the contribution of cytoskeletal force generation for pathogen transfer from one host cell to another [43]. Interestingly, both RickA and Sca2 in R. parkeri participate in actin polymerization in tick cells, yet the absence of these proteins does not alter its dissemination patterns within the tick vector, suggesting the use of other modes of intercellular spread and dissemination [44].

Host–cell interactions & responses

Infection with pathogenic rickettsiae triggers a multitude of signaling mechanisms and downstream responses in host cells. A significant body of research on this important topic directly pertinent to the understanding of disease pathogenesis and immune mechanisms has justifiably focused on endothelial responses to rickettsial infections. These responses include induction of oxidative stress and antioxidant responses; triggering of transcriptional mechanisms as evidenced by the activation of the ubiquitous transcription factor NF-κB, mitogen-activated protein kinases and JAK-STAT signaling pathway; exploitation of antiapoptotic functions of NF-κB by rickettsiae to enable survival and maintenance of host cell niche early during the infection; and nitric-oxide mediated killing of intracellular rickettsiae by cytokine-activated endothelial cells [12,13]. This body of knowledge corroborates the phenomenon of activation of otherwise quiescent endothelial cells in response to infectious or inflammatory stimuli and the now well-appreciated concept of strategic abilities of intracellular pathogens to subvert or hijack cellular defense mechanisms and to evade immune recognition and clearance, favoring their survival and replication in the hosts [45]. As one would expect, the emphasis of both experimental and clinical research has justifiably aimed on enhancing our understanding of rickettsial pathogenesis and host immunity as well as disease transmission from both established and previously unrealized vectors, including the phenomenon of eschar formation during infection with select Rickettsia species, salient characteristics of the host innate, cell-mediated and antibody responses in humans and experimental animal models, biological basis of pathogen virulence and vector competence, and the determinants of disease severity and outcomes. For a detailed discussion on these fundamentally important topics and in consideration of the prescribed focus on the latest breakthrough developments for this article, we refer the readership to a number of recent, comprehensive review articles from international experts in the field of Rickettsiology [2,5,12,46–49].

A relatively new and emerging area of investigation in host–pathogen interplay is the roles of noncoding RNAs (ncRNAs) in the regulation of gene expression in response to infection. As a consequence of genome sequencing efforts including that of humans, ncRNAs including microRNAs (miRs) and long noncoding RNAs (lncRNAs) have emerged as important regulators of gene expression and critical players in the determination of various immune-related processes, including host responses to infection and vaccination. In this regard, interesting recent findings from our laboratory implicate miR424 and miR503-mediated regulation of the expression of FGF2 and its receptor FGFR1 in rickettsial invasion of host endothelial cells [50]. Similar changes in the expression of both of these miRs and FGF2/FGFR1 are also evident in the lungs, a major target organ in a mouse model of R. conorii infection. Manipulation of the levels of miR424 by either a mimic or an inhibitor further implicates its involvement in the regulation of FGF2/FGFR1 to facilitate R. conorii invasion into host endothelial cells. Further studies to determine the precise roles of differentially regulated miRNAs identified in a profiling study in host–pathogen interplay will provide new information about the roles of novel regulatory elements in governing host responses and investigations of their utility as potential biomarkers of rickettsial infections [51]. Similarly, RNA sequencing-based in-depth profiling of pulmonary lncRNAs in an established murine model during R. conorii infection has led to the identification of two enhancer lncRNAs carrying the potential of regulating inhibitor of DNA binding protein-2 (Id2) and apolipoprotein-10b (Apol 10b) in a target-cell-specific manner [52]. Continuation and expansion of studies on these novel regulatory mechanisms of gene expression, although fraught with difficulties associated with the lack of information on the functions of miRNAs and lncRNAs will constitute a major step forward in advancing our existing knowledge of host–pathogen interactions during rickettsial infections.

Rickettsial effectors

Most pathogenic bacteria manipulate their eukaryotic host cells by secreting virulence related proteins, also known as effectors, into the host microenvironment. As of now, several effector proteins have been described in many pathogenic bacteria including enteropathogenic Escherichia coli and Salmonella, Shigella, Listeria and Mycobacterium species. Among obligately intracellular pathogens in the order Rickettsiales, for example, Anaplasma phagocytophilum and Ehrlichia chaffeensis, the secretion and function of multiple effectors have been unequivocally documented. These effector proteins are not only able to interact with their host counterparts but are also capable of performing post-translational modifications, such as phosphorylation, SUMOylation, ubiquitination and acetylation, of host proteins resulting in the activation and/or inhibition of host immune pathways and responses [53]. On the other hand, molecular and functional characterization of bonafide effectors in Rickettsia species has remained a challenge, due mainly to technical hurdles owing to difficulties associated with genetic manipulation and recovery of viable mutants on a consistent basis. Based on the analysis of complete genome sequences of different Rickettsia species, it is now well established that similar to other bacteria, rickettsial genomes also encode protein secretion machineries. Rickettsial protein secretion systems are categorized into Sec dependent (Sec-TolC and type 5 [T5SS]) and Sec independent (twin arginine translocation [TAT], type 1 [T1SS] and type 4 [T4SS]) systems, and detailed information on the involvement of different rickettsial proteins in the assembly of each secretion apparatus has been captured in a thorough recent review on this topic of high importance [24]. Apart from the rickettsial surface cell antigens (except for sca4), a few effectors involved in actin-based motility (RickA and RalF), escape from phagosome (TlyC, Pld, Pat1 and Pat2), ankyrin repeat-containing proteins (RARP-1 and RARP-2) and toxin–antitoxin system (VapC) are encoded and secreted into the host cytosol by many Rickettsia species [24,54]. It has also been demonstrated that VapBC toxin–antitoxin genes of R. conorii are upregulated during in vitro treatment with ciprofloxacin, suggesting a presumptive role for toxin-antitoxin (TA) modules in determining the toxic effect of fluoroquinolones during rickettsial infections [55].

Among several Scas and surface-exposed proteins identified in Rickettsia species, only five Sca proteins, namely Sca0 (OmpA), Sca1, Sca2 and Sca4 and Sca5 (OmpB), are conserved among most Rickettsia species and characterized to contain an N-terminal Sec signal (with the exception of Sca4) and a passenger domain, and a C-terminal autotransporter domain, and secreted via T5SS [24]. Recent studies have determined that Sca4 of R. parkeri lacking the N-terminal secretion signal is secreted into host cytoplasm by an as yet unknown secretion apparatus [43]. The RickA and RalF are now well established for their association with host actin, activation of Arp2/3 complex and actin-based motility, and actin remodeling resulting from the activation of Arf6 and recruitment of phosphoinositide PI(4,5)P₂ at the host plasma membrane [37,56–60]. Although the mechanism by which RickA is secreted remains to be identified, rickettsial RalF is secreted via T4SS based on its interaction with RvhD4, an ortholog of VirD4 acting as a coupling protein with ATPase activity [57]. The R. prowazekii RP534 was identified as an ortholog of Pseudomonas ExoU, a ubiquitin-activated phospholipase A2 secreted via the type III system and involved in the cleavage of membrane phospholipids and escape from the endosome, and secretion of RP534 is documented [61–63]. Ankyrin repeat-containing proteins (Anks) are also known to modulate host responses and many Ank proteins have been well characterized in obligately intracellular pathogens such as Orientia and Ehrlichia. Genome-wide analysis for repeat regions within the coding genes has identified two rickettsial ankyrin repeat proteins (RARP-1 and RARP-2) to be present in a majority of Rickettsia species. Interestingly, while RARP-1 encodes an N-terminal Sec secretion signal and is secreted via Sec-TolC system, RARP-2 localizes to the host endoplasmic reticulum, is devoid of the Sec signal and has been determined to be a T4SS effector [64,65]. The list of rickettsial effectors identified thus far, their function and secretion system utilized are presented Table 1.

Table 1. . List of rickettsial effectors.

Effector	Function	Secretion system	Ref.
Pat1	Phospholipase activity, and host–cell infection	ND	[62]
Pat2	Phospholipase activity, and host–cell infection	ND	[63]
TlyC	Membranolytic activity and likely involved in endosomal escape of Rickettsia	ND	[33,103]
Pld	Endosomal lysis and escape of Rickettsia into host cytosol	ND	[33]
VapC	Exhibits RNase activity	ND	[54]
RalF	Activation of Arf6, actin remodeling and recruitment of phosphoinositide PI(4,5)P2 at the host plasma membrane	T4SS	[56–58]
RickA	Activation of Arp2/3 complex, and actin-based motility	ND	[37,44]
Sca4	Binds to vinculin and helps in intracellular spread of Rickettsia	ND	[43]
RARP1	Unknown	Sec-TolC dependent	[64]
RARP2	Localizes to host endoplasmic reticulum. Function unknown	T4SS	[65]

ND: Not determined.

Outer membranous vesicles (OMVs), defined as small, spherical and bilayered protein-rich protrusions pinching off from the cell wall, are known to play vital roles in host–pathogen interactions. The biogenesis of OMVs in diverse Gram-negative bacteria including E. coli, Salmonella and Shigella has been extensively studied, and OMVs in these pathogens carry toxins, autolysins, adhesins, invasins and virulence factors (e.g., proteases) required for adhesion, invasion and modulation of host defense and disease pathogenesis [66–68]. Despite published evidence documenting that Orientia and Piscirickettsia are capable of forming OMVs, their biogenesis in other obligate intracellular bacteria has not yet been explored [69,70]. We have recently identified the production of OMVs by R. conorii during in vitro infection of host endothelial cells. In this particular study, a rickettsial peptidoglycan hydrolase (RC0497), encoding an AmpD domain, was identified to be localized to the cell wall of OMVs, suggesting the possibility of its delivery into the host cytosol through vesicles [71]. Importantly, R. conorii RC0497 has also been identified as one of the most abundant rickettsial proteins present in the serum of infected patients [72] and its ortholog in Mycobacterium binds to laminin and fibronectin, host extracellular matrix proteins, indicating its secretion outside the bacterium [73]. Furthermore, quantitative proteomic profiling of endothelial cells during R. conorii infection has identified 34 rickettsial proteins in the golgi, plasma membrane and soluble protein fractions, clearly suggesting their potential secretion by an unknown mechanism into the host milieu during host–pathogen interactions [74]. Together, these studies establish the presence of multiple secreted effectors in rickettsial genomes and support the need for further large-scale studies focused on homology-based modeling and determination of secretion signal signatures and eukaryotic-like functional domains to enable the identification and characterization of novel effector proteins.

Riboregulatory mechanisms & adaptation in different host cells

Bacterial small noncoding RNAs (sRNAs) are now identified in all taxonomic phyla and classified into trans-acting, cis-acting, untranslated region-derived (5′-UTR and 3′-UTR) and transfer RNA-derived (tRFs) sRNAs, and riboswitches based on their genomic location. While trans-acting sRNAs originate from the intergenic regions and generally act on distant genes, cis-acting sRNAs are transcribed from the antisense strand of a coding open-reading frame. The riboswitches and 5′-UTR, and 3′-UTR-based sRNAs are located in the untranslated regions of a messenger RNA at the 5′ and 3′ ends of the transcript, respectively [75–77]. Recently, tRNA-derived sRNAs (tRFs) have also been reported from all three domains of life, and at least 12 tRFs of which six are conserved in multiple lineages, have been identified in the Genus Buchnera, a Gram-negative obligately intracellular symbiont of aphids [78,79]. The existence, expression, and function(s) of sRNAs in obligately intracellular bacteria is, thus, well documented and evidence suggests important riboregulatory roles for bacterial sRNAs in host–pathogen interactions.

As a novel direction in the field of Rickettsiology, we have recently established that Rickettsia species also encode both conserved (ssrS, ssrA, ffs and RNaseP_bact_a) as well as species- and/or strain-specific sRNAs. An initial comparative study based on genomic profiling of 16 different rickettsial strains belonging to 13 different species and including all phylogenetically defined groups predicted 1785 trans-acting sRNAs endowed with conserved -10 and -35 motifs similar to the consensus motifs observed in E. coli. However, rho-independent terminators could not be determined for a majority of predicted sRNAs, suggesting the possibility of existence of unknown alternative transcriptional termination mechanisms in bacterial genomes. Interestingly, the number of sRNAs varies between species, the sRNA repertoire does not correspond to the genome size, and the number of sRNAs predicted per million base (Mb) of the genome also varies between Rickettsia species within the same group [80]. This study offers the first glimpse of the influence of changes in genomic architecture attributed to spurious recombination events, pseudogenization, deletion bias and other driving forces of evolution, known to shape bacterial genomes, on the biogenesis and conservation of sRNAs in Rickettsia species. Although noticeable similarities in the sRNA repertoire are evident between different strains belonging to a Rickettsia species, it is unlikely that all conserved sRNAs are functional in all strains as some sRNAs may, in fact, represent transcriptional noise resulting from the conserved genomic synteny between strains [81].

Application of deep RNA sequencing to identify transcriptionally active sRNAs in rickettsial genomes during in vitro infection of human dermal microvascular endothelial cells has corroborated the computational predictive analysis and validated the expression of both well conserved and novel, species-specific trans- as well as cis-acting sRNAs in R. conorii and R. prowazekii representing prototypical spotted fever and typhus group Rickettsia species. Taken together, a total of 43 and 58 sRNAs have so far been identified based on their expression in R. conorii and R. prowazekii, respectively [82,83]. Independent biogenesis of several of these sRNAs has further been ascertained by northern blotting to demonstrate their steady-state expression in host cells and by determination of the 5′-transcriptional start sites of the noncoding transcripts by rapid amplification of complementary DNA ends. Interestingly, further studies aimed at comparative quantification of relative transcript abundance of R. conorii sRNAs during in vitro infection of human endothelial cells as the mammalian host cells and AAE2 (Amblyomma americanum) and RSE8 (Rhipicephalus sanguineus) tick vector cells reveal significant upregulation of two newly identified sRNAs (Rc_sR35 and Rc_sR45) in endothelial cells when compared directly with vector tick cells, providing the first evidence for alterations in the pattern of rickettsial sRNA expression depending on the host niche [82]. Importantly, target genes involved in multiple cellular and molecular pathways, governing cell wall biosynthesis, metabolite transport systems, amino acid biosynthesis, DNA repair and replication, and virulence, are all predicted to carry the potential for regulation by rickettsial sRNAs. In this regard, experimental validation of the interaction between an R. conorii trans-acting sRNA (Rc_sR42) and the mRNA coding for cytochrome d ubiquinol oxidase subunit I (cydA) yields the very first evidence for the existence of riboregulatory mechanism(s) during host–pathogen–vector interactions [82]. On a similar note, comparative analysis of both coding and noncoding RNA transcripts in R. prowazekii during in vitro infection of human endothelial cells vis-a-vis and tick AAE2 cells has also led to the identification of 93 cis- and trans-acting sRNAs that are uniquely expressed in the tick vector cells [84]. About 62% of these R. prowazekii sRNAs represent cis-acting sRNAs originating from the antisense strand of a coding gene. Although a relatively higher number of antisense sRNAs than that of trans-acting sRNAs in rickettsiae is not surprising in light of the evidence for similar trends in other bacterial genomes (Buchnera, for example), it is also important to consider that some of the identified antisense sRNAs in AT-rich genomes may actually represent transcriptional noise due to spurious promoters [85,86]. On the other hand, purifying selection is presumed to play a pivotal role in maintaining functional antisense sRNAs in rickettsial genomes, as in the case of post-transcriptional interactions between Buchnera antisense sRNA carB and its cognate protein CarB involved in arginine biosynthesis [79]. Thus, in consideration that bacterial sRNAs are now emerging as critically important mediators of several metabolic, adaptive and replicative processes relevant to their life cycles in different host environments, their importance in different groups of Rickettsia and involvement in the triad of host–pathogen–vector interplay and modes of action need to be comprehensively understood.

Progress in experimental animal models of infection

Several mouse models based on needle-based inoculation through standard parenteral routes of administration (intravenous, intraperitoneal and intradermal) have been described for different Rickettsia species and employed to investigate the mechanisms underlying host immune responses and pathogenesis (reviewed in [49]). Among these, intravenous infection of C3H/HeN mice-carrying wild-type TLR4 with different doses of R. conorii and R. typhi has been demonstrated to yield murine models closely recapitulating the pathogenesis, respectively, of spotted fever and typhus group rickettsioses in humans as evidenced by disseminated infection of microvascular endothelium, fatal outcomes in animals infected with a lethal dose (generically defined as ten-times LD₅₀), and onset of disease followed by recovery during infection with the sublethal dose (≤0.1X LD₅₀) [87,88]. The cynomolgus monkeys have also been used as the hosts susceptible to infection with both major pathogenic groups of rickettsiae and capable of mimicking the prominent features of clinical manifestations and pathophysiology of human rickettsioses [89]. In this study, killed and undiluted R. rickettsii provided complete protection against subsequent challenge inoculations, providing a basis for the evaluation of vaccine candidates in nonhuman primate models [89]. Recently, the efficacy of two recombinant antigens (Adr2 and OmpB-4) and inactivated whole-cell antigen (WCA) as potential vaccination strategies were tested in a canine model of R. rickettsii infection. This initial study not only documented the susceptibility of experimental canine hosts to severe RMSF disease but also yielded evidence for antigen-specific B-cell responses for both recombinant antigen-based and WCA-based vaccine formulations. Intriguingly, while only the WCA was determined to offer early complete protection against RMSF, the animals receiving the mix of recombinant antigens developed clinical signs of disease similar to the nonvaccinated, R. rickettsii-infected cohort of animals [90]. This study constitutes the development of another large animal model of rickettsial infections amenable for the evaluation of new diagnostics and treatment or vaccination approaches to combat or prevent rickettsial diseases in a biologically relevant host system.

Although needle-based inoculation models can effectively recapitulate the disease in susceptible hosts, the number of rickettsiae injected into the host when compared with natural transmission occurring through arthropod vector bites is likely quite high, thus making tick transmission models an attractive tool to capture the course of disease in humans. Importantly, experimental infection of susceptible dogs using Dermacentor variabilis ticks infected with R. rickettsii has been described to define the progression of disease as it might occur during natural transmission, allowing for the simultaneous evaluation of hematological, pathological and serological parameters, efficacy of antibiotic treatment and convalescence or relapse [91]. Another similar study has also documented the ability of R. sanguineus ticks to acquire R. rickettsii from experimentally infected dogs and subsequent transmission to guinea pigs, confirming the phenomenon of vector competence [92]. A recent study on this aspect investigated the infection dynamics of two Rickettsia species with varying degree of virulence to reveal that more virulent R. parkeri was transmitted at a much higher and faster rate as compared with the less virulent Ca. R. andeanae, suggesting the potential roles of rickettsial load in tick salivary glands and saliva in pathogen transmission [93]. A murine model of tick transmission for R. parkeri, an emerging SFG pathogenic species in the Western hemisphere, should facilitate the conduct of initial small-animal studies aimed at deciphering the complexities of the triad of host–pathogen–vector interactions. This study demonstrated the ability of R. parkeri transmitted by Amblyomma maculatum ticks as the natural vector of disease to produce eschars at the site of tick attachment and feeding, to enable recruitment of polymorphonuclear neutrophils in the bite lesions on host skin interface, and to elicit antibody responses in a murine model of tick-transmitted infection [94]. Collectively, continued progress on the development, in-depth characterization and application of such tick transmission models will be useful to strengthen our understanding of the contributions of tick-derived mediators in the saliva to pathogen transmission and serve as the foundation for focused investigations on identifying transcriptomic changes in the tick vector and associated rickettsial pathogens to design novel tactics to interfere with vector-to-host transmission or subsequent spread from the bite sites.

Roles of the vector microbiome in rickettsial transmission

As hematophagous arthropods serving as transmitting vectors in the natural life cycle, ticks carry a diverse group of commensal and symbiotic microorganisms, likely to play a key role in driving the transmission of a number of Rickettsia species and other pathogens with considerable implications for both human and animal health. Recent years have witnessed increased appreciation for the contributions of microbial composition and diversity in the maintenance or transmission of disease from the ticks of medical and veterinary importance. As such, this notion is supported by evidence suggesting that gut microbiome including resident endosymbionts plays an important role in the uptake of pathogens during the blood meal, their movement into salivary glands and other organs, and transmission into mammalian hosts [95–97]. In the context of rickettsial infection and maintenance, maternally inherited R. peacockii has been shown to interfere with the ability of virulent R. rickettsii to invade ovarial tissues and inhibit the development of R. rickettsii in epithelial and germinative ovarial cells, serving as the basis of the concept of rickettsial exclusion/interference/interspecies competition [4,98]. On a similar note, resistance of ovaries also precludes the coinfection of R. rhipicephali and R. montana in Dermacentor variabilis ticks [6]. In Amblyomma maculatum ticks colonized with R. parkeri, population density of an endosymbiont Francisella-like bacteria is negatively impacted, while that of another predominant endosymbiont Candidatus Midichloria mitochondrii remains unchanged [99]. Further evidence also suggests the presence of inherited or acquired nonpathogenic bacteria as a determinant of vector to host transmission of pathogenic rickettsiae. In the absence of rickettsemia in the host, horizontal transmission of pathogenic R. parkeri from infected to naive ticks during cofeeding is inhibited by the presence of vertically transmitted nonpathogenic Ca. R. andeanae in naive ticks [100]. However, artificial capillary feeding of the same ticks with both Rickettsia species exhibits a synergistic effect and results in increased levels of both Rickettsia species as compared with ticks fed on an individual basis [101]. Overall, these studies provide evidence for the roles of maternally inherited endosymbiotic microbiota in the regulation of transmission of tick-borne pathogens. Further mechanistic studies on the roles of tick microbiome as well as inherent immune system in the acquisition, colonization, maintenance and transmission of Rickettsia species will provide a better understanding of the complex relationship between the resident microbes and rickettsial pathogens in the ticks, opening up novel avenues for controlling pathogen transmission from vector to the host.

Conclusion & future perspective

In recent years, the incidence of tick-borne diseases in general and rickettsial infections in particular has continued to increase across all geographic regions of the world. Warmer temperatures due to global climate changes resulting in increased prevalence and activity of ticks as disease vectors coupled with the findings of rickettsial transmission by previously unrecognized tick species is presumed to contribute to higher incidence of RMSF as a serious infectious disease of humans. Despite significant progress in our understanding of the etiological agents and disease biology of rickettsioses since the discovery of R. rickettsii as a human pathogen in 1906, a complete understanding of the mechanisms by which rickettsial species escape from the phagosome to replicate in the host cytosol as ‘free’ intracellular bacteria is still lacking. Despite having reduced genomes, the function(s) of about 40–50% of the coding repertoire of most rickettsial genomes remains a virtual black box. This knowledge gap acquires additional significance because a number of these proteins are expected to be involved in carrying out diverse functions or activities important for the pathogens’ growth and replication and establishment of infection. It is now appreciated that rickettsiae acquire at least 51 different metabolites from the host cytosol to survive and replicate within the host [102]. Thus, in light of the obligate intracellular parasitism and tendency to evolve by genomic reduction, deciphering the function of entire coding transcriptome and cataloging of essential genes will lead to a much deeper understanding of the pathogens’ biology. Recent studies have also begun to identify rickettsial effector proteins that are secreted into host cytosol, yet much of the rickettsial effectome still remains to be discovered and understood. Application of comparative genomics with related bacterial species and identification of effector homologs in rickettsial genomes may provide valuable information on the functions of secreted effector proteins.

Epigenetic regulation of gene expression in both the host and the pathogen is now gaining ample significance with the identification of host noncoding RNAs as important determinants of immune responses and analysis of the roles of bacterial small RNAs in the regulation of the pathogen’s transcriptome. Recent work from our laboratory has unequivocally established the presence and expression of bacterial small RNAs within the rickettsial genomes and further identified the expression of host–cell-specific expression of enhancer lncRNAs and differential expression of a number of miRNAs in host endothelial cells during rickettsial infections. These foundational findings will pave the way for enhanced understanding of the contributions of riboregulatory mechanisms in host–pathogen–vector interactions and the pathogenesis of rickettsial diseases. Finally, functional genomics involving genetic manipulation of important genes involved in virulence, adhesion, invasion and other critical biological processes or their controlled expression in target host cells should facilitate the identification of potential therapeutic targets or vaccine candidates for the development of effective alternative and/or adjunct treatments or prevention strategies.

Executive summary.

• Members of the family Rickettsiaceae within Genus Rickettsia include both endosymbiotic and pathogenic Gram-negative bacteria that generally have life-cycle association with arthropod vectors.

• Pathogenic rickettsiae primarily infect microvascular endothelial cells lining the blood vessels of target organs, avoid complement-mediated lysis in the blood and may potentially hitchhike circulatory macrophages for dissemination within the host.

• As intracellular energy/metabolic parasites, rickettsiae exploit redundant adhesion-invasion and escape mechanisms to achieve host cell entry and access to nutrient-rich cytoplasm.

• Despite smaller genomes and the tendency to evolve by genomic reduction, the presence of multiple secretion systems suggests important but poorly understood roles for rickettsial effectors in manipulating host cell functions and immune response mechanisms.

• Recent demonstration of the presence of small RNAs in different Rickettsia species, their differential regulation in human host versus tick vector cells, and preliminary evidence for sRNA interactions with rickettsial genes indicates the involvement of riboregulatory mechanism in target niche-specific adaptation.

• Recent description of vector-transmitted infection in susceptible murine hosts should facilitate in-depth investigations at the vector–pathogen–host interface more closely recapitulating the natural disease course and substantiate our existing knowledge from established mouse models of both spotted fever and typhus rickettsioses based on the intravenous administration of varying infectious doses to define mechanisms underlying host immune responses and pathogenesis during severe disease and recovery.

• Furthermore, it is becoming increasingly evident that the resident microbiome of ticks and their tissues (salivary glands, midgut and ovaries) plays a critical role in pathogen acquisition, maintenance and/or transmission into the mammalian hosts. Continued investigations on mechanistic aspects of the complex microbiome–pathogen interplay and the roles of vector immune system should reveal important insights for developing novel strategies to control vector-to-host transmission.